Contributing to the Concept of Sustainable Buildings: Evaluation of the Carbon Emissions of a Solar Photovoltaic Coating Developed in Northeast Brazil

Abstract

Solar coatings have become increasingly relevant as a means to enhance the performance and efficiency of photovoltaic (PV) panels, playing a critical role in advancing sustainable solar energy solutions. This study employs the life cycle assessment (LCA) methodology to quantify the greenhouse gas (GHG) emissions associated with the production process of a coating used on solar PV panels. Actual data were collected for the manufacture of the solar coating, constituted by two layers: (i) tetraethyl orthosilicate (TEOS) and (ii) titanium isopropoxide (TTIP). Data on energy and material flows were compiled. The GHG emissions for the TEOS and TTIP coatings were 1.8977 and 6.3204 g CO₂-eq/mL, respectively. With experimental data demonstrating a 4.5% increase in panel efficiency from the coatings, a simulation was carried out to verify the impact of the solar coating on a 16.4 MW solar power plant. The results indicate lifetime avoided emissions of 98,029,294 kg CO₂-eq over 25 years. Sensitivity assessments verified the impact of shorter lifetimes of the coatings, and even with frequent reapplication—down to monthly intervals—the coating continues to provide net environmental benefits. This robustness reinforces the potential of solar coatings as a complementary strategy for decarbonizing PV systems.

1. Introduction

Installing solar photovoltaic (PV) panels can promote sustainable building design by reducing the building’s operational carbon footprint. By producing low-carbon electricity, solar PV systems can contribute to meeting green building certification standards, enhancing energy efficiency, and lowering long-term energy costs. Additionally, solar PV panels support broader sustainability goals by promoting energy resilience and reducing the environmental impact associated with conventional grid-supplied electricity.

Electricity production by solar PV systems depends on solar irradiance but can be negatively affected by temperature [1,2], as most PV modules perform best at lower temperatures. Improvement of manufacturing processes can enhance the performance of solar absorbers, such as silicon dioxide (SiO₂) deposition on molybdenum (Mo) films to increase absorbance and reduce emittance [3], and Mo and Mo/SiO₂ film deposition on electropolished surfaces to increase absorptance [4].

Shadowing can significantly reduce PV power output [5], along with heavy rain, snow, and strong winds, which can damage panels over time. The accumulation of sand and dust, which can drastically reduce energy production [6,7], is also an issue. The influence of dirt has been reported to decrease the output power of solar PV panels by up to 40% [8].

Silva et al. [9] argued that regular cleaning ensures optimal system performance and can mitigate energy losses, highlighting the effectiveness of cleaning in restoring the parameters of PV systems exposed to environmental conditions. As a result, cleaning practices are frequently employed on solar PV systems. The study of Schultz and Carvalho [10] considered water-based washing of PV panels and mowing of grass and weeds to decrease output losses. Sallamah et al. [11] presented a detailed review on the effects of dust and cleaning methods on the performance of solar PV systems and reported that self-cleaning coatings could dramatically increase the efficiency of solar energy devices. Abdallah et al. [12] mentioned that the waterless cleaning method presents lower costs, but water-based robot cleaning is suitable for large-scale solar plants. However, sometimes the friction associated with cleaning can damage PV panels [13].

Several recent studies have been conducted on the development of coatings for PV panels and the design of mechanical cleaning equipment, aiming to reduce cleaning maintenance without affecting PV efficiency [11]. The main categories of coatings are self-cleaning, anti-reflective, and non-stick coatings. Often hydrophobic, these coatings help repel dust and dirt, improving efficiency and reducing cleaning needs [14]. A scheme of a coated PV panel is shown in Figure 1.

Sarkin et al. [16] reported that SiO₂, MgF₂, TiO₂, Si₃N₄, and ZrO₂ are widely used in anti-reflective coatings. Regarding self-cleaning applications, superhydrophobic coatings were highlighted, with the most suitable materials being Al₂O₃, TiO₂, and Si₃N₄ [16]. Tayel et al. [17] remarked that the effectiveness of PV coatings can be significantly affected by geographical and climatic factors (e.g., dust deposition problems that occur in desert and plateau areas [18]). The nanocoating tested by Tayel et al. [17] was able to reduce the accumulated dust density by 57% compared to an uncoated reference PV panel.

