[Omega Glory]

Got a bunch of questions here, purely out of curiosity and expanding my knowledge.

[I'm not an undergraduate in chemistry anymore -- but I sometimes come up with undergraduate-level questions which I realize I'm not sure of the answers to, either because I've forgotten, or because I never learned the question all that well in the first place. ]

Infrared. Ionic compounds. When I was in school, the professors always taught me, "Ionic compounds don't absorb ANY infrared light, they are INVISIBLE to infrared. That's why we use KBr pellets and NaCl plates." Now, someone let me know me if I'm off-track here, but I think that's quite wrong, isn't it? The correct thing to say would be, "Ionic compounds like KBr and NaCl DO absorb infrared light, they just absorb OUTSIDE the mid-IR (~4000-400 cm-1), which is the region of interest to FTIR analysis... They absorb at very low energies and low frequencies, below 400 cm-1.

Those absorbances -- those are associated with phonon modes, right? The propagation of a transient compression wave through the crystalline lattice?

Off-the-wall question I just thought of. Do organic compounds have similar phonon modes? If I take a crystallite of sucrose and hit it with low-energy IR, will I get a similar low-energy absorbance as the energy causes a compression wave to ripple through the crystal lattice of the sugar? Pure curiosity, phonons are something I don't have any experience dealing with.

Are such phonon modes not accounted for by the old 3N-6 (or 3N-5) formalism? I'm assuming not, since they're lattice vibrations, not molecular vibrations.

Are there any ionic compounds for which you'll see discrete absorbance bands in the Mid-IR range (4000-400 cm-1?). Talking about real absorption bands, not just a slightly narrower spectral window. Will we ever see any absorption bands for metals or ionic salts in this range?

[pgk]

Many mineral salts and acids absorb further than 3,500 cm-1, which is the limit for the most organic compounds.
Organic salts absorb in the organic region (say 3,500-600 cm-1) but the appearance of their IR spectrum is different than their mother neutral compounds. Thus:
Amine salts have an additional weak band at about 2,000 cm-1,
Carboxylate, phenolic, organosulfonic and organophosphate salts have quite different IR spectrums than the corresponding acids, etc., due to the different "reduced mass" μ.
Please, take a look to a organic spectroscopy or IR sprectroscopy book, for further reading.

[Omega Glory]

I shan't be opening any organic spectroscopy textbooks, I don't have the time. And I don't learn well from books anyway... Every time I read one, I think up all sorts of other fundamental questions which the books just never seem to answer, but a person probably could, if they would take the time.

I wouldn't turn away a friendly conversation, if you'd be willing to indulge me.

Amines, carboxylates, phenolic, and organosulfonic compounds all have covalent bonds. Therefore, it's not at all surprising that these would absorb in the mid-IR, as the covalent bonds in organic compounds are also wont to absorb there.

I'm much more curious to know about the ionic bond -- between the carboxylate and the cation, for example. Does *this* bond have a particular absorbance in the infrared? And if so, does it tend to fall high or low, and what are the physical effects associated with it -- vibrational modes, phonon modes, etc.?

I am a dummy, so I would be grateful to anyone willing to dumb it down for me.

[pgk]

1). IR absorbance is due to vibrations of atoms that participate in a functional group. But the bond between the anion and the cation is electrostatic, meaning that anions and cations are separated, they do not share the same orbital and besides, the cation may be quite far away from the corresponding anion. Therefore, ionic bond does not vibrate after IR absorption.
Anyway, exceptions occur in partially ionized salts, where the non-ionized part absorbs IR. This is why a weak peak appears at 2000 cm-1, in the IR spectrum of amine salts.
2). This is mainly an educational forum and therefore, the general rules of education are followed hereby:
Since the antiquity era, education is based on further reading, after the teacher thus said. So, if you do not open books, we cannot help. Sorry! 

[wildfyr]

It seems presumptuous to assume books by highly experienced spectroscopists will be unable to answer your questions. By using books to learn the fundamentals, one is able to answer questions not directly addressed in the volume. I believe pgk applied such knowledge in answering your question in fact.

I venture that every chemist here has used reading to acquire the bulk of their knowledge, through articles and books. I think I read 'Spectrometric Identification of Organic Compounds, 7th Edition" cover to cover. It also feels a little ridiculous to say you don't have the time to open one. You seem to be pursuing abstract knowledge for its own sake, there is no time constraint.

[Irlanur]

you can have a look here

http://matematicas.udea.edu.co/~carlopez/ac60068a007.pdf

of course, many of the transitions are very similar to ''organic'' transitions.

[pgk]

1). I feel glad if having helped you to understand something that is not fully clarified in ordinary IR spectroscopy textbooks.
2). Nevertheless, my replies in this post tried to explain, as simply as possible, why some ionic compounds, whether organic or mineral, are visible in an ordinary IR spectrum and some others are not. But I am afraid that my explanations will be lost soon, if not going back in a textbook and take an idea about, e.g. what does “reduced mass” μ mean in IR spectroscopy.
3). Many thinks and values (e.g. human relations) haven’t change so much, since the antiquity era, education included. As far as known (though I am not a specialist in the issue), modern education methods are usually applied to little pupils and do not still have proven their long term value and strength, through time. But if you have an example of a more advanced and effective education method that doesn’t demand a further reading, please indicate.
4). As far as understood, wildfyr said what he said and nothing more and nothing less and additionally, it is a good advice. Therefore, the corresponding reply ought to be more chic and elegant, at least!

