The growing concern for the diversification of the generating base and the constant search for energy sustainability has driven the search for new energy sources and the spread of unconventional renewable energy use [1], demanding the transition to a green economy with the implementation of specific public policies and regulations, also imposing on the productive sector innovative actions of sustainability [2]. In this regard, Brazil has assumed a prominent role in the world stage due to its predominantly hydro generating base and the implementation of incentive programs, such as the “Incentive Program for Alternative Sources of Electricity”, which aims to encourage investment in new sustainable technologies [3]. However, despite the political and economic efforts in Brazil with the allocation of R $ 226 million for investment in mitigation projects, adaptation to climate change in 2011; and federal renewable energy incentive laws, the diversification of the generating base in relation to the GDP has been late when compared to the Latin American level with countries like Costa Rica and Uruguay [2,4,5].

Currently the Brazilian energy matrix is conditioned to the national hydraulic supply [6], showing a strong dependence on the rainfall, which added the constant climate changes between the years 2012-2017. Resulted in periods of irregular rainfall causing a reduction in the volume of the reservoir and consequently, the significant increase in energy prices by the use of thermal generation as back-up system [7,8].

Among the possibilities of diversification of the Brazilian generating base, alternative sources have emerged, such as wind and photovoltaic energy. It is evident that the wind potential is the second highest in the country, with values ranging from around 60,000 MW (depending on the methodology employed) [9]; together, among unconventional renewables, the cost of wind is one of the lowest due to the return on investment [10].

Regarding the photovoltaic solar potential in Brazil, it has such high values of direct irradiation that reached 2400 kWh/m2.year in regions such as Bahia and Paraiba. In addition, it´s possible to generate more solar electricity than in the sunniest place in Germany which possess large investments in generation with this technology [6]. It is important to highlight that in summer months; the generation is maximum in the extreme south and southeast states of Brazil and coincides with the maximum demand registered by ONS for these regions. In this context, photovoltaic solar generation also has a great potential to contribute to the reduction of the demand peaks of the SIN transmission systems [6,9].

Given the above, and considering the possibilities it presents in the national territory for the massification of alternative sources of energy, photovoltaic solar is promising for the use of alternative energies.

New PV generation techniques include building-integrated photovoltaic generation technologies, also called Building Integrate Photovoltaics (BIPVs), which adapt to the original design of the architectural project. These techniques replace the structure's cladding materials, saving on the total cost of installation. In addition, they contribute aesthetically to the building and provide unique features that can assist in thermal comfort depending on the constituent material. [11].

From the point of view of energy efficiency, BIPVs are "ideal", as power generation and consumption have coincidence, thereby minimizing transmission losses that occur in traditional power systems [12].

The BIPVs have a wide range of methodologies for the installation of photovoltaic systems, among which stand out facades, windows, walls, roofs and shingles for being better known. It is this last classification that presents the real interest of BIPVs in this research, since solar tiles with integrated solar tile are a new product, still in improvement and with ample possibility of improvement.

In Brazil, the residential and commercial sectors are responsible for 23% and 11% respectively of the total national consumption of electricity, where ventilation and air conditioning equipment account for 50% of summer consumption, up to 70% in glazed buildings [12]. In this context, this article presents the analysis of the influence of operating temperature on the performance of a photovoltaic solar tile (TSF) for residential and land use of small incorporated into the construction (BIPVs) to the west of Paraná - Brazil.

Increasing photovoltaic cell temperature is a critical problem that impacts system performance, causing drastic yield reduction, material degradation and reduced lifetime [13]. As cell temperature increases, the mobility of electrons and holes in the material decreases, increasing the resistivity value of the photoconductor according to the mass action law. Moreover, while increasing the operating temperature, there is a drop in the open circuit voltage value, reducing the value of the power delivered to the load. [14].

This problem has been widely studied under various conditions for conventional modules, but the specificity of BIPVs has not been covered, especially TSFs. Given the above, this research aims to investigate the effects of operating temperature on the efficiency of cells incorporated into solar tiles, and verify the behavior and analyze other factors that affect the solar tile and directly to the structure.