Regarding Brazilian studies, these are limited and scarce. Araújo et al. [19] investigated the use of granite industry waste combined with chromium oxide to create selective coatings for solar collectors. The study demonstrated that a 50/50 mix of granite powder and chromium oxide achieved an efficiency comparable to commercial selective surfaces, offering a cost-effective and sustainable alternative. Papakonstantinou et al. [20] developed a scalable, spray-coated, transparent superhydrophobic coating to reduce dust accumulation and water usage in PV systems. Field tests were conducted in various locations, including Rio de Janeiro (Southeast Brazil), to assess the effectiveness of the coating under real-world conditions. Mazur et al. [21] developed self-cleaning, non-fouling, transparent, and weather-resistant coatings for PV modules. These coatings were based on metal oxides and nitrides and produced by plasma magnetron sputtering technology. Finally, Andrade et al. [22] analyzed 23 elastomeric coatings available in Brazil, assessing their chemical composition and optical properties, concluding that coatings with higher oxygen content and lower carbon content exhibited increased solar reflectance, particularly in the near-infrared spectrum.

Life cycle assessment (LCA) is a consolidated methodology to quantify the potential environmental impacts associated with a product, process, or activity [23]. Although there are many studies on the development of PV coatings, especially some recent reviews, there are limited LCAs on the subject. Most LCA studies focus on solar selective thin films, such as Sánchez-Cruces et al. [24], who carried out an LCA for solar selective coatings of black cobalt, obtained by electrodeposition and the sol-gel technique. Four different production lines were studied by Rossi et al. [25]: spin coating, blade coating, spin coating + press, and blade coating in a glovebox, with the blade coating process presenting the lowest environmental impacts. Focusing on PV coatings, the study of Laopreecha et al. [26] focused on a nano-silica coating with superhydrophobic and self-cleaning properties, designed to enhance the efficiency of PV panels by reducing the accumulation of dust and other pollutants. The results revealed that the increase in efficiency due to the coating could reduce the necessary area of PV panels.

There is a wide variability in solar coatings and manufacturing techniques. This study aims to identify the key environmental hotspots in an experimental, lab-scale research project focused on the manufacturing processes of a pre-developed solar coating. This study details the synthesis and deposition procedure of a solar coating that is constituted by two layers: the first based on tetraethyl orthosilicate (TEOS) and the second based on titanium isopropoxide (TTIP). The LCA methodology is applied to quantify the carbon footprint associated with the coating. This study is the first to report a detailed LCA inventory for the manufacturing processes of solar coatings and presents the corresponding GHG emission results.

2. Materials and Methods

2.1. Preparation of the Coatings

The inventory of the solar coating is based on the study by Cavalcante [27], conducted at the Laboratory for the Synthesis and Characterization of Thin Films (LabFilm) at the Federal University of Paraíba (in João Pessoa, Northeast Brazil).

The coating consists of two layers (based on TEOS and TTIP), developed using the sol-gel method and deposited via the dip-coating technique. The coatings were characterized through UV-Vis-NIR spectroscopy analyses, scanning from 220 to 1400 nm using the integrating sphere accessory, photocatalytic tests in a UV-C radiation chamber for 30 min, and wettability tests using an open-source image processing program. Finally, the glass substrates with the coatings were attached to photovoltaic modules, and the assembly was subjected to tests on a prototype produced by Cavalcante [27], after which electricity generation data were collected to assess the performance of the coatings. More details can be found in Appendix A.

For the preparation of the coatings, a magnetic stirrer on a ceramic platform was used (Figure 2), following all instructions provided by Zäll et al. [28]. During the synthesis of the coatings, the temperature variation was measured throughout the process.

For the TEOS-based coating, the reagents used were tetraethyl orthosilicate 99%, deionized water, 99.8% ethanol, and 37% HCl P.A./ACS as the precursor agents, solvent, and catalyst, respectively, at a molar ratio of 1:15.47:31.28:6.87. The solution was stirred for 180 min at 50 °C and then left to rest for 24 h for subsequent application.

For the TTIP-based coating, 1 mole of titanium (IV) isopropoxide 97% P.A. was used, added slowly and continuously under agitation to 4 moles of glacial acetic acid P.A. After homogenizing the mixture, isopropyl alcohol P.A. was added in an amount equal to the mixture to delay gel formation, and finally, 0.1 mole nitric acid was added in a volume equivalent to 5% of the total volume of the mixture. The mixture was stirred for 120 min at 50 °C and left to rest for subsequent application.

