1. Renewable energy technologies
It is well known that the main source of renewable energy is the solar radiation arriving on the Earth's surface, which amounts to a continuous power of 125 000 TW. This value is 10 000 times as large as the current average power of artificial energy consumed by mankind, but it is so sparse that it is useless for practical applications unless some physical mechanisms interact with the original radiation and produce new forms of energy.
Evaporation of water is one of the fundamental processes fed by solar radiation, and it is essential for life on our planet. In addition, it is the origin of hydropower, which is mainly exploited by dams containing large volumes of water from rainfalls and thaw.
Wind power has a similar root: solar radiation interaction with the atmosphere and hydrosphere creates large-scale gradients of pressure in the air, which conveys an important mechanism of energy transformation, from thermal to kinetic energy, i.e. wind. This fact simplifies the machinery to convert wind kinetic energy into rotor kinetic energy.
Solar radiation can also be exploited directly, either through photovoltaics (PVs) or through concentrated solar power (CSP). Other processes, such as Seebeck's effect, have much stronger physical limitations.
PVs are based on the PV principle applied to a semiconductor junction, and have a series of advantages because they do not convey movable parts for the electricity generation process, they do not need a cooling sink, and they are not limited by Carnot's efficiency, although they do have other limitations due to quantum mechanics principles.
In recent years, PVs have experienced an impressive cost reduction, and PV installed capacity has rapidly increased in countries with generous feed-in tariffs, such as Germany and Spain, which has, in turn, reduced furthermore the cost of PV cells (particularly the simplest ones, multi-crystalline silicon cells and thin films).
It seems that CSP has more difficulties to reduce costs and will not have ever the flexibility and the operational simplicity of PVs, which implies that an explanation about why to develop CSP plants is required.
2. Essential features of renewable energy technologies
Each renewable energy technology has its own features that must be properly analysed for understanding the role it can play in the energy sector, for each major application, e.g. transportation, industry, housing, agriculture and so forth. A key application is electricity generation. Experience in this field has already lasted more than one century and includes initial steam-Rankine coal-fired power plants, nuclear power plants and different kinds of renewable energy sources. It also includes the deployment of very large and densely interconnected grids, which are fundamental elements for attaining high reliability in electricity supply. Most of the big blackouts that have already occurred in this century, from California to the east coast, to London, to Rome, have shown very limited interconnection to cope with the problems caused by increases in consumption or defaults in generation units.
Large hydropower has been exploited since the end of the nineteenth century, and one of its main features is energy storage, which implies that it is especially suited for physical regulation (in frequency, voltage and power) of the complex physical domain ranging from generation units to the end consumer.
New renewable technologies, notably wind power and PVs, do not add stability to the system because they do not have inertia of any kind and their original source of energy cannot be stored. This is particularly negative for electricity generation, and it must be underlined that both wind power and PVs are electric generators per se. It is sometimes forgotten that a kilowatt is the shortest lived commercial good because it lasts around 1 m from generation to consumption. Electricity cannot be stored as such in relevant amounts, and this is another essential characteristic of the problem. In the classical electricity sector, hydro-pumping has been the standard method for energy storage, but there are obvious limitations in this field because of the amount of water and the difference in altitude required for having a significant energy storage.
It is worth remembering that storing 1 MWh requires pumping 3000 m3 of water up to 100 m above the original level. Note that this value roughly corresponds to the energy generated over 1 h by a 1 MW windmill.
An alternative for storing energy is thermal energy storage, which is a very powerful method, although it has two main limitations: thermal losses in the storage charge, discharge and proper storage phase; and the need of a thermodynamic cycle for recovering the energy as electricity. The latter is an actual drawback in general terms, but it is not so in CSP because the thermodynamic cycle is already present.
A typical Spanish commercial CSP plant of 50 MW may be given as an example. The capital cost of the plant in southern Spain a few years ago was of the order of €250×106. The addition of a 7.5 h storage based on K and Na nitrate salts has required, within the given balance of plant, an additional cost of €36×106 and a larger solar plant in order to increase the useful time window and hence the amount of produced total energy. Notwithstanding, the overall increments in costs have actually resulted in significant reduction in the levelized electricity cost (LEC) of electricity expressed in cents per kilowatt hour. However, this case is far from being an optimal choice, as due to the relatively low temperature (≈380°C) of the chosen oil storage, the resulting temperature difference (DT) for the thermal storage has been only DT=96°C. Higher temperature plants presently under construction can operate with higher temperatures, for instance, DT=275°C, reducing the volumes and costs by a factor of about three. While the molten salt solution is today a fully operational technology, other forms of storage are under active development, for instance, based on different and cheaper heat-storing media, such as pebble beds or even concrete.
