Currently, the primary way to power MEMS based wireless sensors is a conventional battery. In addition, vibrational and thermal energy harvesters are starting to see limited use in some cases. Other approaches for power generation include micro fuel cells, as well as microscale combustors that burn hydrocarbons and then convert the motion of the micro-rotor to electricity. In this interview, we spoke with Professor Jeongmin Ahn at Syracuse University about the current state of microscale combustor research, existing challenges, and emerging approaches. In addition to discussing microscale combustors, Professor Ahn also provides comparisons to conventional batteries, energy harvesters and micro fuel cell technologies.
MEMS Investor Journal: What are the main types of ways that people are currently powering wireless sensor nodes?
Professor Jeongmin Ahn: Batteries have seen significant advances in recent years, but their power density is still far inferior to combustion devices.
MEMS Investor Journal: In terms of energy density, how do conventional batteries, hydrogen-powered fuel cells and microscale combustion technologies compare? Based on these numbers, for how long can an energy source (say, a cubic centimeter in size) of each type power a typical wireless sensor node?
Professor Jeongmin Ahn: Fuel cells can produce far more electrical energy per unit weight than batteries can. Also, the use of hydrocarbon fuels for electrical power generation provides enormous advantages over conventional batteries both in terms of energy storage per unit mass and in terms of power generation per unit volume; even when the conversion efficiency from thermal energy to electrical energy is taken into account. Hydrogen provides an energy storage density of ~120 MJ/kg and hydrocarbon fuels provide ~50 MJ/kg, whereas even modern lithium ion batteries provide only ~ 0.7 MJ/kg.
For this reason, automotive and aviation vehicles employ internal combustion engines for prime moving and electrical power generation almost entirely to the exclusion of batteries, even in vehicles whose mass may be less than 1 kg or more than 100,000 kg.
Hydrocarbon fuels have approximately 70 times more energy density than the best modern batteries. In other words, even a device with very low energy conversion efficiency of hydrocarbon fuels to electricity, say 10% overall efficiency, still has a more than 5 times longer-lasting power source than can be realized with batteries.
MEMS Investor Journal: Do you think that energy harvesting will have a major role to power wireless sensor nodes? What type of energy harvesting do you think is especially promising? Why?
Professor Jeongmin Ahn: The demand for energy to power miniature applications, including portable electronics, wireless remote sensors, micro-robots, UAV, etc, has increased in recent years and will continue to do so. Unlike in stationary power generation, however, renewable energies such as solar, wind, geothermal, and hydropower cannot be easily adopted in such systems due to the intermittency in their availability and difficulties in storage. Batteries have seen significant advances in recent years, but their power density is still far inferior to combustion devices.
Therefore, the best source to harvest energy is by creating electricity from the chemical energy available in various hydrocarbon fuels. This remains a versatile and cost-effective approach in a wide variety of applications.
MEMS Investor Journal: What are the most common fuel cell technologies? What are their limitations?
Professor Jeongmin Ahn: Let me first address fuel cells, which have both pros and cons when it comes to microscale combustion. Fuel cells can produce far more electrical energy per unit weight than batteries can, and thus are widely used in specialized applications where minimum weight is critical, e.g., spacecraft. Additionally, fuel cells are not heat engines and thus are not subject to the thermodynamic limitations on efficiency that heat engines suffer. Of course, this is not to say that catalysis and membrane-based fuel cells do not have their problems. Fuel cells were invented shortly after the internal combustion engines, so why do they lag so much in development? Listed below are some of the reasons.
Proton Exchange Membrane (PEM) fuel cells require hydrogen, which is an outstanding fuel in many ways. Its energy content per unit weight is nearly three times that of hydrocarbons, it has outstanding electrochemical properties for use in fuel cells, and is an extremely clean-burning fuel.
On the other hand, hydrogen is not a fuel per se because it does not exist in nature in significant quantities. Rather, hydrogen is an energy carrier that must be made from something else, thus cost effective production of hydrogen from energy “feedstocks” (coal, petroleum, nuclear energy, solar energy, biomass, etc.) is a very substantial barrier to any hydrogen-based economy. Additionally, while hydrogen is indeed very clean-burning, the environmental cost of producing the hydrogen must also be considered. Hydrogen is also easy to burn, even in very lean, low-temperature flames, but also has much more serious explosion hazards than hydrocarbons. Also, being a very small molecule, it leaks out of tanks and through valves much more readily than other pressurized-gas fuels such as natural gas. In addition to the production and safety issues associated with hydrogen, its storage is extremely problematic. Therefore, hydrocarbons are more practical than fuels such as hydrogen or alcohols due to their far lower energy density when the weight of the energy storage media (pressure vessel, chemical solution, etc.) is included.
