Irregular baseline noise

[StephenO]

Hi All we have a HPLC and the Chromatogram is showing straight line noise as if the detector is being turned off and on very quickly. Would it be right in thinking this is an electrical issue with the detector. What would the best why be to diagnose the problem? Detector is a Water 2489

Thanks in advance.

[Gerhard Kratz]

Why you do that????????????

[tom jupille]

If you mean "spikes" on the baseline, I've seen them come both from electronic noise and from air bubbles. The diagnostic is to shut off the flow. If the spikes go away, they were air bubbles. If they continue, it's electronic. In that case, the most frequent culprit is a bad ground or a slightly corroded connection somewhere. Try disconnecting/reconnecting all the cables. If that doesn't help, do the same for all the circuit board edge connectors. If the problem *still* persists, it's time for factory service.

If it's not "spikes", but offsets of longer duration, it could be the lamp turning on/and off. I thing most newer detectors would throw an error code, but I have seen that happen years back where the lamp would repeatedly "fire" and then die.

[Dory]

I have to tell you my problem with the baseline... actually i am a plant geneticist trying to use HPLC to quantify sugars in potato. I am using aminex 87H column with RI detector. I started to see in the blanks a peak corresponding to sucrose at a low concentration and then when i saw the samples and mix of standards the baseline just started to drop and then went up again at the end of the last peak. I tried everything to clean what i thought was a contamination, until I change the column I used for a new one... and saw the same pattern of this sucrose ghost peak and the crazy baseline. Then... we realized it was problem of the injector, we perform the mainteinance and the things went out normal again. Apparently it was contaminated, the needle was not straight so it was also matter of a descalibration of the carrousel of the autosampler.

But now... the problem appeared again in a very short time...And I dont know... what the cause could be! I need your help, thanks... should it be a new contamination of the injector?? Or what do you think??

[mattmullaney]

Hello Dory,

Interesting Problem, but I am confused.

1. The Waters 2489 is a dual-lambda UV detector. Is the difficulty with this detector, or the RI detector (a 414 or 2414, if it's another Waters product)?

2. Good News...I've actually used the Bio-Rad Aminex 87H column. The packing material for this column is, shall we say, Challenging to Use. Are you sticking to what Bio-Rad suggests in using this column? The packing is quite fragile...must be treated with great care.

More details concerning the separation and the samples would be helpful...for this work it may pay dividends to ignore Waters' recommendations--if you are using an Alliance HPLC--on composition of needle and seal wash. Aminex 87H will not tolerate any organic eluent at all...as I recall...so any chance of mixing methanol or acetonitrile into the eluent flow path, however small, probably aught to be eliminated.

Anyway, see what you think...good luck! I'll see if I can dig up Bio-Rad's info on the Aminex 87H so I have it when/if you write again.

[Dory]

Hi Matt, thank you for your interest.

By now i am using a RI detector from Shodex and an ultimate 3000 HPLC equipment from Dionex. The samples are methanolic extracts from potato sugars, I rotoevaporate but sometimes are evident methanol traces in the chromatograms. Sulfuric acid 10 mM is the mobile phase, temperature from the oven is 18°C, temperature from the detector is 35°, flux is 0.4 ml/min, injection volume is 20ul and the time per sample is 35 min. Actually, as a cleaning phase the manual suggest to use 5% acetonitrile in 5mM sulfuric acid.

I have been thinking that the problem consist on injector contamination...the coincidence is that as soon as the techniciar check the injector in the autosampler, everything starts to run out again well. When I change the column for a new one... the problem in the baseline and the sucrose ghost peak persisted.

Thank you for your attention, any further question I will be pending.

[mattmullaney]

Hi Dory,

So it's a Shodex RI detector whose celll is maintained at 35 degrees Celsius, a Thermo Dionex UHPLC (or is this an analytical scale instrument?) whose column heater is kept at 18 degrees Celsius and a flow rate of 0.4 mL/min of 10 mM sulfuric acid, and apparently these days one can employ acetonitrile, IPA and ethanol at certain levels and times, although not methanol, ever, with the Bio-Rad Aminex HPX-87H. Got it.

A few questions:

Why do you run your sucrose separations at a temperature so much lower (18 degrees Celsius) than the recommended values (50 - 65 degrees Celsius) for the Aminex HPX-87H?

At this low temp and flow rate, what is your backpressure reading? There's kind of a strict limit on the backpressure these columns can tolerate, 1500 psi as I recall.

