The Problem of Energy.
The world population of 7 billion humans uses energy at a mean rate of 16 TW (16 terawatts). This amounts to an annual 504 EJ (504 exajoules = 5.04 x 1020 J), and is provided from fossil fuels (oil, gas and coal), plus nuclear and hydro (hydroelectric) power along with all other forms of renewable energy1. The breakdown of these various contributions is given in Table 1. Around one third of the total energy used by humans on Earth is provided by crude oil, and close to one quarter each by natural gas and coal, although the amount of coal being burned is rising, particularly in China. Nuclear and hydro-power each contribute around 6% of the total energy mix, while the combination of renewable energy from all sources, wind, wave, geothermal, wood, solar etc. amounts to just a little above 1%. In 2004, humanity used 471 EJ of energy2, and while the relative proportion of each contributing energy source has remained modestly constant, it is clear that at a rise to 504 EJ in 2010, the demand placed upon these energy sources is rising relentlessly. This follows not merely a growing human population per se, but an increasingly affluent industrialised consumer society. We need to consider two aspects: firstly, that the CO2 produced by burning fossil fuels is believed to contribute to global-warming and this may lead to unwelcome or even catastrophic changes to the global climate3. Secondly, and more immediately, the fossil fuels and uranium too (for nuclear power) are in finite supply, and there is compelling evidence that each source will meet its own production peak within the next two decades. Most vulnerable appears to be crude oil (petroleum), world supplies of which are predicted to peak (“peak oil”) probably during the next 5 years4. Irrespective of the exact timing of peak oil, there are salient predictions (see below) that a gap will emerge in the supply of oil against demand for it, from the end of this year (2012), rising to a shortfall of 10 million barrels a day by 20155. This situation has been termed “gap oil”6, and can only be exacerbated by peak oil, when supply must draw-down against rising demand, thus enlarging the gap from both sides. Thus, in order to curb carbon-emissions and to extend limited resources, alternative and ideally renewable sources of energy are needed.
Liquid Fuels and Transportation.
A simple comparison of the energy content delivered from different energy sources, as is made in Table 1, is somewhat misleading, since it seems to imply that if the production of one of them begins to fail, it can be readily substituted by another. The issue of transportation is a singular example where this is not the case, since practically all the vehicles used in the world – cars, lorries, buses, trains, ships and planes - have been engineered to run on liquid fuels that are refined from crude oil. Therefore the likely consequences of “peak oil”, with dwindling supplies and escalating costs of liquid fuels, are very serious. Ward's, the U.S. based publisher, estimated that as of 2010 there were 1.015 billion motor vehicles in use in the world7. This figure represents the number of cars, light, medium and heavy duty trucks, and buses, but does not include off-road vehicles or heavy construction equipment. Between 1950 and 1970, the world vehicle population doubled roughly every 10 years, passing the 250 million mark in 1970, and exceeding 500 million in 1986. It has been estimated that the world's road transportation fleet will reach 2 billion by 2020, of which at least 50% will be cars. China’s and India’s automobile fleets are expected to grow at an annual rate of around 7 or 8%, while in the United States, it will be under 1% a year, and around 1 to 2% in Western Europe, but this depends tacitly on finding an expanding liquid fuel supply, and it is this which is at issue. Indeed, the International Energy Agency (IEA) has issued a report8 to the effect that a shortfall in oil production of 64 million barrels a day (mbd) can be expected by 2030, which represents a loss of 62% of the world supply of conventional crude oil, currently 84 mbd, assuming a demand by 2030 of 96 mbd, a figure significantly downgraded from prior estimates by the IEA of 120 - 130 mbd. At a mean decline rate of 2.9 mbd/year (-3.4%/year) this value accords closely with the prediction in a recent U.S. Army report5 that there will be a deficiency of 10 mbd by 2015, following a loss of any spare capacity for crude oil against demand for it by the end of this year (2012).
While it is possible to run cars and other road vehicles on electricity, provided either from batteries or hydrogen/fuel cells, actually converting their number substantially to these alternative energy carriers (neither electrons nor hydrogen being primary fuels, i.e. they must be created from primary sources) would be such a considerable undertaking that the scheme is not feasible. Vehicles can be adapted to run on gas but a peak in natural gas production is expected within twenty years, following oil, and converting them all would take many decades, so this is no solution either. It is, therefore, a new source of liquid fuels that must be sought, since they would be far more compatible with a transportation fleet and distribution infrastructure designed for liquid petroleum fuels. These, ideally, should be “carbon-neutral” in order to reduce carbon emissions, a condition which certainly does not apply4 to coal-to-liquids processing (with probably twice the carbon emissions overall that are incurred in the production and burning of diesel or petrol derived from petroleum), nor gas-to-liquids either, unless the CO2 is captured and stored in some way that prevents it escaping into the atmosphere. Biofuels are attended by a number of vexed issues6: the competition for arable land between growing crops for fuel or crops for food, the increased amount of freshwater required to grow fuel crops above that already needed for agriculture, the clearing of rainforest to produce high-energy fuel-crops, e.g. palm-oil, and the fundamental EROEI which for bioethanol may only marginally exceed the overall energy costs of its production, or in some cases not quite break-even. Clearly, some other strategy is necessary.
