The future of the hydrogen economy: old promises and serious uncertainties
|février 29, 2016||Posté par Pierre Papon sous Articles||
Captain Nemo, the hero of Jules Verne’s novel “Twenty thousand leagues under the seas”, praised the performance of his submarine, the Nautilus, which was powered with a fuel cell working with hydrogen produced by electrolysis of seawater (cf. the figure representing the Nautilus and captain Nemo). At that time its principle has been known for fifty years (a first stack was built in 1839 by a Welsh lawyer W. Grove), but today the energy transition issue with the perspective of a « low carbon » energy has reopened discussion about hydrogen technology as Jeremy Rifkin, already did it ten years ago, in his book about the hydrogen economy, considering it as the energy vector that would replace oil when it would be exhausted.
Hydrogen has the advantage of being the simplest and lightest molecular compound (two hydrogen atoms), with remarkable energy properties which explains its interest. It is the energy carrier with the largest mass density (2.2 times more energy per kg than natural gas), 1kg of hydrogen being equivalent to 3.8 liters of gasoline (2.75 kg). However, it is a light gas liquefying only at very low temperatures (- 253 ° C) and with10-13 kWh / kg of energy to be liquefied. Moreover, its flammability range with air is very wide (5 times more than that of natural gas) and with very low ignition energy (14 times lower than that of natural gas). Finally, as it is very light it spreads very easily, especially in metals that can be weakened; its widespread use therefore requires significant safety regulations for storing and transport (storage in metal tanks can be dangerous ).
Hydrogen is a gas used in industry (particularly for the desulphurization of fuels, chemistry and electronics), but as it does not exist in nature (except in the gaseous emissions of certain deep marine subduction), it has to be produced by industrial processes. The main one (95%) is by « steam reforming » of hydrocarbons (mostly natural gas) with steam, it is also produced by electrolysis of water with electricity … which is often produced by thermal power plants. The yield of the electrolysis of alkaline water is on average only 70%. Production costs are quite different – by reforming it was in 2014, from 1.5 to 2.5 € / kg in 2014 (it depends on the price of natural gas and has been lowered since), or about € 50 per MWh of thermal heat – by water electrolysis (with an industrial price of 70 € / MWh for electricity), it would amount to a range of 4 to 8 € / kg more than double that by reforming. (France Strategy, Etienne Beeker, « Are there a place for hydrogen in the energy transition? » Notes No. 15, August 2014, www.strategie.gouv.fr ).
Hydrogen technology’s supporters put forward two possible perspectives: its use as a fuel in vehicles and for the electricity storage. Hydrogen can be used directly as fuel instead of gasoline in a conventional internal combustion engine. So BMW has commissioned in Berlin a fleet of cars, Hydrogen7, working with liquid hydrogen (a tank with 8 kg of liquid hydrogen for a 200 km autonomy). This solution has probably no future being uneconomical, the liquefaction of hydrogen consuming a third of the energy stored in the liquid and requires a cryogenic tank. In fact the vast majority of technical efforts are devoted to the electric motor using hydrogen in fuel cells. The first operational battery, besides that one of Captain Nemo, has been used in the 1960s by NASA on the Apollo flights. Schematically, a fuel cell is the opposite of a cell performing the electrolysis of water: an electric current is produced in a cell by recombining hydrogen and oxygen. The hydrogen gas introduced to the anode is ionized, the ions pass through an electrolyte consisting of a membrane (a fluorinated polymer Teflon type) impermeable to electrons, ions recombine at the cathode with oxygen from the air, the reaction produces water and heat. The flow of electrons can power an electric motor. This type of battery is called PEM (Polymer Electrolyte Membrane Fuel Cell, see figure Source: Nature, L.Schlapbach, vol 460, p 810, 2009..). They operate at temperatures close to room temperature (40-85 °C) but with the serious disadvantage of requiring the use of a noble metal catalyst in the electrodes, platinum being the most effective. A stack is formed by the assembly of several cells (200 for a power of 50 kW) and its efficiency is excellent, being 80 to 90% (in nominal operating conditions of an engine at lower power it is only about 55%), it powers an electric motor whose performance is close to 90%. The use of platinum as a catalyst is the critical point for these batteries because it is a rare and therefore expensive metal (28 € / g) and they must operate with hydrogen free of impurities (including carbon monoxide which can be a residue of the production of hydrogen) that can poison the catalyst.
