What’s embodied energy, and is there any reason to pay attention to it? Embodied energy is the energy it takes to manufacture building materials. Until recently, it was safe to advise builders that it wasn’t worth worrying about embodied energy, because the amount of energy (especially heating energy and cooling energy) used to operate a building over the building’s lifetime dwarfed the relatively small amount of energy embodied in the building materials.
As builders choose to install thicker and thicker layers of insulation in their buildings, however, the old rules of thumb aren’t working as well as they used to. Installing 6 inches of rigid foam insulation under a slab on grade may make sense in a cold climate; but if a Passivhaus builder decides to specify an additional 6 inches of foam, bringing the thickness of the sub-slab insulation up to 12 inches, it’s unlikely that the final 6 inches of foam will ever save as much energy as was required to manufacture the foam.
Defining embodied energy
The embodied energy of a building component or of an entire building is calculated by adding:
- The energy required to extract the raw materials used to make the building component (for example, the energy required to mine iron ore and make the steel wire that becomes a nail).
- The energy required to transport the raw materials to the factory where the building components are manufactured.
- The energy required to manufacture the building materials (for example, the energy required to transform steel wire into nails).
- The energy required to transport the building materials to the building site.
- The energy required to put the materials together at the building site, including the energy needed to transport the workers to the site.
Some, but not all, definitions of embodied energy include a few other inputs:
An embodied energy calculation differs from a life-cycle analysis. A life-cycle analysis includes embodied energy, but also encompasses additional environmental impacts, such as the environmental effects of resource depletion and pollution.
Factors affecting the importance of embodied energy
There are several ways to report the embodied energy of a building. Assuming that one is capable of making the necessary calculations, embodied energy can be reported in energy units (for example, in joules or MMBTU). If a building is being demolished (and its age is known), you can report embodied energy as a percentage of the total lifetime energy use of by the building, or in “years of operating energy.”
To take an example: imagine that a 40-year-old building is being demolished. The embodied energy of the building was 1,000 MMBTU, and the building used 100 MMBTU of operating energy every year (or 4,000 MMBTU over its lifetime). If we add the building’s lifetime operating energy use to its embodied energy, we find that the building’s total lifetime energy use was 5,000 MMBTU. We can describe the relationship between the building’s operating energy use and its embodied energy in at least two ways:
- The building’s embodied energy use was equal to 10 years of operating energy.
- The embodied energy amounted to 20% of its lifetime energy use.
So what factors affect the numbers reported above? Compared to operating energy, embodied energy is more important:
- In mild climates than in severe climates (because buildings in mild climates don’t need much energy to operate);
- In buildings with high levels of insulation than in buildings with low levels of insulation (because the amount of operating energy saved with each added inch of insulation is less than the first few inches);
- In buildings that are demolished after a short lifetime than in buildings that are demolished after a long lifetime.
If you are considering increasing the thickness of the insulation in a new building, it makes little sense to add insulation that requires more energy to manufacture that it will save over the lifetime of the building. Unfortunately, you can’t calculate how much energy the insulation will save unless you know when the building will be demolished.
Increasing the energy efficiency of a building is admirable, of course. Bur remember: as you reduce a building’s requirement for operating energy, the embodied energy takes on a larger and larger percentage of the building’s total lifetime energy budget, and there is an increasing burden on the owner to keep the building intact (rather than demolish it) for many decades.
Estimates of the embodied energy of single-family homes
It’s fairly common to see published estimates of the embodied energy for various building components; unfortunately, many of these estimates are based on work performed decades ago.
Twenty years ago, Ray Cole, a professor at the University of British Columbia’s School of Architecture, made some oft-quoted estimates of the embodied energy in residential buildings. Cole estimated the embodied energy of a 3,750-square-foot ranch house. According to Cole, a “conventional house” has 2×4 walls and R-24 ceiling insulation, while an “energy-efficient” house has 2×6 walls and R-42 ceiling insulation.
- For a “conventional” house in Toronto, the heating energy use is 136 MMBTU/year; embodied energy equals 948 MMBTU (or 7.0 years of heating energy).
