This 16-page note models the green hydrogen value chain: harnessing renewable energy, electrolysing water, storing the hydrogen, then generating usable power in a fuel cell. Today’s end costs are very high, at 64c/kWh. Even by 2050, our best case scenario is 14c/kWh, which would elevate average household electricity bills by $440-990/year compared with the superior alternative of decarbonizing natural gas.
Voltaire famously slated the Holy Roman Empire for being neither holy, nor Roman, nor an Empire. The same criticism may apply to the Green Hydrogen Economy. Well over 80% of 2050’s hydrogen market is likely to be blue hydrogen. This is ultimately derived from natural gas energy (increasing gas demand). And it is still not economical, costing 39c/kWh today and 11c/kWh in our best case by 2050.
Our baseline costs to decarbonize natural gas power are presented on page 2.
Renewable energy inputs to green hydrogen production are costed on pages 3-4,
Electrolyser costs, and potential future improvements, are covered on pages 5-7.
Distributing and storing hydrogen is surprisingly challenging. We review 5 key reasons and derive base- and best-case cost estimates on pages 8-11.
Generating electricity from green hydrogen is costed on pages 12-13.
What do you have to believe to be constructive on green hydrogen costs, in the best case scenario in the 2040s and 2050s? We answer this question on page 14.
Relative advantages of blue hydrogen are discussed on pages 15-16, although we still think decarbonized natural gas will be a superior option for the energy transition.
We presented our ‘Top Ten Themes for Energy in the 2020s’ to an audience at Yale SOM, in February-2020. The audio recording is available below. The slides are available to TSE clients, in order to follow along with the presentation.
Please sign up to our distribution list, to receive our best ideas going forwards…
Molten carbonate fuel cells (MCFCs) could be a game-changer for CCS and fossil fuels. They are electrochemical reactors with the unique capability to capture CO2 from the exhaust pipes of combustion facilities; while at the same time, efficiently generating electricity from natural gas. The first pilot plant was due to be tested in 1Q20, by ExxonMobil and FuelCell Energy, but was deferred. Economics range from passable to phenomenal. The opportunity is outlined in our 27-page report.
Pages 2-4 outline the market opportunity for more efficient carbon separation technologies, which can be retrofitted to 4TW of pre-existing power plants, without adding $50/T of cost and 15-30% of energy penalties per traditional CCS.
Pages 5-13 outline how MCFCs work, including their operation, development history, how recent patents promise to overcome reliability problems, and their emergent adaptation to carbon capture.
Pages 14-18 assess the economics, both in absolute terms, and by comparison to new gas plants and hydrogen fuel cells. CCS-MCFC economics range from passable to phenomenal, at recent power prices.
Pages 19-23 suggest who might benefit. Fuel Cell Energy has received $60M investment from ExxonMobil, hence both companies’ prospects are explored.
Appendix I is an overview of incumbent CCS technologies, and their limitations.
Appendix II is an overview of six different fuel cell types, comparing and contrasting MCFCs.
Gas demand could treble by 2050, gaining traction not just as the world’s cleanest fossil fuel, but also the most economical. The ascent would be driven by technology. Hence this note outlines 200MTpa of potential upside to consensus LNG demand, via de-carbonised power and shipping fuels. LNG demand could thus compound at 8% pa to 800MTpa by 2030, justifying greater investment in unsanctioned LNG projects.
[restrict]
Consensus LNG demand?
A simple model of global LNG demand is shown below (and downloadable here). It is created by extrapolating recent trends in key LNG-consuming regions. The total market grew at 5.7% pa in 2013-18. At a 5.4% forward CAGR, it would reach c570MTpa by 2030. These numbers are not far from other LNG forecasters’, and thus serve as a reasonable consensus.
What excites us is the potential for technology to accelerate LNG demand. Markets are slow to reflect technological breakthroughs. Hence these new demand sources likely do not feature in consensus forecasts yet. In our view, this makes them worthy of attention.
Upside from De-Carbonised Power Generation?
