Environment

IBM "sunflowers" to supply off-grid energy, water, and cooling

IBM "sunflowers" to supply off...
Equipped with an array of multi-junction photovoltaic chips, each of the IBM "sunflowers" can supply the energy needs of several homes (Image: Airlight Energy/dsolar)
Equipped with an array of multi-junction photovoltaic chips, each of the IBM "sunflowers" can supply the energy needs of several homes (Image: Airlight Energy/dsolar)
View 7 Images
Equipped with an array of multi-junction photovoltaic chips, each of the IBM "sunflowers" can supply the energy needs of several homes (Image: Airlight Energy/dsolar)
1/7
Equipped with an array of multi-junction photovoltaic chips, each of the IBM "sunflowers" can supply the energy needs of several homes (Image: Airlight Energy/dsolar)
The receiver with its triple-junction PV cells (Photo: IBM)
2/7
The receiver with its triple-junction PV cells (Photo: IBM)
IBM scientist Dr. Bruno Michel holds the receiver design package (Photo: IBM)
3/7
IBM scientist Dr. Bruno Michel holds the receiver design package (Photo: IBM)
HCPVT being tested at Airlight (Photo: Airlight Energy/dsolar)
4/7
HCPVT being tested at Airlight (Photo: Airlight Energy/dsolar)
HCPVT being tested at Airlight (Photo: Airlight Energy/dsolar)
5/7
HCPVT being tested at Airlight (Photo: Airlight Energy/dsolar)
The prototype electronics and cooling system for the HCPVT (Photo: IBM)
6/7
The prototype electronics and cooling system for the HCPVT (Photo: IBM)
A schematic of the HCPVT system (Image: Airlight Energy/dsolar)
7/7
A schematic of the HCPVT system (Image: Airlight Energy/dsolar)

Looking rather like a 10-meter (33 ft) tall sunflower, IBM's High Concentration PhotoVoltaic Thermal (HCPVT) system concentrates the sun’s radiation over 2,000 times on a single point and then transforms 80 percent of that into usable energy. Using a number of liquid-cooled microchannel receivers, each equipped with an array of multi-junction photovoltaic chips, each HCPVT can produce enough power, water, and cooling to supply several homes.

Swiss-based supplier of solar power technology, Airlight Energy, has partnered with IBM Research to utilize IBM's direct wam-water cooling design (adapted from use in IBM’s SuperMUC supercomputer), water adsorption technologies, and leverage IBM’s past work with multi-chip solar receivers developed in a collaboration between IBM and the Egypt Nanotechnology Research Center, to develop and produce the system.

Using a 40-sq-m (430.5-sq-ft) parabolic dish coated with 36 plastic foil elliptic mirrors just 0.2 mm thick, the HCPVT system prototype concentrates the sun’s radiation onto a number of liquid-cooled receivers, each of which contains an array of 1-cm2 (0.39 in2) chips that each generate "up to 57 watts of electrical power when operating during a typical sunny day," combining to produce 12 kW of electrical power and 20 kW of heat.

Micro-structured conduits pump treated water around these receivers to carry away excess heat at a rate that is claimed to be 10 times more effective than passive air cooling. Although the water is still subsequently heated to around 85-90° C (183-194° F), the removal of heat from the chips keeps them at a relatively cool safe operating temperature of around 105° C (221° F). Without this cooling, the concentrated energy of the sun would see the chips reach temperatures of over 1,500° C (2,732° F).

"The direct cooling technology with very small pumping power used to cool the photovoltaic chips with water is inspired by the hierarchical branched blood supply system of the human body," said Dr. Bruno Michel, manager, advanced thermal packaging at IBM Research.

The HCPVT system can also be adapted to use the cooling system to provide drinkable water and air conditioning from the hot water output produced. Salt water is passed through the heating conduits before being run through a permeable membrane distillation system, where it is then evaporated and desalinated. To produce cool air for the home, the waste heat can be run through an adsorption chiller, which is an evaporator/condenser heat exchanger that uses water, rather than other chemicals, as the refrigerant medium.

The creators claim that this system adaptation could provide up to 40 liters (10 gallons) of drinkable water per square meter of receiver area per day, with a large, multi-dish installation theoretically able to provide enough water for an entire small town.

All of these factors, – waste energy used for distillation and air-conditioning combined with a 25 percent yield on solar power – along with the setup's sun tracking system that continuously positions the dish at the best angle throughout the day, combine to produce the claimed 80 percent energy efficiency.

Other sunflower-type heliostats exist, such as those that redirect sunlight in a residential situation or – at the other end of the scale – in industrial applications that produce many megawatts of power. The HCPVT system, however, could be considered more of a super-efficient, multi-house power plant where the claimed efficacy of the sunflower design shows its worth as a medium-scale solution to off-grid sustainable power needs.

Estimations on the operating lifetime for the HCPVT system are around 60 years with adequate maintenance, including replacing the shielding foil and the elliptic mirrors every 10 to 15 years (contingent on environmental conditions) and the PV cells, which will require replacement at the end of their operational life of approximately 25 years.

Preliminary versions of the HCPVT system will be offered with "non-optimized predecessor" (i.e. "basic") distillation and adsorption cooling systems, with optimized desalination and adsorption cooling requiring an extra two to three years to develop and bring to market.

As a gesture to bring such technology to markets unable to afford to buy their own system, Airlight and IBM will also donate a HCPVT system to two communities via competitive application. Each of the successful communities will be awarded a prototype HCPVT system from Airlight Energy and be eligible for no-cost set up and adaptation support from IBM Corporate Service Corps.

