The Energy Collective Group

This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.

9,844 Members

Article Post

OTEC and Energy Innovation: The Willie Sutton Approach

ocean thermal energy conversionWillie Sutton is purported to have claimed he robbed banks “because that’s where the money is.”

Whether or not those were his words; it is the right approach to energy.  

Robert Stewart of Texas A&M points out in Oceanography in the 21st Century,”18 times more heat has been stored in the ocean since the mid 1950s due to global warming than has been stored in the atmosphere.

When a liquid is heated it expands and because its molecules move apart it becomes less dense. As a consequence the oceans are becoming thermally stratified which negatively impacts phytoplankton that are the base of the ocean food chain and the lungs of the planet. They absorb more atmospheric carbon dioxide than the world’s forests.

The Nature article, “Global phytoplankton decline over the past century” by Daniel G. Boyce of Dalhousie postulates the volume of phytoplankton in the world’s oceans, which produce half of the oxygen in the atmosphere by consuming the equivalent amount of carbon dioxide, has been declining steadily for the past half century-down about 40 percent since 1950.

“What we think is happening is that the oceans are becoming more stratified as the water warms,” said Boyce. “The plants need sunlight from above and nutrients from below; and as it becomes more stratified, that limits the availability of nutrients.”

Expanding oceans have no place to go but up onto the land and warming oceans and air melt the polar icecaps which exacerbates the sea level problem.

Contrary to the IEA’s recent report that we have five years to prevent “dangerous” climate change, a Canadian Centre for Climate Modelling and Analysis study concluded that even if we stopped putting CO2 into the atmosphere today the seas may rise by at least four metres, over the next 1,000 years.

The insurance company Allianz has estimated that $28 trillion worth of infrastructure will be at risk by as early as 2050 and the outlook for Small Island States is bleak.

Increasing evaporation is another consequence of warming oceans and the conventional wisdom has been this moisture produces cloud cover and an albdedo effect that will produce ocean cooling.

A recent study however indicates that this may in fact be wrong and instead warming oceans transfer heat to the overlying atmosphere, thinning out the low-lying clouds to let in more sunlight that further warms the ocean. This feedback warms both the air and water and may lead to thermal runaway and catastrophe.

As Will Rogers was wise enough to note:


Unfortunately, many of the actions proposed in response to global warming are likely to do more harm than good.

Fusion has been referred to as the holy grail of energy because it mimics the sun which is the source of virtually all of our power but for fission. The problem is like fission; fusion boils water to produce electricity which is a process that is only about 33 percent efficient.

Richard Smalley, Nobel Laureate in Chemistry, estimated a population of 10 billion by the year 2050 will require as much as 60 terawatts to meet its needs, including massive desalination.

To produce this 60 terawatts with either fission or fusion an additional 120 terawatts of waste heat would be produced, most of which would end up in the ocean, exacerbating thermal expansion and accelerating the collapse of the West Antarctic ice sheet.

Solar panels, wind and hydro do not produce waste heat but neither do they remedy sea level rise, thermal runaway or our dying oceans.

Only one energy source, Ocean Thermal Energy Conversion (OTEC) converts accumulating ocean heat to energy, produces renewable energy 24/7, eliminates carbon emissions, and increases carbon dioxide absorption (cooler water absorbs more CO2).

A NASA study recently published in Nature determined the average amount of energy the ocean absorbed each year over the period 1993 to 2008 was enough to power nearly 500 100-watt light bulbs for each of the roughly 6.7 billion people on the planet.

This 330 terawatts is about 20 times the total amount of primary energy consumed globally every year.

It must be noted that even though the ocean is accumulating more solar energy than we can use, it is the cold denser water available due to the thermohaline circulation that makes conversion of this heat to electrical or mechanical energy possible.

Conventional OTEC would be so effective cooling the ocean, one of its major drawbacks is the potential to overturn the the thermohaline circulation which is vital to the maintenance of the deep water heat sink required to produce energy by this method.

