摘要:

hnexc hnexc hnfluo hnfluo Fluor Fluor Spacer e– e– e– LUMO LUMO HOMO HOMO HOMO HOMO Potential energy Potential energy Receptor Spacer Receptor EXCITED FLUOR EXCITED FLUOR FREERECEPTOR BOUNDEDRECEPTOR (a) (b) Guest Selective fluorescent chemosensor for fructose G. Pina Luisa, M. Grandaa, R. Bad�ýab and Marta Elena D�ýaz-Garc�ýa*b a Department of Analytical Chemistry, University of La Habana, Havana, Cuba b Department of Physical and Analytical Chemistry, University of Oviedo, Juli�an Claver�ýa 8, 33006 Oviedo, Spain A chemosensing system for the selective recognition of fructose based on a reverse photoinduced electron transfer process was developed.A fluorescent boronic acid, m-dansylaminophenylboronic acid, reacts with fructose to produce an electron transfer, which results in the fluorescence quenching of the dye. The addition of the sugar shifted the pKa from 8.13 to 7.80. A possible sensing mechanism is proposed. The analytical figures determined in a batch approach were detection limit 5 3 1026 M, repeatability of 1% at the 1 3 1024 M fructose level and linear calibration up to 3 3 1024 M.A flow injection system was also examined and after the experimental conditions had been optimized a selectivity study showed that only galactose (at a 1 : 2 fructose to galactose molar ratio) gave a positive deviation. Several food samples were analysed by the proposed flow injection procedure and the results agreed with those obtained using an enzymatic kit for food analysis.Keywords: Chemosensor; fructose; photoinduced electron transfer; food analysis; flow injection analysis As a result of research conducted over the last 5 years,1–4 it has become clear that synthetic molecular receptors that bind small molecules with signal transduction provide the basis for the development of selective chemical sensors for biologically important molecules (chemosensing). Chemosensors can possess various signal transduction systems (optical, electrochemical, etc.), fluorescence being one of the most useful response systems for optical readout.There are numerous mechanisms by which fluorescent signal transduction may be effected. The photoinduced electron transfer (PET) mechanism has been used as a tool of choice in fluorescence chemosensor design for metal ions,5–7 protons8–10 and small molecules.11,12 Typical PET sensors are molecules composed of three major moieties: a fluorophore, a spacer and a receptor (Fig. 1). The fluorophore moiety is usually based on a polycyclic aromatic system (e.g., anthracene, naphthalene, pyrene) insulated from a receptor specific for a particular species (e.g., a crown ether, an amine or a boronic acid) by at least one methylene group (the spacer). In the receptor-free situation, the fluorescence of the fluorophore–spacer–receptor system can be ‘switched off’ by the PET process [Fig. (1a)], which, in turn, can be suppressed by the entry of an ion or molecule into the receptor [Fig. 1(b)]. Fluorescence then becomes the dominant decay pathway for the excited fluorophore. There are also other mechanisms for inhibiting electron transfer such as conformational changes, micropolarity modulations and/or hydrogen bonding.4,13,14 The ‘guest’ species induces an increase in the ionization/oxidation potential of the receptor. This could be clarified in terms of orbital energy diagrams (Fig. 1). Continuing our search for sensitive and selective luminescent sensors for saccharides,15,16 we have started to apply the PET concept to the recognition of saccharides using a fluorescent boronic acid, m-dansylaminophenylboronic acid (DAPB), as chemosensor for fructose.Firstly, we addressed basic studies on batch fluorescence measurements and then a flow injection system for fructose determination was developed and applied to real sample analysis. A possible mechanism for the PET process of the chemosensor described here is outlined . Experimental Chemicals and solutions DAPB acid was obtained from Molecular Probes (Leiden, The Netherlands) and used as received.Sugars (fructose, glucose, sucrose and galactose) were obtained from Sigma (St. Louis, USA). Disodium hydrogenphosphate was purchased from Merck (Darmstadt, Germany). All other chemicals were of analytical-reagent grade and were used without further purification. A UV/VIS enzymic test kit for fructose determination in foodstuffs was obtained from Boehringer Mannheim (Mannheim, Germany).All aqueous solutions were prepared using water obtained from a Milli-Q system (Millipore, Bedford, MA, USA). A 1 3 1024 m DAPB stock standard solution was prepared by dissolving 3.7 mg of powder in 1 ml of dimethyl sulfoxide (DMSO) and diluting to 100 ml. Saccharide standards at different concentrations were prepared in the carrier solution (0.1 m phosphate buffer of the appropriate pH). When not in use, the solutions were stored at 4 °C.Instrumentation All fluorescence intensity measurements were made with an LS-50B luminescence spectrometer (Perkin-Elmer, Beaconsfield, UK), which has a xenon discharge excitation source (pulse width at peak half-height < 10 ms); instrumental parameters and processing data are controlled by the Fluorescence Data Manager software. The excitation and emission wavelengths were set at 324 and 508 nm, respectively. For the flow injection system a Hellman Model 176.52 flow-through Fig. 1 Basic photoinduced processes in a fluorophore–spacer–receptor system and