every five conditions when ACN Conc was increased, there was decreased pump pressure (100 bar to 50 bar). i doubt that the pressure is the reason of baseline drift... but TEST 20 also had decreased pressure pattern.... pressure is not the cause

If you use mobile-phase-gradient that's obvious in these wavelengths.

If not- think of increasing column equilib. time.

Always good to use a guard column, but as Zvi indicated, assuming you are using  DAD/UV detection it is not unusual for baseline drift using the kind of mobile phase gradients you've indicated. See http://www.chromacademy.com/chromatography-HPLC-Gradient-Elution-Baseline-Drift.html

Thanks to kindly answers!!

But i'm wondering WHY baseline drift only appeared when i added ammonium acetate or acetic acid or both ?? (without two things there was no drift when i did same gradient)

AND this protocol came from some articles... but there was no baseline drift in their paper's figure. only differential peak size...

In that paper, their protocol is "a linear gradient system from solvent A (20% acetonitrile with 10 mM ammonium acetate) to solvent B (80% acetonitrile with 10 mM ammonium acetate) over 60 min at a flow rate of 1.0 ml/min, with detection at 220 nm"

Am i doing something wrong??

http://www.jmb.or.kr/submission/Journal/026/JMB026-01-09_FDOC_1.pdf

Article Interspecies Complementation of the LuxR Family Pathway-Spec...

First of all, you do not have any "baseline drift". Baseline drift is caused by problems and what you are observing is a result of the additives absorbing light and changing the detector output (RI) with mobile phase composition changes (from your method). 100% NORMAL, not an artifact, not abnormal variation or problem and not "drift". The change is a 100% predictable change in absorbance due to a calculated change in mobile phase composition. The system is doing exactly what you would expect given the condition of your HPLC method.

Your gradient is basically a gradient of water to organic (Water -> ACN). You are decreasing the amount of water (and everything in the water bottle) over time. That is why the baseline level decreases after you start your gradient (which BTW has a HUGE delay volume!). This is clearly seen with all of your chromatograms at 220nm. Most of the decrease in signal over the gradient may be due to the acetic acid alone as this acid is often very dirty (is it clear or brown) and failure to use fresh clear solution can result in strong absorbance and poor chromatography. - As you decrease the amt of acetic acid during the gradient change, the amt of acetic acid seen also decreases so the signal drops too. That is probably what you are observing.

Now the way we deal with this type of physical effect in chromatography is to try and add equal parts of the additives to both mobile phase bottles to keep the total concentration "seen" by the detector and method similar (You did not do this in your "test"). This is not always possible to do when you have a salt buffer and a pure organic, but you can change your bottles (A and B) to be 80/20 and 20/80 and both contain some water to start with (modify your gradient to allow for this change) and then easily add the two additives to each bottle (w/o precip). This will help minimize the baseline slope changes which result during the normal gradient composition change. You should still see some change in slope (esp at 220 nm) because you are running an organic (ACN) against water, which have different properties and this is normal.

One other way we deal with this normal physical change is to also run a blank sample (mobile phase) after the real sample. Next, use the software's built-in "blank subtraction" feature to subtract the "blank" from the actual sample and that will have the effect of cancelling out the slope change over time and provide you with a much flatter baseline which will allow much better integration of the peak(s).

Joon wrote: "every five conditions when ACN Conc was increased, there was decreased pump pressure (100 bar to 50 bar)". - To answer another one of your questions, this too is normal. ACN is about 3x less viscous than pure water so the decrease in viscosity is the reason for the decrease in pressure over time (increase in ACN = lower back pressure). So this too is normal and a fundamental parameter of liquid chromatography.

Thanks to Bill Letter!!! kindly answers

I used fresh acetic acid(SIGMA-Acetic acid eluent additive for LC-MS), and it's clear (i bought that few days ago..).

i test same additive (same conc(v/v) 0.05% acetic acid ) to both solutions (Water / ACN) but when ACN conc increased(over~45%) there was increased baseline.(TEST 32,33)

I can take result from sample peak area(from TEST32 - RED line), but i'm wondering why increased baseline was appeared by increased ACN conc ? (i made same conc acetic acid : 0.05 %)

my thinking : increased baseline by acetic acid??

 water and ACN has same conc acetic acid. The ratio of water and ACN is changed, nevertheless total conc of acetic acid of flow must be constant? 

But the result wasn't agreed with my thnking...

Why are the results different (Peaks are different) for "Test 32 & 33? I thought you said they were both the same sample under the same conditions?

