Date of Award

Summer 8-27-2021

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Civil and Environmental Engineering

Advisor(s)

Bhatia, Shobha

Subject Categories

Civil and Environmental Engineering | Engineering | Geotechnical Engineering

Abstract

Failure of an embankment dam around mine tailings and coal ash slurry impoundments typically results in a fast-moving flow of toxic waste that can carry for several miles with disastrous consequence. The most recent catastrophic tailing-dam failure took place on 25 January 2019, at Brumadinho, Brazil, where the resultant tailings flow killed at least 232 people. A similar impoundment failure also occurred in 2008 at the TVA Kingston fossil power plant in Tennessee, USA, where coal combustion residuals such as fly ash and bottom ash were stored in the impoundments. The catastrophic failure at Kingston released 4.2 million m3 of coal ash slurry and compacted ash and highlighted the fatal potentials of impounded ash failures. Over the past century, more than 300 tailings dam and ash pond failures have been reported around the world. The post-failure investigations have identified several common failure mechanisms: including overtopping due to water mismanagement, shear failure of foundation soils and shear of compressible low-permeability tailings placed near the perimeter of the impoundment. Numerous studies have been carried out to characterize the failed tailings and ash materials and understand the failures of slurries. Tests conducted include rheological, containment loss, plane table and slump testing. However, there is still much uncertainty that remains to be excised before it is possible to fully understand the flow behavior and rheological properties of tailings and fly ash slurries when containment fails. A few studies have been carried out to investigate the flow behavior of fly ash slurries by measuring their rheology properties, but the rheological properties have not been correlated to the flowability of the fly ash slurries. In addition, there has been no systematic study conducted to understand the role of index and chemical properties on the rheological behavior of slurries and measure changes to the degree or amount of flow of tailings and ash slurries. The main objective of this study is to understand the flow behavior of granular soil and fly ash slurries. The current research aims to evaluate the flow behavior of granular soil and fly ash slurries and investigate the use of additives in minimizing the flow and modifying the rheological properties of slurries. The following research questions are posed by this study: (1) Do the index and chemical properties of granular soil and fly ash slurries have an impact on the yield stress (rheological properties) and flow behavior of slurries? (2) How are the rheological properties and flow behavior of slurries related? (3) Can additives modify the rheological properties and flowability of slurries? This research does not focus on the evaluation of stability of tailings and coal ash dams, but on the investigation of rheological and flow behavior of tailings and fly ash slurries which are stored in the impoundments. The rheological properties of slurries are not only important to evaluate the stability of tailings and coal ash dams, but also critical in making major decisions about design and operations of dams. To answer these questions, the following research objectives are established: (1) characterize the physical and chemical properties of representative granular soil and fly ash slurries; (2) understand rheological behavior of slurries and correlate rheological properties with their index and chemical properties of slurries; (3) evaluate the flow behavior of slurries using the containment loss test; (4) correlate flow behavior with index and chemical properties of slurries; and (5) evaluate the effectiveness of additive on the rheological and flow behavior of slurries. For this research, six granular soils and six fly ash materials were selected. Three fly ash materials were collected from impoundments in slurry form. Due to confidential issue, the history of hydration and the accurate location of fly ash impoundments were unknown. The non-hydrated fly ashes were collected directly from hoppers of coal ash thermal power plant, and they have never been exposed to water. The granular soil materials include three Silica products, two Tully soils and one Rock flour. Index and chemical properties of all granular soils and fly ash materials were measured, with Scanned Electron Microscope (SEM) images of fly ash show that most of the fly ash particles are spherical. Both Tully soil and Rock flour are glacially deposited materials with sub-rounded and sub-angular particles. Silica is mostly angular, produced by grinding Midwest sand. The fly ash materials are well graded with a uniformity coefficient of 13.33 – 73.07 and their median particle diameter (D50) ranged between 14 – 40 microns. The granular soils are either well or poorly distributed with median particle diameter (D50) ranging between 14 – 40 microns. The chemical compositions of all granular soil and fly ash materials were measured using Vanta M series X-ray Fluorescence. A rotational rheometer Anton Paar MCR-302 was used to measure the rheological properties of granular soil and fly ash slurries. The rheological tests for all slurries were performed in the range of 40% – 75% solid content and for each slurry; three tests were conducted to check the repeatability. Test results show a high degree of repeatability for all granular soil and fly ash slurries. In total, 170 rheological tests were performed. At lower solid content (10% – 30%), a slurry behaves like a Newtonian slurry, while a slurry behaves like a non-Newtonian slurry at higher solid content. Results show that all granular soils and fly ash slurries in this study behaved like shear thinning non-Newtonian slurries with yield stresses. It was