Benefits of Medium Temperature Solar Concentration Technologies as Thermal Energy Source of Industrial Processes in Spain

energies Article Benefits of Medium Temperature Solar Concentration Technologies as Thermal Energy Source of Industrial Processes in Spain Isidoro Lillo-Bravo 1,* , Elena Pérez-Aparicio 2, Natividad Sancho-Caparrini 2and Manuel Antonio Silva-Pérez 1 1Department of Energy Engineering, Universidad de Sevilla, Camino de los Descubrimientos s/n, 41092 Seville, Spain; [email protected] 2Andalusian Association for Research and Industrial Cooperation, Camino de los Descubrimientos s/n, 41092 Seville, Spain; [email protected] (E.P.-A.); [email protected] (N.S.-C.) *Correspondence: [email protected]; Tel.: +34-95-448-7236 Received: 31 July 2018; Accepted: 23 October 2018; Published: 29 October 2018   Abstract: This paper analyses the possible applications of medium temperature solar concentration technologies, Compound Parabolic Collector, Linear Fresnel Collector and Parabolic Trough Collector in the Spanish industrial sector. Results of this study allow evaluating whether or not solar technologies are an alternative to conventional sources. This possibility is analyzed energetically, economically and environmentally. Results show that the percentage of solar use is decisive in determining the true thermal energy generation cost. The other essential parameter is the solar field area due to produce economy of scale that reduces investment costs. Fluid temperature has significant influence mainly in Compound Parabolic Collector technology. Results obtained in this paper collect multiple alternatives and allow comparing for different scenarios the suitability to replace conventional energy sources by thermal energy obtained from medium temperature solar concentration technologies from an economic perspective. For instance, for percentage of solar use equal to 100%, the lowest thermal energy generation costs for each technology are 1.3 c € /kWh for Compound Parabolic Collector technology, fluid temperature of 100 ◦ C and industrial process located in Seville, 2.4 c € /kWh for Linear Fresnel Collector technology, fluid temperature of 170 ◦ C and industrial process located in Jaen, 3.3 c € /kWh for technology, fluid temperature of 350 ◦ C and industrial process located in Jaen. These costs are lower than conventional energy sources costs. Keywords: solar concentration technologies; solar thermal energy; heat process; thermal energy generation cost; greenhouse gas emissions 1. Introduction World consumption of primary energy is growing as though supplies of fossil energy carriers were unlimited and climate change was not occurring [ 1 ]. Global primary energy consumption increased by 1% in 2016, following growth of 0.9% in 2015 and 1% in 2014. This compares with the 10-year average of 1.8% a year. As was the case in 2015, growth was below average in all regions except Europe and Eurasia. All fuels except oil and nuclear power grew at below-average rates. The analysis of the primary energy consumption distribution shows that oil remains being the most consumed primary energy, 31.7% over the total. Coal remains as the second energy resource with 28.1%. Natural gas appears in the third position; its consumption represents 21.6% of the total [ 2 ]. These data show that fossil fuels are still the most used primary energies. The sum of their consumption is around 80% of the total. Energies 2018,11, 2950; doi:10.3390/en11112950 www.mdpi.com/journal/energies Energies 2018,11, 2950 2 of 30 The analysis of the distribution of energy end-uses shows that 46% of energy consumption is used to satisfy cooling or heating processes requirements at industrial, residential or tertiary levels. Most of this energy is produced from fossil fuels and only 15% comes from renewable energies. The remaining 54% consumed energy is divided between electricity and transport, 32% and 22% respectively [ 3 ]. These figures clearly show the importance of global energy consumption for thermal purposes and the low percentage that is satisfied by renewable energies. Solar thermal energy is one of the alternatives that nowadays present a greater potential to reduce the fossil fuels consumption. Solar technologies can be applied in lots of industrial processes, mainly due to the temperature range that they allow, from 45 ◦ C to 400 ◦ C. Higher temperatures could even be covered if it would be necessary, although this is not usual in industrial processes. To select one or other of the available solar technologies, it is crucial to analyse the industrial processes thermal requirements whose needs are to be met. Table 1shows the industrial processes temperature range susceptible of this study [4,5]. Table 1. Industrial processes temperature range [4,5]. Industry Process Temperature Range (◦C) Dairy Sterilization 100–120 Drying 120–180 Canned food Sterilization 110–120 Agricultural products Drying 80–200 Textile Drying 100–130 Degreasing 160–180 Paper Bleach 130–150 Chemistry Soaps 200–260 Synthetic rubber 150–200 Process heat 120–180 Petroleum 100–150 Wood products Pulp preparation 120–170 Desalinization Heat transfer fluid 100–250 Mining Drying 100–400 Concentrate smelting Heating solutions Washing Plastics Preparation 120–140 Distillation 140–150 Separation 200–220 Extension 140–160 Drying 180–200 Mixing 120–140 Thermal treatment Medium tempering 350–450 Refrigeration Double effect solar chiller 120–190 In addition to the heat transfer fluid temperature, another important issue to assess the suitability of solar technologies as provider of thermal energy for industrial processes are the daily, monthly or annually thermal energy consumption time profiles. In recent years, several studies have discussed the possibilities of jointly using a solar installation and an industrial process. These studies highlight the advances that are still necessary in solar installations to be correctly coupled to industrial processes, analyse the potential in regions like Latin America [ 6 ]. Aristoteles Aidonis et al. [ 7 ] analyse the potential in the Mediterranean region and identify the most promising sectors within industry like food products and beverages and textiles. Pierres Energies 2018,11, 2950 3 of 30 Krummenacher et al. [ 8 ] identify practical constraints and analyse the complexity of heat supply in most industrial processes proposing a methodology to identify these points. N. Cottret et al. [9] evaluate the current market situation and finally identify crucial points yet to be solved, such as high investment costs, the lack of specific skills of many designers and installers, lack of public financing or low cost of conventional energies. In [ 10 ] is shown and overview of selected demonstration projects, proposing some actions, such as increase the demonstration projects to gain more experience, propose financial incentives to companies and promote training course for professionals. As medium temperature solar technologies adapt to industrial processes requirements, solar installations will become viable [ 11 , 12 ]. The literature review indicates that there is a lot of research about the comparison of Parabolic Trough Collector (PTC) and Linear Fresnel Collector (LFC) for electricity applications (Askaru et al. [13] , Sharma et al. [ 14 ], Rovira et al. [ 15 ]), but