Enclosure I;

DESCRIPTION

Polyaromatic hydrocarbons (PAHs) are molecular subunits of graphene and carbon nanotubes (CNTs) [1-3]. They are released into the environment mainly by incomplete combustion of fossil fuels, coal tar, diesel engine exhausts, aviation exhausts, and cigarette smoke [4]. Because of their ubiquitous occurrence, the bioaccumulation potential and carcinogenic activity (especially lung cancer, the principal category of adult cancer death in the U.S. population) of PAHs have generated significant environmental concern [5-7]. Hence, methods for the reduction of PAH toxicity are extremely important for environmental and health reasons. Oxidation of PAHs is a major way to degrade and, inturn, reduce their toxic activities. It is generally known that PAHs are not oxidized readily under mild conditions.

Several stringent chemical decomposition methods have been reported for this purpose, for example (with respect to anthracene oxidation), photocatalytic [8-13], photo-electrochemical [14], and sono-electrochemical oxidation [15], Fenton reagent coupled biodegradation [16-19], electro-enzymatic oxidation [20-23], gamma-ray irradiation [24], the ozonization reaction [25,26], high temperature vapor-phase oxidation on a catalytic surface (300° C) [27,28], and liquid-phase catalytic chemical oxidation at approximately 90° C [29,30]. Electrochemical oxidation of PAHs has rarely been achieved because of their particularly unreactive character. However, electrochemical oxidation of PAHs with boron-doped diamond (BOD) in a non-aqueous/aqueous medium at high voltage (1.5 V vs. Ag/AgCl) [31] and thiol-functionalized self-assembled monolayers (SAMs) on gold surface electrodes [14] have been reported.

Although the methods provide new electrochemical oxidation methods for PAH degradation, the synthetic difficulty in functionalizing PAH and time consuming (SAM) and expensive (diamond) electrode preparation routes restrict the methods for further practical degradation applications. Herein, we are reporting a simple and facile surface-bound electrochemical oxidation of three linear PAHs, namely, naphthalene (NP), anthracene (AN), and tetracene (TC) to surface-confined redox active species on multiwalled carbon nanotube (CNT) modified electrodes (CNT@PAH) under pH neutral conditions. Carbon nanotubes were reported to be an excellent adsorbent for removal of PAHs from the environment [32-36]. The pi-pi interaction between the CNT and PAH is key for the efficient adsorption process over conventional activated carbon. Herein, we are interested in examining the surface of PAHs physisorbed on multiwalled carbon nanotubes (MWCNT@PAHads, ads = physically adsorbed, MWCNT=multiwalled carbon nanotube) by electrochemical methods.

Interestingly, we found that the MWCNT@PAHadS may be oxidized electrochemically, in the case that the PAHads is derivative-free and non-functionalized, to highly redox active and stable polyaromatic hydroquinones or di-hydroxy compounds on a MWCNT surface. Initial experiments were all carried out with anthracene (AN) as system-modeling PAHs. Surface-bound AN was prepared by drop-casting of a solution of AN in ethanol (5 p.L stock solution of 2 mg of AN dissolved in 0.5 mL ethanol) on a clean gold electrode (Scheme 1, step 1). First, bare gold electrode coated with AN (ANadS/Au) was prepared and subjected to cyclic voltammetric (CV) experiments in a potential window of -1.0 to +1.0 V versus Ag/AgCl in pH 7 phosphate buffer solution (PBS). As can be seen in Figure 1 Aa, the AN modification does not show any faradic response on the Au working electrode (Scheme 1, step 2). The response was the same on a glassy carbon modified electrode (GCE) as well (Figure 1 Ab). These observations indicate the impracticality of oxidizing AN under normal circumstances in aqueous solution.

For this reason SAM-PAH and BOD systems were used for the electro-oxidation processes previously described [14, 31]. The surface-bound AN experiment was repeated with pristine MWCNT-modified Au (Au/MWCNT@ANadS)- The Au/MWCNT@ANads was prepared by dispersing CNTs (1 mg) in ethanol (500 uL) and sonicating for 10 min. This mixture was then drop cast (5 uL) onto a clean Au electrode and AN was subsequently adsorbed (Scheme 1, step 3). As can be seen in Figure 1 Ac, in the first cycle a large irreversible anodic peak at 1.0 V is observed. Subsequently, a weak reduction peak at -0.55 V was observed. Interestingly, a new redox peak was generated at an apparent standard potential, E° (Epc + EPJ2), of (-550±5) mV (Al/Cl). The peak grows continuously during successive voltammetric cycles. After the 40th cycle, the redox peak is saturated. After the experiment, the working electrode was washed with copious amounts of water and twenty voltammetric cycles were performed on a blank pH 7 PBS. Surprisingly, the Al/Cl redox was retained without any leaching on the working electrode, as shown in Figure 1 Ba (Scheme 1, step 4).

