Comparing Eight Cold-Climate PH Envelopes
In 2010, Rolf Jacobson received a Fulbright scholarship to study in Norway for a year and complete his Master’s thesis. This was an appropriate destination for his research, considering that he was evaluating Passive House (PH) building envelopes for very cold climates (climate zone 6). In particular he wanted to fully explore to what extent PH envelopes exhibit life cycle energy and carbon savings compared to conventionally built homes. Much of Norway’s climate is a close match with the climate that inspired Jacobson’s research, housed as he was then at the Center for Sustainable Building Research (CSBR) at the University of Minnesota in Minneapolis.
When Jacobson started his research, the PH approach was still relatively new in the United States, so he chose his five U.S. building envelope case studies based on what was available in a cold climate. He studied (1) a TJI frame with blown-in fiberglass insulation, built in Urbana, Illinois; (2) an insulating concrete form (ICF) with exterior expanded polystyrene (EPS), built in southern Wisconsin; (3) a structural insulated panel (SIP) filled with urethane foam with an interior 2 x 4 wall filled with blown-in cellulose, built in Belfast, Maine; (4) an advanced 2 x 12 stud framing filled with open-cell spray foam and insulated on the exterior with either EPS or vacuum insulated panels (VIPs), built in Bemidji, Minnesota; and (5) a double 2 x 4 stud wall insulated with blown-in cellulose, built in Duluth, Minnesota.
R‐Value Drop Due to Wall Framing
Global Warming Potential Per Square Foot Wall Area
Embodied Energy Per Square Foot Wall Area
Life Cycle Energy
Life Cycle Carbon (GWP)
Table 1. Thermal Bridge Heat Losses
Jacobson expanded his case studies to include three types of typical building envelope from Scandinavia. These were (1) an advanced frame with cross-strapping and mineral wool insulation; (2) a mass wall that includes 6 inches of concrete and mineral wool insulation; and (3) a structural panel, known as Massivtre, that is 4 inches thick with a core made from low-grade wood and an interior surface made from higher-grade wood that provides a nice wood wall finish. The Massivtre wall is insulated with exterior mineral wool. Massivtre is not available in the United States, but four demonstration homes in St. Paul, Minnesota, were built with a similar product: a structural engineered panel (SEP), which is a panelized studless wall made of 1-1/2-inch-thick OSB (oriented strand board). For insulation on this house, rigid foam was added to the exterior of the SEP. In his analyses, Jacobson lumped together the Massivtre and SEP constructions.
Each of the eight wall systems, which were of varying thicknesses, achieved an R-value of approximately 60. The floor systems were also insulated to R-60, while the roofs were R-80. While Jacobson knew that these superinsulated homes would save heating and cooling energy, the main question that he wanted to address was whether the embodied energy and carbon would neutralize those savings. He also wanted to model how these envelopes would perform in general, not so much (in his words) because he wanted to pick a winner, but because he wanted to compare their relative strengths and weaknesses.
As a first step, Jacobson modeled each of the eight wall systems using THERM, a two-dimensional heat transfer simulation program developed by Lawrence Berkeley National Laboratory researchers, to assess both perpendicular and lateral heat flow in the walls, in order to account for the extra heat loss through the envelopes due to thermal bridging in the assemblies. This heat loss occurs through, for example, studs, which provide lower-resistance pathways for heat flows compared to a solid section of insulation. Thermal bridges in an assembly diminish the actual, delivered R-value. Heat loss calculated using the typical method of multiplying an assembly’s total U-value by the wall area rarely accounts accurately for thermal bridges, but heat loss through these bridges can be calculated using THERM. Not too surprisingly, the walls with the fewest framing members—the ICF wall, the mass wall, and the Massivtre wall—provided the fewest pathways for heat to travel out of the assembly (see Figure 1).
