Global Warming Problems with Spray Polyurethane Foam Insulation Questioned

The Spray Polyurethane Foam Alliance (SPFA) is a trade association representing the spray polyurethane insulation and roofing industries, including contractors, suppliers, distributors, and consultants. Several of our members have brought to our attention your article “Avoiding the Global Warming Impact on Insulation,” published in the June 2010 issue of Environmental Building News. This article inaccurately targets the high global warming potential (GWP) of blowing agents used by closed-cell spray polyurethane foam (SPF) and extruded polystyrene (XPS) relative to other insulation products, and provides insulation material recommendations based on flawed analyses. SPFA believes the assumptions, analyses, and statements are inaccurate and damaging to the SPF industry.

SPFA considers evaluating the environmental impact of building products to be a valuable and required step in the material selection process. However, this evaluation must be done using established analytical methods. Our industry believes that the analyses used in this article need refinement in several areas to accurately characterize and compare the environmental impact of the different insulation products. Specifically, the analysis is deficient in the following six areas:

We detail our concerns regarding each of these topics below, and suggest improvements for the authors to consider.

Richard S. Duncan, Ph.D., P.E.Technical Director, Spray Polyurethane Foam Association

Editor’s response:

We commend the SPF industry for efforts to address environmental characteristics of polyurethane foam and for providing in-depth responses to our June article. Our responses to the specific points mentioned above are interspersed (in italics) in the detailed discussion below.

Detailed Discussion

Below is a detailed discussion addressing each of SPFA’s six concerns with the referenced article Avoiding the Global Warming Impact on Insulation, as published in Environmental Building News, June 1, 2010.


SPFA Point #1: Amount of HFC-245fa blowing agent in SPF is overstated

The amount of blowing agent used in closed-cell SPF was mistakenly assumed to be two times higher than is actually used in practice. The Harvey paper assumes that HFC blowing agents comprise 12% of the weight of closed-cell SPF. The fact is blowing agents make up about 12% of the B-side (polyol) component. When SPF is manufactured on the jobsite, it is combined with an equal amount of A-side (MDI) to create the finished foam. This 1:1 field mixture of the A-side and B-side reduces the amount of blowing agent by a factor of two, so the blowing agent is approximately 6% of the weight of the finished foam, not 12% assumed by the Harvey study. The analysis should be re-done considering this information, or the payback term should be reduced by a factor of two.


EBN response to Point #1

In writing about the greenhouse gas emissions from SPF and XPS we relied on the best data we had available to us at the time (including the peer-reviewed Harvey paper), but we welcome newer or better data from SPFA and other sources on the exact quantity of HFC blowing agent in SPF as well as more precise estimates of offgassing rates of that blowing agent—something not addressed in your comments. We are also seeking this data from one of the leading manufacturers of blowing agents.

SPFA Point #2: Embodied GWP assumptions may be incomplete or inaccurate

The embodied GWP reported by ICE for the different insulation products in Table 2 may not be accurate or complete and may not follow standard practice. It is important to compare the environmental impact of different materials only when they have undergone a thorough life-cycle analysis (LCA) or life-cycle inventory (LCI).

There are established LCA protocols that must be followed for all products (per ISO 14040/14025) for the results to be meaningful and comparable. These ISO protocols define the boundary for the LCA, such as cradle-to-grave or cradle-to-gate, and include all aspects of producing the product such as the environmental impact of raw material extraction, transportation, and processing, as well as production of the final product. In addition, these protocols establish a functional unit for each product and require an independent review by industry experts. In addition, there is a Product Category Rule for insulation materials established by the ISO that defines the proper boundaries and functional unit for characterizing the environmental impact of all insulation products.

It is not clear from Table 2, or from the analysis in this paper, that the ISO standards were consistently followed for every material. If these standards were not followed, then their environmental impacts cannot be compared.

