Graduation Year

2018

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Civil and Environmental Engineering

Major Professor

A. Zayed, Ph.D.

Committee Member

Rajan Sen, Ph.D.

Committee Member

Gray A. Mullins, Ph.D.

Committee Member

Aydin Sunol, Ph.D.

Committee Member

Kyle A. Riding, Ph.D.

Committee Member

Abdul Malik, Ph.D.

Keywords

Granulated Slag, Internal Curing, Metakaolin, Mix Design, Rigid Cracking Frame, Surface Resistivity, Chloride Ingress

Abstract

Control of early-age cracking in concrete structures due to restrained deformation, thermal and/or autogenous, presents a major durability challenge for state highway agencies. This study details research conducted on improving the early-age cracking resilience of plain concrete and supplementary cementitious materials (SCM) blended concrete. Three specific areas, within concrete structural elements, with high incidence of early-age cracking are addressed:

  • High early-age strength (HES) concrete mixtures used in slab-replacement of Jointed Plain Concrete Pavements (JPCP)
  • Use of metakaolin (MK) in bridge deck mixtures
  • Use of slag blended concrete, at high cement replacement levels, in massive structural elements

Three cracking mitigation measures were evaluated for HES concrete mixtures; namely, aggregate blend optimization, use of pre-wetted lightweight aggregate and the use of shrinkage-reducing admixture. While all methods reduced early-age induced tensile stresses and indicated lower cracking tendencies, mixtures with lower cement content had the lowest tensile stress to strength ratio, while meeting the 6-hour strength thresholds.

The use of internal curing by incorporating pre-wetted lightweight aggregate (LWA) to reduce early-age cracking tendencies of MK-blended concrete was evaluated. The findings indicate a lower cracking temperature and a lower tensile stress/strength ratio at 48 hours when LWA was used in MK blended concrete mixtures. Limited dilatometry testing for chemical shrinkage indicated a shrinkage coefficient of 0.28 ml/g MK.

The performance of five commercial slags with varying chemical, mineralogical and physical properties, typically used in massive concrete elements, was evaluated for early-age cracking tendencies at 60% replacement level. The mixtures were prepared using two cements having different C3A content. Cracking indices obtained from the rigid cracking frame testing under uniaxial restraint correlated well with magnesia to alumina ratio of slag, with a lower MgO/Al2O3 ratio correlating with a higher cracking tendency.

A limited assessment of the long-term durability of these mixtures was conducted through measuring concrete surface resistivity and apparent chloride diffusivity. Three concrete mixtures were prepared; namely, control, slag blended concrete with low alumina slag (8%) and another slag blended mixture with high alumina slag (16%). The total chloride concentration profiles, obtained using ASTM C1556 methodology at 105 days of immersion in 165g/L NaCl solution, indicated better performance of the low alumina slag concrete mixture compared to the high alumina slag concrete or the control. Similarly, resistivity measurements and calculated formation factors were higher for the slag blended systems compared to the control systems, indicating a more disconnected pore structure and therefore higher resistance to chloride ingress. The 8% alumina slag concrete mixture reported highest surface resistivity and formation factor at 28 days relative to the 16% alumina slag concrete mixture as well as the control mixture.

Chloride binding isotherms were determined on the cementitious paste of the same three systems considered for chloride diffusivity measurements. After 48-day immersion in solutions of variable chloride content (0.1-3M), the slag blended paste showed higher binding capacity than the control systems. The mixture with high alumina slag exhibited higher binding capacity at 48 days of exposure to chlorides. An exposure time of 105 days might be necessary to assess the free chloride ion concentration at the same age as the total chloride diffusivity curves.

Share

COinS