When using HPLC, how do you deal with split peaks?

Dear peer,

By making a look on this, u'l gain something re-leaf from your query....

best regards.......!

To understand of your question i need few details. What kind of column and mobile phase using and what u want to analysis.

Peak splitting is synonym of poor chromatography. You better search for possible causes and improve peak shape rather than performing approximate peak integration.

The most common causes for peak splitting are i) too strong injection solvent compared to mobile phase composition at elution, ii) column channelling, iii) partially clogged part of the system (column, filter etc.). A too high acquisition rate may also cause jagged peaks.

Check your medium, smoothness of your medium and try to carefully re run with lesser quantity of injection the problem is with your protocol i think

Can you provide your method conditions please? It will help us give a clearer answer.

Sometimes it is because of fluctuating pressure.. Always check back pressure... If u add areas, it would not be goodbec ur chromatographic parameters will not show good results,You will have unreliable data. In order to fix problem, if indeed there is high pressure, u may change ur flow rate to a slower one, or adjust the composition of your mobile phase. If this is a an official and validated method and u keep getting same results, u may have a clog in ur column or you would need to reasses ur system

It's not okay to add the areas, because you do not know if the peak is intact or perhaps he is adding a compound of degradation or impurity analysis matrix. The most convenient is to repeat the analysis until a single peak. Review by 3D software if there may be more related peaks to the main peak. I agree with what you say Mr. Glauser, regarding analytic review the conditions of the method.

First if all, I think you need to tell us.. what is your desired separation? as in you are working on small molecules or proteins or oligos??because sometimes you get to see split peaks because you have some kind of mixture which are very close to each other.. then you have to work on your method gradient of elution.. and depends on what column you are using.. so if you specify these details .. it will be easy to help you...

Check the HPLC system first (pressure fluctuation, injection, plumbing, flow rate,gradient mixing). After, whether chemistry of your sample is compatible with your eluent (for example solubility, dissociation). Check the column performance according to the column manual . Maybe your compound has an isomer. See Veronika Meyer's book on HPLC.

If it is reversed-phase system, i.e C18 column, use a simple compound such as naphthalene to test. if peak split observed, it is very likely a physical reason, such as cavity of column. You can use some sationary phase material to amend the cavity on the column head.

The chemical reason is the mobile phase pH or solubility of the analytes in the mobile phase, or other reasons indicated by other experters above.

Please inject in the same condition your standard. If the peaks are still split you have a problem with your system. If the standard peaks are not split you have more compounds which co-eluted at the same retention time. In this case you should develop other protocols for analysis by changing the column, mobile phase composition or gradient.

There are several reasons that cause peak splitting among then are:

pressure fluctuation so check the pumps and the instrument and make sure the the flow rate is constant and if you work in gradient mode them make sure that the ration of mixing the solvents are correct as desired. Also make sure that the column is not clogged or there are column channeling. The analyte(s) should be completely soluble in the eluent and mobile phase since partial solubility of the analyte(s) may create a problem and cause splitting. Also read the comments of Mr. Mihail Beldean-Galea regarding the purity of the standard and that the splitting of the peak is not due to impurities or degradation product(s) in the standard.

Mihail has the best advice as to where to start with troubleshooting.

Do you accurately wash the column with the more eluotropic solvent, of your elution method, between two consecutive injection? I'd advise to check it, at first!

Firstly I need to Known about the sample. On the other hand the computer analysis is based on electric signal. Thus, the integration system see this signal. One peak for your eyes is not necessary one peak for electric analysis.

The shortest and best way for this troubleshooting is to inject a standard solution under the same operating conditions as suggested by Mihail.

Injecting in a solvent stronger than the mobile phase can cause splitting. Hence, keep organic concentration in the sample solvent less than Mobile Phase solvent. Kindly, check the pH of both mobile phase and sample solvent.

Most answers which have been recently given are very nice. But those focus only the comparison with the standard compound. Since one peak with split peak close to target peak (possibly like shoulder), This problem could first solved by chromatographic separation. However, using mass spectrometric technique (GCMS or LCMS), one can easily identify or decide whether to sum up the target peak with split peak. Using GCMS or LCMS, selected-ion-monitoring (SIM) mode can easily solve such problem because one just analyze the target compound with a mass-to-charge (m/z) ratio. Then the target peak and split peak could be easily separated. This is one of the advantage of mass spectrometric detection rather than UV or DAD detection methods.

