Carbon Nanotubes with Tailored Density of ... - ACS Publications

Research Article www.acsami.org

Carbon Nanotubes with Tailored Density of Electronic States for Electrochemical Applications Yingpan Song, Huifang Hu, Miao Feng, and Hongbing Zhan* College of Materials Science and Engineering, Fuzhou University, 2 Xueyuan Road, Fuzhou 350116, People’s Republic of China S Supporting Information *

ABSTRACT: The density of electronic states (DOS) is an intrinsic electronic property that works conclusively in the electrochemistry of carbon materials. However, seldom has it been reported how the DOS at the Fermi level influences the electrochemical activity. In this work, we synthesized partially and fully unzipped carbon nanotubes by longitudinally unzipping pristine carbon nanotubes (CNTs). We then studied the electrochemical activity and biosensitivity of carbon materials by means of the CNTs and their derivatives to elucidate the effect of the DOS on their electrochemical performances. Tailoring of the DOS for the CNT derivatives could be conveniently realized by varying the sp2/sp3 ratio (i.e., graphite concentration) through manipulating the oxidative unzipping degree. Despite the diverse electron transfer mechanisms and influence factors of the four investigated redox probes (IrCl62−, [Fe(CN)6]3−, Fe3+, and ascorbic acid), the CNT derivatives exhibited consistent kinetic behaviors, wherein CNTs with a high DOS showed superior electrochemical response compared with partially and fully unzipped carbon nanotubes. For biological detection, the CNTs could simultaneously distinguish ascorbic acid, dopamine, and uric acid, while the three CNT derivatives could all differentiate phenethylamine and epinephrine existed in the newborn calf serum. Moreover, the three CNT derivatives all presented wide linear detection ranges with high sensitivities for dopamine, phenethylamine, and epinephrine. KEYWORDS: carbon nanotube derivatives, density of electronic states, Fermi level, graphite concentration, redox probes, biomolecules, newborn calf serum

■

INTRODUCTION Carbon materials are of significant interest for electrochemical applications because of their structural polymorphism, rich surface chemistry, chemical stability, and the strength of the internal C−C bonds present.1−9 Tailoring of heterogeneous electron transfer (HET) kinetics for carbon materials via rational design and operational manipulation is important in the study of fundamental electrochemistry and in various electrochemical applications, such as biosensors.3,6 Concerning an appropriate carbon electrode material, the key point is its electronic property, that is, the density of electronic states (DOS), which varies considerably for different types of carbon. The DOS of the electrode material dominates the HET kinetics, because of its conclusiveness to the probability that an electron with appropriate energy is feasible to transfer to the redox probe.3,8 HET rate increases while there are sufficient electronic states with energies in the electrode near the formal potential level of the redox probe involved.3,7,8 It is known that, for metals, the DOS is usually very high and depends weakly on the energy, while the DOS for carbon materials is dependent on the carbon microstructure, as well as surface states, such as defects and functional groups. First, the DOS for carbon materials changes remarkably with their © 2015 American Chemical Society

structural polymorphism, which spans a range from evenly distributed DOS for disordered graphitic materials to distinctively structural DOS for carbon nanotubes; from large energy-gapped undoped diamond with no DOS around the Fermi level to highly oriented pyrolytic graphite with low DOS.3 Second, the DOS for carbon materials can be affected by the presence of disorder, such as defects and oxygen-containing functional groups, so that redox probes show varying HET kinetics. For example, with regard to outer-sphere redox probes, disorder facilitates the HET process by disruption of the carbon electronic structure, whereas, for inner-sphere probes, surface chemistry also contributes.8 Because the DOS is an intrinsic electronic property of carbon materials, it is important for the electrochemical activities of these materials.9−13 However, although different theoretical models have been proposed to elucidate HET at carbon-based electrodes,14−22 how the DOS at the Fermi level influences HET rates remains unclear. Hatton et al.9 reported that modifying surface chemistry, varying graphene orientation, and Received: August 19, 2015 Accepted: November 5, 2015 Published: November 5, 2015 25793

DOI: 10.1021/acsami.5b07700 ACS Appl. Mater. Interfaces 2015, 7, 25793−25803

Research Article

ACS Applied Materials & Interfaces

Figure 1. Typical and high-resolution TEM images of the (a, d) CNTs, (b, e) PUCNTs, and (c, f) FUCNTs. with PBS depending on the needs of the measurement. The NCS was diluted 10 times with 0.1 M PBS before electrochemical measurements. PUCNTs and FUCNTs were prepared according to the method proposed by Tour et al.,23,24 with a few modifications. Preparation of CNT-, PUCNT-, and FUCNT-Modified Electrodes. Prior to modification, the GCE was polished to mirror state, and dried at room temperature. CNTs were dispersed into water with ultrasonication to give a 1 mg/mL suspension solution. Eight microliters of CNTs suspension solution was coated onto a GCE surface, and was dried in air, thus to fabricate a CNT/GCE. For comparison, PUCNT/GCE and FUCNT/GCE were prepared using the similar procedure. Apparatus. Field emission transmission electron microscopy (TEM) images were carried out with a Tecnai G2F20 S-TWIN high-resolution transmission electron microscope operating at an accelerating voltage of 200 kV. X-ray photoelectron spectroscopy (XPS, ESCALAB 250, Thermo Scientific, USA) and Raman spectroscopy (inVia+Reflex, Renishaw, UK) were used to analyze the composition of the CNT materials. He(I) ultraviolet photoemission spectroscopy (UPS, ESCALAB 250, Thermo Scientific, USA) was used to detect the DOS for the CNT derivatives. Ultraviolet− visible (UV−vis) adsorption spectra were obtained with a Shimadzu UV-2450 spectrophotometer. Electrochemical measurements were carried out with a CHI 660E electrochemical workstation (Shanghai Chenhua Instrument Co., Ltd., China) and a three-electrode system comprising of a reference electrode (saturated calomel electrode, SCE), a counter electrode (Pt wire), and a working electrode (modified GCE).

