Numerical performance investigation of High Vacuum Flat Plate Hybrid Photovoltaic-Thermal devices.

We propose an innovative flat plate hybrid Photovoltaic-Thermal system under high vacuum (HV PV-T) optimized for solar-to-thermal energy conversion. It consists of a glass cover, metallic vessel, and the actual PV-T device, which englobes a low-emissive Transparent Conductive Oxide (TCO), a perovskite-based PV cell, a Solar Absorber, and a copper substrate. We investigate, through a 1-D model developed in MATLAB, the performances of the proposed PV-T system, still mined by radiative losses, varying the operating temperature (Top) and the emittance of the TCO (εTCO ) in the ranges of (25÷175) °C and (0.05÷0.45) respectively. The annual thermal and electrical productions are evaluated considering the Typical Meteorological Year of Naples, Italy. Specific annual costs and emission savings are evaluated and compared with the ones assured by commercial High Vacuum Flat Plate Solar-Thermal (HVFP ST) and PV collectors. Results indicate that the proposed HV PV-T increases the annual cost savings by 34% and 11% when compared to HVFP ST and PV collectors, respectively. Moreover, the presented HV PV-T increases the annual CO2 emissions savings by 7% and 48% when compared to HVFP ST and PV collectors, respectively.


Introduction
Renewable energies have a key role in the fight against climate change.Since the Earth's surface receives from the Sun enough power to satisfy the demands of the whole world [1,2], solar energy could provide the most relevant contribution, but its true potential is still unexplored.Currently, solar energy is predominantly converted into electrical or thermal energy thanks to PV and Solar-Thermal (ST) collectors: different devices with different conversion efficiencies and different performance decay with increasing operating temperature.ST collectors can reach efficiency values up to 80%, while, on the other hand, the highest PV electrical efficiency value recorded is 29.1% for singlejunction cells made of GaAs [3,4].Hybrid Photovoltaic-Thermal (PV-T) devices combine a solar cell and a thermal collector, exploiting more solar radiation than a typical PV collector due to the recovery of the fraction of incident power not converted into electricity.Several PV-T architectures and layouts have been studied since the 1970s [5,6,7].Concentrating PV-T devices reach the highest operating temperatures, but since they require reflectors, lenses, and a sunlight-tracking mechanism, the flatplate layout is preferred due to its affordability.Classical PV-T applications aim to increase the electrical efficiency of PV panels preventing their heating [8]: these systems are optimized for electrical production, while the thermal output is a secondary product, which can meet loads at temperatures equal to or lower than 50 °C.However, the Heating and Cooling sector accounts for 51% of Total Final Energy Consumption, against the 17% related to the Power sector [9].Moreover, thermal loads require frequently supply temperatures higher than 50°C.
Mellor et al. [10] and Hu et al. [11] proposed Silicon-based PV-T encapsulated in vacuum, demonstrating the thermal efficiency enhancement due to the abatement of convective losses, but the lamination of Silicon Cells on copper doesn't extend the spectral exploitation of the device: copper becomes highly reflective long before the wavelength corresponding to the bandgap of Silicon (1.1 eV = 1029 nm).The effectiveness of high vacuum encapsulation in flat plate architectures has been validated for both Solar-Thermal (ST) [12,13,14] and PV-T devices [10,11,15].For the PV-Ts, the abatement of convective losses can double the thermal efficiency against the non-evacuated architecture [10].Materials with energy bandgap higher than that of Si, like GaAs, exhibit a weaker dependence on the temperature and let a higher portion of the spectrum shine through the PV cell.GaAs-based PV modules are already adopted in space applications, which are similar to terrestrial applications under high-vacuum due to the high operating temperatures and absence of convective cooling.GaAs-based PV cell electrical efficiency experimental values, in the operating temperature range (25÷175) °C [16], will support our modelling.Unfortunately, GaAs PV modules are far more expensive than the ones adopted for terrestrial applications, so they are not suited for industrial-scale applications.However, cheaper materials like Perovskites are emerging in the PV sector.Perovskite PV cells (PSC) reached 20% electrical efficiency even in the inverted configuration [17] and Romano et al. and Hoang et al. [18,19] contemplate the tandem PSC/CIGS cells eligibility in space applications.Considering their positive trend in the last years [20], is safe to assume that PSCs will soon reach the same thermal stability as GaAs-based PV cells.The adoption of low-emitting TCO contributes to the minimization of the predominant loss mechanism which still mines the thermal performance of our device, i.e., radiative losses.Thermal coupling of the PV cell and SA is another means at our disposal to expand solar spectrum exploitation.Thanks to their spectral absorptivity higher than 0.95 in the wavelength range (350÷1500) nm, SAs like the ones proposed by D'Alessandro et al. [14] can enhance the thermal absorption above the energy bandgap of the PV cell, allowing the collection of almost all the radiation transpired through the semiconductor.In this sense, the PV cell acts as a spectral splitter.De Luca et.al [15] performed the simulations of two spectral splitting layouts: thermally decoupled and thermally coupled.In the thermally decoupled layout, the PV cell is placed below the glass cover so that it does not experience an efficiency decay due to high operating temperatures.In the thermally coupled layout, the PV cell is placed directly above the SA so that almost all the radiation that has been absorbed by the PV cell but not converted into electricity (thermalization and extraction losses) is conducted to the SA, providing an additional availability for the solar-to-thermal conversion.They demonstrated that the thermally coupled layout outperforms the decoupled one in terms of thermal and total efficiency.
In this work, we propose a novel HV PV-T layout optimized for solar-to-thermal-conversion which ensures a non-negligible amount of electrical output.This layout is based on the following main concepts: a) High-Vacuum encapsulation (p<0.1 Pa); b) adoption of high bandgap PV cells; c) adoption of low-emitting TCOs; d) thermal coupling of the PV cell on a Solar Absorber (SA) (figure 1).The performances of the proposed device are evaluated through a 1-D model developed in Matlab.Section 2 describes the thermodynamic model and the performance parameters investigated through a parametrical analysis, which is carried out by varying those parameters that govern the radiative losses, i.e., operating temperature (  ) and TCO emittance (  ).The performance parameters are evaluated considering both standard (subsection 2.1) and hourly (subsection 2.2) global irradiation in the city of Naples, Italy.An economic and environmental impact analysis, contextualized in the Neapolitan climate, is also carried out comparing PV, HVFP ST, and HV PV-T devices as proposed systems (subsection 2.3).In Results and discussion, we expose the results of the analysis, while in Conclusions the main achievements are highlighted.

