Anal. Chem. 2000, 72, 1724-1727
Correspondence
Selective Voltammetric and Amperometric Detection of Uric Acid with Oxidized Diamond Film Electrodes Elena Popa,† Yoshinobu Kubota,‡ Donald A. Tryk,† and Akira Fujishima*,†
Department of Applied Chemistry, School of Engineering, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-8656, Japan, and Department of Urology, School of Medicine, Yokohama City University, 3-9 Fukuura, Kanazawa-ku, Yokohama 236, Japan
Electrochemically anodized diamond film electrodes were used for selective detection of uric acid (UA) in the presence of high concentrations of ascorbic acid (AA) by differential pulse voltammetry and chronoamperometry. Because the oxidation peak potential for AA is ∼450 mV more positive than that for UA at anodized diamond electrodes, UA can be determined with very good selectivity. By use of chronoamperometry, linear calibration curves were obtained for UA over the concentration range up to 1 × 10-6 M in 0.1 M HClO4 solution, with the lowest experimental value measured being 5 × 10-8 M. This is consistent with the fact that a statistical analysis of the calibration curve yielded a detection limit of 1.5 × 10-8 M (S/N ) 3). AA in less than 20-fold excess does not interfere. The practical analytical utility of the method is demonstrated by the measurement of UA in human urine and serum without any preliminary treatment.
electrodes modified with polymers2,4-7 or enzymes8,9 or by electrochemical pretreatment.3,10 These voltammetric techniques are more selective, less costly, and less time-consuming than those based on colorimetry or spectrophotometry.10 However, even when very good selectivity and sensitivity have been achieved with the use of modified electrodes of this type, there are often complications, for example, adsorption phenomena2,6 or the need for renewal of the surface after each measurement.3 In the present work, we describe a simple electrochemical method for the sensitive and selective detection of UA, using anodized diamond film electrodes, which are free of such problems. An increasing number of reports have described the usefulness of conductive diamond electrodes for electroanalysis, which results from their unique electrochemical properties.11-24 The low and
Uric acid (2,6,8-trihydroxypurine, UA) and other oxypurines are the principal final products of purine metabolism in the human body.1 Abnormal levels of UA are symptoms of several diseases, including gout, hyperuricemia, and Lesch-Nyan disease.2 Uric acid and ascorbic acid coexist in biological fluids such as blood and urine. In a healthy human being, the typical concentration of UA in urine is in the millimolar range (∼2 mM), whereas that in blood is in the micromolar (120-450 µM) range.3,4 A major obstacle in monitoring this compound is interference from other electroactive constituents, such as ascorbic acid (AA), which oxidizes at close to the same potential as UA on various types of carbon electrodes,2,5 as well as on as-deposited diamond. Several electrochemical procedures for UA detection are based on oxidation at
(10)
(6) (7) (8) (9)
(11) (12) (13) (14) (15) (16) (17) (18) (19) (20)
†
The University of Tokyo. ‡ Yokohama City University. (1) Dryhurst, G. Electrochemistry of Biological Molecules; Academic Press: New York, 1977. (2) Zen, J. M.; Jou, J. J.; Ilangovan, G. Analyst 1998, 123, 1345. (3) Cai, X.; Kalcher, K.; Neuhold, C.; Ogorevc, B. Talanta 1994, 41, 407. (4) Zen, J. M.; Tang, J. S. Anal. Chem. 1995, 67, 1892. (5) Gao, Z.; Siow, K. S.; Ng, A.; Zhang, Y. Anal. Chim. Acta 1997, 343, 49.
