Anal. Chem. 2010, 82, 8367–8370
Single-, Few-, and Multilayer Graphene Not Exhibiting Significant Advantages over Graphite Microparticles in Electroanalysis Madeline Shuhua Goh and Martin Pumera* Division of Chemistry & Biological Chemistry, School of Physical and Mathematical Sciences, Nanyang Technological University, 21 Nanyang Link, Singapore 637371 This report compares the electroanalytical performances of single- (G-SL), few- (G-FL), and multilayer graphene (G-ML), graphite microparticles, and edge-plane pyrolytic graphite electrodes in terms of sensitivity, linearity, and repeatability. We show that in the case of differential pulse voltammetric (DPV) detection of ascorbic acid, the sensitivity of a G-SL electrode is about 30% greater than that of G-ML and about 40% greater than graphite microparticles. However, in the case of DPV determination of uric acid, sensitivity is practically the same for all (G-SL, G-FL, and G-ML) and, importantly, the graphite microparticles do provide higher sensitivity than graphenes do for this analyte. Graphenes also do not provide a significant advantage in terms of repeatability. We pose the question of whether the efforts leading to the bulk method of producing single-layer graphene are justified for electroanalytical applications. Graphene is a two-dimensional sheet of sp2 bonded carbon atoms in hexagonal structures, exhibiting a structure of large polycyclic aromatic molecules.1 Its optical,2,3 mechanical,4 electrical,5 and electrochemical6,7 properties have been widely studied. Graphene-based nanomaterials are also being used in electrochemical applications, particularly in electroanalytical chemistry.8,9 Graphene-based nanomaterials can be classified according to the restriction of the axes perpendicular to the basal plane of the graphene, as single-layer graphene (G-SL), double-layer graphene,
* To whom correspondence should be addressed. E-mail:
[email protected]. (1) Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.; Zhang, Y.; Dubonos, S. V.; Grigorieva, I. V.; Firsov, A. A. Science 2004, 306, 666. (2) Nair, R. R.; Blake, P.; Grigorenko, A. N.; Novoselov, K. S.; Booth, T. J.; Stauber, T.; Peres, N. M. R.; Geim, A. K. Science 2008, 320, 1308. (3) Shi, Y.; Fang, W.; Zhang, K.; Zhang, W.; Li, L.-J. Small 2009, 5, 2005. (4) Bunch, J. S.; van der Zande, A. M.; Verbridge, S. S.; Frank, I. W.; Tanenbaum, D. M.; Parpia, J. M.; Craighead, H. G.; McEuen, P. L. Science 2007, 315, 490. (5) Wallace, P. R. Phys. Rev. 1947, 71, 622. (6) Pumera, M. Chem. Rec. 2009, 9, 211. (7) Pumera, M. Chem. Soc. Rev. 2010, DOI: 10.1039/c002690p. (8) Shao, Y.; Wang, J.; Wu, H.; Liu, J.; Aksay, I. A.; Lin, Y. Electroanalysis 2010, 22, 1027. (9) Pumera, M.; Ambrosi, A.; Bonanni, A.; Chang, E. L. K.; Poh, H. L. Trends Anal. Chem. 2010, DOI: 10.1016/j.trac.2010.04.048. 10.1021/ac101996m 2010 American Chemical Society Published on Web 09/07/2010
few-layer graphene (3-9 layers; G-FL), or multilayer graphene (10 or more layers; G-ML).10 Most of the methods used to prepare bulk quantities of graphene are based on a modified Brodie method, which results in multilayer graphene structures (99%) and only less than 1% in single-layer graphene.11 Only recently, methods for preparing bulk quantities of single layer graphene were discovered; most of the electrochemical and electroanalysis studies undertaken before now have been performed on multilayer graphene.9 Although there are about 100 articles on the electroanalytical chemistry of graphene and related materials,8,9 surprisingly, no study compares the performance of different graphene structures from the viewpoint of analytical chemistry, which relates to sensitivity, selectivity, linearity of response, and repeatability. We have recently demonstrated that the electrochemical response of graphene sheets is independent of the number of layers, from a single graphene sheet to multilayer stacked graphene, in terms of the oxidation potential of dopamine and ascorbic acid.12 This was a logical extension of Compton’s work where he demonstrated that heterogeneous electron transfer is fast at the edges of a graphene basal plane while the rate of heterogeneous electron transfer from the basal plane itself is several orders of magnitude lower.13 Now, the real question is, is there any advantage in using graphene and graphene-related nanomaterials from the analytical chemistry point of view? Here, we wish to report for the first time a comparison of the electroanalytical performances of single-layer graphene, few-layer graphene, and multilayer graphene. We compared their electroanalytical performances to those of graphite and edge plane pyrolytic graphite (EPPG) electrodes using two biologically and clinically important analytes, ascorbic acid and uric acid. EXPERIMENTAL SECTION Single-layer and few-layer graphene was obtained from NanoIntegris, IL, while multilayer graphene and graphite (size of particles 10-20 µm) were from Sigma-Aldrich. According to atomic force microscopy (AFM) analysis, the area of the G-SL and G-FL flakes was ∼10 000 nm2 and the average length of the edge of the (10) Rao, C. N. R.; Sood, A. K.; Subrahmanyam, K. S.; Govindaraj, A. Angew. Chem., Int. Ed. 2009, 48, 7752. (11) Ang, P. K.; Wang, S.; Bao, Q.; Thong, J. T. L.; Loh, K. P. ACS Nano 2009, 3, 3587. (12) Goh, M. S.; Pumera, M. Chem. Asian J. 2010, DOI: 10.1002/asia.201000437. (13) Davis, T. J.; Hyde, M. E.; Compton, R. G. Angew. Chem., Int. Ed. 2005, 44, 5251.