manipulating surface roughness offers little control over the DOS near the Fermi level, and that only changes in the intrinsic structure can alter the DOS for the carbon materials. It is this factor that gives rise to varied electrochemical activities and HET rates. Here, we determine the effect of DOS on the electrochemistry of carbon-based electrode materials by means of three different structures of carbon nanotube (CNT) derivatives. These are pristine CNTs, partially unzipped carbon nanotubes (PUCNTs), and fully unzipped carbon nanotubes (FUCNTs). PUCNTs and FUCNTs were prepared from longitudinally cutting CNTs by controlling the oxidative opening degree.23,24 Compared with CNTs they possess different structures, defects, and sp2/sp3 ratios, all of which can strongly alter the DOS for the CNT derivatives and further influence the electrochemical responses of the modified electrodes. Therefore, we investigated the electrochemistry of CNT-, PUCNT-, and FUCNT-modified glassy carbon electrodes (GCEs) to determine the underlying effect of DOS using four redox probes, five biomolecules and Newborn Calf Serum. Our work provides further insight into the effect of DOS on the electrochemistry of carbon electrode materials.

■

EXPERIMENTAL SECTION

Sample Preparation. Multiwalled carbon nanotubes (MWCNTs, purity >97%, ∼20 nm in diameter and 5−15 μm in length) were obtained from Shenzhen Nanotech Port Co., Ltd. (Shenzhen, China). HCl, HNO3, H2SO4, H3PO4, KMnO4, H2O2 (30 wt % in H2O), K3Fe(CN)6, FeCl3, KCl, K2HPO4·3H2O, and KH2PO4 were of analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd., China. K2IrCl6, dopamine (DA), ascorbic acid (AA), and uric acid (UA) were purchased from Alfa Aesar. Phenethylamine (PEA) was purchased from Sigma-Aldrich. Epinephrine (EP) was purchased from Aladdin. Newborn calf serum (NCS) was purchased from Gibco. All solutions were prepared with ultrapure water of resistivity not less than 18.25 MΩ/cm. K2IrCl6, K3Fe(CN)6, and FeCl3 were utilized at a concentration of 5 mM in 0.1 M KCl supporting electrolyte. PBS (phosphate buffer saline) solution (0.1 M, pH = 7.0) were prepared with KCl, K2HPO4·3H2O, and KH2PO4. Different concentrations of AA, DA, UA, PEA, and EP solutions were prepared

■

RESULTS AND DISCUSSION Physicochemical Characterization. The CNT, PUCNT, and FUCNT samples were first examined using TEM. As shown in Figure 1a and d, the pristine CNTs present an average diameter of ∼20 nm, with the core of the nanotubes being visibly hollow. Partial cutting of the CNTs resulted in PUCNTs consisting of a few layers with widths of 30−50 nm and lengths of a few micrometers; the edges are partly destroyed and rough but the inner walls retain the layered graphitic structure of the pristine CNTs (Figure 1b and e). Finally, for complete opening, the CNTs were sliced either longitudinally or in a spiral manner, affording FUCNTs with an average width of 50 25794


Research Article


Figure 2. (a) Deconvoluted and (b) full C 1s XPS spectra of the CNTs, PUCNTs, and FUCNTs. (1, sp2; 2, C−OH/CO; 3, COOH).

nm (Figure 1c and f). The ribbon-like structures are stacked and entangled with each other because of the strong molecular interactions present. The FUCNT edges are badly destroyed and are not as straight as those of pristine CNTs, indicating introduction of abundant defects during the unzipping process. XPS (Figure 2) were used to quantify the sp2/sp3 ratio in the CNT derivatives and estimate the graphite concentration (i.e., sp2 carbon). Figure 2a shows that, for CNTs, the carbon contents comprise 92.72 atomic% sp2 carbon at 284.5 eV, which is characteristic of graphitic groups; for PUCNTs, the carbon contents comprise 38.46 atomic% sp2 carbon at 284.6 eV, and 34.45 atomic% sp3 carbon (24.2 atomic% at 286.3 eV corresponding to C−OH/CO; 10.25 atomic% at 288.6 eV corresponding to COOH); for FUCNTs, the carbon contents comprise 35.36 atomic% sp2 carbon at 284.7 eV and 32.64 atomic% sp3 carbon (26.05 atomic% at 286.6 eV corresponding to C−OH/CO; 6.59 atomic% at 288.7 eV corresponding to COOH). Figure 2b shows that in going from CNTs to PUCNTs to FUCNTs, the maximum in the C 1s spectrum shifts to higher binding energy, which indicates the decrease of graphite concentration with the oxidative unzipping degree, due to the sp3 bond possessing a higher binding energy than sp2 bond. The sp2/sp3 ratio is very high for CNTs. After oxidative opening, the sp2/sp3 ratio decreases to 1.12 and 1.08 for PUCNTs and FUCNTs respectively, confirming the graphite concentration decreases with oxidative unzipping degree (Table 1). Raman spectroscopy was used to detect the structural characteristics of the samples (Figure 3), providing useful information on the defects (D-band, ∼1350 cm−1), in-plane vibrations of sp2 carbon atoms (G-band, ∼1580 cm−1), and the

