Part 1
Self-powered sensors

It can be inconvenient to replace batteries in devices that need to work over long periods of time. Doctors might have to get beneath a patient’s skin to replace batteries for implanted biomedical monitoring or treatment systems. Batteries used in devices that monitor machinery, infrastructure or industrial installations may be crammed into hard-to-reach nooks or distributed over wide areas that are often difficult to access.

But new technology being developed by MIT researchers could make such replacements unnecessary.

Soon, such devices could be powered just by differences in temperature between the body (or another warm object) and the surrounding air, eliminating or reducing the need for a battery. They would use new energy-scavenging systems being developed by Anantha Chandrakasan, MIT’s Joseph F. and Nancy P. Keithley Professor of Electrical Engineering and director of the MIT Microsystems Technology Laboratories, and Yogesh Ramadass SM ’06, PhD ’09.

Such a system, for example, could enable 24-hour-a-day monitoring of heart rate, blood sugar or other biomedical data, through a simple device worn on a patient’s arm or a leg and powered by the body’s temperature (which, except on a 98.6-degree F summer day, would always be different from the surrounding air). A similarly powered system could monitor the warm exhaust gases in the flues of a chemical plant, or air quality in the ducts of a heating and ventilation system.

The concept of harvesting energy from differences in temperature is nothing new. Many technologies for doing so have been developed, including devices NASA has used to power probes sent into deep space (the probes harvest heat from radioactive plutonium). Certain semiconductor materials, by their nature, will produce a flow of electrical current when one side is hotter than the other — or, conversely, will produce a difference in temperature when a current is run through them. Such materials are already used for solid-state coolers and heaters for food or beverages.

The principle was discovered in the 19th century, but only in recent years has it been seriously explored as an energy source. In thermoelectric materials, as soon as there is a temperature difference, heat begins to flow from the hotter to the cooler side. In the process, at the atomic scale this heat flow propels charge carriers (known as electrons or electron holes) to migrate in the same direction, producing an electric current — and a voltage difference between the two sides.

The key to making this principle practical for low-powered devices is to harness as much as possible of the available energy. Chandrakasan and Ramadass have been working to get as close as possible to the theoretical limits of efficiency in tapping this heat energy.

The higher the temperature difference, the greater the potential for producing power, and most such power-generating devices are designed to exploit differences of tens to hundreds of degrees C. The unique aspect of the new MIT-developed devices is their ability to harness differences of just one or two degrees, producing tiny (about 100 microwatts) but nevertheless usable amounts of electric power. The key to the new technology is a control circuit that optimizes the match between the energy output from the thermoelectric material and the storage system connected to it, in this case a storage capacitor. The findings were presented this week at the International Solid State Circuits Conference in San Francisco.

Because thermoelectric systems rely on a difference in temperature between one side of the device and the other, they are not usable for implanted medical devices, where they would be in a uniform-temperature environment. The present experimental versions of the device require a metal heat-sink worn on an arm or leg, exposed to the ambient air. “There’s work to be done on miniaturizing the whole system,” Ramadass says. This might be accomplished by combining and simplifying the electronics and by improving airflow over the heat sink.

Ramadass says that as a result of research over the last decade, the power consumption of various electronic sensors, processors and communications devices has been greatly reduced, making it possible to power such devices from very low-power energy harvesting systems such as this wearable thermoelectric system.

David Lamb, chief operating officer of Camgian Microsystems, a company that produces a variety of low-power, lightweight semiconductor chips, says that “we believe the wireless sensor products we are developing will all migrate to energy harvesting, as we push their size, weight and power down.” He adds that the research of Chandrakasan and Ramadass “is in the critical path of technologies required by companies such as Camgian that are developing next-generation microsystems.”

Devices to use this power would in most cases still need some energy storage system, so that the constant slow trickle of energy could be accumulated and used in short bursts, for example to operate a transmitter to send data readings back to a receiver. Different ways of storing the energy are possible, such as the use of ultracapacitors, Ramadass says. “These will play a critical role, in order to build a complete energy harvesting system,” he says.

After years of work on these highly efficient energy-scavenging devices, currently funded by a seed grant from the MIT Energy Initiative, Chandrakasan says, “the time has come to find the real applications and realize the vision.”

