The job I need to do right now is to implement Radia kick maps in AT. Its something that's been done before in other codes, and probably even in AT before. I have a supervisor who knows enough about it to help me with it, and is interested that I do it. So why do I go so slowly?
One of the problems is that I see this as an obscure topic that only a few people understand. Further its already been done before, so I am not breaking new ground. So its hard to gather motivation, since there will be few people who can benefit from this and some people who know that its not even new.
What would be valuable about it, is if it allows the code to be easier to use in the process and connects the topic both conceptually and software-wise to a larger community. So, to get myself to do it, I sit in a place which is more well-known and approach the problem with pedagogic bent. Initially, I came to the problem from the beam dynamics side. From here, one is naturally led to the questions of dynamic aperture and lifetime. But the kick maps come from insertion devices which are there for the purpose of creating the x-rays. So now I learn the Radia code, and try to sit on that side. Somehow in the piecing together of these two perspectives, I hope to make it a little more useful or at least understandable, and hopefully not take the rest of my life to do it. (really its not so hard, and perhaps one could say I'm just lazy and procrastinating writing such posts as this)
Monday, November 14, 2011
Thursday, November 03, 2011
physics of photons and needs of SR experiments
On the topic of #3 from my last post, I try to learn a bit of how to understand photons. I posted a question on Stack Exchange here. I referenced a paper by KJ Kim on the Wigner function approach to synchrotron radiation, and asked the question:
I got some links to papers about Wigner functions and about photons. I think that quantum optics has a lot to say on this topic, but unfortunately I seem to have lost my copy of Mandel and Wolf. I was referred to papers on the wave function of a photon, such as this by Iwo Bialynicki-Birula. The wave function defined by Bialynicki-Birula is referenced here in a 2011 New Journal of Physics article by Saldanha and Monken. In this article, they extend the photon wave function approach to "to include the interaction of photons with non-absorptive continuous media."
I was interested in this question because I wanted to understand undulator radiation better. The question is what one should calculate? How should one represent the radiation such that it covers the properties used by the experiments with synchrotron radiation? In "Undulators, Wigglers and Their Applications", edited by Elleaume and Onuki, the brilliance is identified with the Wigner function and is computed for undulators, wigglers and bending magnets. Its a dense book, but contains an up-do-date perspective on these topics. Regarding codes, the synchrotron radiation may be computed in the near field with SRW. But this just computes different components of the Electric field for a given electron beam source and undulator construction. What to do with the output, and are there still open foundational questions?
Once one knows the radiation fields, one can propagate them, and probably ray optics is sufficient for most purposes. The Shadow code may be used for this purpose, and allows one to enter lenses, mirrors and other optical components in the x-ray beamline.
So, this is sort of the the landscape, as far as books, theoretical frameworks, and software for synchrotron radiation. Certainly I'm biased to that used at my institute, but I think it covers a good amount.
For some practical questions, I think one can look at how some different electron beam parameters affect the photon beam. For example, the electron beam energy spread, or the tilt angle of the electron beam. Next, the question is, for the experiments one does with the radiation, what are really the important parameters. Brightness is important, but it seems to be a stand-in for a more detailed case by case examination. Does one need large numbers of photons (flux?), does one want a round beam? Are the coherence properties of the x-rays important? In the latter case, it appears that the Wigner function does not tell all, but in faction one needs to compute the mutual intensity (the argument of the integral in the definition of the Wigner function).
When one represents radiation via a Wigner function, is this really quantum mechanics?
I got some links to papers about Wigner functions and about photons. I think that quantum optics has a lot to say on this topic, but unfortunately I seem to have lost my copy of Mandel and Wolf. I was referred to papers on the wave function of a photon, such as this by Iwo Bialynicki-Birula. The wave function defined by Bialynicki-Birula is referenced here in a 2011 New Journal of Physics article by Saldanha and Monken. In this article, they extend the photon wave function approach to "to include the interaction of photons with non-absorptive continuous media."
I was interested in this question because I wanted to understand undulator radiation better. The question is what one should calculate? How should one represent the radiation such that it covers the properties used by the experiments with synchrotron radiation? In "Undulators, Wigglers and Their Applications", edited by Elleaume and Onuki, the brilliance is identified with the Wigner function and is computed for undulators, wigglers and bending magnets. Its a dense book, but contains an up-do-date perspective on these topics. Regarding codes, the synchrotron radiation may be computed in the near field with SRW. But this just computes different components of the Electric field for a given electron beam source and undulator construction. What to do with the output, and are there still open foundational questions?
