Thrust I: Energy Harvesting and Storage
Goals: The energy thrust of the program entails nano-enabled energy harvesting, energy storage, and energy management components. The energy-harvesting component of ASSIST will focus on motion and temperature. The goal is to achieve a 10 X improvement in thermoelectric and piezoelectric energy harvesting systems, and a 10 – 100 X improvement in energy storage. The challenge is to create efficient and robust energy harvesting approaches that provide continuous power with a wearable form-factor.
Energy Harvesting from Body Heat: The thermoelectrics effort focuses on combining nanostructured, high figure of merit materials with advances in material and device design to minimize parasitic electrical and thermal resistances. This task will focus on fabrication of flexible thermoelectric arrays made of low-dimensional materials. A good thermoelectric material must possess high electrical conductivity, low thermal conductivity and high Seebeck coefficient. Nanostructuring greatly reduces the thermal conductivity by phonon scattering, resulting in a 100-fold improvement in thermoelectric performance. A figure-of-merit value in excess of 1 was shown at 200K, making it a good candidate for thermoelectric applications. Research will be directed towards flexible microchannel heat sinks to minimize parasitic thermal resistance losses, which will utilize grapheme/graphite layers. We will work closely with Phononic Devices, a start-up company at NC State. Extensive modeling of the electron and phonon transport will be undertaken.
Energy Harvesting from Body Motion: The ASSIST program will optimize piezoelectric materials with on-harvester active voltage amplification, and heterogeneous integration with low stiffness elastic layers. Fig. 1 shows a comparison of available thin films for piezoelectric energy harvesting in terms of the in-plane piezoelectric coefficient, e31,f, and the figure of merit for energy harvesting (assuming that the majority of the mechanical energy is stored in the passive elastic layer). ASSIST team members have recently shown that by tailoring the internal electric field and the strain state in PbZr1-xTixO3 films, it is possible to decrease the permittivity by a factor of 10, increasing the figure of merit for energy harvesting by an order of magnitude relative to AlN. We will integrate piezoelectric materials with engineered nano-domain states into highly compliant structures with nonlinear structure dynamics optimized to match biological motions. To further increase the efficiency of the piezoelectric MEMS energy harvesters, optimized piezoelectric materials will be integrated with ZnO interposer electronics for on-MEMS voltage amplification and active switching harvesting circuits. This will allow the energy harvesting system to be reconfigured to optimize energy extraction from an array of cantilevers.
Energy Storage: ASSIST will target the under-researched space between conventional capacitors and battery-like materials. By marrying electrochemical and electrostatic capacitor technologies, capabilities comparable to lead acid batteries will be achieved. Nanostructured C electrodes will provide high surface area, low equivalent series resistance. Two key approaches will be explored to increasing stored energy: a hybrid barrier layer supercapacitor in which ionic, rather than electronic space charge polarizability is employed, and a nano-electrochemical supercapacitor in which Faradaic reactions are used to increase energy densities and slow discharge rates. Preliminary data demonstrate stored energy densities of near 396 F/g while conventional supercapacitors with the same electrodes are 124 F/g. We anticipate a path towards storing up to 1000 J/cc in the ASSIST supercapacitors.
Power Conditioning (in conjunction with Thrusts 2 and 4): The harvested energy from various sources will be used to charge a supercapacitor and/or a rechargeable battery that can then act as a constant power source of power for the load. However, the harvested energy available as AC or DC voltages needs to be converted to appropriate DC levels required for supercapacitor or battery charging. This is accomplished by employing AC-DC or DC-DC power converters. In addition, with multiple energy sources available at the same time, an ultra-low-power Source Power Management system is necessary for achieving maximum power output at high power conversion efficiencies. The load for the energy storage supercapacitor will consists of various functions including RF/analog, digital and sensing systems. Each of these has specific power supply requirements, and they are met by choosing a dedicated DC-DC converter for the three subsystems. Power conditioning and DC-DC conversion are likely to be needed at the energy storage output, especially for supercapacitors with their variable voltage output. For example, RF/analog systems require low supply noise which can be met by a Linear Regulator. The digital systems have higher and more dynamic power requirements, generally met by a step-down switch-mode Buck converter. The sensors may require high/low voltages (compared to VBAT) and thus a Buck-Boost converter may be necessary. Finally, the load current needs to be minimized based on the various operating modes of the WSN. As an example, the digital circuitry may need to be operated at lower clock frequency in standby mode, for which VDIG needs to be dynamically scaled-down for significantly reducing power consumption. These functions and more will be implemented in an ultra-low-power Load Power Management system.