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.

9,753 Subscribers


Negative-CO2-emissions ocean thermal energy conversion

By Greg Rau and Jim Baird

Renewable and Sustainable Energy Reviews Volume 95, November 2018, Pages 265-272

Published online August 1, 2018


• OTEC can generate electricity while cooling the surface ocean/atmosphere.

• Vertically transporting fluids other than seawater increases OTEC efficiency.

• Conversion of OTEC electricity to H2 via electrolysis allows energy transport onshore.

• Modifying electrolysis to consume CO2 produces negative-CO2-emissions OTEC, NEOTEC.

• CO2 is converted to ocean alkalinity for carbon storage and for acidity mitigation.

Conversion of the ocean’s vertical thermal energy gradient to electricity via Ocean Thermal Energy Conversion (OTEC) has been demonstrated at small scales over the past century, and represents one of the largest (and growing) potential energy sources on the planet. Here we describe how OTEC could be modified to provide a large source of CO2-emissions-negative energy while also allowing heat removal from the surface ocean, helping to directly counter ocean/atmosphere warming. Most OTEC energy potential is far offshore, thus the conversion of the produced electricity to a chemical energy carrier such as H2 or derivatives is required. This can be achieved by employing a method of electrochemically generating H2 that also consumes CO2, converting the carbon to a common form of ocean alkalinity. The addition of such alkalinity to the ocean would provide high-capacity carbon storage while countering the chemical and biological effects of ocean acidification. For each gigawatt (GW) of continuous electric power generated over one year by the preceding negative-emissions OTEC (NEOTEC), roughly 13 GW of surface ocean heat would be directly removed to deep water, while producing 1.3 × 105 tonnes of H2/yr (avoiding 1.1 × 106 tonnes of CO2 emissions/yr), and consuming and storing (as dissolved mineral bicarbonate) approximately 5 × 106 tonnes CO2/yr. The preceding CO2 mitigation would result in an indirect planetary cooling effect of about 2.6 GW. Such negative-emissions energy production and global warming mitigation would avoid the biophysical and land use limitations posed by methods that rely on terrestrial biology.

Content Discussion

Schalk Cloete's picture
Schalk Cloete on August 3, 2018

Hi Jim. Congratulations on the publication of your peer-reviewed paper. It is interesting to see some preliminary economic assessment of OTEC. 

After scanning through the paper (mostly the economics section), I have the following comments that would be nice to get your response on:

1. The estimated H2 production cost of $14/kg is very expensive. Clean H2 production from excess wind & solar, nuclear or natural gas with CCS comes in below $3/kg.

2. The paper estimates that 70 kWh of electricity is required for 1 kg of H2. Thus, at the assumed $0.2/kWh electricity cost, the above-mentioned $14/kg H2 is only accounting for electricity costs. The costs of the electrolyser and H2 transport is not included. 

3. Capital costs for the electrolyser can be quite low compared to the large energy cost above, although this electrolyser has to work with seawater very far off-shore, making it a lot more expensive. Maybe we can add $2/kg for this cost. 

4. Transporting H2 from the deep ocean to distributed demand centres will be very expensive. Long distance onshore H2 transport costs about $2/kg. We can double that for offshore, bringing the cumulative H2 cost up to about $20/kg. 

5. Mineral carbonation has been studied for a long time as a potential CO2 sequestration method. It is generally accepted to be much more expensive than CO2 transport and storage, mainly because of the massive bulk of minerals that need to be crushed and transported (about 3 kg of mineral for each kg of sequestered CO2). Moving the whole operation far off-shore will greatly increase the costs. A very high CO2 price will therefore be needed before the addition of this element becomes economically feasible. 

Unfortunately, $20/kg H2 (equivalent to $20/gal gasoline) is just never going to fly, no matter how clean it is. Compared to the internalized cost of $2/gal gasoline, $20/gal represents a CO2 price of about $2000/ton, an order of magnitude more expensive even than direct air capture. 

This paper has therefore significantly increased my scepticism about OTEC. 

