Concrete spheres could deliver feasible energy storage for offshore wind turbines


May 1, 2013

A new system being developed at MIT would store excess energy in concrete spheres on the sea floor

A new system being developed at MIT would store excess energy in concrete spheres on the sea floor

The intermittent nature of wind and solar power generation is one of the biggest challenges facing these renewable energy sources. But this isn’t likely to remain a problem for much longer with everything from flywheels to liquid air systems being developed to provide a cheaper form of energy storage than batteries for times when the wind is blowing or the sun isn’t shining. A new concept out of MIT can now be added to the the list of potential solutions. Aimed specifically at offshore wind turbines, the concept would see energy stored in huge concrete spheres that would sit on the seafloor and also function as anchors for the turbines.

The MIT concept works by using excess energy generated by the wind turbines to pump seawater from a hollow concrete sphere sitting on the seafloor that measures 30 meters (98 ft) in diameter. Then, when the wind dies down and power is needed, a valve is opened to let the water back into the sphere through a turbine that drives a generator to produce electricity.

The MIT researchers say that such a sphere positioned in 400-meter (1,312 ft) deep water could store up to 6 MWh of power, meaning that 1,000 spheres could supply as much power as a nuclear power plant for several hours. They claim this is enough to transform offshore wind turbines into a reliable alternative to conventional on-shore coal or nuclear plants.

Additionally, since the system would be connected to the grid, the spheres could also be used to store energy generated from other sources, such as on-shore wind and solar, or from base load power plants that are most efficient when operating at steady levels. Such a system could reduce the reliance on the generally less efficient peaking power plants that kick in when there is a high electricity demand that base load plants can’t meet.

The spheres with their 3-meter thick concrete walls would weigh thousands of tons each, which would also make them suitable to anchor the wind turbines in place. However, because there is currently no vessel with the capacity to deploy a load of their size and weight, a specially built barge would need to be constructed to tow them out to sea after being cast on land.

This contributes to the preliminary cost estimates of about US$12 million to build and deploy one sphere, with costs gradually declining from there. The team estimates the technology could yield storage costs of around six cents per kilowatt-hour, which is considered viable by the utility industry.

While the team’s analysis indicates the technology would be economically feasible at depths as shallow as 200 m, with costs per megawatt hour of storage dropping as depth increases to 1,500 m before rising again, 750 m is seen as the optimal depth for the spheres. However, Brian Hodder, a researcher at the MIT Energy Initiative, says as costs are reduced over time, the system could become cost-effective in shallower water.

Alexander Slocum, the Pappalardo Professor of Mechanical Engineering at MIT, and his students built a prototype 30-inch (76 cm) diameter sphere in 2011 to demonstrate the feasibility of the system. The team now hopes to scale testing up to a 3-meter sphere and then, if funding becomes available, a 10-meter version that would be tested in an undersea environment.

They estimate an offshore wind farm using the technology could supply an amount of power comparable to the Hoover Dam, while using a similar amount of concrete. The team says that some of the concrete for the spheres could be made using fly ash from existing coal plants to cut the amount of carbon dioxide emissions resulting from production.

The MIT team has filed a patent for the system, which is detailed in a paper published in IEEE Transactions.

Source: MIT

About the Author
Darren Quick Darren's love of technology started in primary school with a Nintendo Game & Watch Donkey Kong (still functioning) and a Commodore VIC 20 computer (not still functioning). In high school he upgraded to a 286 PC, and he's been following Moore's law ever since. This love of technology continued through a number of university courses and crappy jobs until 2008, when his interests found a home at Gizmag. All articles by Darren Quick

@ Nigel Giddings - There are plenty of seaborne wind turbine designs that can float on the surface. Look up the FLIP research vessel to see equivalent technology or 'floating wind turbines'. Once in place, the turbines are tethered securely in place to the seabed. But your point about storage is absolutely correct. Once we solve this power storage problem, and make it portable, we will start to see the technologies we all dream of =).


A rather "heavy-weight" a solution ...

I still think the energy bags as proposed by Seamus Garvey (Nottingham University) are a cheap and elegant - thus lightweight - solution.

Admittedly you need some water depth, but at some places that's not a problem at all. Nice background article:

Jan Vegt

Interested to see the suggestion of using these as anchors for wind turbines, surely if they are hundreds of metres underwater it wouldn't make sense to use them as a form of foundation (anchor) for a turbine as it would need hundreds of meters of structure to get the turbine above the water line?

Storage of energy appears to be the biggest issue for renewables, especially wind based systems, but being able to store a 'few hours' doesn't deal with the extended periods needed for when the wind isn't blowing. I'm not sure about off-shore wind, my knowledge is based more on the variations on-shore but you can see the variations in UK wind produced energy here which highlights the fluctuations we currently see.

I'm trying to be positive but there is still a long way to go to match the on-demand system that we have today. Maybe we should be looking at people using power when it is available and deal with the issues that creates...

Nigel Giddings

I wonder if they considered using a low-density object that could store buoyancy energy when pulled under the surface of the water. The density of concrete is ~2320 kg/m3 and the density of water is ~1000 kg/m3. So, for a 1 meter cube, the downward force, ignoring density changes for water at increased depths, is not much different than the upward force of a void at the same depth. Given that it's less resource intensive to build a void than it is to build a solid, I think they should consider something along those lines.


Looking at various reports and studies, we have a lot of information on the intermittent of wind power. Its pretty site specific, as some places are just more windy then others. In general for the united states it appears that wind turbines run at full power about 25-30% of the time. I didn't look deeper to see how much of that 75% was complete dead wind.

The thing is the spheres don't have to make up for that entire 75%. They will be needed only when demand is high and wind is low. Then you switch on the spheres, which have a known power period.

We already have backup systems in place, with gas turbines. Unfortunately since the power has to be ramped up at a moments notice we have to use open cycle turbines, which are far less efficient then the close cycle versions. Closed cycle can be brought up to 2/3rd power in about 1 hour. So if you have a few hours worth of sphere power online, it gives you time to start up your super efficient backup systems in an organized manner.

Also the system can be expanded as you go to allow for long and longer sphere cycles, using the backup generators less.


Interesting idea but unless it can be done at about the same cost as conventional anchors it's a very expensive option. Since internal pressures need to be the same for the air to displace the water anyway the storage can be a lightweight structure anchored to the seabed and doesn't even need to be at the turbine site. That would permit it to be situated in deeper water and thus have a higher energy density which means for the same storage capacity you can have a smaller less costly unit.

Max Kennedy

A simpler solution that I had been thinking for years would be to pump water onshore at a higher elevation. Low-cost of construction, low cost of maintenance, already developed technology, ideal in any environment and even it can help in harnessing tidal energy.

Hassan Syed

What a waste of time materials and effort. Alternative energy research will never be what it should be until government investments are driven into low cost low maintenance consumer based systems that individually feed back into a rebuilt grid. The wind is always blowing somewhere and the sun is always shining somewhere in the USA for at least 8 hours. 200 million home based units could power all of the USA and probably Canada and Mexico at the same time

Robert Moynihan

There's lots of wild conjecture about intermittent supply, but this is a half-decent study:

It's no surprise that the straightforward solution to intermittency is scale, but most of the fossil fuel brigade try to make the case that if it's not windy in one place then it's not windy anywhere; this study shows that that assumption is wrong.


There are at least TWO existing alternative approaches to making these spheres or any similar really large underwater structure. First, electro-deposition can, and has been used to build large underwater structures. A steel wire and rebar structure defines a skeleton through with a current is run which causes mineral ions, ( calcium & magnesium mostly), in seawater to electro-deposit onto the wire form. These can be made arbitrarily large and thick at the intended depth, literally, deposited/built in place. Alternatively this could be an ideal use for a team of robots to place concrete on a form. The form could be the same kind vinyl sheet used for swimming pools only far larger and inflated with sea water to establish the desired shape. Think something on a scale of a blimp only filled with seawater. Concrete is continuously mixed and placed by many small bots centrally controlled so that the leading, or "wet" edge of concrete is always being added to with more concrete. The objective is wind up with a single monolithic seamless structure cast in place. Third, both of these approaches could be combined in each sphere to gain the best of both construction techniques. This would side step having to build and then cart around an object significantly the size of a cruise ship or oil rig. Remember how well that worked from the news about Carnival cruises many big dead buoy ships and also several oil companies such as the oil rig that broke free in the arctic. Jockeying stuff that large against ocean currents provably has NOT worked well.


I think that more emphasis should be placed upon improvement of the grid system so that power could be moved from one area to another without the need of too much storage. We have many alternative energy sources and more will be forthcoming such as with wave or current (Gulf of Mexico) energy conversion

Adrian Akau

It will also depending on the local undersea topology. You may have to be 30 to 50 miles offshore to get to those depths. Placing turbines that far out is less desirable to start with due to line losses back to shore, maintenance, etc.


Ocean depth of 2460 feet, 750 m = 1107 psi.

On land you could use dead weight that equals 2460 feet, 750 m water to obtain the 1107 psi & save US$12 million.

Flipider Comm

I like the idea, I also like the deep water gasbag energy storage idea. having the windmills pump water (or air) and generate all the electricity from stored would remove the spikes and dips in the current flow introduced by vagaries of the wind. For pneumatic storage power Stirling cycle engines with the heat and cold generated by the compression and expansion of the gas.

re; Nigel Giddings

Given the claim that for an equivalent mass of concrete they can store as much energy as Hover Dam. I would say that they are talking about more than a few hours of storage.

re; StWils

If Judging by news events nothing is safe. Uneventful flight, barges and loads not breaking free, commuters driving home safely never make the news nor should they because they are not news.

re; Synchro, Adrian Akau

How much theoretical generating capacity do we need to build per megawatt of on demand electricity?


New Ideas aren't always Bad...

As someone has stated, you don't need an immensely strong shell to hold air under water, as the whole system will be open to water pressure a bag would do...

@Slowburn... Pumping water to the bottom of the ocean will not result in any net energy...The system only works with air because it is significantly lighter than water.. The pressure at depth is solely related to the density of water. And as the mass of air is "negligible" (ok 1.2 kg/m^3) the pressure at depth results in almost the same pressure at the surface to recover the energy through a turbine etc...

The idea of lowering buoyant thing sin the ocean will run into problems when the ocean crushes them, yes for that one they would need immensely string (and heavy) structures....

For people considering pumping Sea water 700m in the air... have a talk to the envorinmentalists on salinity, ground water contamination issues etc.... WE could do it if there was a net return bu pumping river water back up, and that is exactly what pumped hydroelectric storage is... Also in many countries there aren't a lot of sites suitable for that.... While most countries Do have an ocean nearby..

For real power storage, most people have no idea of the amount of water one would have to displace Per Consumer....

These structures (accumulators) will end up being very big and heavy, because they have to displace an immense amount of water at depth.... also, though many countries are near the sea, finding a trench 700m deep isn't that likely on most of the continental shelf (would work well in Tahiti or Hawaii).. As the distance offshore increases (as already mentioned), the potential financial returns on system reduces and the cost of maintenance will increase exponentially.... Any idea of how big the waves can get on the high seas.. What you (we) see at the beach is nothing... Those waves can really smash a wing turbine about.. and the storm winds that accompany them will have the turbines shut down often...

The problem for wind turbines is that they have to be in a windy area, but not too windy..... its a goldilocks position they have to find. (oh and not cause navigational hazards.) lol.


The solution has always been their! Use the wind to get hydrogen from the water and then release the O2 back into the water to promote life. Now use the hydrogen day or night as your power source.


Hassan & others,

Energy is already stored in ponds on tall hills with the turbines that pump water up later being used to generate energy when the water is run out. Probably the best example on North America is at Niagara Falls. There is a huge pond into which water is run after midnight when the falls are effectively "turned off" by diverting water to the impondment area. During the day and early evening the falls roar along for the valuable tourists and water is discretely drained through the turbines to provide electricity. In Niagara Falls NEW YORK there is an excellent visitor pavilion that shows this. However the best view and nicest parks are across the gorge is Niagara Falls ONTARIO. The issue is geography. No one storage or generating approach is universally appropriate everywhere. Pretty much all the choices are needed to build a robust, stable, and modern electrical grid. Storing energy offshore in submerged spheres as compressed air is a great idea if the spheres can be built in place, and sturdy enough, at a cost that works. And even then this may be an idea that only works economically enough in some places.


re; MD

The concrete spheres are to hold water out creating a huge pressure differential equal to a 400m drop so that large quantities of electricity can be generated on demand. To get the water back out it must be pumped given the difficulties of sucking water from above I could see pumping warm surface water down to provide hydraulic power to the evacuation pump. Since this generates a temperature differential there is no reason not to do a little OTEC Stirling cycle should work quite well under those conditions.

How did you get 700m read the article and add 75%? Not that deeper is not better. At 400m all three coasts of mainland USofA look fairly reasonable for it.

About open ocean waves. Do you think all engineers are idiots or just those that work around wind energy?

I am not a fan of wind energy because of its intermittent nature including the spikes and dips in current that makes integrating it into the grid expensive. But you seam to have a real problem.

re; donwine

Using platinum as the catalysis it take 1.8 kw of electricity to generate 1 kw worth of hydrogen they you have the difficulties of dealing with hydrogen and the energy loss of converting it back to electricity leaving an efficiency of less than 40% hydraulic and pneumatic energy storage are around 90% efficient. If you are after an demand electricity hydrogen generation is a stupid way of going about it.


Slowburn What are your percentage numbers when the wind stops and the sun light ends for the day?


re; donwine

That is the point of energy storage if the sun always shined, and the wind always blew you wouldn't need it. However keeping generators that are driven by gas turbines, diesel, and such running at maximum efficiency or completely off doesn't qualify as need energy storage is highly desirable.


Sorry but from what I understand from reading this we would pump extremely high pressure water out of a large cement sphere???? Then when the energy is needed we use the stored energy to turn a water turbine? Based on my understanding of pumps and water turbines I would think that half of the energy would be lost. I could be completely wrong, perhaps there is a more efficient way to move water and recoup the energy. Seems to me the logical and far more simple solution is to use the depth of the water to suspend a large mass which would be lifted and lowered to store and generate electricity. Mechanical energy storage is far simpler, cheaper, efficient and more dependable. Would you rather service a hydraulic pump and generator on the bottom of the sea, or a simple mechanical or electric motor at sea level? anytime you do anything at the bottom of the ocean you are asking for serious cost and or suffering. People certainly were meant to be at the bottom of the ocean.

Michael Mantion

re; Michael Mantion

Your float system stores only a small fraction of the energy that the Concrete spheres do because it does not benefit from the increasing pressure that comes with increasing depth.

If you were to force the water out with compressed air you would then have a vast supply of high pressure air to generate electricity with before you begin introducing water into the chambers. And either way you have the thermal differential to power a Stirling cycle engine. Also adding a basic solar-thermal collector to the side of the tower would be childishly simple providing even more energy.


re; Justfly25 part 2

Framing and pouring concrete in any shape can not be considered a new industry and neither can anchoring things to the ocean bottom. More importantly doing both is far easier than boring large diameter tubes in rock.


Boston being an seaport it is understandable that MIT did not mention that the are fresh water lakes deep enough to use this as well.


I'm far from an expert here but it seems to me that these concrete balls would want to float really, really bad. How about modifying off shore drilling technology to drill large diameter tubes into the sea floor? This anchors the tubes naturally and solves any depth issues at the same time. Maybe you hit oil in the process? Maybe some kind of geothermal heat exchange is opened up a little farther too? I see this solution being a lot more stable than trying to anchor what equates to massive ships under the ocean and keep them down and in place. Off shore drilling isn't cheap but I'm pretty sure it would be cheaper than developing a completely new industry centered around building and anchoring concrete balls.


re; Justfly25

According to my calculations the 3m concrete shell weighs more than the total amount of water displaced buy the sphere. However a 30m sphere is about the limit of this so if you were to make a bigger sphere you would have to add ballast to make it sink. This is not all bad. If you were to cast larger hollow spheres you could float them to their destination without a barge (just a tugboat) and refloat them for maintenance.


I'm not convinced that this concept uses compressed air to force water out, as Max Kennedy and Slowburn assume. There isn't enough detail in the source article, but their statement seems to say the water is pumped out, not forced out. My assumption is there's air at sea level pressure in the sphere after the water is pumped out. That would be consistent with their statement of using a large hydrostatic pressure differential and explains why they need the ten-foot-thick sphere walls to maintain structural integrity. Deep ocean water at 75 atmospheres of pressure rushing into a vessel with only one atmosphere of internal pressure would explain the large amount of available power storage.


re; Gadgeteer

If you force the water out using compressed air and then run the compressed air out to atmospheric pressure through a energy harvesting device of your choice before using the hydraulic system you have stored more energy. Be sure to use a thermal differential harvesting device on both the heat generated by the compressing the air and the cold by the expansion.


I will defer to the minds at MIT, who say "pump the water out," not "force it out." I'd like to believe they've actually done the math and worked out more details than commenters have.


re; Gadgeteer

Just because the engineers that thought of it didn't think of it, didn't want to confuse the reporters, or didn't think that at their maximum depth of 750m the 7640.19kPa (1108.12psi) worth of energy was worth tapping. You are going ignore the energy storage potential. Incidentally the the pressure stored could be higher but that would have the concrete under tension.


It seems to make more sense to create a vacuum than pump air into the sphere since this would make it easier to fill the vessel back up with seawater. Additionally, the creation of vacuum would seem to avoid issues with cavitation on the pump blades.

To address the issue of maintaining the pumps at those depths, I would have a primary and then secondary pumps available down there, easily detached, swapped and floated/sunk between the surface and cylinder location via tether.

The idea of drilling into the seafloor for storage makes sense if you are primarily doing it to offset oil, geothermal, and mining exploration costs. Our tax dollars already subsidize such activities, and this could result in generating a return on failed taps, spent wells, etc...

One more thing I want to add...... Would it be more beneficial to have extremely brackish water or distilled water void of oxygen pumped in and out to minimize biofouling? Is this a potential option?


re; GRich

After you have vented the air to sea level atmospheric pressure you have about a 1090psi pressure difference between the inside and outside of the concrete sphere.


I understand that air compressors and air motors (or turbines) are horribly inefficient – primarily due to the adiabatic heating and cooling (respectively) in the two processes. A typical air powered motor might use four times as much electricity (via the compressor) as would an electric motor with the same output. With various stratagems such as intercooling or oil injection this can be reduced but not eliminated. Hydraulics do not suffer the same losses – efficiencies close to 100 % being normal. Thus, it seems clear that the MIT proposal is just an “upside down version” of a “simple” pumped storage system with the air having no more effect than it would when water is pumped up a hill in a conventional pumped storage system.

Lindsey Roke

re; Lindsey Roke

The point of the system is that it can be tough to find a several hundred meter tall hill, especially in Florida and other coastal planes. Increasing the speed at which you disperse the waste heat does very little to improve the efficiency of compressing air. Knowing that the heat is going to be generated and attaching it to an appliance that needs heat in its operation is efficient. Just as using the cold produced by the expanding gas boosts total efficiency. Also storing energy as compressed gas is more efficient than in electric batteries especially when you consider the cost and life of the storage medium. Putting compressed air into the concrete containers just adds energy storage capacity.


What's missing from the discussion is the capital cost per kWh of energy stored, and any analysis of the size of the generator system that is activated when the water is allowed back into the shell.

My guess is that capital cost would be more than $1000 per kWh of capacity because of the challenges of the location, whereas compressed air startup Lightsail Energy is targeting the $200-$500 per kWh, for large tubes in 40-foot containers, and they have made improvement in air-pumping efficiency by conquering the heat loss problem.

A new kind of battery will soon bring storage capital cost below $200 per kWh capacity, and last for over 25,000 discharge cycles. This may solve the renewable energy grid penetration problem once and for all.

Mark Roest
Post a Comment

Login with your Gizmag account:

Related Articles
Looking for something? Search our articles