As Dr. Paul Curto, former NASA Chief Technologist, puts it in his Op Ed American Energy Policy V — Ocean Thermal Energy Conversion, last year “The size of the heat sink represented by the “cold ocean mass” in the tropics needs to be more than roughly 300 times or larger resource than that of the OTEC power generation over a year so that OTEC may become a third order effect. If we estimate the total volume of water below 500m depth in the tropical oceans, roughly 500 million cubic kilometers, 5e17 cubic meters, or 5e20 liters, we arrive at an estimated 5e20 joules per degree Celsius differential in the heat sink, or 139e6 TWh, over 317 times that from 2.5 TWe of OTEC each year. The efficiency of OTEC conversion is proportional to the temperature difference (dT) between the surface layer and the mean temperature of the heat sink (~3C). If we assume very large OTEC utilization, say 2.5 TWe as shown, with an average dT of 20C, the average efficiency is roughly 70% of the Carnot efficiency (taking into account parasitic losses), or 4.73%. The amount of heat dumped by that much OTEC into the ocean’s heat sink at depth is therefore just over 50 TWth, and that is also equal to the heat removed from the surface plus the power output, about 53 TW. The heat sink is replenished by cold arctic and antarctic waters sinking to the bottom at the poles. The reradiation from the world’s oceans should also be enhanced by the elevated temperatures due to global warming, but the amount of water sinking to the bottom will likely remain in balance. In other words, as long as the heat sink is replenished by the arctic currents at near to or the same as is done today, the added heat from OTEC will not measurably impact the thermocline for centuries or longer, after which OTEC’s cooling effect on the ocean may enhance the replenishment of cold water at the poles. The surface layers of the ocean have relatively small volume, three orders of magnitude less, compared to that of the heat sink at depth. Therefore, OTEC’s impact on reducing the surface water temperature over time will be much larger, on the order of one degree F per decade at this power level.”

The North Atlantic thermohaline circulation is responsible for much of the total oceanic heat transport towards the north pole, peaking at about 1.2 + 0.3 Peta Watts (1015 Watts) at 24oN latitude.

To produce Smalley’s 60TW with conventional OTEC you would therefore dump 60TW*20 or 1.2 Peta Watts of heat to the depths and remove the same from the surface which would overturn the thermohaline.

GWMM OTEC uses a heat pipe to take exhausted vapors from a turbine to the depths for condensation, instead of using massive and expensive cold water pipes to bring water to the surface, and a counter-current heat transfer system to recirculate the latent heat of condensation back to the surface rather than dumping the heat to the depths. This solves OTEC’s problems of cost, limited potential, efficiency and reduces the environmental impacts on the thermohaline and aquatic life.

To produce 60 TW with this approach you would extract 120 TW from the surface and ideally dump 60TW worth of heat to the depths or about the same as you would to produce 2.5 TW with the conventional approach. (A large hurricane extracts 50 or more terrawatts of heat from the ocean’s surface and on average there are 21category 3 or greater storms around the globe each year plus many smaller storms.)

They do not impact the Thermohaline because most of the heat is returned to the surface in falling rain, which is the same principle GWMM OTEC seeks to employ.

In the process of creating all of the renewable energy mankind needs, you simultaneously draw down the fuel hurricanes thrive on as well as the cause of thermal expansion and prevent the potential for thermal runaway and mass extinctions.

When it comes to energy Mr. Sutton and Mr. Rogers had it right.

Content Discussion

Rob Freda's picture
Rob Freda on May 15, 2013

great article, especially highlighting the shear amount of energy contain in earth’s water and gases.  one minor thing.  any thermal system will reduce the heat in the atmosphere so wind and solar will do the same thing.  Wind will do it at a rate of around 40%.  Not sure what the net on solar is.  

One other thing.  The GWMM OTEC site says that the production can occur within 5000km of the energy markets.  That type of distance is the self-same reason why the Sahara will never be Europe’s energy “basket” and why deepwater wind is not viable or mid-west wind will never be shipped to the northeast be it via HVDC or HVAC.  Absorbing that type of transmission cost on top of the marinized generation and install costs is not economically feasible.  Effective thermosyphon’s (which is what I think this tech is) also generally use some operating fluids that you would not want to risk putting in the ocean. 

nice thought though.  certainly does a great job of elucidating the problems with the current set of solutions.