The spray foam industry, through SPFA and an independent service provider, is undertaking an industry-wide cradle-to-grave LCA to incorporate not only the embodied energy required to make, transport and install SPF, but to also consider the building energy saved during the use phase and proper disposal of the product. We believe that the results will show the GWP impact of energy saved will be 30 to 100 times that of the GWP impact to manufacture SPF. When completed within the next 12-18 months, the SPF industry LCA should provide thorough evidence to challenge many of the statements and conclusion in this article.

Finally, there are several existing LCAs developed for SPF. A summary example of an LCA for the embodied energy of several insulation materials, including SPF made with HCFC-141b blowing agent, is shown in “Eco-Efficiency Analysis of Insulation Products,” by BASF in 2006. This work, using the ISO protocol, should be included or referenced in the article.


EBN response to point #2

We agree about the importance of in-depth LCA data, and we welcome that data as we evaluate the environmental attributes of building materials. Such information often isn’t available, however. In lieu of more rigorous LCA and LCI data on insulation materials, we relied on what we consider to be the best, freely available, academic information on embodied energy—that from the Inventory of Carbon and Energy (ICE) from the Sustainable Energy Research Team at the University of Bath, U.K., Department of Mechanical Engineering.We do not accept that the lack of more comprehensive LCA or LCI data on a material, however, is justification to hold off on making decisions about use of that material or alternatives. While more rigorous analysis may show some differences from the embodied energy assumptions we used in this analysis, we believe that the ICE data we used is in the right ballpark.

SPFA Point #3: Sole consideration of SPF as a secondary or additional insulation over a baseline R-value unfairly biases the conclusions of the study

Considering foam plastics only as a secondary or additional insulation unfairly positions them at a disadvantage in the analysis by dramatically reducing energy savings and increasing the payback period for these products.


Figure 1 – Energy Lost versus R-value

Figure: SPFA
The relationship between conductive energy loss of a building and its envelope R-value is inherently non-linear, as shown in Figure 1. The example in this figure it is assumed that the uninsulated wall has an R-value of 1. At this R-value, the energy loss of the wall is assumed to be 100%. If a continuous R-13 primary insulation is added, the energy loss from the insulated wall will be reduced by about 86% of its uninsulated value. If an additional R-6 of ‘secondary’ insulation is added to the wall with R-13 of primary insulation, an additional 4% energy savings will be realized.

In this paper, it was assumed that a fibrous insulation was the primary insulation, and that foam plastic was the secondary insulation. This assumption greatly reduces any energy savings benefit from the foam plastic. If R-6 of foam plastics was considered as the primary insulation, and a fibrous R-13 was considered as a secondary insulation, the R-6 foam plastic would be credited with about 75% of the savings and the R-13 fibrous insulation with about 17%. The assumption of primary versus secondary position can have a significant effect on the energy savings used to calculate payback in this paper.

Moreover, foam plastics are routinely used as the only insulation in many buildings. For example, XPS insulated sheathing is used along with SPF as an external continuous insulation over walls and low-sloped roofs. SPF is often used as the only cavity insulation in many framed structures. SPF in combination with fibrous insulations is a far less common application than using SPF insulation alone.

To make this analysis more representative and to compare all products on an equal basis, all insulations should be considered alone as the primary insulation.


EBN response to Point #3

Your point about analyzing SPF solely as an added (or secondary) insulation material is a reasonable one. When SPF (or XPS) is the only insulation material being used in a particular wall system or other application, or if the energy savings from this foam insulation is considered primary, then the speed at which the global warming potential from the HFC emissions will be “paid back” from the energy savings resulting from that insulation is far more rapid; I did not make that clear enough in my article, which I regret. The example we used—that of a 2x6 wall insulated with cellulose with added foam insulation—admittedly is a more common construction detail when XPS is added than when SPF is added. We hope to present greater clarity in future coverage of this issue.

SPFA Point #4: GWP impact of fourth-generation blowing agents not fully reported


Table 1 – Environmental Impact (GWP and ODP) of closed-cell SPF Blowing Agents

Figure: SPFA
Fourth-generation blowing agents for SPF, with significant reductions in GWP, are not fully reported. During the past two decades, blowing-agent manufacturers have worked diligently to reduce the environmental impact of these chemicals in regard to ozone depletion and greenhouse gas production. A brief summary of this development is provided in Table 1.