It is an adsorption-controlled electron transfer in behavior (plot of /Pa versus scan rate (v) is linear (data not enclosed), pH dependent Nernstian characteristic (data not enclosed, dEpa/dpH = -60 mV vs Ag/AgCl, Epa = anodic peak potential) [37] and peak-to-peak separation, AEP (Epa-Epc) = 55±2 mV. Calculated surface excess value (/) for the Al process is 23.48 nmol.cm"2 and relative standard deviation (RSD) for the twenty cycles is 2.13 %. These observations indicate facile electrochemical oxidation of surface-bound AN on the MWCNT-modified electrode in contrast to the conventional electrodes like Au and GCE, which provided no response in this work. To obtain precise information about the AN oxidation potential on an Au/MWCNT surface, a discreet potentiostatic polarization experiment was also carried out at different applied potentials CEapp), from 0 to 1.2 V versus Ag/AgCl with a holding time of 120 s. After the electrochemical experiment, the electrode was washed with distilled water and subjected to voltammetric cycles from +0.5 to -1.0 V versus Ag/AgCl, in which the Al/Cl redox peak exists. Figure 1 Ca is a typical plot of/pa (Al peak) against 2sapp. At Em =1.0 V versus Ag/AgCl (i.e., 1.42 V vs. reversible hydrogen electrode), the maximum Al/Cl redox peak response is observed and beyond that potential a decrease in the redox response is observed.

Electrochemical oxidation by both the forty continuous cycles (Figure 1 A) and £aPP (-T=22.1 nmol.cm"2) methods result to qualitatively similar electrochemical Al/Cl responses (±1.61%). Meanwhile, as a control experiment, bare Au/MWCNTs was subjected to £app = 1 V oxidation and the obtained CV graph is given in Figure l(C)b. Fortunately, the bare Au/MWCNT electrode failed to show any redox peak at -0.6 V; however a weak CV peak centered at -0.2 V (quasi-reversible; r= 0.6 ±0.05 nmol.cm"2) that is about 60 times weaker than the peak observed in the Figure IB (b&c), was observed. This minor observation highlights oxidation of the MWCNT; presumably the graphitic edges, to oxygenated species such as >C=0, -OH, >C-0- and >C-OH. Amongst these species, >C=0/>C-OH may be responsible for the weak redox peak observed at -0.2 V [38]. Overall, it can be concluded that the MWCNT surface-bound AN is oxidized markedly at 1 V versus Ag/AgCl under optimal conditions. The electrochemically oxidized Au/MWCNT@ANadS (tentatively designated as Au/MWCNT@ANO, ANO =oxidized form of AN) was characterized (along with control samples) by surface-enhanced Raman, FTIR, and X-ray photoelectron spectroscopy.

Surface-enhanced Raman spectroscopy of MWCNT@ANads and MWCNT@ANO showed marked enhancement in the ratio of the intensities (I) of the D and G bands (7D/IG, in which the G band is an effect of the pure graphite structure and the D band is based on the disorder-induced mode of oxygen) from 1.54 to 2.09 due to the surface-bound oxygenation of AN to quinone (ANO)-like species (Figure 2A). The control experiment with 1 V electro-oxidized Au/MWCNT (i.e, Au/MWCNT*, * = conditioned at 1 V) showed a lower ID/la value (1.73), which might be an effect of the lower degree of oxygenation at the MWCNT surface. FTIR spectra of MWCNT@ANO (Figure Sic in the Supporting Information) showed appreciable signals at 1617 and 1638 cm"1. Assuming an anthraquinone (AQ)-like product is formed on the surface, a standard AQ powder sample, as a control, was also tested. It shows an FTIR response at 1615 and 1642 cm"1, corresponding to v(-C=C) and v(-C=0), respectively (Figure Sle). This is closer to the FTIR response of the MWCNT@ANO. This observation suggests the formation of AQ, as one of the main products, on the surface. X-ray photoelectron spectroscopy (XPS) characterization of MWCNT@ANadS and MWCNT@ANO showed 7.2 fold enhancement in the response of one of the 01 peak binding energies (530.9 eV), which corresponds to the > C=0 system (Figure 2B).

This observation is in agreement with the FTIR results and suggests AQ formation. To authentically confirm the kind of oxygenated species formed at the interface, we have subjected the ethanolic extract of MWCNT@ANO to GC-MS. Several fractions with molecular weights (Mw) of 178, 206, and 208, which correspond to AN, {AQ-2H+}, and AQ species, respectively, were identified (Figure 3). GC-MS retention time values for AN and AQ standards also match the experimental samples. Overall, the characterization presented confirms the presence of AQ as a major species with the MWCNT@ANO, along with unreacted AN, on the surface. It is expected that a fraction of AN species, which are not in contact with MWCNT, might not be involved in the electrochemical oxidation (Figures S2 and S3 in the Supporting Information). Hence, this is the source of the unreacted AN in the GC-MS. Hereafter, MWCNT@ANO will be represented as MWCNT@AQ. Finally, in support of this conclusion, we also studied the electrochemical features of a 5 uL sample of a standard AQ solution in ethanol (2 mg/500 uL) drop-cast on the Au/MWCNT system (Scheme 1, step 5).

