AmateurGenius6
Harmless
Posts: 2
Registered: 23-6-2014
Member Is Offline
Mood: No Mood
|
|
How accurate is chemical analysis, and what differnet methods exist?
If there is an unknown compound, with modern technology, how easy would it be to to exactly identify a chemical?
From the little information I have, I've heard that each compound has its own signature(lack of correct word?), and if the signature matches a known
compound, then that compound can be declared as such with confidence. But when if the chemical signature is not known? How easy is it for an equipped
analysis lab to identify an unknown compound that does not resemble any compound they have worked with before?
I've heard that different functional groups can put off different signatures, allowing for the structure to be guessed....but how scientifically
developed is this technology? What is actually possible and what isn't?
Like for instance how easily could an Analysis lab tell apart relatively inert compounds such as Benzene, Toluene, Xylene, Mesitylene, etc? These
compounds all have the methyl groups (Benzene excluded), but they're all relatively unreactive in a chemical sense. I know a simple boiling point test
is one way to identify them, if the boiling points are known, but what other techniques exist?
|
|
|
alexleyenda
Hazard to Others
Posts: 277
Registered: 17-12-2013
Location: Québec, Canada
Member Is Offline
Mood: Busy studying chemistry at the University
|
|
Many techniques exist for caracterisation :
-Chromatography (related to polarity)
-Infrared spectrum (shows up most function)
-Mass spectrometry (the most awesome one, destroys molecules and give the mass of the pieces and their quantity, which is representative of what is in
the molecule)
Caracteristics of the substance:
-Reaction under UV light
-Boiling point
-Melting point
-Solubility
-Flame tests
-Chemical reaction tests
-Refraction index
-Light absorption test
I guess is miss some, but that is a quite good list. It is hard not to identify something once you passed though all that.
Help us build the Sciencemadness Wiki! Every question and tips about amateur chemistry two clicks away, wouldn't that be awesome?!
sciencemadness.org/smwiki
|
|
|
AmateurGenius6
Harmless
Posts: 2
Registered: 23-6-2014
Member Is Offline
Mood: No Mood
|
|
I suppose what I meant is, is it possible to accurately determine the structure of a compound from those tests alone? In the case of either there is
no documentation on the molecule, or access to known BP, MP, UV reactions, are all unavailible. Imagine you only have the equipment, and no
documentation of past compounds. How accurately could the exact structure of compounds be determined? Is there a way to just put a compound under a
super-power electron scanning microscope, or something even more precise?
|
|
|
alexleyenda
Hazard to Others
Posts: 277
Registered: 17-12-2013
Location: Québec, Canada
Member Is Offline
Mood: Busy studying chemistry at the University
|
|
I guess a mix of infrared spectrum and mass chromatography would allow this most of the time as it really gives you data about what is in the
molecule. With a bit of logic it is possible to rebuild the molecules correctly most of the time. I don't think we're able to see atoms well enough to
identify molecules yet, but i'm not really aware of the progress in microscopy so I can't really tell.
I'm aware that i'm not answering your question completely but this is a good start 
[Edited on 24-6-2014 by alexleyenda]
Help us build the Sciencemadness Wiki! Every question and tips about amateur chemistry two clicks away, wouldn't that be awesome?!
sciencemadness.org/smwiki
|
|
|
kavu
Hazard to Others
Posts: 207
Registered: 11-9-2011
Location: Scandinavia
Member Is Offline
Mood: To understand is to synthesize
|
|
There are a multitude of ways. Most typically structures of organic compounds are solved using a variety of 1D and 2D NMR methods alongside with HRMS
and mass fragmentation patterns. If the compound is crystalline Xray techniques can also be applied. Structures of even very complex natural products
have been solved via these methods. In some cases there are ambiguities and further chemical modifications to the structure are needed to get definite
information. In many cases the atomic connectivity can usually be resolved easily, but absolute (and in some cases relative) stereochemistries posses
more a significant challenge.
For example, I have had a few reactions producing unknown side products. Within just hours from isolating the compound I've managed to figure out the
exact structure based on HRMS, IR, 1H, 13C and COSY. This kind of stuff is done routinely in universities and the industry on daily bases.
[Edited on 24-6-2014 by kavu]
|
|
|
WGTR
National Hazard
Posts: 971
Registered: 29-9-2013
Location: Online
Member Is Offline
Mood: Outline
|
|
This isn't a fancy method, but if a sample can be made to approximate an ideal gas, its molar mass can be determined by weighing a sample at known
volume/temp/pressure/etc. I've done this in a syringe with common hydrocarbon gasses, although it took a very sensitive and accurate balance, as
well as careful handling. Molar mass doesn't tell you everything, but it's one piece of the puzzle.
|
|
|
phlogiston
International Hazard
Posts: 1376
Registered: 26-4-2008
Location: Neon Thorium Erbium Lanthanum Neodymium Sulphur
Member Is Offline
Mood: pyrophoric
|
|
Different analytical techniques yield different pieces of the puzzle. Depending on how big the puzzle is, you may need to gather more information to
solve the puzzle.
