Tikalon Header Blog Logo

Temperature Cycle Energy-Harvesting

September 29, 2014

Now that digital communication, sensing, and computing devices can operate at very low power, environmental energy-harvesting is being pursued as a means to enable wireless sensors and other devices. I've written quite a few articles about environmental energy-harvesting.
• Thermogalvanic Cell, June 25, 2014
• Tidal and Wave Energy, March 24, 2014
• Energy Harvesting Cantilevers, January 14, 2013
• Triboelectric Generators, July 25, 2012
• Electret Energy Harvester, December 7, 2011
• Harvesting Radio Frequency Energy, November 22, 2011
• Cantilever Energy Harvesting, August 16, 2011
• Multiferroic Energy, July 11, 2011
• Acoustic Energy Harvesting, May 5, 2011
• Environmental Energy Harvesting, April 19, 2011
• Green Braking, October 20, 2010
• Pyroelectric Energy Harvesting, October 15, 2010
• Flutter Power, August 17, 2010
• Hot Bodies, July 22, 2010
• Rain Power, January 30, 2008

The most important environmental energy source is solar energy. Solar energy has powered all the biological processes on Earth for more than three billion years. Since photovoltaics generally convert only 20% of solar energy into electricity, direct sunlight will produce about 20 mW/cm2 of electrical power. A cloudy day will give you about 200 μW/cm2, while indoor lighting will produce just about 10 μW/cm2. Solar energy is not available at night, so you need to store some daytime energy or look to other environmental energy sources.

Our environment is populated by machines as well as people, and most of these vibrate. Noisy machines can give you about a 100 μW/cm2. Temperature cycling between day and night will produce about 10 μW/cm2, while a temperature gradient of 10°C will produce 15μW/cm2. Acoustic noise is not a good source of energy, since a 75 dB noise source, equivalent to road traffic, will produce only 0.003μW/cm2.

The human body can be a source of energy for powering medical implants and other devices. A 1996 paper by Thad Starner, who was then at the MIT Media Laboratory, estimated the harvestable energy content for various human activities, as follow:[1]
Body Heat 2.4 - 4.8 W
Blood Pressure 370 mW
Exhalation 400 mW
Breathing (Total) 830 mW
Chest Band (Breathing) 0.42 W
Arm Motion 0.33 W
finger Motion 0.76 - 2.1 mW
Footfalls 5.0 - 8.3 W

Scientists at MIT have developed a thermoelectric energy harvester that extracts up to 100 microwatts of power from the temperature difference between body temperature and room temperature.[2] A piezoelectric airflow energy harvester has produced 296 microwatts from an airflow of 8.0 meters per second,[3] and piezoelectrics have also been used to harvest energy from raindrops. The raindrop energy harvester produces about one Watt-hr per square meter per year for rainfall typical in France.[4]

Recently, electrical engineers and computer scientists at the University of Washington (Seattle, Washington) have developed a novel energy harvester for environmental temperature change. Instead of a direct temperature to electrical conversion, as in thermoelectrics, their device uses the vapor pressure change associated with temperature to expand and contract a bellows. Linear motion harvesters then convert the motion of the bellows to an electrical signal.[5-8]

The research was done by University of Washington associate professors Shwetak Patel and Joshua Smith, Sam Yisrael, an electrical engineering undergraduate student, Sidhant Gupta, a former University of Washington doctoral student, and Eric Larson, a former University of Washington doctoral student and now an assistant professor at Southern Methodist University.[6] The team presented this research at the Association for Computing Machinery's International Joint Conference on Pervasive and Ubiquitous Computing in Seattle, September 13-17, 2014.[6]

As can be seen in the figure, the bellows, which is filled with a gas with a high temperature sensitivity of its vapor pressure, expands considerably, and nearly linearly, when heated through a 31 °C temperature range. In the device, the bellows expansion activates cantilever energy-harvesters to power a wireless sensor. A temperature change of only a quarter of a degree Celsius is enough to wirelessly transmit data over a five meter range. This means that even the normal room temperature changes in a room's air conditioning can power the sensor. Outside temperature changes are generally greater.[6]

Figure caption

Bellows expansion from heating. Shown are actuations at 0°C, 15°C, and 31°C. (Still images from a University of Washington YouTube video.)[8)]

In their experiments, the research team found the following gases to be effective in the bellows actuator (Table I of ref. 5).

