Numerical investigation of the effect of vacuum insulation panels on the thermal bridges of a lightweight drywall envelope

3.1Assessment of building thermal performance

The main concept of the standardized methodology is the partitioning of the building into 1D, 2D and 3D geometrical models, as shown in Figure 2. The 1D geometrical model stands for the centre of the wall assembly, including the steel studs, while the 2D and 3D geometrical models stand for the 2D and 3D junctions of the walls, respectively (e.g. external wall and roof, external wall and inside wall and ceiling etc.), including the steel studs of the junction.

After partitioning the building, the transmission heat transfer coefficient, H_D, is calculated by Equation 1:

3.2Calculation of linear and point thermal transmittance

The linear thermal transmittance, Ψ, concerns the two dimension geometries and is calculated according to Equation 2:

The point thermal transmittance, χ, of a part of the building envelope is calculated according to Equation 3:

The thermal transmittance of the 1D component, U-value, at each component, is calculated according to EN ISO 6946 (EN ISO 6946, 2007) taking into account the physical properties (i.e. thermal conductivity and thickness) of the materials composing the assemblies.

It should be noted that in the above equations (i.e. Equations 1, 2 and 3), the linear and point thermal bridges refer to the steel skeleton and the geometry of the facade, i.e. level 3 according to Quenard (Quenard, 2014), and not to the thermal bridges due to the laminate barrier of the VIP, i.e. level 1 as reported in Brunner, Stahl & Ghazi Wakili (2012a).

3.3Numerical simulation

In the current work, the two and three-dimensional thermal coupling coefficients, L_2D and L_3D, were numerically calculated for all the junctions of the building envelope, covering all the thermal bridges introduced due to the presence of the steel frame. The ANSYS CFX (ANSYS CFX, 2009) commercial CFD package was used to simulate all the configurations and the types of thermal bridges. Regarding the boundary conditions, at the inner side of the walls the ambient temperature and the heat transfer coefficient were assumed to be T_in = 20°C and h_in = 7.69 W/(m²K), respectively, while the respective values at the outside environment were T_out = – 10°C and h_out = 20 W/(m²K). In the case of ground soil, its temperature was considered to be constant and equal to T_soil = – 10°C. Finally, the factor of temperature correction, f, was assumed to be f = 1, apart from the floor assembly and floor’s junctions where it was assumed to be f = 0.6 (Tichelmann & Ohl, 2005).

The thermal properties of the materials were assumed to be constant. Table 1 summarizes the physical properties and the thickness of the utilized materials. The ventilated air cavities were assumed to be stagnant air with an “effective” thermal conductivity calculated according to EN ISO 6946 (EN ISO 6946, 2007) taking into account the convection-radiation phenomena. The equivalent thermal conductivity of the VIP layer has been calculated and found to be equal to 0.007 mW/(m K) taking into account the edge effect according to Brunner et al. (2012b).

The total heat flow, Φ, which passes each configuration, can be calculated by the simulations. Hence, the two and three-dimensional thermal coupling coefficients, L_2D and L_3D, can be calculated via Equation 4:

4.1Central part of the walls

The impact of the metal studs within the central part of the wall elements at the thermal performance of the building envelope is depicted in Figure 3. Figure 3a illustrates the temperature contours of the central part of the external wall for every type of utilized studs, as well as the linear thermal transmittance of each stud, in the cases with and without VIP layer. As it is shown, the presence of the metal studs introduces significant thermal bridges, which have to be taken into account in the overall thermal assessment of the building. Moreover, it can be seen that the thermal transmittance of the I-studs is ca. 35–40% higher than the respective of C-type stud (in both cases without and with VIPs). This is due to the fact that the I-stud is a double C-stud, which means that there is more metal in the same area; increasing the heat losses. Additionally, the impact of the CW-type is not significant compared to the other types of the studs. Finally, considering the impact of the VIPs on the linear thermal transmittance it can be seen that it is reduced by ca. 64% to 83% for all the stud types.

Figure 3b shows the effect of the metal studs, indicated by the U-value for each wall element (i.e. external wall, dry roof and floor) of the building. As it is shown, the presence of the metal studs affects significantly the thermal performance of the wall elements by increasing the U-value. More specifically, the increase of the thermal transmittance of the external wall is ca. 50% for the external wall without the VIP and ca. 27% for the case with the VIP. Concerning the roof and the floor, the inclusion of the metal studs increases the U-value by ca. 169% and 210% respectively.

