The promising perspective of Concentrating Solar Power (CSP)

Paper presented at the conference Futures Power Systems 2005 in Amsterdam, 16-18 November 2005.

Evert H.du Marchie van Voorthuysen
Director  of  GEZEN Foundation for Massive Scale Solar Energy, www.gezen.nl, Nieuwe Kerkhof 30a, 9712 PW Groningen, The Netherlands. E-mail: voorthuysen@home.nl

Abstract--

Index Terms—

Chemical energy storage
Desalination
Desert regions
Hydrogen economy
Energy storage
HVDC transmission lines
Solar power generation
Solar energy
Thermal energy storage
Zinc

I.  Introduction

In the beginning of the twenty-first century the world is confronted with three major problems: i. the global climate change caused by the anthropogenic greenhouse effect, ii. the energy shortage of which the high oil prices are a symptom, iii. the shortage of fresh water in many parts of the world. Four different solution strategies are widely proposed:
·        Increase of energy efficiency and energy saving
·        Massive introduction of renewable energy technologies
·        Coal gasification together with CO₂-sequestration
·        Massive (re)introduction of nuclear energy.
There is a widely felt common understanding that the solution strategies at the top of the list are preferable to those lower down. The massive reintroduction of nuclear energy and the introduction of CO2-sequestration in particular should not be implemented unless it becomes clear that all renewable energy options together would be insufficient to solve the energy and climate problem.
    Of all sources of renewable energy solar energy is by far most abundant. There are two methods to extract electricity from solar radiation: Photovoltaics (PV) and Concentrating Solar Power (CSP). According to a study by the International Energy Agency [1] CSP is a factor 3 cheaper than PV, at present, and in the future. In this paper we show that there are no economic objections against a massive deployment in the world of Concentrating Solar Power.

II.  CSP Technology

A.  Harvesting Solar Heat

A solar thermal power station is a conventional power station in which the burning fossil fuel in the boiler is replaced by the heat from concentrated solar rays. Six different optical methods can be applied to concentrate sunlight, see Table I. The high oil prices and the growing concern about climate change have triggered many new projects for building solar thermal power stations. The table mentions only the existing stations and the most recently planned projects, which are not listed in the overview of Greenpeace [2]. As all relevant information originates from private companies, we give their names in the footnote to Table I enabling the reader to continue his/her investigation.
    In the first two configurations of Table I the mirrors rotate along one North-South oriented axis in order to keep the rays concentrated onto a line-focus. A receiver tube which is located in this line-focus contains a streaming liquid which absorbs the solar heat. The liquid is either oil, a mixture of liquid salts or boiling water, see Fig.1. The receiver tube of a linear Fresnel mirror field is stationary whereas parabolic troughs and linear Fresnel lenses have moving receiver tubes.
The last three configurations of Table I have a point focus and the necessary rotation occurs along two axes. The concentration factor of the solar radiation is higher, leading to higher operating temperatures. The common focal point of the heliostat mirrors is located in a receiver on top of a solar tower.
Solar thermal power stations are located in dry climates, often deserts. During sand-storms and heavy hail-storms the mirrors are turned upside down. The protection strategy for lense systems is still unknown.

B.  Electricity out of Solar Heat

The most common method for producing mechanical power and hence electricity from concentrated solar heat using the first four technologies of Table I is the Rankine cycle.  The basic solar thermal power station consists of a mirror field (or the roof of a greenhouse consisting of linear Fresnel lenses), receiver(s), a heat exchanger, a boiler, a turbine, a generator, and a condenser. The heat exchanger can be omitted when steam is produced directly in the receiver. The condenser is cooled by sea water, a wet cooling tower, or by means of air cooling.  The peak in the electricity output conveniently coincides with the peak demand from air conditioning appliances.  Occasional cloudy days can be dealt with by means of burning gas, oil, or hydrogen in the boiler.