 GasBoiling Point (°C)Range (°C)
 Butane-0.5-8 ~ 12.2
 Acetaldehyde20.212.7 ~ 32.9
 Chloroethane12.34.8 ~ 25
 Trichlorofluoromethane23.7716.27 ~ 36.47

The device was shown to generate up to 21 millijoule of energy per cycle for a temperature change from 5-25 °C.[5] The Washington research team has filed patents on this idea, but they've placed detailed construction information on a web site.[8] Their future plans include making the device smaller, and using a combination of gases so that the device will operate in an extended temperature range.[6]

Typical applications would be water leak detectors and structural integrity monitors that would signal a central monitoring node. For a wireless water detector I designed to operate many years on a single battery, see ref. 9.[9] This research was funded by the Sloan Foundation, and the Intel Science and Technology Center for Pervasive Computing at the University of Washington.[6]

Figure caption

The bellows-powered sensor at an outdoor location.

(University of Washington Image.[6]


  1. Starner, T., “Human-Powered Wearable Computing,” IBM Systems Journal, vol. 35, no. 3&4, 1996, pp. 618-629. Also available here.
  2. David L. Chandler, "Self-powered sensors," MIT News Office, February 11, 2010.
  3. Shuguang Li and Hod Lipson, "Vertical-Stalk Flapping-Leaf Generator for Wind Energy Harvesting," Paper SMASIS2009-1276 of the Proceedings of the ASME 2009 Conference on Smart Materials, Adaptive Structures and Intelligent Systems (Oxnard, California), September 20-24, 2009.
  4. Romain Guigon, Jean-Jacques Chaillout, Thomas Jager and Ghislain Despesse, "Harvesting raindrop energy: experimental study," Smart Mater. Struct. vol. 17 (2008) 015038-9.
  5. C. Zhao, S. Yisrael, J.R. Smith, and S.N. Patel, "Powering Wireless Sensor Nodes with Ambient Temperature Changes," Proceedings of the 13th International Conference on Ubiquitous Computing (UbiComp 2014), Seattle, USA, Sep 13-17, 2014 (PDF file).
  6. Michelle Ma, "Changing temperature powers sensors in hard-to-reach places," University of Washington Press Release, September 3, 2014.
  7. Temperature Power Harvester Web Site, University of Washington.
  8. Powering Wireless Sensor Nodes with Ambient Temperature Changes, YouTube video, August 29, 2014.
  9. Build an Inexpensive Wireless Water Alarm, Circuit Cellar Magazine, February, 2014.

Permanent Link to this article

Linked Keywords: Data transmission; digital communication; sensor; sensing; computer; computing device; electric power; natural environment; environmental; energy-harvesting; wireless sensor; solar energy; biology; biological; Earth; timeline of evolution; three billion years; photovoltaic; electricity; sunlight; Watt; mW; square meter; cloud cover; cloudy day; indoor lighting; night; energy storage; machine; human; people; vibration; vibrate; diurnal temperature variation; temperature gradient; acoustic noise; decibel; dB; road traffic; human body; medical implant; scientific literature; paper; Thad Starner; MIT Media Laboratory; thermoregulation; body heat; blood pressure; exhalation; breathing; thorax; chest; Vitruvian Man; arm motion; finger; walking; footfall; scientist; Massachusetts Institute of Technology; MIT; thermoelectric; room temperature; piezoelectricity; piezoelectric; airflow; meters per second; rain; raindrop; France; electrical engineering; electrical engineer; computer scientist; University of Washington (Seattle, Washington); vapor pressure; bellows; associate professor; Shwetak Patel; Joshua Smith; undergraduate education; undergraduate student; Sidhant Gupta; doctoral student; Eric Larson; assistant professor; Southern Methodist University; research; Association for Computing Machinery; International Joint Conference on Pervasive and Ubiquitous Computing; Seattle; gas; linear function; linearly; Celsius; °C; cantilever; transmitter; transmit; meter; HVAC; air conditioning; YouTube video; experiment; boiling point; butane; acetaldehyde; chloroethane; trichlorofluoromethane; joule; millijoule; patent; water; structural integrity; node; battery; Alfred P. Sloan Foundation; Intel Science and Technology Center for Pervasive Computing.