Abstract
The electronic transport and the sensing performance of an individual SnO2 crossed-nanowires device in a three-terminal field-effect transistor configuration were investigated using a combination of macroscopic transport measurements and scanning surface-potential microscopy (SSPM). The structure of the device was determined using both scanning electron- and atomic force microscopy data. The SSPM images of two crossed one-dimensional nanostructures, simulating a prototypical nanowire network sensors, exhibit large dc potential drops at the crossed-wire junction and at the contacts, identifying them as the primary electroactive elements in the circuit. The gas sensitivity of this device was comparable to those of sensors formed by individual homogeneous nanostructures of similar dimensions. Under ambient conditions, the dc transport measurements were found to be strongly affected by field-induced surface charges on the nanostructure and the gate oxide. These charges result in a memory effect in transport measurements and charge dynamics which are visualized by SSPM. Finally, scanning probe microscopy is used to measure the current-voltage characteristics of individual active circuit elements, paving the way to a detailed understanding of chemical functionality at the level of an individual electroactive element in an individual nanowire.
Original language | English |
---|---|
Article number | 044503 |
Journal | Journal of Applied Physics |
Volume | 98 |
Issue number | 4 |
DOIs | |
State | Published - Aug 15 2005 |
Funding
One of the authors (S.V.K.) acknowledges Eugene P. Wigner fellowship at the Oak Ridge National Laboratory, managed by UT-Battelle, LLC, for the U.S. Department of Energy under Contract No. DE-AC05-00OR22725. Three of the authors (S.V.K., A.P.B., and D.B.G.) acknowledge support from ORNL Laboratory Research and Development funding. The work at UCSB was supported via AFOSR DURINT Grant No. F49620-01-1-0459, and by funding through the Institute for Collaborative Biotechnologies supported by the U.S. Army Research office. FIG. 1. (a) Schematic band diagram of an ideal semiconductor 1D nanostructure joining two highly conducting electrodes. (b) A conductometric device based on two crossed quasi-1D nanostructures. Center: the cross-junction barrier resulting from the partial depletion of the near-surface region. The Schottky barriers at the electrodes are also shown. (c) Experimental setup for carrying out spatially resolved transport measurements. (d) Topographic image of SnO 2 nanowire. The lateral size is 50 × 50 μ m 2 and the vertical scale is 500 nm. FIG. 2. (a) Topographic AFM image of the nanostructure and (b) SEM image of the same system. By combining the vertical dimensions determined by AFM and the lateral sizes determined by SEM, the dimensions of the 1D elements were determined as a 650 × 85 - nm 2 nanobelt and an ∼ 25 - nm -diameter nanowire. FIG. 3. (a) Typical response I DS ( V SD = 2 V ) of the individual homogeneous nanobelt to two sequential ∼ 10 − 3 - Torr oxygen (oxidizing agent) pulses followed by two sequential pulses of hydrogen (reducing agent) at 250 °C. (b) The comparison of the response functions of two nanostructures [ I DS ( V SD = 2 V ) at 200 °C] toward the oxygen pulse. The bold curve (A) corresponds to measurements on a defect-free single nanobelt with Ohmic contacts. The narrow line (B) represents measurements for the nanobelt crossed by a nanowire, which also forms Schottky barriers at the contacts. I DS partially recovers in (A) after oxygen exposure, due to the partial evaporation of chemisorbed oxygen from the nanostructure’s surface at 200 °C. FIG. 4. Surface-potential images for biases of 6 V [(a), (c), and (e)] and − 6 V [(b), (d), and (f)]. Voltages were applied to the bottom [(a) and (b)], top [(c) and (d)], and both [(e) and (f)] electrodes. In each case the back gate electrode is grounded. The vertical scale is 10 V. FIG. 5. (a) The absolute potential along the nanowire for + 6 V applied to bottom and top electrodes. Differential potential profiles for + 6 - V (b) and − 6 - V (c) biases. Note the asymmetry between the curves obtained with top and bottom electrodes positively biased. (d) Potential drop on the top contact, nanowire junction, and bottom contact as a function of lateral bias. FIG. 6. Surface potential on the biased nanowire after (a) 10-min, (b) 20-min, and (c) 1-h scanning illustrating the smearing of potential contrast due to the mobile charge effect. (d) Surface potential at 10-V negative bias and (e) immediately after the bias is off illustrates the formation of charged negative halo. (f) After a week time, the halo disappears. The scale is 6 V [(a)–(c)], 10 V (d), and 1 V [(e) and (f)]. FIG. 7. (a) A comparison of the transient response of the wired nanowire (top curve) and empty pads (lower curve) to a sudden change of the gate potential from − 8.7 to 0 V. The accumulation of the positive mobile charges induced by a negative gate bias gives rise to a positive gating when the bias is off. (b) I - V characteristics of the nanowire [complementary to the top curve in (a)] taken sequentially every few seconds after the sudden change of gate potential. FIG. 8. (a) Macroscopic I − V curve of the device. The inset shows the ratio of currents for opposite bias polarities. (b) I − V curves of the top contact (squares), nanowire junction (diamonds), and bottom contact (triangles) from the combination of SSPM and macroscopic transport data. The inset shows the expanded view of the I − V curves.