I K's picture
I K on May 15, 2013

Richard Smalley, Nobel Laureate in Chemistry, estimated a population of 10 billion by the year 2050 will require as much as 60 terawatts to meet its needs, including massive desalination.

60TW primary energy for 10B people is probably a gross overestimate and the need for massive desalination is also probably something which will not come to pass.

Efficiency of converting energy is always improving. Once upon a time we had coal and gas plants operating at around 30%. Now coal plants hit 45% and gas plants 60%. Going forward these figures will likely continue to rise somewhat and if the new super materials deliver on their promises of much much greater strengths and temps then both coal and gas can be pushed to much higher efficiencies.

With computer cars and better building regulations and more efficiency I would wager 10B people in 2050 will need at most 30TW primary energy and perhaps as low as 20TW

I K's picture
I K on May 15, 2013

“To produce 60 TW with this approach you would extract 120 TW from the surface and ideally dump 60TW worth of heat to the depths”

You have made a mistake somewhere or have discovered a way to get 50% efficency from a temp gradient of 20 centigrade. The former is magnitudes more likely.

Anyway, how does a heat pipe help?
Lets assume for arguments sake you want to produce 60TW of eletricity from OTEC operating at 3.33% net efficency. That means you need to move 1800TW of heat.

No matter what you medium of heat exchange is, 1800TW is a MASSIVE number. You are going to need a HUGE HUGE HUGEEEEEEEEE quantity of your working fluid. Using the water itself is probably best becuase water has quite a high specific heat capacity and well it is free. Using anything else is going to cost you an arm and a leg

You still need a working fluid but if you pump the cold and hot water so they are close your working fluid can be recycled many times over rather than having to travil 2km up and down each cycle.

BTW the 60TW of eletricity you are producing, where is that heat going?

I K's picture
I K on May 15, 2013

Your going to have to explain this once more. This is what I understand so correct the bits where I have gone wrong.


1: you need a cold sink and a hot sink

2: you need a working fluid

3: one method is to bring the cold and hot sink close to each other (ie pump the cold water to the surface) so that your working fluid has not far to travil to do its heat transfer.

4: another method is to use the working fluid and move that the vast distance. So the fluid vaporises at the hot surface and then is moved to the cold ocean depth where it condenses and then you need to pump that fluid up again for it to vaporise at the hot ocean surface.

5: therefore in both cases you have a “heat pipe” the only difference is that in #4 the working fluid needs to travil a much lessor distance and hence can be recycled through the hot and cold sinks many more times per time unit

6: sure a phase change may carry more heat in its liquid stage (pumping liquid CO2 at pressure from the ocean depth up to the surface) but how are you going to move the huge quantity of heat when it is in its gas phase? you just vaporised your liquid CO2 and turned a turbine, its now a gas at some pressure, you need to pump this gas down to the ocean depth for it to turn back into a liquid. What is the diameter pipe you need to move that much heat vs the water pipe? water has a specific heat capacity of 4.2kj/kg while CO2 is around 0.8kj/kg. more importantly water is 1000 kg per m3 while CO2 gas at 5atm is a lot lot less so you are going to need a much bigger pipe moving your CO2 not a smaller pipe


so once of us has something very wrong

I K's picture
I K on May 15, 2013

Don’t underestimate the engineering challenges of making a physically possible idea work. Often there are many unforeseen factors which will add a lot of complexity and cost.


For arguments sake imagine your actual heat exchanger barrier (probably a massive copper wall of some sort). It may get clogged by fish or algae or all sorts of life or chemistry (like the bottom of ships are) and hence need frequent cleaning and scraping or even replacing. No easy task 1,000km out at sea.


Also bear in mind no matter what method you use the surface area of your actual heat exchanger needs to be massive simply because you are trying to transfer 1800TW of heat through a gradient of less than 20 centigrade.


You are looking at around 1KW of heat per square meter of heat exchanger. Therefore to move 1,800TW of heat you need some 1,800 billion square meters of surface contact between your heat sources.