orbital energy diagram: (a) in a free situation and (b) guest species bound. Analyst, January 1998, Vol. 123 (155–158) 155cell (25 ml) was used. All the experiments were carried out at 20 ± 2 °C. pH measurements were made with a MicropH 2002 pH meter (Crison, Barcelona, Spain). Flow injection manifold and general procedure The experimental set up for the system is shown in Fig. 2. A Minipuls 2 four-channel peristaltic pump (Gilson, Worthington, OH, USA) was used to generate the flow stream. A Type 50 PTFE four-way rotary valve (Omnifit, Cambridge, UK) provided with a 150 ml sample loop was used for sample introduction. PTFE tubing (0.8 mm id) and fittings were used for connecting the flow-through cell, the valve, the carrier solution, the DAPB and carrier reservoirs. If not stated otherwise, the carrier buffer consisted of 0.1 m Na2HPO4 buffer (pH 9.0 ± 0.2).The 7.0 3 1025 m DAPB solution and injected samples were always diluted with the carrier buffer. General procedure for fructose determination Calibration Fructose solutions (consisting of carrier solution plus a known volume of the fructose standard solution) were injected through the valve into the carrier flow stream. This stream was merged with the DAPB fluorimetric reagent, mixed in a reaction coil and driven to the detector system. Typical quenching fluorescence signals were obtained and the Stern–Volmer relationship was plotted against molar fructose concentration.Sample preparation A 30 g amount of whole wheat biscuit was homogenized in a mortar and 0.50007 g of the powdered sample was extracted with water (heated to 60 °C). The aqueous extract was transferred quantitatively into a 250 ml calibrated flask and diluted to volume with water. The solution was mixed and filtered to remove any suspended particles. A 1.25 ml volume of this solution was diluted to 25 ml with phosphate buffer.A 0.5082 g amount of homogenized natural jam was weighed into a 100 ml calibrated flask, mixed with water, diluted to volume and filtered. A 1.25 ml volume of clear solution was finally diluted to 25 ml with phosphate buffer. Volumes of 5.0 ml of commercial fruit juices were diluted to 100 ml in a calibrated flask without further preparation and 1.85 ml of this solution (which was clear and colourless) was diluted to 25 ml with phosphate buffer.The fructose content in all the samples was evaluated from a calibration graph obtained using fructose standards. The reference method used for the determination of fructose in foodstuffs was applied using a commercial kit. Results and discussion Spectral properties of DAPB in aqueous solfluorescence spectra of DAPB acid in a 1% dimethylsulfoxide aqueous solution displayed a fluorescence excitation and emission centered at 324 and 508 nm respectively.Upon addition of base, fluorescence enhanced (Fig. 3). From the pH fluorescence profile of the dye the two pKa values were calculated to be 3.50 and 8.13, respectively. The fluorescence-pH emission properties of DAPB acid were sensitive to the presence of millimolar concentration of fructose. In fact, the addition of fructose (Fig. 4) resulted in a shift of pKa2 to a lower pH region (pKA2a = 7.8) and in a dye fluorescence ‘switching off’ upon binding.This effect could be explained taking into account that the boron-nitrogen bond may be strengthened by the sugar uptake which, in turn, shifted the fluorescent reagent pK2a to a lower value. On the other hand, the PET process could occur through an inverse electron interaction along the p- and n-electronic system of the complex in a way such as that given in Fig. 4 (inserted scheme). A few PET sensors displaying this behaviour have been described in the literature.4,17,18 The switch-off factor (Stern-Volmer representation) was greater for fructose than for glucose, galactose or sucrose (Fig. 5), proving the validity of DAPB acid as a chemosensor for fructose.The analytical figures determined in batch were: calibration (Stern-Volmer relationship) was linear up to at least Fig. 2 Schematic diagram of the flow system. Carrier, 0.1 m Na2HPO4 (pH 9.0 ± 0.2); DAPB, 7.0 3 1025 m. Fig. 3 Fluorescence intensity versus pH profile of DAPB (1.0 3 1025 m) in 1% v/v DMSO.Fig. 4 Fluorescence intensity versus pH profile of DAPB (1.0 3 1025 m) in 1% v/v DMSO (~) with and (-) without fructose (1.5 3 1023 m). Reaction scheme: reversal mechanism of PET process. 156 Analyst, January 1998, Vol. 1233 3 1024 m (r = 0.998), the detection limit, based on three times the standard deviation of the fluorescence intensity of blank, was 5 3 1026 m and repeatability (peak height) at 1 3 1024 m fructose level was 1% (n = 10). Flow injection approach to fructose determination In order to explore the analytical possibilities of the fluorescent DAPB chemosensor in a flow injection approach for fructose determination, the flow set up shown in Fig. 2 was used and the experimental conditions were optimized. A compromise flow rate of 0.40 ml min21 in the carrier line and 0.30 ml min21 in the DAPB line, with a reaction coil length of 90 cm (0.8 cm id), were selected for high sensitivity and sample throughput. The influence of the volume of sample injected (1.6 3 1023 m fructose) on peak height and peak width was studied from 25 to 500 ml.In order to obtain the minimum peak width without losing sensitivity, an injection volume of 150 ml was selected for further work. The DAPB concentration