Jacobson went on to calculate the thermal bridge heat losses, or psi values, of ten common thermal-bridge locations in the eight envelopes (see Table 1). Generally, details with psi values below 0.01 watts per meter kelvin (W/mK) lose so little heat that they are considered thermal bridge free and can be ignored in heat loss calculations, while details with psi values above 0.01 W/mK should be included as an additional heat loss. In these cases, a larger negative number indicates reduced heat loss. As long as these details were thoughtfully designed to reduce or eliminate thermal bridges, Jacobson found little difference between envelope types. He found, rather, that specific locations were more problematic for all types. These locations included rim joists, parapets, and footings at or below grade. All the details Jacobson modeled for each envelope type are described in Performance of 8 Cold-Climate Envelopes for Passive Houses, which Jacobson wrote based on his thesis research (see “learn more”). Jacobson also wanted to investigate how these walls performed from a moisture and heat, or hygrothermal, perspective. For stud walls with wood sheathing on the exterior, increasing insulation thickness can increase the risk of mold growth—unless the envelope is also airtight. For these PH envelopes, would there be moisture problems down the road? To fully answer that question, Jacobson conducted detailed hygrothermal analyses of all the envelopes, using WUFI, a moisture simulation tool developed by Oak Ridge National Laboratory.
These WUFI results are discussed at length in Jacobson’s book. Here are some key conclusions that he drew from his analyses.
Installing several inches of continuous exterior insulation over wood sheathing is an effective way to provide a thermal break, minimize thermal bridge losses, and keep the critical moisture-sensitive sheathing layer warm. But installing exterior insulation can be tricky. First, you have to know the vapor permeance of the insulation and of the other materials in your envelope before you install them. Using exterior insulation that is vapor permeable, such as mineral wool, is key to reducing mold growth risks in cold climates. Some foams are more vapor permeable than others; polyiso, for example, is relatively permeable, but extruded polystyrene (XPS) is not. The ratio of exterior to interior R-value is also important. A very thick layer of exterior insulation can allow you to get away with using a vapor-closed insulation, but again the exterior-interior ratio is critical, and Jacobson recommends conducting a WUFI analysis before you build. Another approach is to eliminate altogether the critical moisture-sensitive layer with an assembly, such as an ICF, that does not support mold growth and is relatively impervious to moisture.
Second, the colder the climate, the more important a warm-side vapor retarder becomes, although advice about vapor retarders has shifted in the last five years or so. More experts now recommend using vapor retarders that can change their permeability as a safer approach than using a sheet of polyethylene. With a “smart” vapor retarder, if leaks should develop around the windows, for example, the walls can dry to the interior as well as the exterior when conditions permit. Finally, hit the PH air tightness target (0.6 ACH50) to greatly reduce the chance of moisture traveling into your envelope.
To analyze the life cycle environmental impacts of the eight assemblies, Jacobson used the Athena Environmental Impact Estimator, both because he was familiar with this tool, as the CSBR contributed to its initial development, and because the database of materials is periodically updated. Athena assesses life cycle impacts from raw-material extraction or mining, transportation, processing, product fabrication, distribution, maintenance, and all the way to disposal. Jacobson found that the double-stud wall filled with cellulose insulation came in with the lowest impact, largely because cellulose is a fully recycled product that doesn’t require much energy to produce (see Figure 2). In contrast, mineral wool has a fairly high environmental impact, because the industrial process used to melt the rock and steel slag and spin the molten mixture into fibers is very energy intensive. Because the industrial process to manufacture OSB is also very energy intensive—wood chips are glued together with resins at high temperatures and pressures—OSB also has a high environmental impact—in some cases higher than that of concrete per inch of thickness.