1724 Analytical Chemistry, Vol. 72, No. 7, April 1, 2000
(21) (22) (23) (24)
Zen, J. M.; Chen, P. J. Anal. Chem. 1997, 69, 5087. Zen, J. M.; Hsu, C. T. Talanta 1998, 46, 1363. Gilmartin, M. A.; Hart, J. P.; Birch, B. Analyst 1992, 117, 1299. Nakaminami T.; Ito, S. I.; Kuwabata, S.; Yoneyama, H. Anal. Chem. 1999, 71, 1928. Strochkova, E. M.; Tur’yan, Ya. I.; Kuselman, I.; Shenhar, A. Talanta 1997, 44, 1923. Strojek, J. W.; Granger, M. C.; Dallas, T.; Holtz, M. W.; Swain, G. M. Anal. Chem. 1996, 68, 2031. Martin, H. B.; Argoitia, A.; Landau, U.; Anderson, A. B.; Angus, J. C. J. Electrochem. Soc. 1996, 143, L133. Jolley, S.; Koppang, M.; Jackson, T.; Swain, G. M. Anal. Chem. 1997, 69, 4099. Yano, T.; Tryk, D. A.; Hashimoto, K.; Fujishima, A. J. Electrochem. Soc. 1998, 145, 1870. Goeting, C. H.; Jones, F.; Foord, J. C.; Eklund, J. C.; Marken, F.; Compton, R. G.; Chalker, P. R.; Johnston, C. J. Electroanal. Chem. 1998, 442, 207. Chen, Q.; Granger, M. C.; Lister, T. E.; Swain, G. M. J. Electrochem. Soc. 1997, 144, 3806. Xu, J.; Swain, G. M. Anal. Chem. 1998, 70, 1502. Vinokur, N.; Miller, B.; Avyigal, Y.; Kalish, R. J. Electrochem. Soc. 1996, 143, L238. Vinokur, N.; Miller, B.; Avyigal, Y.; Kalish, R. J. Electrochem. Soc. 1999, 146, 125. Popa, E.; Notsu, H.; Miwa, T.; Tryk, D. A.; Fujishima, A. Electrochem. Solid State Lett. 1999, 2, 49. Fujishima, A.; Rao, T. N.; Popa, E.; Yagi, I.; Tryk, D. A. J. Electroanal. Chem. 1999, 473, 179. Yano, T.; Popa, E.; Tryk, D. A.; Hashimoto, K.; Fujishima, A. J. Electrochem. Soc. 1999, 146, 1081. Rao, T. N.; Yagi, I.; Miwa, T.; Tryk, D. A.; Fujishima, A. Anal. Chem. 1999, 71, 2506. Sarada, B. V.; Rao, T. N.; Tryk, D. A.; Fujishima, A. Anal. Chem., in press.
10.1021/ac990862m CCC: $19.00
© 2000 American Chemical Society Published on Web 03/03/2000
stable background current,11-14 the wide potential window in aqueous solutions,11,12 the poor adsorption of polar organic molecules,17,21 the relative insensitivity to oxygen present in the solution,14,22 and the long-term stability of the response17,20,23 are the most important features for application in analysis. Diamond thin-film electrodes can be conveniently chemically modified by electrochemically oxidizing them, and these anodized diamond electrodes retain the excellent attributes of the as-deposited diamond films. Recently we have shown that anodically oxidized diamond electrodes are useful for the sensitive and selective detection of dopamine in the presence of ascorbic acid.20 EXPERIMENTAL SECTION Uric acid (Aldrich), L-(+)-ascorbic acid (Wako), and concentrated HClO4 (Aldrich) were used as received. All other reagents were analytical reagent grade. All solutions were prepared in Milli-Q water (Millipore). The boron-doped polycrystalline diamond thin films were grown by use of microwave plasma-assisted chemical vapor deposition (CVD), as described elsewhere.14, 24 The electrochemical measurements were carried out in a single-compartment, three-electrode glass cell. A saturated calomel electrode (SCE) was used as the reference electrode, and Pt was used as the counter electrode. The diamond film electrode was mounted with epoxy (exposed area, 0.25 cm2). All of the measurements were carried out without removing dissolved oxygen. The electrochemical measurements were carried out with a HZ-3000 system (Hokuto Denko Corp.). For the electrochemical pretreatment that was