All 16 entries tagged Physics
View all 41 entries tagged Physics on Warwick Blogs | View entries tagged Physics at Technorati | There are no images tagged Physics on this blog
August 21, 2007
In other news, we may have spent a while filling a whiteboard with text, but the project now works! We can generate one million double beta decay events, complete with energies and angular distributions, in a little over a minute (on a 1.86GHz processor).
Today we’re going to tidy the code up a bit and complete all the documentation, then we can turn it into a library and interface with other peoples code, such as the CERN GEANT4 package, which will allow us to track the motion of the particles our program generates, as they pass through various detector materials.
August 11, 2007
After last week’s work on optimising the parameters for the simulations, previous spectral simulations were redone to improve their quality.
At the end of last week my supervisor had asked me to prepare a talk for the next group meeting, which was to be held on the Friday morning. The majority of the week was thus spent creating the associated powerpoint slides, which I found to be a very useful exercise. The first thing I learnt was that you cannot include everything you’ve done into a talk; you have to pick out the sections that worked well and which link together in order to form a story. For this reason the talk omitted the simulations work, focusing only on relevant previous experimental work, my research and what they told about the possible structures of the samples.
I did a run through of the talk with a colleague, which was great for identifying flaws. After this the talk itself seemed to go fairly well :)
On Friday afternoon and the following Monday and Tuesday I started work on my poster.
Back in Devon now and enjoying the sunshine, but still with my poster to do. I’ll let you know when I finish!
Part of one day was spent looking again at the experimental spectral peaks we found and trying to come to some conclusions about whether it is what we expected.
Throughout the week we did absolutely loads of simulations work – some of which took quite a few goes to get:
a) atomic coordinates from molecule modelling programs and
b) input files written so that spectral peaks look as you would expect from experiment (there are several parameters that one can set)!
We tried a 2D simulation and did some 6-spin ones using coordinates from atoms that would be on adjacent molecules in a crystal.
When I started the project I was under the impression that simulations of our samples would have been completed by this stage. However I now realise that these things have to be done systematically and that you have to ensure that the basics are correct before moving onto the more complex.
We met Erwin Hahn! Who honourarily (?) graduated on Monday with the other physicists :) Go here to see him and the rest of us!
Work-wise I did more simulations using Spin Evolution, but this time using different pulse sequences. Amy did the equivalent using Simpson so we can compare.
Experimentally we had planned to run quite a few samples on the 600 MHz spectrometer but we had difficulty with various things so we didn’t get as much done as expected (research is seemingly a very slow process at times!).
On Thursday I did a 4-spin simulation using Spin Evolution. I chose atoms in the penicillin molecule to check that interatomic distances were represented on the spectum as expected. I also looked at the effect of changing the spinning speed and number of propagations.
Friday we ran a 2D experiment on sample B04. We got a lot of noise so left it running over the weekend.
August 03, 2007
Week five in the physics cave (literally – our office has no windows!) and we’ve made significant progress. For those of you reading this who forgot what it is we’re doing, the task is to create a program which simulates double-beta (ßß) decay events and produces a list of energies and angular distributions of the resultant particles.
The last few weeks involved implementing lots of helper functions and classes; numeric integration, a four-vector, various classes to represent different kinds of particles, and other associated trickery. This week, we started the biggest single part of the program; the event generation functionality itself.
At the start of this week we hoped to have this finished by today; unfortunately that did not happen! We’re getting close, though. Understanding what we have to do is half the task!
We’ve also named our project PEBBLE – momentum(p)-Energy ßß (double-Beta) Leptonic Event!
July 22, 2007
This week brought more complex simulations in the form of Post C7 3-spin (3 atom) ones (after working out the problems of last week’s 2-spin ones). With more spins the Simpson simulation becomes a lot slower. I only managed one arrangement of the 3 spins (at different spinning speeds and number of pulse sequences) in one day, although this process had to be done accurately for suitable comparison with the equivalent setup using the Spin Evolution simulation.
Throughout the week I tried three more arrangements of three atoms, modelling them on the compound alanine and looking at both carbon and hydrogen spectral simulations. The amplitudes of each arragement’s spectra after different numbers of pulse sequences produced what is known as a build up curve. They should look fairly similar in shape…some did but some of the later ones didn’t…This is probably a problem with estimating the spectral peak amplitudes and I will try different methods at some point. If I do get good curves, it it a good indication of the simulations working correctly.
Amy put B05 in the spectrometer so now we have all six of the CP spectra!
July 17, 2007
Just a short update from me today, since my last blog i have found a suitable graph producing library for PHP (a web based programming language), and collated the data that needs to be plotted.
In the background of the plots I needed to plot the theoretical allowed values- this presented a challenge, as none of the academics in the department remebered how! However after a trawl through papers, and a week of fiddling and coding i was able to produce the graph below, in the correct form, and generating the curves on the fly, so eventually the user will be able to change the paramters at will!
July 12, 2007
It’s now almost the end of week two and I’ve yet to post anything here, so here goes…
Sam and I have just finished revision 1 of a design document for the program we have to write. Hopefully there won’t be too many changes to make before we can get on with implementing it, but there will inevitably be some!
On a lighter note, we created a new particle; the Duckton, which mediates a vector field known as the Pond. A duckton (see the photo below) is a bound state of three quacks. Each quack has a colour (red, green, blue) and the Duckton is a yellow combination of quacks.
(No, we were not bored, just insane.)
There are some photos of the “office” below, too. It spends most of the year masquerading as a meeting room, so you can imagine the size!
July 09, 2007
To start the week I continued with reading but more concerning the physics behind simulation parameters than how the simulations work.
I tried two simulation programs and tried to get matching spectra on both – usually unsuccesfully! This is a minor setback and I’m sure that as I accumulate a greater understanding of the programs, my use of them will improve. I asked some of the others for help and have a few things to try for next time.
Using a previously found paper, which listed the chemical shifts of a very similar compound to those I am studying, I compared shifts in compounds. I found the expected similarities – indicating that the two samples have common molecular groups. I also found a large anomaly for one of the samples, suggesting that the sample isn’t of the structure we first assumed it to be.