Three major chemical companies have recently announced the impending launch of fourth generation of SPF blowing agents within the next 1–2 years—see “HFO-1234ze(E) Commercial Status, And HFO LGWP Advancements,” Bowman, J. and Williams, D, (Honeywell); “Investigation Of New Low GWP Blowing Agent AFA-L1 For PUR/PIR,” Chen, B., Costa, J., Bonnet, P. (Arkema); and “Development Program Update For Low GWP Foam Expansion Agent,” Loh, G., Creazzo, J., Robin, M. (DuPont), all presented at the Polyurethanes 2009 Technical Conference.

These fourth generation materials have zero ODP and GWP in the range of 6 to 15, representing a GWP reduction of more than 150 times current values reported in this paper. These fourth-generation blowing agents were not included in Table 1 of the article, and should be added in fairness to future of these materials. We believe the continuing evolution of SPF blowing agents should be considered and discussed.


EBN response to Point #4

In our article, we did state that as fourth-generation blowing agents come into use, “the argument for avoiding SPF and XPS on the basis of lifetime GWP will largely disappear.” Our information, like yours, points to hydrofluoroolefins (HFOs) as the likely replacement blowing agents, at least for the polyurethane insulation. With XPS, Europe has already switched to a fourth-generation, non-ODP, low-GDP blowing agent: carbon dioxide (from water). However, the XPS industry has decided that for the North American market, an R-5-per-inch insulation material is needed, rather than requiring designers and builders to increase the thickness of R-4-per-inch XPS (as is available in Europe today). It also bears repeating just how much improvement has been made over the past 15 years by the insulation industry in phasing out the highest-ODP CFC blowing agents—which also had far higher GWP values than the current HFC blowing agents. The industry deserves a lot of credit for these improvements, and the environmental community and government agencies may share some fault for not making a bigger issue of the GWP properties of alternative blowing agents being considered—because the priority was given to dealing with ozone depletion.

SPFA Point #5: Air sealing properties of foam plastics ignored

The U.S. Department of Energy states air leakage can account for 20%–40% of a building’s heating and cooling costs. In the analysis, the effects of energy loss from air leakage were ignored. Foam plastics are air-impermeable, minimizing air leakage and convection effects found in fibrous insulations. Most foam plastics are moisture resistant, and, unlike fibrous insulations, are not degraded by moisture content.

Moreover, SPF expands during installation to seal cracks and gaps in the building envelope. When used with proper sealing around windows and doors, SPF can provide complete air barrier system for most buildings. Fibrous insulations are inherently air permeable and will need an additional air barrier system installed to perform equivalently to SPF. To account for the air sealing benefits, the analysis should have considered SPF to have an additional 20%–40% energy savings, or the embodied energy of the added air barrier system for fibrous insulation should have been included for these products.


EBN response to Point #5

I in our analysis we assumed that in a highly insulated building—which our article was focused on—airtightness will be a priority no matter what the insulation material used, so that additional benefit from SPF’s (superb) air-sealing should not be factored in. Remember, our analysis focused on highly insulated buildings—those approaching net-zero-energy or Passive House performance. In typical construction practices, the levels of insulation provided and the GWP issues associated with SFP and XPS are much smaller.

SPFA Point #6: Inaccurate statements regarding quality, chemical safety and off-gassing of SPF

High-pressure SPF must be installed by a professional installer. Installers undergo many hours of training on chemical safety, equipment operation and material application. Pumps and heaters, using automated controls, are used to dispense and mix SPF chemicals under precisely controlled pressures and temperatures. With proper training and modern equipment, it is very difficult to apply SPF materials improperly. An experienced installer can easily control the thickness of closed-cell SPF to within ¼” to ½". It is more difficult to control the thickness of open-cell foam—where a thickness control of 1” to 2" is typical.

