scenario VIMPA project | partners people | publications project material  
  news media | related links and material | contact   

The challenge addressed by VIMPA is to develop usable microsystems able to effectively power a wide range of devices. There are several fields waiting for high energy density power supply.
The first domain is portable electronics, including for example cellular phones, laptop computers, camcorders and other consumer devices. In fact, we are observing a real discrepancy between what is available to the user and what should be the real performance of power supplies. A laptop or a digital camera, for example, have a typical autonomy of some (few) hours between recharges, while a day-based autonomy should be reached for an effective operation. Optimising power consumption is evidently only an incremental and partial solution, because of the high performances more and more expected from new devices (e.g., wide LCD screens, powerful CPUs). Dramatically increasing the energy density in electrochemical batteries seems not to be a realistic issue, as proven by the trend in results obtained by batteries companies in the last years, where the density of 1MJ/Kg seems to be an asymptotic value.
The second domain of interest regards systems able to perform autonomous tasks without being connected to a power cord or being constrained in a spatial location where energy can be delivered both trough docking stations or via wireless means. A challenging example is the one of “field microrobotics”, where swimming, locomoting or even flying microsystems can be exploited for environmental inspection.
A third field addresses biomedical devices for health assistance requiring high energy density and autonomy. The most challenging example is maybe represented by artificial heart, but also other biomedical systems would take advantage from new, high performance power cells, like autonomous microcapsules.
The above mentioned needs led to a wide research effort in two main areas where a discontinuity in power generation is expected: the one of fuel cells and the one of Power MEMS. Just for reference also nuclear energy has been considered as a long term candidate for power generation in small systems, and in the USA a small category of funding has been allocated for this activity by the Nuclear Engineering Education Research (NEER) at the Energy Department.

State of the Art in Power MEMS
In the area of Power MEMS several micro-engine development programs are underway. These include both microturbines and positive displacement micromachines. These projects, mentioned in the following paragraph, are moved by the common motivation to replace batteries with integrated packages composed of an engine-generator, a control unit and a fuel tank. As already mentioned, the rationale for developing micro-engines is rooted in energy density; hydrocarbon fuels such as propane, have a lower heating value of approximately 46 MJ/Kg. Batteries on the other hand, have energy densities of 1 MJ/Kg at most. Consequently an engine-generator would only need to have an overall fuel conversion efficiency of 2.5 % to surpass any battery. Of note, the overall fuel conversion efficiency of the smallest mass-produced model airplane engine (0.16 cm3 displacement, absolutely not optimized regarding consumption) is better than 4%. In addition, power MEMS can always produce power as long as fuel is available, they can be recharged (i.e., refilled) in a very fast time and they do not pose a disposal problem when they need to be replaced because they do not contain dangerous or polluting components, only simply structural and electrical parts. Finally, microengines can be used in two other ways than as electrical generators: directly as prime movers, thus eliminating the need for electrical actuators, or as a source for compressed gas, enabling power pneumatics in microsystems or microjet technology.
Microturbines have been developed in USA under DARPA funding, in Japan and in Europe, but the problem of high rotating speeds and bearings overheating restricted experimental devices to be mainly operated using cold compressed air.
Regarding burners for turbomachinery, complex combustors are being developed in order to overcome the problem of quenching in the micro domain, consisting in tortuous burners with recirculation of thermal energy from the combustion products to preheat the reactants. Also the use of catalyst has been explored, exploiting expensive platinum coated microstructures.
As a general remark, micro turbomachinery is still in a very early stage, where individual components are being developed but no real power MEMS have been integrated yet.
On the other hand, a very promising approach is represented by positive displacement micromachines, where thermal energy does not have to be converted into kinetic energy of the fluid. This is particularly advantageous when dimensions are scaled down to the sub-millimetre scale, avoiding pressure losses caused by the leading role played by drag forces with respect to inertial forces (low Reynolds numbers), which is a typical problem in microturbines, whose efficiencies are in the order of 15 %.
Furthermore, in these systems combustion is not critical, thanks to the possibility of increasing for a short time the pressure and the temperature of the fluid by exploiting a sudden compression. In fact at the University of Minnesota the feasibility of micro-combustion based on Homogeneous Charge Compression Ignition (HCCI) was proven. It is interesting to observe that, following these considerations, the industrial partner of the research project, Honeywell Inc., filed a patent for a microcombustion engine (US pat. n. 6276313). However the proposed architecture exploited by the inventors consists of a free-piston engine, a working principle derived from macro engines. In this system the moving parts (pistons) are sliding in a frame and they have to provide sealing in order to obtain the adequate levels of compression and to open and close the ports for gas exchange, similarly to two strokes engines. Two requirements are evidently involved: the low friction and the good sealing. It is clear that the only technological way to fit both issues is to fabricate coupling surfaces with extremely good finishing, which is a very challenging if not impossible task in the microdomain. A similar approach, using sliding pistons, has been also adopted very recently in Japan.

Power MEMS vs. Fuel Cells
Fuel cells presently are either very sensitive to fuel impurities (such as CO in polymer-based fuel cells operating on H2) or require very high operating temperatures, which delay startups and cause shortened service life due to thermal cycling stresses. Extensive research has determined methanol to be the best choice but this technology still needs development and currently no solutions for mass production have been developed.
Also from an energetical point of view fuel cells are still not very effective. Last results in industrial research (Toshiba Corp., Japan) allowed companies to give specifications for next generation of fuel cells, to be introduced on the market in 2004. In particular the best product is a Toshiba direct methanol fuel cell, having a weight (excluding fuel tank) of 900 gr., with an output of 12W and a Specific Fuel Consumption (SFC) of 2·10-7 Kg/J. If this result is compared with mini-generators for personal power need based on HCCI, conclusions are very interesting. In fact a prototype of electrical generator exploiting a mini HCCI engines has the following characteristics: weight of 2.3 Kg, output electrical power of 500 W, specific fuel consumption of 1.7·10-7 Kg/J. It is therefore clear that miniaturization is really welcome for HCCI engines.
  Project funded by the European Commission under the NEST (New and Emerging Science and
  Technology) activity of the Sixth Framework Programme (FP6), contract No. 511889