The extent of contamination of samples for carbon analysis by contact with plastic surfaces
Internal Report 254
Speers A & leGras C
Supervising Scientist Division
Department of the Environment, 1997
- The extent of contamination of samples for carbon analysis by contact with plastic surfaces (PDF - 1,745 KB)
About the report
This report describes the effects of storage in plastic bottles, and filtration, on samples for analysis of total organic carbon. The sample containers used were made from high density polyethylene (HDPE). Polycarbonate filtration apparatus with neoprene gaskets were used, together with polycarbonate and nylon <0.45 µm membranes. High-purity water was prepared using Millipore Super-Q equipment.
Six experiments were performed on notionally blank samples (n=10). Unfiltered, unacidified Super-Q water was stored at 4°C for 24 hours and 7 days; unfiltered Super-Q water was acidified 1% v/v with Aristar HN03 and stored at 4°C for 24 hours and 7 days; and unacidified Super-Q water was filtered (<0.45 µm) through polycarbonate and nylon membranes, then acidified as above and stored for 24 hours before analysis. The highest concentration observed, in the unfiltered, acidified sample stored for 7 days, was 0.11 mg/L. Statistical analyses were performed on pairs of treatments, using the Student's t distribution. These showed that all treatments, except filtration through a polycarbonate membrane, yielded mean values which were significantly different from a Super-Q water blank stored in a glass container (p<0.05).
Three experiments were performed on samples containing 25 mg/L dissolved organic carbon. The treatments were: a sample prepared using glass apparatus and analysed immediately; samples stored in HDPE bottles for 7 days at 4°C, and samples acidified as above and stored for 7 days at 4°C. Statistical analysis using the F distribution demonstrated that no significant differences existed among the samples. This implies that carbon contamination, and adsorption and oxidative degradation losses are minimal. However, close examination of the data suggests that a small degree of carbon contamination, similar to that found in the blanks, may occur.
The glass vials used in the autosampler require only simple rinsing in Super-Q water as a cleaning procedure.