-
- Posts: 2
- Joined: Tue Jun 22, 2021 9:01 pm
I’ve been having issues over the past several weeks with a Trace 1310 dual-column GC, using an FID with two ZB-624plus columns. The baseline for the back column - but not the front - slowly but steadily rises from injection to injection, and the noise rises proportionately. There isn't normally any real problem during the temperature ramp of any particular injection, but over a few days of injections the baseline will rise and noise will increase until the column is basically unusable.
Here’s a summary of chromatographic conditions:
Column: Zebron ZB-624plus
Length: 30 m
ID: 0.32 mm
Stationary phase: G43, 1.8 µm.
Split ratio: 10:1
Injection volume: 1 µL
Column temperature: 40°C for 5 min, then ramp 30°C/min for 6 min, then hold at 220°C for 5 min
Injection port temperature: 200°C
Detector: Flame ionization with temperature of 280°C
Carrier gas: Helium
I work at a plant making (normally) high-purity ethanol, and this is almost exclusively the only thing we run on this GC. There is very little in the way of non-volatiles in the samples we run – we’re limited to 25 mg/L total nonvolatile residue – so I do not believe the issue is likely to come down to nonvolatile or semi-volatile residue in the samples. The front column of the same instrument is working normally, as well, and receives the same type of samples.
My boss and I have tried replacing the septum, liner, and needle multiple times with no noticeable change. Changing the split vent filter didn’t do anything either. We also tried replacing the column with a new one of the same type, only to have the issue reappear within a few days on the new column as well.
The only thing we have found that works in the short run, not counting the column change, is to bake out the column around its isothermal maximum of 300 C for 2-3 hours, with the inlet and FID temperature also increased to 300 C. This causes the baseline to decline to near-normal levels and reduces noise temporarily, but both climb back to where they had been after a few injections with the normal method.
I’ve attached a screenshot of the issue. Each injection is just a solvent blank using the same high-purity ethanol. The date is just the date of calibration – these injections actually took place over the course of about three weeks, ending a couple of weeks ago. It’s hard to see the noise on this scale except for the last injection, but it steadily crept up along with the baseline. Lately it has gotten even worse and we’ve had to take the column out of service.
Has anyone else run into an issue like this? Thanks in advance for the help!