Impact of air barriers application in LCA and LCC of naturally ventilated dwellings in mild climate regions

Impact of air barriers application in LCA and LCC of naturally ventilated dwellings in mild climate regions Vitor E.M. Cardoso a,b, ⇑ , M. Lurdes Simões b , Nuno M.M. Ramos b , Ricardo M.S.F. Almeida b,c , Manuela Almeida d , Ricardo Mateus d a BUILT CoLAB, Collaborative Laboratory For The Future Built Environment, Rua do Campo Alegre, 760, 4150-003 Porto, Portugal b CONSTRUCT-LFC, Department of Civil Engineering, Faculty of Engineering, University of Porto, 4200-465 Porto, Portugal c Polytechnic Institute of Viseu, School of Technology and Management, Department of Civil Engineering, Campus Politécnico de Repeses, 3504-510 Viseu, Portugal d ISISE, University of Minho, School of Engineering, Department of Civil Engineering, Campus de Azurém, Guimarães 4800-058, Portugal article info Article history: Received 22 July 2022 Revised 1 November 2022 Accepted 14 November 2022 Available online 17 November 2022 Keywords: LCA application LCCA Airtightness Air change rate Building energy performance Building envelope abstract Assessing singular elements that constitute the air barrier of a building envelope is quite unfeasible in Life Cycle Assessment (LCA). The study of these solutions through this particular scope is often overlooked. Two major aspects contribute to it: the complexity of the relationships between elements and the reduced embodied impact of these materials in the overall construction or retrofitting works. This work uses LCA and Life Cycle Costing (LCC) to study the viability of applying two envelope air barrier solutions in dwellings with excessive air change rates and equipped with different heating systems. The application of air barrier solutions resulted in average energy consumption savings in urban terrain, almost half of those in rural terrain during the heating season. Environmental performance and life cycle costs revealed mechanically (MECH) fastened air barriers to outperform fluid (FLUID) applied ones. The median annualized cost of adopting a FLUID solution was almost four times that of a MECH solution. Dwellings equipped with electric radiators ranked first in the shortest average Energy Payback Period (EPP) and the highest average Reference Service Life (RSL) savings. With the current analysis, the adoption of MECH solutions is recommended, independently of the heating system the dwelling is equipped with. Ó2022 The Authors. Published by Elsevier B.V. This is an open access article underthe CCBY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 1. Introduction The Architecture, Engineering and Construction (AEC) industry has some of the most challenging sustainability issues [1]. Environmentally, the sector is one of the most impactful regarding energy consumption and carbon emissions [2]. Economically, buildings are assets with high initial and operational costs and long life cycles [3], and socially, as humans spend most of their time inside buildings, the numerous challenges have deep implications in societies [4]. Among several reasons for which its structure and inefficiencies can be highlighted [5], the sector has one of the most substantial potentials for reducing greenhouse gas emissions (GHG) and cutting energy consumption [6]. Principles of sustainable construction encompass reusing and recycling resources, reducing their consumption, protecting natural systems, and eliminating toxic materials from the life cycle [7]. Life cycle assessment (LCA) is a major tool driving decisions and policies [8]. In the initial approach of the Energy Performance of Building Directive (EPBD) [9], priority was on energy efficiency measures in the form of design or retrofit packages pointing to costoptimality through the life cycle of reference buildings [10,11].In the current version of the EPBD [12], there is a shift to clean energy adoption, deployment of renewables, and general decarbonization of the building stock [13]. This trend requires the nearly Zero Energy Buildings (nZEB) and ZEB to focus significantly on carbon targets, not solely on energy ones [14–16]. Embodied carbon and energy refer to the CO 2 emissions produced and the primary energy required, respectively, during the resource extraction, transportation, fabrication, assembly, disassembly and end-of-life disposal of a particular product [17]. Operational carbon and energy refer to the resources spent on building usage, such as heating, cooling, and powering. Throughout the literature, between