| Pages:
1
2 |
JohnWW
International Hazard
Posts: 2849
Registered: 27-7-2004
Location: New Zealand
Member Is Offline
Mood: No Mood
|
|
That last post - re neutron scattering properties of D2O - has NOTHING to do with the effect of D on NMR spectra! There is no way that it can be a
legitimate criticism of my last post.
I stick by what I said in my last post. The Hs in alcohols are much less acidic, and hence much less labile (liable to ionization and exchange with D
from D2O), than in carboxylic acids (which have O= on the same carbon), and indeed are less acidic than water. UNLESS, as I said before, there are
strongly electron-drawing groups like -NO2 on either the same carbon or a neighbouring carbon (which also results in deshielding of the alcoholic (or
phenolic) H and hence shift in NMR spectra).
Besides, only a relatively few short-chain alcohols (and phenols) are soluble in water (or D2O). The longer-chain ones are insoluble waxes or oils,
which rules out even trying to exchange their alcoholic Hs with D2O. Carboxylic acids with the same numbers of carbons are much more water-soluble,
due to their greater polarity.
John W.
|
|
|
mick
Hazard to Others
Posts: 338
Registered: 3-10-2003
Member Is Offline
Mood: No Mood
|
|
Carrying out reactions in deuterated solvents and using stuff like deuterated acetic acid, sodium hydroxide, pyridine etc is useful because it can
make it easier to follow some reactions by NMR.
Mick
|
|
|
Proteios
Hazard to Others
Posts: 109
Registered: 7-3-2004
Member Is Offline
Mood: No Mood
|
|
| Quote: | Originally posted by JohnWW
.. has NOTHING to do with the effect of D on NMR spectra!
John W. |
...chuckle... i think u might have missed the point.....D2O is very common. However the statement above is not true... both neutrons scattering
properties and NMR spectra depend upon nuclear spin. Youre right about the acids though.
|
|
|
Marvin
National Hazard
Posts: 995
Registered: 13-10-2002
Member Is Offline
Mood: No Mood
|
|
Disentangling the threads somewhat,
John wrote in the deuterated acetone thread,
"As for alcohols not exchanging Hs with D2SO4, I only went so far as to say that the rates of such an exchange would be much less than that of
carboxylic acids due to the much lesser degree of ionization; and that it would practically not happen for other than the shortest-chain alcohols,
because higher alcohols (which are waxes and oils) are practically insoluble in water."
We wernt discussing alcohols with D2SO4, just plain D2O. Even moderatly concentrated D2SO4 in D2O would be expected to cause oxygen exchange, which
is another level entirly.
In this current thread,
"....I stick by what I said in my last post. The Hs in alcohols are much less acidic, and hence much less labile (liable to ionization and
exchange with D from D2O), than in carboxylic acids (which have O= on the same carbon), and indeed are less acidic than water. UNLESS, as I said
before, there are strongly electron-drawing groups..."
Going from a random webpage from google the timescale for most alcohols are 10^-5 seconds for an exchange. This is not slow.
http://www.chemistry.ccsu.edu/glagovich/teaching/472/nmr/cou...
A textbook grabbed from the shelf behind me, "Spectroscopic methods in organic chemistry" a pretty standard undergrad textbook calls it the
"D2O shake" and its number 3 on the tips for determining an unknown structure right behind identifying solvent peaks and trying to find an
integral for a single proton. How much more basic a method can you get!
The 'method' if you can call something so simple a method says 'if the sample in CDCl3 or CCl4 solution, is shaken with a drop of D2O
the OH NH and SH hydrogens exchange rapidly with the deuterons, the HDO floats to the surface out of the region examined by the spectrometer and the
signal of the OH NH or SH simply dissapears from the spectrum (or more usually is replaced by a weak signal close to d4.8 coming from suspended
droplets of HDO).
and later on,
"Besides, only a relatively few short-chain alcohols (and phenols) are soluble in water (or D2O). The longer-chain ones are insoluble waxes or
oils, which rules out even trying to exchange their alcoholic Hs with D2O. "
Irrelavent, as is commonly known, and mick has already pointed out in the other thread anyway, the solvent for NMR is usually organic and more
frequently chloroform or carbon tet. Pure compounds are not used.
and from much earlier,
"But the shift of carboxylic acid Hs are by far the largest in proton NMR spectra of organic compounds because of having the greatest level of
deshielding, and so have very distinctive NMR spectral signals, anyway, usually easily capable of being distinguished from alcoholic and phenolic Hs.