4.2Two-dimensional junctions

Figure 4a illustrates the temperature contours of an indicative 2D junction, i.e. junction between two external walls for the case without and with the VIPs. It is clearly shown that the metal studs’ junction introduces thermal bridges in both cases defined by the two dimensional temperature field. Comparing the two cases shows that the presence of the VIPs reduces the heat flow through the junction.

The linear thermal transmittance of all the 2D junctions of the building elements examined in this work is presented in Figures 4b and c. Focusing on Figure 4b, it is obvious that the VIP layer improves the impact of the 2D junctions at the thermal performance of the building, decreasing the Ψ-value by 12% up to 92%. It is observed that the linear thermal transmittance has negative values at the junction between external walls (EW+EW), meaning that this thermal bridge has a positive effect on the total thermal performance of the building envelope. Figure 4c presents the 2D intersections that are not affected by the additional insulation. Overall, it can be seen that the most important thermal bridges are the junctions that include the internal walls (i.e. EW+IW, IW+RF and IW+FL), as well as the junction between the external wall and the dry ceiling. Thus, special design modification should be taken into account in order to further reduce the heat losses through these thermal bridges.

4.3Three-dimensional junctions

Figure 5a illustrates the temperature contours of an indicative 3D junction, i.e. between the two external walls and the roof for the cases without and with the VIPs. Once again, the metal studs of the junction introduce significant thermal bridges in both cases defined by the three-dimensional temperature field.

Figure 5b shows the point thermal transmittance for all the 3D junctions of the building examined in this work. It is clearly shown that the presence of the VIPs generally improves the thermal performance of all the 3D thermal bridges. The point thermal transmittances are decreased up to 138% due to the installation of VIP layer at the external walls.

4.4Overall thermal transmittance

The contribution of the thermal bridges and the impact of the VIP layer on the overall thermal performance of the building envelope is depicted in Figure 6. The installed VIPs decrease the total thermal transmission, H_D, by ca. 33%. The contribution of the metal studs of the central part of the wall configurations is ca. 30% of the total transmittance in both cases. Although the additional insulation reduces the linear thermal transmittance of all the metal studs at the centre part of the walls, the percentage impact of them remains the same because of the high Ψ-values of the studs at the roof and floor, where extra insulation is not installed. The total impact of the 2D and 3D thermal bridges is ca. 31% in the case without the VIPs, while in the case with the VIPs it is ca. 25%. The 2D and the 3D junctions contribute the same percentage in the case without VIP, showing that three-dimensional intersections are as important as two-dimensional intersections, and thus they should be taken into account in relevant calculations. In the case with VIP, the transmission of the point thermal bridges is 4 times lower than that of the linear thermal bridges.

5.1Roof

Regarding the roof, two potential design solutions were assumed: the gap of the suspended ceiling was increased in order to increase the thickness of the mineral wool and an XPS insulation layer (80 mm thickness) was positioned on the top of the roof. Figure 7a and b illustrate schematic diagrams of the design solutions described above. Figure 7c and d show the calculated U-value of the dry roof assembly, considering the different types of insulation, as well as the calculated linear thermal transmittance (Ψ-value) of the I-stud inside the dry roof assembly. It should be mentioned that the case “RC” means “Resilient Channel” and the number next to it defines its thickness, while the case “XPS 80 mm” defines the XPS insulation layer (80 mm thickness) positioned on the top of the roof. As it is expected, when the thickness of the suspended ceiling is increased the U-value of the dry roof and the linear thermal transmittance of the I-studs decreases. The 200 mm thickness of the resilient channel achieves ca. 67% and 81% reduction of the U and Ψ values, respectively. Nevertheless, increasing the thickness of the resilient channel means that the “free space” of the room or else the room’s height decreases. On the other hand, the installation of an XPS layer of 80 mm thickness on the top of the roof, which is also an easier design solution from a construction point of view, reduces the U and Ψ values ca. 50% and 64%, respectively. This reduction is equivalent to the reduction that occurs if the thickness of the resilient channel is 100 mm.