Simple small organic molecules are commonly identified conclusively by a combination of
- infrared spectroscopy (IR), which mainly provides information on the different functional groups in the molecule
- mass spectrometry (MS), which mainly yields information on structure and atomic composition (isotope peaks)
- Nuclear magnetic resonance (NMR), which mainly yields information on structure
Nearly always you have some background information available as well on the origin of the molecule that can greatly help guide your guess, but it is
certainly possible to identify a compound by analytical techniques alone with no further information.
There are almost literally countless of other techniques and varieties that can be used if needed, but if you are interested I suggest you start
looking into the above three first.
I don't believe there are actually any molecules that can not be identified at present (provided enough of it is available, etc)
-----
"If a rocket goes up, who cares where it comes down, that's not my concern said Wernher von Braun" - Tom Lehrer |
|
|
ziqquratu
Hazard to Others
Posts: 385
Registered: 15-11-2002
Member Is Offline
Mood: No Mood
|
|
As the previous posters have noted, there are a multitude of techniques available for identification of molecules. And what you need depends on where
your compound came from (for example, if it was extracted from a plant or is as an unexpected product from a reaction).
Chromatography (HPLC, GC, TLC and so on) is useful for separating mixtures of compounds into individual components, and in some cases for measuring
the amounts of each component that is present. there are many different mechanisms that can be used for the separation, but essentially you have an
unmoving "stationary phase" and a moving "mobile phase", and compounds separate based on their relative affinity for each - so something that prefers
the mobile phase will move faster than something that is happy with either, which will move faster than something that prefers the stationary phase.
It provides no structural information, although if you know what compounds are present in a mixture you may be able to identify each component based
on how quickly it moves under a known set of conditions.
Mass spectrometry is an extremely important for determining chemical structures. It provides the molecular mass of a compound - and since it can be
accurate to the mass of an electron, you can reliably calculate the molecular formula (or a small number of probable ones) from this. It also tells
you about isotopes, which can allow you to figure out if a less common atom (i.e. something like a halogen or a metal) is present, and even how many
of those atoms. Finally, by adjusting the conditions, you can make your molecule break up and then detect the pieces, which can give you useful
information about substitutents - for example, if you see a peak that is 29amu less than the molecular ion, you can guess that your molecule contains
an ethyl group. Likewise, a peak that is 31 less would suggest that CH3O is present - perhaps a methyl ether or ester. Furthermore, when
the molecules break up, you can use the pattern of fragments as a fingerprint and, by comparing with databases of known compounds, you can identify an
unknown if someone has identified it previously.
Spectroscopic methods - primarily NMR and IR - are probably the most powerful tools for structural elucidation. IR can tell you what functional groups
are present in a molecule - and provide some limited information about their connectivity (for example, you can often distinguish between o-, m- and
p-substituted benzenes). The vast array of different NMR experiments can give you enormous information. Simple spectra can tell you about the
environment in which an atom finds itself - that is, what type of things are attached to it. Others can tell you how many protons are attached to a
particular carbon atom; or what carbon atoms are attached to each other; or what protons are spatially near each other, even if they aren't directly
attached.
Finally, the pinnacle of structural elucidation is x-ray crystallography, which physically measures the locations of atoms in a crystal. Of course,
you need a crystal for that (and many compounds are anything but crystalline!), and it still takes a lot of work to figure it out.
Ultimately, you usually need a combination of these techniques to identify a compound, depending on what you already know. For example, if you're
looking at an unexpected synthetic product, you have the advantage of knowning what went in, and thus of being able to make predictions about what the
product might be. So, a few NMR experiments might be enough for you to work out what it is. On the other hand, if you have something from a random sea
sponge, you may have no clue about the structure and will need every tool at your disposal to figure it out.
Perhaps the last remaining area in which there are significant limitations is in figuring out the stereochemistry of an unknown product. The tools for
determining absolute stereochemistry (that is, R or S) are actually very limited - particularly if you only have a tiny amount of a sample. And whilst
NMR can be quite good for determining relative stereochemistry (that is, the orientation of two groups relative to one another - for example, do they
both point the same way or is one back and one forward?), it can also be very difficult to interpret the data and get right - particularly for complex
natural products or where you only have a tiny amount of material (and often you can only isolate sub-milligram amounts of a natural product, for
example!). Even if you get lucky enough to obtain an x-ray structure, which will without doubt confirm relative stereochemistry, it can still be
difficult to be sure of the absolute stereochemistry.
|
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