For the PV evaluation purposes, the reference values at Standard Test Conditions (STC) were used for irradiation in 1000 W/m2 and temperatures of 25°C [15]. However, such conditions rarely occur during daily operation, so it is important to evaluate the solar tiles front of actual operating conditions, with daily variations in ambient temperature and quantifying the influence of factors such as wind speeds, thermal properties of materials, TSF geometry and their influence on system behavior.

This paper present of study and verify the influence of temperature on the performance of the Solar Tile incorporated into the building for different technical arrangements (Material - Photovoltaic Solar Arrangement), considering heat transfer conditions with the environment by conduction mechanisms and the convection and radiation from data literature.

The main contribution of this research in modelling is a specific analysis of the performance of the proposed technical arrangement on the influence of operating temperature on the efficiency of cells incorporated in the solar tile, which aims to fill the existing gap on BIPVs studies. It is noteworthy that the importance of this research is to determine the operating temperature of the PV array at STC. Therefore, it is essential to propose studies and analysis of this issue. However, the effects of this problem have not yet been studied for TSF, which have distinct characteristics from conventional photovoltaic systems, as they interact directly with the building structure materials, which need analysis to verify their influence on operating conditions and yield from TSF.

While the contribution of this research to the proposed technical arrangement, the commercial model of flat tile and single crystalline silicon cell has been modified, with an approximate 16% efficiency for conventional cells, to an overlapping format cell, as this configuration has not been studied or commercially developed. Based on this product, it is intended to verify the characteristics while differentiating the influence of thermal effects for different conditions.

Concerning the constituent materials, the research contribution incorporates materials that have not been used for the construction of BIPVs on tiles, reason why the innovation of materials is planted, using concrete, Polypropylene and Polyvinyl Chloride (PVC). These materials were considered to characterize and contribute in the cell as a function of the influence of heat transfer as well as structure as a function of material weight, mechanical strength and thermal comfort.

The Section 2 showed the photovoltaic technologies and arrangements, describing the field of action, the materials that make up the panels and their respective efficiencies. BIPVs will be discussed in more detail, characterizing their uses and aspects for implementation and a brief comparison between the different factors of PV and BIPVs. In addition, the state of the art of BIPVs technologies, current regulations and regulations will be addressed. The Section 3 presents the dimensioning and modeling of the chosen models of photovoltaic solar tile (low cost), the geometry of the studied solar tile, detailing the materials and constituents of the evaluated photovoltaic arrangement. The Section 4 shows the correlation of the results and the comparison of the proposed models with the materials, the cell temperatures and the determined efficiencies. Finally, in Section 5 there are the conclusions of this article on the design of the solar tile and the use of Building Integrated Photovoltaics (BIPVs) technologies.

The photovoltaic technologies can be classified in several ways, one of them is by grouping the range of existing technologies into current and possible future options, classifying them into 1_st, 2_nd and 3_rd generation [16].

The current production of PV technology, corresponding to the 1st generation, is dominated by single-junction cells based on monocrystalline silicon wafers (m-Si) and polycrystalline silicon (p-Si), with laboratory efficiencies ranging from 18% to 21% but high cost [17]. However, the m-Si commercial model is manufactured with efficiencies between 13-14% for a lower market value [18]. Half the cost of 1st generation PV is 200-250 microns thick silicon junction, a superlative cost, as most solar absorption occurs in the first 10 microns. Therefore, reducing “wafer” thickness offers potential cost savings [16,17]. Despite much scientific progress, 1st Generation PV costs an average of US $4/W, which is still about 4 times more expensive for truly competitive commercial production [18].

The 2nd generation of the PV is a reduction of US $/ W, reducing the material, by using thin film devices, while maintaining the efficiencies of the 1st generation [17]. Use Amorphous Silicon (a-Si), a thin film Copper-Indium-Selenium (CIS) association, Cadmium Telluride (CdTe) or Polycrystalline Silicon (p-Si) deposited on low cost substrates such as glass [16]. Cost savings have been made possible by the fact that CdTe, CIS and a-Si technologies absorb the solar spectrum much more efficiently than m-Si or p-Si, using between 1-10μm active material [17].

