To view the figures and tables associated with this article, please refer to the flipbook above.
Boulder, Colorado, is one North American city advancing circular economy practices. Aiming to reduce its emissions by 80% by 2050 and become “zero-waste,” Boulder passed Deconstruction Ordinance 8366 in 2020, which requires residential and commercial deconstruction projects to divert 75% of materials by weight from landfills, via recycling or reuse. This case study describes Boulder’s exemplary circular economy project, focusing on the deconstruction, stockpiling, and reuse of salvaged structural steel from the Boulder Community Hospital for Boulder’s new Fire Station 3.
A circular economy prioritizes sharing, leasing, reusing, repairing, and recycling existing materials and products (i.e. structural components) for as long as possible, reducing environmental impacts compared to the prevailing linear economy, which relies on a limitless supply of cheap, easily accessible materials and energy. Extending the lifecycle of construction products prevents waste to landfill, resource extraction, and embodied carbon emissions associated with manufacturing.
Often, buildings are demolished and their components landfilled— not because they are in deterioration but because they have lost their aesthetic or functional value. Many buildings are deemed too costly to convert to a new occupancy type, requiring major structural modifications and material replacement. The building may no longer have value, but what about its components?
Components such as structural steel beams and columns can be recovered through deconstruction, which is the intentional disassembly of a building or structure to prevent damage to components. Although a typical hot-rolled structural steel section contains 93% recycled content on average, 89% of the product’s manufacturing emissions are attributed to the melting and processing of the raw material, billet casting, and rolling to shape according to the American Institute of Steel Construction’s industry average Environmental Product Declaration (EPD). The emissions from high-energy processes to produce molten steel and form shapes are avoided by reusing components directly.
While some companies such as Rheaply, Building Ease, and All for Reuse have developed material exchange databases for non-structural salvaged components, North America lacks circular economy infrastructure and at-scale salvaged material markets. A few North American municipalities have reuse and landfill diversion legislation, mainly for non-structural materials or dimensional lumber. This policy landscape contrasts with Europe’s recent policies and expansion of commercial markets that enable structural component reuse in London, Belgium, and the Netherlands. The reuse of structural components is unique in that geometry, material properties, and structural integrity must be well documented or at least discoverable, to enable reuse in structural applications due to life-safety requirements. This adds burden to structural materials that is not present for non-structural materials.
Boulder Community Hospital
The city-owned Boulder Community Hospital (BCH), a roughly 250,000 square-foot building, was sustainably deconstructed in 2023. For this building, the goals were to achieve 90% waste diversion, prioritize reuse, and illustrate the potential for a circular economy and long-term material stockpile within the city. The project achieved 93.5% landfill diversion of all interior and exterior materials by weight (98% of the core and shell). This is understood to be the first major commercial building to be entirely deconstructed and the first structural steel stockpile of its kind and scale in North America.
KL&A Engineers & Builders’ Team Carbon was contracted to develop processes, requirements, and documentation for material management and administration to facilitate the deconstruction, stockpiling, and reuse of the steel in BCH. The process was developed in collaboration with Ameresco Inc. (general contractor), Colorado Cleanup Corporation (deconstruction contractor), and the City of Boulder. KL&A was in a unique position to approach this project because of its combined structural engineering, construction management, steel detailing, and embodied carbon expertise. Fundamentally, it was approached from the perspective of the end-user: new construction structural engineers.
What dimensional information is needed to implement salvaged steel? What physical and capacity information is needed for an engineer to incorporate salvaged steel? What legacy information from the original use is important? When does a new construction project need this information?
The general process phasing and flow is illustrated above.
As is typical for innovation in design and construction, significant obstacles were assessing risk, cost, and quantifying uncertainty. Four key logistical factors enabled the project to advance through this uncertainty:
- City of Boulder owned the deconstruction, stockpile, and new construction.
- The deconstruction and new construction schedules aligned.
- Stockpile laydown space was available for roughly 2 years.
- Pieces of salvaged steel were connected to end-uses prior to deconstruction.