Interestingly, the electrode showed a stable, well-defined redox peak (£° = (-0.560 ±5) mV vs. Ag/AgCl) very close to the £° value observed for Au/MWCNT@AQ (£°' = -550 mV vs. Ag/AgCl) (data not shown) [37]. This observation undoubtedly confirms the facile formation of AQ upon surface-bound oxidation of AN at 1 V versus Ag/AgCl in pH 7 PBS (Scheme 1, step 6). Concerning the mechanism for the AN->AQ conversion, electrochemical oxidation of anthracene to a stable divalent anthracene cation at approximately 1 V versus Ag/Ag+ in non-aqueous solution is well-known [39-46]. Depending on the reaction conditions, the anthracene divalent cation is subsequently involved in an addition reaction to either a second anthracene or another species, like pyridine [39], ethanol [40], or acetonitrile[46]. Note that the formation of the solution-phase divalent anthracene cation was previously confirmed by ESR techniques [42, 46]. In this regard, divalent anthracene cation derived compounds like 9,10-dihydroanthranyldipyridinium diperchlorate [39], the 9,10-diethyl ether derivative of anthracene [40], 9-nitro-lO-trifluoroacetoxyanthracene [45], and bianthracene [45] (dimerized anthracene) were electro-synthesized.

On the other hand, in presence of H2O, products like AQ [45, 46] and bianthrone [40. 44] (electro-inactive species) were reported as forming via the divalent anthracene cation by hydroxylation and dimerization reactions, respectively. Paddan et al. reported AQ (5-14 %) and bianthrone (4-6 %) as products upon sono-electrochemical oxidation of AN on a Pt ultramicroelectrode in the presence of 1.4 mM H20 and 02-free non-aqueous media [15]. In fact, our preliminary in situ CV and electrochemical quartz crystal microbalance (EQCM) studies with EQCM-Au/MWCNT@AN showed involvement of the following molecular weights: 17, 19, 32, 36, and 50 g mol"1, which correspond to OH (i.e., H202), H30+, 2H20, 02, and {02+H20} species in the region -1.0 to +1.0 V versus Ag/AgCl (Figure 2C and D). At approximately 1 V the participation of H20 was identified in the CV-EQCM study. However at higher oxidation potentials there is a good chance water is oxidized to H202 and 02 [47,48], and these species may in turn further oxidize the AN to AQ.

To pinpoint the species involved at the optimal oxidation potential in the present work, various control experiments relating to the oxidation of AN under four different sets of conditions were carried out (see Scheme SI in the Supporting Information). Au/MWCNT@AN was tested with and without added H202 in pH 7 PBS after purging with N2 under sub-optimal (0.7 V vs. Ag/AgCl; data not shown), and optimal conditions (IV versus Ag/AgCl; data not shown), the experiments were attempted with Au/MWCNT@AN and oxygen-or N2-saturated pH 7 PBS at the optimal conditions (data not shown), and Au/MWCNT was treated with solution-phase AN in the absence of water in N2-saturated acetonitrile (Figure S4 and Table SI in the Supporting Information). In the first three cases, similar AQ-formation results were observed (in terms of AQ peak currents) with and without the presence of the analyte. In the last case, there was no formation of AQ either in the solution phase or on the CME surface (Figure S4 in the Supporting Information). These control experiments ruled out the involvement of H2O2 and O2 in the conversion of AN to AQ. On the other hand, formation of the divalent anthracene cation and its subsequent hydroxylation seem to be a key step in the overall electrochemical oxidation of Au/MWCNT@AN to Au/MWCNT@AQ at 1 V versus Ag/AgCl in pH 7 PBS, as observed in this work (Scheme SI, step 5 in the Supporting Information).

Purified-MWCNT (p-MWCNT, p=purified), functionalized-MWCNT (f-MWCNT, f=functionalized), and single walled carbon nanotube (SWCNT), along with graphite nano powder (GNP) and activated charcoal were also examined in the surface-bound electrochemical oxidation of AN as studied above. Except for the activated charcoal, the other systems showed qualitatively similar Al/Cl redox peak features to different extents (Figure S5 in the Supporting Information). With respect to the rAQ value of the Al peak, pristine-MWCNT@AQ showed the highest /AQ value (22.1 nmolcm"2), followed by f-MWCNT@AQ (17.04 nmol cm"2), p-MWCNT@AQ (14.48 nmol cm"2), SWCNT@AQ (2.94 nmol cm"2), and GNP@AQ (2.25 nmol cm"2; Figure 4A). The following conclusions can be drawn from the above observation: 1) hexagonal carbon with a graphite-like structure is necessary for the AN electro-oxidation, 2) dense multiwalled CNTs are responsible for the higher electrochemical conversion, 3) the hydrophilic nature of the f-MWCNTs results in a repulsive interaction with hydrophobic PAH and in turn reduction by the electrochemical oxidation, and 4) metal impurities (oxides of iron, cobalt, and nickel)[49-53] might have a specific role in the enhanced surface-bound oxidation of AN. There may plausibly be a Fenton-like reaction (Fe11 with H2O2) [16-19], with an electrochemically formed hydrogen peroxide intermediate and nanometal impurities, which results in significant oxidation of AN to AQ.

