pubs.acs.org/Langmuir © 2010 American Chemical Society
Accurately Tuning the Dispersity and Size of Palladium Particles on Carbon Spheres and Using Carbon Spheres/Palladium Composite as Support for Polyaniline in H2O2 Electrochemical Sensing Lirong Kong, Xiaofeng Lu,* Xiujie Bian, Wanjin Zhang,* and Ce Wang Alan G. MacDiarmid Institute, Jilin University, Changchun, P. R. China 130012 Received November 30, 2009. Revised Manuscript Received January 11, 2010 With an accurate control of the dispersity and size of the palladium nanoparticles (Pd NPs), carbon spheres/Pd NPs composite was prepared without any extra reducing agents. In order to fully understand the formation mechanism and find out the best condition for the fabrication of carbon/Pd composite spheres, the effects of temperature, reaction time, pH value, and the weight ratio of PdCl2 to carbon spheres on the morphology of the final products were investigated. A superior product with small (d = 7.66 nm, σ = 1.94 nm), homogeneously distributed Pd crystals was obtained at pH 7 and a reaction temperature of 70 °C in ethanol. The Pd NPs decorated carbon sphere was used as support for electroactive polyaniline (PANI) in our work because it could enhance their sensing properties which were afforded by catalytic Pd NPs and hydrophilic carbon spheres. The sensor based on carbon/Pd/PANI exhibited a high sensitivity of 656.0693 mA M-1 cm-2 and a detection limit of 5.48 μM toward the reduction of H2O2. In addition, the carbon/Pd/ PANI sensor also showed good selectivity between H2O2 and ascorbic acid.
Introduction Colloidal carbon micro- or nanospheres synthesized from sugar have been a hot topic and trend of researches in the fields of nanoscience because of its unique properties which could not be found in other colloidal spheres such as silica and polymer spheres.1-7 First, carbon spheres can be easily prepared from various kinds of low-cost sugar. Second, the preparative process of carbon spheres is “green chemistry” because the reactant is safe and the preparative process causes no contamination to the environment. In addition, the dispersity and size of carbon spheres can be precisely and easily controlled by adjusting the synthetic conditions.8,9 Up to now, only amorphous silica and some colloidal polymer spheres can be prepared with satisfactorily narrow size distributions as carbon spheres.10-13 However, silica and polymer spheres are always inert when they are used as supports for metal nanoparticles (NPs) or other functional materials; thus, surface modification is almost unavoidable before they are used as supports or templates.14,15 According to the advantages stated above, carbon spheres are more and more investigated and expected to be the substitute of amorphous silica and polymer spheres in the near future. *Corresponding authors: Tel þ86-431-85168924, Fax þ86-431-85168924, e-mail
[email protected] (X.L.); Tel þ86-431-85168924, Fax þ86-43185168924, e-mail
[email protected] (W.Z.). (1) Wang, Q.; Li, H.; Chen, L.; Huang, X. Carbon 2001, 39, 2211. (2) Sun, X.; Li, Y. Angew. Chem., Int. Ed. 2004, 43, 597. (3) Cakan, R. D.; Titirici, M. M.; Antonietti, M.; Cui, G. L.; Maier, J.; Hu, Y. S. Chem. Commun. 2008, 3759. (4) Liang, X. Z.; Yang, J. G. Catal. Lett. 2009, 132, 460. (5) Chen, C. Y.; Sun, X. D.; Jiang, X. C.; Niu, D.; Yu, A. B.; Liu, Z. G.; Li, J. G. Nanoscale Res. Lett. 2009, 4, 971. (6) Lou, X. W.; Chen, J. S.; Chen, P.; Archer, L. A. Chem. Mater. 2009, 21, 2868. (7) Sevilla, M.; Fuertes, A. B. Carbon 2009, 47, 2281. (8) Xia, Y. N.; Gates, B.; Yin, Y. D.; Lu, Y. Adv. Mater. 2000, 12, 693. (9) SchKrtl, W. Adv. Mater. 2000, 12, 1899. (10) Jiang, P.; Bertone, J. F.; Colvin, V. L. Science 2001, 291, 453. (11) Pol, V. G.; Grisaru, H.; Gedanken, A. Langmuir 2005, 21, 3635. (12) Wan, Y.; Min, Y. L.; Yu, S. H. Langmuir 2008, 24, 5024. (13) Phonthammachai, N.; White, T. J. Langmuir 2007, 23, 11421. (14) Caruso, F. Adv. Mater. 2001, 13, 11. (15) Wang, D. Y.; Rogach, A. L.; Caruso, F. Nano Lett. 2002, 2, 857.