Table 1. Fuel type: Average power in TW (1012 W) Energy/year in EJ (1018 J) (2010 figures), (Data from reference 1).
Oil: 5.4 TW (169 EJ)
Gas: 3.8 TW (120 EJ)
Coal: 4.7 TW (149 EJ)
Hydroelectric: 0.8 TW (26 EJ)
Nuclear: 1.0 TW (33 EJ)
Geothermal, wind, solar, wood: 0.2 TW (7 EJ).
Total: 16.0 TW (504 EJ) (2010).
Figure 1 summarises the quantity and fate of solar radiation striking the top of the earth’s atmosphere2. We see that 52 PW (1015 W) is reflected back into space (i.e. 30% of the total). Thus, in outer space, there is more solar energy available to be collected, which has prompted potential schemes to launch photovoltaic arrays into space on satellites2, with which to capture the sun’s energy and then beam it back to earth in the form of microwaves for terrestrial applications. At the top of the atmosphere, with the sun directly overhead, the radiation flux provides around 1.4 kW/m2 of energy, the “solar constant”2. Since the total amount of energy1 (oil, gas, coal, nuclear, hydro, everything) used on earth by humans amounts to a power of 16 TW, at 174 PW, the amount of radiation striking the exposed hemisphere of the earth is well over 10,000 times that. So if we could capture even a small amount of this bounty, and convert it into useable energy, the imminent energy crisis could be averted. Since the production of solar energy (and renewable energy generally, other than hydropower) has an intermittent quality – the sun doesn’t always shine, or not to the same extent throughout the day, and not at all at night – some method of storing it is essential to its use as a serious energy source, to supplant fossil fuels, and the creation by its means of a chemical fuel (solar fuel) would be the best means to achieve this. Furthermore, and as noted earlier, if liquid fuels could be produced in quantity, they would be entirely compatible with the present world transportation fleet based on liquid petroleum fuels, and a distribution network designed to supply the latter.
Means for capturing solar energy2.
Solar energy can be captured by various means, which boil-down to either collecting heat directly from the sun’s rays, or using the wavelengths of the solar spectrum to grow biomass through photosynthesis, or to excite electrons in photovoltaic materials to produce electricity.
(1) Direct heating systems2 à on-roof water heating systems; solar furnaces; concentrating solar thermal power (CSTP) plants etc.
(2) Photosynthesis9 (PS) à creates biomass, and a total of 200 EJ of fuel (Shell estimate) could be produced by hydrothermal conversion. PS is 12% efficient as a theoretical maximum, but most plants give 0.1 - 6%. Growing biofuel crops also suffers from competition with food crops for fertile land, hence if we turned all the available arable land in the U.K. over to biofuel crops (i.e. grew no food at all) we could only match ca 17% of our fuel from rapeseed/biodiesel or ca 50% from sugar beet/ethanol as is currently made from crude oil. These estimates assume that all vehicles are first converted to diesel engines, which are more efficient in terms of tank-to-wheel miles than spark-ignition engines, which burn petrol, by about 40%.
(3) Photovoltaics2,10. In essence, light photons knock electrons into higher energy states in semi-conductor materials such as silicon and cause conduction (electron diffusion), generating electricity. The photo-active material needs a band-gap of < 3.2 eV (i.e. at the edge of the visible spectrum, 400 nm), and ideally down to around 1.0 eV (1250 nm) - which is the “near” IR region. The actual recovered efficiencies for single-junction silicon cells are now approaching the theoretical efficiency of 33%, which is way ahead of that for photosynthesis (6%). In principle, quantum dot (QD) cells might achieve an efficiency of 42% or even 65% as a result of multiple exciton generation2, although the record to date is a rather more modest 6%11.
Fuels Made Using Concentrating Solar Thermal Power12,13.
Those “fuels” that have been most pursued using various solar methods are H2 and CO, and while each may indeed be used either in fuel cells or combustion engines, a mixture of them (known as synthesis gas or syngas), can be converted to liquid fuels, e.g. methanol, or hydrocarbons, using the Fischer Tropsch process4. The latter involves a series of chemical reactions that lead to a mixture of hydrocarbons (CnH(2n+2)) (equation 1):
(2n+1) H2 + n CO → CnH(2n+2) + n H2O (1),
where 'n' is a positive integer. It is mainly straight-chain alkanes that are formed and so the product is quite suitable as a diesel fuel. As we now discuss, there are various routes to forming H2, but CO can be formed essentially either through steam reforming4 carbon-containing materials (methane, oil, coal, biomass) – which also yields H2 - or by dry reforming14 with CO2. Now the latter lends the possibility that CO2, either as a product of steam reforming natural gas or as captured from power stations, might be intercepted from being emitted into the atmosphere and instead converted to useful fuels, in substitution for those presently refined from petroleum.
