FEATURE: Energy storage at sea could enable full transition to renewable power

Dr Jody Muelaner

A huge increase in grid-level energy storage is likely to be required as electricity transitions to renewable sources (Credit: Shutterstock)
A huge increase in grid-level energy storage is likely to be required as electricity transitions to renewable sources (Credit: Shutterstock)

Last year Bloomberg reported that the price of energy produced from wind and solar had fallen below the cost of energy produced from coal.

This was found to be true in Britain, France and Germany. It means that renewable energy can now compete with fossil fuels on price, without requiring government subsidies. 

However, renewable energy is only cheap when it is being produced. When the sun isn’t shining and the wind isn’t blowing, we remain dependent on fossil fuels and nuclear power. It is possible to store energy produced from renewable sources but the current options are limited. A huge increase in grid-level energy storage is likely to be required as electricity is fully transitioned to renewable sources such as wind and solar.

There are a few established technologies for grid-level energy storage. The oldest is pumped-storage hydroelectricity in which energy is stored by pumping water up to an elevated reservoir and then allowed to flow back down through a turbine to generate electricity when required. More recently, grid-scale battery banks are starting to be installed. In their current form, neither of these technologies can be deployed at the scale and rapidity needed to reduce greenhouse gas emissions to safe levels.

An innovative pumped-storage hydroelectricity technology uses a concrete sphere located on the seabed as the lower reservoir. No upper reservoir or transmission pipe is required since the surrounding seawater provides the necessary water pressure. This has significant potential to provide near-term highly scaleable grid-level energy storage, integrated with other offshore power facilities.  

Conventional hydro

The most economical way of storing really large amounts of energy is pumped-storage hydroelectricity (PSH). This stores surplus electricity as potential energy, typically by pumping water from a low-lying reservoir up to another reservoir located much higher up in a mountainous region. When electricity is needed, the water is allowed to flow back down to the lower reservoir, passing through a turbine that generates electricity. Round-trip energy efficiency is typically 70 to 80%. Total global installations of PSH amount to 127GW, over 99% of the world’s bulk-storage capacity.

The nominal capacity of PSH can be easily calculated by multiplying the mass of water by the elevation and the acceleration due to gravity. A number of questionable designs have received considerable interest on the internet. These have proposed small-scale PSH and other potential energy-based systems within buildings. 

Simple ‘back of the envelope’ calculations show that the technology is not well-suited to this type of application. For example, a typical home uses about 1kW of power and its roof space is approximately 5m above the ground. Therefore, to store the 12kWh required for a 12-hour period would require a mass of 880 tonnes. This would require very significant reinforcements to the foundations and structure, making it an uneconomical way to store such a small quantity of power.

For PSH to be economical, it must use existing geology to support the elevated mass of the water. There are examples of relatively small PSH projects working where this support is available. For example, GE produced 40m-high cylinders that contain 40,000 tonnes of water at the base of a wind turbine. These wind turbines have been installed on a mountain ridge, 200m above a natural reservoir, giving each turbine an energy-storage capacity of 18MWh with a peak power output of 3.4MW. 

This type of integration can work well for select sites but the very particular requirements mean that its application is ultimately quite limited.

Where PSH has been most successful is for really large installations in which the reservoirs themselves are contained by surrounding geology and even the high-pressure pipe connecting the reservoirs is drilled through the rock to provide support. These facilities may have elevation differences of more than 500m between the upper and lower reservoirs. 

Potential sites must be located relatively close to population centres while providing the required geology and water supply to create large reservoirs relatively close to one another and yet with sufficient difference in elevation. Sites that combine all of these features are rare. 

The Raccoon Mountain pumped-storage plant in Tennessee has an upper reservoir that covers an area of 214ha elevated 320m above the turbine. It has a capacity of 36GWh, producing 1,652MW for 22 hours

Suitable mountainous sites close to cities invariably have significant cultural value, for example in national parks, making it politically difficult to construct the large dams and other required infrastructure. Difficulty in locating PSH, combined with long lead times for these major projects, means that the technology is not able to keep pace with the speed and scale of the energy transition. 

