log in

Advanced search

Message boards : Science : Please tell me what I am

Author Message
Charles LaDue
Send message
Joined: 1 Dec 12
Posts: 5
Credit: 64,673
RAC: 3
Message 196 - 9 Jan 2014, 15:25:39 UTC

What is Rioja Science, curious minds want to know. Please anyone with more information than myself fill in the many threads that should be here.

Administrador
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 13
Posts: 57
Credit: 0
RAC: 0
Spain
Message 208 - 9 Feb 2014, 21:57:06 UTC - in response to Message 196.

Hi Charles.
Sorry for delaying the answer too much. I was revising forums and realized that although I answered you, the message wasn't envy. It was my fault, not forum problems.

We use boinc platform to study the dynamics of gas phase reactions.
We use a triatomic or pseudotriatomic models, it means that we study reactions like A + BC.
The calculations running now are to study the dynamics of N +NO reaction.
In a very simply way we are reproducing the reaction like three balls colliding on a surface. This surface has barriers, local minima ... This surface is the potential energy surface, that describes how reaction occurs and it is calculated using quantum chemical calculations. The quantum calculations give the potential energy for different A-B-C positions.

The three bal collisions are studied by classical dynamics equations i.e. Newton's movement equations. It is like a three mass point attraction problem but the potential is not gravitational but a chemical potential.
So, a chemical collision is named as a trajectory and to reproduce the “real” reaction we need to calculate thousands or millions of trajectories for statistical purposes (to check all of the potential energy surface). Here we are using BOINC, we send to the volunteers batches of trajectories to be calculated. The number of the trajectories per batch depends on the reaction to study (a fast reaction is fast to be studied).

I don't know if it is clear enough, but if you want more information or bibliography, don't bother to ask.

Best,
R.

Profile Crunch3r
Avatar
Send message
Joined: 8 Mar 14
Posts: 2
Credit: 15
RAC: 0
Germany
Message 235 - 8 Mar 2014, 20:07:07 UTC - in response to Message 208.


We use boinc platform to study the dynamics of gas phase reactions.
We use a triatomic or pseudotriatomic models, it means that we study reactions like A + BC.
The calculations running now are to study the dynamics of N +NO reaction.
In a very simply way we are reproducing the reaction like three balls colliding on a surface. This surface has barriers, local minima ... This surface is the potential energy surface, that describes how reaction occurs and it is calculated using quantum chemical calculations. The quantum calculations give the potential energy for different A-B-C positions.

The three bal collisions are studied by classical dynamics equations i.e. Newton's movement equations. It is like a three mass point attraction problem but the potential is not gravitational but a chemical potential.
So, a chemical collision is named as a trajectory and to reproduce the “real” reaction we need to calculate thousands or millions of trajectories for statistical purposes (to check all of the potential energy surface). Here we are using BOINC, we send to the volunteers batches of trajectories to be calculated. The number of the trajectories per batch depends on the reaction to study (a fast reaction is fast to be studied).

I don't know if it is clear enough, but if you want more information or bibliography, don't bother to ask.

Best,
R.



Hi there,

is the source code going to be available for download ?

Administrador
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 13
Posts: 57
Credit: 0
RAC: 0
Spain
Message 236 - 10 Mar 2014, 8:32:52 UTC - in response to Message 235.

Hi Crunch3er.

Sorry but the code will not be able for the volunteers.
THe code we are using is a modified version of an quite old one, so there have been many authors involved in the developing of different subroutines (is a fortran code) so now is impossible to get the agreement of all of them.
Several years ago there was an attempt to make the code available to the scientific community publishing it, but it was unsuccessful.

Best,
R.

Profile Crunch3r
Avatar
Send message
Joined: 8 Mar 14
Posts: 2
Credit: 15
RAC: 0
Germany
Message 290 - 12 Mar 2014, 18:28:13 UTC - in response to Message 236.

Hi Crunch3er.

Sorry but the code will not be able for the volunteers.
THe code we are using is a modified version of an quite old one, so there have been many authors involved in the developing of different subroutines (is a fortran code) so now is impossible to get the agreement of all of them.
Several years ago there was an attempt to make the code available to the scientific community publishing it, but it was unsuccessful.