H2 by direct thermolysis of H2O.
Although it is conceptually the simplest process, the thermal dissociation of H2O to
H2 and O2 is difficult to achieve in practice, since very high temperatures are required to attain a reasonable degree of dissociation (e.g. 2725°C for 64% dissociation at atmospheric pressure12), and it is necessary to separate the H2 from O2 to obviate the build-up of an explosive mixture. Ceramic membranes, such as those made from zirconia, have been tested but even these usually fail to withstand the thermal shocks incurred when exposed to high-flux solar radiation. Rapid-quench methods, in which a cold gas is expanded in a nozzle, or a solar-irradiated target is immersed in liquid water, are straightforward but cause a fall in the energy efficiency which is compounded by significant energy losses by radiation from the extremely hot reactor. Furthermore, the gas mixture is explosive. There are water-splitting multi-step thermochemical cycles which may be performed at significantly lower temperatures (below 950 oC), expected to be available in the future either from concentrating solar thermal power or from very high temperature nuclear reactors (VHTR), and which further avoid the necessity to separate H2 from O2.The leading candidates are a three-step sulphur iodine cycle (equations 2 - 4)13 based on the thermal decomposition of sulphuric acid at 850°C and a four-step UT-3 cycle (equations 5 - 8)13 based on the hydrolysis of calcium bromide and iron bromide at 750°C and 600°C, respectively. It is proposed that, in the longer run, a complete substitution of fossil fuels by solar H2 might be achieved, while the decarbonization of fossil fuels in the medium term creates a link between today’s fossil-fuel-based technology and the solar chemical technology of tomorrow.
2H2SO4 → 2SO2 + 2H2O +O2 850 oC (2)
2HI → H2 + I2 300 oC (3)
I2 + SO2 + 2H2O → 2HI + H2SO4 100 oC (4)
2Br2 + 2CaO → 2CaBr2 + O2 600 oC (5)
3FeBr2 + 4H2O → Fe3O4 + 6HBr + H2 600 oC (6)
CaBr2 + H2O → CaO + 2HBr 750 oC (7)
Fe3O4 + 8HBr → Br2 + 3FeBr2 + 4H2O 300 oC (8)
H2 by decarbonization of fossil fuels.
There are principally three solar thermochemical processes for H2 production using fossil fuels, namely: cracking, reforming, and gasification. In solar cracking, natural gas, oil, or other hydrocarbons are decomposed thermally into H2 and carbon. The carbon can either be sequestered to avoid the release of CO2, or used for other purposes. Steam-reforming of natural gas or oil, and steam-gasification of coal and other solid carbonaceous materials furnishes syngas, which may be converted to liquid hydrocarbon fuels, vide supra. The chemical properties of syngas accord principally to its differing content of H2, CO, and CO2. The CO content can be reduced in favour of H2 via a catalysed reaction with steam, known as the water-gas shift reaction (equation 9):
CO(g) + H2O(g) → CO2(g) + H2(g) (9)
CH4(g) + CO2(g) → 2CO(g) + 2H2(g) (10)
The product, CO2, can be separated from H2, e.g. by using the pressure-swing adsorption technique, which is a method for the separation of a gas from a mixture of gases under pressure, according to particular molecular characteristics and affinity for an adsorbent material (e.g. through differential molecular electric quadrupole moments). Syngas may also be produced by the solar dry reforming of CH4 with CO2 (equation 10) and transported to locations to be used as a fuel directly. Stored solar energy is released by the reverse exothermic reaction in the form of heat, which can be used for generating electricity via a Rankine cycle. [The Rankine cycle generates about 90% of all electrical power used throughout the world, including that from virtually all solar thermal, biomass, fossil fuel and nuclear power plants. Heat is supplied externally to a closed loop, which normally uses water. The cycle also provides the fundamental thermodynamic basis of the steam engine]. Similarly, ammonia (NH3) can be used in a chemical heat pipe to store and transport solar energy. In one system developed by the Australian National University (ANU), ammonia is dissociated in an energy storing (endothermic) chemical reactor as it absorbs solar thermal energy. At some subsequent place and time, the reaction products of H2 and nitrogen (N2) undergo an exothermic reaction in which ammonia is re-formed, releasing heat. An industrial demonstration plant has been announced using four of ANU’s 400 m2 parabolic dishes15.
Solar steam gasification.
The steam-gasification of coal and oil shale has been investigated using concentrated solar energy. A conceptual design of a solar reactor for the gasification of carbonaceous materials was created using optical fibres to direct the solar radiation into the reaction chamber. In the project SYNPET15 (2003–2009), a solar thermal technology for the steam-gasification of petcoke particles was developed, using a 10 kWth (th = thermal) solar reactor, directly exposed to concentrated solar radiation, with a continuous gas-particle vortex flow confined to a cavity receiver. A scale-up of the vortex solar reactor to 500 kWth at the Plataforma Solar de Almería is currently being undertaken15.
Catalysed thermal methods17 for the production of H2, CO and CH4 from H2O and CO2.