TABLE I   Concentrating Technologies

Refs.[3]-[4].  b. Chris van Felius, Rotterdam, The Netherlands, private communication. c. International Automated Systems, Utah, USA  d. Schott AG, solar, Mainz, Germany. e. Solargenix Energy, Raleigh, NC, USA. f. Solar Heat and Power, Sydney, Australia. g. Solar Power Group GmbH München (successor of Solarmundo, Belgium). h. Stirling Energy Systems, Phoenix, Arizona.  j. FPL Energy, Juno beach, Fl, USA; the plants were built in 1984-1990 by Luz International, USA. k. Solarmillennium AG, Erlangen, Germany and ACS Cobra, Madrid, Spain. l.  Iberdrola, Bilbao, Spain.  n. Kernenergieen, Stuttgart, Germany. o. Abengoa, Sevilla, Spain. p. Ref.[5]

Like many conventional power stations in the subtropics, solar power stations in coastal regions can be combined with desalination of sea water; the temperature of the condenser is increased to about 75 ⁰C, causing a slight decrease in efficiency, but saving the investment of the bulky low-pressure turbine.
In the Multiple-Effect Desalination (MED) process sea water is distilled in a series of 10 to 15 vessels at decreasing temperatures and pressures. In combination with solar thermal power plants, MED desalination is more economical than  Reverse  Osmosis desalination [6].
   In the future solar towers may become equipped with novel hybrid gas turbines employing solar pre-heating [7].  Solar dishes and Fresnel lenses usually have Stirling engines or microturbines in their focal points and are by nature comparatively small (20 - 100 kWe). They do not enjoy the same economy of scale as do the trough and tower systems so it is doubtful whether they will ever form the backbone of multi-GW grid-connected systems.  However, solar dishes and Fresnel lenses could play an important role in the decentralized part of the solar economy, see section II.5

C.  Calculation of the Solar Yield and its Seasonal Variation

The Direct Normal Irradiation (DNI) is defined as the intensity
of the radiation that originates directly from the sun prior to scattering.  Concentrating solar techniques utilize only the DNI, whereas solar cells (PV = Photo Voltaics) and solar boilers also utilize diffuse radiation. The DNI just before entering the atmosphere is I₀ = 1376 W/m². The cloudless atmosphere absorbs and scatters radiation, the remaining transmission fraction Tr depends on the zenith angle z   between  the perpendicular and the direction of the sun and on the altitude. A good approximation for the transmission factor is [8]:

Fig.1. Cross-section of a parabolic trough mirror with the receiver tube isolated by vacuum and covered by a spectral-selective coating, not to scale.

Tr(z) =  a_o + a₁ exp(-k /cos z)                     (1)               

                               
The factors a_o, a₁, and k depend strongly on the altitude and slightly on the climate. In this paper we apply the values for sea level: a_o = 0,13, a₁ = 0,76, and k = 0,39.
The zenith angle z depends on the declination d of the sun  (-23.5⁰ < d < 23.5⁰ , dependent on the date), the geographical latitude ph and the time angle of the day om  (om = (t - 12).15⁰, with t the clock time in hours in the 24 hours system) in the following way:

cos z = sin d sin ph + cos d cos ph cos om               (2)

The DNI at ground level is

DNI (z) = 1376 Tr (z)                                 (3)

The largest cost factor of a solar thermal power station is the collector field of mirrors or lenses, therefore the most relevant parameter is the irradiation per m² of mirror or lens.  Solar dishes and Fresnel lenses are oriented towards the sun, so for them the DNI is the relevant parameter. For collecting systems
that are in a horizontal position such as linear Fresnel mirrors and linear Fresnel lenses the intensity of the solar radiation I_hor    is weakened by a factor cos z:

I_hor (z)  =  DNI (z) cos z                      (4)       

The angular weakening factor for parabolic troughs and heliostat mirrors is less severe.
     For two different locations on the Earth on two different dates we calculated the DNI and the horizontal irradiation  I_hor for each hour using the equations  (3) and (4) and out of these numbers the total daily yield in kilowatt hours.  In the tropics we chose Darwin (Australia), ph = 12⁰ latitude South, and in the subtropics Southern Morocco, ph = 28⁰ latitude North.
    In the summer there is not much difference between the tropics and the subtropics, but at winter time a tropical site is superior.  However, table II does not take clouds into account. In most tropical regions clouds reduce the annual production of CSP plants to such an extent that they probably never will become viable. Detailed DNI-values can be calculated from satellite data of moisture and dust, gathered over many years for all locations on Earth [9].

