• 8 •
by Neil Mosesman, GCColumns Product MarketingManager
ColumnBleed&
SystemContamination
IdentifyingandReducingSourcesof Rising
Baselines inGCAnalysis
Rising baselines are a common occurrence during
temperature-programmed gas chromatography (GC)
(Figure 1). The rise in the baseline can be caused
by several factors: stationary phase bleed from the
analytical column, contamination in the injection or
detection system, and/or a change in the flow rate.
Themagnitude of the baseline rise often depends on
the sensitivity of the detection system. With very sen-
sitive detectors, even a small amount of bleed or
contamination can cause a significant rise in the
baseline. Reducing or eliminating rising baselines
can improve qualitative and quantitative reliability of
your chromatographic analyses.
Because both the column and the system can con-
tribute to rising baselines, it is important to distin-
guish between the two sources when troubleshoot-
ing. The simplest way to do this is to remove the
analytical column from the system, cap off the detec-
tor, and determine the background level during a
temperature-programmed run. If the baseline is
unstable, follow the recommendations in “How can
detector effects be reduced?” If the baseline is sta-
ble, connect a jumper (i.e., a short length of clean,
uncoated fused silica tubing) from the injector to
the detector and perform another temperature-
programmed run to show the effects of the injector
on baseline stability. If the baseline is unstable,
follow the recommendations in “How can injector
effects be reduced?” If the baseline from the injector
and detector is stable, install the analytical column
and perform a temperature-programmed runwith-
out making an injection to determine if addition of
the column increases the background level.
How can injector effects be reduced?
Injector contamination can be amajor cause of
baseline instability. Highmolecular weight, non-
volatile residue from the sample can slowlymigrate
through the analytical column and cause a rise in
the baseline during a temperature-programmed
run. It often is difficult to determine if the baseline
rise is caused by the column or by injector contam-
ination. Removing the column from the GC and
running a jumper (see above) will isolate the
source of the baseline rise. If the injector is con-
tributing to the bleed level, maintenance should be
performed. In particular, replace the septum, liner,
and seal. In cases of extreme contamination, rins-
ing the injection port with solvent may be necessary.
Aftermaintenance, confirm the cleanliness of the
injection port by performing a blank injectionwith
the jumper installed between the injector and the
detector.
How can detector effects be reduced?
Baseline rise from the detector is usually caused by
contamination or impure gas. Proper detectormain-
tenance, including periodic cleaning, is critical to
minimizing baseline rise. Make-up gas and/or fuel
gases also can contribute to bleed. Figure 2 shows
an unstable flame ionization detector (FID) baseline
caused by trace impurities in the compressed air
supply to the flame. Switching to a high-purity air
generator that employs traps to remove trace hydro-
carbons can greatly improve FID baseline stability
(Figure 3). Using high-quality gas purifiers formake-
up gas and fuel gases is critical to reducing back-
ground levels caused by the detection system.
How can column bleed be reduced?
If the baseline rise caused by the injector and
detector has been reduced and the baseline still
remains high, then themost likely cause is column
bleed. Themagnitude of column bleed is related to
the final oven temperature. Higher final tempera-
Figure 1
Rising baselines are a common occurrence during temperature-programmedGC.
Figure 2
Trace impurities in the carrier gas can cause an unstable FID baseline.
Figure 3
Using a high-purity gas generator and traps to remove contaminants can greatly improve
FID baseline stability.
bleed
GC_EX00388
✔
Improve qualitative and
quantitative reliability
✔
Increase column lifetime
0min.
100
200
300
400
500
600
6
4
2
picoamps
0min.
100
200
300
400
500
600
6
4
2
picoamps
0min. 5
10
15
20
25
30
40
34
30m, 0.25mm ID, 0.25µmXTI
®
-5
(cat.# 12223) XTI
®
test mix
injection;
Oven temp.:
100°C to
300°C@ 6°C/min. to 360°C@
15°C/min.
100
80
60
40
20
picoamps
GC_EX00390
GC_EX00391
