The hydrogen economy is a proposed system of delivering energy using hydrogen. The term hydrogen economy was coined by John Bockris during a talk he gave in 1970 at General Motors (GM) Technical Center. The concept was proposed earlier by geneticist J.B.S. Haldane.
Proponents of a hydrogen economy advocate hydrogen as a potential fuel for motive power (including cars and boats) and on-board auxiliary power, stationary power generation (e.g., for the energy needs of buildings), and as an energy storage medium (e.g., for interconversion from excess electric power generated off-peak). Molecular hydrogen of the sort that can be used as a fuel does not occur naturally in convenient reservoirs; nonetheless it can be generated by steam reformation of hydrocarbons, water electrolysis or by other methods.
A spike in attention for the concept during the 2000s has been repeatedly described as hype by some critics and proponents of alternative technologies. A resurgence in the energy carrier is now underway, notably by the forming of the Hydrogen Council in 2017. Several manufacturers have now released hydrogen fuel cell cars commercially, with manufacturers such as Toyota and industry groups in China planning to increase numbers of the cars into the hundreds of thousands over the next decade.
A hydrogen economy was proposed by the University of Michigan to solve some of the negative effects of using hydrocarbon fuels where the carbon is released to the atmosphere (as carbon dioxide, carbon monoxide, unburnt hydrocarbons, etc.). Modern interest in the hydrogen economy can generally be traced to a 1970 technical report by Lawrence W. Jones of the University of Michigan.
In the current hydrocarbon economy, transportation is fueled primarily by petroleum. Burning of hydrocarbon fuels emits carbon dioxide and other pollutants. The supply of economically usable hydrocarbon resources in the world is limited, and the demand for hydrocarbon fuels is increasing, particularly in China, India, and other developing countries.
Proponents of a world-scale hydrogen economy argue that hydrogen can be an environmentally cleaner source of energy to end-users, particularly in transportation applications, without release of pollutants (such as particulate matter) or carbon dioxide at the point of end use. A 2004 analysis asserted that “most of the hydrogen supply chain pathways would release significantly less carbon dioxide into the atmosphere than would gasoline used in hybrid electric vehicles” and that significant reductions in carbon dioxide emissions would be possible if carbon capture or carbon sequestration methods were utilized at the site of energy or hydrogen production.
Hydrogen has a high energy density by weight but has a low energy density by volume. Even when highly compressed or liquified, the energy density by volume is only 1/4 that of gasoline, although the energy density by weight is approximately three times that of gasoline or natural gas. An Otto cycle internal-combustion engine running on hydrogen is said to have a maximum efficiency of about 38%, 8% higher than a gasoline internal-combustion engine.
The combination of the fuel cell and electric motor is 2-3 times more efficient than an internal-combustion engine. Capital costs of fuel cells have reduced significantly over recent years, with a modeled cost of $50/kW cited by the Department of Energy.
Previous technical obstacles have included hydrogen storage issues and the purity requirement of hydrogen used in fuel cells, as with current technology, an operating fuel cell requires the purity of hydrogen to be as high as 99.999%. Hydrogen engine conversion technology could be considered more economical than fuel cells.
Current hydrogen market
Hydrogen production is a large and growing industry, as of 2004. Globally, some 57 million metric tons of hydrogen, equal to about 170 million tons of oil equivalent, were produced in 2004. The growth rate is around 10% per year. Within the United States, 2004 production was about 11 million metric tons (Mt), an average power flow of 48 gigawatts. (For comparison, the average electric production in 2003 was some 442 GW.) As of 2005, the economic value of all hydrogen produced worldwide is about $135 billion per year.
There are two primary uses for hydrogen today. About half is used in the Haber process to produce ammonia (NH3), which is then used directly or indirectly as fertilizer. Because both the world population and the intensive agriculture used to support it are growing, ammonia demand is growing. Ammonia can be used as a safer and easier indirect method of transporting hydrogen. Transported ammonia can be then converted back to hydrogen at the bowser by a membrane technology.
The other half of current hydrogen production is used to convert heavy petroleum sources into lighter fractions suitable for use as fuels. This latter process is known as hydrocracking. Hydrocracking represents an even larger growth area, since rising oil prices encourage oil companies to extract poorer source material, such as tar sands and oil shale. The scale economies inherent in large-scale oil refining and fertilizer manufacture make possible on-site production and “captive” use. Smaller quantities of “merchant” hydrogen are manufactured and delivered to end users as well.