Recent laboratory scale research shows the potential of thin film, with efficiencies of 16.5% CdTe and 18.4% CIGS [19], however, CdTe and CIGS PVs have been slow to scale commercially. This is partly due to the gap between laboratory efficiencies and commercial production module efficiencies of 10.7% for CdTe [19] and 13.4% for CIGS [20] which are low as a result of unresolved issues related to material reproducibility and uniformity over large areas [17]. Given the precursor progress of thin film silicon over the last few years, the trend indicates that the potential of 2nd generation is more likely to be realized by silicon based thin film devices, enhanced by the development of production tools for the flat tiles industry [17].

3rd generation deals with devices that may or may not exceed single-junction limits and feature ultra-high efficiency for the same production costs as 1st and 2nd generation, improving m-Si and p-Si efficiency and reducing cost from US$/W [16]. This technology uses thin films applied to low cost substrates to maintain material economy and increase efficiency.

In addition to using thin and triple junction films, this technology includes organic materials (OPV) and quantum dot products (PQs), with both technologies still under development [19]. Although 3rd generation PV still lacks conversion efficiency, they exhibit great potential and several advantages over established technologies, including low cost processing over large areas, study of semi-transparency effects, mechanical flexibility and low weight [19]. Applications of these devices include low power equipment (for residential use) and building integration known as Building Integrated Photovoltaics. The comparison of the three PV generations in terms of efficiency, cost of US$/W and US$/m2 is shown in Fig. 1.

Figure 1 Source: Conibeer, 2007.

As shown in Fig. 1, 1st and 2nd PV technologies do not reach 30% efficiency for the unit bandgap limit, while 3rd generation PV technologies are close to reaching the thermodynamic limit, with efficiencies close to 70%. Costs of US$/m2 from 1st to 2nd generation decreased by 350 US$/m2, so the transition from 2nd to 3rd generation had a slight increase from 150 US $/m² to 200 US$/m², which can be explained by the huge increase in efficiency from one technology to another. Regarding US$/W costs, 3rd generation has the most cost-effective technologies ranging from US$0.15/W to US$0.30/W, considerably reducing US$3.5/W to US$1.00/W 1st and 2nd generation PV. Therefore, the current tendency is to improve the costs of US$m2, US$1.00/W and the efficiency of the 3rd generation.

This section presents the design of the solar tile showed the hypotheses adopted for conduction heat transfer and the convection and flow data from the literature. The base geometry of the chosen tile is of flat shape with incorporation of superficial PV arrangement.

This format is one of the most common in the market and is easy to replace if material fails or ruptures. It is also easier to build than curved roof tiles, has good roofing and is easily adapted to photovoltaic cells [60-66]. From the standard rectangular flat tile format, two models have been proposed which are: Model 1 - Tile with Overlaid Photovoltaic Arrangement and Model 2 (tile with integrated photovoltaic arrangement). The integrated and overlapping models are shown in Fig. 3 (a) and Fig. 3 (b) respectively.

Figure 3 Source: The authors.

Model 1 is intended to be a different product to the market with a spare photovoltaic tile geometry. This model aims to determine if the overlap of the PV array over the tile has distinctive features to the commercial model (integrated system), because the effects of heat transfer between the tile and the PV arrangement happen differently with less contact area.

On the other hand, in order to determine the characteristics of the standard and commercial model for the proposed TSF materials in Model 2, the insertion of the PV arrangement in the tile is evaluated. By having a larger contact area between the surfaces, the heat transfer effects happen differently, so they reach different temperatures to the overlapping model, which consequently affect the PV arrangement. When modelling and implementing both models in the 3D structural simulation environment in Comsol Multiphysics® software one has that the overlay or insertion of the photovoltaic array in blue color, as shown in Fig. 3.a) for the integrated model and Fig. 3.b) for the overlay model.

For the photovoltaic cell modelling, the same materials and dimensions were kept in both models. It should be noted that the main difference between the two models is in how the temperature will influence the tile and the turn in the photovoltaic cell, as the surface of the integrated model experiences conduction heat transfer influenced by all PV array components and the tile. Whereas in the superimposed model, the heat exchange by conduction takes place only by the last layer of the PV (Elvax) arrangement with the tile, until it reaches the photovoltaic cell arrangement.

