Saturday 8 November 2014
Monday 27 October 2014
Cavity size (Technical)
So I've just finished some calculations that took up a fair amount of computing time of Raijjin. But which came to basically nothing as I just proved my hypothesis wrong. This isn't totally a bad thing as I can rule it out as a possibility. But I can't help but wonder whether someone else has had the same idea tested it realised it didn't work and therefore abandoned it, and I never heard about it as there was nothing to publish. So I ended up wasting my time by checking it. I guess this is a similar problem to what they have in Medicine and Psychology where negative trial results just sit in the bottom drawer and never get published.
It's a little different I guess as there's no way I could write up and publish every idea I had that came to nothing, it would make reading journals a nightmare sorting through all the things people had tried. I guess we could do with some kind of online wiki article where we all describe things we tried that didn't work to solve a problem so others know not to try it as well.
I might as well describe the idea. It was that as two ions come together their electric fields will partially cancel this will mean there will be a weak ion-water electrostatic interaction. The water molecules will therefore move away from ion. This will reduce the ionic solvation energy leading to a repulsive force. So I performed Quantum Chemical geometry optimisation calculations on a sodium chloride dimer with 10 water molecules around them. But in the end the water molecules barely budged at all from where they were around the single ions. I also did MD simulations around the dimer and saw that the peak in the RDF stayed in exactly the same position around the dimer. Although it lowered a little as you'd expect due to the other ion removing some water.
On the plus side I learned the basics of performing MD. Was kind of ridiculously easy, the only difficult thing pretty much was getting the files in the right format. And as a result I got vast amounts of information. I can understand why it is so widely used so easy and satisfying. The only problem is that I barely knew what to do with all that information. It was kind of overwhelming. I had a simple hypothesis that I was testing and it was great for that. But if I was just trying to understand the problem more generally I wouldn't know where to start. I get that feeling a lot reading some simulation papers, that they just throw all this information together then kind of give hand wavy interpretations of what's going on and then write it up. And it's not really clear at the end of it what's been learnt.
It's a little different I guess as there's no way I could write up and publish every idea I had that came to nothing, it would make reading journals a nightmare sorting through all the things people had tried. I guess we could do with some kind of online wiki article where we all describe things we tried that didn't work to solve a problem so others know not to try it as well.
I might as well describe the idea. It was that as two ions come together their electric fields will partially cancel this will mean there will be a weak ion-water electrostatic interaction. The water molecules will therefore move away from ion. This will reduce the ionic solvation energy leading to a repulsive force. So I performed Quantum Chemical geometry optimisation calculations on a sodium chloride dimer with 10 water molecules around them. But in the end the water molecules barely budged at all from where they were around the single ions. I also did MD simulations around the dimer and saw that the peak in the RDF stayed in exactly the same position around the dimer. Although it lowered a little as you'd expect due to the other ion removing some water.
On the plus side I learned the basics of performing MD. Was kind of ridiculously easy, the only difficult thing pretty much was getting the files in the right format. And as a result I got vast amounts of information. I can understand why it is so widely used so easy and satisfying. The only problem is that I barely knew what to do with all that information. It was kind of overwhelming. I had a simple hypothesis that I was testing and it was great for that. But if I was just trying to understand the problem more generally I wouldn't know where to start. I get that feeling a lot reading some simulation papers, that they just throw all this information together then kind of give hand wavy interpretations of what's going on and then write it up. And it's not really clear at the end of it what's been learnt.
Polarizable Water Models (Technical)
One frustrating thing is that I often have ideas about stuff that I can't write up as a paper because it isn't substantial enough. Because theirs no real new results or theories I have its just opinion and speculation. So I figure I could put it here, so its recorded.
So one thing that puzzles me a lot about Molecular Dynamics water models is that as far as I'm aware no one has built one that incorporates induced multipole moments higher than the dipole moment.
So one thing that puzzles me a lot about Molecular Dynamics water models is that as far as I'm aware no one has built one that incorporates induced multipole moments higher than the dipole moment.