Many researchers have considered Direct Methanol Fuel Cells (DMFCs) for the power generation, since methanol is easily stored compared to hydrogen, but it has ~ 6 times lower energy/mass and cannot be used in pure form due to fuel cross-over anode, thus requiring substantial fuel dilution with water which decreases energy storage density still further. In addition, the efficiency of DMFCs is low due to the high permeation of methanol through the membrane, and the dynamic behavior is sluggish.
On the other hand, Solid Oxide Fuel Cells (SOFCs) use hydrocarbon fuels directly and thus do not suffer from these limitations. In particular, SOFCs also have received considerable attention due to their fuel flexibility and use of inexpensive catalytic materials. They, however, work at very high temperatures, which are difficult to maintain at small scale due to significant heat losses. Furthermore, the seals that separate the anode and cathode chambers are subject to failure as a result of thermal shock during on/off cycling. These challenges have limited the applicability of SOFCs in small scale.
MEMS Investor Journal: What are the main limitations of microscale combustion?
Professor Jeongmin Ahn: Several small-scale alternatives to batteries for electrical power generation technologies have been proposed. However, scaled-down conventional power generation devices with moving parts (e.g., internal combustion engines) experience more difficulties with heat and friction losses than their macroscale counterparts due to higher surface to volume ratios. Additionally, sealing, fabrication, and assembly are much more difficult at small scales. For example, the smallest existing combustion engine (e.g. Cox Tee Dee has a weight of 0.49 oz. and displacement of 0.01 in3) produces about 5 watts power, but its emissions and noise are unacceptable for many indoor applications.
MEMS Investor Journal: What has been your research group’s approach to microscale combustion?
Professor Jeongmin Ahn: Because of the limitations of power generation systems requiring moving parts and parasitic electrical power requirements, an ideal small-scale power generation system should have (a) means for pumping with no moving parts; (b) use fuel, not electrical power, as the “energy feedstock” for pumping; (c) produce electrical power with no moving parts and (d) not require high-precision fabrication. We also recently proposed to develop a catalytic combustion-driven thermal transpiration (thus pressurization) based pumping and power source having no moving parts and not requiring ultra high-precision fabrication and assembly. Thermal transpiration occurs in porous materials when the pore size is comparable to the mean free path of gas molecules.
Combustors coupled to thermoelectric generators (TEGs) are robust, durable, and have no moving parts for power generation but suffer from relatively low TE module efficiency, as well as the need for massive heat sinks on the cold side of the TEGs.
MEMS Investor Journal: What about other implementations of microscale combustors?
Professor Jeongmin Ahn: Applications that haven’t worked as well include the MIT and Columbia micro-turbines and the Berkeley rotary engines that have been funded by many DoD grants. Though they showed most advanced MEMS fabrication work, they basically do not work since friction and heat losses are still tremendous. Most of micro gas turbine made from silicon (very high thermal conductivity, k ≈ 120 W/m-K), and therefore, heat transfer along casting and rotor, from turbine to compressor is large. This causes large heat losses and perhaps reaction quenching in some case. It also has low manufacturing tolerances. In addition, it requires very high rotational speed (~2 million rpm) for compression (speed of sound doesn’t scale) and need larger relative chamber size as scale decreases since mixing time scales with combustor size but reaction time does not.
In addition, all of the aforementioned systems, even fuel cells that have no moving parts, require some form of pump since air must be supplied from and reaction products expelled to the ambient environment. In the case of microscale power generation, many ideas have been proposed, but most leave unresolved critical “balance of plant” issues, in particular pressurization for fuel and air. Even successful micromotors and micropumps are notoriously inefficient at converting electrical power into pumping power, meaning that a large fraction of the total system size and weight must be devoted to batteries. The difficulty of reducing the size and/or power consumption of the pressurization or vacuum systems required for a complete system in turn limits the portability and utility of miniature thermochemical systems.
MEMS Investor Journal: Which research groups are leading the development efforts in the area of microscale combustion technologies?
Professor Jeongmin Ahn: DARPA Microsystems Technology Office (MTO) supported many research groups in past 10+ years. Those research groups have been leading the development efforts in the MEMS area. MIT, UC Berkeley, Caltech, CWRU, Yale, UCLA, Carnegie Mellon are among them.
This article is a part of MEMS Investor Journal's ongoing market research project in the area of micro power generation. If you would like to receive our comprehensive market research report on this topic, please contact Dr. Mike Pinelis at firstname.lastname@example.org for more information about rates and report contents.
Copyright 2012 MEMS Investor Journal, Inc.