Figure 3. Raman spectra of the CNTs, PUCNTs, and FUCNTs.

stacking order (2D-band, ∼2720 cm−1).25,26 From CNTs to PUCNTs and FUCNTs, the ratio of the D and G band intensities (ID/IG) increases, reflecting an increase in the defect density and a decrease in the sp2 content for the CNT derivatives. Table 2 also shows that PUCNTs have the highest Table 2. Raman, UV Data, and Band Gap for the CNTs, PUCNTs, and FUCNTs

Table 1. XPS Data for the CNTs, PUCNTs, and FUCNTs

[sp3] (%) sample

[sp ] (%)

CNTs

92.72

PUCNTs FUCNTs

38.46 35.36

C−OH/CO

24.2 26.05

COOH

10.25 6.59

sp2/sp3 ratio very high 1.12 1.08

O 1s (%)

O/C

7.28

0.08

27.1 32

Raman data ID/IG

UV data λmax (nm)

band gap Eg (eV)

0.81 1.11 0.97

259 253 238

3.29 3.67 3.88

ID/IG of those studied, indicating that the highest defect density unexpectedly lies within PUCNTs. We assume that the mixed structure of CNTs and FUCNTs makes PUCNTs highly heterogeneous in nature, giving rise to the largest numbers of defects. Also, going from CNTs to PUCNTs and FUCNTs, the 2D band becomes broader and weaker, reflecting the gradual destruction of the multilayer graphitic structure within the CNTs. UPS is used to detect the DOS for CNT derivatives directly because the normalized intensity of the spectra is dependent on

C 1s (%) 2

sample CNTs PUCNTs FUCNTs

0.37 0.47 25795


Research Article


Figure 4. (a) Full and (b) detailed UPS spectra showing the DOS near the Fermi level of the CNTs, PUCNTs, and FUCNTs.

Figure 5. (a) UV−vis spectra and (b) band gap calculations of the CNTs, PUCNTs, and FUCNTs.

the DOS of the samples studied.9,10 The normalized intensities of full UPS spectra (Figure 4a) reveal that the CNT derivatives exhibit valence-band structures like graphite. The intensities from 0 to 5 eV, 5 to 10 eV, and 10 to 18 eV are ascribed to pπ-, pσ-, and s-like σ-bands, respectively.27,28 Figure 4b displays the enlarged image of the normalized UPS spectra near the Fermi level (EF = 0 eV) for the CNT derivatives, indicating that the DOS near EF for CNTs is greatly higher than those for PUCNTs and FUCNTS, while the DOS for PUCNTs is slightly higher than that for FUCNTs. A higher DOS could effectively fill in the valence and conduction bands near EF of the CNT derivatives, and leads to a higher conductivity. This is consistent with the finding that the band gap of CNTs is smaller than that of PUCNTs, which in turn is smaller than that of FUCNTs (Figure 5b). Figure 5a shows the UV−vis absorption spectra of the samples studied. Notably, the CNTs show a broad absorption peak at 259 nm, from typical π → π* transitions of aromatic CC bonds. In comparison, the absorption peak of PUCNTs is blue-shifted slightly to 253 nm, indicating partial destruction of the electronic conjugated network of the CNT carbon

framework. The absorption peak of FUCNTs is significantly blue-shifted to 238 nm, demonstrating complete destruction of the electronic conjugated network of the CNTs and longitudinal opening from a tube-like to ribbon-like structure. The electronic band structures of the CNTs, PUCNTs, and FUCNTs were also studied using UV−vis spectra (Figure 5b). The optical band gap could be determined from the UV−vis spectra via extrapolation of the equation αhν = A(hν − Eg )1/2

(1)

where A is a constant, h is Planck’s constant, ν is the frequency, α is the weak-field absorption coefficient, and Eg is the optical band gap. By plotting (αhν)2 against hν, Eg is obtained via extrapolation of the absorption edge. As shown in Figure 5b, the values of Eg for CNTs, PUCNTs, and FUCNTs were 3.29, 3.67, and 3.88 eV, respectively. For carbon materials, the energy band gap depends on the extent of oxidation and molecular size. Oxidized sites break the continuity of the π network by forming sp3-hybridized domains. A direct band gap can be obtained at these sp3 domains where the electronic transport barriers appear, and can be tuned by varying the oxidation 25796


Research Article


Figure 6. (a) CNT/GCE, (b) PUCNT/GCE, and (c) FUCNT/GCE: (A) EIS for 5.0 mM [Fe(CN)6]3−/4− in 0.1 M KCl and (B) CVs for 0.1 M KCl. Scan rate: 50 mV/s.

Figure 7. (a) CNT/GCE, (b) PUCNT/GCE, and (c) FUCNT/GCE: CVs for (A) 5.0 mM IrCl62−, (B) 5.0 mM [Fe(CN)6]3−, (C) 5.0 mM Fe3+ in 0.1 M KCl, and (D) 5.0 mM AA in 0.1 M PBS. Scan rate: 50 mV/s. Insets: Corresponding peak current dependence on the square-root of scan rate for (A) IrCl62− and (B) [Fe(CN)6]3−.