Langmuir 2010, 26(8), 5985–5990
Noble metal NPs loaded colloidal carbon spheres have potential use in various aspects such as drug delivery, catalysis, energy storage, and separation technologies because of their large surface area, hydrophilicity, and high catalytic activity.16-20 Among all the noble metal NPs, palladium (Pd) NPs are mostly attached to colloidal carbon nanospheres because they not only have great catalytic, adsorptive, and other useful properties themselves but also can be used as additives for other functional materials to enhance their catalytic properties.21-23 Considering the excellent catalytic activity of Pd NPs and the fact that its price is relatively lower than gold and platinum, several scientists have prepared carbon/Pd NPs composite spheres using different approaches.2 For example, Sun et al. reported the fabrication of carbon/Pd NPs composite spheres in aqueous solution with out any other reducing agent,2 and Xu et al. reported the fabrication of carbon/Pd NPs composite spheres using NaBH4 as the reducing agent.17 However, the investigation on the preparation of carbon/Pd NPs composite spheres is still a great challenge because the dispersity and size of Pd NPs can still not be well controlled, and this will decrease the catalytic activity of Pd NPs. In our work, highly dispersed Pd NPs with small diameter have been successfully prepared on the surface of carbon spheres through an in situ reduction process. Through tuning the reactive conditions such as temperature, reaction time, pH, solvent, and (16) Hu, F. P.; Wang, Z. Y.; Li, Y. L.; Li, C. M.; Zhang, X.; Shen, P. K. J. Power Sources 2008, 177, 61. (17) Xu, C. W.; Cheng, L. Q.; Shen, P. K.; Liu, Y. L. Electrochem. Commun. 2007, 9, 997. (18) Makowski, P.; Cakan, R. D.; Antonietti, M.; Goettmann, F.; Titirici, M. M. Chem. Commun. 2008, 999. (19) Sun, Z. P.; Zhang, X. G.; Tong, H.; Liang, Y. Y.; Li, H. L. J. Colloid Interface Sci. 2009, 337, 614. (20) Zhu, Y.; Kang, Y. Y.; Zou, Z. Q.; Zhou, Q.; Zheng, J. W.; Xia, B. J.; Yang, H. Electrochem. Commun. 2008, 10, 802. (21) Kong, L. R.; Lu, X. F.; Jin, E.; Jiang, S.; Wang, C.; Zhang, W. J. Compos. Sci. Technol. 2009, 69, 561. (22) Gallon, B. J.; Kojima, R. W.; Kaner, R. B.; Diaconescu, P. L. Angew. Chem., Int. Ed. 2007, 46, 7251. (23) Kong, L. R.; Lu, X. F.; Jin, E.; Jiang, S.; Bian, X. J.; Zhang, W. J.; Wang, C. J. Solid State Chem. 2009, 182, 2081.
Published on Web 01/22/2010
DOI: 10.1021/la904509v
5985
Article
Kong et al.
the weight ratio of PdCl2 to carbon spheres, the dispersity and size of the Pd NPs can be well controlled. In addition, we use the asprepared carbon/Pd composite spheres as support for conductive polyaniline (PANI) to enhance the electrocatalytic activity for the reduction of H2O2.