The cerium(IV) oxide–cerium(III) oxide cycle (CeO2/Ce2O3 cycle) is a two-step thermochemical process based on cerium(IV) oxide/cerium(III) oxide, and is normally employed for hydrogen production. One advantage of this approach is that the H2 and O2 are generated in two distinct steps, and so it is unnecessary to separate the components of a high-temperature gas. The process constitutes a redox system:
Dissociation: 2CeO2 → Ce2O3 + 0.5 O2 (11)
Hydrolysis: Ce2O3 + H2O → 2CeO2 + H2 (12)
In the first step (equation 11), which is endothermic, cerium(IV) oxide is dissociated thermally, under an inert gas atmosphere at high temperatures, into cerium(III) oxide and oxygen. In the second step (equation 12), which is exothermic, cerium(III) oxide reacts at lower temperatures with water to produce hydrogen and regenerate cerium(IV) oxide. The strategy is also applicable to the dissociation of CO2 (equations 13 - 14):
Dissociation: 2CeO2 → Ce2O3 + 0.5 O2 (13)
Hydrolysis: Ce2O3 + CO2 → 2CeO2 + CO (14)
If both H2O and CO2 are fed into a suitable solar thermal reactor containing a ceria catalyst, which is cycled within the temperature range, 800 oC - 1500 oC, syngas is produced. Although the yields are low, with only around 0.7 - 0.8% of the solar thermal energy being harnessed by the fuel, it is thought this is only a limitation of the scale and design of the system rather than of the underlying chemistry. Most of the energy is lost as heat through the wall of the reactor, or by the re-radiation of sunlight back through the aperture of the device. But the researchers are confident that efficiency rates of up to 19% can be achieved through better insulation and smaller apertures. Such efficiency rates, they say, could make for a viable commercial device. It is worth noting that of various oxide materials that might be employed as redox catalysts for splitting water or carbon dioxide, ferrite-based oxides show fairly slow reaction rates, whereas oxides such as ZnO and SnO2 sublime during the decomposition stage and require rapid quenching of gaseous products to avoid recombination. In contrast, ceria is stable to volatility and to sintering (which reduces the surface area and activity), and shows relatively rapid reaction rates. This is thought to be due partly to the presence of nonstoichiometric oxidized and reduced phases, a high oxygen diffusion rate, and to its thermal structural stability. By impregnating a samarium/ceria catalyst with nickel11, methane was produced, shown to arise from catalytic hydrogenation of carbon particles formed on the surface of the metal by decomposition of CO2.
Fuels Produced by “Artificial Photosynthesis”.
I have some issue with the accuracy of the term “Artificial Photosynthesis”17, but it is snappy and sounds very “green” so I think it will stick. As a reminder, natural photosynthesis9 is a “water-splitting” process by which green plants and green algae fix CO2 from the atmosphere to build carbohydrate, and grow. The rate of energy capture by photosynthesis is immense, at approximately 100 TW, which is about six times the entire power consumption of human civilization. Photosynthesis is also the sole source of carbon in all life on Earth, and converts around 370 billion tonnes of CO2 into 250 billion tonnes of biomass each year, in which is stored almost 4,000 EJ worth of energy. The counterpart product is O2, and hence photosynthesis is the origin of virtually all atmospheric oxygen, with a roughly 50:50 contribution made by land-based plants and by oceanic phytoplankton. The process may be summarised as follows:
nCO2 + nH2O + hν → (CH2O)n + nO2
Production of ONE O2 molecule requires the transfer of FOUR electrons:
2H2O → O2 + 4e- + 4H+
And FOUR electrons are required to reduce ONE CO2 molecule:
CO2 + 4e- + 4H+ → (CH2O) + H2O
The energy resulting from light adsorption by the chlorophyll photocatalyst is transferred to a manganese-protein complex called “Photosystem II”, which oxidises water:
H2O – e- → [H2O+•] → HO• + H+
2 HO• → H2O2
H2O2 – e- → [H2O2+•] → HOO• + H+
2 HOO• → O2 + H2O2
In a broad analogy with this, the label “artificial photosynthesis” is commonly used to describe any scheme for capturing and storing the energy from sunlight in the chemical bonds of a material that might be used as a fuel (a solar fuel). Photocatalytic water splitting converts water into protons (leading to H2) and O2, and provides a major research topic in the field. Another important area is “light-driven carbon dioxide reduction” which aims to replicate carbon fixation by natural photosynthesis, and might provide a carbon-mitigation strategy, along with the creation of fuels independent of petroleum and natural gas. In the broad classification of “artificial photosynthesis”, accepting that light can be harvested using external PV cells and converted to electrons, we give due mention to electrochemical processes, which may be photo-assisted or direct. Thus, although the primary reduction product of CO2 is CO, a direct photoelectrochemical conversion of carbon dioxide and water to methanol using a p-type semiconductor (GaP) electrode has been reported with faradaic efficiencies of 88 – 100%. The process appears highly selective, since