Most types of spectrometer measure either transmittance or reflectance (to be more precise, they measure the number of photons that reach the detector and not the ones which are absorbed). There are, however, a few techniques available like photoacoustic IR spectroscopy or IR photothermal-induced resonance imaging which really measure a quantity related to absorption. To what quantity is the measured quantity really related, absorptance (the absorbed irradiance, as in 1-R-T) or absorbance (as in -lg T)? What other methods that really measure absorption or a related quantity are out there?
Absorption (A) is I/I0 value, where I and I0 are the light intensities after and before the sample.
Absorbance (optical density) is log(I0/I).
@ Dr. Lokshin: Sorry to say that, but your definition of absorption is actually that of transmittance...
Is the question what you measure with photoacoustics? Well, let's check...
You got a photon getting absorbed by one particle. This process releases joule heat. This thermal energy creates a shock wave traveling to a microphone. And you measure i.a. the intensity of these spatial density variations of your gas. So you don't count photons anymore. To my mind this process is driven by an absorption of your sample as for absorbance you need to measure in- vs output.
https://en.wikipedia.org/wiki/Absorbance
https://en.wikipedia.org/wiki/Absorption_(electromagnetic_radiation)
Dear Thomas Mayerhöfer,
your statement "most types of spectrometer measure either transmittance or reflectance" pertains to spectrophotometers. They measure, as you write, "the number of photons reaching the detector". The connection to absorption is given only by the assumption, that sample absorption is by far the most dominant process. This assumption may hold if you diligently eliminate all other processes reducing the number of photons. Under these circumstances it is, as Boris Lokshin points out, essentially the ratio I/I0 . Unfortunately, not only the phenomenon is called absorption, but also the magnitude, that more correctly is named transmittance. Your question, whether a certain "measured quantity (is) really related to absorption ...or absorbance ... can be answered easily: to both, since they are, as again Boris stated correctly, only mathematically different formulations of the same fact, tied together by a simple negative decadic logarithm. That both names are used is only due to different traditions in different labs. A justification for "transmittance" (or, a little sloppy, "absorption", lies in the fact that it describes the phenomenon in a most straightforward way. On the other hand, "absorbance" has the closer relationship to the Lambert-beer law. To be complete, one should also mention the terms "attenuance" (takes into account also emission) and "optical density" (easily confused with the same term used for the index of refraction).
But these are only semantic subtleties. The interesting part is your query after techniques “which really measure a quantity related to absorption “. In addition to your examples (photoacoustic and photothermal effects), one should mention luminescence spectroscopy (including its time-resolved variants): Energy must first be absorbed to appear as luminescence thereafter.
Thank you for posing these interesting questions. Hope I could contribute a bit to the discussion.
Kind regards
Max
@Alex Risos: In this sense sound should be proportional to the squared electric field of the light. Light intensity would, however, fall off exponentially (or, better, nearly exponentially) within your sample. This nearly exponential behavior is, as far as I know the justification of the quantity absorbance. How would the exponential decrease affect the quantity that is measured? Concerning measuring In- vs. Output, you always have to do that, be it to determine transmittance or reflectance. Accordingly, all quantities are defined in relation to the irradiance bevor the sample I0 and the reflected irradiance IR and transmitted irradiance IT. Accordingly, I0 = IR + IT + IA with IA the absorbed part. If you divide then by I0 you get to the energy conservation law T + R + A = 1 (in the absence of scattering), where A is absorption.
@Max: Thank you for your contribution! I fully agree! In a simple system of a homogeneous sample with plane and parallel interfaces that are smooth everything that is not reflected or transmitted must be absorbed. What I see somewhat ambiguous, however, is the fact that T, R and A are all functions of both, index of refraction and index of absorption. If you define A = -log T, you make A also a function of both. Once you connect, however, the so defined absorbance with the Beer-Lambert law, you deny a dependence of absorbance of the index of refraction and reflectance (which Beer actually did not do... ;-). In my opinion this is a very important difference between absorbance and absorption (in the sense of a quantity). What I am in particularly also looking for by my question is a justification beyond "it is simple and we always used it" on merely physical grounds. Talking about methods that really measure absorption I should probably also mention Laser induced deflection, which was developed in our institute.
Thomas, i was merely pointing out the definition in regard what you are after. Stupidly, I never really thought about the distinction. I use for the physical phenomenon of absorption the symbol A for the IR absorbance.