Experimental Process Materials. Aniline monomer was distilled under reduced pressure and stored below 0 °C. Other reagents were of analytical grade and used without further purification including glucose, PdCl2 (Aldrich), absolute ethanol, FeCl3 3 6H2O, and PVP (K30). The water used in the experiments was distilled water. Preparation of Carbon Spheres. The procedure for preparing carbon spheres was in accordance with the reported method by Sun et al.2 Glucose (4 g) was dissolved in distilled water (40 mL) by stirring for 0.5 h, and then the solution were placed in a 40 mL Teflon-sealed autoclave and maintained at 180 °C for 4 or 6 h. The final product was isolated by centrifugation, cleaned by three cycles of centrifugation/washing/redispersion in water and in ethanol, and oven-dried at 60 °C for 8 h. Preparation of Carbon/Pd Composite Nanospheres. Carbon spheres (50 mg) were dispersed in absolute ethanol (25 mL) by ultrasoncating for more than 20 min. Then PdCl2 (11.1 mg) was added into the above suspension, followed by reacting at 70 °C for 1 h under vigorous stirring. After the reaction, the mixture was centrifuged, washed with distilled water twice, and dried at 60 °C for 8 h. Preparation of Carbon/Pd/PANI Composite Nanospheres. Carbon/Pd composite nanospheres (25 mg) were dispersed in ethanol solution (25 mL) in the presence of PVP (125 mg) by ultrasonication for 0.5 h and stirred for 0.5 h. After the mixture was aged for 24 h, it was centrifuged and redispersed in distilled water (25 mL). Then aniline (0.224 g) was drop added into the mixture. After it was stirred for 0.5 h, 5 mL of distilled water containing FeCl3 3 6H2O (0.6528 g) was quickly added into the above mixture under stirring, and the reaction lasted for 12 h before it was centrifuged, washed with ethanol, and dried at 60 °C for 8 h. Characterization. Transmission electron microscopy (TEM) experiments are performed on Hitachi 800 electron microscopes (Tokyo, Japan) with an acceleration voltage of 200 kV. X-ray diffraction (XRD) patterns are obtained with a Siemens D5005 diffractometer using Cu KR radiation. Inductively coupled plasma atomic spectroscopy (ICP) was performed on a Perkin-Elmer OPTIMA 3300DV. Analysis of the X-ray photoelectron spectra (XPS) was performed on an ESCLAB MKII using Al as the exciting source. A CHI660c electrochemical workstation (Shanghai CH Instruments, China) was used for electrochemical measurements. All electrochemical measurements were performed using a three-electrode configuration, consisting of the glassy carbon (GC) working electrode, an saturated calomel electrode (SCE) reference electrode, and a platinum wire counter electrode.
Results and Discussion Formation Mechanism and Morphology. As no other reducing agent was added into the dispersion, it is believed that Pd2þ ions which were first adsorbed to the oxygen atom on the surface of carbon nanospheres were also reduced to Pd0 by the surface groups such as -OH and -CHO. The weight percentage of Pd in carbon/Pd composite was analyzed to be 3.6% by ICP. In order to fully understand the main factors which affect the diameter and size distribution of the Pd NPs, the reactive conditions including the used kind of carbon spheres, temperature, added amounts of the reactants, reactive time, the kind of solvent, and pH were modified to show their effect on the morphology of the final products. The morphology of the final products was observed by TEM. Though the particle size and distribution may change when they get dried, in our experiments, 5986 DOI: 10.1021/la904509v
Figure 1. Carbon/Pd composite spheres prepared using (A) 6 hcarbon spheres and (B) 4 h-carbon spheres and carbon/Pd composite nanospheres prepared using 6 h-carbon spheres at (C) 20 °C and (D) 50 °C.
