[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math

View post   
File: 96 KB, 1114x603, Butane-3D-balls.png [View same] [iqdb] [saucenao] [google]
10083083 No.10083083 [Reply] [Original]

How do we know what chemicals look like?

>> No.10083087

>>10083083
Define "look like"

>> No.10083089
File: 223 KB, 1013x1100, Beta-D-glucose-6-phosphate-3D-balls.png [View same] [iqdb] [saucenao] [google]
10083089

>>10083087
What experiments allowed us to determine the 3d structure of chemicals?

>> No.10083097

>>10083089
https://en.wikipedia.org/wiki/History_of_molecular_theory

>> No.10083103

>>10083089
Usually some setup of x-rax crystallography.

>> No.10083108

>>10083103
I thought that was only for proteins.

>> No.10083135

>>10083083
This is not how they "look" like. This is how model of them we work with can be visually represented, keep that in mind.

>> No.10083195

>>10083135
Interesting distinction. Who's to say molecules exist at all and that they're not just a model for higher level observations? Anyway I was just wondering what the exact experimental protocol was for determining bond angles and stuff like that. Looks like I'll have to dig into spectroscopy.

>> No.10083218

>>10083083
NMR, xray crystallography/scattering, IR, microwave blablabla

spectroscopy is your friend

>> No.10083291

>>10083089
Geometry and electric field theory after compositional analysis, refraction and crystallography and various light or polarization based techniques, scanning probe microscopy.

>> No.10083323

>>10083135
Good post. Same goes for then internal structure of atoms.

>> No.10083327

>>10083108
works for all crystal lattices, i.e. all solids
If you can crystallize it, you can determine the actual, physical structure using x-ray scattering.

Incidentally, this is the problem with analyzing protein structures using x-ray diffraction in general. It's difficult to crystallize proteins and it is not fully understood how the solvent affects the protein structure.

>> No.10083353

>>10083195
>Who's to say molecules exist at all

What are you, 12?

>> No.10083414
File: 15 KB, 496x456, ibm-microscope-sees-molecules.jpg [View same] [iqdb] [saucenao] [google]
10083414

>>10083083
Quantum shit

>> No.10083419
File: 675 KB, 1360x878, image.png [View same] [iqdb] [saucenao] [google]
10083419

>>10083083
>>10083089
Earliest work was done analyzing crystals utilizing x-ray diffraction since wavelength of x-rays are in a similar range like the atom (or molecule) distances in crystals. Von Laue and Bragg got Nobel prizes in 1914 and 1915 for this. It remains an important method today, but it's greatest limitation is that your sample has to be crystallized.
https://en.wikipedia.org/wiki/X-ray_crystallography

Most important method in Organic Chemistry today is probably NMR (nuclear magnetic resonance spectroscopy). It utilizes the magnetic interaction of the nuclei of certain isotopes that have a magnetic moment (1H, 13C). Certain structural elements show characteristic shifts and couplings in relation to other structural elements allowing you to piece together how the groups are connected in the molecule.

A bit more exotic are scanning probe microscopy methods. But you can produce really nice pictures with these. You can only analyze surfaces though.
Source of picture: Nature Nanotechnology 13, 371–375 (2018) (somehow 4chan is giving me spam alerts, when I try to link nature.com)

>> No.10085183

>>10083414
You can't just say quantum in any context and pretend it means something

>> No.10085211

>>10083353
Even worse, he's a philosopher.

>> No.10085244

Do people actually think that VSEPR theory is hard? Seems a lot like common sense but keep telling me they hate it

>> No.10085259

>>10083195
>Who's to say molecules exist at all
Wake up, Ostwald, your side lost

>> No.10085267

>>10083419
You're just capturing electromagnetic frequencies/vibrations and calling them atoms/molecules

>> No.10085394
File: 17 KB, 334x155, 334px-PTCDA.svg.png [View same] [iqdb] [saucenao] [google]
10085394

>>10085267
Yes. You could say about 19th century chemists that they were just treating matter with selective reagents until they found the most primitive compounds of a reaction and called those elements.
Compare this skeletal formula of PTCDA to the charge density plot above. The formula uses double bonds for simplicity. You could draw a resonance structure placing the double bonds differently. But when you measure the electron density you see there are no double bonds. It is just a formalism. The ring electrons are delocalized throughout the system. You can even see how multiple PTCDA molecules are connected through OH bonds.
Molecules, elements, atoms, electrons, protons, neutrons and other particles are a model. But several experiments provide evidence for the existence of such particles, such making the model useful to explain the structure of matter.
You don't have to do crystallography with x-rays. You could do it with electron or neutron beams.

>> No.10086025

>>10085267
Nothing is being captured. Interactions between the measurement device and the object being measured are recorded, that's how measurement works.

Measuring the forces generated by atoms and their chemical bonds isn't going to give you a clear cut picture of little balls rolling around, but it is going to show you the physical structure of the molecules they form.

>> No.10086046

>>10085183
Really? Well that Quantums.
I quantum that quantum means a lot of quantum.

>> No.10086392

>>10085183
But you can
Just not with experts in the field

>> No.10086484

>>10086392
I mean technically you can, you can do whatever you want, but outside of proper context prefixing random stuff with quantum is pointless.

>> No.10086491

>>10083414
Check this out desu:

https://www.nature.com/articles/nature19816