[Bidagdha_TADIR]

This is my first post, so excuse any faults -

In absorption spectrophotometer, the function measured is absorbance (at least that is what shows up in PC monitor). That means that the spectrophotometer measures both the "intensity of the incident light" and the "intensity of the transmitted light".

But what I understand from instrumentation of a spectrophotometer (be that may be UV-Vis spectrophotometer or AAS), that only the transmitted light will fall on the detector. So the spectrophotometer is only measuring the intensity of the transmitted light; not the incident light intensity. So it can't measure the absorbance.
But when I place a sample inside the spectrophotometer, it shows the "ABSORBANCE" value 0.434... etc.

[Arkcon]

Why not?  I mean, tell me, what bit of information would the spectrophotometer need to measure transmittance?  If you can tell what info is missing, can you also tell how the instrument may be able to compensate?

[Corribus]

Correct, the instrument measures transmittance. It calculates (and typically reports) absorbance as the negative logarithm of the ratio between the recorded intensity and the initial intensity, id est, the transmittance. This makes sense from a practical standpoint because it is easy to measure how much light is left over after absorption but not how much light is actually absorbed. We indirectly measure the quantity absorbed by looking at the difference between what we started with and how much is left over.

The reason we report absorbance rather than (in most but not all cases) transmittance is because absorbance can be directly related to molecular physical information via Beer's Law (the extinction coefficient or absorption cross-section).

[Bidagdha_TADIR]

What I was wanting to know is, how does the machine measure the initial intensity i.e. intensity of the incident light? All the diagrams in the books show that the incident light directly passes through the sample holder and no part goes to the detector. So the machine can't measure the intensity of the incident light (at least that's what I understand - I would love to be proven wrong). Without knowing the intensity of the incident light we can't calculate the transmittance, can we?

[Borek]

We can record a blank to calibrate the instrument, then assume light source is stable enough to work reproducibly for the next sample.

Often we just assume light source to be stable on much longer time scale.

Note that in most cases we use the calibration curve, so even if the instrument is not giving correct results in absolute terms, it is enough that it works stable for several hours.

[Arkcon]

What I was wanting to know is, how does the machine measure the initial intensity i.e. intensity of the incident light? All the diagrams in the books show that the incident light directly passes through the sample holder and no part goes to the detector. So the machine can't measure the intensity of the incident light (at least that's what I understand - I would love to be proven wrong). Without knowing the intensity of the incident light we can't calculate the transmittance, can we?

I was hoping you'd piece this together on your own, but to re-iterate what's buried in the other posts -- the instrument can measure the light out put from the lamp bypassing the sample cell.  It may do this at the start of each run, or at power up.  I f you search around, you'll see more detailed diagrams that support this.  Also, like Corribus: said, the instrument can measure transmittance and report absorbance, since by definition, that is simply the inverse.  We don't have to justify our use of definitions in instrument design.

[Bidagdha_TADIR]

the instrument can measure the light out put from the lamp bypassing the sample cell.  It may do this at the start of each run, or at power up.  I f you search around, you'll see more detailed diagrams that support this.

Thanks. This is what I wanted to know. I am actually studying pharmaceutical analysis - I use the books
1. Introduction to spectroscopy - Donald L. Pavia
2. Spectroscopic methods in organic chemistry - William H. Dudley, Ian Flemming
3. Pharmaceutical Chemistry vol. 2 - Leslie G Chatten

If you could give better references, that will be great.