the Pd NPs were stabilized on the surface of carbon spheres, so after being separated with reactant and solvent by washing with water, their size and distribution would not undergo major changes as pure noble metal NPs when dried. First, different carbon spheres synthesized within different reaction time were used as the supports for Pd NPs to evaluate the size of the Pd NPs. It was found that when the carbon spheres synthesized within 6 h (6 h-carbon) were used, small Pd NPs with average diameter of 7.66 nm formed (Figure 1A) while big Pd particles with average diameter of 15-20 nm would appear on the surface of carbon spheres synthesized within 4 h (4 h-carbon) (Figure 1B). According to the results, it was obvious that the intrinsic properties of carbon spheres could influence the size of Pd NPs. Furthermore, if we observe their TEM images more carefully, it could be found that the 4 h-carbon was wrapped in a thin layer of cotton-shaped material; thus, their surface were not very smooth which was different from the smooth surface of 6 hcarbon. According to the results, the thin layer of cotton-shaped material should be polyelectrolyte which derived from the dehydration of the glucose. However, when the reduction time was prolonged to 6 h, such polyelectrolyte would carbonize into rigid carbon ring structure which has few reductive groups such as -OH and -CHO in it. The carbonization process was also reported by Sun et al.2 As more reductive groups (-OH and -CHO) were formed on the surface of 4 h-carbon, Pd2þ ions were reduced quickly on their surface and then aggregated into big particles. As shown in the electron diffraction pattern in Figure 1B, the diffraction ring was composed of several diffraction image points and the image reflected that the big Pd NPs were composed of several single-crystal Pd NPs which adopted different tropisms. The electron diffraction pattern further verified the explanation about the formation of big Pd particles in Figure 1B. As 6 h-carbon has fewer reductive groups on their surface, Pd NPs slowly formed and grew on it. When the reaction terminated, Pd NPs had not aggregated yet. The reaction temperature was another factor for the size control of Pd NPs on the surface of carbon spheres because the reactive groups were sensitive to the temperature. As a result, when the reaction temperature was tuned to 20 °C, the activity of the reaction groups was greatly decreased, so it is harder for them to reduce the Pd ions and fewer Pd NPs could form on the surface of carbon spheres. In addition, the size of Pd NPs was not very uniform (Figure 1C). Langmuir 2010, 26(8), 5985–5990
Kong et al.
Figure 2. Carbon/Pd composite spheres prepared using (A) double amount of PdCl2, (B) half amount of PdCl2, and carbon/Pd composite spheres prepared when the reaction time is (C) 5 min and (D) 2 h. The scale bar in the inset of (C) is 100 nm.
According to the formation process of Pd NPs, it is deduced that under low temperature the formation and growth of Pd NPs quickly terminated due to the low reductive activity, so the agglomeration of Pd NPs happened earlier than that happened under high temperatures and big Pd particles appeared (Figure 1C). Comparatively, when the reaction temperature is elevated to 50 °C, Pd NPs became smaller and exhibited a narrower size distribution (Figure 1D). However, the activity of the reductive groups were still not high enough because some carbon spheres were only coated with a few Pd NPs, and this would lower the catalytic activity of the nanocomposite. Considering the boiling point of the used solvent, 70 °C was a proper temperature for the reaction because the reaction proceeded mildly and carbon/Pd composite spheres with well morphology could be obtained under this temperature. The effect of the temperature to the morphology of carbon supported Pd NPs are similar to that reported by Phonthammachai et al.13 To further confirm the best condition for the synthesis of carbon/Pd composite spheres with well morphology, the used amount of the precursor PdCl2 and reaction time are tuned to change the morphology of the composite spheres. As expected, more big Pd particles appeared when the used amount of PdCl2 was doubled. From its TEM image (Figure 2A), It could be clearly observed that excess small Pd NPs aggregated together, and this further confirmed the fact that the formation of big Pd particles are due to the agglomeration of smaller Pd NPs, not for Ostwald ripening. On the contrary, when the used amount of PdCl2 was decreased to a half, the density of the Pd NPs was decreased but bigger Pd particles still appeared (Figure 2B). According to above aggregation theory and our observation, it was suggested that when the fewer PdCl2 were used off, the aggregation process was advanced. Investigation on the effect of reaction time to the morphology of the composite spheres could indicate the beginning and termination of formation, growth, and aggregation periods of Pd NPs.13 As shown in Figure 2C, when the reaction time was decreased to 5 min, Pd NPs were already evenly dispersed on the surface of carbon spheres, indicating that the Pd NPs quickly formed in 5 min. However, from Figure 2C, it was also observed that the formed Pd NPs were smaller and much fewer than those formed after reacting for 1 h because the formation process had not finished. After the formation process, the growth process of Pd NPs proceeded and had not terminated Langmuir 2010, 26(8), 5985–5990
Article
Figure 3. Carbon/Pd composite spheres prepared using (A) water, (B) water/ethanol (1:1) as the solvent, and carbon/Pd composite spheres prepared under (C) pH = 2 and (D) pH = 10.