This may however change if it becomes politically acceptable to locate seawater upper reservoirs in mountains close to coastal cities. It would then be possible to use the sea as the lower reservoir – reducing construction costs and water use, and greatly increasing the number of available sites.

Limitations of batteries

Recently installed grid-scale batteries use cobalt-based lithium-ion technology. This application is not sustainable owing to critical shortages of the cobalt required for the electrification of vehicles. Although cobalt supply initially kept pace with the increased demand caused by mobile devices, more recent demand from electric vehicles has driven up the cobalt price by 300% since 2015. 

Approximately 80m cars are produced globally each year. A battery-electric car requires about 10kg of cobalt, although a Tesla uses more than twice that. This means that 800,000 tonnes of cobalt would be required annually if every new car was a small battery-electric vehicle. This alone is eight times the world’s current annual production of cobalt, but the material is also required for other applications such as metal alloys and catalysts. Transitioning to renewable energy will, therefore, require an order of magnitude increase in the supply of cobalt.

Cobalt deposits are mixed with other metals, meaning that it is usually extracted as a by-product of mining other materials, such as copper, nickel or silver. About 60% of the world’s cobalt is currently supplied by the Democratic Republic of Congo. A combination of shortages, instability of supply and ethical issues around Congolese cobalt are driving interest in mining in other countries, closer to major markets. 

However, there is very little chance that these mines will be able to keep up with the demand from rapid decarbonisation. Worldwide cobalt reserves on land are estimated at 7m tonnes.

There is increasing commercial interest in mining ocean nodules lying on the seafloor. It is thought that such nodules will eventually be able to economically supply enough cobalt for widespread automotive electrification as well as other applications. For example, the Clarion-Clipperton Zone is a large area of the Pacific extending from the west coast of Mexico to Hawaii. Its nodules contain an estimated 44m tonnes of cobalt. 

However, deep-sea mining has its own issues. The nodules mostly occur at great depths in the deep ocean. Mining them will require new robotic systems. Developing and deploying them at scale will take time. 

The ecological impact of deep-sea mining is also a significant concern. The richest sites have concentrations of 160 tonnes of cobalt per square kilometre. Extracting the cobalt required for a fully electrified global automotive industry would therefore mean mining about 5,000km2 annually. How this would affect the organisms living there is not yet known, although the research project Blue Nodules is currently investigating it. Since the deep ocean habitat is largely unexplored, it could mean the loss of species before they have been described by science, destroying important resources for pharmaceutical development. It may also affect the oceans’ ability to buffer against climate change.

Supplies in seawater

There is also cobalt within seawater itself. With an average concentration of about 0.39µg/L this amounts to 507m tonnes spread throughout the world’s oceans. Experiments suggest that selective absorption could be used to extract this from water flowing past floating structures such as disused oil rigs and floating wind turbines. In time, this may provide a plentiful supply but, like deep-ocean nodule mining, it is unlikely to be available at the scale and rate required to meet urgent climate objectives.

Batteries that require significantly less cobalt are already coming into the market, but current technologies will still require more cobalt than is available. New battery technologies may one day provide economical grid-scale energy storage without critical metal limitations. However, this technology remains speculative and is unlikely to be widely available within the timeframe required to avert major disruption from climate change.

Is this the solution?

A new form of PSH has been developed by the Fraunhofer Institute for Energy Economics and Energy System Technology in Germany. The project, entitled Storing Energy at Sea (StEnSea), uses concrete spheres anchored on the seafloor. To store energy, water is pumped out of the spheres, against the pressure of the surrounding seawater. When the energy is required, water is allowed to flow back into the spheres, driving turbines. 

How one of the spheres would be constructed. Once finished, water would flow into the sphere to start the turbine and generate power when needed (Credit: Maxine Heath)

How one of the spheres would be constructed. Once finished, water would flow into the sphere to start the turbine and generate power when needed (Credit: Maxine Heath)

No upper reservoir or transmission pipe is required since the surrounding seawater provides the necessary water pressure. When compared to conventional PSH, the need for land is eliminated and structures are minimised. It eliminates the danger of dam collapse and improves storage efficiency since evaporation from the upper reservoir does not reduce the energy stored. 