That is not going to be easy. Lets scale down from 60TW to just 60GW for something maybe more realistic you still need around 1.8 billion square meters of surface contact to generate just 60GW of electricity.


To be frank that is a ridiculously large figure. Effectively you need a heat exchanger with a surface area of 1,800 km2 to produce just 60GW of electricity (about equal to just the nukes in france)


That would be ridiculous on land let alone far offshore

I think I have figured out a strong reason for why its not likely to work

I K's picture
I K on May 15, 2013

Effectively you need a heat exchanger with ~ thousand times the surface area vs high temp methods and that will make it difficult for the tech.

Also if the ocean temp difference you are using is 20 centigrade your actual fluid temp difference will be even lower. (your CO2 vaporising and condensing temps are your hot and cold sinks not the ocean temps).

Using coal analogy the boiler temp is used, not the flame temp of the burning fire fed by the coal.

I K's picture
I K on May 15, 2013

Oh I got it, you think I meant heating the working fluid so it expands as PV=NRT, no that would likely be unworkable you would need far too long a time period for the heat to transfer via your working gas for it to build up pressure.

You will need a phase change in your working fluid from liquid to gas but that does not remove the need to pump water from the depth up to the surface.  The alternative of pumping your CO2 in its gas phase down would likely not work because a given volume of CO2 gas, even at pressure, will carry substantially less heat than water.  You need a phase change fluid, you call it a heat pipe, but the ends of the heat pipe cant be the ocean top to the ocean bottom. It will have to be a heat pipe (phase change working fluid) but it will have to be via a short distance between the cold sink and hot sink. Imagine it at moving only a few meters rather than 2km down and then up.  Anyway none of this gets over the fundamental need to have a huge surface area in contact with the heat sinks. Its actually worse than I originally pointed out because the temp difference between your heat exchanger will not be 20 centigrade but perhaps as low a 1 centigrade . Therefore your heat exchanger will be very ineffective vs heat exchangers in normal power stations which operate at a thousand centigrade difference (and they need to move 10-20x less heat) so your exchanger needs to be perhaps 10,000 x the surface area.

Rob Freda's picture
Rob Freda on May 15, 2013

IK’s point is well made.  For example, I assume you are talking about hydrogen and maybe some type of oxide?  So the reason we do not have a hydrogen economy based on electrolysis is because of the cost of storage and locally produced electricity.  We do not have a clean technology that can cheaply produce hydrogen locally.  Add to that hydrogen is a serious pain to store.  Slippery little bastards.   Ipso no hydrogen economy because if local production is a pain centralized production is a nightmare which is what you are talking about.  Centralized production in a marine environment 5,000 km from the end use.  

So again the problem is getting the energy 5,000 km to the load centers.  Oil transport quite effectively as does coal.   NG not so much but nowhere near as bad hydrogen.  I would point you to offshore wind  costs a lot more per kWh than onshore despite a substantially better energy resource.  If hydrogen does not work onshore the OTEC system would be have to be pretty cheap just to absorb the cost of operating in a marine environment.  

Had a question.  When you are quoting the stored thermal energy I believe you are talking about the total thermal energy contained in the mass correct?  The OTEC as described would only exploit the temperature differential between the surface and the system depth which is a fraction amount of the resource you are describing (at least what I think you are quoting).  The OTEC would only have access to the Hemholtz free energy.  So it would operate roughly in the same type of energy environment as wind or geothermal which also capitalize on thermal differentials. 

I am having a very big problem seeing how with the problems that similar techs have had being cost effective and with the marine environment and remoteness this is going to be cost effective.  

I think your overall thinking is correct but this strikes me a bit like initial estimates and excitement that sent us off down the solar path 30 years ago.  A lot of resource.  A lot less technical resource. A lot less cost effective than everyone thought.

CO2 would certainly be a good operating fluid.  So the differential is not that large?

I K's picture
I K on May 16, 2013

It is potentially a good idea because of its continuous operation. Prototypes should be built and tested that’s the only way to encounter the full range of problems you would not predict.

My assumption would be that with just twenty centigrade between your hot and cold sinks its going to be very difficult