embodied and operational, one finds average values of GHG emissions, over 50 %, and energy https://doi.org/10.1016/j.enbuild.2022.112667 0378-7788/Ó2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). ⇑ Corresponding author at: BUILT CoLAB, Collaborative Laboratory For The Future Built Environment, Rua do Campo Alegre, 760, 4150-003 Porto, Portugal. E-mail address: [email protected] (V.E.M. Cardoso). Energy & Buildings 279 (2023) 112667 Contents lists available at ScienceDirect Energy & Buildings journal homepage: www.elsevier.com/locate/enb consumption, around 80 %, associated with the latter [18]. Lowenergy buildings face a shifting paradigm [19,20]. As the operational phase gets less impactful, the percent contribuition of the embodied energy increases [21–23]. Increasing insulation levels and adopting new technologies on active systems are two examples of the trend of rising embodied and declining operational impacts. Proportions can go as high as 40 % of embodied energy impact and as low as 60 % of operational energy impact [24,25], for considered lifespans of 50 years. However, even in low-energy buildings, the operational energy still contributes the most to the total. The air change rate (ACH) strategy of a building has a significant impact in the operational phase as the annual related heating energy loss represents around 35 % of the delivered energy for space conditioning [26–29].Itis dependent on ventilation effectiveness, which is closely related to and sensitive to the envelope airtightness performance [30– 32]. Designing or retrofitting of the latter often overlooks the first, which translates into disruptions and underperformance of the ventilation strategy as a whole [33–36]. Southern European countries, with mostly mild climates, experience this reality frequently, such as Greece, Italy, Portugal, and Spain. Of these, only Spain instated whole envelope airtightness requirements by the end of 2020, showing that the existing envelopes encounter similar challenges for retrofitting works. Achieving high levels of airtightness often relies on air barrier systems [37] dependent on air barrier materials, components and accessories [38–41]. Mechanically wrapped air barrier materials include polyethene fibre wraps or gypsum plastering. Windows and doors are examples of air barrier components. Air barrier accessories refer to tapes and sealants, elements that maintain the airtightness performance in the penetrations and joints of air barrier materials and the joints between materials and components. Their performance relies on durability, compatibility and constructability as they are often complex regarding proper installation [42–45]. Constructability issues are mostly absent when the barrier is a fluid-applied polymer membrane. This system adapts better than its mechanical alternatives to complex envelope geometries and is less prone to human fault during application. However, the chemical composition of such a membrane highly influences its movement accommodation factors and the need for primers for proper substrate adhesion [46–49]. Partially due to the complexity of these relationships [50–52], partially due to a reduced embodied impact of these materials in the overall construction or retrofitting works, the study of airtightness solutions regarding life cycle assessment is often overlooked. Usually, when reporting airtightness and air infiltration improvements, they come as a by-product of the consideration of works not primarily intended to target the issue [53]. When directly considered, the reductions in air change rates generally result from window replacement [54]. Moreover, Environmental Product Declarations (EPDs) on complete air barrier solutions are scarce compared to the current availability of EPDs in individual elements of an air barrier [55]. Assessing singular elements that constitute the air barrier is quite unfeasible in an LCA approach. The available ones often do not comprise the same LCA information modules in the system boundaryand list of environmental parameters, creating difficulties in correctly summing their impacts. Also, the common unavailability of data regarding use stage modules results in the Reference Service Life (RSL) going unreported in an EPD, further hindering a comprehensive analisis of this stage. Additionally, quantifying singular elements such as point envelope penetrations or envelope elements perimeters is increasingly prone to misjudgement and bias compared to evaluating surface elements such as the envelope