in 1 h. However, further prolonging the reaction time to 2 h, big Pd particles appeared, indicating that the aggregation process happened. According to these results, the growth process terminated at a reaction time of more than 1 h and less than 2 h. After the investigation on the effect of above conditions to the morphology of the final products, it was confirmed that the happening of aggregation process mainly depended on time and any modification of the conditions which may cause the advance or delay of the aggregation process would greatly affect the final morphology. However, in order to make clear that if there were other conditions which could also affect the happening of aggregation process, the used kind of solvent and the pH of the reaction was changed to see the role of hydrogen bond interaction in the aggregation process. Clearly observed from Figure 3A, when changing the solvent from ethanol to water, no small Pd NPs (5-10 nm) could be found and big Pd block appeared even when the reaction time was decreased to 5 min (data was not shown). This phenomenon showed that hydrogen bond played a big role in the aggregation process. When the hydrogen bond interaction was strong between Pd NPs, the aggregation process would proceed during the formation and growth process. As a result, composite spheres with small Pd NPs (5-10 nm) could not be obtained at any time of the reaction period. Using the mixed solvent of ethanol and water (1:1), the aggregation process of Pd NPs was slowed and could be clearly observed. Moreover, as shown in Figure 3C,D, changing the pH of the reaction system to strong basicity would also cause the aggregation process to proceed at the same time with the formation and growth processes of Pd NPs. The result was caused by the increased hydrogen bond interaction between Pd NPs. Under a low pH, the formed Pd NPs adsorb Cl- ions on their surface, while in basic solution, they adsorb fewer Cl- ions and a great many OH- ions which have hydrogen bond interaction between them.13,24,25 As a result, the formed Pd NPs aggregate immediately during the formation and growth processes. However, as the hydrogen bond interaction are weaker in basic ethanol solution than that in aqueous solution, a few small Pd NPs could also be observed in Figure 3D but cannot be observed in Figure 3A. According to the experimental results, it is concluded that the diameter and size distribution of the Pd NPs can be greatly (24) Grisel, R. J. H.; Kooyman, P. J.; Nieuwenhuys, B. E. J. Catal. 2000, 191, 430. (25) Yin, D.; Qin, L.; Liu, J.; Li, C.; Jin, Y. J. Mol. Catal. A: Chem. 2005, 240, 40.
DOI: 10.1021/la904509v
5987
Article
Kong et al.
Figure 4. Carbon/Pd/PANI composite spheres when the molar ratio of FeCl3 to aniline is (A) 1:1 and (B) 2:1.
