i think I get what you're after, and can think of two ways of approaching the issue. The first - and most straightforward - is to do a linear regression.
Use the time as the dependent variable and concentration as the predictor. This will allow you to predict time for any concentration. If you are not familiar with stats, you should go for this option.
A slightly more powerful method is to run an ANOVA (DV = time, fixed factor = concentration) and look at the linear contrast. This method, although not as immediately intuitive, will look for a montonic trend rather than the restricted simple linear trend (but won't allow you to describe the relationship)