used for these measurements, the diamond electrodes were immersed in 0.1 M KOH, and a potential of +2.6 V vs SCE was applied for 75 min.20,21 During this treatment there is active gas (oxygen) evolution. Urine samples were obtained from Yokohama City University, and the serum sample (human, lyophilized powder) was purchased from Sigma; all were stored in the dark at 4 °C. The standard addition method was used to evaluate the UA concentrations in all samples. The urine samples were prepared as follows: 10 µL of urine was diluted in 0.1 M HClO4 solution (dilution factor, 10 000), without either purification or filtration. The electroanalytical technique used was chronoamperometry (same experimental conditions as in Figure 3). RESULTS AND DISCUSSION Differential pulse (DP) voltammograms obtained for 50 µM UA in 0.1 M HClO4 solution on as-deposited and electrochemically treated diamond electrodes are shown in Figure 1A. Voltammetric pulse techniques have been shown to be very sensitive in the determination of micromolar amounts of this analyte.2-4 Even though the anodic peak potential (Ep,a) for uric acid oxidation is shifted by ∼100 mV (from +0.76 ( 0.02 V at as-deposited electrodes to +0.91 ( 0.02 V vs SCE at anodized electrodes), the current response remains sharp and well-defined. Furthermore, the current density is slightly greater at oxidized diamond electrodes than that for fresh electrodes, indicating that the effective electrode area increases after electrochemical treatment. As described in our previous publications,21,22 there are very slight changes in the surface morphology after anodic polarization; SEM images obtained after this treatment show that the diamond
A
B
Figure 1. Differential pulse voltammetric curves for the oxidation of (A) 50 µM UA and (B) 50 µM AA at untreated (dashed lines) and oxidized (solid lines) diamond electrodes in 0.1 M HClO4 solution; sweep rate, 20 mV s-1; pulse amplitude, 50 mV; pulse width, 50 ms; repetition time, 500 ms.
surface is slightly etched.22 The anodic charge passed during the 75-min treatment at 2.6 V was typically ∼625 mC cm2, some of which could be accounted for by the etching process. However, a significant portion of this charge is due to oxygen evolution. This point is being examined in more detail and will be reported soon.25 The calibration curve for the DPV peak current for UA oxidation vs UA concentration shows excellent linearity over a range of 1-75 µM, with a slope (µA/µM) of 0.063 and a correlation coefficient of 0.9989. In the case of AA, the differential pulse voltammetry (DPV) oxidation peak potential was 0.68 ( 0.02 V at untreated diamond electrodes, whereas, on treated diamond electrodes, the peak shifted to +1.15 ( 0.01 V vs SCE (Figure 1B). At the same concentration (50 µM) of both analytes, the oxidized diamond electrodes exhibit a stronger response for UA than for AA. The effects for both compounds are completely negated by hydrogen plasma treatment, for example, 30 min at 800 °C.26 We have examined several types of anodic treatment: electrochemical oxidation in acidic and basic solutions, thermal oxidation (at 500 °C for 60 min), and oxygen plasma treatment (100 W, for 60 s). Among these, the electrochemical treatment in (25) Popa, E.; Kubota, Y.; Tryk, D. A.; Fujishima, A., manuscript in preparation. (26) Notsu, H.; Tryk, D. A.; Fujishima, A., manuscript in preparation.