other reduction products such as formic acid and formaldehyde were not detected, nor hydrogen.18 The direct electrochemical reduction of CO2 has mainly been thwarted by the impractically high overpotentials required to drive the process. In contrast, a direct conversion of CO2 to CO has been achieved at overpotentials of less than 0.2 volt. The medium uses an ionic liquid (1-ethyl-3-methylimidazolium tetrafluoroborate, EMIM-BF4) as its electrolyte, which is thought to lower the energy of the CO2−• intermediate through the formation of a complex of the type “EMIM+-CO2−•”.19 As pointed out previously20, for ionic liquid applications to be used in earnest, they would need to be synthesised on a large scale, and from molecules derived from crude oil. This case is no exception, if it is to provide a source of syngas and hence liquid hydrocarbon fuels, in a quantity that in any way matches the expected loss of petroleum and fuels derived from it. Dye-sensitized solar cells2 (DSSC) have been used in H2 production. A DSSC is a kind of thin-film cell in which a semiconductor, normally TiO2, along with a coating of an organic or inorganic dye to act as a photosensitizer is coated on the anode, while a platinum catalyst is present at the cathode, where H2 is generated. Thin-film cells offer the considerable advantage2,10 over conventional solar cells that perhaps only 1/100th the amount of conductive material is required in their fabrication, and is thought to be critical both to the rate and scale at which PV technology can be implemented. TiO2 is the photocatalytic semiconductor material most studied as a potential solar water-splitting catalyst, and yet despite 4 decades of research the process remains inefficient and economically unsound, due mainly to rapid recombination of holes and electrons, the rapid reconversion of oxygen and hydrogen back to water and poor activation of TiO2 by visible light. The hole-charge recombination can be inhibited by loading the particles at their surface with metals, most commonly platinum, but its limited world supply and high cost imposes a severe limitation on the technology, which is has to be said remains rather inefficient in any case. The implantation of high-energy transition metal ions (accelerated by high voltage) has been shown to modify the electronic structure of TiO2, so that its photo-response is shifted into the visible region (up to 600 nm), which may prove useful in the development of the “second generation photocatalyst”21.
Nocera and his co-workers have developed what they describe as an “artificial leaf”22. In the overall strategy, sunlight can be captured and converted to electricity using a silicon solar PV cell, but the crucial advance lies in the design of the electrode surface to provide a highly efficient anode electrocatalyst (Oxygen Evolving Electrode, OEE) for use in the electrolysis of water employing inexpensive materials. An indium tin oxide (ITO) electrode was immersed in water containing Co2+ cations and potassium phosphate (Pi). By application of a voltage to the electrode, cobalt, potassium, and phosphate accumulated on its surface, to form the catalytically active phase, where water is oxidised yielding O2 and protons18. By reduction of these protons, H2 might result, using a suitable cathode, which originally was made from platinum. During the process, the cobalt-based phase decomposes, but is regenerated (“self-heals”) by cobalt, potassium and phosphate being adsorbed from the solution. To reduce the protons to H2, a cathode made form cheap, readily available (“Earth Abundant”) materials rather than precious metals is required, if a serious scale-up of the technology is to be feasible. To this end, a Co/borate catalyst was deposited on the surface of a triple-junction, amorphous silicon PV semiconductor (which was in contact with a stainless steel plate, to act as the anode (OEE), while a ternary Ni/Mo/Zn alloy was employed as the cathode (Hydrogen Evolving Electrode, HEE)23. The silicon was passivated (protected) by an ITO layer, which protects it from reactive oxidising species formed at the anode, e.g. HO• radicals. The two electrodes were employed in two configurations: (1) in which they were connected by a wire, and (2) in a wireless mode, where the Ni/Mo/Zn was deposited directly onto the stainless steel backing of the silicon wafer. The efficiencies of the two devices are 4.7% for a wired configuration (1) and 2.5% for the wireless arrangement (2). It is thought that the efficiency of (2) is reduced by the relatively greater distance that the protons must travel to the cathode from the front face of the anode, which imposes substantial ohmic losses in the wireless cell, in comparison with the 1 mm gap between the two electrodes in the wired cell (1). It is proposed that this technology could be adapted from a panel geometry to one based on (nano)particles free in solution/suspension22. However, in the in situ arrangements described, as opposed to standard electrolysers where H2 and O2 are generated in different chambers of the device, explosive mixtures of H2/O2 would be produced, and need to be separated probably using some form of membrane technology.