area. Therefore, considering whole envelope airtightness solutions seems to be a more plausible path. The present research’s objective is to assess the life cycle impacts and costs of applying two different envelope air barrier solutions to naturally ventilated dwellings already equipped with different heating systems, which have excessive heat losses due to ventilation. It aims to evaluate: The energy saving potentials during the operational phase by identifying the energy performance variations from a default scenario to one with an envelope air barrier; The environmental and economic viability of applying air barriers to the envelope of these dwellings by assessing the embodied energy payback periods and quantifying life cycle costs of such works. The equipment of the heating system may be physical in the case of retrofitting works or, in the case of new designs, be already planned regardless of the air barrier solution installation. 2. Methodology 2.1. Research strategy The developed research follows a clear flux of events. Fig. 1 portrays a flowchart of the methodology followed. It aims to ease the reader’s comprehension, improve the structuring and encapsulation of topics, and promote reproducibility. The methodology addresses four levels: (1) the sourcing of the case study dataset, (2) considered heating systems and air barrier solutions, (3) LCA and LCC considerations, and (4) heating loads calculation and research assumptions. Each of the following subsections will address these topics in detail. Detailing the heating systems occurs simultaneously with the exposition of life cycle cost methods and primary energy considerations. This way, it eases the comprehension of equations flow and their respective variables. 2.2. Case study dataset The case study dataset derives from a previous work [56], which outputted a large dataset on dwelling characteristics, their respective air change rates (ACHs) time series and labels. These labels represent dwellings with low, adequate, and high ACHs, according to defined lower (0.4 h 1 ) and upper (0.7 h 1 ) limits. The group of dwellings with high ACHs applied to the present research, as they experience a significant amount of time above the defined upper limit. From that work [56], the roof slope (RS), the side ratio (SR), and the number of vertical exposed surfaces (ES) all proved to have residual relative importance in explaining the variability of ACHs. The terrain was the categorical feature with the most relative importance. Because of these findings and for visualization purposes, only the dwellings with a 20°RS, a 2:1 SR, four ES, two floors (NF), and three vertical extraction ducts (ND) undergo evaluation in this research. The dataset for analysis comprises 298 unique dwellings: 171 in rural terrain and 127 in urban terrain. The importance of wind speed at the building site was the major reason for dividing the analysis according to the two types of terrain. Table 1 informs on the centrality and variability of these dwellings’ features. The average dwelling in rural terrain has a volume of 339.7 m 3 and an envelope area of 248.1 m 2 , which translates into an average air permeability of 13.3 m 3 /(hm 2 ) at 50 Pa. In urban terrain, the average dwelling has a volume of 305.1 m 3 and an envelope area of 231.6 m 2 , resulting in an average air permeability of 14.1 m 3 / (hm 2 ) at 50 Pa. Vitor E.M. Cardoso, M. Lurdes Simões, Nuno M.M. Ramos et al. Energy & Buildings 279 (2023) 112667 2 Fig. 1. Flowchart of the proposed methodology. Vitor E.M. Cardoso, M. Lurdes Simões, Nuno M.M. Ramos et al. Energy & Buildings 279 (2023) 112667 3 2.3. Air barrier solutions and dwellings type While research reports significant disparities in materials’ airtightness performance in a controlled environment [40], one must assume that commercial products and systems are completely airtight when properly installed. Therefore, the main goal is not to compare solutions by their ability to perform in conferring airtightness, taken as similar, but by their environmental and cost performance as a function of a dwelling airtightness level and installed heating system. With this rationale, two external air barrier systems, with their own EPDs, were chosen for analysis: a mechanically fastened (MECH) [57] and a fluid applied (FLUID) [58]. These solutions already account for the estimated number of separate elements needed to perform and the detailing between the system and other components, such as windows, and technical penetrations, such as electrical and plumbing