|
|
|
|
|
|
|
|
|
 Posted: Sat May 10, 2008 9:13 pm
Heres to anyone and everyone about Computational Chemistry/ Molecular Modeling. Does anyone else do it? Would you like to learn about it?
Whats your favorite computer build? Software?
Do you do you're own tech support or do you have outside support or student support? (which is REALLY cool advantage of teaching...you can never know what knowledge students bring to the table until you ask)
Anyone?
|
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: Sat May 10, 2008 10:47 pm
Not me. Would you mind telling me a bit about it? I'm always interested in learning something new.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: Sun May 11, 2008 2:07 am
Somehow molecular modelling reminds me of HYSIS and UNIFAC... crying
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: Sun May 11, 2008 11:47 pm
Melchom Not me. Would you mind telling me a bit about it? I'm always interested in learning something new. Ditto, I'm ALWAYS looking to learn something new!
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: Mon May 12, 2008 4:54 am
So, essentially computational chemistry is using the laws of physics we already know that govern molecular interaction, and using them to write a fairly simple computer program.
Then we input a molecule such as, say for example, benzene. Now a single molecular dynamics step would be to move every atom slightly so each one reaches a lower energy state. This is done 1000's to 10's of 1000's of times, and if you watch a kind of replay of the molecules path, it looks a lot like its vibrating.
Molecular dynamics is probably the most simple of the computational chemistry techniques. Its referred to (correctly) as the 'ball and spring' model, because thats exactly how it models molecules. Each atom is a 'ball' that has mass and velocity (and therefore inertia), and the bonds act like springs: they can be twisted, pulled, pushed and bent (to some degree).
Theres actually free software out there if you'd like to try it for yourselves I think...
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: Mon May 12, 2008 10:39 am
That sounds really cool =p I've always wondered whether we can just put everything we know about the universe into a computer and try and simulate it. Just all the laws and random facts or phenomena, the sum of our knowledge.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: Mon May 12, 2008 10:06 pm
This does sound pretty cool!
I was in an organic chem class where we used a program to model some molecules and find their energy configurations but it was before I had more physics so I didn't understand the theory behind it very well... I'm not sure if it is the same as what you are talking about.
I sometimes get intertia and momentum p = mv confused. How do you use intertia in the ball and spring model?
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: Fri May 16, 2008 10:32 pm
Melchom That sounds really cool =p I've always wondered whether we can just put everything we know about the universe into a computer and try and simulate it. Just all the laws and random facts or phenomena, the sum of our knowledge. Unfortunately that doesn't quite work. We have to pick and choose which laws and phenomena we model and which we dont (just approximate instead) even disregarding all electron interactions I still need a supercomputing cluster to run major simulations (major simulations at the quantum level are not possible at this point in time).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: Fri May 16, 2008 10:35 pm
Mecill This does sound pretty cool! I was in an organic chem class where we used a program to model some molecules and find their energy configurations but it was before I had more physics so I didn't understand the theory behind it very well... I'm not sure if it is the same as what you are talking about. I sometimes get intertia and momentum p = mv confused. How do you use intertia in the ball and spring model? Essentially as if it were a literal ball. Because it has mass, it has inertia, and must be accelerated proportionately to the force(s) that are acting on it. I don't think inertia is expressly calculated; what we need to keep track of for each atom is three vectors and a scalar: The scalar is its mass the Vectors are: Position
Velocity and
Force that is acting on the atom.
The rest is force field data.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: Sun May 18, 2008 6:19 am
Paul A M Dirac Melchom That sounds really cool =p I've always wondered whether we can just put everything we know about the universe into a computer and try and simulate it. Just all the laws and random facts or phenomena, the sum of our knowledge. Unfortunately that doesn't quite work. We have to pick and choose which laws and phenomena we model and which we dont (just approximate instead) even disregarding all electron interactions I still need a supercomputing cluster to run major simulations (major simulations at the quantum level are not possible at this point in time). =p well yea, I sort've figured you'd have to leave some out. Definitely have to leave out quantum mechanics. I just mean that it would be really cool if we could.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: Sat May 24, 2008 12:21 pm
Links to this! Links to this now! And marry me, damnit!
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Reply