[Irlanur]

But the bond between the anion and the cation is electrostatic, meaning that anions and cations are separated, they do not share the same orbital and besides, the cation may be quite far away from the corresponding anion. Therefore, ionic bond does not vibrate after IR absorption.

The "therefore" in this sentence is not true. there is no theoretical necessity for it. In the common Born-Oppenheimer description, vibrational modes come from the nuclei that move in some potential. they don't care where this potential is coming from.

[pgk]

1). In theory, may be no necessity for it but in practice, there is. IR radiation has low energy content and thus, it is unlike to be able to generate enough potential that is necessary for the resonant vibration of ionized nuclei, which additionally contain the corresponding amount of their ionization energy. As a comparison, O-H covalent bond energy = 467 kJ/mole, in contrast to the 1st ionization energy of oxygen that is 1314 kJ/mole and hydrogen that is 1312 kJ/mole.
2). If desired, the energy content at a given IR frequency can be calculated by the Plank equation:
E = hv,
where E is the energy, v is the frequency and h is the Plank’s constant. But attention cm-1 is a wavenumber and not a frequency unit.
3). Thank you very much for the link above that is highly useful. But what you observe therein, are the vibrations of the covalent bonds. For example, the IR spectrum of Na2CO3 shows the vibrations of covalent C-O and C=O bonds but not the ionic Na-O bond.

[Irlanur]

3). Thank you very much for the link above that is highly useful. But what you observe therein, are the vibrations of the covalent bonds. For example, the IR spectrum of Na2CO3 shows the vibrations of covalent C-O and C=O bonds but not the ionic Na-O bond.

that's what I meant by

of course, many of the transitions are very similar to ''organic'' transitions.

I didn't check the whole list though.

As a comparison, O-H covalent bond energy = 467 kJ/mole, in contrast to the 1st ionization energy of oxygen that is 1314 kJ/mole and hydrogen that is 1312 kJ/mole.

careful here. For the energy difference between vibrational modes, the important contribution is the force constant (in a harmonic approximation). This describes the steepness of the potential, which is not directly related to the dissociation energy (deepness of the potential).

[pgk]

Of course not! Bond dissociation energy has a little to do with resonant vibration energy, but it is an effective and descriptive measure of comparison, in regards of the issue.
PS: Also, the covalent bonds of crystalline water are clearly visible in the most of IR spectra of this useful list.

[Corribus]

Ionic compounds are indeed often optically active in the IR region - typically in the far IR or THz region. But chemists' (and particularly organic chemists') perspective of vibrations (well everything) is very molecule-centered.  Molecules tend to feature single vibrating bonds or, perhaps, delocalized molecular vibrations that extend over a region (functional group vibration) or in some cases the entirety of large molecules (like benzene, but the aromatic rigidity and atom-to-atom communication makes these exceptional cases). It mostly stops there, though, because in most cases intermolecular interactions are to a first approximation too weak to transmit vibrational energy between molecules even in condensed phases - coupling is just really low. Even in condensed phases, these interactions mostly manifest themselves as small shifts in the energy of in-molecule vibrations. (Hydrogen bonds, among the strongest intermolecular interactions, are an obvious exception.)

Ionic compounds are not molecules, and ionic "bonds" aren't bonds in a molecular sense at all. To wit: in the solid, there is no such thing as NaCl, much less an NaCl vibration. This is because it's actually ..NaClNaClNaClNa.., and in three dimensions. Each ionic "bond" is exactly the same, and keep in mind the glue that holds this solid together aren't localized bonds but rather delocalized interactions between evenly spaced "point charges" - if you're willing to accept that approximation anyway.

Consider just a linear chain of three of the atoms, Na-Cl-Na. Let's say the Na-Cl on the left shrinks as part of a vibration. Well then the adjacent Cl-Na expands. And so on down the chain. What you have then in an ionic solid are coordinated lattice vibrations that extend in three dimensions, quasi infinitely for a perfect infinite crystal. These lattice vibrations, which can be stimulated by light absorption like any coordinated movement of charges, must obey the same kind of selection rules. But the symmetries involved are rather more complex - I haven't ever really tried to derive or understand them, but in principle they should be derived the same way. Also, since the lattice vibrations typically involve many more atoms than molecular vibrations (which as we've said are usually localized to small molecular regions or single bonds), they also tend to be far lower energy. The lattice vibration of NaCl, for instance, absorbs at about 164 cm^-1.

A lattice containing a polyatomic ion I imagine would exhibit a low energy lattice vibration, as well as higher energy vibrations within the "molecular" ion. Lattice vibrations aren't limited to ionic solids, either. Molecular crystals also exhibit lattice vibrations - as an example, the lattice vibration of glucose has a lattice vibration absorption at about 1.45 THz, or ~ 50 wavenumber. I haven't checked, but I'm guessing noncrystalline (amorphous) solids have no analogous transitions because long-range coordinated molecular vibrations isn't possible in disordered atomic arrays. 

[wildfyr]

Corribus, that was a truly excellent and complete response

[pgk]

In addition to the excellent response by Corribus, many mineral salts show IR signals further than 4000 cm-1 and thus, near-IR spectrometry can be used for the identifiication of minerals, as well for the detection of mineral impurities in organic compounds.