Thanks a lot all for your excellent suggestions. I will check the system and do the maintenance then analyze again!!

Pl see points given below:

Complex sample matrix or many samples analyzed - likely column contamination or partially plugged column frit.

2. Mobile phase pH > 7 - likely column void due to silica dissolution (unless specialty column used, Zorbax Extend-C18 stable to pH 11)

3. Injection solvent stronger than mobile phase - likely split and broad peaks, shape dependent on injection volume and k value.

. Column “secondary interactions”

2. Column packing voids

3. Column contamination

4. Column aging

5. Column loading

6. Extra-column effects

Due to long use of column or more number of sample analysis also the reason of peak splitting. Go for column flushing for longer time.

Split peaks can be caused incompatibility of the injected sample and mobile phase or poor sample solubility.If this is the case before injecting ur sample check ur sample with mobile phase whtr getting dissolved in it. If it completely soluble in mobile phase then ur sample compatabile with ur mobile phase. And second thing contamination in guard column. Try to remove all particle.

Split peaks can be the consequence of an erroneous injection. Thus, it is necessary to minimize the injection volume by utilizing loops as little as possible. Fill completely the injection loop when injecting, as split peaks can arise from incompletely filled sample loops.

If you're using an autosampler, split peaks can be the symptom of a failure to the injection system. Check the needle and be sure that the aspired volume of the sample is correct. Differently, split peaks can arise from the incompatibility of the sample solvent and the eluting system. Dissolve sample and inject it in the solvent (mixture) corresponding to the run starting conditions. Example: If you are using a gradient of 5-50 % acetonitrile in water, in RP-HPLC, do not dissolve your sample in a water-immiscible solvent. You should be able to dissolve it in water containing not more than 5% acetonitrile.

For more information you can find several HPLC troubleshooting guides on-line, such as the following:

Francesco Addeo and Gianluca Picariello

Possible resons of peak splitting are:-

1. coeluting compound (matrix component having similar retention time

2. Overloading of sample to coulmn

3. Incompatibility of sample diluent and mobile phase

follow Mihail and Aslam's advice

Well, when a peak is expanded, it is necessary to change the mobile phase, or to choose to quantify only that one that corresponds with its retention time

Also you can use chemometrics to find the peak area

We had such spectra and we could interprete them by the use of calibration-comparison with peaks originated by polymer molecules of a well defined molweight.

Most probable causes have already been well covered above. However, another cause of split peak (e.g., in the absence of above listed causes), which you may look into may be the existence of a compound in more than one form like a compound may partially exist or get converted on column to a mixture of keto and enol form, diasteromeric forms due to presence of an active hydrogen etc. A closer look into the structure may guide you, if this is the case.

Some clues: If the problem is due to a bad column performance you will generally see similar distortion or splitting in most of the peaks in a run. If it is related to high strength of sample as compared to mobile phase, the problem of splitting is generally observed with early eluting peaks.

Well good suggestion have been given by everybody. Well these problems are often encountered. Its better to have your column wash in bidirectional mode as well as the complete washing of the injection port upto the sample loop by acid-methanol cleansing as well as detector washing by the same system and complete washing later by water and methanol. Please do often check the compatibility of your results in at least to duplicate columns.

Peak splitting is a common observation with larger volume injection, the portion reaching the column head undergoes chromatography while the portion behing has still to reach the column. Situation is worsened by column void and tube crimps (by over tightening of nuts).

Please read "behing' as "behind."

....and one thing..Do u use the precolumn? If the answer is yes, then replace it with an other new (or try to inject, for cheking, without precolumn)

Peak spillitng may be due to:

contamination of the mobile phase

partial blocking of the analytical column or guard column

problems in the pumping system

Peak Splitting is a bad chromatogram

You will change your mobile with slight modification in the basis of composition. so it can be removal of peak splitting.

You may also change the pH of mobile phase it may help to solve your problems regarding Tailing or fronting issues.