TABLE  II
Maximum and Minimum Irradiation in the Tropics and the Subtropics
  

  
  A desert location at ph = 28⁰ receives on average 4.65 kWh_th/m²day of horizontal irradiation  I_hor  which is 1697 kWh_th/m²year or 193.8 W_th/m² continuously. The efficiency of collecting the solar heat is typically 50% and the heat-to-electricity efficiency is 30%, resulting in a 15% overall efficiency of the solar power station, so the average electric yield is 29.07 W_e/m² of horizontal area or 29 MW_e/km². A base-load solar power station of 1 GW_e needs 34.4 km² of desert ground. The present continuous primary global energy consumption is about 14000 GW. A desert area of 482,000 km², less than the size of France, would suffice to supply the whole world with solar energy.

   
D.  Storage of Heat and Chemical Potential Energy

An extension of the elementary configuration with extra mirrors and receivers and with heat-storage tanks enables around-the-clock solar electricity generation. The heat-storage tanks may contain a mixture of salts, in fact any material that is able to survive the temperatures will do. A storage capacity of 16 hours is sufficient to compensate completely for the unproductive part of the 24 hours' day. In order to guarantee complete supply security two measures can be taken to compensate for a sequence of cloudy days, i. a substantial increase in storage capacity or ii. addition of the facility for burning gas, oil, or hydrogen in the boiler.
    In the last column of Table II we see a large effect in the solar yield; in the winter a linear Fresnel plant collects only about 40% of the summer yield. It is not viable to average out the seasonal effect by means of thermal storage. What is needed is chemical storage, that is the production of chemicals with a large chemical potential energy out of the concentrated solar heat.
     

I₂ + SO₂ + 2H₂O  —›  2HI + H₂SO₄
H₂SO₄  (850 ⁰C)  —›  H₂O + SO₂ + ½O₂                                                         (5)
2HI (500 ⁰C)  —›  H₂ + I₂

_{Another way to produce hydrogen is the zinc oxide process [11]:}

The extremely high dissociation temperature in (6) can only be obtained by means of solar dishes. An important advantage of the zinc oxide process is the easy and relatively cheap storage of energy in the form of zinc metal. The hydrogen economy could become more realistic when CSP would be the main source of energy, and zinc metal the storage and transportation medium. The hydrogen-producing reaction (7) proceeds exothermally at temperatures above about 400 ⁰C [12]. The losses are still substantial, arising mainly from the quench process after formation of the gaseous oxygen-zinc mixture. However, when the zinc production would be combined with direct electricity generation using the waste heat from the quench process, the overall efficiency would increase substantially.
It is to be expected that continued research and development in this still rather new field will lead to great progress.
If it turns out that there is no viable solution for the long-term storage of solar heat (just as no viable solution exists for storing electricity) it still makes sense to invest in solar thermal power stations. From table II we calculate that on December 21 the production is reduced to 557 MW. Reduced production occurs during the winter half-year, the reduction, averaged over the whole year is 14.6 %.