Applications from communities will be open sometime in 2015 and the winners will be announced in December 2015, with installations beginning in 2016. A fully commercial HCPVT system is expected to be launched to market sometime in 2017.

The video below shows the prototype system in action.

Sources: IBM, Airlight Energy

Update (Oct. 10, 2014): Thanks to the commenters who have pointed out some errors regarding units of measurement. These have now been corrected.

Airlight Energy and IBM Bring Solar Electricity and Heat to Remote Locations

27 comments
Slowburn
How much energy is lost if there is just a lite haze? You want solar power lets do it right. Non concentrating thermal solar collectors driving abortion heat pumps using a "steam" engine as the expander driving a generator, using the waste heat off the compressor for generating hot water for domestic or heating purposes, and the cold used for Refrigeration, air conditioning, or driving a Stirling cycle or thermal couple generator. Plus when the weather does not cooperate the system can be powered by a combustion heat source.
Windsor Wilder
"each HCPVT can produce enough power, water, and cooling to supply several homes." Except when it's cloudy which is a problem with ALL concentrating solar power systems so great if you live in Arizona or Australia but not so great if you're somewhere where it actually rains enough to grow food.
Catweazle
"each HCPVT can produce enough power, water, and cooling to supply several homes"... While the Sun is shining.
LordInsidious
Fantastic! Energy, desalinated water and cooling in one unit will really help a lot of people.
TeeWee
Now much energy will these produce between 1800 (6:00 PM) and 1000 hours? Others remarked about overcast days. How easy will it be to clean snow off the collectors. At night in the bitter cold of winter, what will prevent the cooling water from freezing? Lots of questions lots of Pie the sky.
Jeffrey A. Edwards
Slowburn, the Stirling collector was posted in Gizmag back in 2010 http://www.gizmag.com/sunpower-corp-solar-cell-efficiency-record/15536/ Here is the link to Qnergy partnering with Infinia (Stirling) http://www.qnergy.com/solar-stirling-engine- I saw a 'failure' of a concentrator fire, very enlightening!
BeWalt
It's not a system for regions with lots of annual snowfall. It's not a system for regions with haze, cloudy skies, or lots of rain. Simple as that! There are plenty of people living without power in places where this can be used. It's a really good system, engineering done right. TeeWee: freezing in the winter, are you kidding me? Ever heard of cars and antifreeze? Solar hot water heaters, used for 40 years with glycol preventing freezing? Anything? There are hot water tanks unlike the ones from your average DIY market that can keep the heat for a week, with one degree of temperature lost per night. They have been on the market for 25 years. Battery storage, or grid tie PV, never heard of? Holy moly. No pie, no sky, and ignorance is not bliss.
Mark Windsor
"Combined, the whole system produces a total of 12 kW of electrical power and 20 kW of heat over that same period." Colin, it appears that you don't fully understand the difference between energy and power.
Drum56
Colin, good article, but please don't confuse units of linear measurement with units of area. The chips are 1cm x 1cm in size, and therefore have an area of 1cm2. Also, the energy output could be a maximum of 20kW, or it could average 20kW, but it cannot TOTAL 20kW as you say. This suggests that you are also unsure of the difference between a kW (kilowatt), which is a rate of energy production / use / transfer, and a kWh (kilowatt.Hour), which is an amount of energy, whether produced / used / transferred slowly over a long time or quickly over a short time. You may think I am being overly pedantic in asking for accuracy around units of measurement, but this sort of error perpetuates the confusion the exists in the general public around the whole subject of Energy. It is irritating but understandable when these errors appear in small local newspapers; annoying when seen in major newspapers, but very disappointing when seen in a forum specifically dealing with technical subjects. Also, if my interpretation of the information in the video is correct, the foil IS the mirror surface, not some form of "shielding". The elliptical cross section of the mirror is produced by stretching metallised film across a circular housing, then pulling a slight vacuum behind it, so atmospheric pressure pushes down on the foil and curves it into the required shape. TeeWee: Are you kidding? A gifted backyard tinkerer could design and build a water cooled concentrating solar system that could cope with freezing temperatures, and add some form of energy and heat storage that made it useful outside of the "sunshine" hours of the day. Not "Pie in the sky" even for an amateur!! Do you really think a company with the technical ability and reputation of IBM would publically associate with a device that was unable to meet its designed function? Yes, any solar energy "device" will have to be associated with some form of energy storage if it is to be useful overnight, and concentrated solar collectors are only fully effective in parts of the world that see a lot of cloud-free sunshine hours, but for those parts of the world, this could be useful technology, producing good amounts of usable heat as well as electricity. Energy efficiency and renewable energy generation of various types are a NECESSARY part of our future. There have been a lot of poorly designed and even knowingly fraudulent devices which have clouded the reputation of the field, but there is a lot of genuinely useful research going on, and you should welcome any serious development work being done in those fields.
Mirmillion
This is a great convergence of technologies. Bravo. I wonder though what the real world economics will be after factoring in capital cost, maintenance and efficiency? Will home owners and cooperative power suppliers be able to recoup costs in under 5 years rather than just as the PV module have reached their end of life? We have, in North America and Europe, quasi-government utilities and private operators alike which will attempt to show that it is less risky to spend a few more dollars staying attached to their costly, and upward trending, facilities than to be independent. If economics here are such as they outweigh the "risks" I think IBM/Airlight will have a winner on their hands.