Date of Award


Document Type


Degree Name

Master of Science (MS)


Civil and Environmental Engineering

First Advisor

Matthew J. Scarborough


Glass recycling appears to be a viable way to reduce waste; however, there are many challenges to ensuring the recycled product has an economically and environmentally sustainable end use. Cleaning recycled glass to a standard that allows it to be melted down into new glass is extremely difficult when single-stream recycling is in practice. The recycled glass that is not melted down is processed into different materials, such as processed glass aggregate (PGA). As states implement bans on landfilling recyclable materials (e.g., Act 148 in Vermont), additional uses for PGA are needed. In New England, there are diminishing sources of sand borrow that meet Vermont’s specifications, causing an increase in prices (Hedges, 2009). At the same time, recycling facilities can produce a sand-sized PGA that may be able to replace or supplement sand borrow as a construction material. One major concern for end users of PGA is the deleterious (i.e., non-glass) material present after separating and crushing glass at a recycling facility, which may have a negative effect on geotechnical properties or cause environmental harm. Despite the potential benefits of using PGA as a construction material, limited research has been conducted on how to accurately determine the amount of deleterious materials in PGA. The goal of this research is to develop a protocol for quantifying total deleterious material content of PGA.

The specific objectives were to (1) research, develop, and evaluate a variety of processes to determine deleterious material content in PGA; (2) validate the effectiveness of individual processes using lab-manufactured PGA (LM-PGA) samples with known amounts and type of deleterious material; and (3) recommend a reliable and simple protocol (i.e., sequential processes) to determine deleterious material content for operational purposes. Four main mass removal processes were identified for determining deleterious material content in PGA: (1) magnet process, (2) float and skim process, (3) furnace process, and (4) acid washing process. The precision and accuracy of each process in determining deleterious material content was tested using LM-PGA containing known amounts of deleterious materials (i.e. plastics, papers, metals, ceramics, and food organics). Two protocols combining multiple processes were developed. Protocol 1 determined overall deleterious material content using a magnet process followed by a furnace process. Protocol 2 sought to estimate plastics content only using a magnet process and float and skim process. Protocol 1 and Protocol 2 were tested on LM-PGA containing multiple deleterious materials and recycling facility PGA (RF-PGA), provided by a local recycling facility. Protocol 1 was found to be precise and accurate for determining overall deleterious material content of PGA. While Protocol 2 provided a good estimate of plastic content in LM-PGA, it was unable to determine plastic content in recycling facility PGA due to the floated material containing large quantities of glass, organics, and plastics. Further work should focus on developing an improved method for determining the plastics content of PGA.



Number of Pages

94 p.