Finally, some of the other linear PAHs, including naphthalene and tetracene, were also subjected to the surface-bound 1 V electro-oxidation conditions on Au/MWCNT. The result showed several redox peaks at E°'= -0.1 and -0.2 V versus Ag/AgCl caused by hydroxylation (1,2- and 1,4-dihydroxynaphthalene were formed) on the MWCNT-surface-bound naphthalene (Figure 4B). Similar treatment of tetracene compounds resulted in a stable redox peak at -0.7 V and minor peaks at -0.2 and -0.4 V versus Ag/AgCl, which may be an AQ-like redox couple in addition to dihydroxy derivatives formed on the MWCNT surface (Figure 4C). Finally, continuous CVs of the modified electrodes show unaltered peak-current responses with relative standard deviation (RSD) values for 20 successive cycles of 2.2 and 3.1%, respectively. These observations indicate the stable surface confined redox activity of PAH on chemically modified electrodes at physiological pH in this work. In conclusion, carbon nanotube modified electrodes showed facile surface-bound electrochemical oxidation of linear PAHs to surface-confined redox active species at 1 V versus Ag/AgCl in a neutral buffer solution.

Collective physicochemical characterization of the electro-oxidized AN on MWCNT by in situ EQCM, surface-enhanced Raman spectroscopy, XPS, FTIR spectroscopy, and GC-MS confirmed the formation of AQ, along with some unreacted AN species on the interface. MWCNT-surface-bound electrochemical oxidation of naphthalene and tetracene were also successfully demonstrated. Overall, the electro-oxidation methodology introduced in this work has four important features: 1) it is a simple and easy way to degrade PAHs, 2) PAHs may be recycled as redox active polyaromatic hydroquinone@MWCNT electrodes, 3) it assists understanding of the surface oxidation features of the graphene and CNT subunits, and 5) it would allow the environmental monitoring of PAHs, including AN pollution, through electrochemical detection. Such detection might make use of low-cost disposable screen-printed carbon electrodes modified with a compact MWCNT layer as a detector. Lastly, the system developed may be used for environmental cleanup by converting carcinogenic PAH pollution into non-hazardous quinones with a MWCNT-packed column coupled with an electrochemical system (i.e., under electrified condition).

Acknowledgements Authors gratefully acknowledge the Department of Science and Technology (DST), India, under the Science and Engineering Research Council Scheme for financial support.
References
[1] K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva, A. A. Firsov, Science 306 (2004) 666. [2] A. K. Geim, K. S. Novoselov, Nat. Mater. 6 (2007) 183. [3] Y. B. Zhang, Y. W. Tan, H. L. Stormer, P. Kim, Nature 438 (2005) 201. [4] G. P. Pfiefer, M. F. Denissenko, M. Olivier, N. Tretyakova, S. Hecht, P. Hainaut, Oncogene 21 (2002) 7435. [5] P. Boffetta, N. Jourenkova, P. Gustavsson, Cancer, Causes Control 8 (1997) 444. [6] B. Armstrong, E. Hutchinson, J. Unwin, T. Fletcher, Environ. Health Perspect. 112 (2004) 970. [7] H. Rubin, Carcinogenesis 22 (2001) 1903. [8] A. J. Fatiadi, £/W>WJ. Sci. Technol. 1 (1967)570. [9] Y.Mao, J. K.Thomas,Zawgmwr 8,(1992),2501. [10] C. A. Reyes, M. Medina, C. C. Hernandez, M. Z. Cedeno, R. Arce, O. Rosario, D. M. Steffenson, I. N. Ivanov, M. E. Sigman, R. Dabestani, Environ. Sci. Technol. 34 (2000)415.

[11] M. E. Sigman, P. F. Schuler, M. M. Ghosh, R. T. Dabestani, Environ. Sci. Technol. 32 (1998)3980. [12] R. Dabestani, K. J. Ellis, M. E. Sigman, J. Photochem. Photobiol A 86 (1995) 231. [13] J. Kou, Z. Li, Y. Yuan, H. Zhang, Y. Wang, Z. Zou, Environ. Sci. Technol. 43 (2009) 2919. [14] M. Mazur, G. J. Blanchard, J. Phys. Chem. B 108 (2004) 1038, [15] C. A. Paddon, C. E. Banks, I. G. Davies, R. G. Compton, Ultrason. Sonochem 13 (2006) 126. [16] Z. Qiang, J. H. Chang, C. P. Huang, D. Cha, ACS Symp. Ser. 778 (2000) 187. [17] K. Nam, W. Rodriguez, J. J. Kukor, Chemosphere 45 (2001) 11. [18] S. Jonsson, Y. Persson, S. Frankki, B. Bavel, S. Lundstedt, P. Haglund, M. Tysklind, J. Hazard. Mater. 149 (2007) 86. [19] B. W. Bogan, V. Trbovic, J. Hazard. Mater. 100 (2003) 285. [20] C. E. La Rotta Hernandez, D. S. Werberich, E. D'Elia, Electrochem. com 10 (2008) 108.

