Alternative Cementitious Materials: An Evolution or Revolution?
Taking new approaches to using ACMs will help to develop durable next-generation infrastructure as economically as possible.
The Nation's deteriorating infrastructure requires an estimated investment of $2 trillion, according to the American Society of Civil Engineers. With close to $100 trillion needed to address global
infrastructure demands, a revolutionary approach to infrastructure design, construction, and maintenance is critical. Because of its wide availability at low cost and its use in a vast range of infrastructure applications—including roadways and bridges, pipes and spillways, nuclear power structures and dams, among others—concrete will play a central role in meeting the infrastructure needs for future generations.
The need for infrastructure investment represents an enormous opportunity to rethink concrete design and construction. Will this be an evolution or a revolution?
Increasingly, non-portland cements, or alternative cementitious materials (ACMs), are of interest because of their potential to reduce the environmental footprint of concrete and their potentially advantageous special properties. Traditionally, transportation engineers largely used ACMs in specialty applications that take advantage of the materials' unique performance characteristics. Examples include use in small-scale repairs where rapid set and strength development are needed, for winter construction where high heat of hydration is advantageous, and in industrial settings where chemical resistance is required. The special properties of ACMs, which vary depending on composition, can also facilitate new construction methods and significantly increase service life, achieving enhanced durability under a range of aggressive environments.
"It's important for the transportation industry to examine the evolution of ACMs from materials used in niche applications to materials that can be scaled up and used as alternatives to traditional portland cement concrete," says Richard Meininger, a research highway engineer with the Federal Highway Administration's Pavement Materials Team in the Office of Infrastructure Research and Development. "ACMs also have potential for some revolutionary applications—applications that go beyond conventional construction. These opportunities are ripe for advancement of ACM usage in next-generation infrastructure."
What are ACMs?
Agreement on a comprehensive definition for binders used as alternatives to portland cement has been an evolutionary process itself. In defining "alternative cementitious materials," some of the questions the technical community tackled included: Should this term encompass materials that are produced without any heat-treatment or without calcination? Should it include binders that do not include hydrated minerals among their reaction products, as conventional cements do? Should the term encompass blended cements that contain portland cement as a component, along with other mineral phases not typically found in portland cement? What about blends of phases found in portland cement, but at different proportions?
|FHWA performed both laboratory and field studies on ACMs. These images are from the petrographic analysis of the test specimens, which aimed to identify any internal damage, discoloration, freeze/thaw action, or other modes of deterioration.|
An American Concrete Institute (ACI) committee on ACMs has determined the answer to each of these questions is "yes." ACI now broadly defines an alternative cement as "an inorganic cement that can be used as a complete replacement for portland or blended hydraulic cements, and that is not covered by applicable specifications for portland or blended hydraulic cements."
By the ACI definition, ACMs include: (1) clinkered materials, such as calcium aluminate cement (CAC) and calcium sulfoaluminate cement (CSA), which are thermally processed similarly to conventional portland cement; (2) calcined materials, such as magnesium phosphate cement (MPC) and magnesium oxychloride cement (MO), which require lower production temperatures that do not result in new phase formation during manufacture; and (3) non-clinkered materials, such as alkali- or chemically-activated silicates or aluminosilicates (AA), including geopolymers.
Chemical Compositions And Reactions
ACMs encompass a broad range of compositions and can include materials and mineral phases not found in conventional portland cement, as well as those containing typical portland cement phases at different relative quantities. A phase is a homogeneous substance that has a fixed composition and uniform chemical and physical properties. For example, a recent Exploratory Advanced Research (EAR) project funded by FHWA examined phase compositions for six commercially available ACMs, among others, and compared them to ordinary portland cement (OPC). Some key research results for ACM concrete produced from a subset of these cements are as follows.
In general, like portland cement, clinkered and calcined ACMs react with mix water over time to form various products that lead to set and ultimately bind the aggregates together. While OPC is dominated by tricalcium silicate (C3S) with more minor amounts of calcium aluminate phases and hydrates (reacts with water) to form calcium silicate hydrates (C-S-H) as the primary binding phase, ACM products fluctuate considerably because of their varying composition.