Using D2O to replace these (being the most labile Hs in organic chemistry) with D, which does not show up in proton NMR, would thus be of limited use.
"
Removing these lines is not just about having less lines on the spectrum. Proton exchange also blurs static hydrogen lines making multiplets more
difficult to interpret, and as the molecules get more complex the job of interpreting goes up massively.
I trust this clarifies the matter.
It does work, it is used often and it is not of very limited help.
[Edited on 9-9-2004 by Marvin]
|
|
|
mick
Hazard to Others
Posts: 338
Registered: 3-10-2003
Member Is Offline
Mood: No Mood
|
|
If you run a COSEY and NOESY experiment on a fairly complex compound containing sugar and/or protein residues before and after a D2O shake there is a
good chance that you can work out the 3D structure in solution. It is a fairly powerful tool. The instuments today have superconducting magnets and
work on the FT principle. There is an excellent free FT-NMR conversion program on the net. The old ones had a large magnet and worked with radio wave
frequences, they always looked fairly straight forward to me. Does anyone know the most powerful magnet that can be built at home, I will have a look.
Home built NMR instrument? Some thing to think about.
Thanks
mick
Typo
[Edited on 10-9-2004 by mick]
|
|
|
JohnWW
International Hazard
Posts: 2849
Registered: 27-7-2004
Location: New Zealand
Member Is Offline
Mood: No Mood
|
|
Because of other electron-drawing groups in the molecules, often vicinal, the alcoholic Hs in saccharides and any in peptides should be more labile
(more capable of exchange with D2O) than those in ordinary alcohols, and the molecules themselves would be (for chains of the same lengths) more
water-soluble for the purpose. At the same time, their -OH NMR shifts would be intermediate between those of ordinary alcohols and of carboxylic
acids, and similar to those of -OHs with e.g. alpha-nitro groups, posing difficulty in identifying them. This would be reflected in their acid
dissociation constants. So in this case, exchange of H for D using an excess of D2O, to modify the NMR spectrum so as to determine the -OH
environment, is certainly useful.
As regards making your own NMR spectrometer - you could look up patents (most of which should have expired by now) for them with the US Patent Office,
and other countries' patent offices. Many patents are on-line on their websites. But it would be fairly expensive. Varian Associates seems to be
the longest-established and best-known maker of NMR spectrometers.
John W.
|
|
|
mick
Hazard to Others
Posts: 338
Registered: 3-10-2003
Member Is Offline
Mood: No Mood
|
|
A home made NMR should be possible. You spin the stuff in a big magnet, more of the protons spin in the same direction due to the magnetic field. Hit
them with a broad band radio frequence and some off them will spin the other way absorbing the radio frequency. Measure the output and you could get
an NMR spectra. I know there is a problem with relaxation time and other stuff but if you used modern electrics it could be better than the original,
and no superconducting stuff. I think you could get a good NMR from 40MHz magnet, I think they go upto over 900MHz now.
mick
The edit was because I forgot to sign it
[Edited on 10-9-2004 by mick]
|
|
|
Proteios
Hazard to Others
Posts: 109
Registered: 7-3-2004
Member Is Offline
Mood: No Mood
|
|
| Quote: | Originally posted by JohnWW
Because of other electron-drawing groups in the molecules, often vicinal, the alcoholic Hs in saccharides and any in peptides should be more labile
(more capable of exchange with D2O) than those in ordinary alcohols, and the molecules themselves would be (for chains of the same lengths) more
water-soluble for the purpose. At the same time, their -OH NMR shifts would be intermediate between those of ordinary alcohols and of carboxylic
acids, and similar to those of -OHs
John W. |
a nice piece of specious reasoning there...... peptide hydrogens are really quite non-acidic due to the resonance/hydbridisation.....or whatever
chemical handwaving you want.... but the bottom line is.....peptide hydrogens are relatively very stable/non-acidic (ergo proteins dont fall apart
that easily). Google before flight of fantasy on what you think the universe should look like.
Your hand waving on chemical shifts leaves something to be desired too!
|
|
|
Marvin
National Hazard
Posts: 995
Registered: 13-10-2002
Member Is Offline
Mood: No Mood
|
|
Home NMR is something I've considered. A feasable electromagnet would be limited by iron saturation at (from memory) about 2.2 Tesla, this would
be enough for a 100MHz (proton, as will all the numbers I quote be) NMR machine.