systems, since they include self-adhered flashings and tapes. They share system boundary modules, production and installation phases, and endof-life transport. Their impacts are reported in the functional unit of one square meter (m 2 ), leading to a straightforward comparison of environmental performance [59]. A functional unit is a quantified product description that serves as a basis of reference for the subsequent impact assessment calculations [60]. The scope of the airtightness solutions is closely associated with dwelling construction types. Dwellings of heavyweight construction are mostly susceptible to point penetrations, such as pipes and cables, and elements perimeters, such as windows and doors. The opaque envelope is mostly homogeneous in these, providing one or more highly airtight layers by default. By comparison, most lightweight dwelling construction types experience an increase in the contribution of the opaque envelope to the whole envelope leakage. This is mostly justified by the higher partitioning of envelope elements, e.g., Structural Insulated Panels (SIPs) or wood frames, to comply with geometry requirements. Because of it, most of the buildings of this last construction type require purpose-provided air barriers as a necessary component of an airtight envelope solution. Thus, while the chosen air barrier solutions provide detailing of the opaque envelope with point penetrations and elements perimeters, i.e. there is a benefit for heavyweight constructions as well. They are more suitable for lightweight constructions, which experience a higher payoff. Additionally, the potentially occurring moisture condensation in the envelope inner layers has far less serious consequences in heavyweight constructions compared to lightweight constructions. Hence, the present research is most appropriately applied to this latter construction type. 2.4. Environmental indicators The environmental indicators were extracted from the two used EPDs [57,58] (Table 2). They include: abiotic depletion potential for both fossil (ADPF) and non-fossil resources (ADPE); global warming potential (GWP); depletion potential of the stratospheric ozone layer (ODP); formation potential of tropospheric ozone photochemical oxidants (POCP); acidification potential of land and water (AP); and eutrophication potential (EP). These indicators are assessed using the characterization factors of CML [61]. Two additional indicators related to energy consumption were extracted from the EPDs: non-renewable primary energy (NRPE) and total primary energy (TPE), obtained through the Cumulative Energy Demand (CED) single-issue method [62]. 2.5. Life cycle cost methods and primary energy considerations Life cycle cost (LCC) methods enable the effective evaluation of investment options, considering the impact of all costs instead of initial expenses alone, and planning the management of buildings and infrastructures during their lifetime [13]. As the EPDs of the solutions do not indicate a Reference Service Life (RSL), their lifespan is treated as a constituent of the envelope and similar to insulation materials. The lifespan considered was 40 years, and it is a compromise between: the stated 30 years calculation period, by the European Commission [63], for residential buildings, on establishing a comparative methodology framework for calculating cost-optimal levels of minimum energy performance requirements for buildings and building elements; the used lifespans in the literature, 50 years for envelope components [6465], and more specifically, 40 years for air barrier elements [66]; the 40 years indicated life span of major replaceable components in international standards such as ISO 15686–1:2011 [67]. There are no maintenance costs to be included in the calculation of global costs, as they are elements of the envelope non-accessible after installation. Therefore, a dwelling RSL savings is simply the Table 1 Descriptive statistics of the case study’s dwellings with a 20°RS, 2:1 SR, 4 ES, 2 NF, and 3 ND, by terrain, rural and urban. P(ACH condition) stands for the percentage of time the ACH complies with the stated condition. Feature Metric Rural (N = 171) Urban (N = 127) ACH [h 1 ] Mean 1.11 0.85 Std. dev. 0.39 0.25 P(ACH < 0.4 h 1 ) [%] Mean 2.61 1.76 Std. dev. 4.64 2.88 P(0.4 h 1 < ACH < 0.7 h 1 ) [%] Mean 25.60 38.78 Std. dev. 16.21 25.19 P(ACH > 0.7 h 1 ) [%] Mean 71.79 59.45 Std. dev. 19.05 26.97 Airtightness (n 50 )[h 1 ] Mean 9.08 9.96 Std. dev. 3.64 3.82 Airflow exponent (n) [-] Mean 0.59 0.59 Std. dev. 0.04 