I think column washing may solve the problem. It is first necessary to determine whether all splitted picks giving same pure protein or they contain some contamination. If it is same protein, may be some composition of your buffer changed. If it is contamination then you have to wash you column vigorously.

Just back flush the column for half an hour and reconnect again to the right direction, the splitted peak will become one peak, otherwise repalce column.

If every peak that you see in the chromatogram is a split peak then chances are the column packing has settled and you have a void. My suggestion would be to try a new column. Keep in mind that columns are generally intended for the flow to be in one direction (see arrow) and reversing the flow, while it may help clean the head of the column, it can run the risk of repositioning the particles and actually create voids. Depending on what the stationary phase is, it is possible to purchase the same or equivalent packing material and repack the head of the column with a slurry of the material if you have time and want to save money on columns.

First, I think that is not acceptable to sum the splitted peak as it decrease the repeatability of the method and give unquantitative results. So, you can regard the following: 1- wash your column well 2- study the effect of pH on the resolution of your drug 3- Try to inject another compound that previously injected on this column

4-Confirm that the drug is not diasteriomer 5- Increase the flow rate, this may be improve the resolution 6- Try to apply a reported method for your compound using the same column (find one method that nearly have the same specifications of your column) and compare the results. If the peak still splitted, the authentic sample may be degraded (check if it is photosensitive or hygroscopic).

Good luck

Column washing eliminates the peak splitting, which resulted from a contaminant on the column

As another a suggestion, if you are using PDA or MS detection is to compare the spectra at different time points along the peak(s) by overlaying their spectra to see if they appear to be the same compound or co-eluting compounds. Another thing you could do is to try preparing your samples and standards in mobile phase that is either the same or preferably even weaker than used in your initial HPLC conditions. The idea for doing this is to concentrate your analyte(s) onto the head of the column in a tight band prior to eluting with stronger mobile phase. This technique can be used to improve resolution.

Wash the column thoroughly and repeat your experiment. At least you will be sure that the split peaks are not due to column clogging or other contaminants or left overs from previous experiment.

Use mobile phase used in the HPLC for the final dilution of sample

If your compound is charged (amines or acids), or can form hydrates (like aldehyde vs. geminal diol), they may exist in multiple charge states or chemical forms that can be partially resolved on the column. You could try changing the pH and/or concentration of your buffer, or move to a different mobile phase system (like ACN to MeOH or vice versa).

this 2 links will help you

wash the entrance frit by reversing the column, but consult the column sypplier if this conlumn can be reversed. Usuall it can be done. And teke care about you injection.

I hope these links will be helpful for you,

Eliminate contamination and clogging.

If this is not common practice, it somehow indicates the decreasing separation efficiency which might be caused by many factors. In the meantime, the simplest way is to try to reduce the injection volume to see if the split peak is gone.

Conversely your problem may be chromatography that is too god. In the case of some compounds there may be structural isomers that can be separated only by very careful selection of chromatography conditions (I give perfluoro-octanoic acid (PFOS) as an example). Under normal high-through put chromatographic conditions you will usually only see a single symetrical peak and so no problems - under soecifc conditions you can resolve the isomers and it looks like peak splitting or poor chromatography.

Read this link by John Dolan it is excellent and advise you on how to overcome the splitted peak.

Splitting in peaks means poor separation. That's may be do to filtration, column type (you can revers the column and see) or change mobile phase ratio.

To accurate quantification, you must select the one of these peak. So, you must sure which peak is your analyte's peak. In this cases, you can change your mobile phase, flow rate or even lambda Max. It depends on the nature of your sample and analytes. Anyway, if you wanna more help, please indicate your analytes category and the Elution condition.

Good luck

This might be due to column ageing and diluent incompatibility.

Several possible reasons have been addressed well-one simple issue is there could be degradation of the desired material. Without knowing the type of material and breakdown potential from extraction and storage (before and during the run!) it is hard to say. We add the peaks together-manual integration. Paul

This is not a good sign. of separation chromatography.

