Vacuum insulation panels (VIP) are one of the most promising advanced building insulation materials owing to their low thermal conductivity of about 0.004 W/m·K (Simmler et al., 2005). Traditional thermal insulation materials such as mineral wool, expanded polystyrene, extruded polystyrene, cellulose etc. are in the range of 0.025–0.040 W/m·K , have thermal conductivities 5–10 times higher than that of VIPs (Jelle, 2011). Integration of VIPs into building envelopes can help builders meet increasingly stringent code requirements, particularly when space is limited, and allow for greater aesthetic flexibility in architectural design. In spite of these benefits, VIPs have not been broadly adopted because of their high cost, concerns related to long-term effectiveness, thermal bridge effects at the panel edges, and the risk of condensation if appropriate moisture control is not considered in the design (Kalnaes and Jelle, 2014). It is worth mentioning that integration and adoption of new materials and technologies in the building construction industry takes time and cannot be done prior to rigorous short- and long-term holistic performance evaluation. This is particularly true in Canada¹ or insulation because of the significant economic implications that can be expected should the insulation not perform as expected during Canada's long and very cold winters. In the case of VIPs, there are also related concerns about the effects of damage to VIPs during construction and in service, as well as the unknown service life for VIPs.

VIPs consist of an evacuated open-pore core enclosed within a gas-tight envelope, as shown in Figure 1. The core provides thermal resistance and adequate load bearing capacity to resist atmospheric pressure (Heinemann, 2017) while the single/multi-layer sealed envelope maintains the vacuum inside the panel. Opacifiers can be added to the core to reduce radiative heat transfer through the core, and getters/desiccants adsorb residual or permeating gas and moisture (Simmler et al., 2005) and thus slow the aging rate. In these ways, heat transfer across the panel is minimized and the service life is extended.

To maintain high thermal performance, VIPs require their internal pressure to be kept very low, close to vacuum (zero). Permeation of dry air (e.g., oxygen and nitrogen) and water vapor through the envelope can compromise this vacuum and lead to an irreversible pressure increase and moisture accumulation inside the panel (Mukhopadhyaya, 2006). Temperature and relative humidity (RH) are the dominant factors impacting the rate of gas permeation through the envelope and consequently the VIP's expected service life.

The most commonly used VIP core materials are fumed silica and glass fiber. Fumed silica core VIPs can achieve thermal conductivities of 0.004 W/m·K when the pore pressure is reduced to 2,000–10,000 Pa at room temperature. Fumed silica has a pore size of about 30–100 nm and its thermal characteristics are stable up to about 5,000 Pa (Simmler et al., 2005). Among core materials, fumed silica is least sensitive to increases in pressure, and its internal pressure can be satisfactorily maintained with a moderately effective gas-tight envelope. Fumed silica VIPs with laminated aluminum foil envelopes are expected to have service lives of up to 50 years (Simmler et al., 2005; Fricke et al., 2006). Furthermore, even after complete vacuum failure, fumed silica core VIPs can have thermal resistances twice as high (thermal conductivity = 0.020 W/m·K) as conventional insulation materials (Simmler et al., 2005).

In comparison, glass fiber VIPs are able to achieve thermal conductivities as low as 0.002 W/m·K, the lowest of all core materials (MacLean and Korn, 2011) although they require an internal pressure of about 10 Pa to achieve this performance. In addition, the larger pore size (1–12 μm) of glass fiber cores leads to a rapid increase in thermal conductivity when the internal pressure rises above 100 Pa (Fricke et al., 2006). Nevertheless, glass fiber core VIPs are less expensive to manufacture and therefore remain an attractive alternative as a building material (Simmler et al., 2005). Glass fiber VIPs have been used extensively in the aerospace and refrigerator industries where the expected service life is no more than 15 years (Simmler et al., 2005). While the expected service life of thermal insulation in building construction is 25–50 years, broad industry confidence in long-term thermal performance of glass fiber core VIPs still needs to be established.

Glass fiber core VIPs have been used as thermal insulation in retrofitting an existing wall of a commercial building in Whitehorse, Yukon, a subarctic city in Northern Canada. Buildings in Whitehorse typically have high space-heating demands due to the region's extremely cold winters, so interest in high performing building insulation systems is high. This paper is a continuation of a series of publications (MacLean and Korn, 2011; Mukhopadhyaya et al., 2011, 2013, 2014, 2017; MacLean et al., 2017) and presents the most up-to-date field performance data observed from continuous in situ monitoring of this installation (2011–2018), compares these results to accelerated aging test results presented in the literature, and discusses the lessons learned from VIP handling and installation.

An 8.4 × 3.7 m² wall of an existing building was retrofitted with VIPs sandwiched between layers of extruded polystyrene foam (XPS) attaching to its concrete exterior. The VIP specimens used were 560 × 460 × 12 mm in size. The pre-installation thermal conductivity of the VIPs was measured using a 600 × 600 mm heat flow meter apparatus with 300 × 300 mm heat flow sensor and an accuracy of ± 2%, the thermal conductivity was found to be 0.0034 W/m·K at the center of the panel (Mukhopadhyaya et al., 2014).

The existing exterior concrete block wall (Figure 2) was insulated with fiberglass batts. The thermal resistance of this existing wall was about 3.5 m².K/W, and the retrofitting goal was to increase the wall's thermal resistance to 8.8–10.6 m².K/W (Mukhopadhyaya et al., 2014). The integrity of VIPs was verified by physical examination at the construction site before the assembly. Concerns regarding condensation in additional wall layers and mechanical rubbing on VIP surfaces caused by uneven proximate wall elements were addressed in the design described below (Mukhopadhyaya et al., 2014). In summary, additional layers of insulation were constructed on the exterior of the existing concrete block wall in the following order (Mukhopadhyaya et al., 2014):

Installations proceeded quickly and smoothly as documented in Figures 2–5.

Four thermistors were installed to monitor the temperature at the interface between insulation layers, as shown schematically in Figure 6. These thermistors were set up ~2 m from the south edge of the wall and 1 m down from the top of the wall (Mukhopadhyaya et al., 2014). Three sensors were placed on the existing wall ~2 m from the north edge of the wall retrofitted area. These sensors were located ~1, 1.5, and 3 m from the top of the wall, and they were used to monitor the thermal gradients from the top to the bottom of the wall (Mukhopadhyaya et al., 2014).

Infrared thermal images of the retrofitted wall were taken in 2011, 2014, 2016, and 2018 are shown in Figure 7. These images confirm that all VIPs installed in 2011 remain functional after 8 years. Specifically, it is clear from the temperature of the wall over each VIP that none of the VIPs have lost their insulating value over 8 years (failed VIPs will be indicated by abrupt change of color in the infrared thermal images). This is a very significant finding for the long-term performance of glass fiber core VIPs in an extreme cold climate and demonstrates the effective integration of VIPs in an exterior building envelope retrofit. In order to gain greater insight into the thermal performance of the exterior wall and VIPs, the temperatures at the interfaces between insulation layers of the retrofitted wall were collected over the past 8 years (2011–2018) since the project started, and selected data were analyzed for the following colder winter time periods (summer data were not stable due to factors such as lack of thermal gradient, unstable indoor condition etc.):

The temperature drops across each layer of the composite exterior insulation (XPS-VIP-XPS) were calculated as a percentage of total temperature drop across the composite exterior insulation. A sample calculation can be found in Appendix A. Figure 8 shows the percentage of total temperature drop across each insulation layer averaged over the winter and/or spring months for six selected years. The average temperature drop for the VIP layer is about 66.49%, while that for the first (interior) and second (exterior) layers of XPS are 14.48 and 18.95%, respectively. The exterior layer of XPS is a more effective thermal insulator than the identical interior XPS layer due to the cooler exterior temperature. This observation clearly validates a very basic physical principle of heat transfer and generates confidence in the recorded data and the analysis procedure. The graphical representation in Figure 8 shows that the effectiveness of the VIP layer is decreasing very slowly, from about 69% initially in winter 2011 to 63% in spring 2018, at an average rate of 0.9% per annum, and this rate of aging appears to be linear. However, it is to be noted that the aging rate of VIPs cannot be linear forever and it is expected to increase rapidly as the internal pressure approaches atmospheric pressure. The transition from lower aging rate to higher aging rate depends on the functional relationship between thermal conductivity and internal pore pressure of the VIP core material.