What type of injector does the Ultimate 3000 employ, flow-through needle or fixed loop?

What type of injections are you making, full-loop or partial-loop?

If you exchanged columns, agreed, a possibility is a dirty injector. Are you using some kind of seal wash that contains no methanol and/or a needle wash--I've not the foggiest how the Ultimate 3000 works with respect to these types of solutions. Looks like you can remove the column and purge the injector, and perhaps the sample loop as well, with the Ultimate 3000, most likely warm water (40 degrees Celsius) would be fine for this, to remove sucrose.

Probably an entire system flush with warm water containing 10-15 percent by volume acetonitrile is also in order, again without the column in place.

Please, see what you think.

[Meerkat]

Dory your problem with the analysis of monosaccs and disaccs is typical of what many investigators encounter. The HPLC "sugar" columns are notoriously difficult to equilibrate and prepare for injection because the column phases are plasma gels rather than static phases. Aminex H+ type column is a suitable one for your use, though I'd really suggest the Aminex Ca++ type as it affords much better resolution for sugars.

This will help fix your problem: I can say from experience with the chrom analysis of sugar alcohols (e.g., mannitol, sorbital, xylitol) and using "sugar" type LC columns (with Pb++ as cation exchanger) that it helps tremendously to heat the "sugar" type column up to ~70C. So, go prep a liter of your 5mM acid mobile phase and set on low flow ~0.7 (in fact perfusing your column with mild acid will help to regenerate the resolving capacity of your column), and task your RI detector to "monitor baseline", allow these parameters to run overnight.
At first light notate the apperance of your baseline--if flate/stable, your system is equilibrated, then inject a "blank", then inj your Std at the elevated temp and flow of 1. Things turn out OK?

[Dory]

Hi.

So I guess I need to better prepare the sample in order to eliminate completely the methanol in the rotoevaporation process, but it will be difficult since is an old rotoevaporator and an old pump so it will take a long time to do it.

I run the separation at low temperatures because first in 87H sucrose inverts and in the other hand, at these lower temperatures I get a better resolution. At this low temperature and flow rate, pressure can be no more than 950 psi.

According to the manual, the autosampler uses the in-line split-loop injection principle… as I told you, I am a plant geneticist so I don´t have many concepts inherent to chromatographic equipment clear, I am not sure if the injector is flow-through needle or fixed loop. In fact the seal wash contains 10% isopropilic alcohol…

So I guess It would be fine to use an acetonitrile solution to clean, but at a higher temperature it would be better. I will try it then…

The thing that is happening to me is that sometimes I let the column to equilibrate for quite a long time and the base line works ok, but sometimes I let the column to equilibrate for quite a long time and the base line in the injections don´t work ok. So I haven´t really identifed the detail that really help me out to obtain a nice base line. But of course I have realized that now the column lasts more time to equilibrate than when I started to use the equipment and the column.

Thank you very much for your opinions… I will keep on telling you how the things continue to go on.

[mattmullaney]

Hi Dory,

As to the sample prep...if the methanol peak is resolved from sucrose, are you really sure that you need to bother with tweaking the prep routine? Just asking the question.

Meerkat has a point in choice of column...budgets being what they are these days, I don't know that you can switch over to an Aminex HPX-87C or to a Thermo Dionex column that may be more appropriate for sucrose.

In-line split loop is a form of fixed-loop injector...very good of you to look and thank you. Now, see how large in volume the loop is and ensure that you are injecting no more than one-half the loop volume. Since you're doing work with sugars, it may also be wise to use polymer sample loops and injector assemblies, more in line with what you would see on a typical Thermo Dionex Ion Chromatograph...rather than a typical HPLC. Again...budgets being what they are...the plumbing is relatively easy to change if you desire to and can afford to, the injector, well...that's a bit more money, but polymer won't adsorb sugar or protein to the same degree as stainless steell can even after passivation.

I don't know the HPLC that you are using...in the world of Waters instrumentation, needle wash and seal wash have very different purposes...needle wash helps remove the injected sample solution from the outside of the needle, seal wash protects the pump seals from abrasive damage caused by precipated salts used in the mobile phase. As with the type of injector, see if you can learn what the seal wash does in your UltiMate 3000...most likely Thermo recommends solution composition, as well, for this solution. If the seal wash comes into contact somehow with the eluent flow path...seems that acetonitrile would be a good choice as the column stationary phase is incompatible with alcohols.

Best of Luck...Keep With It!