SPF formulations contain flame retardants for added safety in the event of a fire. However, they do not contain hexabromocyclododecane (HBCD), polybrominated diphenyl ether (PBDE), or tetrabromobisphenol A (TBBPA), chemicals that have been the focus of concern by the U.S. Environmental Protection Agency in recent years. SPF products typically use phosphate-based, chlorinated flame retardants such as TCPP (tris (1-chloro-2-propyl) phosphate), TEP (triethyl phosphate) and TDCPP (tris(1,3-dichloro-2-propyl) phosphate).

Any exposure concerns to chemicals in liquid or aerosolized form during the application of high-pressure SPF are addressed through the use of proper personal protective equipment (PPE). Those in the immediate vicinity of application are trained on the requirements and proper use of PPE such as eye, skin and respiratory protection. Shortly after application, the components react to form the final insulation material, which is highly inert and presents little hazard to anyone who comes in contact with it. As an added precaution, it is also common practice to not allow re-occupancy until 24 hours after SPF is installed. In addition, studies have shown that SPF does not release toxic gases or leach harmful chemicals into the soil. Cured SPF materials are routinely encountered and handled safely, and enhance everyday life. More information on safe application of SPF can be found at www.spraypolyurethane.comIn terms of offgassing, SPF is not known to emit any significant levels of volatile organic compounds (VOCs). All Canadian SPF products must be tested to assure low VOC levels to be compliant with Canadian building codes. These same SPF formulations are used in the U.S. Several U.S. SPF manufacturers have voluntarily evaluated and registered their products with the GreenGuard Environmental Institute to assure that VOCs are below safe levels.


EBN response to Point #6

While it is certainly true that the rigorous training SPF installers receive is very important and helps to ensure quality installations, which comprise to majority of installations, suggesting that “it is very difficult to apply SPF materials improperly” may be overstating the reality. We have been hearing more and more anecdotal reports of problem SPF installations in which shrinkage or other problems occur. We believe that most of these problems stem from formulations and materials selected, ironically, for environmental reasons—such as use of soy oil in the polyol component of the foam or use of water-blown (CO2 blowing agent) formulations. I believe that SPFA is trying to address these installation problems, but there may still be some chemistry issues to work out with the newer materials. As for flame retardants, it is true that SPF does not contain brominated compounds, which are most commonly targeted by health and environmental advocates. There are also concerns with chlorinated flame retardants, however, including TCPP. We look forward to a time when halogenated flame retardants are entirely removed from foam insulation materials.We agree that when properly installed, VOC offgassing from SPF should be negligible. As long as proper protective gear is worn during installation (as SPFA advocates) and building occupants keep out for 24 hours, indoor air quality problems should be rare—and limited to people with extreme chemical sensitivity.

SPFA Summary of Points 1–6

If the issues above are properly addressed, we believe that the payback term for SPF will be dramatically reduced. While a complete re-evaluation of the payback is needed, the reduction of the payback term can be approximated as follows if points 1, 3, and 5 are properly addressed:

• (1) Amount of HFC-245fa blowing agent 0.5 factor

• (3) Secondary vs. primary insulation (4% savings to 75% savings = 19x) 0.05 factor

• (5) Air sealing properties included (30% average savings) 0.70 factor

• 36 year payback (1" of SPF) x 0.5 x 0.05 x 0.70 = 0.63 years actual payback


EBN response to SPFA summary

Regarding your summary calculations that reduce the 36-year payback we calculated to seven and a half months (0.63 years), we’re not in agreement that the SPF should be considered primary (point 3), but we agree that it isn’t fair to consider it solely a secondary insulation material either. I would like to have LCA experts weigh in on this question. We are also not in agreement that in a highly insulated enclosure system, SPF can be given credit for 30% energy savings due to air sealing—though that is a reasonable assumption for standard (leaky) construction.That said, many of your points are well taken. Thank you for taking the time to provide this in-depth response. Our goal at EBN is to improve the environmental performance of buildings, and dialogue such as this will do a lot in moving the discussion forward. I believe that we share the overall goal of reducing the environmental impacts of buildings and the importance of insulation in achieving that goal.– Alex Wilson




Published August 30, 2010

Wilson, A. (2010, August 30). Global Warming Problems with Spray Polyurethane Foam Insulation Questioned. Retrieved from https://www.buildinggreen.com/op-ed/global-warming-problems-spray-polyurethane-foam-insulation-questioned

Add new comment

To post a comment, you need to register for a BuildingGreen Basic membership (free) or login to your existing profile.


