Commercial roof systems represent vast surface areas of building envelopes, and the rate of energy transfer across these roofs significantly influences overall building energy performance. Polyisocyanurate rigid foam has become the predominant insulation choice for low-slope commercial roofing across North America, valued for its higher R-value per thickness compared to alternative insulation materials. However, like all blown insulating foams, polyisocyanurate undergoes a gradual change in thermal resistance over time as its blowing agent diffuses through the cell structure. This phenomenon is known as thermal drift, and understanding it is critical for accurate HVAC load calculations, code compliance, and long-term energy modeling. For builders working on below-grade applications, proper insulation selection for radiant slab assemblies similarly depends on understanding how different foam products perform over time.
Understanding Polyisocyanurate and the Thermal Drift Phenomenon
Polyisocyanurate, commonly called polyiso, is a closed-cell rigid foam insulation that achieves its high thermal performance through the use of low-conductivity blowing agents trapped within its cellular structure. When manufactured, these blowing agents provide significantly better insulation value than still air alone. However, the blowing agent molecules are smaller than the cell walls over time, and they gradually diffuse outward while air diffuses inward to replace them. This gas exchange reduces the overall thermal resistance of the board, a process that is most rapid in the first few years after installation before gradually stabilizing.
The rate and magnitude of thermal drift depend on several factors including the type of blowing agent used, the density of the foam, the quality and type of facer material, and the temperature conditions to which the insulation is exposed. The industry has reformulated blowing agents multiple times in response to environmental regulations. Early formulations used CFC-11, which was replaced by HCFC-141B, and since approximately 2002 most manufacturers have transitioned to hydrocarbon-based agents such as pentane. Each blowing agent formulation has unique thermal properties and diffusion rates, meaning insulation boards produced in different eras may exhibit different aging characteristics. When installing polyiso on walls, attention to proper window integration with exterior rigid foam is essential to avoid thermal bridging and ensure the insulation performs as expected at the critical envelope transitions.
How Long Term Thermal Resistance Is Measured
To address the reality of thermal drift, the concept of long-term thermal resistance was developed in the 1990s through collaborative research led by Oak Ridge National Laboratory in cooperation with the National Roofing Contractors Association, the Polyisocyanurate Insulation Manufacturers Association, and the Society of the Plastics Industry. The resulting standards established a laboratory method for predicting aged R-values using an accelerated aging technique known as the slicing and scaling method. In this procedure, thin slices of foam specimens are cut and tested to accelerate gas diffusion, producing a thermal resistance reduction that simulates years of natural aging in a compressed timeframe.
The LTTR is expressed as a 15-year time-weighted average of the insulation R-value, derived from the first five years of relative aging. This value is what manufacturers publish as the long-term R-value of their polyiso products, and it is the number used for code compliance and design calculations. However, it is important to note that LTTR testing occurs entirely under controlled laboratory conditions at a constant temperature of 75 degrees Fahrenheit. Real roof assemblies experience dramatic temperature swings, freeze-thaw cycles, ultraviolet exposure, and moisture conditions that may accelerate or alter the aging process in ways the laboratory method cannot fully replicate. Understanding thermal barrier compliance requirements for foam plastics is a related consideration, as building codes mandate specific protective barriers between foam insulation and occupied spaces regardless of the insulation aged R-value.
The key elements of LTTR testing include:
- Specimens are conditioned at room temperature for a specified period before testing
- Thin slices of foam accelerate gas diffusion to simulate aging
- Testing is conducted at multiple mean temperatures
- Results are averaged over a 15-year projected service life
- The method assumes constant temperature exposure throughout aging
Previous Research on Polyiso Aging and Performance
Several earlier studies have raised questions about whether laboratory-based LTTR predictions accurately reflect real-world in-service performance. In 2006, the NRCA conducted a limited study involving 20 polyiso specimens that were aged naturally in laboratory conditions for under five years. The results showed that 17 of the 20 tested specimens exhibited R-values below their established LTTR values, suggesting that the accelerated aging methodology may overestimate a product actual R-value at the five-year mark. This finding was significant because it challenged the assumption that laboratory conditions adequately represent field aging.
In 2009, RDH Building Science Inc. of Vancouver conducted a field monitoring study in collaboration with the Roofing Contractors Association of British Columbia. A dedicated building was set up to continuously record the temperature-dependent conductivity of polyiso insulation over five and a half years. The in-service results were compared with samples aged naturally in laboratory conditions at 72 degrees Fahrenheit. The study concluded that in-service aged polyiso consistently measured lower thermal resistance than lab-aged samples, further reinforcing the gap between laboratory predictions and real-world performance. A follow-up study by RDH in 2015 removed three-year-old polyiso specimens from an actual roof assembly in British Columbia and tested them in the laboratory. The aged insulation was found to have up to 25 percent lower R-value at low mean temperatures compared with the same material tested when new. For projects requiring accurate thermal performance data, reviewing thermal insulation fundamentals for buildings provides helpful context on how different insulation materials compare and how aging affects overall envelope performance.
New NRC Canada Field Study Findings
The National Research Council Canada has undertaken the most comprehensive study to date on in-service aging of polyisocyanurate insulation. Researchers collected polyiso samples from 16 low-slope commercial roof assemblies across New Brunswick, Ontario, and Quebec that were undergoing retrofit or replacement. The exposure periods ranged from 13 to 31 years, providing a unique opportunity to evaluate insulation performance at time scales far beyond what laboratory testing can simulate. The buildings represented variable occupancies and were grouped according to ASHRAE climatic zones based on geographic location.
All collected insulation boards had organic felt facers falling into Type II classification under ASTM C1289, with one exception being a Type I board with an aluminum facer. Importantly, the samples were not further conditioned in the laboratory after removal because the objective was to evaluate the insulation in its current in-service state. R-values were determined using a heat flow meter according to ASTM C518 at mean temperatures of 40, 75, and 110 degrees Fahrenheit with a temperature differential of 40 degrees Fahrenheit.
The thermal performance results revealed important trends:
| Sample Age Group | Average R-Value Per Inch | Percentage of Samples Meeting R-5 |
|---|---|---|
| 13-20 years in service | R-4.8 to R-5.5 | Approximately 50% |
| 21-31 years in service | R-4.0 to R-5.2 | Varied by formulation |
| NRCA recommended design value | R-5.0 per inch | For heating climates |
The measured data showed that sample 9, aged for 28 years, recorded the lowest thermal resistance at R-4 per inch, while sample 1 at 21 years recorded the highest at R-5.5 per inch. Because of differences in blowing agent type and manufacturer formulation, no clear relationship could be established between insulation age alone and R-value. When compared against ASTM C1289 minimum requirements for new insulation boards, the in-service aged boards tested 3 to 30 percent lower, which was expected given that the standard tests preconditioned new boards after only 180 days of early aging. However, when compared to the CAN/ULC-S704 minimum LTTR requirement of R-5.2 per inch, five out of ten samples close to the 15-year mark met or were near this threshold. The temperature dependency of the aged insulation was also notable, with most samples showing linear thermal performance where R-values decreased as mean temperature increased. Understanding how buildings manage thermal conditions through thermal mass strategies in passive solar design offers a complementary perspective on how building materials interact with temperature fluctuations over time.
Mechanical Properties of Aged Polyiso Insulation
Beyond thermal performance, the NRC study also evaluated the compressive strength and flexural strength of the aged polyiso samples to determine whether long-term in-service exposure degraded the mechanical properties essential for roof assembly integrity. Compressive strength was tested in accordance with ASTM D1621 using five 2-inch by 2-inch specimens from each sample. The specimens were compressed at a constant rate until reaching 13 percent deformation to determine the load at 10 percent deformation as specified in the standard.
The minimum compressive strength requirement under ASTM C1289 is 16 pounds per square inch for most standard polyiso board types. Every sample tested except one surpassed this minimum requirement, including boards that had been in service for 20 years or more. The overall results indicate that the in-service aging process does not reduce compressive strength below code requirements even after decades of exposure in roof assemblies.
Flexural strength testing was performed according to ASTM C203 Method 1 Procedure B. Specimens measuring 3 inches by 12 inches were tested with load applied at a constant rate of 0.1 inch per minute per inch of thickness. All samples surpassed the minimum breaking load requirement of 17 pound-force, and most met the minimum flexural strength requirement of 40 psi. The oldest specimen tested for flexural strength was 20 years old and still surpassed all current requirements. The mechanical testing results can be summarized as follows:
- Compressive strength exceeded minimums for all but one of the aged samples
- Flexural strength met requirements for all but two samples
- Boards aged 13 to 20 years maintained structural integrity comparable to new material
- Facer type and board density influenced mechanical retention more than age alone
Practical Implications for Roof Design and Installation
The NRC findings have practical significance for roofing contractors, designers, and building owners. The NRCA has recommended using an R-value of 5 per inch in heating-dominated climate zones for roof thermal design, a recommendation that aligns with the study finding that approximately 50 percent of aged samples measured R-5 per inch or more after 13-plus years in service. For the samples nearest to the 15-year LTTR averaging period, the measured values suggest the roofing industry should give serious consideration to the NRCA recommendation. Designers should also consider how wind washing can compromise insulation thermal performance through air movement within the assembly, which can reduce effective R-values independently of the material aging process.
Several samples showed an inverted crown temperature profile with maximum R-value at 75 degrees Fahrenheit and lower R-values at both 40 and 110 degrees Fahrenheit. This profile is characteristic of insulation blown with hydrocarbon and pentane agents that are susceptible to condensation at colder temperatures. Notably, these samples had been in service for 20 years yet had not shown the linear aging characteristics seen in other samples, indicating that blowing agent type may influence both initial performance and long-term aging trajectory. Missing information that would further clarify these results includes the location of the insulation within the roof system whether it was a one-layer or two-layer layout, and whether the sample came from the top or bottom layer in a multi-layer installation.
Future research priorities identified by the study include quantifying the effects of moisture on aging polyiso, composition analysis to measure blowing agent diffusion rates, and investigating the effect of temperature shocks on long-term thermal performance. As the building industry moves toward more rigorous energy modeling and performance-based codes, understanding real-world insulation aging becomes increasingly important for predicting actual versus theoretical energy use in commercial buildings. Even beyond the building envelope, understanding thermal expansion principles in plumbing systems illustrates how thermal effects influence building components throughout the entire structure.
The NRC study confirms that while thermal drift is a real phenomenon that reduces polyisocyanurate R-values over time, the majority of in-service aged insulation boards retain reasonable performance even after decades of exposure. Mechanical properties including compressive and flexural strength remain largely intact well beyond the 15-year LTTR reference period. The key takeaway for practitioners is to use realistic design R-values accounting for in-service aging, maintain proper barrier and covering systems to protect insulation from moisture and mechanical damage, and recognize that insulation performance depends on the specific blowing agent, facer type, and installation conditions of the particular product being specified.