A variant of PEM fuel cells is to replace the hydrogen by methanol, which can be converted into hydrogen in the stack. It allows liquid storage which is an advantage, but methanol is highly toxic … Other variants consist in operating with either liquid electrolytes (potassium, molten carbonate or phosphoric acid) but at higher temperatures (500-600 °C), the liquids being corrosive, or with solid electrolytes (a ceramic zirconia doped with yttrium) operated again at elevated temperature (600-1000 °C), but with a catalyst less expensive than platinum (a cerium and ruthenium oxide) and an efficiency about 40 to 50%. Recent work (Gorte, RJ, « Cooling down ceramic fuel cells », Science, vol. 394, issue 6254, p. 1290, 18 September 2015, www.sciencemag.org ) allows one to consider a ceramic cell with good efficiency in a temperature range of 250-550 ° C, operating with hydrogen being directly produced inside the stack. Batteries operating at high temperatures are more suitable for stationary installations for electricity production.
The use of hydrogen with a fuel cell in a vehicle requires getting overcoming several technical obstacles (see Pierre Papon, « Hydrogen energy sector between promises and Uncertainties » Futuribles Vigie, No. 186, January 5, 2016 , www.futuribles.com ). The first one being its production .The mitigation of climate change policy, supposes that one excludes hydrocarbons as a primary energy to operate with a non-carbon resource, a water electrolysis process with electricity generated by renewable or nuclear energies only remains, which would represent substantial investment (the efficiency of electrolysis should also be increased). For uses related to electric vehicles, it will be necessary to build infrastructures for the distribution and storage of hydrogen in service stations to distribute pressurized hydrogen to vehicles (compression consumes 10% of the energy); the company Air Liquide has embarked into this venture (see picture). As it seems difficult to build a national network of gas pipelines, gas stations will have to be built and being supplied by trucks with cryogenic reservoirs or locally or producing hydrogen in situ by water electrolysis and stored in a pressurized reservoir, solutions that are likely to be costly. The storage of hydrogen in an electric vehicle is a second problem. Liquid storage being uneconomical, the preferred solution is to store it under 700 bars pressure in a tank made of composite material (storing 5 kg of hydrogen at this pressure in a 125 liter tank with a weight of about 100 kg). Stringent security requirements must be met to prevent leaks (hydrogen is very light, it quickly diffuses into the air but it is very flammable). Finally, the main obstacle to the widespread use of cells operating with platinum will be, in addition to its cost, its availability: selling on the market 10 million cars / year on the market would represent, for example, in the present state of the art, a 150 tons / year of platinum demand, equivalent to three-quarters of world production (80% by South Africa) which is unrealistic, although recycling is obviously possible. These are four very serious obstacles.
Research has opened, it is true, new tracks. Storing hydrogen in a solid metal is an alternative to gas storage. Excluding palladium, an expensive metal, a metal hydride can be formed (hydrogen is absorbed by a metal), magnesium hydride is the most favorable material (it stores 7.6 g of hydrogen in 100 g of hydride in the form of wafers). This process is developed by the McPhy Energy company in Grenoble (a start-up created by the Cea and Cnrs). This technique is well suited for fixed installations to supply fuel cells producing electricity to a building or integrated into an electricity storage system associated to an intermittent source. To avoid the use of platinum, nickel organic compounds have been considered (by teams of Cea, the CNRS and the College de France), but their performance is still lower than that of cells operating with platinum. The use of platinum in the form of nanoparticles incorporated into the porous electrodes (made of graphite, graphene being considered) would reduce the weight of cells and one envisage descending today to about 0.3 g / kW (i.e. 15 g of platinum for a 50 kW battery). One can also consider new ways of producing hydrogen. A conventional method would be to degrade biomass in the presence of steam with solar heat (in an oven) in the presence of a catalyst. An alternative would be to achieve artificial photosynthesis assisted by synthetic catalysts, hydrogen is produced from water vapor with solar cells by performing photolysis. Electrons extracted from the silicon by the light can be used in water molecules electrochemical decomposition reaction in the presence of inexpensive catalysts with good efficiency. Work is performed in this area at MIT and by teams of Cea, the University Joseph Fourier and CNRS in Grenoble (Marc Fontecave) who synthesized metal sulfur compounds (based on nickel, and iron) with good catalytic activity. The main difficulty with these techniques being the loss of efficiency of the catalysts over time. Recent studies show that a catalyst made of titanium oxide nanotubes allows good efficiency (JB. Sambur et al. « Sub-particle and photocurrent maping to optimize catalyst-modified photoanodes » Nature, No 530, p. 77, February 2016, www.nature.com ). These are very forward-looking techniques that can create opportunities for solar chemistry.
