In the study by Walsh and Ranney (1998), they
1 and 2, respectively. Analysis of samples with high
injected solvent extracts into the heated injection
concentration of HMX confirmed that the latest
port of a Hewlett-Packard 5890 GC equipped with
eluting peak(s) (retention time > 7.5 min) was in-
an ECD and a 6-m, 0.53-mm-i.d. polydimethyl-
dicative of this compound (Fig. 1). However, ei-
siloxane column. They established that responses
ther the low vapor pressure of HMX caused it to
for analytes with low vapor pressures were en-
rapidly condense once leaving the column, or it
hanced by increasing the linear velocity of the car-
was thermally degraded, and consequently these
rier gas and the injection port temperature. Fol-
late eluting peaks could only be used to qualita-
lowing their lead, we chose on-column injections
tively identify its presence. The GC manufacturer
into a heated injector and a carrier gas flow rate of
suggested that upgrading the NP detector to one
9.5 mL/min that produced a linear velocity of 70
with a heater would eliminate this quantitation