in the carrier was kept constant at 7.031025 m. Analytical features The calibration graph for the system was based on the Stern– Volmer relationship, being the equation of the line y = 0.019c + 0.006, where c is the fructose molar concentration .The Stern–Volmer relationship was linear up to 2 3 1023 m (r = 0.998). The repeatability (peak height) at the 1 3 1023 m fructose level was 1% (n = 10). The detection limit was 5 3 1025 m for an injection volume of 150 ml. It should be possible to lower the linear range by increasing the injection volume to 300 ml or more. The sample throughput (1% carryover) was 50 h21. The interferences of glucose, galactose and sucrose in the flow system were studied.Different molar concentration ratios of fructose (1.6 3 1023 m) to these saccharides (1 : 4, 1 : 2, 1 : 1) and their effects on the relative intensity of the signal are shown in Table 1. Only galactose (at a molar ratio of 1 : 2) gave positive deviations. The proposed method was applied to the analysis of different food samples. As can be seen from Table 2, the results obtained for fructose agree satisfactorily with those obtained by an enzymatic kit for food analysis (Boehringer Mannheim).Conclusions The described chemosensing system can offer a rapid, reproducible method for the specific determination of fructose. Based on the results obtained so far, the chemosensing system developed for fructose shows significant potential for the determination of others sugars liable to be degraded to fructose if enzymes catalysing such degradation are present. This will be the case, for example, for sucrose after inserting in the flow system an invertase minireactor. Research in this direction is in progress.Another field of application, less exploited with boronic chemosensing probes for sugars, is in an optosensing system approach. Of special concern in this area is the development of sensitive analytical layers using different solid supports (zeolites, sol–gels, organogels, etc.). Work in this direction is also in progress. The authors gratefully acknowledge financial support from the CICYT (Comisi�on Interministerial de Ciencia y Tecnolog�ýa, Proyects AMB95/0476 and SAF96/1484).G. Pina Luis and M. Granda thank the University of Oviedo for its collaboration with Havana University. Rosana Bad�ýa thanks ICI (Instituto Iberoamericano de Cooperaci�on Internacional) for a grant. References 1 Rebek, J., Jr., Angew. Chem., Int. Ed. Engl., 1990, 29, 245. 2 Czarnik, A. W., ACS Symp. Ser., 1993, No. 538. 3 Fabbrizzi, L., Licchelli, M., Pallavicini, P., Perotti, A., and Sacchi, D., Angew. Chem., Int.Ed. Engl., 1994, 33, 1975. 4 Fabbrizzi, L., and Poggi, A., Chem. Soc. Rev., 1995, 24, 197. 5 Walkput, G. K., and Imperiali, B., J. Am. Chem Soc., 1996, 118, 3053. 6 Bissell, R. A., Calle, E., De Silva, A. P., De Silva S. A., Gunaratne, H. Q. N., Habibjiwan, J. L., Peiris, S. L. A., Rupasinghe, R. A. D. D., Samarasinghe, T. K. S. D., Sandanayake, K. R. A. S., and Soumillion, J. P., J. Chem. Soc., Perkin Trans. 2, 1992, 9, 1559. 7 Chae, M. Y., and Czarnik, A. W., J. Am. Chem.Soc., 1992, 114, 9704. 8 Shizuka, H., Nakamura, M., and Morita, T., J. Phys. Chem., 1979, 83, 2019. 9 Huston, M. E., Haider, K. W., and Czarnik, A. W., J. Am. Chem. Soc., 1988, 110, 4460. 10 Pardo, A., Poyato, J. M. L., Martin, E., Camacho, J. J., and Reyman, D., J. Lumin., 1990, 46, 381. 11 James, T. D., Samankumara-Sandanayake, K. R. A., and Shinkai, S., Angew. Chem., Int. Ed. Engl., 1994, 33, 2207. Fig. 5 Stern–Volmer plots for different sugars: (-) fructose; (5) galactose; (+) glucose; and (~) sucrose. Table 1 Results of selectivity studies Fructose : sugar Saccharide ratio Recovery (%) Galactose 1 : 2 115 1 : 1 100 1 : 0.5 101 Glucose 1 : 4 100 1 : 2 98 1 : 1 100 Sucrose 1 : 4 102 1 : 2 101 1 : 1 100 Table 2 Determination of fructose in foods Fructose found*/g per 100 g Proposed Sample method Enzymic kit Jam 35.58 ± 1.74 35.80 ± 0.65 Fruit juice 2.71 ± 0.02† 2.69 ± 0.02 Whole biscuit 11.29 ± 0.38 11.24 ± 0.49 * Number of analyses !4. † g per ml. Analyst, January 1998, Vol. 123 15712 Hamasaki, K., Ikeda, H., Nakamura, A., Ueno, A., Toda, F., Suzuki, I., and Osa, T., J. Am. Chem. Soc., 1993, 115, 5035. 13 Fabbrizzi, L., Licchelli, M., Pallavicini, P., Perotti, A., Taglietti, A., and Sacchi, D., Chem. Eur. J., 1996, 2, 75. 14 Czarnik, A. W., Acc. Chem. Res., 1994, 27, 302. 15 Valencia Gonz�alez, M. J, Liu, Y. M., D�ýaz Garc�ýa, M. E., and Sanz- Medel, A., Anal. Chim. Acta, 1993, 283, 439. 16 Valencia Gonz�alez, M. J., and D�ýaz Garc�ýa, M. E., Quim. Anal., 1994, 13, 90. 17 Yoon, J. Y., and Czarnik, A. W., Bioorg. Med. Chem., 1993, 1, 267. 18 De Santis, G., Fabbrizzi, L., Licchelli, M., Sardone, N., and Velders, A. H., Chem. Eur. J., 1996, 2, 1243. Paper 7/03778C Received May 30, 1997 Accepted September 16, 1997 158 Analyst, Ja

ISSN:0003-2654    

DOI:10.1039/a703778c

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

