The method was invalid as shown. The consequences of using instrumentation and techniques without proper training often leads to erroneous data being collected and incorrect conclusions or quantitative results being reported. In your case, there are two reasons for this:

(1) The DAD reference wavelength feature on your Agilent 1260 DAD was turned 'ON'. This must be turned 'OFF', not 'ON'. To learn why, please read the attached article below.

(2) Your signal acquisition wavelength is very close to your Reference Wavelength. This guarantees that your data will be invalid as even the peak of interest will fall under the Reference range you specified (360,10), invalidating the method and data obtained.

• REFERENCE WAVELENGTHS (as used in HPLC UV/VIS): https://hplctips.blogspot.com/2011/03/reference-wavelengths-as-used-in-hplc.html

Those who use HPLC for the first time and/or who have not had any formal training in liquid chromatography (and especially in the use of a diode array detector (aka: PDA)) often make these mistakes. Clients use systems setup just like this one and are unaware of the problems caused until a negative peak shows up (causing them to question the results). The appearance of the new, negative peak is the hint that all of your previous data may be invalid and the only reason you discovered it now is because this time the system had to subtract-out so much data from your main peak that it ended up being negative!

Try to use the mobile phase as final diluent.

Thanks @NidalBatrawi. I tried with mobile phase as diluent but no improvement in peak.

Maybe there is no analytes in Your samples and negative peak is just something thats is coeluting in the same retention time. Try to add some of the standard to Your sample and see what happens.

Hi, could you please tell a bit more about the setup you use? Especially what detector and what sample/standard you use? What kind of column do you use (e.g. C18)? I suppose you use an isocratic method? A chromatogram, possibly negative peak in sample and positive peak in standard would be nice, just to see what you are actually talking about.

Thanks and best regards.

Thanks @Adam, Sample having analyte. And i am agree with you that, coeluting of other analyte may happen.

Thanks @Markus.

Detector is 1260 DAD, Standard is Piperine 95 %, Column is Zorbax Eclipse plus C18. And it is an isocartic method. And the Blank, STandard and sample chromatogram are attached.

Ok, thanks. At first I would turn off the reference wavelength for your acquisition. I can´t read it quite properly on your chromatograms, but does it read 343 nm/Bw 4 nm Ref: 360 nm/Bw 100 nm? This looks like standard settings. The aquisition wavelength shouldn´t be in the bandwidth range of the reference. I recommend to always turn this feature off, especially when developing a new method. The reference gets subtracted from your acquired wavelength and you never know if there might be an analyte that get subtracted from the main chromatogram.

The analyte is piperine which has maximum abs in around 345nm. Turing out the reference wavelength probably should help.

Looks like the Y axis scale in the chromatogram with the negative peak is too low and the negative peak looks like is just noise. The other two graphs (standards?) have a much larger scale and the noise is lost in the baseline. Try increasing the concentration of analyte by injecting a larger volume. Change the scale (zoom in) of your standards to match the scale of your graph with negative peaks, do the standards also show negative peaks at this scale? How long are you equilibrating for before injection? If theres a large negative peak that doesn't mean the data is wrong, as long as the retention time is consistent you can still use the area of the peak.

Could you please tell a briefly about reference wave length. At 380nm BW 2, Ref 380nm BW4 I am getting one analyte. If i change Ref which goes negative.

Hi Sunny,

Try to change the HPLC Mobile phase and the check once your HPLC pump whether its delivering the flow properly what ever your set for system.

What do you mean by blank? are you sure you are using placebo? (i.e. mixture of excipients), where this problem may occur because of excipient interference.

If the peak appears in the samples and not in the standard it may point on a compound, present in your samples, having a lower absorption than that of the mobile-phase at the very moment your detector calibrates. Does your detector enable the postpone of that calibration?

I have seen this before while doing GPC. Unless those negative peaks are interfering with your analysis I wouldn't worry about it. If you try going to something like phosphoric acid in place of acetic acid, an acid that does not absorb UV close to the wavelength of interest, this problem should be resolved.

Try to clean your sample in R.P (C18) then load it again on the hplc, most likely you have an impurity the interfere with the sample absorption

As Markus already said, try to turn of the reference wavelength on the detector, and if you want to use it make sure it does not overlap with the measurement wavelength, if there are any other components similar to the analyte they could cause the signal to show negetive peaks.

i) Try to turn of the reference wavelength

ii) Add some known quantity of Piperine std in sample

iii) While running the sample there may be a chance at the RT of Piperine, there may be exist a compound which does not have absorbance or have less absorbance compare to the baseline which reflects negative peak.

it might sound stupid, but I saw things like this - despite all of the instrument issues discussed before - being caused by a polluted eluent.

