## Semester

Fall

## Date of Graduation

2021

## Document Type

Dissertation

## Degree Type

PhD

## College

Statler College of Engineering and Mineral Resources

## Department

Civil and Environmental Engineering

## Committee Chair

Roger H.L. Chen

## Committee Member

P.V. Vijay

## Committee Member

Karl E. Barth

## Committee Member

Fei Dai

## Committee Member

Hailin Li

## Abstract

In large concrete structures, due to concrete’s low thermal conductivity, the interior temperature rise can approach the concrete’s adiabatic temperature rise. As the concrete’s surface losses heat to the environment and cools rapidly, the interior of the concrete attempts to expand while the exterior provides an internal restraint. The phenomenon causes thermal stresses at the concrete’s surface. If the thermal stresses exceed the concrete’s tensile strength, there is a high probability of cracking. The cracks usually appear during the first few days after construction while the concrete’s tensile strength is low. These early-age cracks can reduce the service life of the structure by allowing the entry of detrimental chemicals such as deicer chemicals and chloride salts that corrode the concrete’s steel reinforcement. The risk of early-age cracking can be alleviated by lessening the temperature gradients inside the concrete structure. This can be accomplished by reducing the cement quantity in the mix design, reducing the placement temperature, adding supplementary cementitious materials, or using insulation blankets. Structures that are at risk of experiencing thermal cracks are usually termed as “mass concrete”. However, the definition of mass concrete is non-uniform and varies from state to state. Thus, a uniform definition is needed. In this study, a methodology to estimate the early-age tensile stresses was developed and used to create definition tables for three common concrete pier stems.

The mechanical and thermal properties of two thermally friendly “Class M” mix designs with supplemental cementitious materials (using either 50% replacement of the cement with ground granulated blast furnace slag (GGBFS) or 30% replacement of the cement with Class F fly ash) were experimentally measured and incorporated into a finite-element analysis. These properties include heat of hydration, adiabatic temperature rise, activation energy, thermal conductivity, static elastic modulus, compressive and tensile strength, creep, and thermal expansion coefficient. A methodology to incorporate the time-dependent material properties of concrete containing GGBFS and fly ash was developed, and a viscoelastic analysis was used to estimate the early-age thermal stresses. A two-term exponential degree of hydration was proposed to better capture the hydration behavior of cement binders with GGBFS and Class F fly ash. The heat rate was derived for the two-term degree of hydration function and incorporated into the finite-element model. The derived heat rate was found to model the hydration of blended binders better than those found in literature. The effect of creep was considered in a viscoelastic analysis where the creep strain in different directions was dependent on the loading direction and magnitude while other studies use an overall element stiffness in a linear elastic static analysis. Thus, the viscoelastic analysis can estimate the stress in displacement induced load and unloading cycles such as non-uniform temperature deformations caused by temperature gradients. The creep increment was derived using a strain super-position method and incorporated into a viscoelastic analysis. The time-dependent material properties were implemented in the commercial finite-element software ABAQUS using user-subroutines. Thus, the concrete’s early-age viscoelastic behavior could be accurately modeled and enabled a better estimation of the early-age stresses. The finite-element analysis was then compared to temperature loggers and strain gages embedded in 4-ft concrete cubes and was found to match well with the experimental measurements.

Local concrete suppliers within West Virginia were selected to produce and deliver 12-yd^{3} to 15-yd^{3} of Class M concrete to check the consistency and production feasibility of Class M concrete. Three districts were selected by the WVDOT where suppliers delivered their own Class M mix to test the concrete’s workability and uniformity. It was found that all the delivered batches met the requirements for slump and air content, however for two districts, D-10 and D-5, the delivered w/cm ratio was significantly higher than expected. The higher w/cm ratio led to low early-age strength. The findings of the feasibility study were used to determine the water-content of the Class M concrete.

The finite-element model was used to develop mass definition tables for three different pier stem geometries with steel formwork, no extra insulation, and a 30 °F daily ambient temperature drop. Mass definition tables for pier stems without insulation were constructed for two daily ambient temperature conditions (Case 1: 60 °F to 90 °F and Case 2: 30 °F to 60 °F). Circular, rectangular, and square pier-stem geometries were evaluated from 2-ft to 9-ft in 0.5-ft increments. The concrete pier-stems were considered non-mass if the predicted tensile stresses did not exceed 80% of the estimated tensile strength. Thus, the definition for mass concrete was uniform. The results were tabulated in “green/red” tables. It was found that the reduction in total cement content and replacement with supplemental cementitious material (using 50% GGBFS or 30% Class F fly ash as Class M concrete) reduced the heat, and the high early-age viscoelastic behavior of these mixes lowered the early-age tensile stresses compared to a typical ordinary Portland cement (OPC) mix design. Furthermore, a 1-in foam board insulation layer (R = 5) was added on the steel formwork to reduce the thermal gradient inside the concrete elements. The insulation layer was assumed to be removed at 10-days in the “worst-case” environmental condition (30 °F to 60 °F). FEM analysis shows that the thermal gradient and the early-age stresses were reduced, however, the removal of the insulation layer needs to be conducted carefully to reduce the thermal shock. The analysis developed in this study is applicable to concrete structure undergoing creep deformation during thermal and shrinkage loading and unloading cycles. Results show that using the measurements of the mechanical properties and the methods proposed in this study, an accurate estimation of the maximum tensile stress can be achieved for a concrete mix containing GGBFS and fly ash replacement. The estimation of the maximum tensile stress can enable engineers to take preventative actions to minimize the risk of thermal cracking.

## Recommended Citation

Leon, Guadalupe, "Estimation and Mitigation of Stresses in Mass Concrete Structures Containing Ground Granulated Blast FurnaceSlag and Fly Ash" (2021). *Graduate Theses, Dissertations, and Problem Reports*. 10245.

https://researchrepository.wvu.edu/etd/10245

## Embargo Reason

Publication Pending