25797


Research Article

ACS Applied Materials & Interfaces degree.29,30 In our case, the sp2/sp3 ratio decreased with the continuous oxidative unzipping process from CNTs to PUCNTs to FUCNTs, leading to an increase in the band gap value. Fundamental Electrochemical Characterization. Electrochemical impedance spectroscopy (EIS) was used to investigate the electrode’s ability to transfer and exchange charges with the analyte solution. Such an ability, the charge transfer resistance (Rct), is strongly influenced by the structural conformation and surface chemistry of the electrode material.31 For the three curves shown in Figure 6A, the slopes of the linear portion at lower frequencies are all approximately 1, confirming the diffusion process in the electrolyte solution. For CNT/GCE (curve a), no obvious bend is observed, indicating a very small Rct and therefore a favorable charge transfer process between the electrode surface and the electrolyte solution. This is because the relatively complete electronic conjugated network and the highest DOS for CNTs endow them with very high electrical conductivity. This allows CNTs to move the electrons to or from the supporting electrode. For PUCNT/ GCE (curve b), the Rct is approximately 103 Ω, indicating a slower charge transfer compared with CNT/GCE. This results from the CNT−graphene complex structure of the PUCNTs, which makes them tend to form a three-dimensional configuration on the electrode surface that can be easily accessed by analytes. However, the functional groups linked with the PUCNTs exhibit electrostatic repulsion to the analyte, so greater oxide content has a negative effect on the chargetransfer process. These two effects counter each other, leading to the medium-sized Rct of PUCNT/GCE. For FUCNT/GCE (curve c), the Rct is approximately 380 Ω, corresponding to the slowest charge transfer kinetics of these systems. This is attributed to the greatest oxide content and lowest conductivity of FUCNTs after oxidative unzipping of the CNTs. We next investigated the intrinsic electrochemical properties of the three electrode materials using cyclic voltammetry (CV). As shown in Figure 6B, in 0.1 M KCl, the background current follows the order of FUCNT/GCE < PUCNT/GCE < CNT/ GCE, which is attributed to the different textures of the three CNT derivatives constructed on the electrode surface. As we know,3 the background current is resulted from the capacitance and therefore is proportional to the exposed electrode materials to the electrolyte solution. In our case, the rigid tube structure of the pristine CNTs makes them easy to build threedimensional spatial configuration on the electrode surface and form interconnected porous structure among the nanotubes, therefore exposing most area to the solution and providing the largest capacitance and background current; as for PUCNTs, the oxidative unzipping process endows them partially remained rigid structure on the electrode surface, therefore the capacitance and background current decrease with the reduced exposed area to the solution; for FUCNTs, the ribbonlike structure of them is indeed two-dimensional, which further reduce the exposed area, and results in the lowest capacitance and background current. Then, we studied the electrochemical behavior of CNT-, PUCNT-, and FUCNT-modified GCEs. We first explored CV characteristics of the samples using an outer-sphere system IrCl62−, whose kinetics depends only upon the DOS of the electrode and therefore could use to benchmark different electrode materials.3,32 CVs and k0 of our three modified GCEs are summarized in Figure 7A and Table 3. k0 decreases from CNT/GCE (6.29 × 10−2 cm/s) to PUCNT/GCE (1.67 × 10−2

Table 3. EIS and CV Data for the CNT/GCE, PUCNT/ GCE, and FUCNT/GCE CV data EIS data electrode CNT/GCE PUCNT/ GCE FUCNT/ GCE a

Rct (Ω)

[Fe(CN)6]3−

IrCl62− ΔEpa (mV)

k0 (cm/s) −2

ΔEpa (mV)

k0 (cm/s)

103

63 72

6.29 × 10 1.67 × 10−2

73 92

1.32 × 10−2 4.11 × 10−3

380

78

7.95 × 10−3

106

2.49 × 10−3

Scan rate: 50 mV/s (vs SCE).

cm/s) and to FUCNT/GCE (7.95 × 10−3 cm/s); this is attributed to the decreasing DOS of the CNT derivatives and is consistent with the UPS and band gap results. CV characteristics were then investigated using the innersphere system [Fe(CN)6]3−. The electron transfer kinetics of [Fe(CN)6]3− may be influenced by edge-plane site density, that is, the defect density.33 From the Raman data, the PUCNTs have the highest defect density, followed by FUCNTs and CNTs. However, in our case (Figure 7B), for [Fe(CN)6]3−, ΔEp follows the order of CNT/GCE < PUCNT/GCE < FUCNT/GCE, while k0 follows the order of CNT/GCE > PUCNT/GCE > FUCNT/GCE, which are not consistent with the defect densities. This is because the DOS of the electrodes still play an important role in the electron-transfer kinetics. The obtained k0 indicate the competing effects of the DOS and defect density of these systems. Meanwhile, as discussed above for EIS, the surface coverage and negatively charged oxygen functional groups are all responsible for the electron transfer process between the modified GCEs and the [Fe(CN)6]3−. To fully understand the competing effects of surface chemistry and DOS, two more inner-sphere systems (Fe3+ and AA) with varied surface sensitivities were further tested. The electrode kinetics of Fe3+ depends on the presence of surface oxides, while the electrode kinetics of AA is surface sensitive but oxide insensitive.3 For Fe3+ (Figure 7C), the cathodic and anodic peak became more and more unidentifiable from CNT/GCE to PUCNT/GCE and to FUCNT/GCE, and the ΔEp also increased, corresponding to a continuously slowed electron transfer process. For AA (Figure 7D), the CVs showed no cathodic peaks. While, with regard to the anodic peaks, it is obvious that CNT/GCE exhibits the most negative peak potential, indicating the fastest electron transfer, followed by PUCNT/GCE and FUCNT/GCE. The consistent results of our CV studies approve that, despite the different surface sensitivities of the four redox probes, DOS plays a dominant effect upon surface chemistry. We also investigated the influence of scan rate on the CV performances of the three modified GCEs. The redox processes with IrCl62− and [Fe(CN)6]3− gave roughly symmetric anodic and cathodic peaks at relatively slow scan rates. When the scan rate increased, the redox potentials (Epa and Epc) shift slightly. The ΔEp also increases with an increase in scan rate, ranging from 25 to 500 mV/s. It is apparent from Figure 7A−B insets that the cathodic and anodic currents of the samples share a linear relationship with the square-root of the scan rate. This agrees with the Randles−Sevcik equation,34 which describes reversible electrochemical reactions controlled by semi-infinite linear diffusion. The k0 values for the three modified GCE were 25798


Research Article


Figure 8. (a) CNT/GCE, (b) PUCNT/GCE, and (c) FUCNT/GCE: CVs of (A) individual 50 μM DA and (B) mixture of 50 μM DA, 50 μM UA, 300 μM AA in 0.1 M PBS (pH 7.0). Scan rate: 50 mV/s.

