I need to measure mid-IR refractive index (imaginary (k) and real (n) parts) of a liquid.
Until now I've only found methods based on using FTIR spectrometer to obtain reflectance and absorbance and then to use Kramers-Kronig relationships to convert them into n and k.
Unfortunately, my FTIR spectrometer does not allow for reflectance measurement.
Are there other ways to find n and k in range 1um-10um?
This is actually a very old problem solved completely in the Mid 70ies, but practically forgotten since: Instead of reflectance you can also use transmittance measurements using cuvettes and convert transmittance to apparent absorbance. Since your cuvettes are usually thin, you then have to correct for the Fabry-Perot effect (= electric field standing wave effect, often indicated by interference fringes in spectra) and calculate the true absorbance and, by that, n and k like e.g. in Article The control of errors in i.r. spectrophotometry---III. Trans...
Infrared spectrum is a composite of absorption and reflection properties. This is true for any type of spectra, such as specular reflection, transmission, ATR, reflection-absorption, and many others, except for photoacoustic spectrum, which involves the thermal conductivity property. Therefore, you can extract absorption and refractive index spectra from any infrared spectrum, except some spectrum has more refractive index contribution than the other and thus more accurate to obtain refractive index spectrum. This is the case because the infrared spectrum usually have greater contribution of the absorption component. Ellipsometry is ideally suited, but is more difficult both instrumentation and analysis. Transmission is the simplest, but the refractive index component appears least. Liquid samples are little more difficult as you need a sample with even thickness. But as Dr. Mayerhöfer indicated, some sample holders that has exact parallel surface is available, usually NaCl or Kir cells, and transmission spectrum of exact thickness sample can be measured, as long as your sample does not contain water. Water dissolves these cells and cannot be used.
So, each approach has pros and cons. Again, you can obtain both absorption and refractive index information from any form of infrared spectra as long as you know the exact optical theory. The following papers give you those exact optical theory analyses of each type of spectrum.
"Optical Constant Determination of Thin Polymer Films in the Infrared," R.T. Graf, J.L. Koenig and H. Ishida, Appl. Spectrosc., 39, 405 (1985).
"Fourier Transform Infrared Ellipsometry of Thin Polymer Films," R.T. Graf, J.L. Koenig and H. Ishida, Anal. Chem., 58, 64 (1986).
"Polarization Modulation Fourier Transform Infrared Ellipsometry," R.T. Graf, F. Eng, J.L. Koenig and H. Ishida, Appl. Spectrosc., 40, 498 (1986).
"Comparison of FT-IR Transmission, Specular Reflectance and Attenuated Total Reflectance Spectra of Polymers," R.T. Graf, J.L. Koenig and H. Ishida in "FT-IR Characterization of Polymers," H. Ishida, Ed., Plenum, New York (1987) p.385.
"Interpretation of Reflection and Transmission Spectra for Thin Films: Reflection," K. Yamamoto and H. Ishida, Appl. Spectrosc., 48, 775 (1994).
"Optical Theory Applied to Infrared Spectroscopy," K. Yamamoto and H. Ishida, Vibrational Spectrosc., 8, 1 (1994).
"Interpretation of Reflection and Transmission Spectra for Thin Films: Transmission," K. Yamamoto and H. Ishida, Appl. Opt., 34, 4177 (1995).
"Kramers-Kronig Analysis Applied to Reflection-Absorption Spectroscopy," K. Yamamoto and H. Ishida, Vibrational Speactosc., 15, 27 (1997).
Here we have a midIR ellipsometer, but I never used it ! If you could make your liquid film thin on a substrate, this could be attempted.
Indeed, IR ellipsometer is the best way to go. Can you deposit your material in a quite thick layer, like 0.1-1µm, on a known material, like Si wafer? Can you share the material and deposition process?
Actually, we just found an alternative way from absorbance spectra, simply via band fitting: Article Quantitative Evaluation of Infrared Absorbance Spectra - Lor...
It works, because band fitting has the same origin as dispersion analysis...
- Badges
- Science topic
- Similar topics
- Correction