0.04 Floor area (AF) [m 2 ] Mean 129.18 116.46 Std. dev. 62.87 59.80 Ceiling height (CH) [m] Mean 2.63 2.62 Std. dev. 0.28 0.25 Table 2 Life cycle assessment indicators considered for the comparison of air barrier solutions by the functional unit (m 2 ). Parameter Unit Mechanically fastened (MECH) [57] Fluid applied (FLUID) [58] ADPE kg Sb eq. 2.64E07 1.61E05 ADPF MJ 1.17E + 01 7.21E + 01 GWP kg CO 2 eq. 5.38E01 3.83E + 00 ODP kg CFC-11 eq. 1.24E06 3.62E07 POCP kg C 2 H 4 eq. 2.90E04 8.15E03 AP kg SO 2 eq. 1.45E03 1.07E02 EP kg PO 4 3 eq. 1.44E04 1.36E03 NRPE MJ 1.24E + 01 7.60E + 01 TPE MJ 1.27E + 01 8.00E + 01 Vitor E.M. Cardoso, M. Lurdes Simões, Nuno M.M. Ramos et al. Energy & Buildings 279 (2023) 112667 4 difference in operational costs between a scenario of absence, and a scenario of application of one of the proposed airtightness solutions, through its RSL (Eqs. (1) and (2)). For each year after, the annual savings in operational costs adjust by a discount factor (Eq. (3)). A discount factor is a weighting term to convert future values into net present values (NPV) [11]. The CYPE software was applied to generate construction prices, by early 2021 in Portugal, and quantify the initial investment by using similar solutions to the ones considered in the present research [68]. The MECH solution was considered a vapor barrier with an air permeability of 0.03 m 3 /(hm 2 ) at 50 Pa, at 7.68 € /m 2 (VAT included). The FLUID solution adopted the compound price of a spray-applied liquid polyurethane-based waterproofing coating with an air permeability of 0.0003 m 3 /(hm 2 ) at 50 Pa, at 20.51 € /m 2 (VAT included). The prices of the solutions already account for the installation labor. Eq. (4) displays the formula for the payback period (PP). The PP is when, the financial savings from a certain improvement scenario equal its initial investment [69].A positive PP indicates that the financial savings exceed the initial cost before the end of the RSL. RSL savings ¼X RSL i¼1 D C a;i R d;i C 0 ð1Þ D C a;i ¼ D Q v;i g k P s ð2Þ R d;i ¼1 1þ r 100  i ð3Þ PP ¼j forC 0 6X j i¼1 D C a;i R d;i ð4Þ RSL sa v ings reference service life savings ( € ). D C a;i difference in operational costs between a scenario of absence or application in the year i ( € ). R d;i discount factor in the year i (-). C 0 initial investment in applying the airtightness solution ( € ). D Q v ;i difference in heat losses between a scenario of absence or application in the year i (kWh). g k nominal efficiency of the heating system k (-). P s energy price by source s ( € /kWh). rdiscount rate in year i (%). PP payback period (year). Table 3 compiles the considered heating systems, their efficiencies, respective energy sources, prices, and conversion factors from final to primary energy. The selected four heating system types are the most common in the Portuguese built stock [73]. National technical documents [71] provided the efficiencies ( g ) of the natural gas boiler, pellet heat recovery, and heat pump for residential buildings. A heating system efficiency translates the ratio between the total energy output and the total energy input. The minimum requirements applied for all the heating systems.For the sake of a top-performing heating system, the research considered an additional heat pump with a higher efficiency. Energy costs on electricity and natural gas refer to Portuguese energy prices (VAT included) for household consumers available at Eurostat [70] in the second half of 2020. For biomass as pellets, the used value refers to the value in Bioenergy Europe [72] for 2018, at 0.27 € /kg. As pellets’ Low Heating Value (LHV) is 18.84 MJ/kg [74], which equals 5.23 kWh/kg, their price conversion equals 0.0516 € /kWh. The discount rate (r) in energy prices is 3 %, in line with European Commission directives [75]. The heating system efficiencies ( g ) and the primary energy factors (F PE ) from final energy to primary energy apply in calculating the primary energy needs (Eq. (5)). These factors account for energy used for transportation, processing, losses, etc. They are sourced from additional Portuguese legislation [76] and are in fair agreement with the factors obtained using the CED v1.09 method [62] in SimaPro with data from the Ecoinvent database [77]. The latter sum up to 2.72 kWh PE /kWh for 1 kWh of electricity, 1.21 kWh PE /kWh for 1 m 3 of natural gas, and 0.96 kWh PE /kWh for 1 kg of wood chips that make up pellets. Eq. (6) gives the Energy Payback Period (EPP). Like PP, the EPP is the time, in years, that the primary energy savings from a before to after intervention scenario equals or overcomes the embodied energy of that same intervention. D PE a ¼ D Q v g k F PE ð5Þ EPP ¼EE 0 D PE a ð6Þ F PE conversion factor between final energy and primary energy for energy source s (kWh PE /kWh). D PE a annual difference in primary energy needs between a scenario of absence or application (kWh PE ). EPP energy payback period (year). EE 0 embodied energy of applying an airtightness solution (kWh PE ). 