High-speed analysis of organic vapors using vacuum -outlet gas chromatography with atmospheric-pressure air as carrier gas and programmable column selectivity.

Abstract

Most lightweight, field-portable gas chromatography instruments are burdened by the need for carrier gas tanks. A solution for this is the use of ambient air as the carrier gas. An instrument was designed using a vacuum pump to draw atmospheric-pressure air and injected samples through a pressure-tunable/programmable ensemble of two capillary columns. Principal issues considered were column efficiency, analysis time, and selectivity. Column degradation in air was studied for several stationary phases. The nonpolar polydimethylsiloxane and polar trifluoropropylmethyl polysiloxane showed good stability up to 210°C. Lengths of 4.5-m, 0.25-mm i.d. polydimethylsiloxane and 7.5-m, 0.25-mm i.d. trifluoropropylmethyl polysiloxane columns were connected in series with an electronic pressure controller at the junction point. The electronic pressure controller was capable of 0.1 psi pressure steps. Pressure changes at the junction point changed the ensemble selectivity. The pressure controller performed adequately under sub-ambient conditions. An environmentally-important mixture of 42 components was targeted for analysis using multiple junction-pressure changes and temperature-programming ramps. An analysis time of 400 s was achieved. High-speed temperature programming with at-column heating technology was studied for column lengths of 6--18 m and temperature programming ramps of 60--600°C/min, on 0.25 mm i.d. polydimethylsiloxane columns. The rate of peak capacity production, the total peak capacity, and the boiling point resolution were determined for C₁₀--C₂₈ n-alkanes. With high temperature programming rates, column lengths of 6--12 m and average linear carrier gas velocities of 100--150 cm/s were satisfactory. The best conditions were to tailor temperature programming rate and carrier gas velocity so that the last component of interest elutes simultaneously with the end of the temperature ramp. It was also concluded that the column temperature change over the duration of the column holdup time should not exceed 25°C. A model of band migration along the column axis with respect to time was developed. Using experimental retention factor values, elution times were predicted with values of relative error from 0.53% to 3.96%. Factors such as column diameter, column length, carrier gas viscosity, inlet pressure, junction pressure, and outlet pressure could be altered in the model, showing resulting band trajectories and elution patterns.