I have a gradient HPLC method that starts with 100% Water/0.05%TFA and goes to 95% ACN/0.05%TFA, over a 20 min. period, using a Luna C18(2) HPLC column. It has been suggested that I should start with 95% Water/0.05%TFA because I may induce phase collapse starting at 100% Water. I would think that with these new column chemistries, staring a gradient with 100% Water/0.05%TFA is acceptable, but I could be wrong. Any opinions/suggetions???

You WILL see phase collapse in 100% water with a Luna C18 column. If you want to use 100% aqueous, use the Aqua column.

Given that you are using an organic solvent in your gradient (which reverses phase collapse) and the fact that phase collapse actually takes a little while, you might be able to get away with what you're doing if the Aqua column doesn't offer sufficient selectivity. What I would do to help ensure reproducibilty is to run a column conditioning (run the gradient without an injection, which is usually a good thing to do anyway), then make your first injection. 5% ACN as a starting point is usually a good way to go, but if what you're trying to separate is too hydrophilic, it'll just slide off in the void.
How's the reproducability of run times for earlier components?

To the Anon before Juddc: Just a question – how do you see phase collapse? To the same Anon - do you think that this is a good idea to suggest Aqua v.s. Luna C18, or any C18? Please think of what do you separate by the help of TFA. Juddc is on the point, but (which reverses phase collapse???)

look back in the archives on "phase collapse". there are some fine discussions by Neue. "Phase collapse" is a flawed concept.

I have not personally used the Luna column and it may indeed suffer from sudden loss of retention, caused by surface tension effects not phase collapse, but there are a good many RP columns that function just fine with 100% water. Spherisorb ODS2 is a notable example. And even ones that do suffer from sudden capacity loss may function just fine under slightly different operating conditions. Anyway, find Neue's discussion and it will all be clear. Bottom line, if you are getting adequate retention, reliably, there is no problem.

If you see reproducible RTs, then dewetting (a more correct term than 'hydrophobic collapse') is not occuring.

Even if you start off at 100% aqueous, you are fine since you are running a gradient that contains an organic modifier.

Dewetting is associated with a sudden, reversible loss of retention. If it ain't happening, don't worry about it.

I agree with juddc, when running at 100% aqueous starting conditions, I never equilibrate for extended periods of time at starting conditions. I will run continuous "Gradient Blanks" until a stable baseline is achieved and I am ready to start a run.

You can try Zorbax SB-Aq column. It is specifically designed to be able to withstand high aqueous condition including 100% water. it is true that there is something called phase collapse in HPLC