Generally, CACs are mainly calcium aluminate (CA) and react to form hydrated calcium aluminate phases. CACs can also be combined with OPC, typically along with some calcium sulfate, in a ternary blended cement.
CSA cements contain relatively large fractions of ye'elemite (C4A3S), along with phases also found in OPC, including belite (β–C2S) and calcium sulfates. Like portland cements, CSA cements react to form C-S-H and ettringite, but the relative amount of ettringite formed in CSAs is considerably greater and the reaction to form C-S-H occurs more slowly. This is why researchers found unhydrated belite persisting in the reacted cement paste microstructure at 56 days of moist curing.
Finally, MPC can have a wide range of magnesium-rich compositions. The MPC in the EAR study was dominated by magnesium oxide and potassium phosphate, but also contained Wollastonite (CaSiO3). That MPC reacted very rapidly to form potassium-struvite (Κ-struvite), a hydrated mineral phase (KPM2H12).
In activated systems, the mix water conveys ions that promote dissolution of minerals and reprecipitation of products. The products of the reaction may or may not be hydrated. ACMs that rely on chemical or alkali activation, rather than thermal processing, are typically largely composed of reactive silicates or aluminosilicate phases (for example, fly ash, ground glass, calcined clay, and slag) that are finely divided and contain glassy or amorphous phases. These precursor materials vary greatly in composition and may or may not contain calcium. In addition, many precursors are byproduct materials, which contribute to the sustainability of activated ACM concrete. In one example from the EAR study, a precursor based on Class C fly ash, a coal-combustion byproduct, was reacted to form sodium-containing aluminosiliceous gels (N-A-S-H) with some containing calcium (C-N-A-S-H) derived from the fly ash. These products bind aggregate together and provide strength and impermeability in activated systems, which is similar to the way C-S-H functions in OPC.
|Calculated phase compositions for ordinary portland cement (OPC) compared with six commercially available ACMs: two calcium aluminate cements (CAC), one ternary blend containing CAC, two calcium sulfoaluminate (CSA) cements, and one magnesium phosphate cement (MPC). The ACM compositions characterized here are examples and do not represent in a general way the compositions of the entire class of that type of ACM or ACMs in general.|
In all ACMs, as in OPC, reactions occur over time and contribute to setting, strength development, and impermeability. The reaction rate depends on ACM fineness, composition, and environmental conditions. The products formed and resulting microstructure depend on these parameters, as well as the water-to-binder ratio. Appropriate water-to-binder ratio to achieve desired performance varies by material. Because the specific gravity of ACMs may be lower or higher than OPC—a range of 2.53 to 3.13 was measured—selecting an appropriate water-to-binder ratio requires strong consideration. Mix proportions by volume may be more appropriate for ACM concrete.
|These micrographs of four commercially produced ACMs show microstructures for the unreacted powders (left) and pastes after 56 days of reaction (right) for (a) a calcium aluminate cement, (b) a calcium sulfoaluminate cement, (c) a magnesium phosphate cement, and (d) a chemically activated aluminosilicate (fly ash). Notations show common phases and oxides identified in each material. The magnification varies from 250 times to 600 times.|
For more details on the chemical composition and reaction of the various common ACMs, see www.fhwa.dot.gov/publications/research/ear/16017/index.cfm or www.fhwa.dot.gov/publications/research/ear/16017/16017.pdf.
Contributions to Sustainable Construction
Many ACMs have an advantage over OPC with respect to their environmental footprint because of their variances in composition and processing. For example, manufacturing CSA and CAC cement clinker can result in perhaps 15 to 50 percent reductions in carbon dioxide (CO2) emissions, compared with OPC, based on the differences in their composition. The reductions derive from decreases in the relative amount of calcium carbonate feedstock and from lower temperatures during clinkering or calcining, or the avoidance of thermal processing.