Analytical Chemistry, Vol. 82, No. 19, October 1, 2010
8367
Figure 1. Differential pulse voltammetric profiles of determination of ascorbic acid at concentrations of 0-10 mM of G-SL-, G-FL-, G-ML-, and graphite microparticle-modified electrodes. Conditions: 50 mM phosphate buffer, pH 7.4, step potential 5 mV, modulation amplitude 25 mV.
graphene was ∼100 nm. The content of the G-SL was 27% of single-layer, 48% of double-layer, 20% of triple-layer, and 5% of four or more layers of graphene. The content of G-FL was 6% of single-layer, 23% of double-layer, 27% of triple-layer, and 44% of four or more layers of graphene. Transmission electron microscopy (TEM) analysis of G-ML showed that the average length of the edge of G-ML was about 100 nm and the layer was ∼5 µm thick, effectively creating a multilayer graphene nanofiber.14 In the graphene based nanomaterials, the electrochemical surface area would be the same so there would only be the effect of the number of layers addressed.12 For electrochemical experiments, a glassy carbon electrode (CH Instruments, TX, product no. CHI104) was completely covered with 0.5 µg of corresponding material (using original colloidal solutions from the provider in the case of G-SL and G-FL or suspension in DMF in the case of G-ML and graphite microparticles). The edge-plane pyrolytic graphite (EPPG) electrode was obtained from Autolab, Japan. Voltammetric experiments were performed in 50 mM phosphate buffer (pH 7.4) using a microAutolabIII, Eco Chemie, The Netherlands. RESULTS AND DISCUSSION First, we focus on the main analytical parameters such as sensitivity of the response to the presence of analyte and the response and correlation coefficient of calibration curves. Figure 1 shows differential pulse voltammetric (DPV) curves of ascorbic acid on G-SL-, G-FL-, G-ML-, and graphite microparticle-modified (14) Ambrosi, A.; Sasaki, T.; Pumera, M. Chem. Asian J. 2010, 5, 266.
8368
Analytical Chemistry, Vol. 82, No. 19, October 1, 2010
electrodes and EPPG while Figure 2 shows similar data for uric acid. The data extracted from Figures 1 and 2 are presented in Tables 1 and 2, respectively. Table 1 shows that sensitivity to the oxidation of ascorbic acid is highest at G-SL (of 2.11 µA/mM) and decreases with an increased number of layers of graphene, from 1.72 of G-FL to 1.62 µA/mM of G-ML. The graphite-modified electrode exhibits sensitivity that is just slightly lower, as well as EPPG. The sensitivity of bare GC electrode was found to be 0.63 µA/mM. There is no dramatic difference between the sensitivities of these materials toward the detection of ascorbic acid; G-SL provides about 30% higher sensitivity than does G-ML. The correlation coefficient is greater than 0.99 in all cases but G-ML, where R is 0.9744. The analytical performance in the case of uric acid is interesting. The sensitivity of G-SL, G-FL, and G-ML is almost indistinguishable at 4.65, 4.54, and 4.47 µA/mM, respectively. Interestingly, the sensitivity of the graphite microparticle-based electrode is even slightly higher (5.11 µA/mM) than that of any of the graphene-based nanomaterials. The sensitivity of the bare GC electrode was found to be 0.96 µA/mM. The only sensitivity that is lower is the sensitivity of the EPPG electrode. In all cases, the linearity of the response is very good, with a correlation coefficient greater than 0.998 except for G-ML, where R is 0.9928. Next, we focused on selectivity. We compared the peak widths of DPV voltammograms because the sharper peaks lead to better peak-to-peak resolution and thus to the possibility of simultaneously determining mixtures of biomarkers and their interferences. We analyzed peak width at the half height (w1/2) of DPV
Figure 2. Differential pulse voltammetric profiles of determination of uric acid at concentrations of 0-1 mM of G-SL-, G-FL-, G-ML-, and graphite microparticle-modified electrodes. Conditions: 50 mM phosphate buffer, pH 7.4, step potential 5 mV, modulation amplitude 25 mV. Table 1. Sensitivity and Correlation Coefficient of Differential Pulse Voltammetric Determination of Ascorbic Acid of Different Electrode Materials Based on Calibration Curves from 0 to 10 mM with 2 mM Stepsa
a
material
slope (µA/mM)
R
G-SL G-FL G-ML graphite EPPG
2.11 (0.061) 1.72 (0.093) 1.62 (0.163) 1.42 (0.037) 1.04 (0.015)
0.9979 0.9942 0.9745 0.9907 0.9996
Standard deviations are stated in parentheses.
