With greater demands on our planet’s resources, there is a need for companies, institutions, and their design partners to rethink their design approach toward the environmental footprint of buildings. It is important to focus on greenhouse gas (GHG) emissions from laboratory buildings and how program selection can affect the amount of carbon produced by these facilities.
Before we begin to detail how programs can affect GHG emissions, it is important to contextualize the issue. For the past 15+ years, the design community has been making significant efforts to cut energy consumption and emissions. A critical part of this push has been the Architecture 2030 Initiative or Challenge, issued by the American Institute of Architects and internationally recognized architect Edward Mazria, FAIA, Hon. FRAIC. Both provide a roadmap to reaching carbon neutral facilities by the year 2030. The challenge sets targeted goals for stepped reduction in five-year increments.
Owners, engineers, and architects have been making considerable efforts to meet this goal. Through 2018, the AIA reports that the industry has reduced 17.8 million metric tons of carbon per year, essentially the equivalent offset from 21 million acres of forest. When placed in context, however, this is the equivalent of only 2.4 percent of the US Direct GHG emissions (per the EPA and using an average of 750 million metric tons of carbon between 1990 and 2017). More so, the reduction does not necessarily lead to a reduction in the total amount of GHG emissions, since this standard is tracking additions or renovations to the total building inventory and does not address existing operational facilities.
If curbing GHG emissions is the goal, then a plausible approach is to try to achieve zero net energy (ZNE). This approach has a slightly different definition depending on the certifying body (e.g. New Buildings Institute, International Living Future Institute, and LEED, to name a few). They all agree, however, that a ZNE building is an extremely efficient building that produces through renewable energy sources (ideally non-combustible) as much energy as it consumes. To achieve this goal, architects and engineers often have to set a target for energy use intensity (EUI) in the 25-35 kbtus/SF range in the US, dependent upon climate zone. This target has been found to be the most applicable range that can be accommodated for on-site energy offsets.
The question, therefore, becomes: how do lab buildings operate relative to this target? One organization that has been helping better understand laboratory facilities within the context of sustainability is the International Institute for Sustainable Laboratories (I2SL). As part of its efforts, I2SL has worked with architects and engineers to benchmark nearly 800 buildings. The takeaway is that the median EUI for lab facilities is 535 kbtu/sq. ft., or nearly 18 times higher than the zero net energy goal. Only one facility had an EUI nearing 50.
The drivers for this high energy output are the mechanical strategies needed to meet the code mandated minimums within these facilities. ASHRAE and NFPA 30 both dictate that a minimum of one cubic feet per minute of outdoor air or six air changes per hour (ACH) are provided within laboratory spaces when occupied. Additionally, certain spaces require a considerable amount of makeup air to support the equipment, such as fume hoods and snorkels, within spaces often doubling the amount of needed air. Other space types may have to meet standards from governmental bodies including the National Academy of Sciences whose guidelines are used for accreditation purposes by organizations such as AAALAC. In the case of the vivarium, the standard dictates 10-15 ACH of constant volume outdoor air to maintain and ensure both macro and micro environmental air quality.
Quantifying energy usage
As noted above, specific programs can have a profound impact on energy due to the spaces' respective ventilation requirements. To better understand this impact, Perkins and Will and Arup decided to take a more nuanced look at the specific EUI associated with programmatic space. This required a significantly different energy modeling approach that studied a building in zones, rather than on the whole. The team started with identifying a target programmatic epidemiology consisting of not only laboratory, core, and support spaces, but also associated spaces often found within academic buildings such as an auditorium and classrooms.
Additionally, the team did not want to start with the baseline building suggested within ASHRAE. Instead, they decided to model a facility that used a number of progressive energy-saving strategies often found within the most efficient facilities. These included chilled beams, an energy efficient envelope, enthalpy wheels/heat recovery, a 30 percent reduction in lighting power density, a reduced plug load dependent on program, high efficiency chillers and boilers, and an occupancy schedule that assumed a nighttime turn down to two ACH within laboratory spaces. We found that the EUI for these ventilation-intensive spaces ranged anywhere from four to 16 times greater than a typical office space.
Additionally, the team decided to study the effect of additional conservation measures. They analyzed the benefit of a sensor-driven air sampling system such as Aircuity, programmatic redistribution by grouping lab write up spaces with office space, a 40 percent overall reduction in glazing versus the initial model, a gravity relief exhaust system or cascade air for non-lab spaces, and external shading optimized to orientation. The modeling showed that the solution with the best benefit was found with the air sampling system, but not nearly close enough to fall within the 25-30 EUI range noted for ZNE potential.
Shifting planning paradigms
Our research has shown the importance of working with our client partners to clearly identify space needs that can accommodate their research goals when trying to effectively manage energy. As part of that conversation, more clients are beginning to consider computational workflows supported by a reduced amount of bench science. It is important to note that computational workflows can often have an unforeseen impact due to the energy intensity of the data centers supporting the high-powered computing. Right sizing the data centers, knowing when to buy time at remote locations—such as IBM’s Deep Computing Capacity on Demand Centers or one of the NSF's sites such as PRObE or Stampede 2—will be critical to help manage and accurately assess energy demand.
Within the context of energy, we are starting to see a shift toward electrification of buildings. This allows for the building to be fully served by the grid, but requires a different approach to the building’s infrastructure by transitioning away from combustion reliant systems, such as steam to hot water produced through heat pumps. By moving to the grid, a building is as carbon neutral as the source energy feeding it. To help better understand that potential impact, our team used EIA conversion values to study the GHG emissions for specific climate zones.
Carbon emissions varied greatly across the studied zones. For instance, the EUI for an open chemistry laboratory in Boston was 236, which was higher than what was found in Denver where we modeled 227. When we compare metric tons of carbon, however, we found that the emissions were 37 percent higher in Denver than Boston.
The data has shown the importance of a right-sizing program to help minimize GHG emissions. Of equal importance is understanding the larger ecosystem supporting the function of the buildings. Having a comprehensive approach—whether toward supporting infrastructure or indirect emissions from electricity production—is critical for effectively curbing the emissions footprint of our laboratory facilities.