extra peak appeared before analyte peak

[tlili]

I am trying to develope an HPLC analysis method for pencillin G.

Mobile phase composition: 60 0.025M KH2PO4 adjusted to pH 3 using phosphoric acid 40ACN. Isocratic.
Injection vol: 5 ul at 1 ml/min
Column: Zorbax SB-C18 4.6 x 75 mm 3.5 um
wavelength: 204 nm
I got the analyte peak at 1.4 min . I have also tried other composition to improve the k' . However, as RT increases, the shape of the peak also got broader.

I have also tried using the above exact composition but at gradient, curve 11. Flowrate: 1 ml/min. Below is the details:


0 to 1 min 100 KH2PO4
1.2 min 60 KH2PO4_40ACN
5 min 60 KH2PO4_40ACN
6 min 100 KH2PO4

I managed to get the RT at 3.8 min. But the strange thing, I get another peak just before my analyte peak. I have tried a few concentrations 500 ug/ml, 250ug/ml, 25 ug/ml, 2.5 ug/ml, but the outcome is always the same. Any suggestions why this is so?

According to Agilent methods, their experimental conditions was gradient: 5% ACN to 60% in 10 min. Column wash: 60% B to 5% B in 2 min. Stop time: 12 min Post time: 5 min. Can anyone enlighten me how to go about doing. The analyte peak elute at about 7 min.

Any suggestions would be much appreciated.

[tom jupille]

What is the sample dissolved in?

[unmgvar]

Try to have a look at the Merck application No. 900246bdh also. they get 5.24 min. for Penicilin G. they use methanol (LiChrosolv® Cat. No. 1.06018) / 50 mM phosphate buffer, pH 3 (55:45).

As Tom suggested be sure that the sample solvent you use as a proper solvent strenght compare to your mobile phase. proper coroletion will improve peak shape overall especially tailing.

the other peak ,might be a degradation from your sample. can you also give more details on your sample preparation?
do you have a PDA? can you perform purity, and library checks? see what spectra you gets from your 1.4 min peak and from your 3.8 peak?

[tlili]

Dear all,

My sample (pure std) was dissolved in MilliQ water.

I tried running just the blank (MilliQ water). There was a peak appeared at the same time as the "extra" peak. But no peak at RT 3.8 min. Any suggestion if there is any method I can use to get rid of this peak or shift it further away fr my analyte peak.



I have PDA. Both the peak at RT 1.4 and RT 3.8 have the similar type of spectrum. Can I conclude that there are the same analyte peak?

I am thinking of trying out the mobile phase composition suggested by unmgar. How do I get a printout of this Merck application No. 900246bdh ?
Can I use KH2PO4 as the phosphate buffer? I am using Empower currently. Any suggestions, how I can go about performing purity and library checks?

Thank you for all your suggestions. Really appreciate it.

[tom jupille]

The reason I asked about the sample diluent was that a too-strong injection solvent can generate all sorts of peak shape problems. Since your sample is dissoved in water, I think we can eliminate that as a problem.

The fact that you see the peak when you inject only water suggests that the peak is either a contaminant in your water or, more likely, a "system peak" resulting from a shift in equililbrium. If that's the case, then dissolving the sample in your initial mobile phase would be the best approach. In fact, try that, and if the interfering peak disappears, then you have confirmed that it was a system peak

[pawan ratra]

Following steps may be taken to determine the origin of the extra peak:

Determine the spectra of the peak eluting just before the analyte peak. Overlay with the spectra of analyte peak. If the spectra are very different the source of peak is other than analyte peak. If the spectra is same the peak may be due to impurity or related substance of the analyte or carry over from the previous injection.

Case 1
If the spectra of the extra peak is different-
Inject 0 ul of injection. If the peak is observed at the same RT of extra peak and the area and spectrum is similar to the extra peak observed in the sample chromatogram, the peak is originated from the either system, method or mobile phase. If changing the mobile phase gives same results, the peak is from system or method. If changing the system gives the same observations. The peak is from method. (I believe this is the most probable cause as the method you are using is started with 100% buffer. Incorporate about 2-5%ACN otherwise C18 column may give you problems with 100% buffer-Column collapse).

Case2
If the spectra is similar to that of analyte peak-
The peak may be due to carry-over or impurity peak. Change the system and inject blank in duplicate. Determine whether the extra peak is observed or not. If no peak is observed. Inject sample and than blank solution again. Check the presence of extra peak. If present in the blank solution, the peak is due to carry-over and you have to use needle wash to minimize the carry-over.
if the peak is only observed in the sample chromatogram, it may be due to impuriy. The linearity solution you injected should give increasing peak area of the extra peak observed in the method.

Although you may have looked into the above strategy I think it is worth mentioning and going through it again.

Determination of peak purity in empower software-
1. Make a PDA method
2. keep the complete UV range
3. Open the chromatogram and integrate with the PDA mthod.
4. Go to integration table and select purity options.
5. Enable peak purity.
6. Select either autothreshold or Noise + Solvent. the default value of Noise +Solvent is 1.0. Ensure 100% of peak is selected.
7. In the chromatogram check the minimum noise region in the sample chromatogram.
8. Enter the noise region in the intergration table. I dont remember which exact sub-window it is. May be in Purity window or in noise window.
9. Reintegrate the chrmatogram and check the purity plot in result window.

[/u]

[unmgvar]

Hello tlili,

the first thing you should do is get the chromcircle CD from Merck or from who, ever represent them in your country.

Do this also from all possible column vendors, like Waters, Dionex Phenomenex, and so on. this way you will have a broader base of applications.

if you wish so you can e-mail me and i will send you this specific application.

[A.Nonymous]

What is the exact name of you analyte?

It could be that its penicilline G sodium, but also penicilline G benzathine, ...

you won't see sodium but surely benzathine.

[tlili]

Dear all,


I have tried injecting water as well as the mobile phase alone in duplicate or triplicate. On top of that I have also tried using the mobile phase to dissolve the Penicillin G. Unfortuately, I am still getting the same "extra" peak at the same RT for both water and mobile phase. The spectra for this peak looks similar to my analyte peak.


The RT for the analyte peak that I obtained from dissolving in mobile phase is almost similar to that of water around RT 3.5 min and RT 3.8 min respectively.
I am wondering if I can just ignore the extra peak since it is about 0.4 min away from my analyte peak? Any suggestions?

By the way, is needle wash = purge injector? May I know how do we go about performing "needle wash"?

Unmvgar had suggested using 55MeOH_45 50mM KH2PO4. I thought of trying out this method. However, I would need more info. such wavelength, isocratic or gradient, column/sample temp. etc. I would appreciate if he can provide more details.

The exact name of my analyte is Benzylpenicillin Potassium salt. Sigma:P8721. Can anyone teach me how to attach the structure to this forum ?


Thank you for all your valuable suggestions.