E.  Centralized and Decentralized CSP

The optimum size of a solar thermal power plant is smaller than that of a fossil-fuel fired or nuclear thermal power station, which is typically 1 GW. The area of the Fresnel mirror field of a 1 GW solar power station would be 34 km², leading to an average transport distance for the steam from the receiver to the turbine of more than 2 km. The inevitable losses of  this transport makes a solar thermal power plant of this size uneconomical. The optimum size is the result of a trade-off between these losses and the optimum size of the power block (turbine + generator). It will probably be around 100 MW. Many of these units will probably be combined and will together form one large solar thermal power station.
    Most of the time that the sun is shining the collector field is gathering more heat than the power block is able to handle, and the extra heat has to be stored. It is possible to transport well-insulated heat storage tanks over many kilometers without losses that are worth mentioning. A certain fraction of the total mirror area can therefore be installed outside of the grounds of the power plant. These mirrors can be owned by independent solar farmers, who pump solar concentrated heat into special containers. At night the containers are transported to the central plant, for instance by means of narrow-track railways, and replaced by cold containers. If the central CSP-station is located at the coast, seawater desalination can be incorporated. The solar farmers receive their payment partially by means of delivery of fresh water, enabling them to grow their own food.
The independent solar farmers are free to construct their own mirrors, allowing for a wide range of solar harvesting procedures, from very basic technology up to a high level of sophistication.
    The most sophisticated small-scale CSP technology is the solar dish. To date the dish has been applied for direct electricity generation only, using Stirling  generators at the focal point. Solar dishes could be used just as well, or even better, for harvesting solar heat in containers and for the production of  energy-rich chemicals, for instance zinc metal, according to (6), see Fig.2. 

Fig.2 Solar dish in the position of harvesting solar heat (left) and in the position for replacing a hot container by a cold one (right). The container can be a well-isolated vessel with for instance molten salts, or a complete chemical reactor.

III.  CSP Economics

A.  Current costs of CSP and subsidy policy.

Like any  starting technology, costs are high due to the absence of optimized, massive production methods. The first products (solar thermal power stations) will be relatively expensive, leading to high kWh-costs in the start-up phase, Fig.3. There are two methods to overcome the difficult market penetration phase: i. operation in niche markets, and ii. subsidy of the technology because of its assumed beneficial effects for society.  
    At the present cost level the production costs of solar heat in a parabolic trough station that is financed at 9%, with a payback period of 20 years, correspond to an oil price of 54 $/barrel [13]. At the oil price level of the second half of 2005, countries that are lying in the sunbelt of the earth and that have their electricity supply based on oil form niche markets where pioneer CSP-projects can start without any subsidy. Other niche markets are hotel parks in sunny regions with a poor supply of electricity and water. Studies performed at the DLR [6][14] have shown that the combined production of electricity and desalinated water is a good method to overcome the first barrier.
   Niche markets may change rapidly. If oil prices were to drop by 50%, most niche markets for CSP would disappear. Investors are therefore reluctant to invest in capital-intensive solar thermal power stations in such markets. An intelligent system for subsidizing solar thermal power is therefore needed. The principle of subsidizing renewable-energy technologies is justified if the subsidy helps to overcome the barrier for market penetration. A necessary condition is that there is a good prospect for the technology to become economically viable. For CSP this prospect is real, see section IIIB.
    The most intelligent subsidizing system is the feed-in method, where the companies operating wind turbines, solar panels, etc. receive a fixed kWh price during a large number of years. The superiority of the feed-in system is widely recognized. It is demonstrated by the much higher density of wind turbines in Germany as compared to Great Britain. In Germany the current feed-in law was introduced in 1991.
    Concentrating solar power needs direct sunshine and is viable in the sun-belt countries only. The best region in Europe for developing CSP is Southern Spain. However, the Southern European countries do not possess the vast unused areas that are needed for the energy need of the European population. Most electricity and most hydrogen or zinc will be imported from North Africa and the Middle East, the so-called MENA region, see Fig.4. Transportation costs of electricity are modest when High-Voltage Direct Current (HVDC) technology is applied, transportation of electricity from Southern Morocco to Central Europe increases the necessary investment costs about 30%, Sec.IIIC.  A super-continental electric grid will level off the fluctations from the wind farms and will bring "green" current to countries that lack sufficient sources of renewable energy.
     The first solar thermal power stations in the MENA region will be built on short term, before completion of the HVDC grid to Europe. They will deliver electricity and water to the local utilities, intended for the local consumers. In the feed-in system, all local consumers together share the costs of the unprofitable part of the pioneering CSP-stations. It would not be fair to burden the rather poor population in most MENA countries with the task of making CSP-technology viable. A new international revolving fund, the International Solar Mobilization Fund (SMF), could steer European tax money to a feed-in subsidizing system for CSP-plants in the MENA countries [15]. Next to its subsidizing task the SMF could become of great help for solving the many problems that are connected to investments in less-developed countries.