The prospect for the hydrogen energy vector is mixed. Several automotive builders sell vehicles with PEM fuel cells platinum (experimental bus ran at Barcelona and Berlin), this is the case, in particular, with the Japanese and the German industry which propose prototype vehicles fleets, some being already on the market. The hydrogen car was thus the star of the Tokyo Motor Show, held in November 2015, which has highlighted the Japanese automotive industry ambitions in electric motorization including fuel cells. Toyota sells since 2014 a model of this type, Mirai, quickly rechargeable (sold in 2015 at a price of € 56 000), it expects to produce 20 000 cars of this type in 2020. The Japanese government, too, has a great ambition for the hydrogen industry and wants to make the Tokyo Olympics in 2020 a showcase for promoting Japanese technology (see Embassy of France in Japan, Department of science and technology, « hydrogen energy for the Tokyo Olympics in 2020, « Antoine Saporta and Sebastian Codina, in August 2015, www.ambafrance-jp.org ). Thus 35 service stations will be built in Tokyo to supply hydrogen cars circulating during the Olympics (a subvention of 22,000 € is granted for the purchase of a car!). Japanese voluntarism is probably explained by the will to rebalance the energy mix in Japan after Fukushima, renewable energy development for electricity production being one option. Observe that in France, the Rhône-Alpes-Auvergne region launched the project HyWay with the support of ADEME (Agency for energy efficiency), which aims to promote electric vehicles with a hydrogen fuel cell commercial vehicle, the Kangoo ZE H2, with a battery built by the Grenoble Symbio FCELL SMEs (it extends the life of its conventional battery). The cost of batteries is an important factor for the future of these vehicles. The US Department of Energy (DOE) has thus estimated in 2015 this cost at $ 280 / kW for a series of 20 000 units (or $ 14,000 for a 50 kW battery, DOE Hydrogen and Fuel cells program record, Fuell cell system cost, September 2014), this cost would fall to $ 55 / kW for a 500 000 units production, it would be nearly € 25,000 for the Kangoo.
The economic viability of hydrogen cars is far from being assured …. All the more according France Strategie, the price to the « pump » of an hydrogen kg would be in the range of 10 to 13 € per kg, that is at least three times the price of conventional fuels. Fuel cells for electric vehicles will be in direct competition with batteries which they certainly need technical progress to improve power density (150 Wh / kg today) and lower their cost but the construction of electric terminal for charging pose fewer technical problems and can be performed virtually anywhere, including in private car parks (excluding a solution for hydrogen charging). This issued was already existing for electric cars in 1900, their lead batteries had very little autonomy and, in the absence at the time of an electrical network covering a whole country, including the United States, the car gasoline engine which is easier to refuel has quickly imposed itself.
The « storage » of electricity, a key point for the energy transition, is possible using fuel cell techniques, it would use electricity generated by intermittent renewable sources (photovoltaic solar or wind powers for example) to produce hydrogen by water electrolysis, stored hydrogen would feed in a fuel cell to produce electricity, battery performance and durability will be key factors. This is an interesting option for hydrogen technology because batteries operating at high temperature could be used in permanent installations (thus avoiding the use of platinum as a catalyst) with a good yield and with storage in solid form in hydrides for example, which poses fewer security problems than high pressure. The experimental solar plant Myrte in the vicinity of Ajaccio in Corsica, carried out by the Cea and Areva, in cooperation with Corte University and Cnrs, is testing this storage solution (solar electricity is used to produce hydrogen by water electrolysis during the day which is stored, it can power a 100 kW fuel cell to restore night electricity).
The transition to a hydrogen economy is very uncertain because there are still many scientific barriers and technical and economic obstacles to overcome in order that hydrogen can establish itself as a credible energy vector, especially in transport. In a recent article, researchers at the JRC (EU Joint Research Centre), believe however that the role of hydrogen in a « decarbonised » Europe could assert in 2050: with 5 to 6% of hydrogen in final energy in industry and transport. It would thus be at best a niche but they express doubts about a possible a stationary batteries breakthrough (Sgobbi A. et al. « How far away is hydrogen? Its role in the medium and long term decarbonisation of the European energy system », International Journal of energy system, 41, p. 19, 2016, www.sciencedirect.com ). It is clear that research progress has been relatively slow since the early success of fuel cells in space and do not perform real breakthrough to overcome the platinum use obstacle. A large-scale use of hydrogen fuel cells in vehicles is problematic because it requires the implementation of a comprehensive technical and complex system that goes from the production of hydrogen to its use in a motor and whose total cost is never worked out. However, this solution may be more easily adapted to electricity storage. Ultimately battery R & D should remain as a true priority for research strategies. Captain Nemo can still consider that its fuel cell is a way forward and is probably destined to remain so.