Analytical Chemistry, Vol. 72, No. 7, April 1, 2000
1725
basic solution (0.1 M KOH or NaOH) was the most effective in terms of time. If the anodization is carried out in acidic solution (0.1 M HClO4 solution), nearly the same effect is obtained, but a longer period of time (∼2 h) is required for the ascorbic acid oxidation peak potential to reach a limiting value, i.e., ∼+1.15 V vs SCE, with DPV and ∼+1.35 V, with cyclic voltammetry (CV). It is expected that a similar treatment of glassy carbon would damage the surface. These types of results will be reported separately.25 In the case of thermal oxidation and oxygen plasma treatment, the oxidation peak potential of AA reaches a limiting value of only ∼+1.15 V vs SCE (as measured by CV). This difference in behavior can perhaps be explained by the presence of different functional groups. In the case of electrochemical treatment, we believe that the major type of oxygen functional group is >CdO, which is believed to exist on (100) faces. These groups, particularly because they are highly oriented, are expected to form a surface dipolar field, which could repel neutral molecules that have oxygen-containing functional groups surrounding a central core, like AA, as noted in our previous publications.20,21 In the case of the thermal oxidation and oxygen plasma treatment, the surface oxygen may be incorporated in the bridged form (C-O-C), because of the higher possibility of the presence of defects. 27 However, interestingly, it appears that this effect (for all types of treatment) can also be neutralized by severe cathodic treatment without significantly affecting the surface oxygen content. We have proposed that the cathodic treatment either (a) converts >CdO groups to -C-OH groups, which would not have as strong a repulsive effect, or (b) injects p-type carriers in the form of subsurface hydrogen, so that electrons can tunnel more easily across the dipolar layer.27 We are in the process of trying to elucidate this effect in greater detail. To confirm that the redox process involves the diffusion of solution-phase UA, CV was measured as a function of potential sweep rate. The peak currents obtained were found to be linearly proportional to the square root of the sweep rate, with a nearzero intercept (plot not shown). In addition, chronocoulometric measurements were used in order to determine whether adsorption is significant. The chronocoulometric Q - t1/2 curves for the oxidation of 1 µM UA indicate a coverage of 1.03 pmol cm-2; on the basis of a molecular area of 0.665 nm2 for a horizontal orientation, and 0.203 nm2 for a vertical orientation of the UA molecule at the electrode, this coverage represents from 0.41 to 0.12% of a monolayer. This result indicates negligible UA adsorption on the oxidized diamond electrodes. Thus, the oxidation of UA can be concluded to involve predominantly the solution-phase species. Various possible interfering substances, such as AA, urea, purine, glucose, and oxalate, were examined for their effects on the electrochemical determination of UA. We have carried out DPV measurements for 5 × 10-3 M UA in the absence and presence of urea, purine, glucose, and oxalate (10-2 M each) and there was negligible interference for all of these compounds. The interference effect of AA was investigated in further detail. Figure 2 shows DP voltammograms for UA in the presence of a 5-fold excess of AA at an oxidized diamond electrode. The oxidation (27) Notsu, H.; Yagi, I.; Tatsuma, T.; Tryk, D. A.; Fujishima, A., Electrochem. Solid State Lett. 1999, 2, 522.
1726 Analytical Chemistry, Vol. 72, No. 7, April 1, 2000
Figure 2. Differential pulse voltammogram for a mixture of 50 µM UA and 0.25 mM AA at an oxidized diamond electrode in 0.1 M HClO4 solution. Other conditions are the same as in Figure 1.
peaks for UA and AA are separated clearly, with a potential difference of ∼ 240 mV. The ability to discriminate between the UA and AA DPV oxidation peaks was examined as a function of pH. The results show that, with increasing pH, a negative shift in Ep,a occurs for both compounds, with a more rapid decrease in the case of AA. The difference between the Ep,a values for AA and UA was on the order of 250 mV in the pH range from 1 to 2.5. For pH values greater than 2.5, the difference decreased to less than 100 mV, as measured in separate solutions, and thus, the UA and AA oxidation peaks could not be resolved in solutions containing both species. The peak potential vs pH dependence appears to be explainable on the basis of the coupling of the electron- and protontransfer steps, as treated theoretically by Laviron,28 but the details remain to be worked out. To extend the detection limits for UA to lower concentrations, chronoamperometric measurements were employed. Figure 3 shows a chronoamerometric curve for 50 nM UA in the presence of a 20-fold excess of AA (curve c). There was a relatively minor change in the current magnitude obtained for a 0.1 M HClO4 solution when AA was added to the solution (∼5%), comparing curves a and b. Linear chronoamperometric calibration curves were obtained over the range 5 × 10-8-1 × 10-6 M in 0.1 M HClO4 solution in the absence and in the presence of 1 and 5 µM AA. In Table 1 the coefficients from the linear regression analyses are given, together with the current densities and the signal-tonoise ratios for the lowest concentration measured (50 nM), which is among the lowest values reported in the literature. To our knowledge, the lowest experimental detection limit (30 nM) was reported by Cai et al.,3 but preanodization was required before each run, and the electrode material was found to be swollen after more than 30 measurements. In the present work, the detection limit was also calculated by statistical analysis29 (15 nM at S/N ) 3), which indicates that concentrations of this order can be (28) Laviron, E. J. Electroanal. Chem. 1980, 109, 57. (29) Skoog, D. A.; Holler, F. J.; Nieman, T. A. Principles of Instrumental Analysis, 5th ed.; Saunders College Publishing: Philadelphia, 1998; pp 12-14.