B.  The learning curve for CSP-technology

The nine existing solar power stations in California sell electricity for fixed prices  under contracts that last 30 years. The oldest plant, built in 1985 delivers electricity for 24 $cents/kWh, the youngest, built in 1990, earns 12 $cents/kWh. These high prices are justified as solar electricity is produced at the time of the day that the demand is at its maximum because of air conditioning installations.  The fast decrease of production costs of solar electricity in the period 1985-1990 is common to all emerging technologies. Costs decrease according to the so-called learning curve if two conditions are fulfilled: i. there is a consistent economic drive for investing in the new technology, ii.there is competition between the producers of the new technology.
      From 1991 onwards no power purchase contracts were closed for new solar power stations, and the producer of the parabolic trough receivers, LUZ International, went bankrupt. The learning curve for CSP is stagnating, and the current costs for solar electricity from solar thermal power stations have remained at the 1990  level. Research and development in CSP has continued , mainly in the USA, Spain, and Germany.
    In 2003 the Sargent & Lundy Consulting group [16] analyzed the scenarios projected by the combined CSP laboratories in the USA (Sunlab). When a consistent program  for building solar thermal power stations would be started in 2004, resulting in a world total CSP capacity of about 3 GWe in 2020, the production costs are estimated to become 6.2 $cents/kWh for CSP-stations using parabolic trough mirrors, and 5.5 $cents for stations using solar towers, Fig.3.
   

The kWh costs of Fig.3 are consistent with the results of a scenario for introducing solar thermal power stations in combination with MED seawater desalination made by Trieb and Knies [14] at the DLR in Germany .

C.  Case study: the investment for a 1 GW_e CSP-station in Southern Morocco for delivery of electricity to Europe.

   The backbone of the future supply of electricity for Europe will be formed by hundreds of solar thermal power stations in the Middle-East and North Africa (MENA) Fig.4. In this paragraph we estimate the investment that is needed for the delivery of base-load solar electricity to Central Europe. We assume that CSP-technology is already applied worldwide and that the learning curves for the different components of CSP have led to saturation values for the costs of linear Fresnel mirrors and receivers and for the thermal storage technology. The costs for the chemical storage are unknown, as it is not clear yet which set of chemical reactions will turn out to be most viable.
   The current square meter price for linear Fresnel receivers, including the receiver tube, Fig.1, varies between 150 $/m² [3],[7] and 106 $/m² [17]. We estimate a final cost price of   50 $/m². The 1 GW_e plant needs 34.4 km² of mirror area, Sec.IIC,  leading to an investment of 1522 M$.
   The power block, consisting of the steam turbine, the generator, the condenser, pumps, etc. is a conventional part of thermal power stations.  We estimate the costs to be 800 $/kWe leading to 800 M$ for the 1 GWe plant.
    A detailed analysis teaches that a storage capacity of 46 GWh_th, which corresponds to 15.3 hours, is sufficient to bridge the evenings and nights. Trieb and Knies [14] estimate a final investment cost of 9 $/kWh_th leading to 414 M$. So the total investment for a base-load 1 GW_e solar thermal power station is 2736 million dollars + the investment costs of the chemical storage, which we don't know yet.
    The electricity has to be transported to Central Europe along 3500 km of overhead High-Voltage Direct Current (HVDC) lines and 50 km of sea cable with power conversion stations at both ends.  The investment costs are 430 M$/GW and the losses are 14.5% [14]. In order to receive 1 GW in Europe, the investment in solar thermal power stations in Southern Morocco must be increased 14.5% resulting in 3133 M$. Together with the 430 M$ investment in transport we arrive at 3563 M$ investment for clean electricity of 1 GW in the period March - September, 600 MW in December and January, and values in between for the remaining months. When investments in chemical storage, for instance zinc metal,  become viable, the production dip in winter time would be eliminated.
    These results are valid for a region without any clouds. If clouds occur for instance 5% of the time, the investment in the mirror field must be increased by 5%, and the investment in the thermal storage by a larger factor in order overcome cloudy days.
   The (rounded) 3.6 G$ investment in a (nearly) complete solar electric infrastructure of 1 GW can be compared with the investment in nuclear energy. The newly planned nuclear power plant of 1.6 GW in Finland will cost 3.2 G€ or about 2.5 G$/GW.  Nuclear energy has special costs which are absent in solar energy: mining of uranium, enrichment of uranium, the absolutely safe management of the nuclear waste, the high decommissioning costs of the worn out reactors. There is no clear economic argument for nuclear energy and against CSP. Any political decision in favor of large-scale reinvestment in nuclear power stations that is not taking CSP into account as a possible alternative is therefore a questionable decision.