2.6. Heating loads and assumptions Being heat transfers by ACHs the scope of this research, and windows opening a prevailing strategy in the cooling season [78], one considers only heat losses through ACHs in the heating season. A value of 1250 °C heating degree-days (HDD) applies to the assessment, corresponding to the Portuguese Porto metropolitan area, which is in line with the region considered in the simulations of the ACHs time series of the case study dwellings. Eqs. (7) to (9) show the calculations required to obtain the heat losses through air change rates during the heating season. They correspond to the adaptation to the Portuguese legislation [79] of the quasi-steady state method of ISO 13,790 [80]. D Q v ¼0:024 HDD ðH v;before H v;after Þð7Þ Table 3 Considered heating systems and efficiencies, their energy sources and respective prices. Heating system Heating systemefficiency ( g ) Energy source Energy price [€/kWh] F PE [kWh PE /kWh] Electric radiator 1.0 Electricity 0.2133 [70] 2.5 Natural gas boiler 0.90 [71] Natural gas 0.0783 [70] 1.0 Pellet heat recovery 0.75 [71] Biomass 0.0516 [72] 1.0 Heat pump (min. efficiency) 3.00 [71] Electricity 0.2133 [70] 2.5 Heat pump (high efficiency) 5.00 Electricity 0.2133 [70] 2.5 Vitor E.M. Cardoso, M. Lurdes Simões, Nuno M.M. Ramos et al. Energy & Buildings 279 (2023) 112667 5 H v;before ¼0:34ACH before A f C h ð8Þ H v;after ¼0:34ACH after A f C h ð9Þ HDD heating degree days (°C). H v ;before coefficient of heat transfer during the heating season in the absence scenario (W/°C). H v ;after coefficient for heat transfer during the heating season in the application scenario (W/°C). ACH air change rate (h 1 ). A f floor area (m 2 ). C h ceiling height (m). One assumes the treatment of the whole opaque envelope area. Self-regulating vents integrate the window frames of the main housing divisions to offset the reduction in ACH from the absence of air infiltration through the building envelope. Since the vents are self-regulating, one assumes that, with the proper dimensioning, average ACHs values become similar to the average of the dwellings with adequate ACHs in the study from which the dataset is sourced [56], 0.55 h 1 and 0.65 h 1 for urban and rural locations. This assumption applies in assessing energy performance. The potential environmental impact of these vents, the embodied energy, and the initial investment are assumed to be residual compared to a window without the vents. Such assumption relies on the elements often integrating across window frame depth and their composition being similar to the frame for functional and aesthetic reasons. This hypothesis is true in design scenarios, as windows installation takes place anyway, and retrofit scenarios where window replacement often occurs without airtightness being the main focus. According to a study of reference buildings in the Portuguese built stock, the glazed area averages 20 % of the envelope area [81]. 3. Results and discussion 3.1. Energy performance In subsection 2.5, one admitted that the maintenance cost of an air barrier implementation would be null since there is no access to it after installation. Therefore, only the expenditure with energy consumption applies to annual costs. Fig. 2 presents several percentiles of the annual savings in energy consumption between the scenarios of absence and application of an air barrier solution. The mean in an urban environment is 1460 kWh, slightly below 40 % of that in a rural terrain, 867 kWh. The standard deviations stand at 1401 kWh and 810 kWh, respectively. As the distributions approach higher percentiles, the differences are emphasized. While these reductions in energy consumption are a positive outcome, a comprehensive outlook on the economic viability of these applications is only achieved by identifying which dwellings with which heating systems have positive PP. When so, one quantifies the savings through the RSL.. Additionally, one must assess the Energy Payback Period (EPP). It allows identifying if the saved primary energy in heating