This is a very good question that is closely related to one I asked two years ago and to which I still got no answer, despite a lot of colleagues were reading it:

https://www.researchgate.net/post/Absorptance_or_absorbance

My opinion is that the spectra must be related to absorptance (1-R-T), because they "reflect" the amount of energy that is absorbed inside the sample, to which absorbance (as in -lg T) is only proportional for very small values.

Thank you! Thomas Mayerhöfer

I don't have an answer, but it might be worth checking whether/how it is done in techniques such as AFM - IR and SNOM

https://www.anasysinstruments.com/afm-ir-overview/

https://www.neaspec.com/applications/

I work with cell wall material from plants. I want to convert photoacoustic units in absorbance or vice versa. I have several spectra data with absorbance value and I need to convert this value in photoacoustic units to compare with other spectra data. My FTIR is a JASCO FTIR spectrometer and the analysis software is called SpectraManager, this software provides me a %T, %R and RK units too.

Thank you! Sandra Teresita Martín del Campo

Do they have a reference standard?

No

Can you measure a known material with FTIR and then with PA-FTIR?

I think it initially depends on the x-axis scale you are working and the type of measurement you are performing. For example, if the x-scale is in time-frequency units, you could transform this into wavelength by a Fourier transform. Then, if you are measuring the material in transmittance mode, you could just apply t = P/Po (P = signal potential for the sample, Po = signal potential for the blank), this would be the transmittance-like spectrum. Absorbance is a result of a mathematical transformation, thus it has no units. To calculate an absorbance-like signal, just apply a = -log10(t).

If your spectrum is in wavelength, then a Fourier transform is not necessary.

If you are measuring in reflectance mode, you can transform the reflectance signal to absorbance using the Kubelka-Munk equation.

I'm not sure if this will work, but you can have an 'absorbance-like' signal. You could then compare this with the actual spectrum generated by the FTIR spectrometer.

This might be a solution, but I agree that this is to a certain degree speculative. We would also need in my opinion more information about the nature of the sample. E.g., if the intensity in PA-FTIR is really proportional to the absorptance, i.e. 1-R-T, then you can obtain either R or T only if you can neglect the other quantity. The sample dependence comes into play, because if you have e.g. a thin layer, you do usually not obtain T by ratioing the signal of the substrate with layer to that of the substrate without layer due to interference enhancement of the signal in one spectral region and signal reduction in others (not even the use of a CaF2 substrate for an organic or biologic layer prevents you from that except for relatively thick layers). Furthermore, reflectance can be either diffuse or specular. Kubelka-Munk only "works" for diffuse reflectance (I have set the word "works" into quotation marks, because what you get from applying Kubelka-Munk is in my opinion not really the absorbance; if you compare the resulting absorbance with that from transmission you know what I mean).

Hugh J Byrne

No :( In my lab we don´t have the PA-FTIR instrument

Ok - now I don't understand why/what you want to convert?

I would suggest that you start to read the literature on photoacoustic FTIR. You can find a VERY good bibliography at http://www.mtecpas.com/pasliterature.html

You should also read chapter 20 in

Fourier Transform Infrared Spectrometry, 2nd EditionPeter R. Griffiths, James A. De Haseth, James D. Winefordner (Series Editor)ISBN: 978-0-471-19404-0 May 2007 560 Pages

It has a very good description of the issues with PAS collecting the spectra in rapid scan and step scan mode.

Hugh J Byrne

Sorry, It's possible that I didn't explain good this issue. I have PA-FTIR spectra data (X axis: wavenumber, cm-1; Y axis: photoacoustic units) and absorbance FTIR spectra data (with/without ATR correction, X axis: wavenumber, cm-1; Y axis: absorbance). I obtained the absorbance FTIR spectra data in the FTIR (JASCO) equipment of my lab. The PA-FTIR spectra data were performed by a collegue of other lab and he sent me this data. I want to transform the absorbance value (of absorbance FTIR spectra data) in photoacoustic units or vice versa to can compare peak heights. I need to compare peak heights to determine differences between my samples and the samples of my collegue.

Thank you for your help and time

This adds another layer of complexity. Absorbance values from ATR do not generally equal those obtained from transmittance, see e.g. Article Expended Formulas for Attenuated Total Reflection and the De...

Thank you for your answer Ellen Miseo

I am reading the bibliography that you has given me.

Before you do any comparisons, you need to go read AT LEAST the chapter in Griffiths and deHaseth. Photoacoustic done in the rapid scan mode will show a wavelength dependence of the penetration and therefore a skewed spectrum due to both saturation as well as varying band intensities. Photoacoustic spectra collected in step-scan using phase modulation and demodulated properly will show a more "transmission like" spectrum. BUT the experiment MUST be done properly. It would be much simpler to collect the spectra in ATR from both labs.

You show me a possible way to "photoacoustic -> absorbance", is it possible the reverse way to "absorbance -> photoacoustic"? The samples are a powder resulting from a cell wall extraction of vegetable material homogenized in mortar with liquid nitrogen. In the FTIR spectrum, the samples are in a thin layer on the window. Thank you for your help Thomas Mayerhöfer and Camilo Morais , I thought that this problem would have an easy solution, but you show me that no.

Ellen Miseo Yes, I agree with you, it would be much simpler to collect the spectra in ATR from both labs. Unfortunately I haven´t the PA-FTIR samples.