Comparative study of the adsorption performance of an active multi-sorbent bed tube (Carbotrap, Carbopack X, Carboxen 569) and a Radiello(®) diffusive sampler for the analysis of VOCs.

A simple comparison is made to evaluate the relative performance of active and passive sampling methods for the analysis of volatile organic compounds (VOCs) in ambient air. The active sampling is done through a multi-sorbent bed tube (Carbotrap, Carbopack X, Carboxen 569) created in our laboratory and the passive sampling through the Radiello(®) diffusive sampler specified for thermal desorption (filled with Carbograph 4). Daily duplicate samples of multi-sorbent bed tubes were taken during a period of 14 days. During the same period of time, quadruplicate samples of Radiello(®) tubes were taken during 3 days, 4 days, 7 days and 14 days. The sampling was carried out indoors during the months of February and March 2010 and outdoors during the month of July 2010 in La Canonja (Tarragona, Spain). The analysis was performed by automatic thermal desorption (ATD) coupled with capillary gas chromatography (GC)/mass spectrometry detector (MSD). The analytical performance of the two sampling approaches was evaluated by describing several quality assurance parameters. The results show that the analytical performances of the methodologies studied are quite similar. They display low limits of detection, good precision, accuracy and desorption efficiency, and low levels of breakthrough for multi-sorbent bed tubes. However, the two monitoring methods produced varying air-borne concentration data for most of the studied compounds, and the Radiello(®) samplers generally gave higher results. Sampling rates (Q(k)) were determined experimentally, and their values were higher than those supplied by the producer. As the experimental calculation of Q(k) values is generally carried out by the suppliers in exposure chambers with only the target compounds present in the air samples, as well as in concentrations dissimilar to those found in ambient air, the use of constant settled Q(k) can lead to inaccurate results in complex samples.