From MIT News, February 16, 2010

The Trans-Alaska Pipeline System, which traverses hundreds of miles of some of the most inhospitable terrain on Earth, must be monitored almost constantly for potential problems like corrosion or cracking. Humans do some of this work — surveying the pipeline from the air and inspecting it more closely in the areas that can be easily accessed by roads — but the bulk of it is done by mechanical “pigs,” sensor-laden robots that travel inside the pipeline looking for flaws.

A simpler process might involve outfitting remote stretches of the pipeline with sensors that would automatically radio a warning of impending problems. But the need to periodically change the batteries on such sensors lessens the appeal of that option. For electronic devices in remote or inaccessible situations like this, including environmental or mechanical monitoring sensors as well as some kinds of biomedical monitors, it can be inconvenient or even impossible to replace batteries.

But what if batteries weren’t necessary?

Systems that could provide power for such sensors just by harvesting the normal vibrations of the pipeline (or bridges or industrial machinery and so on), eliminating or reducing the need for a battery, are being developed by Anantha Chandrakasan, MIT’s Joseph F. and Nancy P. Keithley professor of electrical engineering and director of the MIT Microsystems Technology Laboratories, and his former student Yogesh Ramadass SM ’06, PhD ’09.

They have been working for years on the development of ways to harness small amounts of power from ambient vibrations. A paper describing their latest work on a new control circuit for such systems, which can quadruple the amount of power they produce, appeared last month in the IEEE Journal of Solid-State Circuits.

Big steps toward tiny power

There are a number of different approaches to harnessing vibrational energy, some using magnetic or electric fields. But the new control circuit Ramadass and Chandrakasan developed is designed to work with piezoelectric systems — ones that use voltage generated by stress in a crystalline material, such as lead-zirconate-titanate.

It has been known for well over a century that some materials, including some crystals and ceramics, will produce an electrical current when subjected to stress by squeezing or bending. To harness the energy of motion or vibration, such a material is coupled to a spring, pendulum or other mechanism that converts the motion into pressure.

Chandrakasan and Ramadass envision applications in such things as implantable medical diagnostic or treatment devices that could be powered indefinitely by the person’s own natural movements, or distributed sensors to monitor structural elements on bridges or the pressure in truck tires and transmit the data to a central receiver, powered by the vibrations of ordinary traffic.

Existing devices for harvesting energy from vibrations tend to be tuned to very specific frequencies, Chandrakasan says, but “in many practical applications, we need something more general. That’s still a technical question to be addressed.”

For now, such systems can’t deliver enough power to run consumer devices such as cell phones, Ramadass explains. “The power levels for a cell phone are way up from what we can generate now” from a person’s natural movements, he says, although some simpler devices, such as an mp3 music player, might be within the available range. He is currently working with semiconductor leader Texas Instruments to develop commercial applications of ultra-low power systems and solutions.

David Lamb, chief operating officer of Camgian Microsystems, a company that produces a variety of low-power, lightweight semiconductor chips, says enabling new, low-power distributed sensor and security systems will depend on improving the efficiency of energy-harvesting techniques, including the power-producing system as well as control and storage systems. Because low-power systems are still a relatively new area of research, he says, “typical power management approaches are not well suited to energy harvesters, and there are still a lot of unsolved challenges,” But devices such as the company’s remote surveillance system are designed to operate on very low power, he says, and “if efficient interface and control circuits can be developed, this microsystem can be continuously powered by energy harvesting.”

The U.S. Defense Advanced Research Projects Agency (DARPA) has provided support for this research, which also holds promise for monitoring military equipment in remote locations.

The team has also been developing systems to derive small amounts of power from temperature differences (as described in part one of this series), and Chandrakasan says that in the future, some applications might make use of systems that combine both the heat- and vibration-harvesting devices to produce more power, or to work in situations where these energy sources are variable and one or the other might not always be available.

Some parts of such a system, such as the electronic control circuits and transmitters for relaying the collected data, could be connected to both the heat and vibration generating systems (as well as additional sources of power, such as a solar cell), Ramadass says. “You could have one set of electronics that interfaces” with multiple inputs, he says.

For the future, the researchers are working on ways to improve the integration of the various components, and on making the systems as versatile as possible. “We want to make them adaptable over a broad range” of operating conditions, Chandrakasan says. In addition, they are working on improving the devices’ overall efficiency. “We want to get to the maximum theoretically possible achievable energy,” Ramadass says.