A new, three-step technique for detecting taste and odor sources in bottled water has been shown effective in identifying and quantifying the agents at fault, down to the low parts per billion level. The bottled water market represents big business. U.S. Sales are currently in excess of $4 billion. Globally, the market is estimated to be worth $14 billion at the wholesale level and growing at ~8% per year.

Steam distillation/extraction of the individual packaging components followed by a liquid chromatographic class separation provides a technique for the isolation and identification of these problematic species. Capillary gas chromatography/mass spectrometric (GC-MS) analysis of the fractions enables the identification of several oxidation products that are formed during the fabrication of the components that can significantly impact the integrity of the packaging materials.

Reversed-Phase High Performance Liquid Chromatography (HPLC) along with ultraviolet (UV) and mass spectrometric (MS) detection may be employed to monitor levels of additives in the packaging materials. This generally requires extraction of the bulk material. This extraction is performed on a "ground" sample for solvent insoluble polymers. For solvent soluble polymers, a non-solvent is added to the polymer solution to precipitate the polymer, leaving a solution that contains only the additives. These resulting extracts are analyzed by HPLC and GC-MS to identify all additives and oxidation products that are present.

Analysis of the packaged water can also provide insight into the odor and taste problem. Extraction of large quantities of the packaged water (500 mL) using a Wheaton® purge and trap apparatus followed by thermal desorption-GC/MS analysis of the resulting absorbent provides a way of quantitating any hydrophilic taste and odor components present in the water. The combination of all three of these techniques can identify any problematic species present in the packaged waters.

Sample

Commercially bottled water, which consisted of a poly(ethylene) closure and a poly(ethylene terephthalate) bottle was obtained, and used for this study. The sample was chosen, because the packaged water was found to afford an off odor and taste when sampled by a 5-member sensory panel.

Results

Chromatograms from the combined techniques used to analyze the bottled water sample for odor and taste components are illustrated in Figures 1 through 4. Several represetative compounds were readily identified in these chromatograms. The extract of theground closure (Figure1) analyzed by GC-MS shows the presence of a common slip agent, erucylamide, two polymer additives (benzyl butyl phthalate, 2,6-di-t-butyl phenol), as well as a quinone structure (7,9-Di-tert-butyl-1-oxaspiro(4,5)deca-6,9-diene-2,8-dione). This quinone structure appears to be the result of oxidation of 2,6-di-t-butyl-phenol, which is also detected. Analysis of the extract by HPLC-UV (Figure 2) shows the presence of tris(2,4-di-t-butylphenyl) phosphite (Irgafos 168), as well as the oxidized form of Irgafos 168. Other unknown components were detected as well.

The methylene chloride fraction of the class separation on the steam distillation fraction is included in Figure 3. A series of volatile aldehydes were detected, which have sensory threshold detection limits in the low part-per-billion range. These compounds are characterized by a "rancid" or "musty" odor and typically impact taste and odor.

Figure 4 contains the GC-MS chromatogram for the thermal desorption of the water sample. Several of the aldehydes found in the steam distillation were also detected in the water.

Conclusions

The ability of the combined techniques to identify taste and odor problems in water and food packaging has been demonstrated. The polyethylene closure tested appears to have been oxidized, which is evident by the presence of a quinone in the Figure 1 and the oxidized Irgafos 168 in Figure 2. This oxidation is likely the source of the aldehydes, as well. Oxidation of erucylamide and other components that contain unsaturation can result in the formation of aldehyde moieties. Additional testing on the virgin resin, as well as freshly molded closures, may yield further information as to where the oxidation has occurred.