Chromium(III) hexacyanoferrate(II)-based chemical sensor for the cathodic determination of hydrogen peroxide Meng Shan Lin*, Ta Feng Tseng and Wei Chung Shih Department of Chemistry, Tamkang University, Tamsui 25137, Taiwan. E-mail: mslin@mail.tku.edu.tw A cathodic scheme to measure hydrogen peroxide by utilizing a membrane free chromium(III) hexacyanoferrate(II) based chemical sensor is described. The cluster is prepared simply by generating a chromium(III) hexacyanoferrate(II) cluster electrochemically on a rotating disk glassy carbon electrode. Subsequently, the hydrogen peroxide can be measured at 0 V versus Ag/AgCl. This approach significantly reduces interferences, especially those by easy oxidizable compounds such as catechol and ascorbic acid.Also, no significant oxygen interference is observed at this working potential. Several electrolytes were carefully investigated. The calibration curve was linear up to 1.3 mM (r = 0.9992) with a typical response of the sensor of about 5.6 s with injection of 0.05 mM hydrogen peroxide.The detection limit of this chemical sensor is 3.0 3 1028 M (S/N = 3) with a low-pass filter (time constant 0.1 s). The sensitivity of the sensor is 11.52 3 1026 A l mmol21 mm22 (r = 0.9992). The RSD of this sensor is 1.1%. Keywords: Chromium(III) hexacyanoferrate(II)-modified electrode; hydrogen peroxide; chemical sensors Measurement of hydrogen peroxide is of particular importance in both biomedical and environmental studies, with many applications in the plastics and food processing industries, etc.Hydrogen peroxide also is a by-product of many biological oxidative reactions and is therefore also a valuable parameter for monitoring biological reactions. Numerous schemes based on various principles have been developed for hydrogen peroxide determination, such as fluorimetric,1–3 fiber-optic,4–6 chemiluminescence7–10 and electrochemical11–18 approaches, for both liquid and air samples. Owing to the high overvoltage nature of hydrogen peroxide and possible interferences with detection, a direct oxidative detection scheme for hydrogen peroxide in the presence of easily oxidizable interference does not seem feasible. A catalyst-modified electrode may provide an adequate solution to reducing the high overvoltage problem.Recently, Wang and coworkers19 –21 utilized a series of precious metals, such as Pd and Rh, to reduce the applied potential for the determination of hydrogen peroxide through the coupling of a flavin-containing oxidase.A mixed-valence cluster is a polynuclear compound with two or more metal centers linked by a bridging ligand.22 Usually, a mixed-valence compound is prepared simply by mixing two solutions of an anionic metal ion and a cationic metal ion with a ligand to form an immediate precipitate. The electrons are delocalized over the whole molecule and the proof of this electronic interaction is observed from the separation of the redox potentials of the two metal centers.This characteristic of mixed-valence compounds, such as cobalt(iii) hexacyanoferrate(ii),23 shows catalytic properties for the reduction of hydrogen peroxide. Recently, Chi and Dong24 and Karyakin et al.25 developed a ‘first generation’ of glucose biosensors with Prussian Blue. However, such schemes are subject to interference from either ascorbic acid or oxygen. The basic nature of the redox signal is propagated directionally through chromium(iii) hexacyanoferrate(ii) at the control electrode.The basic electrochemical preparation of chromium( iii) hexacyanoferrate(ii) has been investigated previously. 26,27 In this paper, we report on the catalytic properties of chromium(iii) hexacyanoferrate(ii) obtained by replacing FeIII with CrIII ions during the formation of a mixed-valence compound. The modification procedure results in an energy change of the catalytic surface from the Fe–CN–Fe cluster and lower oxygen interference is experienced.Hence a sensing scheme with a lower overvoltage of hydrogen peroxide may be designed. The characteristic properties of chromium(iii) hexacyanoferrate( ii) and the feasibility of utilizing this new catalytic material in a chromium(iii) hexacyanoferrate(ii)-based sensor at cathodic potentials are demonstrated. In contrast to previous sensors, the significant improvements achieved with this new heterogeneous metal hexacyanoferrate-based chemical sensor offer low overvoltage, low background current, rapid response time, fewer interferences and membrane-free advantages.Experimental Chemicals and reagents All the reagent solutions were prepared with doubly distilled, de-ionized water obtained using a Milli-Q Reagent Water System (Millipore, Bedford, MA, USA). All the measurements were conducted in 0.1 m KCl (Riedel-de Ha�en, Seelze, Germany) (pH 3) or succinic acid (Riedel-de Ha�en) buffer solution (pH 6).A 20 ml volume of plating solution containing both 10 mm chromium(iii) nitrate [Cr(NO3)3·9H2O] (Riedel-de Ha�en) and 5 mm potassium hexacyanoferrate(iii) [K3Fe(CN)6] (Riedel-de Ha�en) was used for the preparation of the chromium( iii) hexacyanoferrate(ii)-modified electrode. Hydrogen peroxide solution (0.1 m), prepared by dissolving 86 ml of 35% hydrogen peroxide (Riedel-de Ha�en) in 10 ml of doubly distilled water, was stored at 4 °C when not in use.Ascorbic acid (0.2 m) (Riedel-de Ha�en), uric acid (Sigma, St. Louis, MO, USA), acetaminophen (Sigma), dopamine (Sigma), 1,4-dihydroxyquinone (Junsei, Tokyo, Japan), spermine (Sigma) and spermidine (Sigma) stock standard solutions were prepared with doubly distilled water just before use. Preparation of modified working electrode The chromium(iii) hexacyanoferrate(ii)-modified glassy carbon electrode was prepared in a 0.1 m KCl solution (pH 3) containing 10 mm Cr(NO3)3·9H2O and 5 mm K3Fe(CN)6.The electrode can be prepared either at a constant potential, 20.1 V, or by cycling the