Please check the quality of reagent and water is it HPLC grade ? Also please conform column details . Most of the time silica based HPLC columns are not comfortable for wide pH range and highly aqueous mobile phase.

Check whether your column is equilibrated. Inject blank. Try to inject smaller amount of sample (5 mcL)

1. Before injecting your sample, allow some time for UV detector to be stabilized maybe for 1 hour. if baseline dirft still persist without the presence of your sample means that it has nothing to do with your analyte.

2. Maybe your "pump-mixer" does not mix your mobile phases pretty well. HPLC grade ACN normally has a very low UV absorbtivity but H2O don't. If these two entities do not mix properly (eg; the 1st second contain 8% ACN but the next 3 sec contain 5% ACN and then go back to 8% again and so on...) baseline fluctuation can occur. What you can do is that, you can pre-mix your mobile phase composition manually (mix 8% ACN in Buffer) and let it flow through single tube channel. Baseline integrity will be restored. That's what i think if your mixer is the culprit.

3. If the baseline still fluctuate, I speculate that something is wrong with your UV lamp or may other factor that have been mentioned by our colleagues.

# For the baseline at 7th min, is it a gradient drift? Are you using isocratic or gradient elution? Please, let us know.

Dear Carlos

Baseline drifts in chromatography depend on several random events related to the experimental conditions applied. We can say (in a very general description of these inopportune events) that baseline drifts depend on uncontrolled changes in the experimental parameters which also determine a reduction of the analytical accuracy because chromatograpic peaks show modified shapes.

The scientific literature describes many mathematic approach to correct (i.e. to reduce) baseline drifts so to obtain the enhancements of analytical accuracy. Some of theme though highly efficient are quite difficult to be applied because they require complex mathematic procedures. Conversely a quite simple and reasonably efficient approach is described in the following reference:

Christensen JH, Tomasi G (2007) J. Chrom. A 1169:1

Basically this method consists of a differentiation of the time series of the n-point xi, yi values present in a chromatogram. We have to save the chromatogram in ASCII format to be retrieved in a common spreadsheet such as Excel where we perform the differentiation.

Good work

Mauro Mecozzi

At 289nm, this shouldn't be an effect from the mobile phase. Both kh2po4 and acn have very low uv cutoff values!

What is you diluent?

Step changes in baseline for absorbance detectors are commonly caused by gas bubbles building up in the detector cell. As the bubble grows, it progressively occludes the optical path. When the bubble gets sufficiently large, it dislodges from the flow cell and the baseline returns to a lower absorbance value. Thorough degassing of all mobile phases each day before operation will minimize bubble problems.

Baseline drift is more difficult to trace and can come from a variety of sources, including gas bubbles. If after degassing, you still have baseline drift, you can try a some of the techniques suggested by others in this thread. Drift can come from highly retained components that slowly bleed off the column. Multiple injections of a strong solvent can help flush out strongly retained compounds.

Curtiss' answer is exactly what I was going to say. Even if you have an inline degasser, you should try sparging the MeCN with He. If you don't have an inline degasser, you should absolutely sparge the organic solvent with He. The random steps you see look just like issues with residual gases in the mobile phase. The baseline drift, on the other hand, looks more like you just need a little more equilibration time on your column before running. In any car, a little baseline drift won't kill your method, but random jumps in the baseline will.

Also, what's the difference between the blue and black lines in the last chromatogram?

Hi Carlos,

The method you are using is a gradient or an isocratic?

Generally this kind of behaviour happens in case of gradient methods.

If the method is an isocratic I will recommend you to run the sequence with premix mobile phase instead of binary.

the sample your a injecting is a high load or low load.

If that is at limit concentration then you may face a problem of base line drift.

Please check the system performance also. Check the delta Psi for your system.

Waiting for your results.

Check the pump, pistons and frits to the column. May be you need to clean the pistons and the filters dipped in mobile phase bottles, sonicate them in MeOH. Then purge the HPLC systhem.

Do not worry about other methods and the drift existing there or not. C-18 column is a very common column in LC analysis. It is always used very extensively. Therefore, more than often it is not properly washed (till you get a stable baseline). See if this is the problem, otherwise the column is heavily contaminated and needs to be replaced.

Have you tried adding TFA to the mobile phase? Up to 1% would help stabilizing the baseline.

The HPLC analysis in blood samples require some experimental condition such as the precolumn to prevent contamination.

To eliminate these baseline problem, you should keep the mobile phase more equilibrium time (24H) at weak flow. think about the purity of the water it must be millpore grade and ultra-pure water and verify the purity of phosphoric acid.

I the baseline is generally not very stable if you inject a serum samples.