Figure 9. (a) CNT/GCE, (b) PUCNT/GCE, and (c) FUCNT/GCE: (A) DPVs of 50 μM DA, 50 μM UA and 300 μM AA in 0.1 M PBS (pH 7.0). DPV, pulse period, 0.2 s; amplitude, 50 mV. (B) Linear relationship between the anodic peak current and the concentration of DA.

estimated by fitting the ΔEp and Nicholson’s kinetic parameter ψ versus the reciprocal of the square root of scan rate (v−1/2),35,36 and are listed in Table 3. Electrochemical Response of Biomolecules. To gain further insight into the physicoelectrochemical characterization of the CNT derivatives, we next explored the electrochemical response of five different biomolecules: AA, DA, UA, PEA, and EP, which are of significant importance in a plethora of areas where electrochemistry is used. Figure 8A presents the CVs of 50 μM DA in 0.1 M PBS (pH 7.0) with three modified GCEs. Two redox peaks were observed for the three electrodes, suggesting that a twoelectron reaction process of DA oxidation to dopaminequinone and dopamine-quinone reduction to DA occurred. For FUCNT/GCE, there was a weak electrochemical response for DA with a ΔEp of 172 mV. In contrast, well-defined and more pronounced redox peaks appeared for PUCNT/GCE and CNT/GCE because PUCNTs and CNTs have more extensive π−π interactions between the phenyl structure of the DA

molecules and the hexagonal carbon structure of the samples. For PUCNT/GCE, Epa and Epc were found to be 177 and 144 mV, respectively, and ΔEp was found to be 33 mV. This is attributed to the combined effect of the greater electrochemically accessible area and higher sp2 content of PUCNTs compared with FUCNTs.37 For CNT/GCE, ΔEp decreased to 14 mV and the anodic peak current (Ipa) was about 2.18 and 4.18 times higher than those of PUCNT/GCE and FUCNT/ GCE, respectively (because of the high conductivity of CNTs available for the electron transfer). Additionally, the relatively rough coverage of CNTs on the electrode increased the possibility of adsorption of DA for effective electron transfer,38 because it provides strong electrostatic interactions with DA.39 Therefore, a synergistic effect between DOS and the adsorption force results in an increasing ΔEp for DA moving from CNT/ GCE to PUCNT/GCE to FUCNT/GCE. Figure 8B displays the CVs of 50 μM DA, 50 μM UA, and 300 μM AA in 0.1 M PBS for three modified GCEs. Oxidation of DA, UA, and AA was observed on both the CNT/GCE and 25799


Research Article


Figure 10. (a) CNT/GCE, (b) PUCNT/GCE, and (c) FUCNT/GCE: Linear relationship between the anodic peak current and the concentration of (A) PEA and (B) EP.

electrons were fixated to the oxygen atoms, obstructing direct electron transfer.40 Therefore, the high DOS and low oxygen content of CNTs in part facilitated the direct electron transfer between the electrode and biomolecules, establishing an abnormal sensitivity. Electrochemical response of other more biomolecules was further investigated. By using DPV, the influence of PEA or EP concentration on the anodic peak current of the three modified GCEs was studied. The anodic peak currents all increased linearly with the increasing PEA or EP concentration for the three electrodes (Figures S4 and S5). For PEA (Figure S4), with the addition of increased concentrations from 1 to 120 μM, the changing trends of the three electrodes are not the same, indicating their different sensitivities to the target biomolecule. As shown in Figure 10A, the linear range was 8−120, 10−120, and 20−120 μM with the sensitivity of 0.086, 0.067, and 0.024 μA/μM for CNT/GCE, PUCNT/GCE, and FUCNT/GCE, respectively. For EP (Figure S5), it underwent a similar detection process, and the linear range was 15−120, 40−120, and 20−120 μM with the sensitivity of 0.076, 0.062, and 0.425 μA/μM for CNT/GCE, PUCNT/GCE, and FUCNT/GCE, respectively (Figure 10B). These results demonstrated that CNTs with high DOS and low oxygen content manifested a novel sensitivity as well. Real Sample Analysis in Newborn Calf Serum. To illustrate the applicability of the modified GCEs for real sample analysis, newborn calf serum (NCS) was selected as real samples for analysis, which is regarded to resemble human serum. Prior to measurement, the NCS was diluted 10 times with 0.1 M PBS solution (pH 7.0). Usually, the serum contains some neurohormones, such as DA, EP, and PEA. In this work, we intend to use our modified GCEs to detect the amount of these molecules. The diluted NCS was used as blank electrolyte solution, and was subjected to continuous addition of undiluted NCS solution during the whole detection process. The influence of NCS volume on the anodic peak current using DPV was shown in Figure S6. It could be seen that, the three modified GCEs all exhibited a well resolution between the anodic peak potential of EP (around 0.24 V) and PEA (around −0.17 V), which is consistent with the individual detection of EP and PEA, indicating the great sensitivities of the three

