Physics Stack Exchange is a question and answer site for active researchers, academics and students of physics. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I am writing an application that should allow to view and manipulate spectra (Raman).

I am only given one normal spectrum consisting of 1024 coordinate points. Plus I have a background spectrum that is supposed to be subtracted from the first spectrum.


The first point in spectrum_1 is: -55.988 , 9373 The according point in background spectrum: -55.988 , 9382. After subtracting the two in a third-party software, the first point of the resulting spectrum is: -55.988 , -641.968

So I asked myself how this is calculated. According to the Documentation of this software, it is using the following formula:

result spectrum(n) = multiple component spectrum(n)

– Factor1 * single component spectrum (n)

– Factor2 * single component spectrum (n)

Unfortunately I can't do anything with it. What is meant by "component spectrum(n)"? Is it the Spectrum equation? Also what could the both factors be and what does the (n) mean?

Here is an image of all three spectra that are involved in the calculation.

Here is an image of all three spectra that are involved in the calculation.

From Top to bottom:

  1. First spectrum
  2. Background spectrum
  3. The resulting spectrum after the subtraction

I simply can't imagine or reproduce the calculation that leads to this result.

Any help and/or advice is welcome. I am highly willing to learn and understand so please don't hesitate to share some links or other references that could help.

Thanks in advance, BC++

share|cite|improve this question
I think this question needs a little more details to be answerable. We're just as stumped as you. One thing that may help is to post screenshots of the software you use, including plots of the signal and background spectrum. This may shed hints on the meaning of "multiple component" and "single component". – Greg Graviton Mar 1 '12 at 11:34
The frequency range spans from 400 to 1800 cm$^{-1}$. Where does your value of -55.988 come from? Also, how does the spectrum look like if you subtract it manually? (for the point you gave: 9373-9382=-9). You could do this subtraction in Excel if you don't have anything else. – gigacyan Mar 2 '12 at 12:00
@gigacyan The frequency range was cut to those values as everything else is considered junk. – Käsebrot Mar 2 '12 at 12:34
up vote 3 down vote accepted

Background subtraction of Raman spectra is as simple as regular subtraction (assuming that both spectra are measured in the same way, same wavelength step size etc.)

Your third-party software does something else. The formula from the Documentation looks like subtraction of different spectral contributions. Suppose that you have, say, water and ethanol - two components. Observed multi component Raman spectrum will be a linear combination of water and ethanol Raman spectra. If you subtract spectra of these compounds (measured separately) from your total spectrum, you should get a straight line or a Raman spectrum of some other substances that you might have in the mixture.

Factor1,2 are just proportionality coefficients that you would need to adjust to match a sum of individual spectra to the total spectrum. I suppose that (n) here means that you perform this operation for every data point in a loop.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.