| Author | Message | ||
     Robin Gruna (robinaut) New member Username: robinaut Post Number: 2 Registered: 10-2009 |
Thank you very much for your quick and detailed answers! Best regards, Robin | ||
| Howard Mark (hlmark) Senior Member Username: hlmark Post Number: 276 Registered: 9-2001 |
Robin - let me expand a bit on what Tony said. To condense it down as much as possible, it's all about the physics. There are two main aspects to this. The first part mainly covers what Tony has said: atoms in a molecule vibrate at specific frequencies, which depend on their atomic mass, their bonds, and secondarily on the surrounding molecular environment. Most of these vibrations occur at frequencies corresponding to the mid-infrared, which are lower frequencies (and hence longer wavelengths) than we usually consider. Most of the atoms are so heavy that even their overtones don't fall into the near infrared region until you get to fairly high-order overtones. Since the absorbance strength of an overtone decreases with its order, most of these overtones are so weak that cannot be observed in the near infrared. The main exception to this is hydrogen. Hydrogen, having the lowest mass of all atoms, vibrates at the highest frequency, corresponding to wavelengths, in fact, just beyond the end of the near infrared region. Therefore, even the lowest overtones of hydrogen, the second and third overtones, and others, do fall into the near infrared region. Because of this, most of the spectra that we observe and can measure involve hydrogen vibrations and their overtones. The overtones of hydrogen vibrations also decrease with the order of the overtone. That means that the closer you get to the visible region, the weaker the overtone absorbances are. The consequence of weaker absorbance is that the material becomes more transparent. Since the differences in absorbance over the IR spectral region can be orders of magnitude, a material that is nearly transparent in the portion of the NIR spectrum near the visible can be very opaque at the long-wavelength end of the NIR region. Now we come to the second important aspect of physics that affects our measurements: the question of optical scatter. To put it in more familiar terms, a clear liquid, such as water, does not scatter light, and therefore it is transparent and we can see clearly through it. Even a colored liquid, such as apple juice, you can see through since it does not scatter the light. In contrast, a scattering liquid, such as milk scatters the light severely, and therefore we cannot see through it. When you look at a glass of milk, what you see, you see by reflection, mainly from the surface of the milk although some light does penetrate below the surface. So it is in the NIR region. Both absorbance and scattering effects determine how the light at different wavelengths interacts with the sample, and therefore this determines the nature of the measurements we can make. There is also interaction between the absorbing properties and scattering properties of a given sample; this very much complicates attempts at trying to analyze the behavior of the system mathematically. We can very roughly associate the measurement techniques to sample characteristics this way, keeping in mind that there are many exceptions: 1) clear liquid samples, no absorbance, no scattering: measure by transmission 2) scattering liquids: measure by transflectance 3) powdered solids, low absorbance (also some scattering liquids): measure by diffuse transmission 4) powdered solids, high absorbance, high scattering: measure by reflection As you can see, the decision of which measurement technique to use is not cut-and-dry. Many types of samples can be measured in more than one way. For example, "high absorbance" and "low absorbance" are very vague terms, at best. You may have to do some experimentation with your own samples, to find out what works best. In addition to Tony's suggestion about studying the spectra of your samples, I would also recommend that you get at least one good book on NIR spectroscopy and study that, in addition. One book that is very popular is the "Handbook of Near Infrared Analysis", CRC Press. \o/ /_\ | ||
| Tony Davies (td) Moderator Username: td Post Number: 202 Registered: 1-2001 |
Hello Robin, Welcome to the Group! Yes, it can be VERY confusing! Overtones are (approximate) multiple of fundamentals so you can calculate where they will come if you have an IR spectrum of your sample. Combination bands arise from combinations of two or more fundamentals; only one of them needs to be infrared active so they are much more difficult to predict. You might like to read through the recent discussion further down the "I need help" list: "fundamental ,overtone , combination bands". Regions The NIR region extends from 780nm to 2,500nm. The sub-divisions are mainly about detectors. Silcon are optimum for 780-1100 (I have proposed the name "Herschel Region" because this is what William Herschel discovered in 1800). The other main detector is the lead sulphide detector which is mainly used from 1100- 2,500 nm (I have called this the "Norris Region" after Karl Norris who is the "father" of modern NIR spectrocopic analysis. He is still alive and writing for this Forum - Hello Karl!). There are other detectors such as the InGAs (Indium gallium arsenide) which will have different ranges. It would be very useful if there were detectors that would only respond to specific vibrations but this is not the case (not yet anyway). All these detector will detect all absorption that occur in their wavelength range. Reflection and transmitance measurements can be made at all wavelengths except that reflection is more common in the 1100-2,500 region. The 1800-2500 range for hyperspectral imaging? Apart from a few exceptions you will be measuring only combination bands. There are a lot of them and they may be considerably over-lapped. I hope this will be of some help. Going straight into hyperspectral imaging is a BIG step. I suggest it would be a good idea to study the NIR spectra of your analytes before you start serious hyperspectral imaging as I think you want to understand what you are doing. Don't worry if you find it confusing. There are stil many unknowns in NIR spectroscopy but that hasn't stopped it being a very useful method of analysis! Best wishes, Tony | ||
| Robin Gruna (robinaut) New member Username: robinaut Post Number: 1 Registered: 10-2009 |
Hello all, I'm a totally newbie to the field of NIR spectroscopy and NIR hyperspectral imaging. I'm a little bit confused about the definition of different NIR spectral regions. The defenition of different combinational and overtone regions is clear to me but I do not understand why there are also different meassurment modes like reflectance (1100 to 1800 nm) or transflectance (800 to 1100 nm) assigned to them. Does this mean that one can meassure only certain overtone vibrations in certain meassurement modes? What does this especally mean to hyperspectral imaging in the 1800 to 2500 nm region? Regards, Robin |