Once one knows the radiation fields, one can propagate them, and probably ray optics is sufficient for most purposes. The Shadow code may be used for this purpose, and allows one to enter lenses, mirrors and other optical components in the x-ray beamline.
So, this is sort of the the landscape, as far as books, theoretical frameworks, and software for synchrotron radiation. Certainly I'm biased to that used at my institute, but I think it covers a good amount.
For some practical questions, I think one can look at how some different electron beam parameters affect the photon beam. For example, the electron beam energy spread, or the tilt angle of the electron beam. Next, the question is, for the experiments one does with the radiation, what are really the important parameters. Brightness is important, but it seems to be a stand-in for a more detailed case by case examination. Does one need large numbers of photons (flux?), does one want a round beam? Are the coherence properties of the x-rays important? In the latter case, it appears that the Wigner function does not tell all, but in faction one needs to compute the mutual intensity (the argument of the integral in the definition of the Wigner function).
Tuesday, October 25, 2011
research topics/work to do.
I've been continuing with this effort of finding links between beam dynamics and other areas. I met with some theorists to try to explain what are some of the problems here, and see if I could find stuff in common. In general, I've been trying to find other areas of physics that could have things in common with beam dynamics and synchrotron radiation physics. But in order to do this, I need to identify what areas of beam dynamics are really research topics that could use developing. So here is my attempt to point out what I think are still somewhat hard, but maybe tractable problems to be worked on.
On the software end, there's work to be done on the electron side and on the code side. On the electron side, we have many codes. Personally I think AT is good to develop, since its in Matlab and easier to add to. Elegant is also nice, though open and extendable in the same way. On the radiation side, we have SRW to compute the radiation in the near field, right out of the bending magnets or undulators. Then there's Shadow which tracks the radiation further down the beam line. There may be some work underway to combine these together in some way.
For myself, I just like to see this big picture. Of course I can only work on a very small amount of this. But I'd at least like to work on something within a bigger picture that makes sense to me.
- Single particle dynamics :Given a one turn map- find the long behavior. What is the stable region ? What happens when damping is added ? Can we find the right parameters from the one-turn map that effect the long term behavior ? Has this problem been solved in chaotic map literature?
- Collective effects: Distribution with diffusion, damping, feedback. Instability due to one effect may be limited by other effects. Is there a theory in this range ?
Want to know emittance growth, non-Gaussain distributions.
Analysis of Fokker-Planck. Maybe more terms as in Master equation is needed sometimes? - Radiation from electron beam:Interface between electron beam and photons. Definition of Wigner function. Sometimes negative. Not fully understood.
Need to define photon beam operationally based on how the photons will interact, be focussed by optics, etc.
Radiation from spin-polarized electrons?
On the software end, there's work to be done on the electron side and on the code side. On the electron side, we have many codes. Personally I think AT is good to develop, since its in Matlab and easier to add to. Elegant is also nice, though open and extendable in the same way. On the radiation side, we have SRW to compute the radiation in the near field, right out of the bending magnets or undulators. Then there's Shadow which tracks the radiation further down the beam line. There may be some work underway to combine these together in some way.
For myself, I just like to see this big picture. Of course I can only work on a very small amount of this. But I'd at least like to work on something within a bigger picture that makes sense to me.
Tuesday, October 18, 2011
electron properties (AT) and x-ray properties
Well, the talk went pretty well. People told me they understood stuff and hadn't heard a comprehensive talk on this topic before. I have a few new contacts.
Now, I need to figure out what to do next. Basically, I see two areas. One is to start to really learn about radiation in undulators and keep in mind the question of whether the details of the electron distribution may yield something potentially interesting. The second is to be able to use AT better and make it better. Perhaps try to encourage and help write a non-linear dynamics optimization package? But this is still the ugly topic. Maybe this can be a foundation to do it better? We might as well at least implement some of the stuff that is known to work. And probably people already have. There's also the kick maps for tracking. Do I really recapitulate this history and keep going with the power series maps?
Hard to get motivated with this again, but I feel like its necessary (at least once more? probably multiple times...) When this problem gets too small, I completely lose interest. Need to remember some of the big questions, and think about clean implementations.
Overview:
1)Radiation brightness and beam size, 2) physics of one electron, 3) electron beam emittance, size and shape, 4)beam lifetime
The other issue here is the return to "work". I have to get back to concrete work with respect to understanding how the machine works, where the files are, and how to model it using AT. Expansion of networks and joining with other areas is important, but there has to be something there to join with. What is useful to model and understand? Can one be in some way comprehensive and systematic, or is this hopeless?