[21] K. D. McCoull, D. Rindgen, I. A. Blair, T. M. Penning, Chem. Res. Toxicol. 12 (1999) 237. [22] D. E. Dodor, H-M. Hwang, S. I. N. Ekunwe, Enzyme Microb. Technol. 35 (2004) 210. [23] X. Hu, P. Wang, H-M. Hwang, Bioresour. Technol. 100 (2009) 4963. [24] R. L. Clough, J. Am. Chem. soc. 102 (1980) 5242. [25] S. N. Chu, S. Sands, M. R. Tomasik, P. S. Lee, V. F. McNeill, J. Am. Chem. Soc. 132 (2010) 15968. [26] S. P. Bailey, P. Kolsaker, B. Sihna, J. B. Ashton, F. Dobinson, J. E. Batterbe, J. Org. Chem. 29(1964)1400. [27] P. Subramanian, M. S. Murthy, Ind. Eng. Chem., Process, Des. Develop. 13 (1974) 112. [28] E. A. Cepeda, Ind. Eng. Chem. Res. 26 (1987) 2401. [29] J. Koshitani, T. Kado, Y. Ueno, T. Yoshida, J. Org. Chem. Al (1982) 2879. [30] A. Sorokin, B. Meunier, Eur. J. Inorg. Chem. 1998 (1998) 1269.

[31] P. Bouvrette, S. Hrapovic, K. B. Male, J. H. T. Luong, J. Chromatogr., ^4 1103 (2006) 248. [32] K. Yang, L. Zhu, B. Xing, Environ. Sci. TechnolAQ(2006) 855. [33] J. Ma, R. Xiao, J. Li, J. Yu, Y. Zhang, L. Chen, J. Chromatogr., A. 1217 (2010) 5462. [34] K. Yang, B. Xing, Chem. Review. 110 (2010) 5989. [35] H. Bagheri, Z. Ayazi, A. Aghakhani, Anal. Chim. Acta. 683 (2011) 212. [36] L. Guo, H. K. Lee, J. Chromatogr., ^. 1218(2011)9321. [37] A. S. Kumar, P. Swetha, Colloids Surf., A 384 (2011) 597. [38] R. L. McCreery, Chem. Rev. 108 (2008) 2646. [39] H. Lund, Acta chem. Scand. B., (1957) 1323. [40] K. E. Friend, W. E. Ohnesorge, J. Org. Chem., 28 (1963) 2435.

[41] P. A. Malachesky, L. S. Marcoux, R. N. Adams, J. Phys. Chem., 70 (1966) 2064. [42] J. Phelps, K.S.V. Santhanam and A.J. Bard, J. Am. Chem. Soc, 89 (1967) 1752. [43] M. E. Peover, B. S. White, J. Electroanal Chem. Interfacial Electrochem. 13 (1967) 93. [44] Edward J. Majeski, James D. Stuart, William E. Ohnesorge, J. Am. Chem. Soc, 90 (1968)633. [45] 0. Hammerich, V. D. Parker, Acta chem. scandB.,36 (1982) 519. [46] S. Fukuzumi, I. Nakanishi, K. Tanaka, J. Phys. Chem. A 103 (1999) 11212. [47] Z. D. Wei, M. B. Ji, S. G. Chen, Y. Liu, C. X. Sun, G. Z. Yin, P. K. Shen, S. H. Chan, Electrochim. Acta 52 (2007) 3323. [48] C. G. Hu, W. L. Wang, K. J. Liao, G. B. Liu, Y. T. Wang, J. Phys. Chem. Solids 65 (2004) 1731. [49] C. E. Banks, A. Crossley, C. Salter, S. J. Wilkins, R. G. Compton, Angew. Chem. Int. Ed. 45 (2006) 2533. [50] T. Kolodiazhnyi, M. Pumera, Small 4 (2008) 1476.

[51] A. Ambrosi, M. Pumera, Chem. Eur. J. 16 (2010) 1786. [52] A. Ambrosi, M. Pumera, Chem. Eur. J. 18 (2012) 3338. [53] A. Senthil Kumar, P. Gayathri, P. Barathi, R. Vijayaraghavan, J. Phys. Chem. C, 116 (2012) 23692. Scheme 1. General illustration of the experiments used to establish the Au/MWCNT surface-bound electro chemical oxidation of a PAH. Anthracene (AN) is physically adsorbed onto A) an Au/MWCNT electrode from a solution in ethanol, forming B) Au/MWCNT@ANa(is (as identified by Raman, FTIR, and X-ray photo electron spectroscopy and GC-MS) and is then electrochemically oxidized by 20 CV cycles from +1 to 1 V or at an applied potential (Em) of 1 V versus Ag/AgCl for 120 s in pH 7 phosphate buffer solution to form C) surface-confined anthraquinone (Au/MWCNT@ANO or, as identified for AN, Au/MWCNT@AQ). Attempts to electrochemically oxidize D) anthracene adsorbed directly onto a gold electrode (Au/ANads) from a solution in ethanol were ultimately unsuccessful, as were attempts to F) immobilize AQ directly onto a gold electrode. E) However, AQ could be directly adsorbed onto an Au/MWCNT electrode.