Methods
This section describes the thermodynamic model implemented in Matlab to evaluate the standard and annual energetic performances of PV, HVFP ST, and HV PV-T collectors.Moreover, the parameters and the equations adopted to translate the energetic performances in terms of specific annual costs and emission savings are exposed.

Standard Performance
Standard Performance indicates the performance that the three types of collector exhibit when exposed to the standard AM1.5GSpectrum (I = 1000   2 ).For the PV collector, we assume the electrical efficiency is independent of the incoming radiation and equal to the value obtained in the Standard Test Condition (STC) [21].Instead, as regards the HVFP ST and HV PV-T collectors, the performances have been simulated through a 1-D model implemented in MATLAB.Equations (1-3) describe the model concerning the HV PV-T, which is based on the power balances related to the glass cover (equation (1)), the PV cell (equation (2)), and the Solar Absorber (equation (3)): where is the equivalent emittance of the two parallel plane surfaces, i.e., glass cover and TCO.The same goes for . All the parameters are described in detail in Table (1).  and   are the input variable of the model and the numerical analysis is performed by varying them in the ranges of (25÷175) °C and (0.05÷0.45) respectively.For each combination of   and   , our model evaluates the performance parameters of the HV PV-T device in terms of thermal, electrical, and total (thermal + electrical) output, i.e.,   Results show that the temperatures of the PV cell and SA differ by less than 1/100 °C, therefore, from now on,   =   =   =   (as was easily expected due to the low thickness of the layers composing the absorbent surface).
The model implemented for the HVFP ST collector is like the set of equations 1-3, except for the absence of the PV cell and the electrical output.In this case, the absorbent surface is a Selective Solar Absorber (SSA) deposited on an aluminium substrate and its thermal properties have been experimentally evaluated in [13].

Annual Performance
Annual performances have been evaluated considering the hourly variation of the incoming Global Irradiation in the city of Naples [22].Hourly thermal efficiencies are calculated according to equation ( 5) and equation ( 6) for HV PV-T and HVFP ST: where   is the mean absorption coefficient of the whole PV-T absorbent surface (PV + SSA).
Radiative losses towards the glass and the vessel are governed solely by the   and the emittance of the surface, therefore they have a different weight on the thermal efficiency depending on the hourly global irradiation I(h).On the contrary, the electrical efficiencies of both HV PV-T and PV collectors can be considered independent of the hourly irradiation because the electrical output is directly proportional to I(h).Hourly thermal and electrical output in   2 are evaluated by simply multiplying the hourly efficiencies by the I(h).Their values eventually allow the evaluation of annual thermal and electrical output in terms of ℎ  2  , i.e.,   ,   , and   , and annual efficiency (̅ ).