Blending with other less energy-intensive mineral phases (for example, calcium sulfates, or limestone)—such as in CSA or ternary blends of CAC, OPC, and calcium sulfate (CACT)—may result in further reductions in embodied CO2 because of dilution. Other ACM formulations, such as geopolymers or other activated aluminosilicates, do not require calcination. As a result, the embodied CO2 in these systems can be quite variable, even among ACM classes. The embodied CO2 for these mixtures is primarily a function of the activating solution, which varies considerably based on type and quantity of aluminosilicate precursors (and whether they include byproduct or recycled materials).
Additional opportunities exist to further enhance the sustainability of next generation infrastructure through the use and development of ACMs. For example, advances in the manufacturing and composition of ACMs can lead to further savings in greenhouse gas emissions and energy required to produce ACMs. More immediately, increased strength can result in decreased section size, with a potential for decreased materials usage. In addition, increased durability further contributes to enhanced sustainability and can help offset initially higher materials costs through longer service life.
The special properties of ACM compositions vary considerably by material. Rapid setting, rapid strength development, higher ultimate strength, improved dimensional stability, reduced heat of hydration, and increased durability in aggressive environments are some of the special properties found in ACMs.
For example, alkali-activated materials are known for their thermal stability and fire resistance, and MPCs exhibit rapid set and high early strength. Some CAC formulations are known for their superior resistance to acid, sulfate attack, and mechanical abrasion. Some CSA systems are associated with improved dimensional stability and resistance to freeze-thaw and sulfate attack. Other ACMs, such as some CSAs rich in belite, may exhibit substantially slower strength development than OPC. However, some ACMs may not possess adequate corrosion, carbonation, or sulfate resistance, and so users must be aware of the associated strengths and weaknesses of each cement in addition to the needs of the project at hand.
FHWA expects to publish a project report that will provide substantial new information on ACM performance later in 2019. The information will aid materials and transportation engineers in the selection of appropriate ACMs for specific applications.
CO2 Emissions for ACM and Other Binder Technology Relative to OPC
|Cement or Binder||Relative CO2 Emissions||Source|
|Portland Cements||1||Scrivener, 2004|
|Calcium Sulfoaluminate Cements||0.5–0.85||Roux, 2013 and Chen, 2012|
|Calcium Aluminate Cements||0.53||Bizzozero, 2014|
|Magnesium Phosphate Cement*||0.55||Wagh, 2013|
|Activated Aluminosilicate Binders**||0.44–0.64||Mclellan, 2011|
*High emissions associated with dead-burned magnesia products result from the very high calcination temperatures required (2800°C), as well as the large quantity of carbon dioxide resulting from chemical transformation of MgCO3 precursor material into MgO. However, magnesia cements derived from other raw materials sources may result in reductions in carbon dioxide emissions, even relative to portland cement.
**Emissions result from chemical activator manufacture and transportation only.
In a survey conducted by the American Association of State Highway and Transportation Officials in 2014, about half of the State departments of transportation (DOTs) responding reported they had field experience using ACMs. CSA cements were the ACM most commonly used, as reported, and applications primarily were for repair as patches, closure pours, joint repairs, partial to full-depth concrete pavement repair, and overlays. These applications take advantage of the rapid set and shrinkage compensation effects associated with many CSA formulations.
According to the same survey, CACs were the next most commonly used ACMs. Example CAC applications included full-depth pavement replacement and joint replacement—applications facilitated by CAC's rapid strength development. However, States considering using CAC for high-strength applications should exercise caution. Unlike OPC, which gains strength over time, CAC early strength can be decreased by half. This significant strength loss is because of conversion of original metastable hydration products to more stable hydration products, leading to an increase in porosity and permeability.
Some State DOTs also reported using MPCs and AAs, primarily for repair because of their rapid set characteristics (that is, their ability to set more rapidly than OPC). MPCs are among the most expensive of the ACMs, and therefore, some States reported a preference for other ACMs in applications where feasible.
Generally, most DOTs responding to the survey reported that their ACM repairs were performing well. For example, full-depth CSA concrete pavements placed in the Los Angeles, CA, area showed only minimal abrasion and joint spalling after 17 years in service. However, some DOTs noted cracking and scaling issues in some locations. Because the DOTs primarily used the materials in repair applications, it is difficult to assess if the observed distresses are related to the material itself or to underlying design, overloading, site, or environmental issues.