584 wide-flange and tube members (HSS as they are known today) were successfully recovered and stockpiled, totaling 161 short tons. At the time of this writing, one new construction project, Boulder Fire Station 3, has successfully installed 89 salvaged members, nearly 25% of the inventory by weight. An additional 298 steel pieces have been claimed and, in some cases, retrieved by new projects, some owned by the city. There are 197 remaining pieces available for procurement, resting at the BCH site, which are intended to be used in the site redevelopment and other construction projects across the state. The stockpile is owned by the city and managed by KL&A.
Material Source
BCH was primarily comprised of cast-in-place concrete structural systems, constructed in 1957 with numerous additions and renovations through the early 1990s. Two areas were steel systems: Source A and Source B. Concrete components were conventionally demolished, however, the material was processed, sorted, crushed, and reused onsite as basement fill.
Source A represents the 1986 and 1989 additions, totaling 18,000 square-feet, consisting of three levels above grade, utilizing non-composite steel wide-flange beam and column framing, steel open-web bar joists, concrete on metal deck floors, and metal deck roof. The framing and deck were covered entirely in spray-applied fireproofing. The Source A structural and architectural drawings were available, which were utilized to create a digital inventory. The timing of the digital inventory is noteworthy, as this allowed the Fire Station 3 team to select individual pieces before deconstruction and early in their design process.
Source B represents the 1982 and 1989 single-story additions, totaling 28,000 square-feet. The structure is one level above grade, utilizing non-composite steel wide-flange beam and column framing, HSS columns, and metal roof deck. Like Source A, the framing and deck were covered entirely in spray-applied fireproofing. Architectural drawings were available but did not include any structural system, member, or material information.
All wide-flange and HSS sections were targeted for deconstruction. The steel bar joists were recycled because the probability of damage during deconstruction was high due to their light shapes and welded seat connections, although steel joist reuse has been demonstrated in projects by KL&A and others. Many steel bar joists were originally optimized for localized loading conditions and are therefore inappropriate for general structural reuse, requiring reverse engineering. Steel members that have experienced cyclical loading, plastic deformation, or other unique loading conditions may not be suitable for structural reuse. Component characteristics and installed conditions affect the ease and challenges of recovery and reuse. A member may be easy to recover and lack reuse application. Inversely, a member may be challenging to recover but have high value.
Deconstruction
KL&A authored a Steel Deconstruction Specification that detailed cut locations, level of cleanup, and piece-marking requirements. It defined the resulting condition of the steel pieces which allowed new construction projects to bid the fabrication scope appropriately. The purpose of defining cut locations was to balance the ease of deconstruction with recovering as much of the members’ length as possible. It was anticipated that 12 inches would be removed from each end at some point during the process. An average of 27 inches total was removed from the original length of Source A pieces.
The deconstruction team recovered 30 beams per day. The “flying time” of the Source A beams was 12 minutes, from removal to setting on the ground. Source B recovery was much faster, primarily because it was a single-story structure, and the team was familiar with having completed Source A. It is estimated that fewer than 30 pieces (5%) were damaged and therefore unsuitable for structural reuse. These pieces were either recycled, used for material testing then recycled, reserved for non-structural applications, or used as stockpile dunnage.
At the time of this writing, the BCH deconstruction costs are still being analyzed and cannot be reported in detail. The total cost of deconstruction (interior plus core and shell) was $9.2 million, compared to an estimated demolition cost of $7.7 million, a $1.5 million (19%) premium. The floor area of Source A and B was roughly 18% of BCH.
Component Processing
The order of operations varied between the two sources, primarily due to the availability of structural drawings for Source A and lack thereof for B. For example, fireproofing was removed from Source A steel, and each component was piece-marked to match the digital inventory before deconstruction. However, Source B fireproofing was removed in situ, further cleaned off while on the ground, then piece-marked, and then inventoried.
Member cleanup consisted of removing fireproofing and accessory material (plates, appendages, etc.). 90% of fireproofing, within any 6 square-inches of surface area, was removed onsite. This is not considered unique to deconstruction versus demolition, as metals recycling facilities typically do not accept steel with fireproofing, or they accept it at a lower value because it must be removed eventually.