By analyzing Fig. 4 it can be seen that the photovoltaic cell was approximated by the geometry of five coupled plates, with 220 mm of square surface and 2.5 mm of thickness each, totaling 12.5 mm system thickness. The PV array was designed according to the typical components of a photovoltaic system composed of a layer of glass, Ethylene Vinyl Acetate (EVA) and a layer of Vinyl Polifloururo (Tedlar). The thermal properties used in the simulation of the PV array (N^o. 1 to N^o. 5) and tile (N^o. 6 to N^o. 8) array components are listed in Table 2.

Figure 4 Source: Adapted from Leow et al., 2016.

Table 2

Source: Adapted from Leow et al., 2016; BRASKEM, 2012 and IMCPLAS, 2018.

A number of literature and studies were revised, such as: Radziemska on the effect of temperature on crystalline silicon cells [13]; Barring the influence of temperature on photovoltaic solar cells [14]; Omer et al on poor PV cell performance in monitoring a BIPV system in the UK [60]; Agrawal & Tiwari in Energy Analysis and Exergy in a BIPV System [61], Portolan & Rüther characterizing the potential of BIPV and Building Applied Photovoltaics (BAPV) in a Brazilian home, demonstrating that temperature affects cell power [65], Essah et al in Performance Evaluation of a BP m-Si Integrated PV System [64]. Those authors demonstrated that temperature has a strong influence on module/cell efficiency and consequently on system power.

In order to analyses the technical arrangement operating under average conditions, it was decided to perform the tests for the average minimum and maximum summer temperatures of Foz do Iguaçu from 21/12/2017 to 20/03/2018. According to data from the National Institute of Meteorology (INMET), the average maximum and minimum temperatures measured for the summer months was 30,6 ºC and 22,1 ºC which corresponds to the hours of 17h and 6h respectively [71].

It is important to highlight that the territory under analysis presents extreme temperatures of maximum and minimum divergent to the average. In the sample universe, the average values of maximum and minimum temperatures represent 71% of the annual samples [71], the remaining 29% corresponds to extreme sample values that occur in isolated cases of maximum or minimum temperatures.

In this section, the results of the temperature distribution on the tiles and the PV arrangement will be presented and discussed. Subsequently, the temperature of the PV cells in the tile is determined to define the efficiency in simulated models using Eq. 2.

Conduction heat-transfer study conditions applied in Comsol Multiphysics® and literature data for convection and radiation for Superimposed (CS) and Integrated (CON) Concrete, Superimposed PV (PS) and Integrated Array Polypropylene (PI), and the overlap model (PVCS) and integrated PVC (PVCI) at the mean maximum temperatures, 30.6 ºC and the average minimum temperature of 22.1ºC [71], Fig. 7 and Fig. 8 show the result of the tile temperature distribution and the photovoltaic arrangement for the superimposed and integrated configurations for steady state simulation.

Figure 7 Source: The authors.

Figure 8 Source: The authors.

Comparing the results of the overlapping TSF models, it was shown that the upper part of the PV arrangement is the component with the highest radiation exposure corresponding to glass, which has a similar color in the three materials, varying the temperature between 95 °C and 82 °C. On the other hand, comparing the color achieved on the surface exposed to 1000W/m² flow and convection, the model/material that experienced the highest temperature increase is the CS; oppositely, PS had the lowest temperatures on the exposed surface, ranging between 72.23°C and 80.67 °C.

Comparing the isolated surfaces, PS reaches the lowest temperature in the lower (inner) part of the TSF with temperatures ranging from 72.23 °C to 80.67 °C for studies at 22.1 °C and 30.6 °C respectively.

In the study of the integrated TSF model, it is noticeable the difference of the IC compared to the PI and PVCI, because the concrete tile is strongly affected by temperature, considering its high conductivity (1.80 W/m°C). On the tile surface exposed to 1000 W/m2 flow and convection, PI and PVCI act as a thermal insulator, reaching maximum temperatures of 75ºC (test at 30.6ºC) in almost all of the tile geometry.

