Using a isocratic method : 50% aqueous (pH 2.5, 25mM Phosphate, 25mM SDS) 50% ACN mobile phase and sample solvent, I have notice a negative peak early in the chromatogram (approx 4mins). Do anyone know what this negative peak is?

Telling us it comes at 4 minutes doesn't help unless we know the column dimensions and flow rate. We also need to know the detector type, but since you don't mention it I presume you are using a UV detector.

A negative peak can be due to a pressure upset from injection. Such an upset would be expected only very early in the chromatogram -- roughly at the void volume of the column.

Otherwise, a negative peak on a UV detector represents a band of liquid which has less absorption than your mobile phase. One common way of getting such a band is to inject a solvent with less absorption than the mobile phase. This band will then "move down the column" at the rate corresponding to that of the colored species absent from the band. Hence, it will not generally come at the void volume, but usually later.

UV-absorbing compounds in your mobile phase can be from any of the mobile phase components, and could be due to contamination, for example, of one bottle of solvent. If you are recycling mobile phase, then you can expect negative peaks corresponding in retention time to your sample components.

Hope this helps.

Yes, I am using a UV detector, column c8, 25cm*4.6mm, 1.5ml/min. The mobile phase and sample solvent are the same. Is it common to see a negative peak in SDS mobile phase (probably cause by metal ions impurities)? By the way, the negative peak does not come out at the void volume.

Another point, I am not recycling mobile phase.

Negative peaks are common in ion-pairing. Most of the time, they are retained. They are typically the ion-pairing reagent itself. However, you have not told us at what wavelength you are monitoring the chromatogram. You would see SDS only in the low UV.

With the parameters given one would suspect the void volume to be close to 4 min. A negative peak near the void volume often has a positive peak overlapping, which can shift the peak´s minimum considerably. In short, one should be careful in regard to the void volume. To find out where a negative peak comes from one can actuate the inj. valve without injecting anything, inject water, inject ACN (in your case), add a relevant salt to extra mobile phase and inject it . . . etc. Also, one should not forget that the mobile phase can change in the process of sample preparation.

I think the void volume (time of) should be around 1.4 minutes? so I think this is not due to the void volume.

Bryan is correct in regard to the void volume.

I'm using a C18 3.9 X 300 with pore size of 10 micron. My mobile phase is a mixture of phosphate buffer (pH 3) & MeOH in ratio of 50:50. I only expect one peak of interest and this peak coming out about 7 min. However, it appears as the negative absorbanced peak. Could anyone tell me what's wrong?
Thanks.

Forgive me if this is insultingly obvious, but have you checked that the electrical connections from the back of your detector to your integrator/computer are the right way around?! ;o)

Other than that, could this first negative peak be due to something else in your sample/sample matrix? Perhaps your true analyte has not yet eluted (or is eluting with the solvent front?)?

Just a few ideas, good luck.