Table 2. Determination of UA in Real Samples with Oxidized Diamond Electrodesa parameter
urine 1
urine 2
serum
original value (nM) spike (nM) recovery (%) total value (mM)a
205 200 102 2.05
198 200 97.7 1.98
587 600 98.4 0.120
a The total value was obtained by multiplying the detected value by the appropriate dilution factor (10 000 for urine and 200 for serum samples).
Figure 3. Chronoamperometric responses at an oxidized diamond electrode for (a) 0.1 M HClO4, (b) 1 µM AA, 0.1 M HClO4, and (c) 1 µM AA + 50 nM UA, 0.1 M HClO4. The initial potential was 0.0, and the potential was stepped to +0.93 V; sampling time, 500 ms. A magnified view of the two curves (a and c) at the current maximum is shown in the inset. Table 1. Chronoamperometric Data for Uric Acid in 0.1 M HClO4 at Oxidized Diamond Electrodesa parameter (nA/cm2)b
i S/Nc slope (nA/nM) intercept (nA) correlation coeff
without AA
with 1 µM AA
with 5 µM AA
2.4 48 0.0122 -0.195 0.9991
2.56 52 0.0125 -0.143 0.9992
2.64 43 0.0131 -0.072 0.9996
a The experimental conditions were the same as in Figure 3. Current densities have been background-subtracted by subtraction of signals obtained with 0.1 M HClO4 alone (column 1), 0.1 M HClO4 + 1 µM AA (column 2), and 0.1 M HClO4 + 5 µM AA (column 3) and were sampled at 10 s after the beginning of the potential step. c S/N ) ifaradaic/ibaseline noise. b
measured if multiple runs, in this case 20, are averaged. The response variability, RSD, for the series of 20 measurements was 1.15%. The selectivity was also checked in the presence of 5 µM AA, and good linearity was obtained over the same concentration range for UA concentration as a function of the chronoamperometric current. However, in this case, the current magnitudes were an average of ∼28% larger, compared with those obtained with HClO4 alone.
Two human urine samples and one serum sample were examined with the present method. To fall within the linear range, the urine samples were diluted by a factor of 10 000 and the blood sample was diluted by a factor of 200 (Table 2). The dilution process greatly decreases possible matrix effects for biological samples.2 To confirm the validity of the results, the samples were spiked with UA at levels similar to those found in the samples themselves. The recoveries for the spiked samples were excellent, varying between 97.7 and 102%. It is interesting to note that the uric acid concentrations in urine (as measured in more than 20 clinical samples) vary over a relatively wide range, between 1.5 and 10 mM, depending on patient health and diet. These results will be reported in a separate publication.25 CONCLUSIONS This study has demonstrated that oxidized diamond electrodes can be successfully applied for the detection of uric acid with excellent sensitivity and selectivity. AA in less than 20-fold excess does not interfere. Significant advantages of these electrodes are their excellent stability (repeatedly used for more than 3 months), very good reproducibility, and ease of preparation, offering a good possibility for extending this technique to the routine analysis of uric acid in clinical samples. ACKNOWLEDGMENT This research was supported by the Japan Society for the Promotion of Science (JSPS) Research for the Future Program (Exploratory Research on Novel Materials and Substances for Next-Generation Industries). Received for review August 2, 1999. Accepted January 24, 2000. AC990862M
Analytical Chemistry, Vol. 72, No. 7, April 1, 2000
1727