only a few researches for heat production for industrial processes. The application of these solar technologies for electricity production has thermal temperature level, control of the system, equipment and costs very different that for heat production for industrial processes. Accordingly, results and conclusions are not comparable. For instance, Rovira et al. [ 15 ], compare the annual performance and economic feasibility of integrated solar combined cycles, with PTC and LFC, using different gas turbines and different pressure levels that feed the steam turbine to produce electricity. They found that the PTC produces more useful energy but the LFC is more sustainable choice financially. Sharma et al. [ 14 ] compare PTC, LFC and Compact Linear Fresnel Reflector (CLFR) fields in terms of energy losses, net energy collection by fluid, electricity generation and cost of electricity for the location of Murcia (Spain). They found that there is no significant difference in the performance of LFC and CLFR field and the PTC is generally a better choice than the LFC financially. Daniele Cocco et al. [ 16 ] combined production of electricity and heat in the dairy sector using an Organic Rankine Cycle. They found that PTC and LFC could be a promising option if electricity and heat are both required. In this case, a suitable energy storage section that provides flexibility to the installation is required. Solar energy possibilities as source of energy supply for industrial processes have aroused the interest of many countries and several authors. There have been initiatives for the analysis of these possibilities in different countries, among which highlight the studies carried out in Australia [ 17 , 18 ], Germany [ 19 ], Tunisia [ 20 ] or Mexico [ 21 ]. Although the common objective of these studies is to analyse the viability of solar technologies as energy supply source for industrial processes, each study has been focused from the particular point of view of each country, that is, each study analyse the solar concentration technologies potential related to the predominant industrial process of the considered country. As consequence of the positive results of these studies and of the expectation created in the industrial sector there are a huge number of specific applications that are in the development process to achieve that solar energy technologies cover the industrial process thermal requirements [ 22 ]. There are also several reports that analyse, regardless the country, the solar technologies possibilities as thermal energy supplier for industrial processes, from the oil industry to the paper, textile or pharmaceutical industry [ 23 – 29 ]. Evangelos et al. [ 30 ] compare and evaluate energetically, exergetically and financially the performance of PTC and LFC for the climate conditions of Athens (Greece) for electricity and heat production. Results show the higher optical performance of PTC. During winter, LFC presents extremely low optical performance due to the low values of the IAM. Among the hypotheses made by Evangelos et al. stand out that they do not consider the Compound Parabolic Collector (CPC) technology as an alternative for the production of thermal energy, they evaluate the facility energy production at the solar field output without considering energy losses or thermal costs of distribution, exchange and storage system, they do not consider the operation and maintenance costs during the facility useful life and finally, they consider that the industrial process use all the annual thermal energy produced by the solar facility. As is already known, Spain was one of the pioneering countries in the development and implementation of solar energy as source of energy supply, in electrical or thermal energy form. The developments that were initially carried out focused on the photovoltaic solar energy, Energies 2018,11, 2950 4 of 30 low-temperature solar thermal energy and solar thermal energy sectors aimed primarily at generating electricity. Proofs of this golden age are the huge number of photovoltaic parks and solar thermal power plants that are currently working in Spain. In the specific case of solar thermal energy, it should be noted that there are three central receiver plants, two linear Fresnel plants and forty-five parabolic trough plants. Among them, they add up to a total of 2300 MW of installed power [ 31 ]. In the specific case of solar energy applied to the industrial sector, there have been several initiatives that, although they have not had the expected success, were useful to establish the bases on which work is currently being done. In recent years, the industrial sector has shown great interest in potential applications of solar energy for different industrial processes. Proof of this is that there are many companies that have focused their activities on obtaining new developments to take advantage of the solar sector in different industrial processes. The objective of this paper is to highlight the benefits of the use of solar thermal energy of medium temperature solar concentration technologies as thermal energy source of industrial processes. In addition to summarizing the potential industrial processes that can be used as thermal energy, all the necessary information about the most appropriate solar technologies is collected. For the specific case of Spain, the potential of thermal energy production for different locations, solar concentration technologies, plant sizes, thermal levels and percentages of use of the generated thermal energy is evaluated. After that, the thermal energy generation cost of medium temperature solar concentration technologies is compared, from the economic point of view, to conventional energy sources. Natural gas, electricity, gas oil and fuel oil cases are considered. A time horizon of 20 years and three different scenarios for the evolution of conventional energy source prices are evaluated. Finally, Greenhouse Gas emissions (GHG) avoided by using solar technologies instead of conventional energy sources are quantified. We have not found studies that analyse the medium temperature solar concentration technologies potential from technical and economic perspective that have into account the parameters included in this paper. This study aims to analyse the influence of the location, the medium temperature solar concentration technologies, the temperature level required by the industrial process, the percentage of used solar energy and the costs in the development of medium temperature solar concentration technologies. 