The first opportunity is in de-carbonised power generation, as we have discussed in our deep-dive report, ‘de-carbonising carbon‘. We think novel technologies are reaching maturity, which can generate cost-competitive electricity (chart below) alongside an exhaust stream of pure CO2, for use in industry or for immediate sequestration. The full details are in our report.
Let us now make some approximate calculations: The world consumes 7.7bn tons of coal per annum. In energy terms, this is equivalent to c165TCF of gas, or 3,300MTpa of LNG. We believe it would be economic, and achievable, to convert c5% of this coal power to gas by 2030. Converting it to decarbonised gas could save c1bn tons of CO2 emissions per annum. In turn, this could be achieved by 200GW of de-carbonised gas-power, in 500 x 400MW power plants, each burning c50mmcfd of input gas, fed by 165MTpa of LNG. This is the first area where technology can greatly accelerate LNG demand.
Upside in Shipping?
The second opportunity is in LNG as a shipping fuel, which will become increasingly economical after IMO 2020 sulphur regulations re-shape the marine sector. The economics are shown below and modelled here.
New technologies in small-scale LNG will accelerate adoption in smaller ports, moving beyond the large port-sizes required for bunkering. The technologies and economics are explored in detail, in our deep-dive note, LNG in Transport. The economics are modeled here. To assist, Shell is also pioneering new solutions for LNG in transport.
The upshot could be 40MTpa of incremental LNG demand in the maritime industry by 2030. This is the second area where technology can greatly accelerate LNG demand.
Less positive on trucking
Is there further upside? One might expect, in an overview of LNG technologies, to find incremental upside in road vehicles: either directly in LNG-fired trucks, in gas-fired vehicles, or to produce hydrogen for fuel-cells. None of these opportunities are yet captured in our models.
The reason is economics. Compared to diesel-powered trucks, we find compressed natural gas to be c10% more expensive, LNG to be 30% more expensive and hydrogen to be around 4x more expensive (model here, chart below). We also find hydrogen to be 85% costlier than gasoline, to powers cars in Europe (model here). In most cases, electrification is the better option, as superior vehicle concepts emerge.
Our numbers do not include any incremental LNG demand in the road-transportation sector. However, it is noteworthy that replacing 1Mbpd, or c2% of the world’s road fuels with LNG would consume an incremental 50MTpa of LNG. This could cushion delays or shortfalls in decarbonised gas-power.
Potential supplies can meet the challenge.
It is only possible for the world to consume 800MT of LNG in 2030 if it is also possible to supply 800MT. While our risked forecasts are for c600MT of LNG supply in 2030 (chart below), our numbers are including just c60% of the 230MTpa of LNG capacity that is currently in the design phase, and just 15% of the 180MTpa that is currently in the discussion phase. In a generous scenario, our forecasts rise close to the 800MTpa level that is required. Please download our risked, LNG supply model to see our scenarios, and the LNG projects included.
LNG technology could thus unlock incremental LNG facilities. We are most positive on low-cost, low-CO2 sources of gas, particularly in stable and low-tax countries. To help assess the potential, we have therefore compiled a data-file of the world’s great gas resources and their CO2 content, downloadable here. Our positive outlook on US LNG is further underpinned by our positive outlook on US shale.
Conclusions: path dependency?
The numbers above are not hard forecasts. We do not believe hard forecasts are possible in a market that is shaped by unpredictable geopolitics, technologies, weather and its own price-reflexivity. However, we have argued that new technologies may unlock materially more LNG demand than is currently embedded in consensus expectations. Leading companies with leading LNG projects may benefit.
We define a “good battery” as one that enhances the efficiency of the total energy system. Conversely, a “bad battery” diminishes it. This distinction matters and must not be overlooked in the world’s quest for cleaner energy. Electric Vehicles are most favoured, while grid-scale hydrogen is questioned.
[restrict]
As renewable energy ramps up into the global energy mix (chart below, model here), the energy system will grow increasingly intermittent. This is unavoidable, because sunshine and wind speeds vary, and these variations are correlated over wide geographic areas.