Greg Rau's picture
Greg Rau on August 4, 2018

Thanks for the feedback, Schalk. I’ve really enjoyed your past, penetrating analyses of various energy and C management schemes, so I welcome the opportunity to respond to your observations here. First off, I agree that NEOTEC is at present no energy bargain and is way down the list as an energy priority in the present market. However there are 3 things that NEOTEC and no other single technology does – consumes excess CO2, cools the surface ocean and exploits vast, stranded, continuous ocean energy. So depending on how much you value these things, NEOTEC may or may not be worthwhile to consider.  Some details: 1) As for the $14/kg H2 production cost, that includes an electrolyzer cap cost of $0.6/kg H2 as contained in our ref 16 – OK maybe a little optimistic.  We also factor in mineral and water pumping costs. If we were only interested in negative emissions H2, certainly there are plenty of much cheaper renewable energy sources that could drive the cost down to a few dollars/kg – see our recent paper on this:

2) We assume that H2 can be transported 1000 km by ship at $0.07/kg H2 – transportation costs would seem the least of our concerns. Hopefully, the ships that deliver alkaline rock (total cost = $20/t rock) to the NEOTEC platform can also transport the H2 back to land, presenting an interesting challenge to the naval architects designing such ships (in addition to NEOTEC itself).

3) If you are just interested in mineral weathering NETs, then certainly NEOTEC is not your first choice, but then what other NETs also makes H2 and cools the surface ocean? And yes a very high value for NETs (and ocean cooling) would be needed for such H2 to compete in the present market. Yet here in CA the low carbon fuel standard credit of some $100/t CO2 avoided would, if my math is corrected, yield a NEOTEC revenue of $5/kg, for a net H2 cost of $9/kg. Current pump price of CA H2 is $6-$13/t kg. Why anyone would pay the equivalent of $6-$13/gal of gasoline is another matter, but in fact there are soon to be 1000s of H2 cars here in CA whose owners are willing to pay that price.  They might be willing to pay even more if they knew that the H2 they were using was also helping conserve marine life and reducing the severity of tropical storms by cooling the surface ocean – they might line up for such “supergreen” H2, adding a whole new meaning to “premium” grade fuel. Let’s find out.  Thanks again for your interest, and I look forward to further discussion.

Jim Baird's picture
Jim Baird on August 4, 2018

Schalk, we reference the MIT thesis of Muralidharan that shows the following chart for the levelized cost and capacities of various technologies.

The OTEC figures are for conventional OTEC and as Muralidharan points out heatpipe OTEC comes in at 66% of the figures in this table.

Wind and solar have come down in price in the past six years but for the reasons Greg has itemized OTEC provides other benefit and looks to be competive with nuclear, which some environmentalists look to for salvation.

As our paper  points out the economics of NEOTEC are somewhat speculative but the technology deserves consideration.



Greg Rau's picture
Greg Rau on August 4, 2018

And just to followup on the implications of Muralidharan's $0.086/kWhe, this is 57% lower than the $0.2/kWhe we cautiously assumed in our NEOTEC paper. If the former is a valid number, that means NEOTEC H2 could be produced and delivered at about $6/kg H2. Subtract the $100/t CO2 LCFS credit and you've got $1/kg H2, lower than gasoline or fossil H2 on a per energy basis(!) 

Schalk Cloete's picture
Schalk Cloete on August 5, 2018

Thanks for the reply, Greg. Given that this is such a new idea, we are afforded some liberties in cost estimation. However, my feeling is that the cost of the specialized offshore electrolyser with enormous mineral throughput, mineral transport to the OTEC site, and H2 transport to shore will more than eat up the $4/kg revenue you can get from a $100/ton CO2 credit. 

I followed your reference for the $0.07/kg shipping cost and just found a blog post that seems to be shipping costs for general commodities. Shipping H2 at 700+ bar will require very expensive and heavy tanks made from specialized alloys. My guess is that H2 shipping will be at least 10x more expensive than LNG because of these high-pressure tanks and the fact that the energy density of compressed H2 is 4x lower than LNG. Maybe NH3 will be a better option. 