If energy for hydrogen production were available (from wind, solar, fission or fusion nuclear power etc.), use of the substance for hydrocarbon synfuel production could expand captive use of hydrogen by a factor of 5 to 10. Present U.S. use of hydrogen for hydrocracking is roughly 4 Mt per year. It is estimated that 37.7 Mt/yr of hydrogen would be sufficient to convert enough domestic coal to liquid fuels to end U.S. dependence on foreign oil importation, and less than half this figure to end dependence on Middle East oil. Coal liquefaction would present significantly worse emissions of carbon dioxide than does the current system of burning fossil petroleum, but it would eliminate the political and economic vulnerabilities inherent in US oil importation before the commercialization of tight oil in North America.
As of 2004 and 2016, 96% of global hydrogen production is from fossil fuels (48% from natural gas, 30% from oil, and 18% from coal); water electrolysis accounts for only 4%. The distribution of production reflects the effects of thermodynamic constraints on economic choices: of the four methods for obtaining hydrogen, partial combustion of natural gas in a NGCC (natural gas combined cycle) power plant offers the most efficient chemical pathway and the greatest off-take of usable heat energy. (needs reference)
The large market and sharply rising prices in fossil fuels have also stimulated great interest in alternate, cheaper means of hydrogen production. As of 2002, most hydrogen is produced on site and the cost is approximately $0.70/kg and, if not produced on site, the cost of liquid hydrogen is about $2.20/kg to $3.08/kg.
Fuel cells as alternative to internal combustion
One of the main offerings of a hydrogen economy is that the fuel can replace the fossil fuel burned in internal combustion engines and turbines as the primary way to convert chemical energy into kinetic or electrical energy; hereby eliminating greenhouse gas emissions and pollution from that engine. Although hydrogen can be used in conventional internal combustion engines, fuel cells, being electrochemical, have a theoretical efficiency advantage over heat engines. Fuel cells are more expensive to produce than common internal combustion engines.
Some types of fuel cells work with hydrocarbon fuels, while all can be operated on pure hydrogen. In the event that fuel cells become price-competitive with internal combustion engines and turbines, large gas-fired power plants could adopt this technology.
Hydrogen gas must be distinguished as “technical-grade” (five nines pure, 99.999%), which is suitable for applications such as fuel cells, and “commercial-grade”, which has carbon- and sulfur-containing impurities, but which can be produced by the much cheaper steam-reformation process. Fuel cells require high-purity hydrogen because the impurities would quickly degrade the life of the fuel cell stack.
Much of the interest in the hydrogen economy concept is focused on the use of fuel cells to power electric cars. Current hydrogen fuel cells suffer from a low power-to-weight ratio. Fuel cells are much more efficient than internal combustion engines, and produce no harmful emissions. If a practical method of hydrogen storage is introduced, and fuel cells become cheaper, they can be economically viable to power hybrid fuel cell/battery vehicles, or purely fuel cell-driven ones. The economic viability of fuel cell powered vehicles will improve as the hydrocarbon fuels used in internal combustion engines become more expensive, because of the depletion of easily accessible reserves or economic accounting of environmental impact through such measures as carbon taxes.
Other fuel cell technologies based on the exchange of metal ions (e.g. zinc-air fuel cells) are typically more efficient at energy conversion than hydrogen fuel cells, but the widespread use of any electrical energy → chemical energy → electrical energy systems would necessitate the production of electricity.
Since the 2003 State of the Union address, when the notion of the hydrogen economy came to national prominence in the United States, there has been a steady chorus of naysayers. Most recently, in 2013, Lux Research, Inc. issued a report that stated: “The dream of a hydrogen economy… is no nearer.” It concluded that “Capital cost, not hydrogen supply, will limit adoption to a mere 5.9 GW” by 2030, providing “a nearly insurmountable barrier to adoption, except in niche applications”. Lux’s analysis speculated that by 2030, PEM stationary market will reach $1 billion, while the vehicle market, including forklifts, will reach a total of $2 billion.
Use as an automotive fuel and system efficiency
An accounting of the energy utilized during a thermodynamic process, known as an energy balance, can be applied to automotive fuels. With today’s technology, the manufacture of hydrogen via steam reforming can be accomplished with a thermal efficiency of 75 to 80 percent. Additional energy will be required to liquefy or compress the hydrogen, and to transport it to the filling station via truck or pipeline. The energy that must be utilized per kilogram to produce, transport and deliver hydrogen (i.e., its well-to-tank energy use) is approximately 50 MJ using technology available in 2004. Subtracting this energy from the enthalpy of one kilogram of hydrogen, which is 141 MJ, and dividing by the enthalpy, yields a thermal energy efficiency of roughly 60%. Gasoline, by comparison, requires less energy input, per gallon, at the refinery, and comparatively little energy is required to transport it and store it owing to its high energy density per gallon at ambient temperatures. Well-to-tank, the supply chain for gasoline is roughly 80% efficient (Wang, 2002). Another grid-based method of supplying hydrogen would be to use electrical to run electrolysers. Roughly 6% of electricity is lost during transmission along power lines, and the process of converting the fossil fuel to electricity in the first place is roughly 33 percent efficient. Thus if efficiency is the key determinant it would be unlikely hydrogen vehicles would be fueled by such a method, and indeed viewed this way, electric vehicles would appear to be a better choice. However, as noted above, hydrogen can be produced from a number of feedstocks, in centralized or distributed fashion, and these afford more efficient pathways to produce and distribute the fuel.
A study of the well-to-wheels efficiency of hydrogen vehicles compared to other vehicles in the Norwegian energy system indicates that hydrogen fuel-cell vehicles (FCV) tend to be about a third as efficient as EVs when electrolysis is used, with hydrogen Internal Combustion Engines (ICE) being barely a sixth as efficient. Even in the case where hydrogen fuel cells get their hydrogen from natural gas reformation rather than electrolysis, and EVs get their power from a natural gas power plant, the EVs still come out ahead 35% to 25% (and only 13% for a H2 ICE). This compares to 14% for a gasoline ICE, 27% for a gasoline ICE hybrid, and 17% for a diesel ICE, also on a well-to-wheels basis.
Hydrogen has been called one of the least efficient and most expensive possible replacements for gasoline (petrol) in terms of reducing greenhouse gases; other technologies may be less expensive and more quickly implemented. A comprehensive study of hydrogen in transportation applications has found that “there are major hurdles on the path to achieving the vision of the hydrogen economy; the path will not be simple or straightforward”. Although Ford Motor Company and French Renault-Nissan cancelled their hydrogen car R&D efforts in 2008 and 2009, respectively, they signed a 2009 letter of intent with the other manufacturers and Now GMBH in September 2009 supporting the commercial introduction of FCVs by 2015. A study by The Carbon Trust for the UK Department of Energy and Climate Change suggests that hydrogen technologies have the potential to deliver UK transport with near-zero emissions whilst reducing dependence on imported oil and curtailment of renewable generation. However, the technologies face very difficult challenges, in terms of cost, performance and policy.
There are many concerns regarding the environmental effects of the manufacture of hydrogen. Hydrogen is made either by electrolysis of water, or by fossil fuel reforming. Reforming a fossil fuel leads to a higher emissions of carbon dioxide compared with direct use of the fossil fuel in an internal combustion engine. Similarly, if hydrogen is produced by electrolysis from fossil-fuel powered generators, increased carbon dioxide is emitted in comparison with direct use of the fossil fuel.
Using renewable energy source to generate hydrogen by electrolysis would require greater energy input than direct use of the renewable energy to operate electric vehicles, because of the extra conversion stages and losses in distribution. Hydrogen as transportation fuel, however, is mainly used for fuel cells that do not produce greenhouse gas emission, but water.
There have also been some concerns over possible problems related to hydrogen gas leakage. Molecular hydrogen leaks slowly from most containment vessels. It has been hypothesized that if significant amounts of hydrogen gas (H2) escape, hydrogen gas may, because of ultraviolet radiation, form free radicals (H) in the stratosphere. These free radicals would then be able to act as catalysts for ozone depletion. A large enough increase in stratospheric hydrogen from leaked H2 could exacerbate the depletion process. However, the effect of these leakage problems may not be significant. The amount of hydrogen that leaks today is much lower (by a factor of 10–100) than the estimated 10–20% figure conjectured by some researchers; for example, in Germany, the leakage rate is only 0.1% (less than the natural gas leak rate of 0.7%). At most, such leakage would likely be no more than 1–2% even with widespread hydrogen use, using present technology.