Making a magnet at home would be a nightmare though, you need field homoginity in the 1 part in 100 million, or better, 1 billion range for decent NMR
and theres only so far winding your own shim coils will go. The magnet itself would take a lot of power unless the sample area is tiny. Moving down
to 60MHz would help a lot. Perminant magnet systems are then feasable, or electromagnet systems become a lot easier. Very very accurate machining of
the iron core would be needed to even attempt this, an off the shell magnet not specifically designed for NMR would be useless.
If you can find a secondhand NMR magnet (not superconducting obviosly) this cuts out a lot of problems.
Youd need to be a good radio engineer to build a 100MHz or 60MHz setup electronics. Youd have to superheterodyne the thing into low noise filters and
the whole project is more like a lifes work if you add a stable magnet to the construction list.
Seems like a reasonable point in a relevent thread to talk about my own plans for a NMR spectrometer. Audio frequency!
Now why the hell would anyone want an audio frequency NMR spectrometer you ask, and you ask it because you'd lose all the information regarding
molecules. The answer is that its still useful because you can still determine isotopes and their concentrations. For those people with little
imagination, call it a 20th century tricorder. You put the sample of a few hundred ml solution inbetween a Helmholtz magnet, you have a pick up and
drop (in nuclear terms) magnetic field switching on and off, the audio coil wrapped round the sample is connected to an amplifier, to a filter, to
another amplifier (gain of a million or more needed) and then into a PC soundcard and signal avaraging coaxes a spectrum out of the noise over time,
(say an hour or so).
From that you get a list of concentrations of all 'visible' isotopes (with a non zero magnetic moment, ie all with odd numbers of neutrons
or an odd number of protons).
Cheep, cheerful and lets you try isotopic enrichment at home, want to know how much deuterium is in that solution after electrolysis? Shout at it and
for some basic electronics (in addition to a PC), less complicated than a ham radio you can hear the reply.
There are only a few elements that wouldnt show up at all if you want to do elemental analysis and you are happy to assume natural abundances. Cerium
is one. There are a fair few elements that might be impractical to see in reasonable time though, being a very low field instrument sensitivity
suffers horribly.
And I know what some people would be thinking, why cant we use a nice stable Helmholtz field to do high field NMR. The answer is with only a few
kilograms of copper (reasonable assumption I think) in coil windings and a reasonable sized coil (circa 10cm I think my math was for), youd be talking
a kilowatt+ to get more than a few thousand gauss (1000 gauss = 0.1Tesla) and with power usage going up with I^2R but field strength only going up
with I, this is close to limiting for this idea, 5 to 10MHz proton NMR not being useful for molecules. I do have some tricks up my sleeve for
increasing the resolution of a 60 or 100MHz instrument, but it couldnt be used commercially as it would take far to much time. I did think about
patenting the idea for an NMR device that did mineral analysis or isotopic analysis with audio frequency, but again only amateurs would be willing to
wait hours or overnight for a single sample result.
|
|
|
mick
Hazard to Others
Posts: 338
Registered: 3-10-2003
Member Is Offline
Mood: No Mood
|
|
Thank
Point taken. The idea came from looking at Mr Fusions stuff with the nice array of magnets. A couple of years ago an old 60 MHz NMR got skipped. The
worst one I think I have missed is all the old PYE 104 GLC's. They seemed to be so straight forward and repairable. You could plug bits into them
to up grade them and unplug the bits and the thing would still work.
I like the idea of isotope detection, you are looking for the the isotope and not environment. Analysis is 99% knowing what you are looking for.
mick
A quick edit on analysis
A chap thought his radio receiver on the roof had been sabotaged and it was personal. It looked like silicone sealant. A quick test in a bunsen flame
proved it was bacon rind. I think it was security or some one eating a bacon sandwich.
mick
[Edited on 12-9-2004 by mick]
|
|
|
mick
Hazard to Others
Posts: 338
Registered: 3-10-2003
Member Is Offline
Mood: No Mood
|
|
I was thinking that if you got some of these NeFeB magnets and clamped them in the right position then something must happen to the nuclei , it is
just measuring it that is the problem.
mick
Typo
Spelling is not very good, chemistry OK
[Edited on 12-9-2004 by mick]
|
|
|
| Pages:
1
2 |
![[*]](images/xpblue/default_icon.gif){kind=icon}