All accessory material that extended beyond the boundary of the member’s cross-sectional area was removed onsite, except for material that extended a significant length along the member, such as welded deck-edge angle. The purpose of cleanup was to ease handling during recovery, vertical stacking at the stockpile, transportation, and processing by fabrication equipment which often uses rolling conveyors. A stockpile endeavor should balance cleanup effort against the intended end-use, time, cost-sharing, and ease of handling. New projects may be unconcerned about extraneous material or even decide to highlight those features as part of their aesthetic and project brand.
A physical and digital inventory was created to document, track, and identify individual pieces of steel. The digital inventory housed component characteristics, as well as logistical information like the specific location in the physical stockpile, if it was to be sampled for material testing, and claims by new projects. The inventory tracked total tonnage, embodied carbon savings, and estimated cost compared to standard new steel.
Measurements and photos were taken onsite, and descriptive information was recorded, such as general condition (finish, corrosion), accessory material and holes (connections, penetrations), observed damage (excessive bend, missing material), and observed geometry (camber, sweep, tilt). This was used to develop individual cut sheets, assign historic AISC Manual of Steel Construction shape categories (e.g. W14x22), and flag any noteworthy conditions for the end-user.
Material Testing
There is no industry standard or code requirements for testing and validating salvaged steel for structural reuse in North America, although protocols have been developed for other regions. KL&A developed a testing protocol, based on recommendations from ASCE 41-13: Seismic Evaluation and Retrofit of Existing Buildings and the Steel Construction Institute’s Structural Steel Reuse: Assessment, Testing and Design Principles.
Samples were taken and tested to estimate the properties of size categories within each source rather than from every piece. ASTM A370 tension testing was performed on 10% of the pieces in each size category and ASTM A751 chemical analysis testing, to verify weldability, was performed on 10% of the total quantity of pieces. The third-party laboratory required 1×8” samples and cost about $100 per sample. Additional testing may be necessary based on the source and end-use.
Based on the dates of construction and available documentation, it was anticipated and validated by testing that the wide-flange members are equivalent to ASTM A36 and tube members equivalent to ASTM A500 Gr. B. All the samples confirmed weldability, with some results recommending preheating procedures.
Reuse in New Construction
The financial feasibility of deconstruction is directly tied to connecting an end-use of the recovered material, whether it be specific or a general resale market opportunity. Understanding the ability of Fire Station 3 to utilize pieces from BCH was critical to moving forward with recovery.
The uncertainty of logistics, cost, and risk is a significant barrier to innovation. The cited motivations of projects that engaged with the stockpile were sustainability goals (embodied carbon reduction), appetite for innovation, project brand differentiation, and material cost savings. It was evident that the reliability of member sizes, geometry, and material testing results of salvaged structural components are highly desirable to new projects before design implementation.
The Fire Station 3 structural design team claimed pieces from the inventory during Design Development to incorporate into roof framing over the apparatus bay and mechanical screen framing atop the roof. The sizes and anticipated strength were incorporated into the analysis model. The structural Construction Documents noted in plan the locations for the intended use of the steel. Full Metal Iron retrieved and transported the pieces from the BCH stockpile to their fabrication shop. After a final cleaning, the fabrication process was reported to be seamless and like that of new steel. It was successfully installed in 2023. There were no reported differences in fabrication or installation costs and schedules, and the steel was procured at zero cost to Fire Station 3. This resulted in net cost savings of 0.5% of the total steel contract because the material savings outweighed the fabricator’s cost to transport and clean the steel.
In the future, policy requirements and consumer demand will be strong motivations to consider salvaged structural materials. It is reasonable to speculate that the use of salvaged steel in new construction can be cost-competitive, even when considering a resale value.
Conclusion
This project illustrates that at-scale deconstruction and reuse of structural steel components is possible and financially feasible. The necessary change in behavior, priorities, and incentives is more challenging to overcome than the technical and logistical aspects. To advance structural material recovery and reuse, several topics arise – policy and incentives, protocol and code development, Environmental Product Declaration development, collaboration among building stock and new construction, specialty business opportunities, technology opportunities, and design for deconstruction (DfD). Through concerted efforts, strategic initiatives, collaboration, and reinvention the construction industry can pave the way for circular economy, embodied carbon reductions, and environmental stewardship.