the indoor environment compensates for the embodied energy of each air barrier solution by the end of the RSL. 3.2. Environmental performance From the analysis of Table 2, only ODP presents lower impacts by square meter with FLUID instead of MECH. For all the other environmental indicators, the FLUID impacts significantly exceed MECH’s. The number of years the reduced primary energy from heating the dwellings take to equalize the embodied energy of the application, the EPP, present very different outcomes depending on the airtightness of the dwelling, the heating system, and the solution applied to the envelope (Fig. 3). Table 4 further informs EPP means and standard deviations for the considered air barrier solutions, terrains, and heating systems. Treating a dwelling envelope with the FLUID solution confers an EPP 6.32 higher than the MECH solution when comparing within the same terrain and heating system. On average, for the same heating system and air barrier solution, the EPP of a dwelling in an urban location is 1.63 higher than the EPP of a dwelling in a rural terrain. For the MECH solution, independently of the heating system and terrain, all the dwellings justify its implementation regarding primary energy, as the EPP never surpasses the respective RSL. For the FLUID air barrier, in rural terrain, only a portion of the dwellings equipped with a high efficiency heat pump does not have a positive EPP, being 6.4 %. In urban terrain, this percentage increases and is joined by a small percentage of dwellings equipped with minimum efficiency heat pumps and natural gas boiler, 18.1 %, 5.5 %, and 1.6 %, respectively. 3.3. Life cycle costs results Addressing the embodied energy payback does not provide the complete picture. One needs to appraise each dwelling’s savings through the RSL to assess the economic viability. Examining the dwellings with a positive and negative PP through their RSL savings by the heating system, airtightness solution, and airtightness performance (n 50 ) provides further understanding. Table 5 presents these data. While the great majority of the dwellings with electric radiators experience a positive PP, with the MECH solution, this percentage drops to 19.1 % in the ones equipped with high efficiency heat pumps. The natural gas boiler, the minimum efficiency heat pump, and the pellet heat recovery stove are in between by descending PP. The order of PP by heating system is unchanged in the FLUID solution, but the percentages drop drastically, ranging from 45.3 % to 1.0 % of dwellings with a positive PP. With MECH, in the groups with positive PP, the average RSL savings range from 5438 € to 1123 € . With FLUID, for the negative PP groups, the RSL savings range from 2437 € to 3853 € , showing the greater financial burden of this airtightness solution than MECH. Fig. 2. Percentiles of annual savings in energy consumption between absence and application scenarios for the dwellings of the case study dataset. Vitor E.M. Cardoso, M. Lurdes Simões, Nuno M.M. Ramos et al. Energy & Buildings 279 (2023) 112667 6 Fig. 3. Energy Payback Period (EPP) as a function of the airtightness performance (n 50 ) of the considered dwellings, by terrain, air barrier solution, and heating system in both scatter plots (left column) and boxplots (right column): a) rural terrain – MECH solution; b) urban terrain – MECH solution; c) rural terrain – FLUID solution; d) urban terrain – FLUID solution. Vitor E.M. Cardoso, M. Lurdes Simões, Nuno M.M. Ramos et al. Energy & Buildings 279 (2023) 112667 7 The lowest average n 50 in the dwellings with positive PP is 10.2 h 1 , in the MECH solution and 12.0 h 1 , in the FLUID solution. Still, they experience substantial variability, as the standard deviation ranges from 3.8 to 4.5 h 1 in MECH and from 2.1 to 5.0 h 1 in FLUID, respectively. To assess this relationship, one addresses the annualized savings through the RSL as a function of the airtightness performance (n 50 ). Since natural gas boiler, pellet heat recovery, and minimum efficiency heat pump have similar percentages of dwellings with positive PP, only the minimum efficiency heat pump is assessed (Fig. 4). These results portray the information with a higher degree of interpretability. For MECH, in a rural location, all dwellings have positive RSL annualized savings with an n 50 over 6.9 h 1 when equipped with electric radiators, over 9.7 h 1 with minimum efficiency heat pump, and over 11.7 h 1 with high efficiency heat pump. With the FLUID solution, these n 50 need to be over 8.9 h 1 , 14.6 h 1 , and 20.7 h 1 , respectively. The lower bound n 50 values increase in an urban location. For MECH, they are 9.1 h 1 , 13.5 h 1 , and 16.1 h 1 , and for FLUID, they reach 13.5 h 1 , 25.5 h 1 , and 26.4 h 1 , for the same order of heating systems. 