Figure 1. (A) Forty continuous CV responses of AN modified (physically adsorbed) on Au (a), GCE (b) and Au/MWCNT (c) in pH 7 phosphate buffer solution at a scan rate of 50 mV.s"1. (B) Twenty continuous CV response of electrochemically oxidized AN (ANO) (Fig.l(A)c) modified electrode, i.e., Au/MWCNT@ANO (a). Control CV experiments relating to the Au/MWCNT (without any PAH), priorly, conditioned at an applied potential (Em) 1 V vs Ag/AgCl for 120 s (b) or CV cycled (n=20) at -1 to +1 V vs Ag/AgCl (c), in pH 7 PBS. (C) Plot of Al peak current vs Eapp data for the Au/MWCNT@ANO (a) and unmodified Au/MWCNT* (no Al peak, * = Conditioned electrode at different applied potentials) (b).

Figure 2. (A) Comparitive surface enhanced Raman Spectroscopy response of SPE-Au/MWCNT@ANO (a), SPE-Au/MWCNT@ANads (b) and SPE-Au/MWCNT* (c) (* = Conditioned at 1 V vs Ag/AgCl for 120 sec). (B) XPS results of Au/MWCNT@ANO (a) and SPE-Au/MWCNT@ANads(b). (C and D) In-situ EQCM-CV response of EQCM-Au/ MWCNT@ANadS in pH 7 PBS at v = 50 mV.s"1. SPE-Au = Screen-printed gold electrode and EQCM-Au = Gold single crystal-electrochemical quartz crystal microbalance electrode.

Figure 3. GC-Mass responses of organic (ethanol) extract of MWCNT@ANO film. (A) Typical chromatogram and (B and C) are Mass spectrometric responses of the various GC fractions.

Figure 4. (A) Plot of surface excess value of AQ (rAQ) vs various Au/carbon@AQ modified electrodes, CV responses of naphthalene (B) and tetracene (C) adsorbed Au/MWCNT electrodes, which are preconditioned at 1 V vs Ag/AgCl for 120 s (a) in pH 7 PBS, at v = 50 mV.s"1. Control CV data of the PAHs modified on Au and preconditioned at 1 V vs Ag/AgCl for 120 s (b) and MWCNT modified on Au and preconditioned at 1 V vs Ag/AgCl (c). Note ANO = AQ.

Supporting Information

1. Experimental section

1.1 Chemicals and reagents Anthracene (AN), Benz (b) Anthracene (Tetracene), Naphthalene, MWCNT (>90% carbon basis, outer diameter: 10-15 nm; inner diameter: 2-6 run; length 0.1-10 urn) and SWCNT (50-70% carbon basis, outer diameter is 1-1.5 nm) were purchased from Sigma-Aldrich. Other chemicals used were all of ACS-certified reagent grade and used without further purification. Screen-printed gold electrodes (SPE-Au) were purchased from Zensor R&D, Taiwan. Aqueous solutions were prepared using deionized and alkaline potassium permanganate distilled water (designated as DD water). The supporting electrolyte pH 7 phosphate buffer solution (PBS) of ionic strength, /= 0.1 mol L"1 was used throughout this work.

1.2 Instrumentation

Voltammetric measurements were all carried out with CHI Model 660C electrochemical workstation (USA). The three electrode system consists of Au and it's chemically modified electrodes (CMEs) as a working electrode (0.0312 cm2), Ag/AgCl as a reference and platinum wire as the auxiliary electrodes. The surface of the Au electrode was cleaned both mechanically and electrochemically by polishing with 0.5 |im alumina powder, washing with DD water and sonicating for 5 min followed by performing cyclic voltammetry (CV) for 10 cycles in the potential window from -0.2 to 1 V vs. Ag/AgCl at a potential scan rate (v) of 50 mV s"1 in pH 7 PBS. In situ CV-EQCM to determine the amount of AQ species immobilized on the CNT was carried out with gold single crystal electrode (EQCM-Au) of geometric surface area = 0.19 cm2 with a CHI 440 B workstation (USA). FTIR analysis was carried out by JASCO 4100 Spectrophotometer using KBr method. XPS analyses were done using Omicron ESCA spectrometer (Germany) with a monochromatic Al Ka X-ray source. The C Is binding energy (BE) peak at 284.6 eV was taken as an internal reference for correcting the XPS spectrum. XPSPEAK41 Software program was used for the deconvolution of the XPS peaks.

1.3 CNTs modified electrodes preparations

Functionalized MWCNT (f-MWCNT, where f= functionalized) sample was prepared by treating the pristine MWCNT with concentrated 13 N HNO3 as per the literature procedure [SI]. X-ray photoelectron spectroscopy (Ulvac-PHI, PHI500, Versaprobe) of the f-MWCNT powder sample showed nil response at 708.6 eV (Fe 2p3/2, (data not enclosed), indicates absence of any impure nano iron oxide with that sample. At the same time pristine-MWCNT shows marked XPS signal for iron impurity. Thermogravimetric analysis (Perkin-Elmer Pyris 1 TGA, USA) of the sample (in O2 atmosphere at a rate 10°C/min.) showed 7.1% residual impurities at 800°C which may primarily due to presence of Fe2C«3 along with oxides of Ni and Co [S2-S5]. Removal of metal impurities from the MWCNT without activating the carbon nanotube was done similar to the earlier work by Compton group [S2],where the commercial MWCNTs were stirred with 2 M dilute nitric acid for 35 h at room temperature and washed thoroughly with DD water yielding a highly purified MWCNT, designated as p-MWCNT.