Table 2. Sensitivity and Correlation Coefficient of Differential Pulse Voltammetric Determination of Uric Acid of Different Electrode Materials Based on Calibration Curves from 0 to 1 mM with 0.2 mM Stepsa
a
materials
slope (µA/mM)
R
G-SL G-FL G-ML graphite EPPG
4.65 (0.094) 4.54 (0.082) 4.47 (0.182) 5.11 (0.143) 2.73 (0.044)
0.9992 0.9985 0.9928 0.9984 0.9989
Standard deviations are stated in parentheses.
peaks of 10 mM ascorbic acid on different surfaces and found that w1/2 for G-SL was 240 mV, for G-FL 260 mV, for G-ML 270 mV, for graphite microparticles 270 mV, and 282 mV for the GC bare electrode. Similar experiments for 1 mM uric acid
indicated a w1/2 of 125 mV for G-SL, 110 mV for G-FL, 105 mV for G-ML, 110 mV for graphite microparticles, and 45 mV for GC bare electrode. The relative standard deviation was for all w1/2 values lower than 3%. Therefore, there is no real advantage in using single-layer graphene over other forms of graphene, either G-FL or G-ML and graphite microparticles. We also studied the repeatability of DPV measurements on the different electrodes for both ascorbic acid and uric acid. We found that the relative standard deviation for 10 measurements of 5 mM ascorbic acid on G-SL is 7%, on G-FL 10%, on G-ML 21%, on graphite microparticles 4%, and on EPPG 11%. Therefore, of all the materials tested, it is obvious that graphite microparticles do provide the highest level of repeatability and the lowest relative standard deviation (RSD), while multilayer graphene exhibits the lowest level of repeatability. Because the acceptable RSD of analytical techniques is typically ∼5%, it is clear that the RSD of all graphene-based nanomaterials is analytically unacceptable. Next we studied the RSD for 10 repetitive measurements of 0.5 mM uric acid, where the RSD for G-SL was 13%, G-FL 6%, G-ML 10%, graphite microparticles 12%, and EPPG 8%. Again, there is no significant positive effect of graphene-based electrodes. We performed experiments for 0.1, 0.5, and 1 µg loadings, and there is no difference to our conclusions as the differences and similarities between the materials are consistent in this range. CONCLUSIONS In conclusion, we have compared the performances of single-, few-, and multilayer graphene and compared them to graphite Analytical Chemistry, Vol. 82, No. 19, October 1, 2010
8369
microparticles and EPPG. We have shown that while, in the case of ascorbic acid, G-SL is about 40% more sensitive than that of graphitic microparticles, in the case of uric acid, the sensitivity of graphenes of different thicknesses is comparable and is actually lower than the sensitivity of graphite microparticles. As far as repeatability is concerned, the RSD of graphene-based electrodes is comparable or worse than that of graphite microparticles. The performance of the nanomaterials was consistent in the whole tested range of 0.1-1.0 µg loading. Therefore, we would like to
8370
Analytical Chemistry, Vol. 82, No. 19, October 1, 2010
suggest that the electrochemist should carefully compare the analytical performance of graphene nanomaterials with graphite before claiming its advantageous properties.
Received for review July 28, 2010. Accepted August 25, 2010. AC101996M