Nocera envisages a widescale future application of such devices for energy production say at the level of individual homes, with H2 acting as an energy storage medium so that power can be provided even at night (using a fuel cell) when the sun is not shining. He sees the most fruitful regions of the world, for the technology to be developed, as the “non-legacy”(i.e. developing) nations, which are less entrenched by tradition and vested interest in centralised, large-scale power production from fossil fuels and nuclear, than is the case in the “legacy” (developed) nations. Thus, remote communities could be provided with electricity at a local level. In October 2010, Nocera signed with the Tata Group of India to commercialize his research, a country in which there could be great demand for such decentralized, “personal” energy production, which he sees as key to the future of humanity. As we note later, due to the failing supply of crude oil and the absence of other fuels on a matching scale to run the global transportation network, a relocalisation of human civilization and its societies appears inevitable, as access to cheap and extensive personalised transport (mainly cars) will be prohibited both by cost and actual fuel shortages. In such localised communities, small-scale power generation will be necessary, and perhaps Nocera’s invention (or something like it) might contribute to a de-industrializing West, along with sustaining the existing non-legacy nations. However, many daunting challenges remain, including the instability of the cells which lose their activity over periods of hours and that there is a world shortage of indium24 (for ITO), which it is thought may run-out within 5 years, due to other pre-existing demands on it, e.g. for thin-film solar cells and light-emitting diodes (LEDs). Thus, any substantive technology is probably many decades away, along with the putative hydrogen economy itself, if that will ever arise in the full-ascension that some envisage. Meanwhile, we will need to address a more immediate and drastic failing of our energy supply, particularly that from crude oil and natural gas, most likely by transforming from a global village to a globe of villages.
Liquid Solar Fuels Through Industrial Photosynthesis25.
It has been estimated that to match the entirety of 16 TW of energy, as used by humans on Earth, in the form of hydrogen would require splitting some 18 billion tonnes (18 km3) of water annually, using photoelectrochemical methods or other means, e.g. concentrating solar thermal power, all as yet to be fully developed let alone proliferated on this scale. However, there is an established solar technology already available to us, namely photosynthesis. The basis of photosynthesis, as it occurs in Nature, is outlined in the previous section, which captures solar-energy at a rate of around 100 TW. Of the resulting 4,000 EJ worth of stored energy, clearly only a fairly small proportion might be sensibly harvested, and a deliberate growing of particular energy crops is necessary, much as we grow crops for food, while allowing animals to graze on available common land and set-aside pasture. There is, as already noted, competition between the use of arable land for food crops or fuel crops, and to place this into perspective, if in the UK we turned over all our crop-land to growing sugar beet for bioethanol production, and grew no food at all, we might match just half of our national fuel demand as is currently met from crude oil. If we chose rapeseed for the same purpose, a mere one sixth of that demand could be supplied in the form of biodiesel. There is the further issue of the freshwater demand, of which agriculture already struggles to secure enough to meet its needs, and in a sustainable picture of the future, supplies of water appear uncertain against the countenance of climate change. It is in the light of these considerations that algae/algal fuels have begun to look very appealing6, especially given the claimed very high yields that can be obtained per hectare as compared say with rapeseed and biodiesel. From a survey26 of the results from different studies, it has been estimated that between 40 – 90 tonnes of biodiesel/hectare can be produced from algae. However, the yields reported have often been extrapolated from growing algae over far smaller areas than a hectare, and there seem to be problems encountered when actual scale-ups are attempted. To date, no one has yet succeeded in producing fuel commercially and at scale, and indeed many small firms that started out to do this have stopped trading, one of which being the MIT spin-off Greenfuel Technologies which closed in 2009 after receiving £44 million ($70 million) of investment to build its own mini-algae plant. The algae programme of the UK’s Carbon Trust has been scrapped in the wake of the government austerity measures, as it attempts to cut the massive debt incurred in having to bale-out the banks after the 2008 economic crash. One problem is that the algae cannot be grown very densely, because those closest to the surface screen the sunlight from reaching those below. Another difficulty is to actually extract the oil from algae, and it appears that producing highly oil-yielding strains en mass is more difficult than previously thought. There are other means to process algae than by oil extraction/transesterification to biodiesel, including hydrothermal liquefaction26, where the water is not removed (a highly energy intensive process) but the material is heated wet under pressure (perhaps with a catalyst) to decompose it into liquid hydrocarbon fuels and methane. Conventional algae production can be combined with water clean-up strategies6,24,26, to remove N and P from agricultural run-off water and sewage effluent, both to prevent eutrophication (nutrient build-up in water), which causes algal blooms, and to conserve the precious resource of phosphate, since a peak in world phosphate rock production is expected around 203024. “Peak phosphate” is connected to “peak oil” since phosphate is mined using oil-powered machinery, and in the absence of sufficient phosphorus, we will be unable to feed the rising global human population, since modern industrialised farming depends on heavy inputs of phosphate, along with nitrogen fertilizers. Pesticides, too, derived chemically from crude oil, are essential, along with oil-refined fuels for farm machinery. It is, nonetheless, doubtful that the world’s liquid transportation fuel requirements can be met through standard methods of algae cultivation entirely, though fuel production on a smaller scale seems thus feasible. I gave an analogy for the latter as growing algae in a “village pond” for use by a community of limited numbers.
