Dear Sangita,

Are all solvents HPLC grade? Is water Milli Q purity?

I have done lipidomic studies using aqueous ammonium formate in 20% isopropanol and acetonitrile in 20% isopropanol, with no problems. 

My guess is that you should run your solvent mix (using the gradient elution program) several times until the whole system (injection system, injection line and column) are cleaned. Probably those parts were left dirty from previous users and now you're cleaning the system and hence why you're seeing noisy spectra.

Hope this helps.

Best regards,

Ana

I agree with Ana, especially if you are using high res/high mas accuracy instruments. You are going to want to use high purity solvents for LC-MS. We use Honeywell LabReady Pre-blended for LC-MS solvents on our Sciex 5600TT.

Thank you for your suggestions. I am using HPLC grade solvents. Right now I am cleaning online with gradient run. I do not see anything getting improved. The noise is not improving. I figured out that the noise starts coming off once the percentage of organic (B) increases (>35%). I individually checked the contaminants on the solvents. However the gradient run is giving me many unknown mass spectral peaks. I am just wondering if the problem is with gradient solvent system or in line contamination. In line contamination should be flushed by now.

I am using 5500 QTRAP.

Any more suggestions to look for?

Thank you all again.

Sangita

Two ideas comes to my mind, first one is the column release some trace of a past analyte at a certain solvent hydrophobicity or with a particular solvent (that's would be isopropanol, apparently) so is a not so clear column but if you say it you cleaned the system with the solvent mix and all is well this isn't the case

So the other idea I have is at a certain point in the gradient the analysis mixture isn't fully soluble into the solvent mix anymore so the noisy data. It could also be the sample isn't fully soluble only at some elution solvent mixture and is fully soluble with the starting and ending solvent mix. If anything else fail I suggest to prepare some solvent mixture starting with 90%A-10%B as first mix and then 80%A-20%B and so on up to 10%A-90%B and check the solubility and turbidity of your sample in all those mix.

Hi Manilio Calvi,

The second one looks more convincing to me. I am still unsure what's going wrong. I checked the solvent mixture in a test tube as you suggested. I have not tried mixing of solvent percentage-wise though. It all comes after your organic percentage reaches >35%. I am guessing the same if something is happening to solvent mixing or I do not know. However the peak shows constantly without any change in intensity. The peaks are in the range of 500-750 and 850-1000. The two clusters of peaks are always their with the gradient. 

I am trying to figure out reason though unsolved yet.

And to let you know, even those peak show up without column.

Thank you for your suggestion.

Hi Sangita,

I suggest you just take trial on other instrument. The tubing, nuts or joints what you are using in HPLC may too old and contribute high baseline on increase of %B.

Thank you Nitesh. This could be one possibility though my instrument is new (few months). This should be other good suggestion. 

A few considerations: passivate your system; ensure you mobile phases (particularly MPA) is thoroughly degassed and don't rely on the instrument degasser to be sufficient (are you observing stable system pressure?) and use a  new column.  If you are seeing ghost peaks without a column likely you have contamination in your flow path. Generally there will be a solvent inline filter post-pump and pre-injector/auto sampler; inspect the filters for debris and sonicate the parts in 50/50 IPA/water for  5 to 10 minutes and replace the filter if possible. I had a similar problem when another chemist used a high salt MP and did not flush properly.

have you tried to direct infuse the a solvent mix over 35%B directly into the MS to see if the noise happens even this way? If so it is either a problem with the solvents or the MS. In that case you could try to do again the solution starting with new bottles and try to reinfuse.

Hi Calvi all,

This is an update. I have been trying all this before with phase A and B direct injection. It did not show such noise in direct injection. It is amazing. Today I tapped out the exact solvent that showed noise through LC but not throug direct injection. 

Now,

the same solvent showed noise with LC-MS however the same solvent tapped out real time and injected directly, the noise is not there. 

So through LC there is noise, but no noise through direct injection. It is mess now. Where is the problem it is hard to tell? Everything same, noise through LC (with or with out column). Direct injection does not show the noises.