potential between 20.2 and +0.95 V versus Ag/AgCl for 1 h at 400 rev min21. Subsequently, the modified Analyst, January 1998, Vol. 123 (159–163) 159electrode was conditioned in 0.1 m KCl electrolyte (pH 3) for an additional 1 h. Apparatus For the steady-state amperometric measurement of hydrogen peroxide, a bipotentiostat (Model PAR 366A, EG&G Princeton Applied Research, Princeton, NJ, USA) was used to control the applied voltage in a three-electrode system for all amperometric experiments.The detection temperature of the electrochemical cell was maintained at 25 °C with a circulator (Model B402, Firstek Scientific, Taipei, Taiwan). The measurement cell containing a glassy carbon ring electrode (RDE 0032, EG&G Princeton Applied Research) with a motor-controlled rotor (Model 636, EG&G Princeton Applied Research), a laboratorymade 3 m Ag/AgCl reference electrode and a platinum wire counter electrode (local supplier) were used for rotating disk experiments. A cyclic voltammetric analyzer (BAS 100W, Bioanalytical Systems, West Lafayette, IN, USA) was used to check the glassy carbon electrode after polishing and to conduct all the cyclic voltammetric experiments.The data from steadystate amperometric experiments were recorded with a stripchart recorder (Linear Chart Recorder MF 1201, Alltech, Deerfield, IL, USA).pH measurements were carried out with a Suntex pH meter (local supplier). Elemental analysis of chromium(iii) hexacyanoferrate(ii) was conducted with a Heraeus (South Plainfield, NJ, USA) CHN-O Rapid Element Analyzer. An infrared spectrometer (FTD-40, Bio-Rad Labs., Richmond, CA, USA) was used to characterize the characteristic absorption of the chromium(iii) hexacyanoferrate(ii). Procedure The glassy carbon working electrode surface was polished with 1 3 1026 m diamond solution (Bioanalytical Systems) and then sonicated for 5 min in de-ionized water.The working electrode surface was polished with 1 31025 m alumina powder and then sonicated for 5 min twice. Subsequently, the electrode was checked with the cyclic voltammetric analyzer. A commercial glassy carbon electrode of 3 mm diameter (MF2012, Bioanalytical Systems) or a 0.221 in diameter commercial rotating glassy code (RDE0032, EG&G Princeton Applied Research) was used to study the electrochemical characteristics on a glassy carbon electrode on a BAS100W cyclic voltammetric analyzer.Solutions of 10 mm Cr(NO3)3·9H2O and 5 mm K3Fe(CN)6 were prepared in 0.1 m KCl solution (pH 3). The potential range of the working electrode was scanned between 2200 and +950 mV at a scan rate of 50 mV s21. Results and discussion In the proposed scheme, a membrane-free chromium(iii) hexacyanoferrate(ii)-based chemical sensor was prepared on both a laboratory-made rotating graphite disk electrode and a conventional glassy carbon electrode (from Bioanalytical Systems).This (CN)5–Cr–NC–Fe–(CN)5 cluster possesses semiconductor-like properties. The responses at various carbon electrodes, such as glassy carbon, graphite and carbon ink electrodes, are similar. However, the response on the platinum electrode is significantly lower than that on carbon, which may be attributed to the difference in the electronic structures of the electrode materials.Chromium(iii) hexacyanoferrate(ii) was prepared by two different methods. In the first, the cluster was prepared by mixing chromium(iii) nitrate and potassium hexacyanoferrate(iii). In contrast, the second method utilized a plating solution containing chromium(iii) chloride and potassium hexacyanoferrate(iii). The first results indicated that the former solution provides a better response to the injection of 5 3 1025 m hydrogen peroxide. In subsequent investigations, the deposition solution always involved a mixture of chromium(iii) nitrate and potassium hexacyanoferrate(iii) to generate the compound.Electrochemical characteristics of chromium(III) hexacyanoferrate(II) In our recent studies of chromium(iii) hexacyanoferrate(ii) and cobalt(ii) hexacyanoferrate(ii), we found that these clusters possess the unique property of acting as a catalyst for the cathodic reduction of hydrogen peroxide at a controlled working potential.This property of chromium(iii) hexacyanoferrate( ii) is of particular interest in chemical sensor development. In this compound, the hybrid orbitals provide suitable new orbitals, based on crystal field theory, for the catalytic reduction of hydrogen peroxide. In other words, the properties of this newly generated compound may provide a suitable energy surface for the catalytic path to occur. Chromium(iii) hexacyanoferrate(ii) {KCrIII[FeII(CN)6]· 8H2O} modified rotating glassy carbon disk chemical sensors can be prepared by scanning the potential, between 20.2 and +0.95 V (scan rate 50 mV s21) or by applying constant voltage of 20.1 V in 20 ml of 0.1 m KCl electrolyte (pH 3) containing both 10 mm Cr(NO3)3·9H2O and 5 mm K3Fe(CN)6. However, potential cycling provides a less noisy surface.Hence, subsequently, all the electrode preparations were carried out by the cyclic voltammetric method. A typical cyclic voltammogram characteristic of the chromium( iii) hexacyanoferrate(ii) modified glassy carbon electrode in 0.1 m KCl electrolyte is shown in Fig. 1(a). The voltammogram indicates that the two oxidative peaks of CrII[FeII(CN)6] are at +225 and +880 mV. The two reductive Fig. 1 Typical cyclic voltammograms for