More recently, ACM use in transportation infrastructure has evolved, beyond the relatively small-scale and repair applications most common in recent decades. For example, researchers from the University of Arkansas and University of Oklahoma have cast structural prestressed concrete sections with CSA cement. The cement's rapid early strength development makes earlier prestress release possible, providing time savings that could translate into faster bed turnover. Improved dimensional stability with CSA cements can also benefit prestressed concrete construction by minimizing prestressing losses associated with shrinkage in OPC.
|The California Department of Transportation (Caltrans) placed ACM pavement on California State Route 60W to the State Route 71S interchange in 1997. After 17 years of continual heavy traffic loads, the ACM pavement showed good performance. Caltrans noted some surface wear and minor joint spalling in some sections.|
Others have explored ACM use in ultra-high performance concrete (UHPC). CAC in UHPC is envisioned for applications requiring high temperature resistance. As an example, foamed geopolymer concrete has been developed in the lab and deployed in the field using similar technology as used for foaming OPC concrete. The highly aerated geopolymer provides good fire resistance and acoustic properties with the potential for operational energy savings and a lower environmental footprint compared to conventional foamed concrete.
Special properties found in ACMs also have facilitated exploration of more revolutionary applications. For example, ACMs with advantageous rheological properties (such as high viscosity, high-yield stress, and shear thickening), rapid set, and rapid strength development may be well-suited for emerging construction methods such as three-dimensional (3D) printing. CSAs, CACs, and geopolymers have each been used in 3D printing demonstrations in concrete and mortars. Some MPCs show good biocompatibility, and researchers are exploring 3D printing approaches to develop synthetic bone scaffolds with them.
Shortages and increasing variability in fly ash supplies may require alternative means for enhancing durability in new construction. In particular, additional means are needed to guard against alkali-silica reaction (ASR) and ACMs may play an important role. Despite containing substantial amounts of alkali as reactants, geopolymers have demonstrated resistance to ASR. To mitigate damage in an existing ASR-affected bridge deck in Alaska, contractors placed a 0.75-inch (1.9-centimeter) MPC overlay on top of the damaged concrete. Because of the very rapid—almost instantaneous—set associated with MPC, the contractors employed a specialized pumping and mixing setup for the installation.
In addition, ACMs may play a role in meeting infrastructure needs as humans become a multiplanet species. The National Aeronautics and Space Administration (NASA) has already funded research, performed by astronauts aboard the International Space Station, exploring how OPC hydration is affected in microgravity. NASA recently concluded a nationwide 3D printed habitat challenge, inspiring university-based teams to conceptualize and demonstrate concrete construction on other planets. Many of the teams formulated their concrete compositions to make use of the available extraterrestrial minerals, including silicon-rich basalt, magnesia, and alkalis common in the Martian crust.
ACMs with Varying Dosages of Set Modifiers
|Cement or Binder||Retarder (dosage)||Setting Time (minutes)|
|Calcium Sulfoaluminate Cement||-||16||28|
|Citric acid (0.5%)||67||125|
|Calcium Aluminate Cement||-||456||550|
|Magnesium Phosphate Cement||-||11||15|
|Sodium borate (14%)||60||105|
Researchers added retarding admixtures, including citric acid and sodium borate, to slow hydration in CSA and MPC formulations to meet the desired 1-hour set time. CAC required no retardation.
The recent FHWA EAR study explored more "earthbound" expanded applications for ACMs. In this project, researchers comprehensively assessed and compared commercially available CSA, CAC, MPC, and AA to OPC. The researchers compared the materials in terms of their early age behavior, mechanical properties, and durability. The objective of the project was the upscaling of ACM concrete for transportation infrastructure applications, particularly where construction could be possible using conventional batching, transportation, placement, and finishing methods to minimize the need for additional capital investments.
|This graph shows the compressive strength development in CAC, CSA, and AA concrete mixtures compared to OPC.|
Performance goals for more than a dozen commercial ACMs that researchers initially considered included: (1) at least 1 hour working time before set, (2) maintaining 3-inch (7.6-centimeter) slump after 1 hour of mixing, and (3) a 3500 psi minimum compressive strength at 7 days and at least 5000 psi at 28 days of age.