Accelerated aging tests were conducted, at the National Research Council (NRC) Canada, in 2011 to establish long-term thermal performance of the glass fiber core VIPs used in this study (Mukhopadhyaya et al., 2014). Three VIP specimens (560 × 460 × 12 mm) were tested under cyclic exposure conditions. In one cycle, VIPs were kept at 23°C, 95% RH for 1 week and then changed to 70°C, 5% RH for the next week. The specimens went through 14 weeks of total exposure (i.e., a total of seven cycles). The thermal resistance of the panels was measured at the end of each week using a 600 × 600 mm heat flow meter. One specimen failed after 8 weeks, and the average linear aging rate of the other two was found to be about 3.29% (Morlidge, 2012) per cycle as shown in Figure 9. The normalized in situ VIP aging rate (expressed as the percentage of its initial thermal resistance) was also plotted with the average of accelerated aging test results in Figure 9. It can be observed from the plot, that the 2-year in situ aging rate is nearly equal to one cycle (two weeks) of the accelerated aging rate for the time period where field data are available. Based on this correlation, the linear extrapolation of the in situ data predicts that the total reduction of thermal resistance in 2026 (14 years since construction) will be about 21% (thermal conductivity increased from 0.0034 to 0.00429 W/m·K).

The small divergence between the accelerated aging test and the actual field performance of VIPs could be attributed to the different environmental (boundary) conditions that the VIPs were exposed to. Accelerated aging tests were conducted with the exposure in cyclic conditions of extreme high temperature and relative humidity (23°C, 95% and 70°C, 5%), while the in situ performance study was carried out in the subarctic climate of Whitehorse, Yukon with lower temperatures and high RH [an average of −13°C and 75% RH over winter months and an average of 11.5°C and 62% RH over summer months]². Since the permeability of gas through the VIPs envelope increases with temperature, the VIP aging rate is slower when subjected to lower temperatures. It is also to be noted that the XPS foam on both sides of the VIPs resisted water vapor movement and absorbed some amount of water vapor, thus buffering the VIP's exposure to moisture. Overall, the linear aging trends from in situ measurements and laboratory accelerated aging test results are quite similar. Additional field and accelerated aging test observations are required to elaborate on the relationship between VIP type, environment and comparison to laboratory tests and to establish more general rules or protocols for the prediction of long-term performance of glass fiber core VIPs.

The typical relationships between the thermal conductivity and internal pore pressure for glass fiber and fumed silica core VIPs are shown in Figure 10 (Fricke et al., 2006). These relationships clearly indicate that the rate of increasing thermal conductivity with pore pressure (i.e., the aging rate) of VIPs is not linear during their service life. Derived from Figure 10, 11 shows the aging of glass fiber and fumed silica core VIPs in terms of their initial thermal resistance as a function of internal gas pressure.

Figure 11 indicates that glass fiber core VIPs lose about 20% of their initial thermal resistance when the internal pore pressure reaches about 10 Pa and that this change of thermal resistance is almost a linear function of time. However, Figure 10 also shows that the thermal conductivity of glass fiber core VIPs will rise at a much faster rate between an internal pressure of 10 Pa and 100 Pa, but still thermal conductivity stays below 0.007 W/m·K at 100 Pa internal pore pressure. These observations further reinforce the relationship between accelerated aging and in situ performance of glass fiber core VIP presented in the previous section. Though in situ decrease in the thermal resistance is found to be slow and vary linearly with time until now. It is expected that the thermal resistance of glass fiber core VIPs will eventually decrease at a much faster rate. Hence, the in situ performance observations in the coming years will be of particular interest and will help us to establish the true relationship between accelerated aging test results and the long-term in situ performance of glass fiber core VIPs.

The results from the energy retrofit project in the Canadian subarctic climate in Whitehorse, Yukon, using a foam-VIP-foam sandwich (i.e., XPS-VIP-XPS) insulation have demonstrated the promising and satisfactory performance of glass fiber core VIPs over a period of 8 years. Analysis of the temperature data from monitoring each insulation layer over a period of 8 years (2011–2018) has shown a <0.9% per year change in the field thermal performance of the VIPs, which indicates that aging of glass fiber core VIPs in cold climate proceeds at a very slow rate. Moreover, this project has shown that many perceived challenges (i.e., handling, installation, etc.) relating to the application of VIPs in the construction industry could be addressed through careful planning and detailing. Ongoing in situ temperature monitoring of this project provides valuable field data that can be used for validating theoretical VIP aging predictions and laboratory accelerated aging results.

All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.