After the thermal study of the models under conduction heat transfer conditions, the convection and radiation data from the literature, together with the data provided by Eq. 4 and the simulation conditions specified for each mechanism, the temperature of the photovoltaic cell was determined by locating the center point of the structure that represents only the "Solar Cell" layer as shown in Fig. 4 and Table 2 by "N°. 3". This would determine the average temperature at which the cell was subjected under simulated conditions.

With the Tcel cell temperature, the cell efficiency for this temperature is determined by means of Eq. 2. The results for each model in the maximum and minimum summer temperature conditions of Foz do Iguaçu and their respective power reduction as a function of efficiency are shown in Table 3.

Table 3

Source: The authors.

The Table 3 shows the temperatures found in the cell (Tcel) for the maximum and minimum temperature conditions of the external environment (Text), the η cel component is determined using Eq. 2 and substituting Tcel for each study, the difference in efficiencies represented by η fab - η cel determined by the subtraction of the efficiency indicated by the manufacturer (η fab = 16%) and the efficiency determined in the study (η cel), which represents the percentage of efficiency decrease under simulated conditions.

The correlating the results and comparing the proposed models with the materials, the cell temperatures and the determined efficiencies. Subsequently, the results will be approached according to the civil-structural characteristics, characterizing the thermal properties of the tile materials and their influence on the roof structure. Finally, the results obtained will be compared with some conventional photovoltaic systems with different efficiencies, but maintaining the same simulation conditions.

In order to quantify the efficiency performance in simulated models for temperatures between 22.1°C and 30.6°C, Table 4 and Fig. 9 show the temperature interpolations, verifying the change efficiency by increasing the cell temperature by 1°C.

Table 4

Source: The authors.

Figure 9 Source: The authors.

By analyzing the data contained in Table 4 and Fig. 9, it can be shown that the decrease in efficiency occurs linearly by approximately 0.05%/°C for Concrete, Polypropylene and PVC, both in the integrated and overlapping model. Comparing the six test, the IC is the model with the highest performance, with efficiency of 13.42% measured at 22.1°C. On the other hand, when buying the yield on overlapping models, the CS showed the best performance, with an efficiency of 12.93% at 25°C (STC).

When evaluating the materials according to the integrated and overlapping models, it was found that the integrated models showed greater efficiency for the same test conditions; which has a direct relation with the convective effect present in the overlapping models, negatively influencing the cell performance and consequently in the system. The models most affected by the increase in cell temperature in the test condition at 30.6 °C were PS and PVCS, which can be explained by the action of the convective effect of air, with consequently increases the temperature in the cell.

Due to the very characteristic of BIPV and specifically the solar tiles evaluated in the research, it is interesting to analyses the materials comparatively from the standpoint of efficiency individually. It also characterizes the civil-structural advantages that these materials can bring to the structure, since the use of a certain material in the roof structure brings individual and specific characteristics that will be decisive when specifying the installation location. Some of the features are thermal comfort, hardness of the roof, less weight to the structure, environmentally friendly, etc.

By implementing mesh refinement for four gradient points on each plate, it is possible to see the temperature gradient in the PV arrangement represented by the color change in the results of Fig. 7 and Fig. 8, which varies up to 10°C when compared in simulated studies at 22.1°C and 30.6°C, reflecting the importance of a correct simulation.

The Building Integrated Photovoltaics (BIPVs) technologies have a promising outlook in the near future. As the cost of manufacturing is expected to decrease, it will increase competitiveness and consequently add new TSF models and manufacturers, thus making this technology extremely competitive with conventional PV systems applied to building systems.

However, in order to properly assert its use, market insertion and competitiveness with conventional PV systems, there is a need, as evidenced in the research, to take into account several factors such as: (i) the materials as a function of the thermal properties that they can provide both in PV arrangement influencing the temperature of cells, cooling the PV system, providing characteristics of thermal comfort, among others; (ii) civil and structural advantages such as lower weight, greater roof resistance to impacts and extreme climates; (iii) characteristics of the installation site, factor directly related to temperature, wind speed, relative humidity and irradiation; (iv) lowering the cost of integrated building technology by combining factors/advantages beyond the power supply function.