As far as thermal losses are concerned, strong insulation and adequate sizing of the equipment can reduce the losses to moderate values. Storing 1 MWh in a tank of molten salt with a temperature jump of 200°C requires 6 m3of salts. Even if an overall thermal efficiency of 50 per cent is assumed, the size is 12 m3 MWh−1. This fact changes the perspective for CSP because it can generate electricity with a sound capability to stabilize a futuristic electric system with a high share of renewable energy sources.
Of course, this opportunity will be restricted to locations with high levels of direct solar radiation, typically deserts, which are very far from the high consumption zones of central and northern Europe. This implies the need to send electricity along several thousand kilometres, but this is a challenge that could be answered with an adequate effort on superconductivity, although other alternatives based on high-voltage DC transmission lines must also be taken into account. This fact also implies that air cooling will be mandatory for CSP plants because there will not be any water available for it.
3. A new electricity expansion
Electricity as an energy and economical sector is bound to undergo paramount transformations in this century, owing to new paradigms such as electric vehicles, distributed generation, smart grids and a new portfolio of generation technologies, including new thermodynamic cycles especially suited for CSP. Although these novelties will convey new technical features and will impose new requirements, the main issue will continue to be ‘reliability’. Our way of living and our way of working (in industries, services and so on) critically depends on a reliable supply of electricity, which will be an enormous task in a generation scenario with a majority of renewable energy technologies that are unreliable because their ‘fuel’ suffers from natural shortages, some of them anticipated, some of them totally unexpected.
Reliability will be improved with strong interconnections between regions with different and complementary meteorology, such as central and southern Europe, but such a complex system will require some important generation points, with one or several big power plants with capabilities for stabilizing the system at the onset of a collapse.
Otherwise, a blackout will occur in some parts of the system (if the perturbing places can be isolated promptly) or will dominate the whole of the system, thus entailing a tremendous negative impact, both at personal level and at macroeconomic level. The former cannot be measured in quantitative terms, but the latter can be estimated after a number of blackouts of different duration that have happened in our recent history. In general, blackouts of a few minutes have a moderate impact, which increases exponentially with the blackout time span. A blackout of one day will likely produce economic losses twice as large as the domestic gross product associated with one workday. Longer blackouts can be economically deleterious.
Therefore, the electric sector in the future will need to have a generation system with a proven capability to react against disturbances, and CSP plants will have to be an important part of it. This means that thermal energy storage will be a must, and the unitary power will have to grow from the current values of around 50–200 MW or so.
CSP is only appealing in locations with high values of direct normal irradiation. In the most illuminated sites (such as deserts), the impinging solar radiation conveys a thermal energy content that is equivalent to a ‘flood’ of petroleum 20 cm high or more. Finding technical solutions for profiting such huge amounts of ‘solar petroleum’ is a challenge from many points of view, and it is a must for sustainable development, which is considered the key for a brighter future.
As an example, the amount of primary renewable solar energy from Saudi Arabia is equal to about 1000 times the whole fossil energy presently produced there from oil and natural gas. Evidently, although such an amount of ‘solar petroleum’ is freely available over the many deserted areas of the huge Sun Belt, its actual transformation into a practical energy source, either electric or otherwise (e.g. hydrogen) requires formidable technological developments. Even if the conversion has an inevitable loss factor, current efficiencies in the region between 10 and 25 per cent have already been currently achieved industrially, both for PVs and CSP for electricity production.
In addition to being brighter, solar thermal energy must also be cheaper. The cost of energy is one of the critical factors for social and economic welfare, and sustainable development could not be sustained indefinitely if clean, renewable energies would require permanent subsidies. Solar energy may become, in the future, a main element of progress only if novel methods are developed that are capable of bridging the gap in the relative costs with respect to other energy alternatives.
CSP and PVs are clearly complementary options. Considerable developments are forthcoming in order to reduce the costs of CSP to a level that is quite comparable with the present and projected costs of PVs that is now reaching the LEC level of about 3 € W−1 peak. For instance, with a ‘lifetime’ of 2200 h yr−1, an equivalent 50 MW PV plant will require a capital cost of about €150×106. The capital cost of a 50 MW conventional but optimized CSP plant in Spain is now as low as €140×106, of which €94×106 is for the solar field. New solutions based on Fresnel lenses might reduce the cost of the solar field to €46×106. Future CSP plants are therefore expected to have specific investment costs of the order of 2 € W−1 peak.
These reasons should help guide research and development activities in this field, which are full of options concerning the geometry of the solar field, the thermodynamic unit, the thermal energy storage and so on. CSP has some obvious limitations, but it also has some unique features that will be highly valuable in the future.
One contribution of 15 to a Discussion Meeting Issue ‘Can solar power deliver?’.
- © 2013 The Author(s) Published by the Royal Society. All rights reserved.