found that yield stress increased exponentially with the increasing solid contents of slurries, however, the increasing rate was different for granular soil and fly ash slurries. For 40% – 75% solid contents, fly ash slurries provided higher yield stresses as compared to granular soil slurries. For granular soil slurries, yield stress increased with the increasing median particle diameter (D50). Effect of particle shapes was also observed. For granular soil slurries, angular particle slurry (Silica 70) had higher yield stress as compared to sub-rounded and sub-angular particle slurry (Rock flour slurry). Yield stress is an important property of slurry, which is used not only to evaluate the flow mechanism of slurries but also used to take critical decisions in fly ash/ mine tailing industries. The determination of fly ash and tailings disposable system design and operating conditions require a thorough understanding of yield stress of slurries with high solid contents. The success of fly ash and tailings management system depends on implementing and operating a safe and environmentally responsible system at minimum cost. The measurement of yield stress of a slurry helps to determine the solid contents required to achieve optimum characteristics for storage impoundments, optimum conditions for pipeline transport and dewatering of slurries at required solid contents. A flow test was developed which resembles a modified version of miniature cylindrical (2.5 cm inner diameter) slump test. The test results were used to measure the flow area and slump height for slurries at 40% – 60% initial solid contents. The flow area of slurries were measured after containment loss and applied vibration. Three tests were performed for each slurry for statistical significance. In total, 126 flow tests were conducted with all granular soil and fly ash slurries. The effect of containment size was evaluated with Silica 400, Silica 300 and Silica 70 slurries and no significant difference was noticed in the flow behavior of silica slurries if containment size was altered. Four different types of flows were observed in the flow test: (a) shear flow (where a slurry completely changed from its original shape and spread laterally over the base), (b) elongational flow (a combination of yielded and unyielded portion), (c) materials remain in cylindrical containment shape for a while and then fall over the base due to its own weight after support removal, and (d) no flow (a slurry did not come out of the cylinder, i.e. was stuck). It was found that the types of slurry flow were not only a function of chemical composition of fly ash, but also initial solid content (%) of the slurry before sedimentation. Most of the granular soil and fly ash slurries showed shear flow (spread) after containment removal. It was found that the dimensionless area of both granular soil and fly ash slurries after containment removal decreased with the increasing solid contents of slurries. The effect of chemical compositions on rheological and flow behavior of fly ash slurries was also investigated. It was believed that the amount of calcium-silicate (CaO+SiO2) played an important role in the rheological and flow behavior of fly ash slurries. Two Hydrated fly ash slurries in both rheological and flow tests and one Non-hydrated fly ash slurry in flow test showed some levels of cementitious and stiff behavior. Along with the amount of calcium-silicate, the degree of hydration greatly influenced the rheological and flow behavior of fly ash slurries. The dimensionless slump height showed a decreasing trend with the increasing yield stress of slurries; however, the decreasing rate was different for granular soil and fly ash slurries. The correlation between dimensionless slump vs. solid content and dimensionless yield stress vs. dimensionless slump were observed mostly based on shear flow and elongational flow. It was found that the median particle diameter, specific gravity, chemical compositions, particle shapes, state and degree of hydration, initial and final solid contents of slurries showed significant influence on the rheological and flow behavior of granular soil and fly ash slurries. The use of polyacrylamide EM 533 increased the yield stress by 350 times for flocculated Rock flour slurry, 10 – 50 times for flocculated Hydrated fly ash -1 and Hydrated fly ash -3 slurries and 190 times for flocculated Non-hydrated fly ash -3 slurry, as compared to the slurries without EM 533. The polyacrylamide EM 533 is used in the coal refuse impoundment for sedimentation. Synthetic fibers increased the yield stress of Hydrated fly ash (1, 3) slurries by a mere two times, as compared to the yield stress of slurries without synthetic fibers. In the flow test, the use of EM 533 and synthetic fibers with fly ash and granular soil (Rock flour) slurries decreased the flow areas by 10% – 50%. This research is a unique approach to understand the effect of nature of soil/ fly ash, particle shapes, index and chemical properties, degree of hydration, initial and final solid content of testing, on the rheological and flow behavior of granular soils and fly ash slurries. It was found that the particle size distribution and specific gravity played an important role on the rheological and flow behavior of granular soil slurries, while the behavior of fly ash slurries mostly depends on particle shape, chemical properties, degree of hydration and solid contents. It was also found that slump test could be performed in a small-scale with limited amount of materials and yield stress could be calculated from a small-scale slump test. Researchers and practitioners use different numerical modelling to predict the flow behavior of slurries, where yield stress of slurry is used. This research will help researchers and practitioners understand the effect of basic properties on the yield stresses and flow behavior of slurries.

Access

Open Access

Download

Share

COinS