chromium(iii) hexacyanoferrate( ii) (a), potassium hexacyanoferrate(iii) (b) and chromium nitrate (c) on glassy carbon electrode in an electrolyte solution containing 0.1 m KCl (pH 3). Scan rate, 50 mV s21; operating temperature, 25 °C. 160 Analyst, January 1998, Vol. 123peaks are at +185 and +840 mV, respectively, which are significantly different from the responses from the voltammograms of either potassium hexacyanoferrate(iii) or chromium( iii) nitrate in Fig. 1(b) and (c), respectively. These two peaks are attributed to different mixed-valence stages such as (2,2) to (2,3) or (3,2) to (3,3). The structure of chromium(iii) hexacyanoferrate( ii) is face-centered cubic, based on a previous report.28 In contrast, we found that the response of iron hexacyanochromate-modified glassy carbon electrode is less sensitive to the injection of hydrogen peroxide if the N-terminus of the ligand faces the iron atom.We also confirmed that the composition of this cluster is KCrIII[FeII(CN)6·8H2O] with a Heraeus CHN–O Rapid Analyzer and from the typical infrared absorptions at 2083 cm21 and 530/539 cm21 for CN and FeCN/ FeC, respectively. These results indicate that the cluster we obtained is a Cr[Fe(CN)6] species, which is in agreement with previous studies.29 Catalysis of hydrogen peroxide The feasibility of using the chromium(iii) hexacyanoferrate(ii) based chemical sensor to measure hydrogen peroxide was investigated.The increasing responses, between +0.3 and 20.20 V upon five successive additions of 0.25 mm hydrogen peroxide, in the cyclic voltammogram in 0.1 m KCl electrolyte, are shown in Fig. 2. These successive cyclic voltammetric responses indicate the feasibility of utilizing this chromium(iii) hexacyanoferrate(ii) compound to develop an effective electrochemical scheme for hydrogen peroxide determination at potentials between +0.3 and 20.2 V.The cluster of chromium(iii) hexacyanoferrate(ii) is at its reduced stage (2,2) when the potential of the modified electrode is held at 20.1 V. The hydrogen peroxide oxidizes chromium( iii) hexacyanoferrate(ii) on the electrode from a (2,2) trapped state cluster into a (2,3) mixed-valence state.Subsequently, an electron is forced to add to the oxidized cluster through the mixed-valence state at the applied reductive potential. Hence, the characteristics and the optimum conditions for the chromium( iii) hexacyanoferrate(ii)-based electrode for hydrogen peroxide measurement were investigated. Optimization study of hydrogen peroxide chemical sensor The sensitivity of the chromium(iii) hexacyanoferrate(ii)- modified chemical sensor to the determination of hydrogen peroxide is dependent on the cluster amount.The deposition time for generating sufficient KCrIII[FeII(CN)6] was studied in 0.1 m KCl solution (pH 3). Fig. 3 shows the response current from the modified electrode to the addition of 5 3 1025 m hydrogen peroxide at various deposition times at 25 °C. The optimum deposition time of the modifying process is about 1 h and the sensitivity soon reaches the steady state thereafter. Also, the effect of the rotation speed was investigated.The Koutecky– Levich plot [current versus rotation speed in (rev min21)21 2] indicates a linear relationship up to 400 rev min21 (data not shown). This result shows that the response current approaches the steady state after 3600 rev min21 and indicates that the electrode reaction was maintained in the mass transportcontrolled region at 900 rev min21. Subsequently, all the experiments were carried out at 900 rev min21. Other parameters that affect the amperometric response of the chemical sensor are electrolyte, pH and applied potential.In the systematic studies, we found that the sizes of the cation and anion have profound effects. In the latter instance, chloride ion provides a greater significant current response than other potassium salts, such as potassium nitrate, acetate and phosphate. In the former instance, potassium was found to be the most suitable cation, followed by lithium, sodium, rubidium, cesium and ammonium ions.The anion effect may result from the size of the anions. In contrast, the potassium ion has the most compatible size for the counter ion, chloride. The electrolyte studies indicated that KCl electrolyte (16.38 3 1026 A) provides the best response to the injection of 5 3 1025 m hydrogen peroxide (pH 3). Fig. 4 shows the dependence of the solution pH on the injection of 5 3 1025 m hydrogen peroxide. The results indicate that the optimum pH range is > 6; the noise level gradually increases at pH > 7, which may be attributed to the conversion of chromium(iii) hexacyanoferrate(ii) into a chromium gel.Table 1 summarizes the various electrolytes’ effects. The results are in agreement with previous investigations.30 Other electrolytes investigated included succinic acid buffer (pH 6) Fig. 2 Six typical successive cyclic voltammograms from the mixedvalence cluster of the chromium(iii) hexacyanoferrate(ii)-modified glassy carbon electrode (blank) and five successive additions of 0.25 mm hydrogen peroxide (1–5).Other conditions as in Fig. 1. Fig. 3 Electrodeposition time study. The optimum deposition time for the chromium(iii) hexacyanoferrate(ii) modified glassy carbon electrode was evaluated with the steady-state response to the injection of 5 31025 m H2O2 at various deposition times. The potential of the rotating disk electrode was scanned between 20.2 