6890N GC, with 1/4" packed column injection

[pochengjean]

Does anyone has the same system as i do? Agilent 6890N with Injection port for 1/4" column or 1/8" column. I just realize that agilent revolutionalized the injection port design.

I noticed that there exist a huge dead volumn between the injection needle and the column.

Instead of having the needle inject directly inside the packed column, agilent introduced a glass liner and a sleeve between the injection needle and the column, which introduced huge dead volumn to the system.

My question is, does the system perform well???

[Russ]

Not sure if the 6890N is any different from the 6890 / 5890 but on those systems you can remove the sleeve containing the liner, if desired, and insert a glass column into the injection port so you can do on-column injection. We use the packed column inlet with megabore columns and have not had any problems with the design. One advantage of the liner is that it will collect any non-volatile residue and keep it off your column. This is not very new, I used a Perkin-Elmer instrument back when dinosaurs were roaming the earth that used injection port liners. I seem to remember that it was not very easy, and may not have been possible, to remove the sleeve containing the liner to perform on-column injections on that instrument.

[pochengjean]

I never worked with 6890 series with the packed column before.

However, what i know is that the 6890N series require a packed port which only support 1/8" packed column, and requires an additional 1/4" adaptor (~170US dollars) to work with my column. The 1/4" adapter does not have any glass insert, it combines both functions as a sleeve and an insert, which creates huge dead volumn.

The worst is the detector assembly. The TCD detector requires an additional adapter to work with my 1/4" column. However, unlike the S/SL injection capillary system, the detector adapter is loosely assembled to fit the whole apparatus. I will say, approximately 4 mL of empty space before the analytes reach the detector.

My result is acceptable, since we are using "loosely" defined acceptance criteria. The peak shape is broad, the baseline response is much higher compared to the same system using capillary system.

I can't seem to identify the problem source, maybe it is due to the copper-made 1/4" injection port adapter or the stainless steel 1/4" detector adapter.

What do you suggest to isolate and identify where the problems actually come from???

[Russ]

I haven't used a glass column on the 6890, but I would think you could get one made that went into the injection port such that an injection was made on column. You would then remove the adaptor and tighten the column with a nut where the adaptor is connected. On the 5890 / 6890, the same nut and Vespel-graphite ferrule that are used to connect the adaptor to the port will fit onto a 1/8" glass column and can be used to connect the column directly to the injection port. DANGER WILL ROBINSON: You do not want the column to be so long that it "bottoms out" at the top of the injection port or you could block carrier flow to the column (sad experience talking). Back in prehistory, I used 1/8" metal columns mounted in a much older HP instrument in a similar manner. We had to get a needle guide that fit over the column and an adaptor to connect the 1/8" column to the larger fitting. Don't know if those needle guides are still available, or how well it would work for an autosampler.

[Ron]

Packed columns run much higher flow rates than capillary columns, so dead volume is not as serious issue as it is for capillary work. 1 mL dead volume is a problem at 1 mL/min flow rate, but no big deal at 30 mL/min flow rate. Most of the time you will have a much broader peak on a packed column than on a capillary column, this is simply a function of column efficiency. Higher baseline is not uncommon, there is more stationary phase to bleed off.

[pochengjean]

Packed columns run much higher flow rates than capillary columns, so dead volume is not as serious issue as it is for capillary work. 1 mL dead volume is a problem at 1 mL/min flow rate, but no big deal at 30 mL/min flow rate. Most of the time you will have a much broader peak on a packed column than on a capillary column, this is simply a function of column efficiency. Higher baseline is not uncommon, there is more stationary phase to bleed off.

Thank you.

anothe question is that, how do define a peak as too broad? and how do you define a peak as too unsymmetric?

[jcstrides]

I used PE's before and even after the KT period when the dinosaurs, for the most part, trundled off this mortal coil I actually really liked them. These did have a glass sleeve and it really did extend the life of columns, since all of the garbage stayed on the sleeves. You only had to be carerul of a few things, such as cleaning them on a regular basis, not chipping them and we actually treated them with I think it was tetramethyl silane, or something like that to deactive the glass after cleaning. They were also a lot cheaper to through away than a column, if the glass had any garbage on them.

I also know that, at least, older varians, had the same stuff on them.

[pochengjean]

I used PE's before and even after the KT period when the dinosaurs, for the most part, trundled off this mortal coil I actually really liked them. These did have a glass sleeve and it really did extend the life of columns, since all of the garbage stayed on the sleeves. You only had to be carerul of a few things, such as cleaning them on a regular basis, not chipping them and we actually treated them with I think it was tetramethyl silane, or something like that to deactive the glass after cleaning. They were also a lot cheaper to through away than a column, if the glass had any garbage on them.

I also know that, at least, older varians, had the same stuff on them.

Could you tell me how to silanize the glass column? such as the temperature, the solvent?

[jcstrides]

Supelco used to sell a solution in, I believe, toluene, although I could be mistaken. After cleaning the liners, we simply soaked them in the solution for a short while and then dried them in an oven. It worked just fine.

You need to be careful though, that when you clean the liners, you don't scratch them. Usually a solvent soak, or a detergent soak in an ultrasonic bath.

hope that helps.