There were many reasons behind it like: 1) Column material was degraded 2) column material will be eluted out with sample 3) might be problem with mobile phase 4) tray to use some PH modifier to solve the peak splitting problems 5) Problem associated with fluctuating of pressure 6) might be an isomers of the sample to be separated together and 7) try to use mobile phase solvent for sample preparation. Try one by one you will surely get a success.

Perhaps this article will help you understand how poor quality method development can result in peak splitting. "HPLC Peak Splitting. Common Reasons For It ".

Clean your column with magic mixture of solvent for at least 30-50 column volume. i.e. ACN-MEOH-IPA-Water (30:30:30:10 v/v/v/v). 

I think you shoud deal with the splitting thing before the data analysis.We always would like to gain a single peak in most of the time , rather than a pooly separated peak which means it is not singular，even the area the same or similar to your job before.In this case,splitting is unacceptable to me,the data means nothing.

1.make sure your column is good.maybe you can use something else to test the column perform well or not.

2.Is your targe an ionic compounds which shows different form depending on pH？So,maybe the pH is beyond your expectations.

3.Sometimes strong solven(Methanol，acetonitrile) used for sample prepairing may cause the peak to split,especially in a  weak elution capacity moblie phase in RP-HPLC.

other reasons to consider next time...

I faced a similar challenge when analyzing for methyl orange. When the Injection volume is 5 microliters I obtain a nice, sharp and symmetrical peak. Doubling the injection volume gives me a 60:40 split peak. This happens with when I use amine containing compounds. Peak splitting seems to be independent of the initial concentration of your analyte. As long as I keep the injection volume low, I have no problem.

 Hope this was helpful?

Dear ,

Splitting come from the back pressure asserting by the flowing mobile phase if something try to stop it . Splitting of peak does not happen suddenly; it happen first as a tailing on the peak (fronting),  then as shoulder and finally appear as splitting of the peak . The reason usually happen when dirt accumulated on the column frit on-lit .

To repair splitting , back flush the column with suitable mobile phase , usually acetonitrile/water  while the oul-let not connect to the detector .2- use less viscose mobile phase. 

Regards

Here are some educated guesses on dealing with split peaks:

1. Wash the column. 80%H2O, 20%ACN v/v 15min. 50%H2O, 50%ACN v/v 15 min. Wash w/ % of H2O matching your buffer % of your mobile phase for 3 min and then run your mobile phase (helps avoid salt precipitation in column).

2. Flip the column.

3. Adjust/Change mobile phase ratio or mobile phase and change diluent

4. Lower injection volume. Lower concentration of sample. 

5. Check if split peak might be a degradant that is forming and not just a split peak w/ PDA peak UV profile.

The reason may be the formation of the void at the inlet of the column or blockage of the inlet. Try to reverse the column and wash with 70:30 acetonitrile:water for 10 min with 1 mL/min. It worked for me. the column working as it worked when it was new.

https://www.silicycle.com/faq/hplc/why-is-my-chromatogram-showing-peak-tailing-ghost-peaks-fronting-peaks-split-peaks-shoulder-peaks-or-rounded-peaks

https://www.chromacademy.com/chromatography-HPLC-column-abuse.html

go through these links it might help you

@ Venkatesh Mandari It should be done very carefully. Reverse flow on the column may damage powder filling. It may be disturbed and column may loose its separation performance. Reverse flow should be apply very gently and very low flow. And it should be done when other methods failed. Sometimes, exchange frits in the ends of column may help.

I would suggest to watch closely the mobile phase pH and make sure that your column is in a good condition. More details are needed, if you want a more specific answer.

Best!

Several factors may cause this problem among of these are:

- change of pH of the mobile phase specially in revered phase elution.

- the column may be partially damaged. Try to clean it by reverse flow of approriate mobile phase at very low flow rate.

I tried many of the other suggestions that did not help, however I did not try reducing the concentration of the component although that probably would have done the trick. Before that, I tried reducing the flow rate from 1 mL/min to 0.5 mL/min and that worked great, possibly the higher flow rate was overloading the column since it was an early eluting peak. Still needling to do method validation which may show this "trick" was not good enough. Just saying . . . .

you are adding contaminants to your "peak" doing this. Reduce volume, and understand why two peaks- have you done ms on them to know one is not a degradation issue, or Na adduct?