- For a “conventional” house in Vancouver, the heating energy use is 101 MMBTU/year; embodied energy equals 948 MMBTU (or 9.4 years of heating energy).
- For an “energy-efficient” house in Toronto, the heating energy use is 78 MMBTU/year; embodied energy equals 1,019 MMBTU (or 13.1 years of heating energy).
- For an “energy-efficient” house in Vancouver, the heating energy use is 57 MMBTU/year; embodied energy equals 1,019 MMBTU (or 17.9 years of heating energy).
An Australian organization, the Commonwealth Scientific and Industrial Research Organization (CSIRO), estimated that the embodied energy of a single-family detached house is about 1,000 GJ — a number that corresponds well with Coles’ estimates, which ranged from 948 to 1,014 GJ. According to a website maintained by the Australian government, 1,000 GJ “is equivalent to about 15 years of normal operational energy use. For a house that lasts 100 years this is over 10 percent of the energy used in its life.”
In a 1995 article in Home Energy magazine, Tracy Mumma reported that the embodied energy of a typical house in Toronto amounts to 10 years of operating energy use — halfway between Cole’s estimate for a “conventional” Toronto house (7 years) and Cole’s estimate for an “energy-efficient” Toronto house (13 years).
An estimate in the same ballpark was make by B. Lippke in a 2004 paper, “Life-Cycle Environmental Performance of Renewable Building Materials.” Lippke concluded that embodied energy accounts for about 15 percent of a building’s lifetime energy use.
What about buildings with above-average amounts of insulation?
Summarizing various studies, architect Bruce Coldham of Amherst, Massachusetts, estimated typical ranges for embodied energy. Because Coldham pays attention to buildings that have better-than-average levels of insulation, he reported that embodied energy is more important than many designers realize. Coldham wrote that a building’s embodied energy “likely amounts to roughly the one quarter (25%) of the operational energy requirement over the life of the structure. This is quite a bundle, and the bundle gets proportionally bigger if the building is designed, constructed, and managed efficiently. Such an operationally efficient building may see more than half of its total lifetime energy requirement committed before the occupants even move in.”
In a paper titled “Energy and resources, material choice and recycling potential in low-energy buildings,” Catarina Thormark looked at buildings with above-average levels of insulation, where embodied energy can be assumed to take on a greater importance than in conventional buildings. She noted, “Studies have been carried out on low-energy buildings and show that embodied energy can account for as much as 40-60% of total energy use.” (This calculation depends, of course, on assumptions about building longevity.)
Thormark set to make her own calculations based on the assumption that buildings last 50 years. After looking at three multifamily low-energy residential buildings in Sweden (including the famous 20-unit Passivhaus project in LindÃ¥s), she concluded, “For a service life of 50 years, production and transportation of materials accounted for 60-75 % of the [buildings’] total impact. These results completely change the prevailing picture that operation accounts for the main part of energy use.”
The reason that embodied energy loomed so large in the buildings examined by Thormark is that she chose to look at low-energy buildings.
Many authors have tried to quantify the degree to which energy-efficiency improvements affect embodied energy calculations. In a 2006 paper by I. Sartori and A.G. Hestnes titled “Energy use in the life cycle of conventional and low-energy buildings: A review article,” the authors introduced their topic with this overview: “Until a few decades ago it was known that operating energy represented by far the largest share in the life cycle energy bill, ranging to about 90-95% even when accounting only for the heating demand.” To determine whether that rule of thumb still holds, the authors analyzed 60 buildings in nine countries (Australia, Canada, Germany, Japan, New Zealand, Norway, Sweden, Switzerland, the U.S.). They concluded, “For those cases that matched the definition of low-energy [buildings], the embodied energy’s share of the total ranged between 9% and 46%. … Conventional buildings had shares ranging between 2% and 38%. … When compared with an equivalent conventional building instead, the passive house demanded only slightly more embodied energy while it reduced the total energy need by a factor of three, when operating energy was expressed as primary energy and the lifetime assumed to be 80 years.”
