Best Practice in Glass Recycling

Click here for printable PDF version


Contaminant Levels in Recycled Glass

Material: Recycled Glass

Issue: Every use of recycled glass, from container manufacturing through construction aggregate, requires specifications that include definitions of acceptable levels of contamination. For the development of reasonable yet effective specifications, an understanding of the nature of typical contaminants in glass is critical. This Best Practice describes some procedures for judging levels of contaminants in recycled glass and presents some of the issues to be considered when developing contaminant guidelines.

Best Practice: Visual Classification. A visual classification procedure has been developed to judge the level of contaminants in recycled glass to be used as a construction aggregate. See Visual Inspection for Glass Construction Aggregate Best Practice. The classification procedure was developed during the Glass Feedstock Study(1) conducted by Dames & Moore consulting engineers in Seattle. The classification method facilitates a judgment of the percentage of contaminants in a two-dimensional area covered by a sample. Model specifications developed in that same study recommended applications for glass in construction aggregate allowing up to 10% contaminants by visual classification. While 10% may sound like a large percentage of contaminants, most of the contamination in recycled glass is usually paper labels, which actually constitute a very small percentage of the mass of a recycled glass stockpile.

For example, one square foot of 20# paper weighs approximately 7.5 grams. In contrast, one square foot of glass 1/4" thick weighs 1500 grams. If a 660 square foot surface were covered 90% with 1/4" thick glass and 10% with paper, the weight of the glass would be about one ton and the weight of the paper would be about one pound. In addition, on the surface of any pile, labels tend to orient themselves so that they lay flat, showing their largest surface. Similarly, aluminum rings tend to lay flat on the pile, appearing to be disproportionately large for their actual weight.

Organic Contamination: During the Glass Feedstock Study, extensive analyses were performed on 29 separate sources of recycled glass from all types of programs (e.g., attended and unattended dropboxes, commingled and source separated curbside, etc.). Leachate testing over time was performed for both organic and inorganic contaminants. Of all the sources tested, the only result of concern was an elevated level of polycyclic aromatic hydrocarbons (PAH's) in one sample coming from a recycling program which commingled other recyclables with used oil containers. No levels of concern were found for organic leaching in any other sample. These results were confirmed in the work of the Florida Institute of Technology (FIT), although the FIT study recommends stockpiling glass to be used as a construction aggregate above ground long enough for biological degradation to begin tapering off before using it in underground projects in contact with groundwater, in order to prevent a concentration of biological activity in a confined space underground.

Inorganic Contamination. In testing for inorganics, one issue of concern was found in the Glass Feedstock Study. Of the 29 sources tested, one was found to have a level of lead in excess of the EPA regulatory threshold (5 ppm by TCLP). Upon further analysis, the source of lead was found to be pieces of lead neck wraps from wine bottles. Although lead neck wraps have not been used on wine bottles produced

in the United States or in the European Economic Community since 1993, some lead foil still enters the recycling stream from old wine bottles and miscellaneous liquor products like single malt scotch. After a total of 50 TCLP tests and 50 additional total lead tests of a number of sources in order to determine a statistical baseline, a determination was made by the Washington State Department of Ecology that any processor of recycled glass for construction would be required quarterly to perform five total lead tests according to EPA method 3010/6010. The mean of those five tests must be less than 80 ppm, which constitutes a 90% confidence interval, given the statistical distribution of the results of the field tests, that the actual stockpile has a TCLP level blow 5 ppm.

Types of Programs. In an overall classification of the levels of debris in collected materials, the Glass Feedstock Study divided the sources of glass into three debris categories: low, medium, and high. In general, the study found that the following debris levels correlated with the following types of collection programs:
High debris
· blue bag commingled programs
· unattended drop boxes
Medium debris
· attended drop boxes
· negative sort MRF glass
Low debris
· positive sort MRF glass
· beneficiated glass
· curbside source-separated
These results support the intuitive proposition that the earlier in the collection process that sorting takes place, the cleaner is the resultant glass.

Implementation: Groups designing collection programs or developing specifications for recycled glass should consider the issues in this Best Practice.

Benefits: Often in the past, collection strategies have been driven by collection costs without regard to quality and value. Understanding contamination issues in recycling should enable programs to design collection systems to optimize the profitability of the overall program.

Application Sites: Materials recovery facilities, glass processors, recycled materials collectors.

Contact: for more information about this Best Practice, contact CWC info@cwc.org.

References:
1. Glass Feedstock Evaluation: Environmental Suitability Evaluation, Clean Washington Center, Rpt #GL-93-3.
2. Developing Specifications for Waste Glass and Waste-to-Energy Bottom Ash as Highway Fill Materials - Volume 2 of 2 (Waste Glass), Florida Institute of Technology, June, 1995.
Issue Date / Update: November 1996