Au/CNT was prepared by drop casting of 5 uL of an aliquot from 1 mg CNT dispersed in 500 uL ethanol, 10 min sonicated stock solution on a cleaned Au electrode, and dried the electrode in air for 3±1 min in room temperature followed by drop casting of 5 uL of 2 mg AN dissolved in 500 uL ethanol on the surface of Au/CNT. Then the Au/CNT@ANads electrode was potential cycled in the window from -1.0 V to +1.0 V vs. Ag/AgCl at v = 50 mV.s'1 for twenty continuous cycles (n=20, n= no. of cycles) or Au/MWCNT@ANads was subjected to Em = +1.0 V vs Ag/AgCl anodic oxidation method in pH 7 phosphate buffer solution (PBS) (Scheme 1), followed by washing with copious amount of water and then moved to a fresh blank solution again for the stabilization of electrochemically oxidized AN (represented as ANO) on Au/MWCNT in the potential window from -1.0 V to 0.5 V vs. Ag/AgCl (n = 20). The surface coverage of the ANO was determined from the CV graph by integrating the anodic peak area (qa) of the respective surface-confined redox peak at v = 50 mV.s"1 and calculated using the equation: JAQ = qJnFA, where n = no. of electrons (n =2, AQ/quinone redox reaction and A = geometrical surface area)

1.4 Sample preparations for the physicochemical characterizations

For the FTIR analysis, as prepared SPE-Au/MWCNT@ANO film, dried in a desiccator for overnight, was carefully separated out from the surface using a doctor's needle (0.5 mm * 3.5 cm) and subjected to further examination. XPS and Raman spectra characterizations were carried out using SPE-Au/MWCNT@ANO modified electrode. For the GC-MS analysis, the ethanolic ANO solution was extracted from GCE/MWCNT@ANO as ethanolic suspension, after 5 min sonication of the electrode in ethanol, using syringe filter (Nupore, 0.22uM). In the EQCM experiments, the change in mass at the quartz crystal was calculated from the change in the measurement frequency using the Sauerbrey equation [S6]: Mass change (Am) = (-l/2)/0"2 Af. A. k.pm -(1) Where A is the area of the gold quartz crystal (0.196 cm2), p is the density of the crystal (2.648 g cm"3), k is the shear modulus of the crystal (2.947x10" g cm S*2), A/is the measured frequency change, and f0 is the oscillation frequency of the crystal (8 MHz). A frequency change of 1 Hz is equivalent to a change of 1.34 ng in mass. In addition, the mass change of inserting/ejecting species during electrochemical oxidation of Au/MWCNT@AN can be evaluated using the following formula: Mw = F x Am/Q, where, Q is the charge passed, A/w is the molar mass per electron (g-mol"1 = Molecular weight), and F is Faraday's constant.

Scheme SI. Plausible mechanistic pathway for electrochemical oxidation of Au/MWCNT@AN to Au/MWCNT@AQ. Note ANO = AQ. Table SI: Various control experiments carried out relating to understanding of plausible mechanism for the electrochemical oxidation of Au/MWCNT@AN-»Au/MWCNT@AQ in aqueous solution. Figure S1. FTIR/KBr responses of MWCNT* (* = conditioned at Em - 1 V vs Ag/AgCl for 120 s in pH 7 PBS) (a), MWCNT@ANads (b), MWCNT@ANO (c), standard AN (d) and standard AQ (e) samples. Note ANO = AQ.
Notes:
FT-IR response of MWCNT@ANO selectively shows major signal at 1638.2 cm"1 which is the characteristic response of >C=0 functional group and it is comparable with a standard AQ system. Figure S2. Effect of AN loading: Plot of surface excess (I) of anthraquinone (AQ) obtained from Au/MWCNT@AQ against different ANads concentration physisorbed Au/MWCNT@ANads electrodes. The Au/MWCNT@ANads electrodes were uniformly conditioned at Em = 1 V vs Ag/AgCl for 120 sec in pH 7 PBS. rAQ values were obtained from the Al redox peak of the respective CVs.

Notes:
Upto 2.8 ppm of surface bound AN, ANads (region-1) slow electrochemical oxidations were noticed. But in the region-II ([AN] =2.8-285.7ppm), a steep increase in the electrochemical oxidation and after that concentration ([AN] >285.7 ppm), a saturation in the response, were noticed. Presumably, certain minimum amount of AN concentration can be reached to underlying surface (Au), which are not fully accessible by the electrochemical oxidation procedure and hence slow AN->AQ conversion at low regions. At the optimal region-II, there will be a maximum contact of the ANads on MWCNT for the electrochemical oxidation. Saturation in the oxidation at region III.

