SPF and the Need for Vapor Retarders in Above-Grade Residential Walls

By John Straube, Ph.D., P.Eng.
Download pdf (712 KB)
This short report summarizes research conducted by John Straube, Rachel Smith, and Graham Finch for the Canadian Urethane Foam Contractors Association (CUFCA). The full report is available at buildingscience.com.

The Question:

Theory and experience suggest that closed-cell spray polyurethane foam (ccSPF) is sufficiently vapor impermeable to control diffusion condensation and that low-density open-cell foam (ocSPF) applications may require additional vapor diffusion control in some extreme environments. However, SPF contractors, building designers, and code officials continue to encounter the question of whether an additional vapor barrier or retarder is needed in various wall assemblies.

The Goal:

To provide recommendations, based on sound scientific evidence, of the need for additional vapor control for both classes of SPF installed in framed walls of a wide range of building occupancy types and cold climates.

Materials and Methods:

A combination of full-scale natural exposure field tests, climate chamber measurements, and hygrothermal computer modeling was applied.

Field Tests. Eight test walls were constructed and installed in the University of Waterloo’s BEGHut test facility, maintained at a high (50% RH) interior humidity level. The moisture content of the exterior wood sheathing and wood studs were monitored for a period of over two years and the results used to assess performance. Exterior temperature, relative humidity, and environmental conditions (including rainfall, wind speed, wind direction, and solar radiation) were measured at the roof of the BEGHut.

Figure 1: Close-up of sensors on steel stud wall assembly.

Figure 1: Close-up of sensors on steel stud wall assembly.

Three different SPF products were selected for use in the study. Both distinct classes of SPF commonly used in construction were represented, i.e. high-density (2 pcf) closed-cell and low-density (0.5 pcf) open-cell foam.

All spray foams were installed by a licensed applicator under normal interior conditions. The open-cell foam (Type C) was sprayed to a full stud cavity depth of 140 mm (5.5 in.). Excess foam was removed to allow drywall installation (the surface skin provides little resistance for open-cell foams).

The closed-cell foams (Types A and B) were sprayed to an average depth of 130 mm (5.1 in.) within the 140 mm (5.5 in.) stud bay to allow flush placement of the drywall over the uneven surface of the foam. This method maintained the surface skin integrity of the closed-cell foam.

Climate Chamber Measurement. The field tests used a relatively small sample of SPF products from a single manufacturer. To provide side-by-side performance measurements of a range of different SPF products, a less expensive and more controlled laboratory experiment was undertaken. The scope of the experiment was to test the most common types of open- and closed-cell spray polyurethane foam insulation used in Canadian residential and commercial construction. Fibreglass batt insulation was included in the test as a reference case.

A 2.4 m x 2.4 m (8’x8’) frame was sub-divided into compartments to allow for simultaneous testing of test wall samples. This frame was inserted into a climate chamber. One side of the climate chamber was conditioned to simulate room temperature with a high humidity load and high temperature (25° C [77° F] and 50% RH); the other side was conditioned to simulate cold outdoor conditions of about -10° C (14° F) and 60% RH. The test boxes were subjected to large, steady gradients for temperature and relative humidity over a period of 57 days. The intention, as in the test house study, was to stress the samples and observe performance in demanding conditions. Water accumulation was observed through periodic mass gain measurements and moisture content readings in the exterior oriented strand board (OSB) sheathing of each test box. Twenty test boxes were tested. Test variables were manufacturer (BASF, Dow, Demilec, Icynene), foam type/thickness and presence/absence of a polyethylene vapor barrier.

Figure 2: Typical construction of test boxes used in climate chamber testing.

Figure 2: Typical construction of test boxes used in climate chamber testing.

Hygrothermal Computer Modeling. A WUFI hygrothermal model was developed and validated using data from the field testing described above. The model was then used to extend the results from the experimental program to a wide range of wall types and climates.

Seven different wall assemblies were parametrically modeled in different Canadian climates. The worst-case scenario for cold-weather diffusion wetting was used (a north-facing orientation with light-weight and light-colored claddings).

Figure 3: Wall assemblies used in hygrothermal modeling.

Figure 3: Wall assemblies used in hygrothermal modeling.

Every simulation case was run for seven different Canadian climates. The climates were categorized according to the number of heating degree days below 18° C (64.4° F). Heating degree days (HDD) are calculated by summing the number of degrees each average daily temperature is below 18°C for a full year of historical temperature data. Table 1 shows the categories chosen, with a representative city for each.