Fig.4. The 100% renewable economy in Europe and surrounding countries in which CSP-desalination plants along the costs of the Mediterranean, the Atlantic Ocean and the Red Sea are playing a major role [13][14].

IV.  CSP Politics

Some densely populated regions in the world are inadequately supplied with direct solar radiation, especially the European Union and Japan. In the solar age they will import a major part of their electricity directly from North Africa and China. The geopolitical risks for Europe will be less than the current risk associated with the increasing dependence of Europe on oil from a decreasing number of oil-exporting countries. Politicians who tend to reject CSP because of the necessity to cooperate with "unreliable" African countries should realize that i. Europe can never be self-sufficient with respect to energy, ii. co-operation on energy between the former enemies Germany and France became a big success, which sets a good example.
    The development of Concentrating Solar Power is stagnant since 1990. This has led to a neglect of CSP by practically all scientists who build energy scenarios. Concentrating Solar Power was therefore absent in nearly all representative advisory reports on energy, resulting in ignorance among governments, politicians and companies. Until recently this ignorance consolidated the stagnation. This neglect of CSP by the institutes of science policy has no scientific justification.
     In the immediate future important decisions have to be made on investments in the energy infrastructure. These investments can be justified only when all energy options, and especially all renewable-energy options are taken into account. Knowledge of the CSP-option is therefore mandatory for all decision makers.  

V.  Conclusion

Concentrating Solar Power is a technique for generating renewable electricity with clear advantages with respect to other sources of renewable energy:
·        On a global scale there are no physical limits to its growth
·        Heat storage enables the delivery of base-load solar electricity
·        Seawater can be desalinated in a sustainable way
·        In principle, hydrogen and other energy-containing chemicals like zinc can be produced in a sustainable way
·        The CSP costs are sufficiently low to enable deployment in existing niche markets without subsidy.
For countries situated at high latitudes the main disadvantage is the necessity to operate the solar thermal power stations in countries far away. The transport costs of electricity over thousands of kilometers do not form a serious objection, however.
    Dedicated research and development is needed for solving the problem of chemical energy storage. European countries should start to cooperate with MENA countries in order to facilitate investments in CSP on MENA territory. An international conference should be held with the purpose of founding  the International Solar Mobilization Fund SMF.