Extending set to at least 1 hour proved challenging for some formulations. Extending set time is important for use of conventional, rather than volumetric, mix trucks and for transportation of fresh concrete from batch facilities to job sites. Some ACMs, such as some of the CACs considered, set after 1 hour, without the need for set-retarding admixtures. For other ACMs, with appropriate admixture use, researchers were able to achieve the set and slump retention goals for CSA, AA, and MPC materials. All ACM concrete, with the exception of MPC, achieved the target slump. Overall, the MPC required substantial intervention to slow its very rapid hydration kinetics, making it challenging to achieve all performance metrics with the same mixture design.
Rate of compressive strength development varied considerably among the materials examined, but all six materials that met the workability requirements also met the required strength targets at both 7 and 28 days. CAC achieved the highest early strength, as well as highest strength until 28 days. The slightly lower CAC strength at 56 days signals the onset of the conversion reaction, which will lead to substantial strength loss over time in CAC. Conversion occurs when the initially formed hydration products revert to more stable, more dense phases, leading to increases in porosity, which is detrimental to strength and increases permeability. The AA composition was most similar to the OPC in strength development, both in terms of rate and in ultimate strength. CSA formulations showed the greatest variability, likely because of the varied proportions of ye'elemite, belite, and calcium sulfate among the CSAs tested.
Future Outlook for ACM Usage
Magnesia-based ACMs, such as Sorel cements, have existed since the mid-1800s. But, advances in low-energy magnesia-based cement production are among the most promising developing ACM technologies. The dichotomy between the historical use of ACMs and the potentially "game-changing" innovative uses demonstrate the intrinsic conflict for growing utilization of all ACMs. That is, while ACMs are familiar, they are also ever-changing, with new formulations facilitating advances in rheology and mechanical properties, in durability, and in sustainability.
The technological advances increase the value for ACM usage in a broader range of applications, including those involving emerging construction methods and in extremely aggressive environments. These advances then facilitate new applications—even potentially in extraterrestrial environments.
What is holding back the broader acceptance and application of ACMs? For conventional infrastructure applications, ACM usage is limited because it is not well-addressed in existing standardized tests or in design guidelines, standards, specifications, or codes. Until ACMs are addressed by these means, the risk associated with their use will limit their application to conventional small-scale applications and to demonstration and pilot projects.
"The growing acceptance of performance-based specification is one avenue for broader acceptance and utilization of ACMs," says Gina Ahlstrom, leader of the Pavement Materials Team in FHWA's Office of Infrastructure, Office of Preconstruction, Construction, and Pavements. "With the information gathered from FHWA's EAR project and ACI's ACM committee and ASTM International's new subcommittees addressing ACM usage, there's promise that new ACM technology will increasingly find a more facile path to applications in infrastructure construction."
|Researchers tested the ACM concrete slump for these seven specimens, which include OPC, two CSAs, CACT, two CACs, and MPC.|
Richard Meininger, Pavement Materials Team, Office of Infrastructure R&D, is the FHWA point of contact for this research. For more on ACM, contact Richard.Meininger@dot.gov or 202–493–3191.
Kimberly Kurtis is a professor in Georgia Tech's School of Civil and Environmental Engineering and School of Materials Science and Engineering. Kurtis leads the FHWA EAR project described in this article.
Prasanth Alapati is a Ph.D. candidate in the School of Civil and Environmental Engineering at Georgia Tech.
Lisa Burris, Ph.D., is an assistant professor in the Department of Civil, Environmental, and Geodetic Engineering at Ohio State University.
The authors gratefully acknowledge the contributions of EAR team members Tyler Ley at Oklahoma State University, Robert Moser at the U.S. Army Engineer Research and Development Center, and Neal Berke at Tourney Consulting Group, along with their teams.
Please contact Richard Meininger or Kimberly Kurtis for a list of available publications, including reports, executive summaries, and technical papers. Kimberly Kurtis can be reached at firstname.lastname@example.org.