PUCNT/GCE. However, only oxidation peaks of DA and UA were seen on the FUCNT/GCE. This is because the poor electrical conductivity and selectivity of FUCNTs when used as electrode materials. To provide improved resolution of the peak potentials, differential pulse voltammetry (DPV) was applied under the same conditions (Figure 9A). For CNT/GCE and PUCNT/ GCE, oxidation of UA, DA, and AA could be distinguished; the separation potentials for UA−DA and DA−AA were 117 and 188 mV (CNT/GCE) and 125 and 183 mV (PUCNT/GCE), respectively. In contrast, for FUCNT/GCE, only the oxidation of UA and DA could be differentiated; the separation potential for UA−DA was 138 mV. Meanwhile, the peak currents decreased from CNT/GCE to FUCNT/GCE as a result of the electrochemical property changes that followed the variation in structure. Next, in the presence of 10 μM UA and 500 μM AA, the influence of DA concentration on the anodic peak current of the three modified GCEs was investigated using DPV. As shown in Figure S3, with addition of increased concentrations of DA from 1 to 100 μM, the anodic peak currents all increased gradually for the three electrodes. For CNT/GCE, the anodic peak potentials and currents of AA and UA both shifted slightly with the continuous addition of DA, indicating a good selectivity of CNT/GCE for AA, DA, and UA, mainly because the higher electrical conductivity and higher π−π interaction between the phenyl structure of DA and the hexagonal carbon structure of CNTs. In contrast, the anodic peak potentials and currents of AA and UA at PUCNT/GCE and FUCNT/GCE changed more or less, suggesting they have poorer selectivity compared with CNT/GCE. Then we studied the relationship between the DA concentration and the anodic peak current, as depicted in Figure 9B, the detection range was from 1−100 μM (R2 = 0.997) and the sensitivity determined by the slope of the linear fitting was 2.262 μA/μM for CNT/GCE. The linear detection range of PUCNT/GCE was 5−90 μM (R2 = 0.993) with a sensitivity of 0.759 μA/μM, while that for FUCNT/GCE was 10−80 μM (R2 = 0.992) with a sensitivity of 0.252 μA/μM. Along with the increased oxidative unzipping degree from CNTs to FUCNTs, sp2-hybridized carbon transitioned to sp3bonding, thus lowering the sp2/sp3 ratio. Furthermore, the 25800


Research Article


Figure 11. (a) CNT/GCE, (b) PUCNT/GCE, and (c) FUCNT/GCE: Linear relationship between the anodic peak current of (A) PEA or (B) EP and the volume of NCS.

■

CONCLUSIONS We have demonstrated a new strategy for studying the electrochemical activity and biosensitivity of carbon nanotube-based electrodes. The DOS for CNT derivatives can be easily altered by varying their graphite concentration (i.e., the sp2/sp3 ratio) through manipulating the oxidative unzipping degree. The CNT derivatives exhibited consistent electrochemical behavior for four investigated redox probes, despite their diverse HET mechanisms and influence factors. That is to say that the DOS play a dominant role in the electron-transfer kinetics. EIS and CV results reveal that CNTs with a high DOS exhibited superior electrochemical behavior to PUCNTs and FUCNTs for IrCl62−, [Fe(CN)6]3−, Fe3+, and AA. When using in biological detection, CNT/GCE could simultaneously detect and distinguish between AA, DA, and UA, while the three modified GCEs could all differentiate PEA and EP existed in the newborn calf serum. In addition, the three modified GCEs all provided the wide linear detection ranges with high sensitivities for DA, PEA, and EP. Our work broadens novel perspective of studying the electrochemistry of carbon derivatives and point toward new applications to electrochemical biosensing and electrocatalysis.

modified GCEs. Meanwhile, with the addition of NCS solution, the anodic peak current of PEA and EP both increased linearly; therefore, we studied the relationship of the anodic peak current of PEA or EP with the volume of NCS. As depicted in Figure 11A, for PEA, the linear range was 0.2−120, 1−110, and 5−100 μL with the sensitivity of 0.041, 0.030, and 0.024 μA/μL for CNT/GCE, PUCNT/GCE, and FUCNT/GCE, respectively. While for EP (Figure 11B), the linear range was 0.5− 120, 1−120, and 1−120 μL with the sensitivity of 0.002, 0.011, and 0.010 μA/μL for CNT/GCE, PUCNT/GCE, and FUCNT/GCE, respectively. Overall, the CNT derivatives could be used effectively for the determination of PEA and EP in real samples. Table 4. Comparison for Determinations of DA, PEA, and EP at CNT/GCE, PUCNT/GCE, and FUCNT/GCE analyte DA

PEA

EP

PEA (NCS)

EP (NCS)

electrode CNT/GCE PUCNT/ GCE FUCNT/ GCE CNT/GCE PUCNT/ GCE FUCNT/ GCE CNT/GCE PUCNT/ GCE FUCNT/ GCE CNT/GCE PUCNT/ GCE FUCNT/ GCE CNT/GCE PUCNT/ GCE FUCNT/ GCE

linear range (μM or μL)

sensitivity (μA/μM or μA/μL)