The depositions were carried out using the dip-coating process, with glass substrates being immersed in the solution for 5 s under ambient conditions of 25 °C and 30% humidity. The samples were left to air dry for 1 min and then placed in an oven for 24 h at 40 °C for drying. Figure 3 shows a flowchart for the preparation of the coatings.

2.2. Life Cycle Assessment

LCA is a consolidated and well-known methodology to quantify the potential environmental impacts of processes and activities. The main standards for LCA are ISO 14040 [29] and ISO 14044 [30]. ISO 14040 [29] defines the principles and framework for LCA, outlining the systematic approach for the evaluation of environmental impacts. ISO 14044 [30] builds upon ISO 14040 [29] by providing specific requirements and guidelines for conducting an LCA study, including the selection of impact categories and methods.

LCA basically has four inter-related steps [29,30]: the definition of objective and scope, construction of inventory, analysis of inventory, and finally, the results and interpretation. The first step defines the purpose of the LCA study, sets its boundaries, and defines the functional unit (to which all inputs and outputs are related). The second step collects and quantifies data on all inputs and outputs of the system depending on the life cycle stages included (raw material extraction, production, use, disposal, etc.). The third step translates inventory data to environmental impacts, based on the selection of an environmental impact assessment method. The last step analyzes the results from the inventory and impact assessment phases to draw conclusions and make recommendations.

The LCA was developed within SimaPro software v.9.6.0.1 [31], using the Ecoinvent version 3.8 database [32]. The environmental impact assessment method used was the IPCC 2021 GWP 100y [33], which groups carbon emissions in terms of carbon dioxide-equivalents (CO₂-eq) over 100 years. Due to widespread public concern about climate change (and global warming), GHG emissions have become an effective means of conveying environmental impacts [34].

The functional unit used herein, to which all input and output flows are related, was the preparation of one batch of coating (consisting of two solutions—one based on TEOS, one based on TTIP). The LCA was prospective (based on experiments at a small scale with primary data).

Table 1. Inventory of the antireflective coating: material flows.

TEOS Coating
Material	Amount (mol)	Molar Mass (g/mol)	Density (g/mL)	Volume (mL)	Chemical Formula
Tetraethyl orthosilicate (TEOS), 99.0%	1	208.33	0.94	221.63	Si(OC2H5)4
Deionized water	15.47	18.02	1	278.77	H2O
Ethyl alcohol, absolute, 99.8%	31.28	46.07	0.789	1826.45	C2H6O
Hydrochloric acid	6.87	36.46	1.18	212.27	HCl

and

Table 2. Inventory of the self-cleaning coating: material flows.

TTIP Coating
Material	Amount (mol)	Molar Mass (g/mol)	Density (g/mL)	Volume (mL)	Chemical Formula
Titanium (IV) isopropoxide	1	284.22	0.955	297.61	TI[OCH(CH3)2]4
Glacial acetic acid	4	60.05	1.049	228.98	CH3COOH
Isopropyl alcohol	Total volume of the above two items	60.1	0.789	526.59	C3H7OH
Nitric acid	0.1 molar solution of nitric acid in distilled water	-	-	0.22	HNO3
Distilled water		-	-	52.44	H2O

detail the material flows involved in producing the solar coating solutions.

Electricity consumption was based on the most recent Brazilian electricity mix available [35]: 62.18% hydroelectric, 15.32% wind, 10.60% solar, 3.67% biomass, 4.38% natural gas, 2.36% nuclear, 1.20% coal, and 0.19% oil, and the methodology proposed by Carvalho and Delgado [36]. The Ecoinvent database [32] required adaptation to obtain titanium isopropoxide, formed by the reaction of titanium tetrachloride with isopropanol.

The two coating layers were deposited on glass substrates (36.6 cm × 22.3 cm), with the following thicknesses: 328 nm for the TEOS coating layer, then 1055.5 nm for the titanium coating layer, totaling 1383.5 nm (corresponding to volumes of 0.0268 mL of TEOS coating and 0.0861 mL of titanium coating). The electricity consumption for the deposition of each layer was allocated based on their thicknesses: 23.71% for the TEOS coating and 76.28% for the titanium coating.

Table 3. Electricity consumption to produce the solutions.