Figure S3. Effect of MWCNT loading: (A) Comparative CV responses of increasing loading of MWCNT on Au/MWCNT@AQ films.(B) Plot of surface excess CTAQ) of anthraquinone (AQ) obtained from the CV redox peak against MWCNT loading. The MWCNT loading amount referred interms of different volumes (2-40 uL) of 2000 ppm MWCNT of stock transferred to Au followed by drop-casting of fixed amount, 287.5 ppm of AN on the Au/MWCNT. Em = 1 V vs Ag/AgCl (120 sec) based oxidation method was adopted for the Au/MWCNT@AN-» Au/MWCNT@AQ conversion. rAQ values were obtained from the Al redox peak of the respective CVs.

Notes: If 2D surface interaction between Au/MWCNT and AN might be involved then there wouldn't be any substantial increase in the AN oxidation to AQ upon the MWCNT loading effect. On the other hand, if 3D surface interaction between Au/MWCNT and AN layers will be happened, where the thickness effect of the Au/MWCNT layer and diffusion of the AN to inside the film and inner walls of the MWCNT might be considered, then there will be considerable increment in the AN oxidation current and -TAQ values. Interestingly, a steep increase in electrochemical oxidation current and TAQ values were observed upon varying the loading of MWCNT on the chemically modified electrode, i.e., Au/MWCNT@AQ. This observation suggests 3D interaction between MWCNT and AN and in turn to enhanced AQ formation (within Au/MWCNT@AQ) in this work.

E/VvsAg/AgCI

Figure S4. Effect of water on the AN-»AQ formation at the optimal oxidation potential (Eapp = 1.0 V): Comparative CV responses of GCE (A) and GCE/MWCNT (B) without (a) and with (b) 2 mM AN dissolved 0.05 M TBAPFe/acetonitrile (N2 purged) electrolyte solution. (C) Control CV response of GCE and GCE/MWCNT with 2 mM standard AQ dissolved 0.05 M TBAPFe/acetonitrile (N2 purged) electrolyte solution, v = lOOmV/s.

Notes:
• To check the water (H20) contribution, the Au/MWCNT@AN modified electrode was electrochemically oxidized in non-aqueous medium (absence of water).
• If considerable AN->AQ formation will occur with significant current signals and TAQ values in absence of water, i.e., non-aqueous medium, then it can be considered as in indication for absence of water contribution. On the other hand, if no AN-»AQ formation will be occurred, where there will be absence or poor AQ current signal and TAQ values, which can be taken as a proof for the water contribution.
• Interestingly, as can be seen in Figure S4 B, if the Au/MWCNT electrode was subjected to CV with AN in tetrabutylammonium hexafluorophosphate (TBAPF6)/acetonitrile at -2 V to +2 V vs Ag/AgCl, no AQ formation was noticed. A feeble peak at -1 V vs Ag/AgCl was appeared, which is due to traces of dissolved oxygen. Note that control AQ dissolved in non-aqueous medium showed couple of redox peak due to AQ/AQ2" electron transfer process. No such peak responses were noticed in the present case (Fig. S4 C). On the other hand, this observation authentically confirms influence of water on the electrochemical oxidation of AN to AQ on MWCNT modified electrode.

Figure S5. Effect of various carbons on the electrochemical response: CV responses of Au/Charcoal@AQ (A), Au/Graphite Nanopowder(GNP)@AQ (B), Au/SWCNT@AQ (C), Au/p-MWCNT@AQ (D), Au/f-MWCNT@AQ (E) and Au/MWCNT@AQ (F) modified electrodes in pH 7 PBS at v = 50 mV.s"1. rAQ values were obtained from the Al redox peak of the respective CVs.

References

[SI] A. S. Kumar, P. Barathi, K. C. Pillai, J. Electroanal. Chem. 654 (2011) 85.
[S2] C. E. Banks, A. Crossley, C. Salter, S. J. Wilkins, R. G. Compton, Angew. Chem. Int. Ed. 45 (2006) 2533.
[S3] T. Kolodiazhnyi, M. Pumera, Small 4 (2008) 1476.
[S4] A. Ambrosi, M. Pumera, Chem. Eur. J. 16 (2010) 1786.
[S5] A. Ambrosi, M. Pumera, Chem. Eur. J. 18 (2012) 3338.
[S6] Q. Xie, Z. Li, C. Deng, M. Liu, Y. Zhang, M. Ma, S. Xia, X. Xiao, D. Yin, S. Yao, J. Chem.Educ. 84(2007)681.

Enclosure-II; Claim

We claim the

I) • Electrochemical oxidation of polyaromatic hydrocarbons (PAHs) such as Anthracene, Tetracene and Naphthalene on multiwalled carbon nanotube (MWCNT) modified electrode in pH 7 phosphate buffer solution.

2.) • Oxidation potential, 1 V vs Ag/AgCl at holding time of 120 sec is specific for the conversion of the PAHs to surface confined redox active species

S) • Electrochemical oxidation of anthracene on multiwalled carbon nanotube modified electrode yielded anthraquinone surface confined redox active species.

\jf) • Electrochemical oxidation of tetracene and naphthalene on MWCNT modified electrode yielded redox active quinone and di-hydroxy species respectively.