Hence, change the eluent, if that does not help, get your hands on a different batch of organic solvents, and don't forget the DI water you might use to prepare the eluents.

Just to add another possibility to the discussion.

Cheers,

Detlef.

The method was invalid as shown. The consequences of using instrumentation and techniques without proper training often leads to erroneous data being collected and incorrect conclusions or quantitative results being reported. In your case, there are two reasons for this:

(1) The DAD reference wavelength feature on your Agilent 1260 DAD was turned 'ON'. This must be turned 'OFF', not 'ON'. To learn why, please read the attached article below.

(2) Your signal acquisition wavelength is very close to your Reference Wavelength. This guarantees that your data will be invalid as even the peak of interest will fall under the Reference range you specified (360,10), invalidating the method and data obtained.

• REFERENCE WAVELENGTHS (as used in HPLC UV/VIS): https://hplctips.blogspot.com/2011/03/reference-wavelengths-as-used-in-hplc.html

Those who use HPLC for the first time and/or who have not had any formal training in liquid chromatography (and especially in the use of a diode array detector (aka: PDA)) often make these mistakes. Clients use systems setup just like this one and are unaware of the problems caused until a negative peak shows up (causing them to question the results). The appearance of the new, negative peak is the hint that all of your previous data may be invalid and the only reason you discovered it now is because this time the system had to subtract-out so much data from your main peak that it ended up being negative!

The negative peak may be due to your solvent or detector low performance.

I totally agree with William.

To me negative peak means no absorption, there might be an impurity in the sample which is making the analyte of interest to absorb below the cutoff wavelength of the mobile phase. Try to clean the sample before injection.

Festos, "no absorption" equals NO CHANGE in the relative signal, not no actual absorption. So-called "negative" absorption normally occurs when the incoming signal is less than that of the reference signal, resulting in a negative change relative to the reference signal used as the baseline. This is because the displayed signal is the product of an acquired new signal minus a stored reference value.

Please review the previously cited article on the misuse of the DAD/PDA "Reference Wavelength" software feature to better understand signal subtraction [REFERENCE WAVELENGTHS (as used in HPLC UV/VIS) ]. Please note that this software "reference" signal is not the same as the true Reference signal which is stored by the HPLC detector at the start of the analysis.

Following

Tips_and_Tricks_HPLC_Troubleshooting.pdf

"HPLC HINTS and TIPS FOR CHROMATOGRAPHERS"; https://hplctips.blogspot.com/

I agree with William about the negative signal

it may be not suitable method for purification

Teun Veldkamp I have answered this very question several months ago.

As far as I remember my answer was recommended.

It seems, that this mostly important service asks for some improvement

you can use other chemical test to ensure your result such s UV absorbance.

Negative peaks are most often caused by difference in refractive index between the sample solvent, sample and mobile phase. They are also caused after routine maintenance when the system has not been reconfigured correctly.

Gunel, I must disagree. As observed when performing HPLC analysis, negative peak changes (or 'dips') in a UV/VIS signal are most often NOT due to differences in Refractive Index for isocratic UV/VIS analysis methods. Any RI changes are balanced out initially and if present (i.e. from use of a different injection solvent, they would appear at the Tzero point). Negative peaks which are observed during the analysis portion are far more commonly due to absorbance differences ( which is critical to be aware of when developing HPLC methods ). Mobile phase compositional changes, poor retention (e.g. poor quality K prime values or Separation) sample matrix changes and/or analysis method wavelength selection are far more common reasons for negative peaks to appear. The user selected signal values determine the types of absorbance subtractions which may be observed (i.e. wavelength, bandwidth and any secondary software features such as the "Reference Wavelength" feature).

Routine Preventative Maintenance (PM) should NOT result in the appearance of any unexplained negative peaks. As noted, such peaks are usually due to the specific method parameters used and signal acquisition settings (which are training issues). They are not due to changing seals, lamps, rotors or pistons in an HPLC system (which are PM items). Maintenance should only be performed by individuals trained in the proper techniques and procedures which should include performance tests to insure they are done properly.

Below is a link to an article which emphases the importance of UV/VIS wavelength selection in HPLC analysis methods.

• "Proper Wavelength Selection for HPLC Method Development (or Purity Determination)"; https://hplctips.blogspot.com/2013/08/wavelength-selection-for-method.html

Negative peak appears in sample due to your sample is contaminant . it is not pure in sample preparation. Please follow sample preparation accurate. in case positive peak for standard appears