New construction versus rehabbing existing buildings
Every time a building is demolished, it’s possible to calculate the ratio of the building’s embodied energy to its total lifetime energy use. The longer the day of demolition is delayed, the smaller that ratio becomes.
The National Trust for Historic Preservation, a nonprofit organization with an undisguised agenda, released a paper that shows the energy advantages of preservation compared to demolition and new construction. The paper notes, “Savings from reuse are between 4 and 46 percent over new construction when comparing buildings with the same energy performance level. The warehouse-to-multifamily conversion — one of the six typologies selected for study — is an exception: it generates a 1 to 6 percent greater environmental impact relative to new construction in the ecosystem quality and human health impact categories, respectively. This is due to a combination of factors, including the amount and types of materials used in this project. This study finds that it takes between 10 to 80 years for a new building that is 30 percent more efficient than an average-performing existing building to overcome, through efficient operations, the negative climate change impacts related to the construction process.”
Needless to say, the analysis performed by the National Trust for Historic Preservation is based on a great many assumptions, and individual buildings may not fall neatly into any of the categories that the analysts created. Suffice it to say that it sometimes makes perfect sense (and it is good for the environment) to bulldoze an old, inefficient building and replace it with a new, energy-efficient building.
There are many uncertainties
Unfortunately, there are so many uncertainties in embodied energy calculations that designers can be forgiven for just throwing up their hands and giving up.
What follows is a list of some of the factors affecting the accuracy of embodied energy calculations.
Some analyses are biased. According to a 1993 article in Environmental Building News, “Industry associations representing wood products, plastics, and concrete have all commissioned studies in the hopes of showing the energy advantages of their materials. As might be expected, the conclusions released from those studies that have been completed vary tremendously.”
Calculations by different organizations can’t be compared because they don’t all include the same energy inputs. Some analysts include the energy used to transport building materials and construction workers to the building site, while others omit these inputs. Some analysts include the energy used to make the machines and to build the factories that are used to manufacture building materials, while others omit these inputs.
Transportation costs vary widely. If a building site is located near a lumber mill or building component factory, energy inputs for transportation might be low. If the building site is remote, or if building materials are shipped to the U.S. from Germany, energy inputs for transportation might be high.
Recycling rates are hard to predict. If aluminum components are recycled, some of the components’ embodied energy can be recovered. The problem is that it’s hard for designers to predict whether aluminum components will be recycled when a building is demolished.
Published data are often out of date. In his 1995 Home Energy article (“Reducing the Embodied Energy of Buildings”), Tracy Mumma wrote, “Part of the challenge of assessing and making decisions based on embodied energy is the lack of current data. The definitive U.S. study on embodied energy was produced under the auspices of the Energy Research and Development Administration and dates from December 1976. Many of the statistics it includes are of 1967 vintage, and most current papers and references on embodied energy still cite data drawn from this old study. While some of the data may still be relevant, the tremendous advances in processing technology and recycling during the past 20 years limit the applicability of this information. Tools, transportation, and installation methods have changed, and most significantly, some building materials in widespread use today didn’t even exist at the time the report came out.”
What’s the bottom line? According to a web site maintained by the Australian government, “Estimates of embodied energy can vary by a factor of up to ten. As a result, figures quoted for embodied energy are broad guidelines only and should not be taken as correct.”
Other sources provide similar warnings:
- The U.S. Department of Energy notes, “Due to the complexity of calculations and the wide range of production methods, transportation distances and other variables for some building products, exact figures for embodied energy vary from study to study.”
- Tracy Mumma wrote, “The quantification of embodied energy for any particular material is an inexact science, requiring a long view look at the entire manufacturing and utilization process, and filled with a large number of potentially significant variables. Consequently, the complexity of embodied energy calculations is frustrating even for scientists, and it is easy for the individual homeowner, builder, designer or government specifier to become discouraged at the difficulty of obtaining accurate figures.”