Noting the troubles and limitations of growing algae by conventional means and producing fuel from it on a grand scale, a breakthough does appear possible through the use of genetically modified cyanobacteria. The company, Joule Unlimited, Inc., has proposed a new high-productivity solar-to-fuels platform that uses direct product synthesis and continuous production25. The process is claimed to produce 15,000 gallons (US) of diesel/acre/year, to be compared with 3,000 gallons of biodiesel/acre/year. Now, this is hydrocarbon diesel that is being produced, with an energy content of (based on the alkane, heptadecane, C17H36) 47.2 MJ/kg, as opposed to 41 MJ/Kg for biodiesel. Noting that the respective densities of these fuels are 777 kg/m3 and 890 kg/m3, this amounts to 110.7 tonnes/hectare/year compared with 25.4 tonnes/hectare/year for the diesel and biodiesel. When the differential energy content of the fuel is taken into account, it can be seen that the direct, continuous process is almost exactly five times as productive in its energy yield/hectare than the conventional batch method. Through advances in genome engineering, solar energy capture and bioprocessing by the organism, a photosynthetic efficiency of 7.2% is obtained, to be compared with the theoretical limit of 12.1% of available sunlight at the ground being used for photosynthesis. The process is innately less complex in terms of avoiding much of the downstream processing attendant to conventional algae-to biodiesel conversion, and the product, being hydrocarbon diesel, is completely compatible (fungible) with the existing engine and liquid fuel distribution infrastructure. It is stressed that the “platform converts sunlight and waste CO2 (at a concentration some 50 – 100 times that in the atmosphere, in a closed system) directly into liquid fuels in a continuous process that is not limited by costly biomass intermediates, processing or the use of natural resources. According to the sales-pitch: “this platform can yield renewable diesel fuel in unprecedented volumes with a fraction of the land use incurred by current methods, leapfrogging biomass-dependent approaches and eliminating the economic and environmental disadvantages of fossil fuels.” The technology does indeed sound promising, but it is going to have to be proliferated on a massive scale across the world, and rapidly, if it is to cope year on year with a failing conventional oil supply5,8 of 2.9 mbd/year. This means that by 2015 – a mere 3 years time – daily production by such unconventional means needs to be around 10 million barrels5, and 64 million barrels8 by 2030, to offset the fall in conventional crude oil production and meet a projected increase in demand from the current 84 mbd to 96 mbd. It is worth noting that this amounts to a very modest anticipated growth of 0.8%/year. Previous estimates for growth during this period were up to 3%/year, reaching a production of 130 mbd, but this has been reckoned-down significantly.
Overall summary and outlook.
In conclusion, we are faced with an overall serious energy problem, and most pressingly the challenge of how to fill the enlarging hole created by a declining production of conventional crude oil. It appears almost certain that there will be profound efforts made in obtaining “unconventional oil” from shale and in liberating gas from various geological formations by “fracking”; the production of “synthetic crude” from tar-sands will doubtless increase too. Noting that world light crude oil production peaked in 2005, it is increasingly the heavy oils, e.g. from the Orinoco Belt in Venezuela, that will need to be recovered and processed, requiring the building of a new swathe of oil refineries that can handle this kind of material. Thus, not only are supplies of conventional crude oil going to fall, but what is recovered will be increasingly difficult to process. How difficult it is to produce an energy resource is usually expressed by the Energy Returned on Energy Invested (EROEI). Thus in the halcyon days of the Texan “giant gushers”, 100 barrels of crude oil could be recovered using the energy equivalent to that contained in one barrel of crude oil, which gives an EROEI = 100. The figure has fallen since then, and presently EROEIs in the range 11 -18 are obtained for North Sea (Brent Crude) oil, and as low as 3 – 5 for heavy oil and tar sands “oil”.
Oil shale should be distinguished from shale oil, though both are misnomers. Oil shale contains no oil, as such, but a solid primordial material called kerogen, which must be thermally cracked to obtain a liquid that resembles crude oil. The production and use of oil-shale is hardly environmentally “clean”, taking account of its carbon emissions (both in the retorting of shale and in burning the final fuel) and large water demand (3 – 10 barrels of water to produce each barrel of oil), and as yet there is no serious commercial production. Nontheless, becasue vast resources of oil-shale can be claimed, it is trumpeted in some quarters that the US will become self-sufficient in “oil” by 2020. In contrast, shale oil (the correct term is tight oil) is actual crude oil (petroleum), but is trapped within impermeable rock. The rock is broken open using hydraulic fracturing ("fracking"), which allows the oil to flow out, whereupon it may be refined in the normal way. Current US production of shale-oil is around 0.5 mbd, and which is predicted to rise to 3 mbd by 2020. However, this must be gauged against a loss of conventional oil by 27 mbd across the world. Can solar fuels fill the gap? As we have seen, much of the solar fuel technology is very much at the research stage. Most of what is ongoing aims to produce H2, but