Coating Layer	Equipment	Time (Hours)	Power (W)	Allocation (%)	Electricity (kWh)
TEOS	Magnetic agitator	3	750	100%	2.25
	Muffle furnace	24	750	23.71%	4.27
	Total				6.52
TiO2	Magnetic agitator	2	750	100%	1.50
	Muffle furnace	24	750	76.28%	13.72
	Total				15.22

summarizes the consumption of electricity in the procedures.

3. Results and Discussion

The yields of the TEOS and TTIP solutions were 2539.12 mL and 1105.84 mL, respectively.

Table 4. Carbon footprint for the TEOS coating.

Antireflective Coating
Material	Amount	Emission Factor	Carbon Footprint (kg CO2-eq)
Tetraethyl orthosilicate	221.63 mL	4.78 kg CO2-eq/L	1.0594
Deionized water	278.77 mL	0.000311 kg CO2-eq/L	8.67 × 10−5
Ethyl alcohol	1826.45 mL	1.17 kg CO2-eq/L	2.1369
Hydrochloric acid	212.27 mL	1.05 kg CO2-eq/L	0.2229
Electricity	6.52 kWh	0.152 kg CO2-eq/kWh	0.9910
		TOTAL	4.4103 (1.7368 g CO2-eq/mL)

and

Table 5. Carbon footprint for the TTIP coating.

Self-Cleaning Coating
Material	Amount	Emission Factor	Carbon Footprint (kg CO2-eq)
Titanium isopropoxide	297.61 mL	6.03 kg CO2-eq/L	1.7946
Glacial acetic acid	228.98 mL	1.67 kg CO2-eq/L	0.3824
Isopropyl alcohol	526.59 mL	4.74 kg CO2-eq/L	2.4960
Nitric acid	0.22 mL	2.87 kg CO2-eq/L	0.0006
Distilled water	52.44 mL	0.00601 kg CO2-eq/L	0.0003
Electricity	15.23 kWh	0.152 kg CO2-eq/kWh	2.3153
		TOTAL	6.99 (6.3204 g CO2-eq/mL)

show the carbon footprint associated with producing the corresponding volumes of coating. The emissions associated with the consumption of 1 kWh from the Brazilian electric grid at low voltage are 0.152 kg CO₂-eq.

The carbon footprint associated with the two batches of coatings was 4.82 + 6.99 = 11.81 kg CO₂-eq. Considering that the thickness of the TEOS coating layer was 328 nm, this is equivalent to 0.0268 mL on the glass substrate. For the titanium coating layer, the thickness was 1055.5 nm, equivalent to 0.0861 mL. Considering these layer thicknesses on the glass substrate, the emissions were 0.0465 g CO₂-eq for the TEOS coating layer and 0.5445 g CO₂-eq for the titanium coating layer, totaling 0.5910 g CO₂-eq.

A simulation was then carried out, based on the 16.4 MW solar PV system proposed by Shultz and Carvalho [10], which contains 59,248 solar panels. The experimental results of Cavalcante [27] (Appendix A) indicate that there is a 4.5% increase in electricity production by the coated PV panel. Simulating the application of both coating layers on the Canadian Solar PV panel CS6K-280M (1.6368 m²), 0.5368 mL and 1.7276 mL were applied, respectively, for the TEOS and titanium coatings. The emissions associated with coating one PV panel were, therefore, 0.9324 and 10.9193 g CO₂-eq, totaling 11.8518 g CO₂-eq per PV panel.

Table 6. Carbon footprint of the 16.4 MW solar-coated PV system (based on Schultz and Carvalho [10]).

Emissions (kg CO2-eq)
Item
	Coated PV panels	20,054,310
	Cables, grounding	1,048,161
	Inverters	395,442
	Transformers	440,978
	Administrative building	115,522
	Junction boxes	15,426
O&M
	Replacement of coated PV panels	250,354
	Replacement of inverters	98,860
	Washing of panels	57,339
	Mowing	59,094
TOTAL		22,535,486

details the emissions of a 16.4 MW solar PV system when the two coating layers were applied.

Although there was an increase in emissions due to the application of the solar coating, the production of electricity also increased.

Table 7. Annual electricity production and avoided emissions, and overall balance of emissions of the updated PV system.