and +0.95 V (versus Ag/AgCl) at 400 rev min21. Other conditions as in Fig. 1. Analyst, January 1998, Vol. 123 161(16.13 3 1026 A), phosphate buffer (pH 6) (13.63 3 1026 A), imidazole buffer (pH 6) (9.82 3 1026 A) and tris(hydroxymethyl) aminomethane (pH 6) (11.43 3 1026 A). In order to maintain pH stability throughout the experiments, a succinic acid buffer was used as electrolyte/buffer in subsequent studies. The ionic strength of the buffer solution was also investigated. The results indicated no significant response current change between 0.05 and 0.1 m.The response decreased gradually at electrolyte concentrations > 0.1 m (data not shown). In subsequent studies, 0.05 m succinic acid buffer (pH 6) was used. The solution pH was maintained at 6 in the succinic acid buffer (pKa1 = 4.2 and pKa2 = 5.6) for further investigations. The detection potential of this sensor is of particular importance since both the oxygen and hydrogen peroxide can act as oxidizing agents in this scheme. Hence proper selection of the potential to distinguish these two species is extremely critical, and a series of investigations were conducted.Fig. 5 shows the results for the chromium(iii) hexacyanoferrate(ii)- based chemical sensor with respect to both oxygen (B) and hydrogen peroxide (C). Fig. 5 indicates that there is no interference from oxygen at potentials > 0.0 V and the reduction response of hydrogen peroxide decreases as the applied potential becomes more positive (oxidative). Hence a detection potential of 0.0 V was adopted in subsequent studies.Analytical preference The deposited chromium(iii) hexacyanoferrate(ii) possesses a unique catalytic property which is of paramount importance for developing a rapid electrochemical scheme. The calibration plot was conducted by sequential injection of 1 3 1025–3 mm hydrogen peroxide and the linear range was evaluated from the calibration curve. Fig. 6 indicates that the linear range of the calibration plot is up to 1.3 mm.A typical response time of 5.6 Fig. 4 Effect of pH on the chromium(iii) hexacyanoferrate(ii)-modified electrode sensor for determining 5 3 1025 m H2O2. Electrolyte, 0.05 m succinic acid buffer; applied potential, 0.0 V (versus Ag/AgCl); rotation speed, 900 rev min21; operating temperature, 25 °C. Table 1 Electrolytes studied Response Response pH 3 (1026 A) pH 6 (1026 A) Acetate buffer 12.63 Succinic acid buffer 16.13 Glycine buffer 10.58 Succinic acid + acetate 11.78 Phosphate buffer 11.25 Succinic acid + 15.00 ammonium chloride Citric acid 13.54 Succinic acid + 16.40 potassium chloride Boric acid 12.38 Ammonium chloride 26.13 Succinic acid buffer 13.88 Ammonium chloride + 14.05 phosphate Ammonium chloride 21.67 Ammonium chloride + 19.63 acetate Ammonium chloride + citrate 18.35 Phosphate 13.63 Ammonium chloride + acetate 21.04 Imidazole 9.82 Tris(hydroxymethyl)- aminomethane 11.43 Response current was obtained on injection of 5 3 1025 m H2O2 under various buffer conditions; applied potential, 0.0 V (versus Ag/AgCl); other conditions as in Fig. 4. Fig. 5 Effect of potential on the chromium(iii) hexacyanoferrate(ii)- modified electrode sensor for determining 5 3 1025 m H2O2 (A) and ambient O2 (B) and the bare electrode (C) sensor for 5 3 1025 m H2O2, using 0.05 m succinic acid buffer (pH 6). Other conditions as in Fig. 4. Fig. 6 Typical calibration curve for the determination of H2O2 using the chromium(iii) hexacyanoferrate(ii)-modified glassy carbon electrode cathodically. Other conditions as in Fig. 5. 162 Analyst, January 1998, Vol. 123s was observed. This result indicates that a rapid measurement scheme for hydrogen peroxide and oxidase-based biosensors is possible for future development. A least-squares treatment of the standard addition data yielded a slope of the initial portion of the calibration plot of 285.2 3 1026 A l mmol21 (y = 285.24753x + 3.2588, with a correlation coefficient r = 0.9992) or a current sensitivity per unit area of 11.52 3 1026 A l mmol21.The sensitivity is comparable to that with Prussian Blue25 (10.00) or our previous result for cobalt(ii) hexacyanoferrate(ii)-based sensors23 (11.80). However, this study does not just concern an additional mixed-valence compound-based sensor. This report indicates that the operational voltage can be adjusted through the proper selection of the metal ion pairs to modify the surface energy of the catalyst in the measurement schemes.Hence this surface modification reduces the measurement voltage effectively down to 0.0 V, which results in few interferences being observed. In contrast to Prussian Blue, the significance of the chromium(iii) hexacyanoferrate( ii) scheme is the effectively decreased applied voltage and minimization of easily oxidizable compounds, such as ascorbic acid. Injection of 5 3 1028 m hydrogen peroxide was used to evaluate the detection limit in the presence of a low-pass filter (time constant = 0.1 s).The signal-to-noise characteristics (S/ N = 3) indicate that a detection limit of 3.0 3 1028 m can be achieved. In contrast, on injection of 1 3 1026 m hydrogen peroxide in the absence of a low-pass filter, the detection limit is 4.6 3 1027 m. This chemical sensor shows few interferences in the presence of 5 3 1025 m hydrogen peroxide. The interferents investigated included easily oxidizable compounds, such as l-ascorbic acid, cysteine, uric acid, tyrosine, acetaminophen, dopamine, 