Degree Type

Dissertation

Date of Award

2020

Degree Name

Doctor of Philosophy

Department

Civil, Construction, and Environmental Engineering

Major

Civil Engineering (Geotechnical Engineering)

First Advisor

Jeramy C. Ashlock

Abstract

Annual maintenance of granular roadways in cold regions costs millions of dollars to the local counties and requires a large portion of county budgets for purchase of new aggregates, equipment use and labor cost. Maintenance of granular roadways commonly includes blading the surface materials and spreading of new aggregates on existing surface aggregates. Freezing and thawing, high traffic loads, the presence of frost susceptible soils, and the use of low-abrasion resistance aggregates reduce the serviceability of unbound surface courses by decreasing the stiffness, strength and ride quality, while increasing aggregate and thickness loss, fines content, and dust emissions. However, utilizing high-abrasion resistant surface aggregates, even if associated with higher material and hauling costs, could be cost-effective by reducing the rate of deterioration of these roadways and consequently minimizing the required number of maintenance cycles.

In this study, a variety of aggregates were collected from four different locations in Iowa to construct three field test sections with standard “Class A” aggregates, and four sections with mixtures of locally available “Class A” and “Clean” aggregates. The optimum mixture proportions were determined based on a previously developed gradation optimization tool. This tool determines the optimum aggregate mixture design for up to five different materials to achieve the optimum particle packing in terms of shear strength. Each test section was constructed with a four-inch thick surface layer of aggregates distributed over 3,300 ft of a granular roadway.

A benefit-cost analysis (BCA) model was developed to evaluate the cost-effectiveness of using six aggregate alternatives with different material and hauling costs. Results were compared with a seventh control section, which was built with the locally available Class A surface aggregate that is most commonly used in the region. This local Class A aggregate was collected from a quarry in Decatur County, Iowa. The performance of all seven test sections was compared to each other over a period of three years. Performance indicators were measured at regular intervals, including surface shear strength, stiffness, gradation changes (e.g. total breakage and breakage potential), dust emission, and ride quality. Maintenance scenarios were determined based on the level of serviceability and performance for the six alternative sections, while the control section was assumed to have maintenance every year. To define the maintenance scenarios, it was decided to delay maintenance procedures for one year for the sections with medium performance, and two years for the sections with the best performance. The results of the BCA illustrated that combining local aggregates with higher quality aggregates would be beneficial in lowering the overall long-term costs of the project compared to building granular road surfaces with locally available Class A materials.

Multichannel analysis of surface waves (MASW) and falling weight deflectometer (FWD) tests were conducted in this study as non-destructive tests to evaluate the elastic modulus of surface and subgrade layers. Two numerical models of road layer systems including a two-layered (surface and subgrade), and a three-layered (surface, subbase, and subgrade) model were selected, based on the results of dynamic cone penetration (DCP) tests. The software programs BAKFAA and Modulus 7.0, as well as a combined Odemark’s- Boussinesq’s (OB) theory were used for FWD back-calculation of the elastic modulus profiles of the layered systems. Linear regression models were developed to compare the back-calculated elastic moduli calculated from the different methods. In addition, the results of FWD and MASW tests were compared based on the dependency of the back-calculated elastic modulus moduli on the stress and strain levels associated with each test by using the KENLAYER program. The results showed that normalizing elastic moduli of surface and subgrade layers by the square root of the confining stress reduces the scatter of the data. A better agreement was observed between the back-calculated results of OB method versus BAKFAA compared to the OB method versus Modulus 7.0.

In this research, performances of the entire test sections was monitored for changes in stiffness, dust emission, total breakage, and fines content, and all of these performance indicators were linked to the abrasion resistance of the surface aggregate materials used in each test section. Abrasion resistance of the surface aggregates was measured by Micro-Deval and Los Angeles (LA) abrasion tests. A pairwise multiple-comparison procedure was used to make correlations between the abrasion resistances and performance measures. Results showed that mixing low-abrasion resistant Class A surface aggregates from locally available sources with high-abrasion resistant clean aggregates were beneficial by reducing fines content and total breakage, while increasing the overall stiffness of sections. Pairwise-correlations showed higher p values for the average results of performance measures and abrasion losses from Micro-Deval tests than those from LA abrasion tests.

Surface aggregate materials deteriorate under traffic and compactor loads, and such deterioration alters their gradation. The Micro-Deval and LA abrasion tests are the two most commonly used tests to evaluate the quality of aggregate materials. However, neither of these tests properly simulate the types of abrasion applied to granular road surface materials in the field. Therefore, this study used the gyratory compaction device to better evaluate the durability of aggregate materials, and comparisons were made between aggregate gradation changes resulting from gyratory compaction, Micro-Deval, and LA abrasion tests. Moreover, the shear resistances void ratios, and maximum dry unit weights of each material were measured before and after each gyratory compaction test and tracked during the tests. The results showed that gyratory compaction produced the greatest change in the gravel and sand contents of the aggregates, and the fines contents were not affected as much during this test relative to the other two test types. Moreover, the energy levels required to gain the optimum void ratio and dry unit weight in the gyratory compaction tests were always greater than those required for modified and standard Proctor compaction tests.

The results of this study concluded that mixing coarse aggregates from sources with greater distances to the site location could be beneficial by reducing the construction and maintenance costs depending on the material and hauling costs. Utilizing aggregate options with higher abrasion resistances than locally available aggregates would decrease the overall thickness and material loss and increase the performance and serviceability by increasing average elastic modulus and decreasing gradation change and total aggregate breakage. Moreover, the MASW method was proved to provide more reliable elastic moduli compared to FWD, where FWD back-calculation methods did not have close agreements on their moduli results. Pearson correlations demonstrated significant correlations between Micro-Deval abrasion losses and performance measures such as composite elastic modulus, fines content, and total breakage. Total breakage and thickness loss also correlated well with each other. Gyratory compaction tests with a pressure distribution analyzer were used to evaluate gradation and mechanical properties of aggregates under simulated compaction or traffic loads. Based on the results of gyratory compaction tests, the required energy to achieve certain degrees of dry density, shear resistance, and void ratio of the aggregate materials were investigated.

DOI

https://doi.org/10.31274/etd-20200902-136

Copyright Owner

Sajjad Satvati

Language

en

File Format

application/pdf

File Size

142 pages

Share

COinS