Seamless Three-Dimensional Carbon Nanotubes-Grown Graphene Hybrid Films Modified with Heme for Electrochemical Biosensing with High Stability and Sensitivity

INTRODUCTION Seamless three-dimensional (3D) sp2 carbon-based nanomaterials, such as foam-like graphene and carbon nanotubes-grown graphene (CNTs/G) hybrid films, are one of the promising candidates as new electrode materials because of their high electrical conductivity and electroactive surface area per unit planar/footprint area. In particular, charge carriers are movable in all three dimensions without significant contact resistance. Seamless 3D sp2 carbon-based nanomaterials would therefore be applied to highly sensitive sensors and high power fuel cells. The determination of H2O2 using heme-containing proteins, such as horseradish peroxidase (HRP), has widely been employed for pharmaceutical and environmental analyses and immunoassay. Although HRP exhibits high catalytic activity, its stability is low. Meanwhile, hemin, which is one of the heme molecules and the active site of peroxidase, exhibits low catalytic activity, but high stability because of their simple structure without long chain polypeptide, compared with HRP. Even if the catalytic activity per one molecule of hemin is low, the surface coverage of hemin would be significantly enhanced because of its small molecular size, reflecting that the total catalytic activity at the CNTs/G film surface should be improved. In the present work, we therefore constructed seamless 3D CNTs/G film-coated electrodes modified with hemin toward the development of not only highly stable but also highly sensitive electrochemical biosensors. By comparing with seamless two-dimensional graphene film-coated electrodes modified with hemin, we confirmed the superiority of the hemin-modified CNTs/G film in electrochemical biosensing. EXPERIMENTAL The graphene film was prepared on a copper foil in an electric tubular furnace at 1030 ˚C under flowing argon, hydrogen, and methane. After the decoration of iron film at the graphene-formed copper foil by e-beam evaporator, the resultant copper foil was further annealed at 750 ˚C under flowing argon, hydrogen, and acetylene to obtain the seamless 3D CNTs/G film. After the removal of the copper foil and iron film in both FeCl3 and HCl aqueous solutions completely, CNTs/G or graphene films were transferred to the glassy carbon electrode surface. Hemin and HRP were immobilized at each surface of CNTs/G and graphene electrodes via 1-pyrenebutytic acid N-hydroxysuccinimide ester (hemin/CNTs/G, hemin/G, HRP/CNTs/G, and HRP/G electrodes). The H2O2 reduction at hemin-modified CNTs/G and graphene electrodes was evaluated by amperometry. After the working electrode was polarized at +150 mV vs. Ag|AgCl and steady-state current was obtained, H2O2 solution was added into phosphate buffer (pH 7.4) as an electrolyte and steady-state current was monitored. RESULTS AND DISCUSSION Cyclic voltammetry was performed at the hemin/CNTs/G, hemin/G, HRP/CNTs/G, and HRP/G electrodes in deaerated phosphate buffer. Redox peaks appeared at hemin/CNTs/G and hemin/G electrodes at approximately -310 mV, corresponding to (P)Fe2+/3+ (P = porphyrin ring), whereas no redox peaks were clearly observed at HRP/CNTs/G and HRP/G electrodes probably due to deeply embedded redox site in insulating polypeptide. We therefore used hemin/CNTs/G and hemin/G electrodes for further studies. The redox peak current for both electrodes was proportional to the scan rate below at least 500 mV s-1, indicating that the redox reaction was a surface controlled process. In addition, the redox peak current at the hemin/CNTs/G electrode was approximately six-folds larger than that at the hemin/G electrode. This is attributed to the larger electroactive surface area of the CNTs/G film because of 3D structure. Based on the relationship between redox peak currents and scan rate, the surface coverage of hemin at the CNTs/G film was determined to be about ~10-10 mol cm-2. This value was about six times larger than that at the graphene electrode and two orders of magnitude larger than the theoretical value of monolayer HRP (ca. ~10-12 mol cm-2) at the flat surface. We next measured cathodic current responses to H2O2 at CNTs/G and graphene electrodes, on which hemin was immobilized. The response at both electrodes increased with increasing H2O2 up to about 1 mM and then leveled off. Since the current response was not influenced when the electrolyte solution was stirred, the catalytic cathodic currents obtained here were kinetically controlled. All electroactive hemin molecules should therefore contribute to the current. The cathodic current for the hemin/CNTs/G electrode was about six-folds lager than that for the hemin/graphene electrode, reflecting that the surface coverage of electroactive hemin at the CNT/G electrode was about six times larger than that at the graphene electrode, as mentioned above. Thus, the CNTs/G film modified with hemin would be applicable to highly stable and sensitive electrochemical biosensors to H2O2.