If yours injections are reproductible and and you obtain a good linear regression of your detector so it will be ok . You must do a statistic study of the precision of your method.

Good work.

Ok, Thank you every one.

My injection volume is about 30 mcL. The sample preparation is based on a protein precipitation with cold ACN (ratio sample:reagent = 1:1), and another step based on a dilution of the sample with mobile phase (ratio sample:diluent = 1:3).

I think I know what´s the problem. It has been a long time since I use my UV-vis detector and I think air bubbles have generated.

If my theory is correct, air bubbles will dissapear with time.

Looking at your chromatogram, I think that you need to use an on line degasser to avoid your troubles.

Charles, the last two chromatograms are a sample of drift and unstable baseline effect on my chromatograms. They are not as stable as the other chromatograms.

Hitesh, is an isocratic method and I think there is no problem with pressure (delta PSI is at low and stable values). I think pump works well.

Nadezhda, i will proceed to clean all pump components like pistons, filters and sonicate them.

Boubakeur, I work with milli-Q water from a water purification system verified every week. ALL my solutions (aqueous and organic) used prior to LC are filtered with a Nylon filter 0.22 mic.

My LC system have a vaccum degasser and I think it works OK.

I´ll tell you. Thank you very much.

Oscar let me make you one question please.

Why TFA up to 1% helps to stabilize baseline?

I squggest also to test your UV lamp. Such troubles can occur when it becomes old.

the chrom you posted look different

in the third I see a very unstable baseline. I had similar things:

1- because of the samples (very dirty matrices such as soil samples)

2- because of non miscible solvents in the system (e.g. running toluene-AcCN as a mobile phase with traces of water stuck in the systems)

The second option could also explain some random steps.

If this is the case I recommend flushing the system with isopropanol.

Sorry Carlos, TFA concentration should be up to 0.1% and not 1% as I wrongly indicated in my original post. When running a gradient water/acetonitrile. the baseline rise is the consequence of difference in absorbance between acetonitrile and water and is visible especially at low detector wavelength (check that if you have a diode array detector). The absorbance difference can be compensated by adding more TFA to the water/aqueous reservoir in relation to the acetonitrile bottle, in a range between 10 to 25% more (fine tune to find out which level flattens the baseline better).

Hi everyone. The drif, unstability and steps that appears on my chromatograms have dissapeared. I don´t know why...but I think about air bubbles on mi cell detector that finally scape. I show you two chromatograms, one is a blank of treated plasma (blue color) and the other is spiked plasma at 35 ppm of meropenem(black color). Perfect Baseline.

Thank you very much.

There are two things can you do to eliminate the unstability of baseline

1- degassing for mobile phase ( hellium gas)

2- try to chack the flow cell of detector if it has contamination

Hi Carlos, I recommend you this document for future problems with the HPLC.

It's good to monitor the pump pressure as well because if you have accidental leaks or clogging somewhere in your system, it is easier to track the problem. I have noticed drifts, steps and weird shapes of UV-absorption baseline along with the anomalies in pump pressure that were caused by minor leaks that were hard to notice just by routine checking. The statistics of pump pressure for a certain column enables to estimate the health status of that column/pre-column as well - when the pressure starts to rise higher than normal, it is a good indicator to clean your system or replace the pre-column.

I think the cause of your cause is buffer and air bubble in mobile phase. Did you test with another buffer at the same pH? I also have the same problem when i use phosphate buffer for mobile phase, especially when the concentration of buffer is high. another problem i guess your degasor did not work properly. You can observe much air bubble when you mix ACN with phosphate buffer.

Check pressure ,degassing of mobile phase or leakage of valves

It seems it was only a matter of time and that now your system is well conditioned. however, I think it is a good practice to:

- perform a system conditioning with the mobile phase (to completely flush the system and let air bubble exit the system, they guves problem especially when stuck-up on the cell detector or in the pump)

- use a proper reference wavelength. It helps both when you are using mobile phase gradients and when you could have some absorption from the carrier

The seconds is often the solution to baseline problems when the first thing are fixed

The system should be at equilibrium before starting the analysis then if such this problem appear again that is possible due to changing the room temperature or and unstable mobile phase composition or/and its pH. to overcome this problem perform conditioning for your system at higher flow rate for 5 to 10 minutes the get back for your interested flow rate and be sure the pump's pressure is stable.

N.P: normally you can see this phenomena in gradient HPLC system

Did You make properly centrifugation of the blood or not (because the proteins in blood can absorbed about 280 nm in very little concentration)?

I suggested two steps centrifugation (about 15 000 rpm 30 min., and obtained supernatant again at 15 000 rpm 10 min.).

I supposed that meropenem is low-molecular weight substance (I didn't check).

Also, is purity of Your water sufficient or not?