Specific Annual Costs and Emissions Saving evaluation.
In the industrial sector, the reference solution to satisfy the annual thermal demand is burning Natural Gas with furnaces in loco.Instead, the electrical demand is satisfied by linking to the national electric ) and Specific Annual Emissions Saving (SAES, ) are predicted starting from the annual energetic performances of the named solar devices.A system based on PV collectors produces    ℎ  2  , avoiding, an annual cost of and an annual emission of both related to the link with the national electric grid, which is characterized by the specific cost of electricity   = 0.292 € ℎ and the emission factor of the national thermos-electric park   = 0.483   2 ℎ .For both proposed systems based on HVFP ST and HV PV-T collectors, we assume: • The integration of a thermal storage system large enough to fully decouple thermal production and demand, which are supposed to be at   = 100 °; • A heat Removal Factor (HRF) of 0.82, was evaluated from a commercial HVFP ST collector datasheet [23].A system based on HVFP ST collectors produces effectively    •  ℎ  2  , obtaining: where   is the combustion efficiency of industrial furnaces (assumed equal to 0.8), LHW= 9.59 is the Lower Heating Value of natural gas, and   =0.7 €  3 is its specific cost.Considering a natural gas emission factor   = 0.202   2 ℎ (referred to as primary energy), we obtain: Similarly, a system based on HV PV-T collectors provides the following annual specific savings:

RESULTS AND DISCUSSION
The colourmap in figure 2(a) shows, for the proposed HV PV-T device, the standard thermal output (subsection 2.1) dependence on   and   .The higher these parameters, the higher the thermal losses, and the lower the thermal output.The vertical black dotted line in figure 2 Increasing the operating temperature, the thermal production of the HV PV-T decreases faster than that of the HVFP ST due to the electrical power extraction and an   three times greater than the emittance of the HVFP ST collector.With   > 25 °, the HV PV-T device has a lower total output than that of an HVFP ST collector, but always higher than that of a PV collector.This trend is also observed on an annual basis, as shown in figure 3(a), where the annual energetic producibilities are compared.In Naples,   = 100 °, which is a common temperature for several industrial processes, leads to interesting results since the HV PV-T collector provides a thermal output of 941 ℎ  2  : this amount corresponds to 68% of the annual production of the commercial HVFP ST at the same temperature (green and orange columns in figure 3

Conclusions
In this paper, authors proposed a novel layout of an HV PV-T device optimized for solar-to-thermal energy conversion based on the following concepts: high-vacuum encapsulation, adoption of perovskite PV cells, low-emitting TCO coating, and thermal coupling of the PV cell on a Solar Absorber.Performances have been simulated under standard global irradiation varying the TCO emittance (0.05÷0.45) and the operating temperature (25-175) °C.Referring to the Typical Meteorological year of Naples, fixing the   at 0.21 and   at 100°C, the specific annual cost and emissions saving parameters have been calculated for all three collectors.HV PV-T collector increases the specific annual cost saving by 34% and 11% when compared to HVFP ST and PV collectors, respectively.Moreover, it increases the annual CO2 emissions saving by 7% and 48% when compared to HVFP ST and PV collectors, respectively.Future research on how to reduce TCO emittance would lead to further improvements in the performance of our device.In conclusion, the results suggest that HV PV-T collectors could play a key role in the renewable energy market.

Figure 1 .
Figure 1.(a) Schematics of the proposed HV PV-T; (b) Exploitation of the solar spectrum enabled by thermal coupling of high bandgap PV cell and SA.
(a) represents the case where the PV-T absorbent surface is coated with the low-emissive TCO proposed by Alonso-Alvarez et al. (  = 0.21) [24].This case is highlighted in figure 2(b), where the standard energetic performances of the proposed HV PV-T and commercial PV and HVFP ST collectors are compared.

Figure 2 .
Figure 2. Standard performances under I=1000W/m 2 .(a) Colourmap of the HV PV-T standard thermal output in dependence of   and   .The vertical black dotted line (  ) represents the case where the PV-T absorbent surface is coated with the low-emissive TCO proposed by Alonso-Alvarez et al. (  = 0.21).(b) Comparison among standard energetic performances of HV PV-T, HVFP ST, and PV collectors.The PV collector is supposed to work in STC.
Figure 3. (a) Comparison among annual energetic performances of HV PV-T with   = 0.21, HVFP ST, and PV collectors.The PV collector is supposed to work in STC (b) SACS and SAES parameters evaluated for the three solar field types.

Table 1 .
̇ ,  ̇ , and  ̇ in Geometric, thermal, and optical parameters of the proposed PV-T device [16]uated by normalizing the output with respect to global irradiation.The electrical efficiency, as well as the electrical output, is evaluated according to the experimental data provided by Maros et al.[16]: ̇ − (  ) =   (  )   ( −     (  ) − In this section, systems based on PV, HVFP ST, and HV PV-T solar collectors are proposed to partially satisfy thermal and electrical demands and contain running costs and emissions.Thermal and electrical demands are contextualized in the climate of Naples.Specific Annual Costs Saving (SACS, Lauterbach C, Schmitt B, Jordan U, Vajen K 2012 The potential of solar heat for industrial processes in Germany Renewable and Sustainable Energy Reviews 16 5121-5130 [2] Shahsavari A, Akbari M 2018 Potential of solar energy in developing countries for reducing energy-related emissions Renewable and Sustainable Energy Reviews, 90:275-291 [3] Green M, Dunlop E, Hohl-Ebinger J, Yoshita M, Kopidakis N, Hao X Solar cell efficiency tables (version 57) Progress in Photovoltaics: Research and Applications 29 (1) 3-15