R2

1−100 5−90

2.262 0.759

0.997 0.993

10−80

0.252

0.992

8−120 10−120

0.086 0.067

0.998 0.994

20−120

0.024

0.988

15−120 40−120

0.076 0.062

0.996 0.996

20−120

0.425

0.996

0.5−120 1−110

0.041 0.030

0.996 0.992

5−100

0.024

0.991

0.2−100 1−120

0.002 0.011

0.993 0.997

1−120

0.010

0.995

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.5b07700. (1) Full XPS spectra of the three CNT derivatives (Figure S1), (2) dispersity of the three CNT derivatives (Figure S2), (3) DPV shows the addition of increased concentrations of DA for the three CNT derivativemodified electrodes (Figure S3), (4) DPV shows the addition of increased concentrations of PEA for the three CNT derivative-modified electrodes (Figure S4), (5) DPV shows the addition of increased concentrations of EP for the three CNT derivative-modified electrodes (Figure S5), and (6) DPV shows the addition of increased volumes of Newborn Calf Serum for the three CNT derivative-modified electrodes (Figure S6) (PDF) 25801


Research Article


■

Electrodes with Different Doping Levels. J. Phys. Chem. B 2004, 108, 15117−15127. (16) Gong, K.; Chakrabarti, S.; Dai, L. Electrochemistry at Carbon Nanotube Electrodes: Is the Nanotube Tip More Active than the Sidewall? Angew. Chem., Int. Ed. 2008, 47, 5446−5450. (17) Holloway, A.; Wildgoose, G.; Compton, R.; Shao, L.; Green, M. H. The Influence of Edge-Plane Defects and Oxygen-Containing Surface Groups On the Voltammetry of Acid-Treated, Annealed and “Super-Annealed” Multiwalled Carbon Nanotubes. J. Solid State Electrochem. 2008, 12, 1337−1348. (18) Chua, C. K.; Sofer, Z.; Pumera, M. Graphene Sheet Orientation of Parent Material Exhibits Dramatic Influence on Graphene Properties. Chem. - Asian J. 2012, 7, 2367−2372. (19) Shinde, D. B.; Pillai, V. K. Electrochemical Resolution of Multiple Redox Events for Graphene Quantum Dots. Angew. Chem., Int. Ed. 2013, 52, 2482−2485. (20) Gao, H.; Duan, H. 2D and 3D Graphene Materials: Preparation and Bioelectrochemical Applications. Biosens. Bioelectron. 2015, 65, 404−419. (21) Wang, Z.; Dai, Z. Carbon Nanomaterial-Based Electrochemical Biosensors: An Overview. Nanoscale 2015, 7, 6420−6431. (22) Vashist, S. K.; Luong, J. H. T. Recent Advances in Electrochemical Biosensing Schemes Using Graphene and Graphene-Based Nanocomposites. Carbon 2015, 84, 519−550. (23) Kosynkin, D. V.; Higginbotham, A. L.; Sinitskii, A.; Lomeda, J. R.; Dimiev, A.; Price, B. K.; Tour, J. M. Longitudinal Unzipping of Carbon Nanotubes to Form Graphene Nanoribbons. Nature 2009, 458, 872−876. (24) Higginbotham, A. L.; Kosynkin, D. V.; Sinitskii, A.; Sun, Z.; Tour, J. M. Lower-Defect Graphene Oxide Nanoribbons From Multiwalled Carbon Nanotubes. ACS Nano 2010, 4, 2059−2069. (25) Ferrari, A. C.; Robertson, J. Interpretation of Raman Spectra of Disordered and Amorphous Carbon. Phys. Rev. B: Condens. Matter Mater. Phys. 2000, 61, 14095−14107. (26) Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; Piscanec, S.; Jiang, D.; Novoselov, K. S.; Roth, S.; Geim, A. K.; Ferrari, A. C. Raman Spectrum of Graphene and Graphene Layers. Phys. Rev. Lett. 2006, 97, 187401. (27) Kowalczyk, S. P.; Ley, L.; Cavell, R. G.; Pollak, R. A.; Shirley, D. A.; McFeely, F. R. X-Ray Photoemission Studies of Diamond, Graphite, and Glassy Carbon Valence Bands. Phys. Rev. B 1974, 9, 5268−5278. (28) Royea, W. J.; Hamann, T. W.; Brunschwig, B. S.; Lewis, N. S. A Comparison between Interfacial Electron-Transfer Rate Constants at Metallic and Graphite Electrodes. J. Phys. Chem. B 2006, 110, 19433− 19442. (29) Xian, L.; Chou, M. Y.; Yan, J. Structural and Electronic Properties of Oxidized Graphene. Phys. Rev. Lett. 2009, 103, 86802. (30) Luo, Z.; Vora, P. M.; Mele, E. J.; Johnson, A. T. C.; Kikkawa, J. M. Photoluminescence and Band Gap Modulation in Graphene Oxide. Appl. Phys. Lett. 2009, 94, 111909. (31) Ambrosi, A.; Bonanni, A.; Sofer, Z.; Cross, J. S.; Pumera, M. Electrochemistry at Chemically Modified Graphenes. Chem. - Eur. J. 2011, 17, 10763−10770. (32) Ranganathan, S.; Kuo, T.; McCreery, R. L. Facile Preparation of Active Glassy Carbon Electrodes with Activated Carbon and Organic Solvents. Anal. Chem. 1999, 71, 3574−3580. (33) Banks, C. E.; Compton, R. G. New Electrodes for Old: From Carbon Nanotubes to Edge Plane Pyrolytic Graphite. Analyst 2006, 131, 15−21. (34) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 2nd ed.; Wiley−VCH: New York, 2000. (35) Nicholson, R. S. Theory and Application of Cyclic Voltammetry for Measurement of Electrode Reaction Kinetics. Anal. Chem. 1965, 37, 1351−1355. (36) Li, W.; Tan, C.; Lowe, M. A.; Abruña, H. D.; Ralph, D. C. Electrochemistry of Individual Monolayer Graphene Sheets. ACS Nano 2011, 5, 2264−2270.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This study was financially supported by the National Natural Science Foundation of China (Nos. 51172045 and 51402051). The authors thank Xiuxiu Dong from The Guangdong Provincial Key Laboratory of Food Quality and Safety, College of Food Science, South China Agricultural University, Guangzhou 510642, for kindly providing the Newborn Calf Serum. The authors also thank the associate researcher Xinglin Li from Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, for UPS testing.