In 2004, the production of unit of hydrogen fuel by steam reformation or electrolysis was approximately 3 to 6 times more expensive than the production of an equivalent unit of fuel from natural gas. When evaluating costs, fossil fuels are generally used as the reference. The energy content of these fuels is not a product of human effort and so has no cost assigned to it. Only the extraction, refining, transportation and production costs are considered. On the other hand, the energy content of a unit of hydrogen fuel must be manufactured, and so has a significant cost, on top of all the costs of refining, transportation, and distribution. Systems which use renewably generated electricity more directly, for example in trolleybuses, or in battery electric vehicles may have a significant economic advantage because there are fewer conversion processes required between primary energy source and point of use.
The barrier to lowering the price of high purity hydrogen is a cost of more than 35 kWh of electricity used to generate each kilogram of hydrogen gas. Hydrogen produced by steam reformation costs approximately three times the cost of natural gas per unit of energy produced. This means that if natural gas costs $6/million BTU, then hydrogen will be $18/million BTU. Also, producing hydrogen from electrolysis with electricity at 5 cents/kWh will cost $28/million BTU — about 1.5 times the cost of hydrogen from natural gas. Note that the cost of hydrogen production from electricity is a linear function of electricity costs, so electricity at 10 cents/kWh means that hydrogen will cost $56/million BTU.
Demonstrated advances in electrolyser and fuel cell technology by ITM Power are claimed to have made significant in-roads into addressing the cost of electrolysing water to make hydrogen. Cost reduction would make hydrogen from off-grid renewable sources economic for refueling vehicles.
Hydrogen pipelines are more expensive than even long-distance electric lines. Hydrogen is about three times bulkier in volume than natural gas for the same enthalpy. Hydrogen accelerates the cracking of steel (hydrogen embrittlement), which increases maintenance costs, leakage rates, and material costs. The difference in cost is likely to expand with newer technology: wires suspended in air can use higher voltage with only marginally increased material costs, but higher pressure pipes require proportionally more material.
Setting up a hydrogen economy would require huge investments in the infrastructure to store and distribute hydrogen to vehicles. In contrast, battery electric vehicles, which are already publicly available, would not necessitate immediate expansion of the existing infrastructure for electricity transmission and distribution. Power plant capacity that now goes unused at night could be used for recharging electric vehicles. A study conducted by the Pacific Northwest National Laboratory for the US Department of Energy in December 2006 found that the idle off-peak grid capacity in the US would be sufficient to power 84% of all vehicles in the US if they all were immediately replaced with electric vehicles.
Different production methods each have differing associated investment and marginal costs. The energy and feedstock could originate from a multitude of sources, i.e. natural gas, nuclear, solar, wind, biomass, coal, other fossil fuels, and geothermal.
Natural Gas at Small Scale
Uses steam reformation. Requires 15.9 million cubic feet (450,000 m3) of gas, which, if produced by small 500 kg/day reformers at the point of dispensing (i.e., the filling station), would equate to 777,000 reformers costing $1 trillion and producing 150 million tons of hydrogen gas annually. Obviates the need for distribution infrastructure dedicated to hydrogen. $3.00 per GGE (Gallons of Gasoline Equivalent)
Provides energy for electrolysis of water. Would require 240,000 tons of unenriched uranium — that’s 2,000 600-megawatt power plants, which would cost $840 billion, or about $2.50 per GGE.
Provides energy for electrolysis of water. Would require 2,500 kWh of sun per square meter, 113 million 40-kilowatt systems, which would cost $22 trillion, or about $9.50 per GGE.
Provides energy for electrolysis of water. At 7 meters per second average wind speed, it would require 1 million 2-MW wind turbines, which would cost $3 trillion, or about $3.00 per GGE.
Gasification plants would produce gas with steam reformation. 1.5 billion tons of dry biomass, 3,300 plants which would require 113.4 million acres (460,000 km²) of farm to produce the biomass. $565 billion in cost, or about $1.90 per GGE
FutureGen plants use coal gasification then steam reformation. Requires 1 billion tons of coal or about 1,000 275-megawatt plants with a cost of about $500 billion, or about $1 per GGE.
