In the "Mystery Peak" thread I.D. says not to use commercial grade HPLC H2O with an MeCN gradient, and to clean all the glassware thoroughly? Why? How? And what do you use for H2O? I am mainly a GC chemist, but run PNAs on an HP 1050 GC and am having some troubles with my UV detector that I believe are related to my H2O, and would appreciate any help.
Thanks

Hi Craiq
Basically there is nothing wrong with the commercial HPLC water, but with MeCN gradient I stick to Milli-Q water (Milli-Q connected to the line of already deionized water). With an isocratic separation I usually don’t care too much about the glassware as well, but traces of detergents are not so helpful in a gradient elution. I do not know what troubles you with the UV detector of HP 1050 HPLC (not GC I guess), and why do you blame H2O. A good start for HPLC of PNAs is to assure ambient, or better sub-ambient column temperature with HIGH PRECISION. This could be the reason why they equipped HP 1100 Series HPLC with the Peltier­-thermostatted column compartment.

Thanks I.D. The problem I see with my UV detectors is sinusoidal a baseline and some ghost peaks. The baseline disturbance starts as the gradient changes and then goes away when the gradient hits 100% acetonitrile. The UV detector is in series with a fluorescene detector, which is fine. I suspect the water because it is a different, but supposedly comparable, brand than usual. No other reason.

The ghost peaks are proably attributed to the water, diluent, and/or dirty glassware.

Is this a low preasure mixing system or a high? It sounds as though the pump delivering the changed composition is not doing it smothly.

As I said in my original post, I am a GC chemist who als uses an HPLC. The mixing is done by a quaternary valve before the pump, so I am assuming this means it is low pressure mixing.
We are in the process of cleaing all the glassware and runnig with new water. Hopefully that will clean up the baseline.

On low preasure mixing systems the proportioning valves have problems accuratly metering out concentrations of 1 to 5% of any one resevoir. If this is the range you are working in try to change your mixtures. The low preasure system can more accurately deliver 42:58 than 2:98.