Next, Jacobson examined the global warming potential (GWP) of the different materials in the eight assemblies. He found that concrete, EPS, brick, and mineral wool, as used in the concrete mass wall, the ICF wall, and the three Scandinavian walls, all have a relatively high GWP compared to, for example, cellulose or wood, as used in most of the assemblies. However, their GWP is tiny compared to that of XPS and closed-cell spray polyurethane foam, as used in the advanced frame with SPF. The GWP of this foam is almost 100 times greater than that of fiberglass per unit area per R-value (see Figure 3). The drastically high GWP of these foam products is due primarily to their hydrofluorocarbon (HFC) blowing agents. Spray foam that didn’t rely on HFC blowing agents would have a much lower GWP. Honeywell has recently developed an environmentally friendly blowing agent that has a GWP of less than 1, which can be used to blow polyurethane foams. As far as Jacobson knows, though, not many installers have experience with this blowing agent at present. For a lower-GWP insulation to use below grade, foam glass can replace XPS. High-density EPS may also be used, depending on moisture conditions.
Given that there is generally more material, and particularly more insulation, in PH envelopes than in conventionally built walls, it is not surprising that there is generally more embodied energy and carbon in PH walls. This greater embodied energy is typically offset by savings in annual energy use to heat and cool the structure. What Jacobson wanted to know was how long that would take. In other words, when would the additional embodied energy of the beefed-up wall assemblies be paid back by these houses’ annual energy savings? As shown in Figure 4, the annual energy use of the conventionally built house (base case standard frame) quickly causes its total embodied and operating energy to outpace that of all the other assemblies. The energy payback time for the other wall assemblies ranged from immediately for the double-stud wall to 4.4 years for the mass wall—not a big chunk of a building’s expected lifetime. When the assemblies are compared on the basis of total carbon emissions, which includes both embodied carbon and carbon emissions from operating energy use calculated using Minnesota’s electricity generation mix, the picture changes slightly (see Figure 5). Because of the HFC blowing agent, the advanced frame with spray foam envelope has a carbon payback of 23 years, while the double-stud wall still has an immediate payback.
Although the double-stud wall comes out smelling of roses in these comparisons, Jacobson points out that designing a wall based on environmental impact alone isn’t always the wisest choice. As long as you avoid specifying insulation made with an HFC blowing agent and minimize the use of energy-intensive materials, such as concrete and OSB, all of these envelopes would have a good energy and carbon payback.
As for double-stud walls, Jacobson says that although they may have the lowest environmental impact, they would not be his first choice as a wall assembly, because they are subject to moisture problems. Conventional walls rely on escaping heat to drive off excess moisture, but in a thick double-stud wall there is so much insulation that very little heat is available to help dry out the wall. The exterior sheathing remains very cold, and moisture can condense on the inside of this sheathing, leading to mold growth and reducing durability. The way to address this is to add exterior insulation to warm up the sheathing, but then you are losing the cost advantages and relative simplicity of the true double-stud wall.
Jacobson, Rolf. Performance of 8 Cold-Climate Envelopes for Passive Houses. Munich: GRIN Verlag, 2013 is available as a paperback through these outlets:
The e-book version is $10 cheaper and is available directly through GRIN.
In the end, each envelope has its own strengths and weaknesses, says Jacobson, and you have to match those to your project, your budget, and the experience of your contractor. For builders who want to achieve really good performance but aren’t comfortable experimenting with too new an approach, the Scandinavian stud wall with the interior cross-strapping is a good option. It’s easy, and the cross-strapping can protect the air barrier and vapor retarder membrane from puncture. Then vapor-permeable insulation is added to the exterior side. If the builder uses 2 x 8 studs that are 24 inches on center, the wall will be cost-effective and will deliver good R-value performance, while minimizing the chance of moisture problems.
Jacobson’s analyses, although they were conducted for homes built in climate zone 6, can be extrapolated to apply to homes built in slightly warmer climates, such as those in climate zone 5. In climates warmer than that, the hygrothermal issues change. The recommendation for an interior vapor retarder for cold climates shifts to an exterior vapor retarder in warmer climates, although which climates, exactly, remains debatable. Still, Jacobson’s analyses will extend to all climates, although the required amounts of insulation will be different in warmer climates.
Enter your comments in the box below:
(Please note that all comments are subject to review prior to posting.)