Quoting Wikipedia is maybe a bit rough though:
The term absorption refers to the physical process of absorbing light, while absorbance does not always measure absorption: it measures attenuation (of transmitted radiant power). Attenuation can be caused by absorption, but also reflection, scattering, and other physical processes.
Measuring absorption
The absorbance of an object quantifies how much of the incident light is absorbed by it (instead of being reflected or refracted). This may be related to other properties of the object through the Beer–Lambert law.
Alex, I lived myself for many years in a kind of schizophrenic state never thinking of what the existence of absorption (not the microscopic process of photon absorption, but the part of the incident irradiance that is absorbed) and absorbance ( as - lg T) really means. Since there are different ways of naming one or the other out there, one has to be very cautious. Your last definition about "measuring absorption" in my eyes confuses absorption (as in A = 1 - R -T), which is the part of the irradiance that is not reflected or transmitted and the absorbance (as in A = -lg T and in Beer-Lambert).
I guess this really shows the ambiguity of the situation through using absorbance, the more as in spectrophotometry, it is used differently than defined as A = -lg (T/T0), where T0 is the transmittance of a blank...
You got me. I don't know anymore :) I guess in spectroscopy "A" is known to be -lg (T/T0), especially since you always should state this explicitly in a paper. This reminds me, that I may not have done this. In one paper I simply wrote that "A" is the absorbance based on the BL law. When you think you got a proper definition, you might add something to the Wikipedia article since this is what most people do for research and definition.
No worries here! The scientific community completely falls apart when it is about absorbance and absorption. If you look in a spectroscopy related book, in particular in infrared spectroscopy, you think there is only absorbance (but defined as -lg(I/I0) aside from reflectance and transmittance. When you read a book about spectrophotometry absorbance is still defined as -lg(I/I0), but you are being told everywhere that it is actually -lg (T/T0). Funny enough, if you look into a textbook of optics, there is no such thing as absorbance, it is all about absorption in the meaning A = 1 - R - T. At the moment I'm going back in literature and time to find out how and where the universes started to split... ;-)
P.S.: There is already a corresponding entry in Wikipedia under "Beer-Lambert-law" under the heading "Validity"...
Back to the original topic: The last days I went through a book about Photoacoustic IR Spectroscopy (see link). It seems that the spectra ordinates are always given in "Photoacoustic Intensity" and no attempt was made to compare at least band intensity ratios to those from conventional IR spectroscopy. So it seems that at least concerning PA IR spectroscopy the matter to which the measured quantity is related (more), absorption or absorbance, cannot be decided.
Article Photoacoustic IR Spectroscopy: Instrumentation, Applications...
Just a short note to let you know that I have edited the question as I think the term "absorptance" for 1-R-T might be more common than the term "absoprtion", which also refers to the process itself, and helps to avoid misunderstandings.
Well, by definition absorptance (and not absorbance...) is proportional to the electric field intensity of the incoming wave, something which puts it on par in this respect with Raman intensity... ;-)
Absorbance = log in base 10 (I0/I)
Absorption = 1- (I/I0)
--------------
Absorption to Absorbance converter formula:
Absorbance in base 10 = log in base 10 (1/(1-Absorption))
Khalil Eslami Jahromi Certainly correct, but no answer to the original question... ;-)
Both are terms for the same thing, which is the flux (energy) ratio of energy absorbed, or Iabs/Itot. Thus, it's a dimensionless quantity. The complete energy balance requires A+R+T=1 where A is the absorbance, R is the reflectance and T is the transmittance (neglecting directionality, i.e. scattering is either contained in those values or negligible).
There is also the absorption coefficient according to Beer-Lambert's law A~e^(-a*x) where x is depth in the material. The term a*x is sometimes, confusingly, also called absorbance, although it is a completely different thing. a is inherently a material property while A is setup dependent (source, angle of incidence, interference with other objects and so on).
The formula a=-ln(I/I0) assumes either reflection or transmission are zero (depends where you put your detector and what you measure). It's is not general. It is very common in chemistry relating to liquids, where you neglect the reflection and measure only the transmission, thus the full equation reduces to what you wrote.
Omer Luria Thank you for your contribution! Please note that the IUPAC has defined the quantity "absorbance" as " Logarithm of the ratio of incident to transmitted radiant power through a sample". This is the point of the discussion and why "absorptance" is actually defined as the ratio of the incident to absorbed radiant power.