2. Solar Thermal Energy Solar thermal energy (STE) allows solar radiation to be harnessed to generate thermal energy through the use of a heat transfer fluid. Subsequently, the thermal energy generated can be used in different processes, whether industrial, residential or commercial. One of the main advantages offered by the substitution of conventional energy sources by solar technologies is the contribution to the mitigation of climate change. The most widely used solar technologies are Flat Plate (FP), Compound Parabolic Collectors (CPC), Linear Fresnel Collectors (LFC) and Parabolic Trough Collectors (PTC). CPC, LFC and PTC technologies are the most used in the case of industrial processes. CPC vacuum tube collector is a system composed of a few rows of transparent glass tubes connected to a head pipe. Each tube contains therein an absorption tube coated with selective paint. Inside this pipe runs the heat transfer fluid. Vacuum is produced to minimize conduction and convection heat losses. Solar radiation passes through the glass over the tube, strikes the absorber tube and finally is transformed into heat. Overall performance of vacuum tube collector is higher than the conventional collector and maintains more constant behaviour. CPC collector includes annular reflectors that allow greater concentration of solar radiation onto the absorber tube. LFC is based on the idea of simulating a continuous concentrator, in this case a parabolic trough collector, as a set of elements. The costs associated with LFC technology are lower than the typical costs of PTC technology. These systems are composed of long parallel rows of mirrors of relatively small width which can rotate about its longitudinal axis. These mirrors concentrate solar radiation Energies 2018,11, 2950 5 of 30 on a fixed central receiver suspended at a certain height. The main element of this technology is the absorber tube, which is similar to the one used in parabolic trough collector systems. PTC, one of the most mature Concentrated Solar Power (CSP) technologies, consists of a series of parabolic reflectors that concentrate solar radiation on receiving pipes containing the heat transfer fluid that is heated throughout the process. These collectors are placed in parallel rows that make up the solar field aligned in a north-south or east-west axis. Receivers have a special coating to maximize energy absorption, minimize infrared re-irradiation and work in an evacuated glass envelope to avoid convection heat losses. In these cases solar heat is moved by a heat transfer fluid flowing in the receiver tube and transferred to a steam generator to produce the super-heated steam that runs the turbine. This section focuses on describing the instantaneous thermal efficiency and the cost structures of CPC, LFC and PTC technologies since these are the three alternatives considered in this paper. 2.1. Efficiency Characterization of CPC, LFC and PTC Technologies To quantify the thermal energy production is required to know the performance behaviour of technologies considered in this paper. The instantaneous thermal efficiency used for each medium temperature solar concentration technologies are described in detail below. 2.1.1. CPC Technology The compound parabolic collector characteristic efficiency equation is as follow: ηs f =k(θ)·η0−a1· ∆T Ig −a2· ∆T2 Ig(1) ηsf: Instantaneous efficiency [◦/1]. k: Incident angle modifier, where θis the incident angle. η0: Optical efficiency [◦/1]. a1: First order heat loss coefficient [W/K·m2]. a2: Second order heat loss coefficient [W/K2·m2]. ∆T: Difference between the mean fluid collector temperature and the ambient temperature [◦C]. Ig: Incident global radiation [W/m2]. To obtain the parameters that define the instantaneous efficiency curve described by the equation above, the information provided by several manufacturers is analyzed. Table 2summarizes the information collected. Table 2. CPC efficiency equation parameters. Technology η0a1a2 CPC-1 [32] 0.642 0.885 0.001 CPC-2 [33] 0.641 0.850 0.010 CPC-3 [34] 0.605 0.850 0.010 Figure 1shows the efficiency curves obtained using the information previously collected. A new curve named “average” is added; this has been calculated theoretically from values in Table 2. To take into account the effect of the incident angle modifier, the information provided in Table 3 has been considered. The parameter k( θ ) of Equation (1) is obtained as the product of k θb ( θT ) and k θb ( θL ). Energies 2018,11, 2950 6 of 30 Energies2018,11,xFORPEERREVIEW 5of30 maximizeenergyabsorption,minimizeinfraredre‐irradiationandworkinanevacuatedglass envelopetoavoidconvectionheatlosses.Inthesecasessolarheatismovedbyaheattransferfluid flowinginthereceivertubeandtransferredtoasteamgeneratortoproducethesuper‐heatedsteam thatrunstheturbine. Thissectionfocusesondescribingtheinstantaneousthermalefficiencyandthecoststructures ofCPC,LFCandPTCtechnologiessincethesearethethreealternativesconsideredinthispaper. 2.1.EfficiencyCharacterizationofCPC,LFCandPTCTechnologies Toquantifythethermalenergyproductionisrequiredtoknowtheperformancebehaviourof technologiesconsideredinthispaper.Theinstantaneousthermalefficiencyusedforeachmedium temperaturesolarconcentrationtechnologiesaredescribedindetailbelow. 2.1.1.CPCTechnology Thecompoundparaboliccollectorcharacteristicefficiencyequationisasfollow: 2 01 2 ()    sf gg TT kaa II (1) ŋsf:Instantaneousefficiency[°/1]. k:Incidentanglemodifier,whereθistheincidentangle. ŋ0:Opticalefficiency[°/1]. a1:Firstorderheatlosscoefficient[W/K∙m2]. a2:Secondorderheatlosscoefficient[W/K2∙m2]. ΔT:Differencebetweenthemeanfluidcollectortemperatureandtheambienttemperature[°C]. Ig:Incidentglobalradiation[W/m2]. Toobtaintheparametersthatdefinetheinstantaneousefficiencycurvedescribedbythe equationabove,theinformationprovidedbyseveralmanufacturersisanalyzed.Table2 summarizestheinformationcollected. Table2.CPCefficiencyequationparameters. Technology η0  1 a2 a  CPC‐1[32]0.6420.8850.001 CPC‐2[33]0.6410.8500.010 CPC‐3[34]0.6050.8500.010 Figure1showstheefficiencycurvesobtainedusingtheinformationpreviouslycollected.A newcurvenamed“average”isadded;thishasbeencalculatedtheoreticallyfromvaluesinTable2.  Energies2018,11,xFORPEERREVIEW 6of30  Figure1.CPCefficiencycurves. Totakeintoaccounttheeffectoftheincidentanglemodifier,theinformationprovidedinTable 3hasbeenconsidered.Theparameterk(θ)ofEquation(1)isobtainedastheproductofkθb(θT)and kθb(θL). Table3.Incidentanglemodifier[35]. θ(°)01020304050607090 kθb(θT)1.001.051.081.201.361.531.812.030.00 kθb(θL)1.001.000.990.990.980.960.920.840.00 2.1.2.LFCTechnology TheCPCtechnologicalmaturityanditsmarketpenetrationarequitehigherfromthesituation inwhichLFCtechnologyislocated;asconsequencetheLFCtechnologyavailableinformationis muchscarcer.TheinstantaneousefficiencyofFresneltechnologyisasfollow[36]:  012 ·· () sf bc T ccT I     (2) ŋsf:Instantaneousefficiency[°/1]. ŋ0:Opticalefficiency[°/1]. c1:Linealheatlosscoefficient[W/K∙m2]. c2:Quadraticheatlosscoefficient[W/K2∙m2]. ΔT:Differencebetweenthemeanfluidtemperature(Tm)andtheambienttemperature(Ta)[K]. Ibc(θ):Incidentdirectnormalradiationonthecollector,whereθistheincidentangle.[W/m2]. Theincidentdirectnormalradiationonthecollector(Ibc)usedintheefficiencyexpression aboveisthatresultingfromtheproductofdirectnormalradiationandtheincidentanglecosine.The incidentangleforthecaseinwhichthetrackingsystemisNorth‐Southisdeterminedaccordingthe followingexpression[37].  22 cos cos( ) cos( ) cos( ) tan( ) sin( ) sin ( ) hor hor a decli lat ang decli lat ang         (3) decli:Declination[°]. lat:Latitude[°]. anghor:Hourlyangle[°]. Efficiency Figure 1. CPC efficiency curves. Table 3. Incident angle modifier [35]. θ(◦) 0 10 20 30 40 50 60 70 90 kθb(θT) 1.00 1.05 1.08 1.20 1.36 1.53 1.81 2.03 0.00 kθb(θL) 1.00 1.00 0.99 0.99 0.98 0.96 0.92 0.84 0.00 2.1.2. LFC Technology The CPC technological maturity and its market penetration are quite higher from the situation in which LFC technology is located; as consequence the LFC technology available information is much scarcer. The instantaneous efficiency of Fresnel technology is as follow [36]: ηs f =η0−[c1+c2·∆T]· ∆T Ibc(θ)(2) ηsf: Instantaneous efficiency [◦/1]. η0: Optical efficiency [◦/1]. c1: Lineal heat loss coefficient [W/K·m2]. c2: Quadratic heat loss coefficient [W/K2·m2]. ∆T: Difference between the mean fluid temperature (Tm) and the ambient temperature (Ta) [K]. Ibc(θ): Incident direct normal radiation on the collector, where θis the incident angle [W/m2]. The incident direct normal radiation on the collector (Ibc) used in the efficiency expression above is that resulting from the product of direct normal radiation and the incident angle cosine. The incident Energies 2018,11, 2950 7 of 30 angle for the case in which the tracking system is North-South is determined according the following expression [37]. θ=acoscos(decli)·q(cos(lat)·cos(anghor) + tan(decli)·sin(lat))2+sin2(anghor)(3) decli: Declination [◦]. lat: Latitude [◦]. anghor: Hourly angle [◦]. Figure 2shows the proposed LFC efficiency curve, considering direct steam generation, a 20 ◦ C degrees ambient temperature (Ta) and 1000 W/m 2 incident radiation on the collector. The parameters of the equation above η0,c1and c2are 0.576, 0.000 y 0.0004 respectively. Energies2018,11,xFORPEERREVIEW 7of30 Figure2showstheproposedLFCefficiencycurve,consideringdirectsteamgeneration,a20°C degreesambienttemperature(Ta)and1000W/m2incidentradiationonthecollector.Theparameters oftheequationaboveη0,c1andc2are0.576,0.000y0.0004respectively.  Figure2.LFCefficiencycurve,Ta=20°C,Ibc=1000W/m2. TheLFCtechnologyefficiencycurveshapeisconsistentwiththeoneproposedbyEvangeloset al.[30],althoughitshowsslightlylowerefficiencyvalues. 2.1.3.PTCTechnology AsmentionedaboveaninstantaneousthermalefficiencycurvehasbeendefinedforCPCand LFCtechnologies.InthecaseofPTCtechnologyitisnotadvisabletouseanadjustmentlikethatdue tothisisasignificantlymorecomplextechnology.Onthisoccasion,anenergybalancewhichaimis toknowthethermalenergyproductionbythesolarinstallationfromtheincidentsolarradiationis made.Thelossesinvolvedintheprocessoftransformingsolarradiationintothermalenergyare dividedintogeometric,opticalandthermal[38].CurrentlythereisquitereliableinformationofPTC technologyusedforelectricalgenerationusingthermalfluidtemperaturesaround400°C.Asthe thermalanalysislevelchosenforthisstudyis350°C,ithasbeendecidedtousetheavailabledata fromPTCtechnologyforelectricalproduction.Theexpressiontocalculatethethermalenergy productionbythesolarinstallationisasfollow: __ _ mod _solar field output incident solar shadow soiling peak optical thermal EEFFk t    (4) _cos incident solar c bn ESI   (5) sin 2       ec shadow track ccp L Fteta aper (6) _    peak optical (7) 442 63 84 mod 1 2.23073 10 1.1 10 3.18596 10 4.8509 10            k(8) Esolar_field_output:Energyattheoutputofthesolarfield[Wh]. Eincident_solar:Energysolarradiation[W]. Fshadow:Shadowfactor[°/1]. Fsoiling:Soilingfactor[°/1]. kmod:Incidenceanglemodifier[°/1]. ηpeak_optical:Peakopticalefficiency[%]. Figure 2. LFC efficiency curve, Ta = 20 ◦C, Ibc = 1000 W/m2. The LFC technology efficiency curve shape is consistent with the one proposed by Evangelos et al. [30], although it shows slightly lower efficiency values. 2.1.3. PTC Technology As mentioned above an instantaneous thermal efficiency curve has been defined for CPC and LFC technologies. In the case of PTC technology it is not advisable to use an adjustment like that due to this is a significantly more complex technology. On this occasion, an energy balance which aim is to know the thermal energy production by the solar installation from the incident solar radiation is made. The losses involved in the process of transforming solar radiation into thermal energy are divided into geometric, optical and thermal [ 38 ]. Currently there is quite reliable information of PTC technology used for electrical generation using thermal fluid temperatures around 400 ◦ C. As the thermal analysis level chosen for this study is 350 ◦ C, it has been decided to use the available data from PTC technology for electrical production. The expression to calculate the thermal energy production by the solar installation is as follow: Esolar_f ield_output =Eincident_solar ·Fshadow ·Fsoiling ·kmod ·ηpeak_optical ·ηthermal ·∆t(4) Eincident_solar =Sc·Ibn ·cos φ(5) Fshadow =sinπ 2−tetatrack ·Lec aperccp (6) ηpeak_optical =ρ·α·τ·γ(7) kmod =h1−2.23073 ×10−4·φ−1.1 ×10−4·φ2+3.18596 ×10−6·φ3−4.8509 ×10−8·φ4i(8) Energies 2018,11, 2950 8 of 30 Esolar_field_output: Energy at the output of the solar field [Wh]. Eincident_solar: Energy solar radiation [W]. Fshadow: Shadow factor [◦/1]. Fsoiling: Soiling factor [◦/1]. kmod: Incidence angle modifier [◦/1]. ηpeak_optical: Peak optical efficiency [%]. ηthermal: Thermal efficiency [%]. ∆t: Time interval [h]. Sc: Reflective surface opening area [m2]. Ibn: Direct normal radiation [W/m2]. Φ: Incidence angle [◦]. tetatrack: Parabolic trough collector track angle [◦]. Lec: Distance between rows of collectors from center to center [m]. aperCCP: Opening width of the collectors [m]. ρ: Reflectance [◦/1]. α: Interception factor [◦/1]. τ: Transmittance [◦/1]. γ: Absorption [◦/1]. To advance in the analysis that is intended to be carried out in the framework of this study, a 0.93 thermal efficiency and the approximate values of the following variables are considered: •Mirrors reflectance: 0.92 [◦/1]. •Cover transmittance: 0.965 [◦/1]. •Receiver absorption: 0.96 [◦/1]. •Interception factor: 0.95 [◦/1]. The tracking system considered in this case, as above, is North-South. Figure 3shows as example the hourly performance curve in terms of thermal energy production for the three technologies and thermal levels considered in this paper for the particular case of Seville and the 20 June. The radiation data used are those corresponding to the representative solar year extracted from Meteonorm. Energies2018,11,xFORPEERREVIEW 8of30 ηthermal:Thermalefficiency[%]. Δt:Timeinterval[h]. Sc:Reflectivesurfaceopeningarea[m2]. Ibn:Directnormalradiation[W/m2]. Φ:Incidenceangle[°]. tetatrack:Parabolictroughcollectortrackangle[°]. Lec:Distancebetweenrowsofcollectorsfromcentertocenter[m]. aperCCP:Openingwidthofthecollectors[m]. ρ:Reflectance[°/1]. α:Interceptionfactor[°/1]. τ:Transmittance[°/1]. γ:Absorption[°/1]. Toadvanceintheanalysisthatisintendedtobecarriedoutintheframeworkofthisstudy,a 0.93thermalefficiencyandtheapproximatevaluesofthefollowingvariablesareconsidered:  Mirrorsreflectance:0.92[°/1].  Covertransmittance:0.965[°/1].  Receiverabsorption:0.96[°/1].  Interceptionfactor:0.95[°/1]. Thetrackingsystemconsideredinthiscase,asabove,isNorth‐South. Figure3showsasexamplethehourlyperformancecurveintermsofthermalenergy productionforthethreetechnologiesandthermallevelsconsideredinthispaperfortheparticular caseofSevilleandthe20June.Theradiationdatausedarethosecorrespondingtotherepresentative solaryearextractedfromMeteonorm.  Figure3.Thermalenergyproductionhourlycurves. 2.2.CPC,LFCandPTCEconomicParameters Thepurposeofthissubsectionistodeterminethecoststructureofsolarinstallationsinwhich CPC,LFCandPTCtechnologiesareemployed.Thisisacomplextasksincethiskindofeconomic informationisnotusuallyavailable,itsreliabilityisnotassuredanditisnotcertainlyknownifthis informationisproperlyupdated.Thisdifficultyincreasesevenmorewhentryingtoobtainthese costsdependingonthesizeofthesolarinstallation.Tables4–6summariseinvestmentcosts(CI), annualoperationandmaintenancecosts(COM)andreplacementcosts(CR)foralltechnologies.CIis Figure 3. Thermal energy production hourly curves. 2.2. CPC, LFC and PTC Economic Parameters The purpose of this subsection is to determine the cost structure of solar installations in which CPC, LFC and PTC technologies are employed. This is a complex task since this kind of economic Energies 2018,11, 2950 9 of 30 information is not usually available, its reliability is not assured and it is not certainly known if this information is properly updated. This difficulty increases even more when trying to obtain these costs depending on the size of the solar installation. Tables 4–6summarise investment costs (C I ), annual operation and maintenance costs (C OM ) and replacement costs (C R ) for all technologies. C I is expressed as a function of the solar field area, C OM and C R are expressed as a percentage of the C I . Costs described in this section do not include auxiliary energy or industrial processes costs. Table 4. CPC technology costs. Solar Field Area Size- Solar Field Area [m2] CI[€/m2]COM [%CI] CR[%CI] Small 50 325 2.5% 10% Large 2000 225 1.5% 10% Table 5. LFC technology costs. Solar Field Area Size- Solar Field Area [m2] CI[€/m2]COM [%CI] CR[%CI] Small 100 425 5% 10% Large 15,000 260 4% 10% Table 6. PTC technology costs. -Solar Field Area Size Solar Field Area [m2] CI[€/m2]COM [%CI] CR[%CI] Small 100 560 5.5% 10% Large 15,000 330 4.5% 10% 2.2.1. CPC Technology Several studies provide information about CPC technology costs [ 5 , 36 ]. The first study indicates that CPC technology costs ranges from 400 € /m 2 to 800 € /m 2 . This is an average value for the entire European market. The second study indicates that the complete installation cost varies from 857 € /m 2 to 730 € /m 2 if the solar field area ranges from 50 m 2 to 5000 m 2 . In addition to the information provided by these studies sector experts have been consulted. They indicate that in both cases these cost reflect specific situations and that in a market with a representative demand, for sizes over 50 m 2 and for updated prices, costs are significantly lower. Based on the gathered information, it has been estimated that the investment cost ranges between 325 € /m 2 and 225 € /m 2 if the solar field area varies from 50 m 2 to 2000 m 2 . Once this size has been reached, the investment cost per solar field area unit remains constant. These figures include investments relating to the storage system. 2.2.2. LFC Technology To determine the LFC technology installation cost two studies are considered [ 36 , 39 ]. Although these studies aim to determine the installation cost of facilities in which electricity is generated, solar field area costs are used as reference. The first study estimates that the solar field area cost is about 156 € /m 2 , in the second one this parameter is about 217 € /m 2 . This paper considers the information provided by the first study since it focuses on the Spanish market. Since there is no economical information about the storage system, exchanger, control system and other elements included in the solar installation group, this cost is estimated about 100 € /m 2 . Table 4shows the costs associated with a small and a large size solar field area. This paper considers that a LFC technology installation is large if its solar field area is equal or greater than 15,000 m 2 . This is not comparable with those installations whose objective is the generation of electrical energy. Energies 2018,11, 2950 16 of 30 Cli f espan = CI+20 ∑ i=1 COM+CR (1+r)n(13) The thermal energy unit cost (C thermal_emergy_unit ) is obtained according to the following expression: Cthermal_energy_unit =Cli f espan ATE (14) 4.6. Analysis of Environmental Advantages The last stage of the proposed methodology quantifies the GHG emission avoided by the use of solar concentration technologies instead of conventional sources of energy. For this purpose it is essential to obtain the quantity of conventional source of energy that produce an equivalent amount of thermal energy to the one generated by the medium temperature solar concentration technology installation. To evaluate the equivalent amount of electricity (E e ) is considered Joule effect. The GHG emissions avoided by the use of a solar system instead of electricity (GHG e ) are calculated using the electricity conversion factor (FPe): GHGe=Ee·FPe(15) In the case of natural gas it is considered the use of a boiler. Thermal energy is generated by a combustion process. The natural gas lower heating value (LHV) and the efficiency boiler ( ηb ) are 8.18 kWh/m 3 and 96% respectively [ 48 ]. The volume (V) of natural gas used is calculated according to the following expressions: V=EIP/(LHV·ηb)(16) GHG emissions avoided by the use of a solar system instead of natural gas are obtained as follow, taking into account that FPng represents the natural gas conversion factor: GHGng =V·FPng (17) Similar expressions are used for the cases of fuel oil and diesel oil, when LHV values are 11.08 kWh/kg and 10.28 kWh/l respectively [49,50]. 5. Application and Results Throughout this section, the application of the methodology previously described is detailed and the results obtained are shown. 5.1. Site Selection Table 14 lists the sites selected for this study, these are also shown in Figure 8. Two cities have been chosen for each climatic zone. It is considered that this selection will provide representative results. For each location, the name of the city, the climatic zone, the latitude, the longitude and the height above the sea level have been included. Energies 2018,11, 2950 17 of 30 Table 14. Selected sites geographical data. Site Climatic Zone Latitude (◦) Longitude (◦) Height (m) La Coruña I43.367 −8.417 67 Vitoria 42.850 −2.670 550 Barcelona II 41.283 2.067 6 Valladolid 41.650 −4.767 739 Salamanca III 40.970 −5.670 823 Teruel 40.260 −1.105 954 JaénIV 37.770 3.800 697 Valencia 39.480 −0.380 13 Cáceres V39.467 −6.333 405 Sevilla 37.410 −6.010 7 Energies2018,11,xFORPEERREVIEW 17of30 kWh/m3and96%respectively[49].Thevolume(V)ofnaturalgasusediscalculatedaccordingtothe followingexpressions:  /? I Pb VE LHV  (16) GHGemissionsavoidedbytheuseofasolarsysteminsteadofnaturalgasareobtainedas follow,takingintoaccountthatFPngrepresentsthenaturalgasconversionfactor: · ng ng GHG V FP(17) Similarexpressionsareusedforthecasesoffueloilanddieseloil,whenLHVvaluesare11.08 kWh/kgand10.28kWh/lrespectively[50,51]. 5.ApplicationandResults Throughoutthissection,theapplicationofthemethodologypreviouslydescribedisdetailed andtheresultsobtainedareshown. 5.1.SiteSelection Table14liststhesitesselectedforthisstudy,thesearealsoshowninFigure8.Twocitieshave beenchosenforeachclimaticzone.Itisconsideredthatthisselectionwillproviderepresentative results.Foreachlocation,thenameofthecity,theclimaticzone,thelatitude,thelongitudeandthe heightabovethesealevelhavebeenincluded. Table14.Selectedsitesgeographicaldata. SiteClimaticZoneLatitude(°)Longitude(°)Height(m) LaCoruñaI43.367−8.41767 Vitoria42.850−2.670550 BarcelonaII41.2832.0676 Valladolid41.650−4.767739 SalamancaIII40.970−5.670823 Teruel40.260−1.105954 JaénIV37.7703.800697 Valencia39.480−0.38013 CáceresV39.467−6.333405 Sevilla37.410−6.0107  Figure8.Selectedsites.Source:GoogleEarth. Figure 8. Selected sites. Source: Google Earth. 5.2. Solar Resource Evaluation Tables 15 and 16 summarize the accumulated monthly and annual global horizontal radiation (I g0 ) and direct normal radiation (I bn ), in both cases for each one of the sites listed in Table 14. Annual global horizontal radiation ranges from 1315 kWh/m 2 to 1927 kWh/m 2 while annual direct normal radiation ranges from 1220 kWh/m 2 to 2329 kWh/m 2 . Monthly and annual accumulated radiation values shown in tables below have been calculated from hourly values obtained throughout Meteonorm software. Although only a summary of these values have been included in this section, radiometric and meteorological hourly values have been used for all calculations. For instance, all solar collector performance values have been calculated from hourly data. Table 15. Global Horizontal Radiation. Site Monthly Global Horizontal Radiation (kWh/m2)Annual Accumulated (kWh/m2) 1 2 3 4 5 6 7 8 9 10 11 12 La Coruña 43 68 109 138 168 185 191 172 130 83 48 38 1373 Vitoria 43 62 105 127 163 178 187 157 125 83 47 38 1315 Barcelona 64 83 131 162 194 202 217 184 139 104 67 56 1603 Valladolid 51 80 128 158 182 223 229 200 146 97 57 44 1595 Salamanca 60 81 132 163 199 222 239 205 155 102 65 53 1676 Teruel 67 84 133 164 201 220 249 211 158 116 76 59 1738 Jaén 84 83 143 180 212 240 261 229 166 121 93 76 1888 Valencia 67 91 135 167 188 203 209 179 137 110 72 60 1618 Cáceres 68 91 142 173 205 225 240 211 154 110 74 54 1747 Sevilla 85 95 151 182 222 240 257 227 170 127 95 76 1927 Energies 2018,11, 2950 18 of 30 Table 16. Direct Normal Radiation. Site Monthly Direct Normal Radiation (kWh/m2)Annual Accumulated (kWh/m2) 1 2 3 4 5 6 7 8 9 10 11 12 La Coruña 61 80 101 115 137 161 173 157 132 91 61 51 1320 Vitoria 48 63 93 109 141 154 153 144 127 94 48 46 1220 Barcelona 104 108 141 157 180 174 213 166 146 124 94 88 1695 Valladolid 78 108 146 157 167 239 247 221 166 115 75 58 1777 Salamanca 103 111 156 178 185 234 270 234 191 120 91 83 1956 Teruel 101 111 152 165 199 229 292 237 197 156 119 104 2062 Jaén 142 124 156 187 211 265 310 280 199 147 162 132 2315 Valencia 97 126 144 153 170 178 195 152 139 134 94 92 1674 Cáceres 101 132 170 183 200 225 261 239 185 135 110 72 2013 Sevilla 146 122 174 192 224 250 298 276 196 158 159 134 2329 Among the meteorological variables that the Meteonorm software provides, it stands out by its influence in this analysis the ambient temperature. Table 16 shows the average monthly ambient temperature (T a ). As in tables above, the average monthly ambient temperature values included in Table 17 have been calculated from the hourly values obtained throughout Meteonorm software. Table 17. Ambient temperature. Site Average Monthly Ambient Temperature (◦C) 1 2 3 4 5 6 7 8 9 10 11 12 La Coruña 10.8 10.9 12.3 12.5 14.9 17.4 18.9 19.3 18.1 16.2 12.8 11.0 Vitoria 5.2 5.9 8.6 10.1 14.0 17.9 19.3 19.4 16.3 13.3 8.2 5.3 Barcelona 9.0 9.9 12.2 14.0 17.5 21.5 24.2 24.5 21.3 18.1 12.6 9.5 Valladolid 3.8 5.3 8.7 10.5 14.8 20.2 22.0 21.6 17.6 12.9 7.0 4.2 Salamanca 3.5 5.0 8.3 10.0 14.8 19.7 21.2 20.8 16.7 12.8 6.7 4.1 Teruel 4.0 5.6 9.1 11.1 15.9 21.9 24.6 23.8 19.0 13.7 7.4 4.3 Jaén 5.9 8.5 11.6 13.5 18.2 23.9 26.3 25.6 20.8 16.2 9.7 6.8 Valencia 9.9 11.1 13.8 15.5 19.1 23.3 26.0 26.1 22.6 19.2 13.6 10.9 Cáceres 7.8 9.6 12.7 14.0 18.9 24.7 26.9 26.9 22.7 17.4 11.2 8.3 Sevilla 11.3 13.2 16.1 17.8 22.1 26.5 28.8 28.7 24.8 20.9 14.9 12.0 5.3. Estimation of Thermal Energy Available at the Solar Installation Output for Each Site, Technology and Thermal Level Analysed in This Paper This subsection shows the results of the estimation of thermal energy available at the solar installation output for each option considered in this paper. To achieve this purpose, the associated information to each solar technology efficiency (Section 2), the equations proposed to estimate the generated thermal energy (Section 4.4) and, naturally the radiometric and meteorological information summarized at the beginning of this section are used. Table 18 summarizes above-mentioned results for the three thermal levels in which it is considered that the use of a CPC technology solar installation can be beneficial, 100 ◦ C, 125 ◦ C and 150 ◦ C. These temperatures refer to the average fluid temperature. This table also summarises the average efficiency for each case. As the table above, Table 19 shows the results associated with the thermal energy available in the storage system of a Fresnel technology solar installation and the average efficiency. In this case, two thermal levels are considered, 170 ◦C and 220 ◦C. Lastly, Table 20 summarizes the generated thermal energy per unit area in a PTC technology solar installation and the average efficiency. Energies 2018,11, 2950 19 of 30 Table 18. Thermal energy available per solar field area and average efficiency for each site and thermal level, CPC technology. Site Thermal Energy Available in the Storage System by Solar Field Area [kWht/m2]Average Efficiency 100 ◦C 125 ◦C 150 ◦C 100 ◦C 125 ◦C 150 ◦C La Coruña 672 557 424 0.43 0.36 0.28 Vitoria 613 506 382 0.42 0.34 0.26 Barcelona 863 731 578 0.46 0.39 0.31 Valladolid 816 691 546 0.45 0.38 0.30 Salamanca 895 762 609 0.46 0.39 0.31 Teruel 923 788 631 0.46 0.39 0.31 Jaén 1058 916 751 0.48 0.42 0.34 Valencia 873 743 592 0.47 0.40 0.32 Cáceres 937 802 645 0.47 0.40 0.32 Sevilla 1095 951 783 0.49 0.42 0.35 Table 19. Thermal energy available per solar field area and average efficiency for each site and thermal level, LFC technology. Site Thermal Energy Available in the Storage System by Solar Field Area [kWht/m2]Average Efficiency 170 ◦C 220 ◦C 170 ◦C 220 ◦C La Coruña 478 465 0.36 0.35 Vitoria 437 426 0.36 0.35 Barcelona 623 607 0.37 0.36 Valladolid 676 660 0.38 0.37 Salamanca 744 728 0.38 0.37 Teruel 797 780 0.39 0.38 Jaén 896 877 0.39 0.38 Valencia 622 606 0.37 0.36 Cáceres 779 762 0.39 0.38 Sevilla 913 893 0.39 0.38 Table 20. Thermal energy available per solar field area and average efficiency for each site, PTC technology. Site Thermal Energy Available in the Storage System by Solar Field Area [kWht/m2]Average Efficiency 350 ◦C La Coruña 496 0.38 Vitoria 466 0.38 Barcelona 632 0.37 Valladolid 696 0.39 Salamanca 757 0.39 Teruel 793 0.38 Jaén 869 0.37 Valencia 628 0.38 Cáceres 794 0.39 Sevilla 918 0.39 Tables above show that all technologies show common operation standards for all sites, mainly depending on their characteristic solar resource available. CPC technology shows general downgrade of thermal energy generated by solar systems when working temperature increases, a similar behaviour, although softer, is observed in the case of LFC technology. These results are close to the expected. At low fluid temperatures (around 100 ◦ C) the most recommended technology from the thermal Energies 2018,11, 2950 20 of 30 energy generation point of view is CPC, as this temperature increases; it goes to PTC technology, going through LFC technology. Between sites considered in this paper Sevilla stands out as the site which greater generated thermal energy values. The results of Vitoria place it at the other extreme. 5.4. Thermal Energy Unit Cost (c€/kWht) Tables 21–23 summarize the thermal energy generation cost for the different sites and each of the medium temperature solar concentration technologies analysed in this paper. These tables also differentiate results depending on the percentage of solar use, the average fluid temperature and the solar field area size. The percentage of solar use parameter is related to the coupling in time between the thermal energy generation and the demand by the industrial process. Accordingly, the role of the storage system is essential due to this is the component of the solar installation that allows decoupling supply and demand. The average fluid temperature is defined, as mentioned above, by the industrial process requirements. The last parameter considered in this analysis is the size of the solar field area, it affects mainly from the economic point of view, due to the reduction of costs that usually occurs when the solar field area is increased. Table 21. Thermal energy unit cost, CPC technology. Site Annual Global Horizontal Radiation (kWh/m2) Average Ambient Temperature (◦C) PSU (%) Thermal Energy Unit Cost (c€/kWht) Solar Field Area (m2) 50 2000 Average Fluid Temperature (◦C) 100 125 150 100 125 150 La Coruña 1372.5 14.6 100 3.5 4.2 5.5 2.2 2.6 3.4 75 4.7 5.6 7.4 2.9 3.5 4.6 50 7.0 8.4 11.1 4.3 5.2 6.9 Vitoria 1315.5 12.0 100 3.8 4.6 6.2 2.4 2.9 3.8 75 5.1 6.2 8.2 3.2 3.8 5.1 50 7.7 9.3 12.3 4.8 5.8 7.6 Barcelona 1600.5 16.2 100 2.7 3.2 4.1 1.7 2.0 2.5 75 3.6 4.3 5.4 2.3 2.7 3.4 50 5.4 6.4 8.1 3.4 4.0 5.1 Valladolid 1594.0 12.4 100 2.9 3.4 4.3 1.8 2.1 2.7 75 3.8 4.5 5.7 2.4 2.8 3.6 50 5.8 6.8 8.6 3.6 4.2 5.3 Salamanca 1674.2 12.0 100 2.6 3.1 3.9 1.6 1.9 2.4 75 3.5 4.1 5.1 2.2 2.6 3.2 50 5.3 6.2 7.7 3.3 3.8 4.8 Teruel 1738.0 13.4 100 2.5 3.0 3.7 1.6 1.9 2.3 75 3.4 4.0 5.0 2.1 2.5 3.1 50 5.1 6.0 7.4 3.2 3.7 4.6 Jaén1897.6 15.6 100 2.2 2.6 3.1 1.4 1.6 1.9 75 3.0 3.4 4.2 1.8 2.1 2.6 50 4.4 5.1 6.3 2.8 3.2 3.9 Valencia 1616.6 17.6 100 2.7 3.2 4.0 1.7 2.0 2.5 75 3.6 4.2 5.3 2.2 2.6 3.3 50 5.4 6.3 7.9 3.3 3.9 4.9 Cáceres 1742.2 16.8 100 2.5 2.9 3.6 1.6 1.8 2.3 75 3.3 3.9 4.9 2.1 2.4 3.0 50 5.0 5.9 7.3 3.1 3.6 4.5 Sevilla 1926.1 19.8 100 2.1 2.5 3.0 1.3 1.5 1.9 75 2.9 3.3 4.0 1.8 2.0 2.5 50 4.3 4.9 6.0 2.7 3.1 3.7 Energies 2018,11, 2950 21 of 30 Table 22. Thermal energy unit cost, LFC technology. Site Annual Direct Normal Radiation (kWh/m2) Average Ambient Temperature (◦C) PSU (%) Thermal Energy Unit Cost (c€/kWht) Solar Field Area (m2) 100 15,000 Average Fluid Temperature (◦C) 170 220 170 220 La Coruña 1320.1 14.6 100 8.1 8.3 4.5 4.7 75 10.8 11.1 6.1 6.2 50 16.2 16.6 9.1 9.3 Vitoria 1219.7 12.0 100 8.8 9.1 5.0 5.1 75 11.8 12.1 6.6 6.8 50 17.7 18.1 9.9 10.2 Barcelona 1694.1 16.2 100 6.2 6.4 3.5 3.6 75 8.3 8.5 4.6 4.8 50 12.4 12.7 7.0 7.2 Valladolid 1777.9 17.6 100 5.7 5.9 3.2 3.3 75 7.6 7.8 4.3 4.4 50 11.4 11.7 6.4 6.6 Salamanca 1955.2 12.0 100 5.2 5.3 2.9 3.0 75 6.9 7.1 3.9 4.0 50 10.4 10.6 5.8 6.0 Teruel 2061.0 16.8 100 4.8 5.0 2.7 2.8 75 6.5 6.6 3.6 3.7 50 9.7 9.9 5.4 5.6 Jaén2314.6 15.6 100 4.3 4.4 2.4 2.5 75 5.7 5.9 3.2 3.3 50 8.6 8.8 4.8 4.9 Valencia 1674.5 12.4 100 6.2 6.4 3.5 3.6 75 8.3 8.5 4.7 4.8 50 12.4 12.8 7.0 7.2 Cáceres 2012.4 13.4 100 5.0 5.1 2.8 2.8 75 6.6 6.8 3.7 3.8 50 9.9 10.1 5.6 5.7 Sevilla 2328.3 19.8 100 4.2 4.3 2.4 2.4 75 5.6 5.8 3.2 3.2 50 8.5 8.7 4.8 4.9 Table 23. Thermal energy unit cost, PTC technology. Site Annual Direct Normal Radiation (kWh/m2) Average Ambient Temperature (◦C) PSU (%) Thermal Energy Unit Cost (c€/kWht) Solar Field Area (m2) 100 15,000 Average Fluid Temperature (◦C) 350 La Coruña 1320.1 14.6 100 10.7 5.8 75 14.2 7.7 50 21.4 11.6 Vitoria 1219.7 12.0 100 11.4 6.2 75 15.2 8.2 50 22.7 12.3 Barcelona 1694.1 16.2 100 8.4 4.6 75 11.2 6.1 50 16.8 9.1 Valladolid 1777.9 17.6 100 7.6 4.1 75 10.2 5.5 50 15.2 8.3 Energies 2018,11, 2950 22 of 30 Table 23. Cont. Site Annual Direct Normal Radiation (kWh/m2) Average Ambient Temperature (◦C) PSU (%) Thermal Energy Unit Cost (c€/kWht) Solar Field Area (m2) 100 15,000 Average Fluid Temperature (◦C) 350 Salamanca 1955.2 12.0 100 7.0 3.8 75 9.3 5.1 50 14.0 7.6 Teruel 2061.0 16.8 100 6.7 3.6 75 8.9 4.8 50 13.4 7.3 Jaén2314.6 15.6 100 6.1 3.3 75 8.1 4.4 50 12.2 6.6 Valencia 1674.5 12.4 100 8.4 4.6 75 11.3 6.1 50 16.9 9.2 Cáceres 2012.4 13.4 100 6.7 3.6 75 8.9 4.8 50 13.3 7.2 Sevilla 2328.3 19.8 100 5.8 3.1 75 7.7 4.2 50 11.5 6.3 To analyse in a simple way the results shown in the tables above, Figures 9and 10 have been included. These graphs show the thermal energy generation cost for each medium temperature solar concentration technology depending on the average fluid temperature and the percentage of solar use. The thermal energy generation cost range represented by each bar of these graphs is related to the solar resource variability, which in turn is connected with the sites selected at the beginning of this section. As already stated, the solar field area is a significant parameter from the economic point of view, thus the graphic representation has been broken down into two graphs. Figure 9represents the thermal energy generation cost for small size solar field areas and Figure 10 for large ones. Energies2018,11,xFORPEERREVIEW 23of30  Figure9.Thermalenergygenerationcost—Smallsizesolarfieldarea.  Figure10.Thermalenergygenerationcost—Largesizesolarfieldarea. ThepreviousgraphsshowthePSUparameter,fluidtemperatureandsolarfieldarea importancehavesignificantinfluenceonthethermalenergygenerationcost.Ascanbeobservedfor allbargroups,thethermalenergygenerationcostincreasesproportionallyasthePSUdecreases.If CPCtechnologyresultsareanalysed,theextremesituationisobservedwhentheaveragefluid temperatureisaround150°C.Theinfluenceoftheaveragefluidtemperatureislowerinthecaseof LFCtechnology.Theseresultsshowagaintheimportanceofthecouplingintimebetweenthe thermalenergygeneratedbythesolarinstallationandthethermalenergydemandedbythe industrialprocess.Ingeneral,termsandaccordingtotheexpressionsproposedinSection4,itcanbe statedthatbyreducingthePSUfrom100%to50%,doublethethermalenergygenerationcost. Moreover,whencomparinggraphsabove,theinfluenceofthesolarfieldareaisalsoevident. AsexpectedbythecoststructureshowninSection2,thecostsofthermalenergygeneration(directly dependentoninvestmentcosts)arereducedbyincreasingthesolarfieldarea.Thisreductionis especiallystrikinginthecaseofPTCtechnology. RegardingtheparticularcaseofSpainandconverselyofwhathappenedwhenanalyzingthe thermalenergygeneratedbythesolarinstallation,highercostsarelinkedtoVitoria,whilelower costsforSeville. Finally,Table24summarizesthethermalenergyunitcostsofconventionalenergysources analysedinthispaperconsideringthethreescenariosdescribedinSection3.Giventhewiderange ofexistingelectricityandnaturalgasratesdependingtheconsumerscharacteristics,thetwoextreme groupshasbeenincludedinthistable,I1andI6GroupsandIAandIGGroupsrespectively.Asin Figure 9. Thermal energy generation cost—Small size solar field area. Energies 2018,11, 2950 23 of 30 Energies2018,11,xFORPEERREVIEW 23of30  Figure9.Thermalenergygenerationcost—Smallsizesolarfieldarea.  Figure10.Thermalenergygenerationcost—Largesizesolarfieldarea. ThepreviousgraphsshowthePSUparameter,fluidtemperatureandsolarfieldarea importancehavesignificantinfluenceonthethermalenergygenerationcost.Ascanbeobservedfor allbargroups,thethermalenergygenerationcostincreasesproportionallyasthePSUdecreases.If CPCtechnologyresultsareanalysed,theextremesituationisobservedwhentheaveragefluid temperatureisaround150°C.Theinfluenceoftheaveragefluidtemperatureislowerinthecaseof LFCtechnology.Theseresultsshowagaintheimportanceofthecouplingintimebetweenthe thermalenergygeneratedbythesolarinstallationandthethermalenergydemandedbythe industrialprocess.Ingeneral,termsandaccordingtotheexpressionsproposedinSection4,itcanbe statedthatbyreducingthePSUfrom100%to50%,doublethethermalenergygenerationcost. Moreover,whencomparinggraphsabove,theinfluenceofthesolarfieldareaisalsoevident. AsexpectedbythecoststructureshowninSection2,thecostsofthermalenergygeneration(directly dependentoninvestmentcosts)arereducedbyincreasingthesolarfieldarea.Thisreductionis especiallystrikinginthecaseofPTCtechnology. RegardingtheparticularcaseofSpainandconverselyofwhathappenedwhenanalyzingthe thermalenergygeneratedbythesolarinstallation,highercostsarelinkedtoVitoria,whilelower costsforSeville. Finally,Table24summarizesthethermalenergyunitcostsofconventionalenergysources analysedinthispaperconsideringthethreescenariosdescribedinSection3.Giventhewiderange ofexistingelectricityandnaturalgasratesdependingtheconsumerscharacteristics,thetwoextreme groupshasbeenincludedinthistable,I1andI6GroupsandIAandIGGroupsrespectively.Asin Figure 10. Thermal energy generation cost—Large size solar field area. The previous graphs show the PSU parameter, fluid temperature and solar field area importance have significant influence on the thermal energy generation cost. As can be observed for all bar groups, the thermal energy generation cost increases proportionally as the PSU decreases. If CPC technology results are analysed, the extreme situation is observed when the average fluid temperature is around 150 ◦ C. The influence of the average fluid temperature is lower in the case of LFC technology. These results show again the importance of the coupling in time between the thermal energy generated by the solar installation and the thermal energy demanded by the industrial process. In general, terms and according to the expressions proposed in Section 4, it can be stated that by reducing the PSU from 100% to 50%, double the thermal energy generation cost. Moreover, when comparing graphs above, the influence of the solar field area is also evident. As expected by the cost structure shown in Section 2, the costs of thermal energy generation (directly dependent on investment costs) are reduced by increasing the solar field area. This reduction is especiallystrikingin the case of PTC technology. Regarding the particular case of Spain and conversely of what happened when analyzing the thermal energy generated by the solar installation, higher costs are linked to Vitoria, while lower costs for Seville. Finally, Table 24 summarizes the thermal energy unit costs of conventional energy sources analysed in this paper considering the three scenarios described in Section 3. Given the wide range of existing electricity and natural gas rates depending the consumers characteristics, the two extreme groups has been included in this table, I1 and I6 Groups and IA and IG Groups respectively. As in the case of medium temperature solar concentration technologies installations the energy costs of this table have been calculated taking into account a time horizon of 20 years. Table 24. Thermal energy unit costs, conventional energy sources (c€/kWht). Conventional Energy Source Scenario Average Low High Natural gas I1 Group 6.1 5.0 8.4 I6 Group 3.6 2.9 4.8 Electricity IA Group 33.3 26.7 46.5 IG Group 6.4 5.4 8.5 Fuel oil 4.4 3.6 6.1 Diesel oil 8.1 6.9 10.6 Finally, Table 25 summarises, as example, the internal rate of return on investment considering the following hypotheses: Energies 2018,11, 2950 24 of 30 •Site: Seville. •Amortization period: 20 years. •The initial investment does not require financing. •PSU = 100%. •Average scenario for conventional energy sources. Table 25. Internal Rate of Return. Technology Solar Field Surface (m2) Average Fluid Temperature (◦C) Conventional Energy Source Natural Gas I1 Group Natural Gas I6 Group Electricity IA Group Electricity IG Group Fuel oil Diesel oil CPC 50 100 14 6 78 16 9 20 125 11 3 65 12 6 16 150 7 - 49 7 2 11 2000 100 23 13 >100 24 16 31 125 19 10 94 20 13 25 150 13 6 71 14 8 19 LFC 100 170 9 - 66 10 4 15 220 9 - 64 10 3 14 15,000 170 20 10 >100 21 13 28 220 19 9 >100 21 12 27 PTC 100 350 4 - 50 5 - 9 15,000 14 5 85 16 8 21 5.5. Analysis of Environmental Advantages To evaluate the environmental advantages, it is necessary to know the conventional energy sources conversion factors; Table 26 shows these parameters for the Spanish case [51]. Table 26. Conventional energy sources conversion factors. Conversion Factor Electricity FP e0.392 kgCO2/kWh Natural Gas FP ng 0.203 kgCO2/kWh Fuel oil FP f3.127 kgCO2/kg Diesel oil FP d2.868 kgCO2/L Lastly GHG emissions avoided by the substitution of conventional sources of energy are summarized in Tables 27–30. Table 27. GHG emissions annually avoided by the use of solar technologies instead of electricity. Site GHG Emissions Avoided [kgCO2/(m2·year)] CPC LFC PTC Thermal Level (◦C) Thermal Level (◦C) Thermal Level (◦C) 100 125 150 170 220 350 La Coruña 263 218 166 187 182 194 Vitoria 240 198 150 171 167 183 Barcelona 338 287 227 244 238 248 Valladolid 320 271 214 265 259 273 Salamanca 351 299 239 292 285 297 Teruel 362 309 247 312 306 311 Jaén 415 359 294 351 344 341 Valencia 342 291 232 244 238 246 Cáceres 367 314 253 305 299 311 Sevilla 429 373 307 358 350 360 Energies 2018,11, 2950 25 of 30 Table 28. GHG emissions annually avoided by the use of solar technologies instead of natural gas. Site GHG Emissions Avoided [kgCO2/(m2·year)] CPC LFC PTC Thermal Level (◦C) Thermal Level (◦C) Thermal Level (◦C) 100 125 150 170 220 350 La Coruña 136 113 86 97 94 101 Vitoria 124 103 78 89 86 95 Barcelona 175 148 117 126 123 128 Valladolid 166 140 111 137 134 141 Salamanca 182 155 124 151 148 154 Teruel 187 160 128 162 158 161 Jaén 215 186 152 182 178 176 Valencia 177 151 120 126 123 127 Cáceres 190 163 131 158 155 161 Sevilla 222 193 159 185 181 186 Table 29. GHG emissions annually avoided by the use of solar technologies instead of fuel oil. Site GHG Emissions Avoided [kgCO2/(m2·year)] CPC LFC PTC Thermal Level (◦C) Thermal Level (◦C) Thermal Level (◦C) 100 125 150 170 220 350 La Coruña 198 164 125 141 137 146 Vitoria 180 149 112 128 125 137 Barcelona 254 215 170 183 178 186 Valladolid 240 203 161 199 194 205 Salamanca 263 224 179 219 214 223 Teruel 271 232 186 234 229 233 Jaén 311 269 221 263 258 255 Valencia 257 218 174 183 178 185 Cáceres 275 236 190 229 224 233 Sevilla 322 280 230 268 263 270 Table 30. GHG emissions annually avoided by the use of solar technologies instead of diesel oil. Site GHG Emissions Avoided [kgCO2/(m2·year)] CPC LFC PTC Thermal Level (◦C) Thermal Level (◦C) Thermal Level (◦C) 100 125 150 170 220 350 La Coruña 195 162 123 139 135 144 Vitoria 178 147 111 127 124 135 Barcelona 251 212 168 181 176 184 Valladolid 237 201 159 196 192 202 Salamanca 260 221 177 216 212 220 Teruel 268 229 183 232 227 230 Jaén 307 266 218 260 255 253 Valencia 254 216 172 181 176 183 Cáceres 272 233 187 226 221 231 Sevilla 318 276 228 265 260 267 Kilograms of CO 2 reduction, by the use of medium temperature solar concentration technologies instead of electricity, stands out above other options. At the other extreme is natural gas, showing the lowest values. In the middle and showing very similar kilograms of CO2are fuel oil and diesel oil.