DOE Cost targets
Examples and pilot programs
Several domestic U.S. automobile manufactures have committed to develop vehicles using hydrogen. The distribution of hydrogen for the purpose of transportation is currently being tested around the world, particularly in Portugal, Iceland, Norway, Denmark, Germany, California, Japan and Canada, but the cost is very high.
Some hospitals have installed combined electrolyser-storage-fuel cell units for local emergency power. These are advantageous for emergency use because of their low maintenance requirement and ease of location compared to internal combustion driven generators.
Iceland has committed to becoming the world’s first hydrogen economy by the year 2050. Iceland is in a unique position. Presently, it imports all the petroleum products necessary to power its automobiles and fishing fleet. Iceland has large geothermal resources, so much that the local price of electricity actually is lower than the price of the hydrocarbons that could be used to produce that electricity.
Iceland already converts its surplus electricity into exportable goods and hydrocarbon replacements. In 2002, it produced 2,000 tons of hydrogen gas by electrolysis, primarily for the production of ammonia (NH3) for fertilizer. Ammonia is produced, transported, and used throughout the world, and 90% of the cost of ammonia is the cost of the energy to produce it. Iceland is also developing an aluminium-smelting industry. Aluminium costs are driven primarily by the cost of the electricity to run the smelters. Either of these industries could effectively export all of Iceland’s potential geothermal electricity.
Neither industry directly replaces hydrocarbons. Reykjavík, Iceland, had a small pilot fleet of city buses running on compressed hydrogen, and research on powering the nation’s fishing fleet with hydrogen is under way. For more practical purposes, Iceland might process imported oil with hydrogen to extend it, rather than to replace it altogether.
The Reykjavík buses are part of a larger program, HyFLEET:CUTE, operating hydrogen fueled buses in eight European cities. HyFLEET:CUTE buses were also operated in Beijing, China and Perth, Australia (see below). A pilot project demonstrating a hydrogen economy is operational on the Norwegian island of Utsira. The installation combines wind power and hydrogen power. In periods when there is surplus wind energy, the excess power is used for generating hydrogen by electrolysis. The hydrogen is stored, and is available for power generation in periods when there is little wind.
United States has a hydrogen policy with several examples. A joint venture between NREL and Xcel Energy is combining wind power and hydrogen power in the same way in Colorado. Hydro in Newfoundland and Labrador are converting the current wind-diesel Power System on the remote island of Ramea into a Wind-Hydrogen Hybrid Power Systems facility. A similar pilot project on Stuart Island uses solar power, instead of wind power, to generate electricity. When excess electricity is available after the batteries are full, hydrogen is generated by electrolysis and stored for later production of electricity by fuel cell.
The UK started a fuel cell pilot program in January 2004, the program ran two Fuel cell buses on route 25 in London until December 2005, and switched to route RV1 until January 2007. The Hydrogen Expedition is currently working to create a hydrogen fuel cell-powered ship and using it to circumnavigate the globe, as a way to demonstrate the capability of hydrogen fuel cells.
Western Australia’s Department of Planning and Infrastructure operated three Daimler Chrysler Citaro fuel cell buses as part of its Sustainable Transport Energy for Perth Fuel Cells Bus Trial in Perth. The buses were operated by Path Transit on regular Transperth public bus routes. The trial began in September 2004 and concluded in September 2007. The buses’ fuel cells used a proton exchange membrane system and were supplied with raw hydrogen from a BP refinery in Kwinana, south of Perth. The hydrogen was a byproduct of the refinery’s industrial process. The buses were refueled at a station in the northern Perth suburb of Malaga.
The United Nations Industrial Development Organization (UNIDO) and the Turkish Ministry of Energy and Natural Resources have signed in 2003 a $40 million trust fund agreement for the creation of the International Centre for Hydrogen Energy Technologies (UNIDO-ICHET) in Istanbul, which started operation in 2004. A hydrogen forklift, a hydrogen cart and a mobile house powered by renewable energies are being demonstrated in UNIDO-ICHET’s premises. An uninterruptible power supply system has been working since April 2009 in the headquarters of Istanbul Sea Buses company.
Hydrogen-using alternatives to a fully distributive hydrogen economy
Hydrogen is simply a method to store and transmit energy. Various alternative energy transmission and storage scenarios which begin with hydrogen production, but do not use it for all parts of the store and transmission infrastructure, may be more economic, in both near and far term. These include:
An alternative to gaseous hydrogen as an energy carrier is to bond it with nitrogen from the air to produce ammonia, which can be easily liquefied, transported, and used (directly or indirectly) as a clean and renewable fuel. For example, researchers at CSIRO in Australia in 2018 fuelled a Toyota Mirai and Hyundai Nexo with hydrogen separated from ammonia using a membrane technology.