Now, I need to figure out what to do next. Basically, I see two areas. One is to start to really learn about radiation in undulators and keep in mind the question of whether the details of the electron distribution may yield something potentially interesting. The second is to be able to use AT better and make it better. Perhaps try to encourage and help write a non-linear dynamics optimization package? But this is still the ugly topic. Maybe this can be a foundation to do it better? We might as well at least implement some of the stuff that is known to work. And probably people already have. There's also the kick maps for tracking. Do I really recapitulate this history and keep going with the power series maps?
Hard to get motivated with this again, but I feel like its necessary (at least once more? probably multiple times...) When this problem gets too small, I completely lose interest. Need to remember some of the big questions, and think about clean implementations.
Overview:
1)Radiation brightness and beam size, 2) physics of one electron, 3) electron beam emittance, size and shape, 4)beam lifetime
The other issue here is the return to "work". I have to get back to concrete work with respect to understanding how the machine works, where the files are, and how to model it using AT. Expansion of networks and joining with other areas is important, but there has to be something there to join with. What is useful to model and understand? Can one be in some way comprehensive and systematic, or is this hopeless?
Monday, September 26, 2011
talk
I'm preparing a talk for this Friday. Its been a long time coming, but somehow my preparation is not where it should be. Its ok.
I will talk about the physics of electron beams in a synchrotron. The goal is to explain the origin of the equilibrium beam sizes and the beam lifetime. What I'd like to say is that these are two basic things everyone should understand if you want to know where your photon beam comes from. From a web perspective, I think these topics are not well covered. Here is beam emittance on wikipedia. The fact that an equilibrium emittance exists for an electron ring, independent of the initial distribution isn't mentioned. The perspective is entirely from non-radiating (or not much) hadron storage rings.
The first question is how do you store an electron. Basically, you create a 6-D harmonic oscillator. Transversely this comes from quadrupole magnets. Longitudinally, this comes from an RF cavity. Using quadrupoles to create a stable quadradatic potential was not discovered immediately. It is referred to as the concept of Strong Focusing, and was found by Courant and Snyder and by Cristofolis independently. The problem is that a quadrupolar magnet field focuses in one direction, but is defocusing in the perpendicular direction. However, by creating a system of several quadrupoles of alternating polarity, one can create a net focusing effect.
So, we have a stable bucket both transversely and longitudinally. Now what? Throw an electron in there! Next, we observe that the energy loss through radiation depends on the energy in such a way that a higher energy electron loses more, and a lower energy electron loses less. Thus we have a damping effect where the electron will head towards a fixed value. This damping effect also comes into the transverse dimensions, and thus overall a bunch of electrons wants to spiral down to a single point in phase space.
What stops the spiraling process? What comes into play to limit the beam size. First, one may imagine it is the coulomb repulsion. However this turns out to be extremely small for a relativistic beam. Basically, the electric repulsion is cancelled (reduced by gamma squared) by the magnetic attraction.
The effect that does set the beam size is the quantum nature of the radiation. The radiation is emitted in photons, and indeed a rather small number of them. This causes a randomness in the energy change that results in a diffusion process. This diffusion, together with the radiation damping effect mentioned cause the equilibrium beam sizes.
The interaction between the electrons is important however. The typical interactions are quite small, but the less frequent short range scattering causes large energy changes which may result in an electron being lost. This is the source of the beam lifetime (actually there's also a part from scattering off the gas in the chamber.).
That's the basic story I want to get across.
I will talk about the physics of electron beams in a synchrotron. The goal is to explain the origin of the equilibrium beam sizes and the beam lifetime. What I'd like to say is that these are two basic things everyone should understand if you want to know where your photon beam comes from. From a web perspective, I think these topics are not well covered. Here is beam emittance on wikipedia. The fact that an equilibrium emittance exists for an electron ring, independent of the initial distribution isn't mentioned. The perspective is entirely from non-radiating (or not much) hadron storage rings.
The first question is how do you store an electron. Basically, you create a 6-D harmonic oscillator. Transversely this comes from quadrupole magnets. Longitudinally, this comes from an RF cavity. Using quadrupoles to create a stable quadradatic potential was not discovered immediately. It is referred to as the concept of Strong Focusing, and was found by Courant and Snyder and by Cristofolis independently. The problem is that a quadrupolar magnet field focuses in one direction, but is defocusing in the perpendicular direction. However, by creating a system of several quadrupoles of alternating polarity, one can create a net focusing effect.