5.2Floor

Regarding the floor, the potential design solution that was assumed was the installation of an EPS insulation layer of two different thicknesses, 100 mm and 200 mm, respectively, between the cement slab and the ground soil, as shown in Figure 8a. Figure 8b shows the calculated U-value of the floor assembly, considering the different types of insulation (i.e. EPS of different thickness), as well as the calculated linear thermal transmittance of the I-stud inside the floor assembly. As it is expected, when the thickness of the EPS insulation layer between the foundation slab and the ground soil is increased, the U-value of the floor and the linear thermal transmittance of the I-studs decreases. Moreover, the installation of the 100 mm thick EPS results in a ca. 60% reduction of the U-value. Additionally, a further increase of the EPS’s thickness from 100 mm to 200 mm results in ca. 15% more improvement. Considering the ratio between the cost of the EPS and the U-value reduction percentage for each thickness, the solution of the 100 mm thick EPS is a reasonable solution.

5.3External wall-ceiling junction

In order to improve the thermal performance of the external wall-ceiling junction, a potential solution would be the installation of a Super Insulation Material (SIM) inside the ventilated part of the external wall near the junction, as shown in Figure 9a. In this study, two different SIMs (aerogel and VIP) of two different thicknesses (10 mm and 15 mm) and five different lengths (308 mm, 338.8 mm, 369.6 mm, 462 mm and 616 mm) were assumed. The 308 mm length corresponds to the minimum length, inside the ventilation part of the external wall, which covers the distance between the metal studs of the junctions (see Figure 9a).

Figure 9b shows the linear thermal transmittance for the external wall-ceiling junction for all the examined cases. It is observed that the two examined SIMs, i.e. VIP and aerogel, have significant differences as far as their thermal performance is concerned. The VIP is observed to perform better than the aerogel due to the significant difference in the thermal conductivity values. When the length of the SIM increases, the effect of the thermal bridge also decreases. The length of 616 mm is almost the maximum length that could reach the maximum thermal bridge reduction. Moreover, when increasing the thickness of the SIM, the effect of the thermal bridge of the junction decreases. The reduction of the thermal bridges by incorporating the proposed solutions ranges between ca. 18% and 61%. It is clearly shown that the aerogel cannot reach the insulation capability of the VIP. The bigger aerogel sample examined (i.e. 15 mm thickness and 616 mm length) reduces the thermal bridge by ca. 32.5%. On the other hand, the smaller VIP sample examined (10 mm thickness and 308 mm length) reduces the thermal bridge by ca. 29%, which is very close to the reduction percentage obtained by the bigger aerogel sample. Another significant issue that must be taken into consideration is the thickness of the SIM due to regulation restriction for the thickness of the ventilation facade. As it is shown, by increasing the length of the SIM, the thickness of the SIM could be compensated with respect to the reduction of the thermal bridge effect. Thus, the 10 mm thick and 616 mm length VIP has an almost equivalent insulation capability as the 15 mm thick and 462 mm length VIP.

5.4External wall-internal wall junction

The potential design solution that was assumed for the external wall-internal wall junction was the total separation of the two walls, as shown in Figure 10a. Figure 10b shows the linear thermal transmittance of the junction for the reference and the new design solution. As it is shown, the linear thermal transmittance for the reference design is ca. Ψ= 0.153 W/m K, while the respective value for the new design solution is ca. Ψ= 0.002 W/m K, which corresponds to ca. 98% reduction of the specific thermal bridge. The respective values for the heat transmission through the junction were 0.39 W/(m K) for the reference design solution and 0.24 W/(m K) for the new design solution. The latter indicates that the overall reduction of the thermal losses of this junction was ca. 38.5%.

5.5Overall thermal transmittance

As mentioned above, different design modifications were assumed in order to further reduce the increased thermal bridges of the building. Table 2 summarizes the final solutions that were assumed to quantify the overall thermal transmittance of the building taking into account the applicability (ease of construction), as well as the cost effectiveness of the solution. Figure 11 shows the transmission heat transfer coefficient of the building for the different design solutions. As it is shown, each solution contributes positively to the reduction of the thermal losses compared to the reference case (i.e. building with VIPs installed at the external walls). More specifically, the larger improvements are obtained due to the design modifications in the roof (ca. 11.5% thermal loss reduction) and the floor (ca. 8.9% thermal loss reduction). This is due to the fact that the two wall assemblies were not insulated in the reference case. Moreover, the local design modifications, i.e. installation of a VIP strip inside the ventilation facade and the decoupling of the external wall-internal wall junction, have a smaller, but still significant impact on the reduction of the overall thermal losses. The most important conclusion of the analysis is that when all these design solutions are implemented the overall thermal loss reduction reaches ca. 27.5%. It should be mentioned that this reduction is achieved on the building with the VIPs installed inside the external walls. Compared to the initial building (without the VIPs at the external wall) the overall reduction of the thermal losses reaches ca. 60.5%.