Inlet Systems

Split/splitless capillary inlet

Figure 6-10

Area

Counts

+1.2% Deviation

~20 ppm n-C14(Cold Trapped)

+ 1.2% Deviation

~10 ppm n-C11(Solvent Effect)

Solvent: Isooctane

Column: 16.5 m x 0.25 mm SE-54

80^C (0.5 min) 170 C @ 15 /min Sample Size: 1.3 ml (Manual Injection) Injection Rate: 1 ml/sec

Purge Flows: 5 ml/min, Septum: 60 ml/min, Inle

10

20

30

40

50

60

Purge Activation Time, Sec

Effect of Inlet Purge Activation Time on Area Counts

Noting Figure 6•10,waiting too long does not increase component peak areas, but does increase interference by the solvent tail. Purging too early risks venting light components, not allowing sufficient time for heavier components to enter the column, and/or not having sufficient solvent enter the column to ensure good reconcentration.

A recommended procedure is to analyze a known standard using conditions identical to those to be used for later sample analyses. Ideally, the standard should be representative (in both components chosen and their respective concentrations) of unknown samples to be analyzed.

Perform a series of analyses with increasingly delayed insert purge times: the optimal time is where maximum area counts are obtained with least interference by the solvent.

119