even if half the “new” platinum recovered annually were used to fabricate fuel cells, only something like 1% of the billion road vehicles currently in existence could be so substituted by “hydrogen cars” over the next 10 years. Hence, a global transportation network based on hydrogen/fuel cells, let alone a full-scale solar hydrogen economy, is a pipe-dream. If hydrogen can be made renewably on the grand scale, as an energy carrier (it is not really a fuel, since it must be created from primary energy sources), it will probably need to be used by combustion. The fabrication of electric cars runs into similar resource difficulties, especially in terms of rare earth metals, and so a strategy based on liquid fuels would seem most sensible. Liquid fuels are furthermore entirely compatible with the prevailing transportation infrastructure, in regard to the distribution of fuels and their deployment in internal combustion engines. The Fischer-Tropsch (FT) process is a well-established technology for converting syngas to liquid hydrocarbons, but the means to obtain H2 + CO on a large scale without using fossil fuels is not. Even when (or if) those clean technologies based on artificial photosynthesis are developed, a whole new generation and scale of FT plants will need to be installed, which at the level envisaged would take decades. Any such timescale must be judged against that for the depletion of conventional crude oil. Of those approaches considered here for the production of liquid fuels, the use of genetically engineered cyanobacteria looks the most promising, but even so, meeting the global demand for them seems to be a bridge too far. Producing millions of electric cars is just not a practical proposition, and the only realistic means to move people around in number using electrical power is with light railway and tram systems. The notion of personalised transport will be relegated to history by massive fuel prices, and an absence of any cheaper “car ownership” option. Our global civilization is underpinned almost entirely by crude oil – as refined into liquid fuels for transporting people and consumer goods around nations; for growing and distributing food; for mining coal, shale and all kinds of minerals, including metallic ores and rock phosphate for agriculture; and as a raw feedstock for the chemical industry, to make pharmaceuticals and to support healthcare. If our stalwart “black gold” is set to abandon us over the next few decades, and it is not possible on that same timescale to produce alternative liquid fuels – “the supply side” - we can only address the problem from the demand side. This means a substantial curbing of transportation and a relocalisation of society, to become more locally sufficient, e.g. in food production, at the community level. Such are the aims of the “Transition Town” movement27. It is likely that energy production will become increasingly decentralized, and done at the smaller scale, to power such communities. Fuel too, e.g. for local agriculture, might be produced from algae at least on a regional scale, as integrated with water treatment schemes6,24,26 to conserve the resource of phosphate, and to avert algal blooms. Methods of regenerative agriculture, including permaculture, provide means to food production that demand far less in their input of fuels, fertilizers and pesticides, and actually rebuild the carbon content of soil. It is thought that 40% of anthropogenic CO2 emissions might be sequestered in soil using no-till practices, if practiced across all the Earth’s 3.5 billion acres (14 million km2) of arable land28. Solar energy may also be harvested usefully and directly in the form of heat2 (rather than converting it to a fuel), at greater efficiency than through PV, using concentrating solar thermal power plants, roof-based water heating systems, solar cookers, solar stills and water sterilization units, and homes especially designed to absorb and retain thermal energy. Though the foreseeable transition to a lower energy and more localised way of life is unequivocally daunting, we should remain optimistic.
“We act as though comfort and luxury were the chief requirements of life,
when all that we need to make us really happy is something
to be enthusiastic about.” – Charles Kingsley (1819 – 1875).
when all that we need to make us really happy is something
to be enthusiastic about.” – Charles Kingsley (1819 – 1875).
(1) BP Statistic Review of World Energy 2011. http://www.bp.com/liveassets/bp_internet/globalbp/globalbp_uk_english/reports_and_publications/statistical_energy_review_2011/STAGING/local_assets/pdf/statistical_review_of_world_energy_full_report_2011.pdf
(2) Rhodes, C.J. (2010), Sci. Prog. 93, 37.
(4) Rhodes, C.J. (2008), Sci. Prog. 91, 317.
(6) Rhodes, C.J. (2009), Sci. Prog. 92, 39.
(9) Rhodes, C.J. (2011), Sci. Prog., 94, 339.
(10) McIntyre, R.A. (2010), Sci. Prog. 93, 361.
(11) Tang, J. et al. (2011), Nature Materials, 10, 765.
(12) Steinfeld, A. (2005), Solar Energy, 78, 603.
(16) Chueh, W.C. and Haile, S.M. (2009), ChemSusChem. 8, 735.
(17) Styring, S. (2012), Faraday Discussions, 155, 357.
(18) Barton, E.E., Rampala, D.M. and Bocarsly, A.B. (2008), J. Am. Chem. Soc.130, 6342.
(19) Rosen, B.A. et al. (2011), Science 334, 643.
(20) Rhodes, C.J. (2011), Sci. Prog. 94, 211.
(21) Anpo, M. (2004), Bull. Chem. Soc. Jpn. 77, 1427.
(22) Reece, S.Y. et al. (2011), Science 334, 645.
(23) Pijpers, J.J.H. et al. (2011) PNAS 108,10056.
(24) Rhodes, C.J. (2011) Sci. Prog. 94, 323.
(25) Robertson, D.E. (2011) Photosynth. Res. 107, 269.
(26) Rhodes, C.J. in Algal Fuels: Phycology, Geology, Biophotonics, Genomics and Nanotechnology, J.Seckbach (ed.), Springer, Dordrecht, in press.