Table 1. Climate Categories Used for Hygrothermal Modeling
HDD Climate Category (with range) Representative Location (with HDD)
HDD 3000 (Up to 3500) Vancouver (2926)
HDD 4000 (3501 to 4250) Toronto (4065)
HDD 4500 (4251 to 4750) Ottawa (4602)
HDD 5000 (4751 to 5500) Calgary (5108)
HDD 6000 (5501 to 7000) Winnipeg (5777)
HDD 8000 (7001 to 9000) Yellowknife (8256)
HDD 10,000 (9001+) Inuvik (9767)
 

The temperature for interior conditions in all simulations was set at 22°C with an annual variation of 1°C (1.8° F). Each climate category was modeled with three interior climate conditions – low, medium and high indoor relative humidities. The actual number used for the indoor climate settings depended on the climate category. For example, a low interior RH (30%) in a warm, rainy climate like Vancouver is higher than what would be considered a low interior RH (20%) in a cold, northern climate like Yellowknife.

The modeling period ran for one year from August 1, 2007 to August 1, 2008 in time steps of one hour. The simulations were run until the moisture content in the sheathing was equal to the moisture content in the previous year.

Results:

The measured and modeled moisture content of the OSB and studs in the closed-cell SPF walls were little affected by changes to the interior relative humidity or vapor control layer permeance. Modeling showed that even in climates as cold as Edmonton (about 6500 HDD), interior RH levels of 50% can be accommodated with no additional vapor control layer. The wood studs have sufficient inherent vapor resistance that they do not require a supplemental vapor control layer. The wood studs remained dry both winter and summer without the need for a polyethylene sheet vapor diffusion retarder.

Field measurements showed that the OSB and wood stud moisture contents of the open-cell SPF walls during the winter were significantly impacted by interior relative humidity and interior vapor control layer permeance. Using standard interior latex paint (in the order of 300 ng/Pa•s•m2) and an interior relative humidity of greater than 40% during the winter in a cold climate (over about 4000 HDD) can result in dangerously high moisture contents of the sheathing as a result of vapor diffusion. However, because of the sensitivity of the wall to changes in interior relative humidity, additional vapor control is recommended with open-cell SPF in climates of more than about 4000 HDD. A vapor retarding paint (in the order of 30 ng/Pa•s•m2), smart retarder, or polyethylene sheet are better choices for vapor control in such cold climates.

Field measurements showed, and modeling confirmed, that when SPF is installed inboard of hygroscopic sheathing, moisture accumulation can occur due to solar driven moisture from brick cladding especially if relatively vapor impermeable SPF is used. Basic building physics suggests that if this moisture increase is excessive, it can be controlled by installing the closed cell foam on the exterior of the sheathing to both increase the sheathing temperature and provide resistance to vapor flow.
Climate chamber vapor diffusion tests on a range of different products confirmed the performance noted in the field tests and demonstrated that different commercial products of the same class (closed cell or open cell) performed in a very similar manner.

Conclusions & Recommendations:

The research results are limited to walls with exterior layers of sheathing, membranes, cladding and other layers with a permeance of more than about 60 ng/Pa•s•m2. Within these limitations, it can be concluded that:

  • Closed-cell (about 2 pounds per cubic foot density or more) spray foam applied in thicknesses of over 50 mm (2 in.) will control vapor diffusion to safe levels in all climates up to 10000 HDD and interior winter-time relative humidities of up to about 50%RH. As thickness increases the level of diffusion control increases. The diffusion control is equivalent to walls with the traditional fiberglass batt and polyethylene.
  • Open-cell (0.5 pcf) foam can control diffusion in climates that are not too cold (eg under 4500 HDD) and when the interior winter RH level is controlled by appropriate ventilation to below about 40%. Open-cell foam does not have sufficient vapor control for use in very cold climates (4500 HDD to 5000 HDD) unless the interior winter-time RH is strictly controlled (to below about 30% RH).
  • For either type of foam, the wood framing provides sufficient inherent vapor resistance to maintain the moisture content within the safe range even in very cold exterior climates (10 000 HDD) and very humid interior conditions (50% RH in winter).

As for all walls made of all materials, a functional air barrier assembly must be provided, as well as rain control, fire control, structural sufficiency, etc.

The one-D WUFI Pro hygrothermal modeling program was validated as an effective and accurate tool for predicting the moisture content of the sheathing in the field tests. It can be used to predict the performance of other wall assemblies in other climates if care is taken to define the material properties and boundary conditions.

Learn More:

Send this to a friend