Higher peak height, for the same sample

[asiswanto]

Hello all,

I injected 100% methanol to my HPLC for calibration purpose. It usually has peak height around 250 mAU, but yesterday I got result of peak height around 500 mAU (all HPLC set up is exactly same). What it does mean?I have same sample, but give different peak height. The retention time is approximately similar to current result and 2 weeks result. Is there any effect of temperature?Yesterday, the lab was colder than 2 weeks ago.
Thank you.

[HPLCaddict]

What exactly are you trying to calibrate? Using 100% Methanol as "sample" is, well let's say a bit extraordinary. If we're talking about RP-HPLC, I doubt that that there's really a retention (!) time measurable.
If you're trying to do something like OQ'ing the autosampler, you should use a decent sample.

Nevertheless, some thoughts:
- What about the peak area? Is it also doubled? If the peak shape changed, the height might do as well without the actual area changing.
- Are the conditions EXACTLY the same (injection volume, mobile phase composition, detection wavelength, flow-rate, column)? All of these might have an influence on peak height/area...

[tom jupille]

Additional thoughts:
- what wavelength are you looking at (mAU for units implies UV detection). Methanol does not really have a chromophore in the UV, but it does show "end absorbance", with a nominal UV cutoff of 205-207 nm. The problem is that you would be working on a steep portion of the spectrum so that a minor wavelength variation can make a big difference in absorbance.
- there is quite a bit of lot-to-lot variation in MeOH.

[asiswanto]

What exactly are you trying to calibrate? Using 100% Methanol as "sample" is, well let's say a bit extraordinary. If we're talking about RP-HPLC, I doubt that that there's really a retention (!) time measurable.
If you're trying to do something like OQ'ing the autosampler, you should use a decent sample.

Nevertheless, some thoughts:
- What about the peak area? Is it also doubled? If the peak shape changed, the height might do as well without the actual area changing.
- Are the conditions EXACTLY the same (injection volume, mobile phase composition, detection wavelength, flow-rate, column)? All of these might have an influence on peak height/area...

I used the RP-HPLC for ester analysis. The methanol injection (peak appeared around 2.1 minutes) actually aims to check the stability of HPLC. I do this check in two steps, first check with methanol injection then compare the peak height with previous result (which is done around 1-2 weeks before) and check with injection of one of data point in calibration stage.
1. The peak area changed, (previous around 1500 mAU*sec, current around 2400 mAU*sec).
2. All conditions are EXACTLY the same.
Do you have any idea, why the peak height change?

[asiswanto]

Additional thoughts:
- what wavelength are you looking at (mAU for units implies UV detection). Methanol does not really have a chromophore in the UV, but it does show "end absorbance", with a nominal UV cutoff of 205-207 nm. The problem is that you would be working on a steep portion of the spectrum so that a minor wavelength variation can make a big difference in absorbance.
- there is quite a bit of lot-to-lot variation in MeOH.

1. I used wavelength 232 nm which is gave result of my previous work a week ago that the methanol peak appear around 2 minutes.
2. I use exactly the same methanol.
Then, how can the peak height different?is there any correlation with temperature?all operating conditions are the same.

[HPLCaddict]

At that wavelength the "peak" that you're seeing is mainly due to a refractive index change methanol vs. mobile phase and not due to the UV absorbance of methanol. That's the peak usually referred to as "solvent front", "t0 peak", "t0 noise" or whatever. This "peak" is highly dependend on the *exact* circumstances and there's no chance that it will be absolutely reproducible over weeks. Guess why it's often called "t0 NOISE" .
Get a decent sample to do the calibration! When using a UV detector "decent sample" means something that shows decent UV absorption .

[asiswanto]

At that wavelength the "peak" that you're seeing is mainly due to a refractive index change methanol vs. mobile phase and not due to the UV absorbance of methanol. That's the peak usually referred to as "solvent front", "t0 peak", "t0 noise" or whatever. This "peak" is highly dependend on the *exact* circumstances and there's no chance that it will be absolutely reproducible over weeks. Guess why it's often called "t0 NOISE" .
Get a decent sample to do the calibration! When using a UV detector "decent sample" means something that shows decent UV absorption .

Great!!!!
Many thanks.
I am a new user of HPLC, so need more your "lights"

[DJ]

For QC? Why the insistence on such an unorthodox approach?

When I want a quick assessment of plate count, I use acetophenone + propiophenone as standards an a 60/40 MeCN/water (isocratic).

Make the sample volume as small as possible (10 or less micro liters).