Making your own embodied energy calculations
Fortunately, most green builders don’t need to calculate the embodied energy of every component of their buildings. It’s possible to use general guidelines for most materials — for example, the guideline that wood-framed walls have lower embodied energy than concrete walls — and to narrow one’s focus to insulation. It’s particularly important to focus on the incremental embodied energy and incremental operational energy savings associated with beyond-code levels of insulation.
After all, the most common embodied energy question is, “How much insulation is too much?”
There are several online sources of information on the embodied energy of insulation materials. Environmental Building News lists the following embodied energy values for R-20 insulation covering one square foot:
- Cellulose: 600 BTU
- Mineral wool: 2,980 BTU
- Fiberglass: 4,550 BTU
- Polyisocyanurate: 14,300 BTU
- EPS: 18,000 BTU
While Environmental Building News reports that the embodied energy of EPS is somewhat higher than that of polyiso, information compiled by Mike Eliason for his Brute Force Collaborative blog shows that the embodied energy of EPS is somewhat lower than that of polyiso. Let’s call it a tie. Eliason reports that the embodied energy of XPS is higher than either EPS or polyiso.
One source reports that the embodied energy of spray foam insulation is about the same as that of EPS.
Every source reports that the insulation material with the lowest embodied energy is cellulose.
Here’s the bottom line: if you are installing cellulose in an attic, pile it on. You don’t have to worry about the embodied energy in cellulose.
If you are building walls with EPS insulation (for example, ICF or SIP walls), or if you are installing EPS or XPS insulation under your slab — especially if you are choosing very high R-values, as might happen at a Passivhaus project — you may want to perform some embodied energy calculations to see if your investment in thick foam makes sense.
One example of such a calculation was posted by Mike Eliason as a comment to a GBA blog. Eliason wrote, “We’ve just calculated the numbers on an Ã¼ber-small Passivhaus (reference area = 577 square feet) we’re working on.” He compared the use of 6 inches of subslab EPS foam to 4 inches of subslab EPS foam, noting that the extra 2 inches of EPS would save 81 kWh (292 MJ) per year. At an electricity cost of 9 cents per kWh, the added insulation would save only $7.29 per year. The extra 2 inches of EPS would cost $325, and would have 2,706 MJ of embodied energy. The embodied energy payback for the extra foam would occur in 9.27 years (2,706 MJ divided by 292 MJ per year). The financial payback period would be much longer, however; it would stretch to 44.5 years ($325 divided by $7.29 per year).
Eliason’s calculations show how hard it is to justify the installation of 12 inches of subslab EPS.
What about building assemblies?
In the 1970s, the U.S. Department of Commerce published data on the embodied energy of building materials. Architect Bruce Coldham used that data to calculate the embodied energy of several different wall assemblies.
Coldham reported the data in BTU per square foot of wall assembly:
- A double-stud wall or Larsen truss wall with fiberglass insulation = 56,400 BTU/sf
- A fiberglass-insulated 2×6 wall with 2 inches of interior polyiso foam = 72,700 Btu/sf
- A structural insulated panel with 6 inches of EPS = 122,900 Btu/sf
- A concrete foundation wall insulated with 4 inches XPS = 141,500 Btu/sf
Coldhman notes, “In all the wood-framed wall sections, it is the insulation that comprises the bulk of the energy embodiment.”
A few simple rules of thumb
Here are a few rules of thumb that emerge from my examination of the world of embodied energy:
- In most cases, recycled building materials will have less embodied energy than new materials.
- If possible, choose wood framing over steel framing or concrete.
- The level of embodied energy in cellulose insulation is very low, so you can use it with abandon.
- Because foam insulation has a high level of embodied energy, it is hard to justify the installation of very thick layers of foam insulation.
If your construction project requires rigid foam insulation, consider using recycled rigid foam recovered from demolition sites or roofing jobs that require old insulation to be removed. When making embodied energy calculations for recycled materials, it’s fair to ignore the energy used to manufacture the materials and to calculate only the energy used to remove, transport, and reinstall the materials.
Martin Holladay’s previous blog: “Can Solar Power Solve the Coal Problem?”