Best,
R.


That's to bad...

considering that the code is a couple of years old, it's most likely not vectorized at all...

Good luck with your project, but that's a no go for me.



____________

Improving science by joining team BOINC United.

Aurel
Send message
Joined: 15 Apr 13
Posts: 1
Credit: 436
RAC: 0
Germany
Message 538 - 16 Mar 2014, 17:39:40 UTC - in response to Message 208.

Hi Charles.
Sorry for delaying the answer too much. I was revising forums and realized that although I answered you, the message wasn't envy. It was my fault, not forum problems.

We use boinc platform to study the dynamics of gas phase reactions.
We use a triatomic or pseudotriatomic models, it means that we study reactions like A + BC.
The calculations running now are to study the dynamics of N +NO reaction.
In a very simply way we are reproducing the reaction like three balls colliding on a surface. This surface has barriers, local minima ... This surface is the potential energy surface, that describes how reaction occurs and it is calculated using quantum chemical calculations. The quantum calculations give the potential energy for different A-B-C positions.

The three bal collisions are studied by classical dynamics equations i.e. Newton's movement equations. It is like a three mass point attraction problem but the potential is not gravitational but a chemical potential.
So, a chemical collision is named as a trajectory and to reproduce the “real” reaction we need to calculate thousands or millions of trajectories for statistical purposes (to check all of the potential energy surface). Here we are using BOINC, we send to the volunteers batches of trajectories to be calculated. The number of the trajectories per batch depends on the reaction to study (a fast reaction is fast to be studied).

I don't know if it is clear enough, but if you want more information or bibliography, don't bother to ask.

Best,
R.


Thats sounds interesting!
But, it seems a little bit like QMC@Home. (quantum physics and science).
Whats about an Batch Monitor? Will you make an, or not? Who many Workunits are needed...
There are so many questions...

Aurel

Administrador
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 13
Posts: 57
Credit: 0
RAC: 0
Spain
Message 539 - 17 Mar 2014, 9:12:13 UTC - in response to Message 538.

Hi Aurel!
It seems that QMC@home project is related to medicinal chemistry, so large biological and organic compounds are involved. It is a very interesting branch but different to our project.
Our project is related to atmospheric and space chemistry where gas-phase reactions are important.
It is true that in both projects Quantum Chemistry (QC) is used, that is because QC is a powerful tool used quite often in Theoretical or Computational Chemistry.

About the Batch monitor, it is difficult to say now because we can not to predict when a mini-project is finished, it depends on the desired statistics, calculated properties, studied reaction ... But it sounds interesting to be commented by the team.

There are so many things to do ... ;)

Best,
R.

Profile [VENETO] boboviz
Send message
Joined: 31 May 12
Posts: 4
Credit: 128
RAC: 0
Italy
Message 620 - 23 Oct 2014, 7:41:30 UTC - in response to Message 539.

Scientific results? Publications?

Administrador
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 13
Posts: 57
Credit: 0
RAC: 0
Spain
Message 623 - 23 Oct 2014, 15:13:04 UTC - in response to Message 620.

Working on it!
In this month we are submitting a paper to a Physical Chemistry journal. The work is about O+CH4 combustion reaction. I hope in few months the reference is on the web page!

We are also preparing a work "related" to green chemistry where Hg atom is involved.

The calculations running now for the N + NO reaction, also interesting in combustion chemistry and the beta version of the program is to calculate oxygen atoms reactions with surfaces.

Not too much, I know, but this is.

R.

Profile [VENETO] boboviz
Send message
Joined: 31 May 12
Posts: 4
Credit: 128
RAC: 0
Italy
Message 625 - 23 Oct 2014, 20:35:02 UTC - in response to Message 623.

Not too much, I know, but this is.
R.


Not so bad!!!! :-)


Post to thread

Message boards : Science : Please tell me what I am


Main page · Your account · Message boards


Copyright © 2017 Knet comunicaciones S.L.