	Annual Electricity [MWh/Year]	Avoided Emissions [kg CO2-eq/Year]	Balance of Emissions [kg CO2-eq]
Year			22,535,486
1	33,969.64	5,163,385	17,372,101
2	33,731.85	5,127,242	12,244,860
3	33,495.73	5,091,351	7,153,509
4	33,261.26	5,055,711	2,097,797
5	33,028.43	5,020,321	−2,922,524
6	32,797.23	4,985,179	−7,907,703
7	32,633.25	4,960,253	−12,867,957
8	32,470.08	4,935,452	−17,803,409
9	32,307.73	4,910,775	−22,714,183
10	32,146.19	4,886,221	−27,600,404
11	31,985.46	4,861,790	−32,462,194
12	31,825.53	4,837,481	−37,299,675
13	31,666.40	4,813,293	−42,112,968
14	31,508.07	4,789,227	−46,902,195
15	31,350.53	4,765,281	−51,667,476
16	31,193.78	4,741,454	−56,408,931
17	31,037.81	4,717,747	−61,126,678
18	30,882.62	4,694,158	−65,820,836
19	30,728.21	4,670,688	−70,491,524
20	30,574.57	4,647,334	−75,138,858
21	30,421.69	4,624,098	−79,762,955
22	30,269.59	4,600,977	−84,363,932
23	30,118.24	4,577,972	−88,941,905
24	29,967.65	4,555,082	−93,496,987
25	29,817.81	4,532,307	−98,029,294
Total	793,189.34

shows the annual electricity produced by the coated PV system, the avoided emissions in comparison with the electric grid (0.152 kg CO₂-eq/kWh), and the overall balance of emissions.

The data in Table 7 indicate increased electricity production in the coated PV system, which increased from the original 759,032.86 MWh [10] to 793,189.34 MWh over 25 years. The center column of the table displays the emissions avoided by self-producing the electricity instead of consuming from the electric grid. The rightmost column shows the balance of emissions over the lifetime of the system, indicating that between years 4 and 5, the solar PV system achieved overall negative emissions, with lifetime avoided emissions of −98,029,294 kg CO₂-eq.

The lifetime avoided emissions metric helps to identify which products and services can deliver a positive environmental impact and clearly quantifies that impact. The overarching aim is to move beyond simply addressing the risks associated with climate change and instead capture the opportunities presented by the transition to a low-carbon economy.

The basis of any avoided emission calculation is a business-as-usual (baseline) scenario, or what would happen in the absence of action (Figure 4) [37]. In the case of the coated solar PV system, the baseline scenario is the consumption of electricity from the Brazilian electric grid.

It must be highlighted that the solar coatings are still in the experimental phase, with currently insufficient information regarding their durability. In this study, it was assumed that the coating would have the same service life as the PV modules (25 years), without the need for reapplication.

A sensitivity assessment was carried out to verify the impact of shorter lifetimes of the solar coating on the overall avoided emissions.

Table 8. System emissions and overall avoided emissions based on the lifetime of the coating.

Lifetime of Coating [Years]	System Emissions [kg CO2-eq]	Avoided Emissions [kg CO2-eq]
None	22,534,773	−92,838,222
25	22,535,486	−98,029,294
10	22,536,556	−98,028,224
5	22,538,340	−98,026,440
1	22,552,606	−98,012,174
0.5	22,570,439	−97,994,341
1/12	22,748,766	−97,816,014

shows the results considering lifetimes of 1 month, 6 months, 1 year, 5 years, and 10 years on the avoided emissions of the solar-coated PV system.

The initial assumption of a 25-year coating lifetime without reapplication was intended as a baseline “best-case” scenario to isolate the maximum potential benefits of the technology and simplify the initial system boundary. Such a lifetime is unlikely without degradation from weathering, abrasion, or UV exposure. The results of the sensitivity analysis, shown in Table 8, incorporate the environmental impacts of reapplication, thereby reflecting more realistic operational conditions. This addition enables a clearer understanding of how durability assumptions influence the overall results and ensures that the findings remain robust across a range of plausible scenarios. Even when the solar coating is applied once a month, there are still significant environmental benefits being realized in terms of overall avoided emissions.

Nevertheless, it is important to acknowledge that the adopted model assumes a constant rate of environmental impact per application, based solely on the amount of coating material. If the reapplication process involves more complex operational steps, such as intensive pre-cleaning, use of solvents, specialized transportation, or additional energy consumption, the impacts per reapplication can be underestimated.