■

REFERENCES

(1) Banks, C. E.; Davies, T. J.; Wildgoose, G. G.; Compton, R. G. Electrocatalysis at Graphite and Carbon Nanotube Modified Electrodes: Edge-Plane Sites and Tube Ends are the Reactive Sites. Chem. Commun. 2005, 829−841. (2) Wang, J. Carbon-Nanotube Based Electrochemical Biosensors: A Review. Electroanalysis 2005, 17, 7−14. (3) McCreery, R. L. Advanced Carbon Electrode Materials for Molecular Electrochemistry. Chem. Rev. 2008, 108, 2646−2687. (4) Pumera, M. The Electrochemistry of Carbon Nanotubes: Fundamentals and Applications. Chem. - Eur. J. 2009, 15, 4970−4978. (5) Yang, W.; Ratinac, K. R.; Ringer, S. P.; Thordarson, P.; Gooding, J. J.; Braet, F. Carbon Nanomaterials in Biosensors: Should You Use Nanotubes or Graphene? Angew. Chem., Int. Ed. 2010, 49, 2114−2138. (6) Pumera, M. Graphene-Based Nanomaterials and their Electrochemistry. Chem. Soc. Rev. 2010, 39, 4146−4157. (7) Ratinac, K. R.; Yang, W. R.; Gooding, J. J.; Thordarson, P.; Braet, F. Graphene and Related Materials in Electrochemical Sensing. Electroanalysis 2011, 23, 803−826. (8) Brownson, D. A. C.; Kampouris, D. K.; Banks, C. E. Graphene Electrochemistry: Fundamental Concepts through to Prominent Applications. Chem. Soc. Rev. 2012, 41, 6944−6976. (9) Mao, X.; Simeon, F.; Rutledge, G. C.; Hatton, T. A. Electrospun Carbon Nanofiber Webs with Controlled Density of States for Sensor Applications. Adv. Mater. 2013, 25, 1309−1314. (10) Wiggins-Camacho, J. D.; Stevenson, K. J. Effect of Nitrogen Concentration on Capacitance, Density of States, Electronic Conductivity, and Morphology of N-Doped Carbon Nanotube Electrodes. J. Phys. Chem. C 2009, 113, 19082−19090. (11) Patten, H. V.; Meadows, K. E.; Hutton, L. A.; Iacobini, J. G.; Battistel, D.; McKelvey, K.; Colburn, A. W.; Newton, M. E.; Macpherson, J. V.; Unwin, P. R. Electrochemical Mapping Reveals Direct Correlation between Heterogeneous Electron-Transfer Kinetics and Local Density of States in Diamond Electrodes. Angew. Chem., Int. Ed. 2012, 51, 7002−7006. (12) Nissim, R.; Batchelor-McAuley, C.; Henstridge, M. C.; Compton, R. G. Electrode Kinetics at Carbon Electrodes and the Density of Electronic States. Chem. Commun. 2012, 48, 3294−3296. (13) Mao, X.; Yang, X.; Rutledge, G. C.; Hatton, T. A. Ultra-WideRange Electrochemical Sensing Using Continuous Electrospun Carbon Nanofibers with High Densities of States. ACS Appl. Mater. Interfaces 2014, 6, 3394−3405. (14) Valentini, F.; Orlanducci, S.; Terranova, M. L.; Amine, A.; Palleschi, G. Carbon Nanotubes as Electrode Materials for the Assembling of New Electrochemical Biosensors. Sens. Actuators, B 2004, 100, 117−125. (15) Holt, K. B.; Bard, A. J.; Show, Y.; Swain, G. M. Scanning Electrochemical Microscopy and Conductive Probe Atomic Force Microscopy Studies of Hydrogen-Terminated Boron-Doped Diamond 25802


Research Article

ACS Applied Materials & Interfaces (37) Hernández-Ferrer, J.; Laporta, P.; Gutiérrez, F.; Rubianes, M. D.; Rivas, G.; Martínez, M. T. Multi-Walled Carbon Nanotubes/ Graphene Nanoribbons Hybrid Materials with Superior Electrochemical Performance. Electrochem. Commun. 2014, 39, 26−29. (38) DuVall, S. H.; McCreery, R. L. Self-Catalysis by Catechols and Quinones during Heterogeneous Electron Transfer at Carbon Electrodes. J. Am. Chem. Soc. 2000, 122, 6759−6764. (39) McDermott, M. T.; Kneten, K.; McCreery, R. L. Anthraquinonedisulfonate Adsorption, Electron-Transfer Kinetics, and Capacitance On Ordered Graphite Electrodes: The Important Role of Surface Defects. J. Phys. Chem. 1992, 96, 3124−3130. (40) Zhang, X.; Liao, Q.; Chu, M.; Liu, S.; Zhang, Y. Structure Effect On Graphene-Modified Enzyme Electrode Glucose Sensors. Biosens. Bioelectron. 2014, 52, 281−287.

25803


Carbon Nanotubes with Tailored Density of ... - ACS Publications

Recommend Documents