affected by the aggregation process, the happening of which is mainly determined by time and the hydrogen interaction between Pd NPs. Any conditional modification which will cause the change of the two factors will greatly affect the diameter and size distribution of the Pd NPs. Though PANI can exhibit high performance in catalysis, especially when composed with noble metal NPs, their application are still limited because pristine PANI cannot be easily dispersed in water and common organic solvents. As reported, acid doping is an effective way to make them dispersible,26,27 but in this way, the asprepared PANI can only be used in acidic environment. As a result, other methods need to be developed to disperse PANI in water. In our work, carbon/Pd composite spheres were used as support for the formation of PANI which can be easily dispersed in water for the strong interaction between carbon spheres and water. As shown in Figure 4A, a thin layer of PANI was coated onto carbon/Pd composite spheres by the PVP-mediated method and Pd NPs could be clearly observed. Moreover, the weight percentage of Pd in carbon/Pd/PANI composite spheres was analyzed to be 0.92%, which is lower than that in carbon/Pd composite spheres, indicating that PANI was successfully wrapped on the carbon/Pd spheres. In addition, the thickness of the PANI layer could be tuned by controlling the amount of added aniline or oxidant in the reaction. For example, when the used amount of oxidant was doubled in the reaction, composite spheres with thicker layer of PANI were observed (Figure 4B). However, as the layer was too thick, Pd NPs could not be clearly observed. In addition, when the amount of added aniline or oxidant was doubled, small isolated PANI which was not coated onto carbon/Pd spheres would appear, and this would cause the decrease of the catalytic activity. Structural Characterization of Carbon/Pd and Carbon/ Pd/PANI Composite Spheres. In the preparative process, PdCl2 was reduced by the surface groups on the carbon spheres. In order to confirm the existence and the kinds of these surface groups, FTIR spectra of carbon, carbon/Pd, and carbon/Pd/ PANI composite spheres were performed, and the existence of (26) Kuo, C. W.; Wen, T. C. Eur. Polym. J. 2008, 44, 3393. (27) Taylor, K. K.; Cole, C. V.; Berry, B. C.; Tito, V. J. Appl. Polym. Sci. 2007, 103, 2113.
5988 DOI: 10.1021/la904509v
Figure 5. (A) FTIR spectra and (B) XRD patterns of (a) carbon spheres, (b) carbon/Pd composite spheres, and (c) carbon/Pd/ PANI composite spheres.
PANI in carbon/Pd/PANI composite spheres could be also identified from the spectra (Figure 5A). As shown, the spectra of carbon and carbon/Pd spheres exhibited similar peaks at 1710 and 1620 cm-1, which attributed to CdO and CdC vibrations, respectively, supporting the concept of aromatization of glucose during hydrothermal treatment. Moreover, the bands in the range 1000-1300 cm-1 are corresponding to C-O stretching and O-H bending vibrations, implying the existence of large numbers of residual hydroxy groups on their surface.2 As Pd NPs do not exhibit characteristic peaks in the FTIR spectrum, there is no additional distinctive peak in the spectrum of carbon/ Pd when compared with the spectrum of carbon. After the coating of PANI, the characteristic peaks at around 3428 (N-H stretching), 1571, 1492 (CdC stretching deformation of quinoid and benzenoid ring respectively), 1300 (C-N stretching of secondary aromatic amine), 1123, and 832 cm-1 (out-of-plane deformation of C-H in the 1,4-disubtituted benzene ring) could be found in the spectrum of carbon/Pd/PANI composite spheres.21,28,29 The crystallite phase of carbon, Pd, and PANI were characterized in the XRD spectra (Figure 5B). From the spectrum of carbon spheres, a wide peak centered at 2θ = 20.55° was observed, indicating that the carbon matrix was amorphous.3 As expected, the characteristic peaks of Pd at 2θ = 40.1°, 46.6°, and 68.1° which corresponded to (111), (110), and (100) crystalline planes of Pd (JCPDS 5-0683) could be observed in the spectra of carbon/Pd and carbon/Pd/PANI composite spheres.30 (28) Lu, X. F.; Yu, Y. H.; Chen, L.; Mao, H.; Zhang, W. J.; Wei, Y. Chem. Commun. 2004, 1522. (29) Wang, Z. J.; Yuan, J. H.; Li, M. Y.; Han, D. X.; Zhang, Y. J.; Shen, Y. F.; Niu, L.; Ivaska, A. J. Electroanal. Chem. 2007, 599, 121. (30) Chang, G.; Oyamaand, M.; Hirao, K. J. Phys. Chem. B 2006, 110, 20362.
Langmuir 2010, 26(8), 5985–5990
Kong et al.
Figure 6. XPS spectra of (A) carbon spheres, (B) carbon/Pd composite spheres, and (C) carbon/Pd/PANI composite spheres.