Hence, batteries will be needed, to store up excess renewable generation for when the sun is not shining and the wind is not blowing. Most commentary on batteries focuses on their costs. But there is also an enormous opportunity in their efficiency…
Designed correctly, batteries can improve the efficiency of the global energy system, and accelerate the energy transition by lowering the total amount of energy that needs to be generated. Designed incorrectly, however, these variables are all worsened. The distinction is the topic of today’s note.
To make our distinction clearer, we have created a new data-file estimating the “net round trip efficiency” of different battery types. The calculation has two steps:
First, we measure the energy efficiency of an energy storage system (kWh given out divided by kWh put in).
Second, the storage system’s energy efficiency is compared with the most likely energy source that storage system will displace.
Interpretation. A score over 100% indicates a “good battery”. It is more efficient than the energy source displaced. A score below 100% indicates a “bad battery”. It is less efficient than the energy source displaced.
Examples to illustrate good vs bad batteries
Electric cars are our top example of a “good battery”, with potential to uplift energy efficiency by 3.5x. This is because electric vehicles achieve c60-80% energy efficiencies. An electric vehicle, in turn, is most likely to displace an internal combustion engine, which typically achieves 15-20% energy efficiency (chart below, model here). Mop up c100 units of excess renewable energy with an electric vehicle battery, and it therefore displaces the equivalent of 350 units of oil-energy.
The same 3.5x uplift applies to the battery in an ‘aerial vehicle‘, as we recently reviewed in depth, with flying cars set to achieve the equivalent of 140mpg (chart below).
Grid-scale batteries can also achieve impressive uplifts in efficiency, when inefficient fuel use is displaced. As a general rule of thumb, a power plant might be c50% efficient, hence replacing the power plant with renewables plus batteries can achieve a c2x efficiency uplift.
Additional opportunities are emerging to uplift system efficiency using batteries. One recent example was described by ConocoPhillips, at the Darwin LNG plant, where the gas turbines have a “sweet spot” of maximum efficiency. Battery storage allows Conoco to avoid low-efficiency turbine usage, which will will cut emissions by 20%.
Demand shifting, in the middle of our chart, deserves special mention because it is practically free and can also arguably uplift efficiency by 1-2x. It constitutes moving demand to the times when electricity is flowing abundantly into the grid. Examples range from backups at data-centers to washer-driers in homes. The practice can be encouraged by variable electricity prices, incentivising consumption when electricity supply is abundant and disincentivising consumption at times when it is scarce.
Now we arrive at the right-hand-side of our graph, where we are more cautious. Many commentators have proposed using hydrogen, molten salt or photo-electro-chemical cells as storage mechanisms, to absorb excess renewable power, for later usage…
We calculate that these “batteries” have c35-40% round-trip efficiencies: i.e., one third of the energy is lost to “charge” the battery (e.g., hydrolysing water) and another third is lost discharging it (e.g., burning hydrogen). Mop up 100 units of of renewable energy with one of these “bad batteries” and only c35-40 units can be recovered later. This means the world’s installed renewable capacity achieves less decarbonisation.
Decision-makers may wish to consider system efficiency in these terms, to maximise the impacts of both their renewable and battery investments. Efficiency and economics tend to overlap in all the models we have built.
[/restrict]
Cookies?
This website uses necessary cookies. Our cookies are simply to improve your experience. We do not undertake any advertising or targeting via our cookies. By clicking 'accept' or continuing to use the website, you consent to our use of cookies.AcceptTerms & Conditions
Privacy & Cookies Policy
Privacy Overview
This website uses cookies to improve your experience while you navigate through the website. Out of these, the cookies that are categorized as necessary are stored on your browser as they are essential for the working of basic functionalities of the website. We also use third-party cookies that help us analyze and understand how you use this website. These cookies will be stored in your browser only with your consent. You also have the option to opt-out of these cookies. But opting out of some of these cookies may affect your browsing experience.
Necessary cookies are absolutely essential for the website to function properly. This category only includes cookies that ensures basic functionalities and security features of the website. These cookies do not store any personal information.
Any cookies that may not be particularly necessary for the website to function and is used specifically to collect user personal data via analytics, ads, other embedded contents are termed as non-necessary cookies. It is mandatory to procure user consent prior to running these cookies on your website.