Hopefully, in the coming decades, conventional and unconventional photovoltaic technologies can make a substantial contribution to global and centralized energy production, although currently its cost is up to five times more expensive than conventional power grid due to low production costs from non-renewable sources.

According to the analysis of the simulations performed in the integrated and superimposed PV arrangement models, for Concrete, Polypropylene and PVC materials, by means of simulated conduction heat transfer in the Comsol Multiphysics^® software, data from convection and radiation literature, shown in Fig. 7 and Fig. 8, to determine the influence of cell operating temperature by means of the efficiency calculation using Eq. 2, and the results shown in Table 3 and Table 4, it was proved that the integrated and overlapping models for the Concrete, Polypropylene and PVC have clearly differentiated cell performances, with efficiency variations between 12.53% and 13.42% for the analyzed models.

Thermally, the simulated PVC and Polypropylene models, comparing the superimposed and integrated PV array configurations, experienced similar results in the temperature reached in the tiles, with a difference of approximately 3°C more for polypropylene, which is reasonable for the higher material thermal conductivity value. On the other hand, the concrete presented different values to PVC and Polypropylene, reaching the highest temperatures in the tile geometry.

Considering that the use of TSF using as concrete base material Concrete, Polypropylene and PVC with the thermal characteristics mentioned in Table 2 has different behaviors, it is essential to verify the contribution of the civil and structural characteristics that support these products.

As found in the study, the variation in efficiency for all models and materials studied presented in, happens in 0,05%/°C. Although this value is considered as minimal and negligible, it is essential to verify how economically this data represents in the TSF, because in terms of comparing annual consumption and savings against the efficiency of a given model and material with a conventional PV system, it could mean real savings in consumption, and consider the civil and structural advantages that the materials bring.

Therefore, according to the analysis of the results obtained in this research it can be stated that: For the Concrete, Polypropylene and PVC studied, both for the integrated and superimposed PV arrangement model for the TSF, when simulating the product in STC, the efficiency decrease with the temperature increase happens in a value of 0,05%/°C. The diminishing effect of 0,05%/°C happens for the thermal properties of the tile materials with varying thermal conductivity of the tile materials. 0,1W/m°C to 1,80W/m°C, specific heat the constant pressure of 850J/kg°C to 1700J/kg°C and specific mass between 910 kg/m³ to 2300 kg/m³. Therefore, for TSF studies under STC conditions for the characteristics of Foz do Iguaçu and varying the thermal conditions of thermal conductivity, specific heat at constant pressure and specific mass of any tile material, within the range of the specified values and maintaining the PV array characteristics resulted in efficiency behaviors between the 12,53% to 13,42%.

Bearing in mind that the effect of decreasing efficiency on 0,05%/°C Both in the integrated and overlapping models for Concrete, Polypropylene and PVC happens differently. These proves that it is relevant to check the input of materials from civil and structural perspective that they can bring to the roof, as the same simulated materials -but with different models- show different TSF thermal behaviors in relation to the temperature reached by the tile. In the simulated CS and IC the 30,6°C, the minimum temperature reached by the tile is 81,36°C and 71,80°C respectively, which evidences that the CS model brought a higher temperature to the internal environment of the house, directly influenced on the thermal comfort.

The same effect happens with the study of simulated PS and IP at 30.6°C, where the minimum temperature reached is 80,67°C and 70,90°C respectively; therefore, the PI will point out the best thermal comfort features that the PS. As in previous cases, PVCI has better thermal comfort characteristics when compacted with PVCS as it has temperature differences of up to 10°C when analyzing the same point of geometry.

Finally, as a continuation of this research, it is suggested to broaden the scope of the simulation including new materials, different geometric shapes of tiles, irradiation conditions and economic factors that influence the commercialization and replacement of the conventional photovoltaic technologist for the BIPV´s range.

In addition, has the development of hybrid generation systems for harnessing thermal energy based in energy harvesting [95-99].