January 16, 2013 - 7:19 am

It seems to me that adding R-19 to the wall discussed cut heat loss 90%. Of the R-19, R-6 was SPF and R-13 was fiberglass, so 6/19 of the 90% was due to SPF, or 28%. This is 7 times the original articles 4%, so the payback would be 36/2 = 18 years due to the error on how much blowing agent was used, and 18/7 = 2.6 years if you do not count either as primary or secondary.

April 22, 2011 - 6:40 am

I gather from LCA folks that the impacts over the assumed service life vary from product to product and have different intervals or bursts--the early extraction and manufacturing impacts, the installation impacts, the slow release in use, the repair interval etc. They normally sum them up over the entire service life but factoring them out is complicated because it does not follow a neat straight line or other depreciation-like curve.

April 10, 2011 - 12:08 pm

at the BE11 presentation a gentleman in the audience suggested that the GWP tool use depreciation as a way to make the life span a less important input (and in his opinion make the analysis more correct). anybody know how i can contact this guy?

April 10, 2011 - 6:53 am

The optimistic result the SPF industry expects from a rigorous LCA analysis is dependent on the assumption of a service life for the assemblies (usually 40-70 years)that is significantly longer than the critical period for climate change in which we have to achieve net reductions of global warming impacts. A shorter and more appropriate assumed service life increases the importance of the initial embodied GWP.

As for dense-pack, papers at the recent NESEA conference document problems with its use in ceiling applications in very cold or humid applications: caution!

March 25, 2011 - 12:53 pm

Dense-pack cellulose is certainly a "greener" product that SPF, and it does tend to resist air infiltration. It is a little fussier in the appplication of it compared to SPF, and there is a higher potential for unintended voids. However, I have seen it applied successfully in walls, and it works well, though I don't think the R-value per inch is as high as some of the SPF. On the plus side, it is very vapor permeable, which can be a problem with the high R-value SPF products. That does help with making the wall assembly more forgiving to varying temperature and water vapor differentials across the assemblies.

March 25, 2011 - 12:42 pm

James, if you re-read my comment you will find that I did not say that more insulation is always better. I am a little alarmed that you read that into my comment! I don't think that's true. I agree with what you said about ROI getting longer.

Regarding convection, I was trying to respond to John Sesic, and so I don't know what he had in mind in broaching the subject. I was thinking of convection through the insulation.

I disagree that SPF is unique in resisting air infiltration. Dense-pack cellulose, for example, is highly resistant to air movement. I have concerns about overreliance on SPF as an air barrier, in part because of the potential for cured insulation to break away from rafters, framing members, sheathing, etc., as those members shift due to wind, drying, long-term building movement, etc.

March 25, 2011 - 12:31 pm

Sorry to intrude into your dialog, but I am curious: when you are discussing convection, are you discussing thermal convection through the insulation, or convection by air circulation within the insulation cavity? If I remember my insulation primer correctly, the heat transfer is retarded by the air spaces between the insulating materials, not the materials themselves. If that is the case, then heat convection is the primary mechanism that is being retarded, assuming no air circulation within the insulating layer.

Also, you make a general claim that more insulation is always better (assuming there is room). If one does not have to consider life cycle cost, that is true. However, once you get beyond a certain thickness (depending upon the type of insulation), the return on investment gets exponentially longer.