This doesn't make sense to me, it has been shown that the electric field varies substantially over the range of a single water molecule. That means it will have significant derivatives and hence significant induced quadrupole and octupole moments. There are models that use Drude oscillators but this is not the same thing as I'm pretty sure these will be dominated by the dipole moment as the point charges are so close together.
I think this causes the common problem with water models. That the polarisability has to be artificially reduced from 1.44 cubic angstroms to about 1 as otherwise the dielectric constant of water comes out too large at around 100. (paper here) This paper argues that on average the electric field is lower around a water molecule than it is at the centre, and that that is why you should lower the polarisability. But strictly speaking the correct way to account for that is to use higher order induced multipole moments as you are taking a taylor expansion about the centre of the molecule. Additional evidence for this comes from a relatively newer water model which has Drude oscillators on the hydrogens as well as the oxygens. This model seems to able to reproduce the dielectric constant of water with the vacuum polarisability value of 1.44 angstroms. That implies that it is the variation in the electric field that explains this effect, and the correct way to account for this is with a multipole expansion. I guess it's too computationally expensive though, although it looks like a quantum drude oscillator model might work.
They also apply their model to ion solvation energies and get nice results for the salt pairs. But they don't calculate the surface potential of their model. So there is no way to know what they're getting for the real or intrinsic values. Their single ion values in periodic boundary conditions are really close to Tissandier's values, which is surprising but I guess you can't read too much into that as the quadrupole trace of their water molecule could be anything. Surely it's not too hard to calculate this, would be really useful for working out what the real and intrinsic values are, which is super important.
Saturday 30 August 2014
Justifications
I'm thinking of updating my three minute thesis talk to make roughly the point I make here.
I'm trying to work out how to justify my research. So I've heard that there are two general ways to do this. The first is by appealing to to the wonder factor: Aren't you just dying to know the answer to this question? This is how you justify the Large Hadron Collider or space expeditions. These experiments can tell us about how the universe came to be or what the origin of life is. The second option is to appeal to practical applications: We are looking for a treatment for cancer or unlimited clean energy.
These are certainly worthy topics of study, but it seems to me that there are a third class of research topics that can fall through the gap between these two justifications. They are questions that are slightly obscure and seemingly boring. But which also can not be justified by any single direct and practical application.
A nice example of this is the Schrödinger equation. Schrödinger discovered this equation while trying to discover a mathematical model that would explain the light emission of hydrogen gas. Now this sounds like an obscure, boring and not very useful topic to study. But Schrödinger's equation has turned out to be one of the most important and useful scientific equations of all time. It describes the properties of the periodic table with amazing accuracy, explaining the vast majority of the physical world with just a few symbols strung together.
It is obviously much harder to justify this kind of research, but I think it is just as, maybe even more important than the other two kinds. Wonder inducing research is certainly very nice to read about but it is very limited in terms of actual implications for our lives. Research into direct applications is fundamentally limited in usefulness as it tends to be so specific that often the results are not generalisable for any other applications.
Often the answers that are most useful are to questions about the simplest possible systems. How do we describe hydrogen, the simplest possible atom? How do we describe a ball flying through empty space. What is the link between electricity and magnetism? These questions aren't very exciting and it's not obvious what use the answers are. And yet these answers, once they were discovered, changed the world more profoundly then anyone could have imagined.
I want to work on questions like this. I think "What is the interaction of any two molecules in water?" is one of them, if we had an answer to it I think there would be massive implications for vast areas of science. But few people seem interested in this question. I had never even heard of it until I started my PhD. Popular science articles seem to focus almost exclusively on research that fits into one of the two justifications I mention above.
So I find it difficult to work out how to sell my research. On the one hand it's kind of a mundane topic, and on the other hand there's no direct applications I can point to. All I really have is a kind of long and involved argument about why answering questions about very simple physical systems is really important, which doesn't seem especially suited to the three minute thesis format.
I'm trying to work out how to justify my research. So I've heard that there are two general ways to do this. The first is by appealing to to the wonder factor: Aren't you just dying to know the answer to this question? This is how you justify the Large Hadron Collider or space expeditions. These experiments can tell us about how the universe came to be or what the origin of life is. The second option is to appeal to practical applications: We are looking for a treatment for cancer or unlimited clean energy.