Hydrogen production of greenhouse-neutral alcohol
The methanol economy is a synfuel production energy plan which may begin with hydrogen production. Hydrogen in a full “hydrogen economy” was initially suggested as a way to make renewable energy, in non-polluting form, available to automobiles. However, a theoretical alternative to address the same problem is to produce hydrogen centrally and immediately use it to make liquid fuels from a CO2 source. This would eliminate the requirement to transport and store the hydrogen. The source could be CO2 that is produced by fuel-burning power plants. In order to be greenhouse-neutral, the source for CO2 in such a plan would need to be from air, biomass, or other source of CO2 which is already in, or to be released into, the air. Direct methanol fuel cells are in commercial use, though as of August 2011 they are not efficient.
The electrical grid plus synthetic methanol fuel cells
Many of the hybrid strategies described above, using captive hydrogen to generate other more easily usable fuels, might be more effective than hydrogen-production alone. Short term energy storage (meaning the energy is used not long after it has been captured) may be best accomplished with battery or even ultracapacitor storage. Longer term energy storage (meaning the energy is used weeks or months after capture) may be better done with synthetic methane or alcohols, which can be stored indefinitely at relatively low cost, and even used directly in some type of fuel cells, for electric vehicles. These strategies dovetail well with the recent interest in Plug-in Hybrid Electric Vehicles, or PHEVs, which use a hybrid strategy of electrical and fuel storage for their energy needs. Hydrogen storage has been proposed by some to be optimal in a narrow range of energy storage time, probably somewhere between a few days and a few weeks. This range is subject to further narrowing with any improvements in battery technology. It is always possible that some kind of breakthrough in hydrogen storage or generation could occur, but this is unlikely given that the physical and chemical limitations of the technical choices are fairly well understood.
Captive hydrogen synthetic methane production (SNG synthetic natural gas)
In a similar way as with synthetic alcohol production, hydrogen can be used on site to directly (nonbiologically) produce greenhouse-neutral gaseous fuels. Thus, captive-hydrogen-mediated production of greenhouse-neutral methane has been proposed (note that this is the reverse of the present method of acquiring hydrogen from natural methane, but one that does not require ultimate burning and release of fossil fuel carbon). Captive hydrogen (and carbon dioxide from, for example, CCS (Carbon Capture & Storage)) may be used onsite to synthesize methane, using the Sabatier reaction. This is about 60% efficient, and with the round trip reducing to 20 to 36% depending on the method of fuel utilization. This is even lower than hydrogen, but the storage costs drop by at least a factor of 3, because of methane’s higher boiling point and higher energy density. Liquid methane has 3.2 times the energy density of liquid hydrogen and is easier to store compactly. Additionally, the pipe infrastructure (natural gas pipelines) are already in place. Natural-gas-powered vehicles already exist, and are known to be easier to adapt from existing internal engine technology, than internal combustion autos running directly on hydrogen. Experience with natural gas powered vehicles shows that methane storage is inexpensive, once one has accepted the cost of conversion to store the fuel. However, the cost of alcohol storage is even lower, so this technology would need to produce methane at a considerable savings with regard to alcohol production. Ultimate mature prices of fuels in the competing technologies are not presently known, but both are expected to offer substantial infrastructural savings over attempts to transport and use hydrogen directly.
It has been proposed in a hypothetical renewable energy dominated energy system to use the excess electricity generated by wind, solar photovoltaic, hydro, marine currents and others to produce hydrogen by electrolysis of water then combine it with CO2 make methane (natural gas). Hydrogen would firstly be used onsite in fuel cells (CHP) or for transportation due to its greater efficiency of production and then methane created which could then be injected into the existing gas network to generate electricity and heat on demand to overcome low points of renewable energy production. The process described would be to create hydrogen (which could partly be used directly in fuel cells) and the addition of carbon dioxide CO2 possibly from BECCS (Bio-Energy with Carbon Capture & Storage) via the (Sabatier reaction) to create methane as follows : CO2 + 4H2 → CH4 + 2H2O.
Source from Wikipedia