Best regards

Thanks.

So, the problem is resolved now. There were air bubbles in UV-vis Detector.

OK.

Best regards

Probably air bobles or any kind of impurity in the system

Hi Carlos

This might be caused by several factors. Please check the following points:

-First, try to monitor the pressure

- Second, make sure you give your instrument enough time to equilibrate your column.

- Third, try to filter your samples before analysis and use a freshly prepared mobile phase.

I hope this can help you.

Regards.

Thoroughly degas your mobile phase, this may be outgassing during the run. Baseline drift is generally column temperature, you didn't say whether you were using a column heater to maintain constant temperature.

Air entrapment in detector may leads to base line drift. So, check it

From my experience, using PC, TLC and HPLC, the fine and fixed adjustment of internal and atmospheric (conditioned) temperature is a dominant crucial factor, perhaps this help you in solving the problem.

Another factor is the readjustment of the dilution percentage of your mobile aqueous phase

These effects look like the result of gradient elution and refractive index effects on the baseline of the detector. Repeatable patterns like fig 1 and 2 make this a likely effect, an

lthough it could involve some elution of a multitude of materials present in the sample.. It is easy to eliminate bubbles with degassing and backpressure on the detector. Bubbles do not generally produce reproducible effects.

I agree with the previous suggestions.....I think the main problem is sample filtering.....other source of problems: Check the actual pH, do you measure the pH in the aqueous mobile phase? after adding ACN the pH changes (I do not remember if up or down) buy anyway, you are very close to the minimum limit of acceptable pH for that column.....you could have blooding of the C18 ligands....

I feel that the drift is detector drift. The flow-through cell needs cleaning or micro air bubble sitting in the pump display head.

manohar

I fully agree with the proposed solutions. You can also check the lamp of ​​your detector

This looks as if there was a gradient flow which was suddenly altered. But assuming that that is not the case and you were using a isocratic mode of elution, I would suggest a. Purging your system; b. Degassing the solutions; c. give a cleaning step in between runs. Also, you may go for a blank run(if on a gradient mode) to see if the pattern continues. You may consider monitoring the back-pressure during the runs as an alteration of the back-pressure is at times indicative of the cause for a baseline drift. I hope this helps! 

I do the followings.

1st; fitration of the eluent by using aspirator such as Eyela (A-1000S).   Use cellulose acetate for aqueous and PTFE for hydrophobic eluents, respectively.    Please, wet the PTFE membrane with ethanol before use.   Please, use the better reagent; i.e., non-ionic detergent Brij58 is better of Wako than of Sigma.   Filtrated eluent is transparent.   Please, use the good pump; i.e., Shimadzu pump is reliable to me.

2nd; degass the eluent with aspirator for 10 min.   Glass ware of DURAN is good.  

3rd; please use the good injector.   I do not like Rheodyne, but it works relatively good under the pressure of 12 MPa.   If higher pressure is to be used, plaese use the high pressure six-bored high-pressure valve (GL sciences Inc., Japan) .

4th; please use the good columns.   I like Machery-Nagel's Nucleosil and Nomura's Develosil.

Please, use the good tubes (plastic tube is not good) with good stainless (SUS316).

I hope you your good job by using HPLC.

.         

Degas  mobile phase throughout the run, check your column  conndition and pH of of the solvent. Make 15% ACN and let know the results

I think the main problems caused by the sample filtering, Back-pressure on the detector or eliminate bubbles. It is nessesary to degassing and backpressure on the detector. Bubbles do not generally produce reproducible effects.

good luck

hi your column might be ruined

I fully agree with Sukanta Bhattacharya. Once you have clean perfectly well your system check the pressure after each injection. Is it possible to have any precipitation inside your column? Do you inject blood directly or do you make an extraction before?

there must be an extraction procedure before injection into hplc system.  Its crime to inject blood directly into the system. Finally check your system pressure if it is stable than ok otherwise replace the check value.

It can be the problem of sample concentration. How about your retention time, does the peak appear in the correct retention time you had done before? The cause of this kind of problem based on my own experience:

1. Low sample concentration

2. Flow rate (it looks more than 0.7ml/min)

3. Wavelength 

4. Solvent gradient

I have never encountered problems with injection of  low sample amounts. In my opinion only gradient elution methods cause problems with baseline stability. If that is not the case, i propose you:-

1. Equilibrate the column with mobile phase for 30-60 min

2. Adjust the pH of the buffer with an equimolar concentration of phosphoric acid. In your case use a 0.03 M phosphoric acid.

All the best

I agree with Alex. Moreover check the purity of the mobile phase starting with "whether the phase is properly degassed" etc. If required open a fresh bottle of the mobile phase solvent.