3.4. Ranked heating systems Overall, for the research case study’s scope, for dwellings equipped with electric radiators, adopting the MECH solution almost always justifies the financial investment, especially in rural locations, as even for those with negative RSL savings, they are quite low. The opposite occurs with a high efficiency heat pump, especially in urban locations. For all the other dwellings, the relationship between airtightness performance and the heating system will condition the RSL savings to be positive or negative. As a more expensive solution, FLUID experiences even greater costs. The greater majority of the dwellings compute negative RSL savings. The exception is in dwellings equipped with electric radiators as a heating system, where the results are quite mixed. Table 6 indicates the heating systems ranked by EPP and RSL savings from applying an airtightness solution. The first ranked has the shortest average EPP and the highest average RSL savings. While scoring second in average EPP, the pellet heat recovery scores fourth in RSL savings, meaning that despite being environmentally superior, it comes with an aggravated financial burden compared to the natural gas boiler and the minimum efficiency heat pump. Table 4 EPP mean and standard deviation by air barrier solution, terrain, and heating system. EPP [years] Electric radiator Natural gas boiler Pellet heat recovery Min. efficiency heat pump High efficiency heat pump Air barrier solution Terrain MECH Rural Mean 0.46 1.03 0.86 1.37 2.29 Std. dev. 0.39 0.88 0.73 1.17 1.95 Urban Mean 0.75 1.68 1.40 2.24 3.73 Std. dev. 0.66 1.48 1.23 1.97 3.29 FLUID Rural Mean 2.89 6.50 5.42 8.67 14.45 Std. dev. 2.46 5.53 4.61 7.38 12.29 Urban Mean 4.71 10.61 8.84 14.14 23.57 Std. dev. 4.16 9.35 7.79 12.47 20.79 Table 5 Mean and standard deviations for the variables of interest of the dwellings with positive and negative PP, by heating system and airtightness solution. Pos. PP [%] n 50 [h 1 ] PP [years] RSL savings [€] Neg. PP [%] n 50 [h 1 ] RSL savings [€] Electrical radiator 80.2 Mean 10.2 11.0 5438 19.8 Mean 6.5 569 MECH Std. dev. 3.8 8.3 6192 Std. dev. 1.2 400 FLUID 45.3 Mean 12.0 18.2 5385 54.7 Mean 7.3 2437 Std. dev. 3.9 9.1 6327 std 1.6 1449 Natural gas boiler 49.7 Mean 11.8 17.9 2161 50.3 Mean 7.2 902 MECH Std. dev. 3.9 9.4 2567 Std. dev. 1.6 526 FLUID 11.1 Mean 16.4 25.8 2301 88.9 Mean 8.6 3124 Std. dev. 4.8 8.2 2856 Std. dev. 2.5 1592 Pellet heat recovery 39.3 Mean 12.5 19.5 1730 60.7 Mean 7.5 960 MECH Std. dev. 4.0 9.0 2048 Std. dev. 1.7 552 FLUID 5.0 Mean 18.7 26.1 2242 95.0 Mean 9.0 3326 Std. dev. 5.0 8.7 2468 Std. dev. 3.0 1592 Min. eff. heat pump 40.9 Mean 12.3 19.4 1775 59.1 Mean 7.5 955 MECH Std. dev. 4.0 9.2 2112 Std. dev. 1.7 546 FLUID 5.7 Mean 18.3 26.2 2228 94.3 Mean 8.9 3295 Std. dev. 4.8 8.5 2521 Std. dev. 2.9 1594 High eff. heat pump 19.1 Mean 14.5 23.2 1123 80.9 Mean 8.3 1076 MECH std 4.5 8.2 1335 Std. dev. 2.3 600 FLUID 1.0 Mean 24.8 27.7 1670 99.0 Mean 9.3 3853 Std. dev. 2.1 7.6 1446 Std. dev. 3.4 1482 Vitor E.M. Cardoso, M. Lurdes Simões, Nuno M.M. Ramos et al. Energy & Buildings 279 (2023) 112667 8 4. Conclusions The developed research explored the life cycle impacts of applying air barrier solutions in a set of naturally ventilated dwellings with excessive heat losses by air change rates. The life cycle assessment contributed to picturing and contextualizing the impacts of including air barrier solutions. While expectedly, the benefits of installing an air barrier system are larger in less airtight dwellings Fig. 4. RSL annualized savings, by terrain, as a function of airtightness performance (n 50 ): a) electric radiator – MECH; b) electric radiator – FLUID; c) min. efficiency heat pump – MECH; d) min. efficiency heat pump – FLUID; e) high efficiency heat pump – MECH; f) high efficiency heat pump – FLUID. Table 6 Ranked heating systems by shortest average EPP and highest average RSL savings. Rank EPP [years] RSL savings [€] 1 Electric radiator Electric radiator 2 Pellet heat recovery Natural gas boiler 3 Natural gas boiler Min. efficiency heat pump 4 Min. efficiency heat pump Pellet heat recovery 5 High efficiency heat pump High efficiency heat pump Vitor E.M. Cardoso, M. Lurdes Simões, Nuno M.M. Ramos et al. Energy & Buildings 279 (2023) 112667 9