The $0.07/kg shipping cost is more applicable to the huge quantity of minerals required. If we produce 40 kg of CO2 per kg H2 and need 1.5 kg of mineral per kg of CO2, the mineral shipping costs amount to 0.07x40x1.5 = $4.2/kg H2.

Getting OTEC electricity costs down will certainly help. I'm just finding it hard to believe that a heat engine operating at 3% efficiency 1000 km offshore can result in an $86/MWh LCOE (due to the enormous size, the great distance and hostile offshore environment). 

Schalk Cloete's picture
Schalk Cloete on August 5, 2018

The idea of using hydroxides to increase ocean alkalinity and increase the ocean's role as a carbon sink is interesting though. As an example, carbonation of olivine can be done with a power plant flue gas onshore and the resulting MgCO3 can then be dumped in the ocean to form Mg(HCO3)2 with dissolved CO2.

Of course, this will only be economically interesting if the plant carbonating the olivine is close to the olivine mine and the ocean, but this can theoretically turn any fossil fuel combustor into a CO2-negative plant: produced CO2 is captured to form MgCO3 and each mol of MgCO3 then captures another mol of CO2 in the ocean, which will eventually absorb that mol of CO2 from the atmosphere again.

Do you think something like this is possible?

Jim Baird's picture
Jim Baird on August 5, 2018

Schalk, the efficiency of the Heat Pipe OTEC heat engine is 7.6% in our paper as calculated by Melvin Prueitt of Los Alomos labs. Conventional OTEC loses half its heat through the evaporator and the condenser. With the heat pipe design the warm surface water is adjacent the evaporator and the cold is adjacent the condenser so more of each can be moved efficiently through each dropping the heat loss of each to about 2C compared to 4C with the conventional approach. Also the weight of a column of the working fluid vapor on a the 1000 meter long column of gas increases the temperature at the bottom by about 5C.  

The diameter of the heat pipe is 1/10th a cold water pipe thus its cost and that of the entire system is reduced by at least 30%.

As to distance, the warmest surfaces temperatures exist in the Eastern Pacific, adjacent the major hydrogen markets of the region.

Hydrogen produced at 1000 meters arrives at the surface at 100 bar, which if I not mistaken, is logarithmecally 70 percent to 700 bar.   

Jim Baird's picture
Jim Baird on August 5, 2018


Schalk, after unloading bulk hydrogen ships require bast for the return leg of their journey.

Greg Rau's picture
Greg Rau on August 5, 2018

Schalk, Roger on the NEOTEC costs. But I still think we deserve some brownie points for inventiveness. As for minerals consuming CO2 at power plants, I wouldn't recommend olivine, Mg2SiO4, due to very slow kinetics.  Use of CaCO3 (limestone) as the feedstock would provide orders of magnitude faster kinetics and probably lower cost:     The end product is primarily Ca(HCO3)2aq (already the largest C reservoir on the Earth's surface).  This won't represent negative emissions unless the power plant is fueled by biomass or derivatives. ;-)

Schalk Cloete's picture
Schalk Cloete on August 6, 2018

Jim, how does the OTEC efficiency match the Carnot limit? Normally a heat engine finds it difficult to get to 60% of Carnot efficiency. 

Greg, I agree that the idea is innovative :-) Have you assessed the possibility of simply transporting electricity to shore via offshore cables? This is expensive, but over 1000 km, I think it could be significantly cheaper than H2. Compared to offshore wind, the cabling will be longer, but will also be used at a much higher capacity factor. Electrolysis can then be done on-shore during times of high wind/solar output with much lower costs for mineral transport and H2 distribution. 

Jim Baird's picture
Jim Baird on August 6, 2018

Schalk, Vicente Fachina, in his paper Deep-Subsea OTEC, simulates an OTEC concept with a threefold increase in exergy efficiency over the topside version. This is in line with the finding of 7.6% with the program OTEC.exe used by Prueitt in his patent application.

Alex Michaelis of EnergyIsland produced a map showing how OTEC can service the globe with DC power but I can't lay my hands on it.

If and when it comes available I will post it. I think I once posted  it in a comment on TEC but when?