1,4-dihydroxyquinone, spermine and spermidine.The experiments were conducted by two different procedures, first, the addition of 2 3 1024 m interferent to a solution containing 5 3 1025 m hydrogen peroxide was investigated, second, the investigations were carried out in the absence of hydrogen peroxide. The results from both experiments indicated that no significant interferences were observed. This may be attributed to low operational voltage of 0.0 V (versus Ag/AgCl). The results of a series of 20 successive injections of 5 31025 m hydrogen peroxide solution were utilized to evaluate the precision of the response currents (conditions as in Fig. 6 ). The mean peak current was 16.10 3 1026 A, with a range of 15.80–16.44 31026 A and an RSD 1.1%. The half-life, t50%, of the chromium(iii) hexacyanoferrate(ii)-based chemical sensor preserved in the dry state after use and at room temperature, was found to be 14 d.However, if the chemical sensor was kept under vacuum the half-life was 16 d. The difference may be attributed to the gradual loss of chromium(iii) hexacyanoferrate( ii) during day-to-day operation. Conclusion We have presented a simple scheme to create a membrane-free mixed-valence based chemical sensor. Chromium(iii) hexacyanoferrate( ii) possesses catalytic properties with respect to hydrogen peroxide which can be utilized to design an interface for a chemical sensor.The chromium(iii) hexacyanoferrate(ii) containing interface was oxidized on contact with hydrogen peroxide. Subsequently, the oxidized cluster received an electron from the electrode at a reductive potential. This interface offers hydrogen peroxide specificity and without any significant oxidative interference from oxygen or other easily oxidizable compounds. Additionally, this work has successfully demonstrated the advantages of a low overvoltage for hydrogen peroxide determination, which may benefit the further implementation of this technique in both biological and environmental applications.It is important to investigate the nature of this catalytic reaction with hydrogen peroxide to design a better interface for hydrogen peroxide detection. We have utilized chromium(iii) to replace iron(iii) in Prussian Blue and successfully decreased the catalytic potential from +100 down to 0.0 mV. The small decrease in potential, however, significantly lowers the potential interference from easily oxidizable compounds, such as ascorbic acid.The characteristic properties of various mixedvalence clusters and their catalytic characteristics were carefully investigated before coupling an additional biological recognition element, flavin-containing enzymes, to this hydrogen peroxide-specific chemical sensor to develop chromium(iii) hexacyanoferrate(ii) biosensors. We gratefully acknowledge support from the National Science Council, Taiwan (Grant No.NSC 86-2113-M-032-013). We also thank Professor W. J. Wang and J. S. Lin for helpful discussions. References 1 Lazrus, A. L., Kok, G. L., Gitlin, S. N., Lind, J. A., and McLaren, S. E., Anal. Chem., 1985, 57, 917. 2 Peinado, J., Torbio, F., and Perez-Bendito, D., Anal. Chem., 1986, 58, 1725. 3 Holm, T. R., George, G. K., and Barcelona, M. J., Anal. Chem., 1987, 59, 582. 4 Abdel-Latif, M. S., and Guilbault, G. G., Anal. Chem., 1988, 60, 2671. 5 Fernandez-Romero, J. M., and Luque de Castro, M. D., Anal. Chem., 1993, 65, 3048. 6 Zhou, X., and Arnold, M. A., Anal. Chim. Acta, 1995, 304, 147. 7 Hoshino, H., and Hinze, W. L., Anal. Chem., 1987, 59, 496. 8 Ingvarsson, A., Flurer, C. L., Riehl, T. E., Thimmaiah, K. N., Williams, J. M., and Hinze, W. L., Anal. Chem., 1988, 60, 2047. 9 Hool, K., and Nieman, T. A., Anal. Chem., 1988, 60, 834. 10 Brestoviskky, A., Eisner, E. K., and Osteryoung, J., Anal. Chem., 1983, 55, 2063. 11 Hanaoka, N., Anal. Chem., 1989, 61, 1298. 12 Smit, M. H., and Cass, A. G., Anal. Chem., 1990, 62, 2429. 13 Pan, S., and Arnold, M. A., Anal. Chim. Acta, 1993, 283, 663. 14 Oungpipat, W., Alexander, P. W., and Southwell-Keely, P., Anal. Chim. Acta, 1995, 309, 35. 15 Vreeke, M. S., Yong, K. T., and Heller, A., Anal. Chem., 1995, 67, 4247. 16 Wright, J. D., Rawson, K. M., Ho, W. O., Athey, D., and McNeil, C. J., Biosens. Bioelectron., 1995, 10, 495. 17 Tatsuma, T., Ariyama, K., and Oyama, N., Anal. Chim. Acta, 1996, 318, 297. 18 Khayyami, M., Johansson, G., Kriz, D., Xie, B., Larsson, P. O., and Danielsson, B., Talanta, 1996, 43, 957. 19 Wang, J., and Angnes, L., Anal. Chem., 1992, 64, 457. 20 Wang, J., and Chen, Q., Anal. Chem., 1994, 66, 1007. 21 Wang, J., Liu, J., Chen, L., and Lu, F., Anal. Chem., 1994, 66, 3600. 22 Brown, D. V., Mixed-Valence Compounds, Reidel, Boston, 1980. 23 Lin, M. S., and Jan, B. I., Electroanalysis, 1997, 9, 340. 24 Chi, Q., and Dong, S., Anal. Chim. Acta, 1995, 310, 429. 25 Karyakin, A. A., Gitelmacher, O. V., and Karyakina, E. E., Anal. Chem., 1995, 67, 2419. 26 Jiang, M., Zhou, X., and Zhao, Z., J. Electroanal. Chem., 1990, 287, 389. 27 Gao, Z., J. Electroanal. Chem., 1994, 370, 95. 28 Brown, D. B., Shriver, D. F., and Schwartz, Inorg. Chem., 1968, 7, 77. 29 Shriver, D. F., Shriver, S. A., and Anderson, S. E., Inorg. Chem., 1965, 5, 725. 30 Itaya, K., Ataka, T., and Toshima, S., J. Am. Chem. Soc., 1982, 104, 4767. Paper 7/05207C Received July 21, 1997 Accepted September 24, 1997 Analyst, January 1998, Vol. 123 163

ISSN:0003-2654    

DOI:10.1039/a705207c

出版商:RSC    

年代:1998

数据来源: RSC