CitationDownload as .RIS
Emerald Group Publishing Limited
Copyright © 2012, Emerald Group Publishing Limited
Nanoscale energy harvesting and nanophotovoltaics
Article Type: Nanosensor update From: Sensor Review, Volume 32, Issue 2
This nanosensor update first considers nanoscale energy harvesting technologies and then, in line with the main theme of this issue, examines the role of nanomaterials in photovoltaics.
Nanoscale energy harvesting
Recent years have seen growing interest in developing energy harvesting devices based on nanomaterials, so-called “nanogenerators” (NGs), which could act as power sources for nanosensors and other electronic devices. The majority exploit the piezoelectric effect (Figure 1) and are based on materials such as zinc oxide (ZnO), barium titanate (BaTiO3) or gallium nitride (GaN). The first was reported by Professor Wang’s group at Georgia Institute of Technology in 2006 (Wang and Song, Science, 312, pp. 242-6) and this group continues to drive progress in this field. Most piezoelectric NGs can be categorised according to their geometery and there are three main types: vertical nanowire integrated nanogenerator (VING), lateral nanowire integrated nanogenerator (LING) and nanocomposite electrical generators (NEG). VING is a 3-D configuration, generally consisting of three layers: the base electrode, the vertically grown piezoelectric nanostructure and the counter electrode. The nanostructures are usually grown from the base electrode and then integrated with the counter electrode in full or partial mechanical contact with its tip (Figure 2). LING is a 2-D configuration comprising the base electrode, the laterally grown piezoelectric nanostructure and the metal electrode for Schottky contact. In most cases, the substrate is much thicker than the diameter of the nanostructure, so the individual nanostructure is subjected to a purely tensile strain (Figure 3). LING is effectively an expansion of a single wire generator, where a laterally aligned nanowire is integrated on the flexible substrate. NEG is another type of 3-D structure with three main parts: the metal plate electrodes, the vertically grown piezoelectric nanostructure and a polymer matrix.
In 2010 the Wang group reported the first use of piezoelectric NGs to power a nanosensor (Xu et al., Nature Nanotechnology, 5, pp. 366-73). It was shown that the integration of VING and LING NGs are capable of generating sufficient power to operate real devices. The lateral integration of 700 rows of ZnO nanowires produced a peak voltage of 1.26 V at a strain of only 0.19 per cent, which is potentially sufficient to recharge an AA battery and in a VING device, the vertical integration of three layers of ZnO nanowire arrays yielded a peak power density of 2.7 mWcm−3. A 4 mm2 VING with an output of 20\-40 mV was used to power ZnO nanowire-based pH and UV sensors and thus demonstrated the feasibility of creating wholly self-powered nanosensor systems composed entirely of nanowires. In 2011 the same group reported that it had taken the self-powered sensor concept a step further by demonstrating a self-powered wireless sensing device (Hu et al., Nano Letters, 11, pp. 2572-7). This comprised az piezoelectric NG which charged a capacitor to power a sensor and an RF transmitter. The NG comprised a free cantilever beam that consisted of a five-layer structure: a flexible polymer substrate, ZnO nanowire textured films on its top and bottom surfaces and Au/Cr contact electrodes on the outer surfaces. The nanowires were about 150 nm in diameter and 2 μm long. When the NG was strained to 0.12 per cent at a rate of 3.56 per cent per second, the output reached 10 V and the output current exceeded 0.6 μA, representing a volume current density of 1 mA/cm3 and a power density of 10 mW/cm3. Storage of the harvested energy was achieved by using an integrated low-loss, full-wave bridge rectifier connected between the NG and a capacitor. To demonstrate wireless data transmission, a single transistor RF transmitter was used, with the oscillation frequency tuned to around 90 MHz, and a portable AM/FM radio received the signal. The sensor used to demonstrate the concept was an IR LED and a silicon phototransistor and the photocurrent generated was periodically transmitted using the energy stored in the capacitor. Because the power consumption of the phototransistor was 100 mW, it required the energy harvested from 1,000 cycles of the NG to power the phototransistor and the transmitter simultaneously. Each time it was triggered, the signal received by the phototransistor modulated the transmitting signal and the information was received by the radio. The system operated over a range of 5-10 m. This work demonstrates the feasibility of using a ZnO nanowire-based NGs in self-powered sensor systems with the capability of remote data transmission. The vibrations used to power the NG could be scavenged from a wide variety of sources ranging from a person’s pulse, breathing or walking motion, to air currents or cars driving over a bridge. Potential applications are in wireless sensor networks and as stand-alone devices in fields such as clinical biosensing, structural monitoring and environmental sensing. A schematic of this concept, together with the technological evolution of NGs, is shown in Figure 4.
A further development by the Wang group is the piezotronic memory. Reported in 2011 (Wu and Wang, Nano Letters, 11, pp. 2779-85), this is a piezoelectric resistive switching device, fabricated from ZnO nanowires, in which the write-read access of the memory cells is controlled by electromechanical modulation. Adjusted by the strain-induced polarisation charges created at the semiconductor/metal interface under externally applied deformation by the piezoelectric effect, the resistive switching characteristics of the cell can be modulated in a controlled manner and the logic levels of the strain stored in the cell can be recorded and read out. Fabricated on flexible substrates, arrays of these devices could provide a new way of interfacing the mechanical activity of the biological world with conventional electronics. Wang argues that these devices provide another of the components needed for fabricating complete self-powered nanoelectromechanical systems on a single chip. This work was sponsored by the Defence Advanced Research Projects Agency, the National Science Foundation, the US Air Force and the US Department of Energy. The device is shown in Figure 5.
A very different approach to powering a sensor has been demonstrated by a group from the Nanoscale Synthesis and Characterization Lab at the Lawrence Livermore National Laboratory who have developed a battery-less chemical nanosensor that relies on the interaction between the analyte and semiconducting nanowires (Wang et al., Advanced Materials, 23, pp. 117-21). Two different configurations have been studied: the first is fabricated from vertically aligned, single-crystal ZnO nanowires (n-type) with diameters in the range of 60-120 nm and the second utilises 30-55 nm diameter, randomly aligned, boron-doped silicon nanowires (p-type). The working principle of the nanosensor relies on the partial exposure of the nanowires to chemical species and the use of a non-ohmic contact that is necessary for the nanosensor to function. On adsorption of molecules such as ethanol, the carrier density (electrons for n-type and holes for p-type nanowires) of the exposed nanowire segment is altered, leading to a modification of the conduction band’s profile. This process metastably tilts the Fermi level of the nanowires and leads to a detectable potential offset between the top (gold) and ground (silver) contacts. This suggests that the magnitude and shape of the voltage signals observed are closely allied to the electron donating/withdrawing ability of molecules; the nature of the non-ohmic contact, which can be modified by molecules; and the charge carrier density within the nanowires. To test the ZnO nanosensor’s response to chemical species at room temperature, a small volume (∼2.2 μL) of ethanol was applied to the surface and the change in the voltage between the two ends of the nanowires was monitored. A sharp rise in the voltage with the peak value in the order of ∼170 mV was observed. The rise was almost instantaneous but decayed slowly to zero as the ethanol evaporated, indicating that the reaction is reversible. A minimal response arose from the application of hexane, suggesting that the sensor is chemically selective. The magnitude of the peak voltage scales with the dipole moment and the surface coverage of the solvents and as different solvents have differing dipole moments, it should be possible to detect selectively a wide range of compounds. Further work with over 15 other organic compounds confirmed this and also that the sensor can detect differing solvent concentrations. The next step is to test the sensors with more complex molecules such as those associated with explosives and biological systems. Patent applications have been filed for this research.