Where I find most insulation systems come up short is that they do not resist air infiltration, should the primary air barrier fail. The exception to that is SPF. One can assume a good air barrier can be installed on the exterior of the building envelope, but my experience is the installation of those barriers are inconsistent, and tend to separate over time as the building moves. The spray foam insulation provides a pretty secure means of air sealing the building without reliance upon highly skilled workers, or clean building surfaces (building paper wraps with tape seals is an example of where clean surfaces are a must). Ideally, you seal the building then insulate the exterior of the wall or roof. In practice, that is very hard to do successfully.

March 25, 2011 - 11:41 am

John, I'm curious if you can say more about that logic. I don't think it holds up under examination.

Let's say you have an SPF product that offers R-5 per inch. That is a good value on a per-inch basis, but does it make sense to stop insulating after one inch? No—for an efficient building you'd want to target a high R-value for the assembly, something like R-40 or higher. R-value measures resistance to heat flow (mostly via conduction) and higher R-value will slow down heat flow more effectively.

Foam is good at stopping convection through the insulation material, but it is not unique in this way. Many other insulation materials are also good at this, particularly when paired with an effective air barrier.

March 10, 2011 - 6:17 pm

Mr White,
You are right of course. There are other mechanisms for stopping infiltration, in addition to SPF. They have been used for decades with varying degrees of success. The reason I prefer SPF is the application does not rely on consistently applied tape or liquid coatings to insure uncontrolled infiltration does not occur. Both tape and liquid applied coatings have limitations which tend to show up entirely too often. For instance, the tape adhesive relies on an application to a clean surface; something one finds rarely on a construction site. Liquid applied coatings rely less on clean surfaces, but do require a minimum coating thickness consistently applied. This is difficult to achieve, particularly at the intersection of the wall sheathing and the windows or roof. Tight drywall application last for maybe the first month; imperceptible cracking occurs due to the building shifting and settling as the soil pad adjusts to the weight of the structure.
SPF is not a cure-all by any means, but it does resolve many of the problems we find with most of the typical methods of stopping infiltration.

March 10, 2011 - 4:15 pm

Dear Mr. Thornton,

Superior air tightness may be achieved using SPF, or a variety of other approaches that are independent of the insulation (taped sheathing, air tight drywall, etc). Most importantly, these techniques need not impose significant GWP. To me the issue is not black and white, but the stance I took on this is that, given good design, superior air tightness can be achieved regardless of insulation material, without significant rise in embodied GWP; therefore it drops out as a basis of comparison. Note that infiltration was also neglected by Harvey in his paper, and by EBN/BSC in their study (see "EBN response to Point #5" above).

March 10, 2011 - 1:24 pm

It is assumed Mr. White believes the thermal properties of any type of insulation is the only valid criteria when evaluating actual received R value versus GWP. It is one factor. With SPF products, air leakage and its associated effects in wall assemblies as it pertains to both thermal transfer and infiltration are no longer in the equation. With other types of insulation products, air leakage is problematic even with the best installers and is a factor in realizing the actual thermal resistance anticipated. The effect of air infiltration on a building's heat loss or load cannot be understated. One should take that factor in consideration when evaluating SPF products versus your more typical insulation products used in shelter construction. And to the question in your mind, no, I am not an employee or otherwise associated with SPF manufacture or application. Just an engineer that has seen it used in commmercial construction and have seen the effects of its air barrier properties first hand. Do the math.

September 7, 2010 - 1:32 pm

I believe point #3 is misleading. At low R values, any insulation has a very short payback, and for most insulations, it is still far shorter than that of XPS and SPF.

The most fundamental point is that high-GWP insulations are limited in their potential to minimize the net impact associated with heating and cooling a building. This means that for high R-value assemblies, they substantially underperform compared to lower GWP insulations.

It also means that beyond a certain R value, adding more insulation will do more harm than good from a climate change standpoint. For instance, there is no place for XPS in an assembly over R20 in a Boston climate. Over R20, the presence of XPS will increase, not lower, the net impact. By the way, it doesn't matter if you added the XPS first or last. In either case, you'd be saving the planet by removing it.

The point at which net impact "bottoms out" (or net benefit tops out) exists for all insulations; it's just that for most insulations, it requires assembly thicknesses that are impractically high, so it never really happens.