[1]   "World Energy Outlook 2004",  International Energy Agency, Paris.
[2]   R. Aringhoff, G. Brakmann, C. Aubrey, and S. Teske, "Solar thermal power 2020, Exploiting the heat from the sun to combat climate change",  Greenpeace International, Amsterdam (2003). Available: http://www.greenpeace.nl/raw/content/reports/solar-thermal-power-2020-en.pdf
[3]   S. Bockamp, T. Griestop, M. Fruth, M. Ewert, H. Lerchenmüller, M. Mertins, G. Morin, A. Häberle, J. Dersch, "Solar Thermal Power Generation",  presented at  Power-Gen Europe 2003, Düsseldorf, Germany.
[4]   A. Häberle, C. Zahler, H. Lerchenmüller, M. Mertins, C. Wittwer, F. Trieb, J. Dersch, „The Solarmundo line focussing Fresnel collector. Optical and thermal performance and cost calculations."  presented at  SolarPaces Conference 2002, Zürich, Switserland. Available: http://www.solar-power-group.de/cms/upload/pdf/Fresnel_trough_cost_compare.pdf
[5]    E. Zarza, K. Hennecke, and O. Goebel, "Project DISS (Direct Solar Steam). Update on Project Status and Future Planning"  presented at ISES Solar World Congres 1999, Jerusalem, Israel.
[6]   R. Wilde, "Case study of a concentrating soaler power plant for the cogeneration of water and electricity." Undergraduate thesis at Deutsches Zentrum für Luft- und Raumfahrt, Stuttgart and Rheinisch-Westfälische Technische Hochschule Aachen, Germany, 2005.
[7]   R. Pitz-Paal, J. Dersch, and B. Milow, „European Concentrated Solar Thermal Road-Mapping, Roadmap Document", final report of EU contract SES6-CT-2003-502578, DLR 2005.
[8]   J. A. Duffie and W. A. Beckman, Solar Engineering of Thermal Processes, .  New York: Wiley, 1980.
[9]   STEPS-Strahlungskarten; Deutsches Zentrum für Luft-und Raumfahrt (DLR), Stuttgart 2005.
[10]           S. Kasahara, S. Kubo, K. Onuki, and M. Nomura, "Thermal efficiency evaluation of HI synthesis/concentration procedures in the thermochemical water splitting IS process," International Journal of Hydrogen Energy vol.29, pp.579-587 (2004).
[11]           A. Steinfeld, "Solar thermochemical production of hydrogen - a review," Solar Energy      (2004)
[12]           A. Berman and M. Epstein "The kinetics of hydrogen production in the oxidation of liquid zinc with water vapor," International Journal of Hydrogen Energy vol.25, pp.957-967 (2000).
[13]            G. Knies and M. Kabariti, "Memorandum" (2004). Available:http://www.trec-eumena.net/documents/memorandum_20041108.pdf
[14]           F. Trieb and G. Knies, "A Renewable Energy and Development Partnership EU-ME-NA for Large Scale Solar Thermal Power & Desalination in the Middle East and in North Africa." DLR, Stuttgart, Germany,  2004. Available: http://www.trec-eumena.org/documents/sanaa_paper_and_annex_2004_04_15.pdf
[14]           F. Trieb and G. Knies, "A Renewable Energy and Development Partnership EU-ME-NA for Large Scale Solar Thermal Power & Desalination in the Middle East and in North Africa." DLR, Stuttgart, Germany,  2004. Available: http://www.trec-eumena.org/documents/sanaa_paper_and_annex_2004_04_15.pdf
[15]           E. H. du Marchie van Voorthuysen, "The International Solar Mobilization Fund,"  presented at the Second MENAREC Conference, Amman, 2005. Available:  http://www.gezen.nl/index.php?option=com_content&task=view&id=60&Itemid=68
[16]           Sargent & Lundy Consulting group, Chicago, Illinois, October 2003. Available: www.osti.gov/dublincore/gpo/servlets/purl/15005520-kLbVbt/native
[17]           D. Mills, G. L. Morrison, and P. le Lièvre, "Design of a 240 MWe Solar Thermal Power Plant", Proc. Eurosun Conference, Freiburg, 2004
http://www.gezen.nl/images/stories/treckaarttekstklein.jpgI.  Biographies