Also you should be aware that your "a" in e^(-a*x) might be seen by some as a material property, while in fact it is not. Even for the simplest medium, i.e. one that is isotropic and perfectly homogenous, "a" is a function of the absorption index divided by the index of refraction, which are both wave properties (for anisotropic media it is more obvious that the indices of refraction and absorption are wave properties, since material properties are then tensors like the dielectric tensor).
In more general (but still macroscopically isotropic) media, "a" also depends on particle shape and size, even when the latter is small compared to wavelength and scattering does not play a role. This can be shown if you consider "a" as the product of number density of absorbers per unit volume multiplied by the absorption cross section of the absorber. The latter is also not a specific property - this becomes most obvious when it is stated that the absorption cross sections are enlarged through plasmonically enhanced electric fields... see e.g.
Article Deviations from Beer's law on the microscale - Nonadditivity...
Article Beer's law derived from electromagnetic theory
Indeed, IUPAC's definition is confusing and does not align with the one used in optics, or heat transfer for example. As I said, it is really a matter of discipline and conventions. When I learned radiative heat transfer, A was defined as either "absorbance" or "absorptance").
Concerning "a", it is a medium property. The wavelength dependence (whether it is by a=4*pi*k/lmbda or anything else) does not make it "not a property", and the same goes to anisotropy/composites and so on. Properties can be functions or tensors as well as constants. You are correct that if your media is made of several materials (metamaterials\plasmonics), you can define an effective "a" (as is done for n and k). Another example is Bruggeman's effective medium approximation for composite thin films. a is still a medium property in this case.
And concerning the first, I forgot to mention that IUPAC's definition is not very general, since if you have significant reflection, the ratio of incident to transmitted radiant power does not express absorption at all.
Omer Luria I absolutely concur, IUPAC's definition needs to be reworked and to include some of the new findings about what they call absorbance. On the other hand, there is no consistent naming of Ia/I0 as absorbance (what I call absorptance) in optics in contrast what you claim.
Let's play a little game - I start with stating a textbook (about optics or heat transfer), where Ia/I0 is not called absorbance and then you state one where it is.
Here is my first:
Principles of Optics : Max Born & Emil Wolf
Concerning calling "a" a medium property, is definitely better than calling it a material or a specific property, but for me this is still not proper. Ask yourself, what would "a" be exactly in a anisotropic crystal (not a tensor!) and what would it be in a effective medium? For isotropic media everythings seems to be simple, but quite the contrary.
Take a look at the spectra I uploaded. In all three cases it is the same isotropic material (= has the same elemental composition, the same unit cell (XRD cannot distinguish one from the other), the same density which is equal to that of the corresponding single crystal, the same appearance etc.).
Nevertheless reflectance in the MIR spectral region is obviously different, i.e. the reflected waves have different properties. So if you would determine the refractive index function and the absorption index function from the reflectance they would be different and since "a" is proportional to the latter, so would "a" be as a function of frequency. Would you consider it meaningful to tabulate "a"? In particular because you can calculate it from the (material property) dielectric function tensor?
Thomas Mayerhöfer, you are truly a champion of the people. Thank you for laying it out so clearly, and especially for clarifying my confusion. Whenever I learn about one thing, like absorbance, a new term pops up, like absorptance.
My pleasure and thanks for the kind words! In any way, I am still working on reducing my own confusion... ;-)
Hi Thomas Mayerhöfer !
Very interesting discussion about spectroscopy in general and the absorption coefficient in particular. I have some interesting questions regarding the absorption coefficient that I would like to discuss with you. I am preparing a paper where the concept of absorption coefficient plays a central role. I would like to take it outside of this forum. Would you be available for this discussion? I am thinking of an email correspondence.
Most of the problems and ambiguities with regard to absorbance and absorptance result from the ambivalent use of the former which is in many cases not conform with electromagnetic theory. We compiled these cases together with some explanation in a review: Article The Bouguer-Beer-Lambert Law: Shining Light on the Obscure
For everyone who is interested in more theoretical background, here is a feature-complete version of my script belonging to a lecture series called "optics for spectroscopists". It is certainly not limited to the infrared spectral range... Preprint Wave optics in Infrared Spectroscopy
Absorbance (A) is states how much of the light the sample absorbed.
while absorptance is absorbed radiation and incident radiation in a ratio format
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs