A different look at polyiso

• Photo courtesy of CRS Inc., Monroe, N.C.
• Photo courtesy of CRS Inc., Monroe, N.C.
• Photo courtesy of CRS Inc., Monroe, N.C.
• Photo courtesy of CRS Inc., Monroe, N.C.
• Photo courtesy of CRS Inc., Monroe, N.C.
• Photo courtesy of CRS Inc., Monroe, N.C.

The roofing industry has been monitoring changes in polyisocyanurate foam technology since the federal government mandated changes in blowing agents used in the manufacture of polyurethane and polyisocyanurate foam materials. The change to pentane blowing agents has raised questions about insulation values, fire resistance and dimensional stability of the foam in roof insulation. At the same time, polyisocyanurate insulation manufacturers have changed the facers on their roof insulation products. Almost by accident, my employees and I discovered facers may contribute to decreased fire resistance and increased dimensional instability of roof insulation boards.

Current facers

A recent project of ours brought the subject of roof insulation board facers into perspective. The roof assembly consisted of an intermediate rib steel roof deck, one layer of mechanically attached 25-pounds-per-square-inch (172-kPa) polyisocyanurate insulation, a polymer-modified asphalt base ply set in adhesive with heat-welded side laps and end laps, and a mineral granule-surfaced polymer-modified asphalt cap sheet set in adhesive with heat-welded laps. Roof insulation boards blown with both HCFC-141b and pentane were used on the project.

As the laps of the base ply were being heat-welded with a torch, a fire broke out under the base ply along a longitudinal insulation board joint. When the base ply was cut open, the facers on the bottom side of the insulation board were burning. The fire was extinguished, and the condition was thought to be an aberration. On the following day, fire broke out again under the base ply—in a similar location as the first fire at longitudinal insulation board joints coincident to a side lap in the base ply. Now, the condition no longer was isolated, and we notified the polyisocyanurate insulation manufacturer about the problem.

Following a field audit by the manufacturer's technical representative, we received a written response to the problem. The response said, in part: "The facer of the [insulation board] is glass-reinforced recycled paper and will burn when exposed long enough to an open flame. Be advised that in our precautions section of the [manufacturer's] technical information sheet, it states: ‘Flammable. Keep away from fires and ignition sources during storage and installation. Do not smoke or expose to flame sources during application.'"

It is important to note the warning information is contained only in a technical information sheet and does not appear on the product labels of materials delivered to job sites. When was the last time your foremen were provided with a technical information sheet for a polyisocyanurate foam roof insulation product? And how does a "flammable" insulation board qualify for a fire-resistant roof system?

Replicating the flame

To determine the degree of susceptibility of roof insulation boards to burn when exposed to "fires and ignition sources," a sample of the insulation was cut and subjected to virtually instantaneous exposure to flame from a small propane torch. The facers on both sides of the insulation board, as well as the foam, ignited immediately. But the foam self-extinguished, leaving the facers on both sides of the board burning without any indication of extinguishing, just as they had done over the steel roof deck.

About a 4-inch (102-mm) square section of the facer was placed in a crucible in an ash oven to determine the approximate quantity of cellulose fibers in the recycled paper facer. Seventy percent of the facer material "ashed out" of the sample, leaving a residue of what appeared to be clay filler (probably residue from the coated recycled paper).

We then searched the ash for the glass reinforcements mentioned in the correspondence. None were located under strong magnification, so the ash was placed on a microscope slide and examined under 20X magnification. Three glass fibers were located in the ash—hardly what I would refer to as "glass-reinforced recycled paper."

Given the quantity of cellulosic material in the facers, it only appeared prudent to determine what kind of moisture absorption could be expected from the "glass-reinforced recycled paper" facer. The facer was removed from the foam, scraped clean of residual foam and placed in an autoclave for 60 minutes. The facer absorbed 47 percent of its dry weight in water and turned into something resembling a potato chip, being generally distorted following exposure to moisture. (See Photo 1.)

So what about the insulation board as a whole? One cycle in the autoclave produced an ugly, distorted sample. The facer on one side of the sample curled upward, and the facer on the other side shrank and tore loose from the foam core. Photo 2 shows the extent of the foam core's distortion caused by the facer. The whole board absorbed 28 percent of its dry weight in water. Was the foam to blame for the distortion, or was it only the facer?

A sample of the foam core was prepared by removing the "glass-reinforced recycled paper" facers from both sides of the board sample and subjecting them to the same conditions as the whole board. The foam absorbed 19 percent of its dry weight in water but remained flat and otherwise unaffected by the high-temperature and humidity exposures. (See Photo 3.)

The problem

By this time, it was obvious the "glass-reinforced recycled paper" facers were the major culprits in distorting the foam core and contributing a large degree of moisture absorption to the whole insulation sample. There did not appear to be any significant difference in the properties (including flammability) of the core foam manufactured with either HCFC-141b or pentane blowing agents; the facers were the determining factor in the "whole board" stability.

A sample of another manufacturer's polyisocyanurate insulation board randomly was selected from stock to determine whether the physical properties of the whole board were similar to those of the project sample. Facer weight and composition virtually were identical, and test results indicated the properties essentially were similar to the project sample. (See Photo 4.)

Facer distortion caused separation of the foam core in close proximity to the board's surface, but the second sample did not react as violently as the first sample.

A third sample of pentane-blown polyisocyanurate foam roof insulation, surfaced with 100 percent fiberglass mat facers, was subjected to similar exposure as the first two samples.

There was no distortion of the whole board or facers. However, there was some separation of the foam from the facer at the adhered interface as the foam expanded from temperature and moisture exposure. (See Photo 5.) But for proof that fiberglass mat facers are not created equally, see Photo 6.

Ironically, while visiting another job site, we found pallets of newly produced insulation boards (dated Jan. 12, 2003, on the label) improperly stored in the shipping wrappers. The site visit was on Jan. 30, so the maximum exposure time for the material was about two weeks. There was extensive separation of facers from the foam core; boards were warped and distorted while still in the shipping wrappers; the foam was cracked between facers; and in one instance, the facer was about 1/2 of an inch (13 mm) longer than the foam core. It obviously wasn't manufactured that way—there still was foam residue on the facer's bottom side.

Guidelines for installers

Just what does all this boil down to for installers? Following are my recommendations and key findings:

• One hundred percent fiberglass facers currently are available from some insulation manufacturers. They should be the choice of anyone interested in quality polyisocyanurate insulation. They will be more expensive than the "glass-reinforced recycled paper" facer products presently available, but overall dimensional stability should be significantly better.
  
  
• Given the properties of roof insulation with "glass-reinforced recycled paper" facers, use extreme care in the storage and handling of these materials. If facers are exposed to high humidity during storage, facer separation appears to be a given, and either blistering of hot-applied roof membrane systems or general adhesion of adhesive-applied roof membrane systems may be a significant problem.
  
  
• Using a torch to seal laps in polymer-modified asphalt membranes may cause ignition of "glass-reinforced recycled paper" facers, and torch applications must be monitored continuously for fire from burning facers. Even a hot-air welder left in place too long may result in autoignition of the "glass-reinforced recycled paper" facers on current polyisocyanurate insulation boards.
  
  
• Use small boards—48 inches by 48 inches (1219 mm by 1219 mm)—to minimize boards' potential distortion during storage and installation. Extraordinary precautions may be necessary when paper-faced polyisocyanurate insulation boards are to be installed in adhesives, including low-rise foams, to maximize the surface contact areas between foam and substrate.
  
  
• Distortion of polyisocyanurate insulation with "glass-reinforced recycled paper" facers can best be controlled by mechanically attaching roof insulation according to current installation recommendations on steel roof decks.
  
  
• Boards that exhibit facer separation must be rejected and discarded, especially when they are to be installed under "adhered" roof membranes.
  
  
• Polyisocyanurate insulation manufacturers should include information as important as product flammability on shipping containers or product labels, not just in technical bulletins.
  
  
• Roofing contractors must be more careful than ever when choosing and evaluating which products to use before beginning each project. Fiberglass-mat-faced polyisocyanurate insulation should be the product of choice in any application where insulation boards are to be set in any type of adhesive, including asphalt, or where torch or hot-air welders are to be used to form laps/seams in roof membranes directly applied over polyisocyanurate insulation.

Ask for quality

Foam structure in board thickness in excess of 2 inches (51 mm) can contribute significantly to dimensional stability. Single-layer polyisocyanurate foam roof insulation should not exceed a 2-inch (51-mm) thickness. Multiple insulation layers should be used for higher R-value requirements.

Unfortunately, fiberglass mat facers are not created equally, and some perform much better than others. Some facer mats are permeable, and others are relatively impermeable. Foam insulation faced with less-permeable facers will retain its insulating properties longer than insulation faced with permeable facers. But that is a long story to be told at a later time.

The better quality polyisocyanurate foam roof insulation products have been the insulation industry's best-kept secret. "Glass-reinforced recycled paper" facers make the foam products more competitive, not better. If you want better insulation, specify it or ask for it by name. If you don't care or only wish to be competitive, the industry's problems with polyisocyanurate insulation only will increase. Kudos to the manufacturers that are paddling upstream to make better insulation products.

Dick Baxter is president of CRS Inc., Monroe, N.C.