[jay]

I've got a calibration curve but some of my samples have absorbances higher than the curve goes - how can I calculate the concentration of the sample?  I've done 1:2 dilutions for them, but still can't work out what to do.  The calibration curgve is not linear so I can't just extrapolate. *delete me*!  Text books never tell you this sort of stuff!

[lemonoman]

I think that the reason textbooks don't tell you is that it would eliminate the point of doing your own calibration.

Depending on what you're using as equipment, you could have different 'errors' associated with the measurements.  One of the most common, is that the absorabances given are some fixed amount more or less than they actually are.  Sometimes though, the error can be much more involved than that.

Standards are prepared exclusively so you can place your sample among them (as I'm sure you already know).  Since your sample isn't among your points, you have no way of knowing for SURE if your number will be correct (it will be a guess)

One way you can 'guess' is to look up exactly what the absorbance curve looks like for your compound, then do some fidgeting around with your own curve to make it look like this 'reference' curve...and then you can place your point.  

If you can't get your hands on a curve, then you might have to look for patterns in your curve, and predict what it looks like.  Again, this is just a guess...and is STRONGLY ADVISED AGAINST in the scientific world!

But good luck  

[sdekivit]

maybe you can do some axis transformations on your curve, depending on the relation of x and y to get a straight curve you can extrapolate.

[Winga]

If you plot A vs. conc., the curve should be linear (assume there is no deviation).

A = ebc

Plot the best straight line!

[Dude]

Dilute it more.  1-10 and 1-100 dilutions are common practice.  Ensure that your dilutions are accurate to minimize error and that they are not contaminated.  Some calibration curves will never be linear depending on what type of analysis you are doing.  Your thought of staying within your calibrated range is correct.  Do not try to extrapolate beyond the highest calibrant.  Do dilutions and then multiply by the dilution factor.