You should have seen a change in the baseline slope when you ran the same gradient with just ACN and Water too. *Are you displaying the overlaid chromatograms in normalized mode? ANY change in composition over time should result in a change in detector output at 220 nm (baseline change). Three ways which could mask this effect are: using a Reference Wavelength feature (which if available should always be turned 'OFF'); displaying the traces as normalized and blank subtraction.

The slope changes observed would of course be dependent on the 'Y' scale (Abs) viewed at 220 nm. As you increase the 'Y' scale, the baseline slope will appear to flatten out and as you decrease the scale it will be more pronounced. These are basic optical principles. The addition of light absorbing additives can make the changes more pronounced. At such low wavelengths, you should always observe some changes in slope which directly relate with the compositional changes made to the mobile phase.

**Could you tell us: What are the dimensions of your C18 column (Length x ID x particle size)? What is the flow rate used? What brand and exact model of HPLC system are you using?

I'm using Hamilton column : PRP-1 10 µm 100 Å 4.1 x 250 mm

Flow rate : 1 ml / min (all)

HPLC system : agilent 1200 series(Degasser, pump),but detector is old 'HP 1100')

The drift in baseline at 220 nm using gradient methods with ammonium acetate and acetic acid conditions is normal. 

You better further decrease the ammonium acetate content to 0.025M and acetic acid to 0.025% and try with

1) MP-A: 0.025% acetic acid in 0.025M ammonium acetate: ACN (90:10) and                        MP-B: 0.025% acetic acid in 0.025M ammonium acetate: ACN (10:90).

If possible start with at least 30% MP-B.

2) MP-A: 0.025M ammonium acetate in WATER: ACN (90:10) and                     

     MP-B: 0.025% acetic acid in WATER: ACN (10:90).

Could you please tell us the gradient also.

You stated in your first post that you are using a "C18" column. The Hamilton PRP-1 column is not a C18 column, it is a hybrid column made from polystyrene-divinylbenzene, not silica and has different functionality. This difference is important when developing methods and also in washing the column after each run (requires different optimization and techniques).

We have provided many suggestions to you in how to reduce the baseline slope change during the gradient portion of your run. Again, this type of change is 100% normal. Baseline subtraction and/or reducing the amount of additives (only if applicable) can reduce the effect somewhat, but at 220 nm you should always see it during a gradient (again, esp at such a low UV wavelength).

I strongly suggest you take some time to read one of the many wonderful books on liquid chromatography before actually running HPLC methods. Running methods without a basic understanding of the chromatography fundamentals is bad science. You need to understand the mechanisms, process and hardware first to understand what you are observing. With time, much study, a mentor and lots of patience you will get there.

BTW: The 1100/1200-series HPLC is an outstanding system to learn on. An "HP/Agilent 1100" is the same as an "Agilent 1200". The manufacturer just changed the front plates to show a different # (they call it marketing!), so no worries, you are not using an older detector.

Oh......... I don' t know my column is not 'C18' ....I just get this column from my lab... this is my big mistake... i'll check my column first!! 

' to read one of the many wonderful books on liquid chromatography ' 

would you please, recommend any good books for studying about chromatography!!

I realized many things... Thanks for answering my question! especially Bill Letter !

Here are two good recommendations for novice to intermediate level users.

"Introduction to Modern Liquid Chromatography"; by Snyder, Lloyd R./ Kirkland, Joseph J./ Dolan, John W.   ISBN 0470167548

"Practical HPLC Methodology and Applications"; by Brian A. Bidlingmeyer. ISBN 9780471572466

• Additionally, here is a link to a website with lots of free HPLC "Hints & Tips" to help you learn about and become a better chromatographer.  http://hplctips.blogspot.com/

http://hplctips.blogspot.com/

Thanks for kindly recommendation... I have another questions!!

i had been used NOT C18 column PRP-1. Hamilton homepage subscribes it packed by only PSDVB.

AND i find other column PRP-C18, it is subscribed "packing material : PS-DVB functionalized with C18"

then what is "functionalized with C18" meaning? C18 is attached with PS-DVB like tailing ? Is this right?

when i refer to some articles, they used 5 µm Discovery HS C18

reversed-phased column (25 × 4.5 cm; Supelco Analytical). that column is packed by silica-based C18. but Hamilton column is packed 'PS-DVB functionalized with C18'.

Are these two kinds of matrix(Silica VS PS-DVB, when both attached with C18) have significantly different functionality?

I only have to join Bill's offer, since it appears that you really know nothing about chromatography: "PS-DVB" are the initials of  polystyrene-divinylbenzene, which is a an extremely stable polymer, which may be derivatized with various functional moieties, to get specific properties usable in separation, like ion-exchange, hydrophilic -hydrophobic  affinity etc.

In short- you must follow Bill's recommendationeven even in order to ask intellugent questins.