It is proposed that spheres with a diameter of 30m would be located at a depth of 700m, giving a nominal storage capacity of 27MWh. When the actual internal volume and operating efficiency of the system are taken into account, this configuration is expected to yield 18.3MWh storage capacity with a peak power output of 5MW.

A technical complication with pumping water out of a submerged sphere is that it reduces the internal pressure. If sufficient air is retained within the vessel to maintain atmospheric pressure when it is pumped empty, this would only occupy 1% of the volume when filled with water at a depth of 700m. However, the researchers found that to prevent cavitation in the main pump-turbine it is advantageous to use a feed pump. This fills a cylinder which then feeds into the pump-turbine. Both pumps must have an input pressure above the net positive suction head to avoid cavitation while pumping water from the inner volume into the cylinder or from the cylinder out of the sphere.

An initial four-week test of the system was completed in 2016 using a 1:10 scale model operated at a depth of 100m in Lake Constance. Economic analysis has been carried out showing that, when located within an energy park containing over 100 units, at an optimum location the capital cost per unit would be approximately €8m. The storage cost would be around six or seven euro cents per kWh, making this system more economical than many conventional PSH systems. 

However, at current energy trading prices this would not be commercially viable without subsidy. To profitably operate a commercial system, the energy arbitrage, or difference between purchase and sale price, would need to reach at least 4 €ct/kWh and potentially as high as 20 €ct/kWh for poorly utilised systems. At the lower end, this seems plausible with current trends in the energy market.

Although analysis has shown that StEnSea could be operated at depths of between 200m and 1,500m, it would be most economical where the concrete wall thickness required to withstand the hydrostatic pressure provides just enough ballast mass. Optimum locations were found to be water depths of between 600 and 800m, with a seafloor slope of less than one degree. They should also be within 100km of the electrical grid and a maintenance base, and within 500km of an installation base. 

A survey found that these optimal sites alone could provide a total global storage capacity of 817TWh. This represents approximately two days of the world’s total energy consumption. The top five countries (US, Japan, Saudi Arabia, Indonesia and the Bahamas) represent 39% of this capacity. However, many countries have significant capacity to meet their own storage requirements. For example, France has a potential capacity of 5.7TWh. If the system were extended to all economically viable water depths the capacity would be many times higher.

It is envisaged that the storage units would be located close to offshore wind turbines, further reducing the need for transmission infrastructure. Floating wind turbines will be particularly well suited due to the requirement for deep water and the fact that the concrete spheres can serve a double duty as anchor points for the turbines.

Proven technology

The StEnSea system does not impact on unspoiled mountainous landscapes and it doesn’t require any rare metal resources. The technology has already been proven with a scaled-down demonstrator and sufficient global locations have been identified. Economic analysis has also shown that it would be possible for such energy-storage systems to be run commercially. 

Combined with floating wind turbines, the system could provide near-baseload-quality utility-scale renewable energy and do double duty as the anchoring point for the generation platforms. It therefore has the potential to provide the scale of grid-level energy storage needed for a full transition to renewable energy with minimal negative impacts.

Want the best engineering stories delivered straight to your inbox? The Professional Engineering newsletter gives you vital updates on the most cutting-edge engineering and exciting new job opportunities. To sign up, click here.

Content published by Professional Engineering does not necessarily represent the views of the Institution of Mechanical Engineers. 


Read more related articles

Professional Engineering magazine

Professional Engineering app

  • Industry features and content
  • Engineering and Institution news
  • News and features exclusive to app users

Download our Professional Engineering app

Professional Engineering newsletter

A weekly round-up of the most popular and topical stories featured on our website, so you won't miss anything

Subscribe to Professional Engineering newsletter

Opt into your industry sector newsletter

Related articles