These results confirmed that the Pd2þ ions have been already reduced to Pd0. The existence of PANI could also be confirmed by XRD spectrum in which PANI exhibited a broad band at a 2θ value of 25°, which was ascribed to the periodicity parallel to the polymer chains of PANI.31 However, the peak was partly overlapped by the peak of amorphous peak because their positions were too close. In order to confirm that the carbon/Pd spheres were fully wrapped in PANI layer, the surface elements of carbon, carbon/ Pd, and carbon/Pd/PANI spheres were analyzed on XPS spectra. In Figure 6B, the characteristic peak of Pd appeared while it could not be observed in the XPS spectrum of carbon/Pd/PANI composite spheres because the thick layer coating of PANI. Instead, in its spectrum, a new peak corresponding to nitrogen in PANI can be observed, and this further identifies the coreshell structure of the carbon/Pd/PANI composite spheres. (31) Pillalamarri, S. K.; Blum, F. D.; Tokuhiro, S. T.; Storyand, J. G.; Bertino, M. F. Chem. Mater. 2005, 17, 227.
Langmuir 2010, 26(8), 5985–5990
Article
Electrochemical Characterization of Carbon/Pd/PANI Composite Spheres. Detection of hydrogen peroxide (H2O2) using amperometric electrochemical methods is important because of the following reasons. First, H2O2 is an important analyte in the environmental and bioanalytical fields.32 Moreover, it is also a side product of some enzymatic reactions.33,34 Therefore, the accurate determination of H2O2 it can be used to indirectly detect other substances such as glucose in the presence of corresponding oxidases. Figure 7A,B shows the cyclic voltammograms (CVs) of bare GC electrode and carbon/Pd/PANI modified GC electrode in the absence and presence of H2O2 in 0.1 M PBS with pH 7.0. The bare GC electrode displayed only a small current response to H2O2 in Figure 7A. As shown in Figure 7B, the modified GC electrode exhibited a pair of redox peaks at 0.18 and -0.37 V which could be attributed to the redox process of PANI on the surface of GC electrode. However, after the addition of H2O2, the peak at -0.37 V gradually moves to -0.78 V, and the peak at 0.18 V moves to 0.30 V. The shift of the redox peaks of PANI were caused by the addition of H2O2 which retarded the redox process. Moreover, the peak value at -0.78 V increased after the addition of 1 mM H2O2, indicating that the anodic peak was caused by the oxidation of H2O2 at the surface of modified electrode. The increase in peak value on modified electrode was a lot more than that on bare GC electrode. In order to quantitatively determine the amount of H2O2 in practical sensor application, the amperometric responses of the modified electrode to successive injections of H2O2 at -0.7 V was recorded (Figure 7C). According to the values of amperometric responses and the surface area of the working electrode which was 0.070 65 cm2, the sensitivity of the modified GC electrode could be calculated to be 656.0693 mA M-1 cm-2 while the sensitivity of the bare GC electrode was calculated to be only 26.426 mA M-1 cm-2 according to the peak value in Figure 1A. Moreover, the sensor exhibited a linear dependence on H2O2 concentration (r = 0.994 17) with a detection limit of 5.48 μM (S/N = 3). The sensitivity is better than the sensitivity of 625 mA M-1 cm-2 reported by Liu et al.,35 but the detection limit is a little higher than 1.25 μM reported by Chen et al.36 The selectivity of the sensor was also evaluated. When 0.1 mM ascorbic acid (AA) was added, no response could be observed in the current-time curve. These results confirmed that the as-synthesized carbon/Pd/PANI composite spheres possessed efficient electrocatalytic activity toward H2O2, which provided a way of using