These are certainly worthy topics of study, but it seems to me that there are a third class of research topics that can fall through the gap between these two justifications. They are questions that are slightly obscure and seemingly boring. But which also can not be justified by any single direct and practical application.
A nice example of this is the Schrödinger equation. Schrödinger discovered this equation while trying to discover a mathematical model that would explain the light emission of hydrogen gas. Now this sounds like an obscure, boring and not very useful topic to study. But Schrödinger's equation has turned out to be one of the most important and useful scientific equations of all time. It describes the properties of the periodic table with amazing accuracy, explaining the vast majority of the physical world with just a few symbols strung together.
It is obviously much harder to justify this kind of research, but I think it is just as, maybe even more important than the other two kinds. Wonder inducing research is certainly very nice to read about but it is very limited in terms of actual implications for our lives. Research into direct applications is fundamentally limited in usefulness as it tends to be so specific that often the results are not generalisable for any other applications.
Often the answers that are most useful are to questions about the simplest possible systems. How do we describe hydrogen, the simplest possible atom? How do we describe a ball flying through empty space. What is the link between electricity and magnetism? These questions aren't very exciting and it's not obvious what use the answers are. And yet these answers, once they were discovered, changed the world more profoundly then anyone could have imagined.
I want to work on questions like this. I think "What is the interaction of any two molecules in water?" is one of them, if we had an answer to it I think there would be massive implications for vast areas of science. But few people seem interested in this question. I had never even heard of it until I started my PhD. Popular science articles seem to focus almost exclusively on research that fits into one of the two justifications I mention above.
So I find it difficult to work out how to sell my research. On the one hand it's kind of a mundane topic, and on the other hand there's no direct applications I can point to. All I really have is a kind of long and involved argument about why answering questions about very simple physical systems is really important, which doesn't seem especially suited to the three minute thesis format.
Solvation Energy
This frustrates me a lot.
So the logical next step is to direct a vast amount of effort into understanding this problem. This lack of understanding should be like a sputnik moment. It is criminal we have no idea how the simplest possible case of something so incredibly important works.
Instead you hear things quite frequently, like there is no funding in this kind of totally fundamental research. You have to tie your grant proposals to direct applications to get funding. But the exact point of government funded research is to build this fundamental science, which doesn't have direct applications but may one day. How does the LHC get funded to discover something of no practical importance? When there are problems that are so fundamentally crucial and so neglected.
Instead you have vast amounts of money thrown at biomedical research trying to discover new drugs and understand biology when the fundamental underlying mechanisms aren't fully understood. It costs billions to develop a drug, normally by an incredibly tedious and costly experimental trial and error. This is analogous to trying to land a man on the moon, without Newton's theory of gravity, or without programs that can predict orbits. You might eventually hit the moon by trial and error or extrapolating from past patterns. But a lot of people are going to die in the process.
I would really love to continue working on this problem and ones like it. Both because of how interesting they are and because I know how important they are. And it is so disheartening to hear that I may not be able to. It would be OK however, if it were because there are too many competent and smart people who will do a better job than me. I would happily find a different career if I knew this problem were in good hands. But instead the reason I can't work on this problem is that there's not enough funding. That is so infuriating, especially considering the costs of theoretical work are so cheap. Just an academic salary and some computing power, are all you need. As well as the fact that vast amounts of money is poured down the drain doing experiments that would become redundant if we had good theoretical models of molecular interactions in water.
It's from an old post of Ashutosh Jogalekar at Curious Wavefunction. He's is talking about what we need to be able to design drugs rationally i.e., using computers. The key quote is:
The basic science is also going to involve the accurate experimental determination of solvation energies. Such measurements are typically considered too mundane and basic to be funded. And yet, as the authors make clear in the paragraph quoted at the beginning, it's only such measurements that are going to aid the calculation of aqueous solvation energies. And these calculations are going to be ultimately key to calculating drug-protein interactions. After all, if you cannot even get the solvation energy right...Once you start learning about chemical modelling you realise how incredibly important it is to be able to calculate the interaction of molecules in water. A vast amount of amazingly important chemistry happens in water. Including all of life. If you can't understand the interaction of molecules in water properly then you have no chance of understanding biology comprehensively. The second thing you realise is that we have no idea how to do this. When you look at the simplest possible cases, a sodium ion interacting with a chloride ion in water. We have no models that can satisfactorily reproduce this interaction.