A potential indirect environmental benefit of the coating is the reduced frequency of PV module cleaning, which could translate into decreased consumption of water and cleaning agents. Nonetheless, quantitative data on the number of cleaning cycles that could be avoided over the panels’ lifetime are not yet available.

Incorporating these aspects could further enhance the environmental performance of the system, thereby reinforcing the benefits of the coating from a broader sustainability perspective. The gaps identified in this analysis highlight important avenues for future research, including experimental studies to evaluate the actual durability of the coating and the effective savings in water and cleaning resources under real operating conditions. Also, as mentioned by Abreu et al. [38], the LCA results can be combined with thermoeconomics, and the emissions associated with internal flows and products can be determined by adapting thermoeconomic equations: monetary costs are replaced by environmental costs, while capital, maintenance, and operational cost rates are reinterpreted as life cycle emission rates. Acknowledging these limitations does not undermine the robustness of the current analysis; rather, it strengthens the argument by demonstrating methodological transparency and the developmental potential of the technology under study.

Conducting an LCA of solar coatings is relevant, even before commercial deployment, for the early identification of environmental hotspots. LCA allows for identifying which stages of the process, such as raw material sourcing, energy use, or waste generation, contribute the most to the environmental impact. In the work by Carvalho et al. [39], a pilot biodiesel plant was studied, and it was verified that refined Brazilian soybean oil accounted for most carbon emissions across the entire lifecycle of the biodiesel production system (the primary environmental impacts were associated with land use and fertilizer application). Menezes et al. [40] focused on two deposition techniques for the production of selective surfaces and found that laboratory-scale LCAs are particularly relevant for the comparison of early-stage processes, helping to identify the most advantageous route for upscaling from an environmental perspective. The authors verified that electricity consumption was responsible for most emissions, being a hotspot of the production process. Early identification of these hotspots enables the development of more sustainable processes before scaling up.

The prospective LCA presented herein helped identify the hotspots of the processes that have margins for improvement, providing guidance regarding process upscaling.

4. Conclusions

The findings demonstrate the potential of solar coatings to enhance the environmental performance of solar energy systems, thereby reinforcing their relevance in climate change mitigation strategies. However, the scarcity of Brazilian national studies on life cycle assessment and solar coatings in photovoltaic systems underscores the need for further investigation.

A detailed inventory is presented for the preparation of a solar coating consisting of two layers: one based on tetraethyl orthosilicate (TEOS), and one based on titanium (IV) isopropoxide (TTIP).

The carbon footprint of the TEOS-based coating is 1.7368 g CO₂-eq/mL, while the TTIP-based coating presented emissions of 6.3204 g CO₂-eq/mL. When considering application on a commercial PV panel (1.6368 m²), the associated emissions for coating a single PV panel are 0.9324 g CO₂-eq for TEOS and 10.9193 g CO₂-eq for TTIP, resulting in a total of 11.8518 g CO₂-eq per panel.

Considering the average increase in electricity production (4.5%) obtained experimentally, a 16.4 MW PV power plant can reach lifetime avoided emissions of −98,029,294 kg CO₂-eq. The sensitivity analysis considered variations in reapplication of the coating throughout the lifetime of the system. Even under frequent reapplication scenarios—down to one-month intervals—the solar coating continues to deliver net environmental benefits, with significant avoided emissions compared to the uncoated reference system. This robustness across scenarios underscores the potential of the coating to contribute meaningfully to emission reductions, even under conservative durability assumptions.

Despite the consistency of the results obtained, this study also revealed areas for future research. Long-term evaluation of coating durability and the quantification of reductions in PV module cleaning frequency, along with the associated environmental benefits, could complement the present analysis and further deepen the understanding of the advantages provided by this technology. These aspects represent promising avenues to support the large-scale adoption of self-cleaning coatings and strengthen their integration into policies and projects aimed at a sustainable energy transition. In the production of solar coatings, it is common for analyses to be limited to laboratory applications. This study contributes to advances in this type of research by presenting a detailed inventory of material and energy flows.

Finally, coupling environmental and economic evaluations would be a valuable avenue for future research to support more holistic decision-making. Future research can verify the economic aspect, regarding the cost of producing the solar coatings and the ultimate carbon emission reduction benefits, including considerations on potential construction cost savings.