it as a sensor to detect H2O2. Furthermore, the electrocatalytic behavior of the carbon/ Pd/PANI/GC electrode to H2O2 has been studied further with the change of scan rate. As shown in Figure 7D, with the increase of the scan rate from 5 to 200 mV/s, the cathodic peak current increased linearly with the square root of scan rate (the inset of Figure 7D), indicating that it was a diffusion-controlled process. Such carbon/Pd and carbon/Pd/PANI spheres can not only be applied in electrocatalysing the reduction of H2O2 and O2 and the oxidation of alcohol37 and formaldehyde38 but also has potential (32) Wolfbeis, O. S.; Durkop, A.; Wu, M.; Lin, Z. H. Angew. Chem., Int. Ed. 2002, 41, 4495. (33) Jia, J. B.; Wang, B. Q.; Wu, A. G.; Cheng, G. J.; Li, Z.; Dong, S. J. Anal. Chem. 2002, 74, 2217. (34) Guo, S. J.; Wang, E. K. Anal. Chim. Acta 2007, 598, 181. (35) Liu, Y.; Chu, Z. Y.; Jin, W. Q. Electrochem. Commun. 2009, 11, 484. (36) Chen, C. X.; Sun, C.; Gao, Y. H. Electrochem. Commun. 2009, 11, 450. (37) Hua, F. P.; Wang, Z. Y.; Li, Y. L.; Li, C. M.; Zhang, X.; Shen, P. K. J. Power Sources 2008, 177, 61. (38) Gao, G. Y.; Guo, D. J.; Li, H. L. J. Power Sources 2006, 162, 1094.
DOI: 10.1021/la904509v
5989
Article
Kong et al.
Figure 7. (A) CVs of GC electrode in PBS (pH = 7) with different concentrations of H2O2. (B) CVs of GC electrode (a, b) and modified GC electrode (c, d) in PBS (pH = 7) without (a, c) and with (b, d) 1 mM H2O2. (C) Successive amperometric response of the modified GC electrode to H2O2 in PBS (pH = 7) at -0.7 V vs SCE (inset: calibration curve for H2O2). (D) CVs of the modified GC electrode in PBS (pH 7.0) in the presence of 1 mM H2O2 at different scan rates (inset: plot of electrocatalytic current of H2O2 at -0.75 V vs v1/2).
applications in catalyzing various reactions such as Suzuki and Heck reactions39,40 and the reduction of organic dye or waste.21,23 Moreover, when carbon/Pd spheres are composed with photocatalytic materials such TiO2 and ZnO2, their photocatalytic activity can be greatly enhanced.41,42
Conclusion In summary, a simple method has been demonstrated to prepare carbon/Pd composite spheres in which the surface groups on carbon spheres acted both as adsorbing and reducing agent. In the final product, the dispersity and size of the Pd NPs could be precisely controlled by easily controlling the synthetic conditions such as temperature, reaction time, pH value, and the weight ratio (39) Corma, A.; Garcia, H.; Leyva, A. J. Mol. Catal. A: Chem. 2005, 230, 97. (40) Gallon, B. J.; Kojima, R. W.; Kaner, R. B.; Diaconescu, P. L. Angew. Chem., Int. Ed. 2007, 46, 7251. (41) Ge, L.; Xu, M. X.; Fang, H. B. Appl. Surf. Sci. 2006, 253, 2257. (42) Formo, E.; Lee, E.; Campbell, D.; Xia, Y. N. Nano Lett. 2008, 8, 668.
5990 DOI: 10.1021/la904509v
of PdCl2 to carbon spheres. We investigated the morphology changes caused by these modifications and constructed the aggregation theory which could explain the formation process of bulk Pd particles on the surface of the carbon spheres. Moreover, when the carbon spheres were changed to polymer or other inorganic colloid spheres, the theory was still applicable. At last, the as-prepared carbon/Pd composite spheres were used as support for electroactive materials PANI to enhance the electrocatalytic activity toward the reduction of H2O2 and help them to be dispersed in water. Note Added after ASAP Publication. This article was published on the web on January 22, 2010. Figure 7 has been modified. The correct version was published on March 18, 2010. Acknowledgment. This work was supported by a research grant from the National 973 project (No. 2007 CD936203) and National 863 project (2007 AA03z324).
Langmuir 2010, 26(8), 5985–5990
