E nhanced water flow through atomic smooth and hydrophobic carbon nanotubes (CNTs) have been demonstrated by both theoretical calculations and experiments. 1À5 There is, however, a great controversy between theory and experiments and even between experiments. The very limited experiments using CNTs membrane demonstrated enormous water flow velocity up to 5 orders of magnitude faster than predicted from conventional fluid-flow theory with three orders of deviation from different sources.1,2 In contrast, molecular dynamics (MD) calculation only gives a rate enhancement of 47À6500 for CNTs with diameters of 4.99À0.81 nm.3,5 One more general debate is whether there exists a clear transition from continuum to subcontinuum transport as the tube diameter shrinks to subnanometer regime. 5 The bottleneck for experimental attempts arises from fabrication of CNTs membrane with well-defined structures and the rational estimation of the available flow area.1 Here we show a single-tube level approach for elucidating such fundamental nanofluidic issues. The unique field effect transistors (FETs) array-based experimental design enables a direct measurement of water flow velocity inside individual CNTs. Our work demonstrates a rate enhancement of 51 to 882 for CNTs with diameters of 1.59 to 0.81 nm, which supports the MD calculation.3,5 Additionally, we achieved the first experimental evidence for the transition from continuum to subcontinuum flow by varying the diameters of CNTs.The key of our approach is to trace the water flow "front" inside an individual millimeter long CNT electrically with a configuration of three FETs in series (Figure 1a,b). The FET1 is used to "in-situ" open the tube end under water droplet by electrical breakdown, 6,7 and the synchronous FET2 and FET3 to detect the water front flowing in based on its influence on the current flow (Figure 2). 8,9 It should be emphasized that opening the tube end under water is a determinant factor for the success of this experimental design. A bias voltage of 0.01 V was applied on FET2 and FET3 (no gate voltage) all the time to detect current change. Simply by measuring the time delay of current signal jumps between FET2 and FET3 with a given interspacing, we can then estimate the average water flow velocity inside the nanotube. The CNT-FETs structure was constructed through directly growing ultralong CNT on SiO 2 /Si substrate with predesigned Pt-pattern (Figure 1b,c). Carbon nanotubes were synthesized by gas flow-directed chemical vapor deposition (CVD) method. 10À12 The catalysts pattern was made on growth substrate using PDMS stamp from the ethanol solution of 0.01 mol/L FeCl 3 . The typical growth conditions are 930À950°C, 3 sccm CH 4 and 5 sccm H 2 . Pt was sputtered and patterned as electrodes on SiO 2 /Si substrate by standard technique of photolithography and magnetron sputtering. The as-grown CNTs were characterized by scanning electron microscopy (SEM) followed by gold-wire wedge bonding, water filling and velocity measurement. A drop of pure water (18.2MΩ ...

Understanding of the continuity and the microscopic structure of as-grown graphene on Cu foils through the chemical vapor deposition (CVD) method is of fundamental significance for optimizing the growth parameters toward high-quality graphene. Because of the corrugated nature of the Cu foil surface, few experimental efforts on this issue have been made so far. We present here a high-resolution scanning tunneling microscopy (STM) study of CVD graphene directly on Cu foils. Our work indicates that graphene can be grown with a perfect continuity extending over both crystalline and noncrystalline regions, highly suggestive of weak graphene-substrate interactions. Due to thermal expansion mismatch, defect-like wrinkles and ripples tend to evolve either along the boundaries of crystalline terraces or on noncrystalline areas for strain relief. Furthermore, the strain effect arising from the conforming of perfect two-dimensional graphene to the highly corrugated surface of Cu foils is found to induce local bonding configuration change of carbon from sp(2) to sp(3), evidenced by the formation of "three-for-six" lattices.

An electrochemical DNAzyme sensor for sensitive and selective detection of lead ion (Pb(2+)) has been developed, taking advantage of catalytic reactions of a DNAzyme upon its binding to Pb(2+) and the use of DNA-Au bio-bar codes to achieve signal enhancement. A specific DNAzyme for Pb(2+) is immobilized onto an Au electrode surface via a thiol-Au interaction. The DNAzyme hybridizes to a specially designed complementary substrate strand that has an overhang, which in turn hybridizes to the DNA-Au bio-bar code (short oligonucleotides attached to 13 nm gold nanoparticles). A redox mediator, Ru(NH3)6(3+), which can bind to the anionic phosphate of DNA through electrostatic interactions, serves as the electrochemical signal transducer. Upon binding of Pb(2+) to the DNAzyme, the DNAzyme catalyzes the hydrolytic cleavage of the substrate, resulting in the removal of the substrate strand along with the DNA bio-bar code and the bound Ru(NH3)6(3+) from the Au electrode surface. The release of Ru(NH3)6(3+) results in lower electrochemical signal of Ru(NH3)6(3+) confined on the electrode surface. Differential pulse voltammetry (DPV) signals of Ru(NH3)6(3+) provides quantitative measures of the concentrations of Pb(2+), with a linear calibration ranging from 5 nM to 0.1 microM. Because each nanoparticle carries a large number of DNA strands that bind to the signal transducer molecule Ru(NH3)6(3+), the use of DNA-Au bio-bar codes enhances the detection sensitivity by five times, enabling the detection of Pb(2+) at a very low level (1 nM). The DPV signal response of the DNAzyme sensor is negligible for other divalent metal ions, indicating that the sensor is highly selective for Pb(2+). Although this DNAzyme sensor is demonstrated for the detection of Pb(2+), it has the potential to serve as a general platform for design sensors for other small molecules and heavy metal ions.