So the logical next step is to direct a vast amount of effort into understanding this problem. This lack of understanding should be like a sputnik moment. It is criminal we have no idea how the simplest possible case of something so incredibly important works.
Instead you hear things quite frequently, like there is no funding in this kind of totally fundamental research. You have to tie your grant proposals to direct applications to get funding. But the exact point of government funded research is to build this fundamental science, which doesn't have direct applications but may one day. How does the LHC get funded to discover something of no practical importance? When there are problems that are so fundamentally crucial and so neglected.
Instead you have vast amounts of money thrown at biomedical research trying to discover new drugs and understand biology when the fundamental underlying mechanisms aren't fully understood. It costs billions to develop a drug, normally by an incredibly tedious and costly experimental trial and error. This is analogous to trying to land a man on the moon, without Newton's theory of gravity, or without programs that can predict orbits. You might eventually hit the moon by trial and error or extrapolating from past patterns. But a lot of people are going to die in the process.
I would really love to continue working on this problem and ones like it. Both because of how interesting they are and because I know how important they are. And it is so disheartening to hear that I may not be able to. It would be OK however, if it were because there are too many competent and smart people who will do a better job than me. I would happily find a different career if I knew this problem were in good hands. But instead the reason I can't work on this problem is that there's not enough funding. That is so infuriating, especially considering the costs of theoretical work are so cheap. Just an academic salary and some computing power, are all you need. As well as the fact that vast amounts of money is poured down the drain doing experiments that would become redundant if we had good theoretical models of molecular interactions in water.
Papers
So over the past year or so I've had some papers published in the Journal of Physical Chemistry B, on the solvation properties of ions. I thought I would explain in simple language what I've done and why I think it's important.
They can be found at here, here and here. I might discuss each paper in its own post but first I thought I'd just explain some background concepts.
The basic idea is that when you dissolve salt in water there is some energy change, normally the ions like to go into water and so energy is released partly in the form of heat. It's just like how a ball will naturally roll down a hill and release some energy in the form of motion, which then turns into heat.
Roughly speaking this is why it takes so much energy to desalinate water, it costs energy to take ions out of pure water as they like to be there.
In technical terms we say that the salt dissolving has a negative free energy change. Because energy is conserved there must be a positive energy increase somewhere else which is why we get our heat released. There have even been attempts to harness this energy, where fresh water rivers mix with the very salty sea.
The calculation of free energies is a centrally important problem in Physical Chemistry. It is normally referred to with the letter G. Ashutosh Jogalekar at Curious Wavefunction does a great job of explaining in a more general context what is is and why it's important.
Here's a nice video of the process of water dissolving salt. The energy change between the start and end of that process is what I'm trying to calculate.
They can be found at here, here and here. I might discuss each paper in its own post but first I thought I'd just explain some background concepts.
The basic idea is that when you dissolve salt in water there is some energy change, normally the ions like to go into water and so energy is released partly in the form of heat. It's just like how a ball will naturally roll down a hill and release some energy in the form of motion, which then turns into heat.
Roughly speaking this is why it takes so much energy to desalinate water, it costs energy to take ions out of pure water as they like to be there.
In technical terms we say that the salt dissolving has a negative free energy change. Because energy is conserved there must be a positive energy increase somewhere else which is why we get our heat released. There have even been attempts to harness this energy, where fresh water rivers mix with the very salty sea.
The calculation of free energies is a centrally important problem in Physical Chemistry. It is normally referred to with the letter G. Ashutosh Jogalekar at Curious Wavefunction does a great job of explaining in a more general context what is is and why it's important.
Here's a nice video of the process of water dissolving salt. The energy change between the start and end of that process is what I'm trying to calculate.
Subscribe to:
Posts (Atom)