So, we have a stable bucket both transversely and longitudinally. Now what? Throw an electron in there! Next, we observe that the energy loss through radiation depends on the energy in such a way that a higher energy electron loses more, and a lower energy electron loses less. Thus we have a damping effect where the electron will head towards a fixed value. This damping effect also comes into the transverse dimensions, and thus overall a bunch of electrons wants to spiral down to a single point in phase space.
What stops the spiraling process? What comes into play to limit the beam size. First, one may imagine it is the coulomb repulsion. However this turns out to be extremely small for a relativistic beam. Basically, the electric repulsion is cancelled (reduced by gamma squared) by the magnetic attraction.
The effect that does set the beam size is the quantum nature of the radiation. The radiation is emitted in photons, and indeed a rather small number of them. This causes a randomness in the energy change that results in a diffusion process. This diffusion, together with the radiation damping effect mentioned cause the equilibrium beam sizes.
The interaction between the electrons is important however. The typical interactions are quite small, but the less frequent short range scattering causes large energy changes which may result in an electron being lost. This is the source of the beam lifetime (actually there's also a part from scattering off the gas in the chamber.).
That's the basic story I want to get across.
Monday, August 29, 2011
Becoming comfortable with
Both of my parents have moved to remote places, with unique communities, and are rather isolated in some ways. Had I grown up in either of these places, I think I would have wanted to escape in some way. I might have longed for something different and looked at how to get out, geographically, socially, and regarding lifestyle and mentality.
At this point in my life, I am only a visitor in these places. But arriving in each one, I get the feeling of being pulled into a black hole, with little communication to the rest of my life, and not sure how to connect the experience to who I am. At the same time, I am adaptable, and arriving, and within the experience, it has an integrity and a quality to it that is quite nice. But the boundaries are difficult. I can't think my way into it from outside, or out of it from inside.
So I have slowly worked on the problem over the years by looking carefully at the boundaries. In the remote town in California where my father lives, for example, I look to see whether there are surrounding communities that might have some life to them. I try to find things in common to other places regarding environment and landscape. I plan trips there with an exit strategy, and friends and other family flanking it. Certainly this is also quite personal, and relates to my own experience of family and who I am there, and who I am seen as. I am starting to try the same strategy with my mother's house in Iowa. There, the boundaries are physical, but there is also a strong ideological barrier that is uncomfortable to me. Is there something within that I can relate to? I find pieces of interest to that community that I might interpret in a different way, but still find interesting.
Is such an elaborate process necessary? Maybe I will reach a point where it will seem smaller and less important, but somehow this work is necessary. The other option is to say that visiting my family is too difficult, and no common ground can be found, but I don't want to do that.
On the Petrolia side, there is the natural environment. The trees, the river, the ocean.
On the MUM/Fairfield side, there is the nearby Mississippi river. There are coffee shops in Fairfield. Ideologically, MUM is more challenging. The Maharishi is a figure that I just have a very hard time appreciating. And the closed mentality fosters an inside/outside split that is hard to overcome. One of the Maharishi's main texts he interpretted and based his power around has been the Baghavad Gita. I think this is something I could become interested in.
At this point in my life, I am only a visitor in these places. But arriving in each one, I get the feeling of being pulled into a black hole, with little communication to the rest of my life, and not sure how to connect the experience to who I am. At the same time, I am adaptable, and arriving, and within the experience, it has an integrity and a quality to it that is quite nice. But the boundaries are difficult. I can't think my way into it from outside, or out of it from inside.
So I have slowly worked on the problem over the years by looking carefully at the boundaries. In the remote town in California where my father lives, for example, I look to see whether there are surrounding communities that might have some life to them. I try to find things in common to other places regarding environment and landscape. I plan trips there with an exit strategy, and friends and other family flanking it. Certainly this is also quite personal, and relates to my own experience of family and who I am there, and who I am seen as. I am starting to try the same strategy with my mother's house in Iowa. There, the boundaries are physical, but there is also a strong ideological barrier that is uncomfortable to me. Is there something within that I can relate to? I find pieces of interest to that community that I might interpret in a different way, but still find interesting.
Is such an elaborate process necessary? Maybe I will reach a point where it will seem smaller and less important, but somehow this work is necessary. The other option is to say that visiting my family is too difficult, and no common ground can be found, but I don't want to do that.
On the Petrolia side, there is the natural environment. The trees, the river, the ocean.
On the MUM/Fairfield side, there is the nearby Mississippi river. There are coffee shops in Fairfield. Ideologically, MUM is more challenging. The Maharishi is a figure that I just have a very hard time appreciating. And the closed mentality fosters an inside/outside split that is hard to overcome. One of the Maharishi's main texts he interpretted and based his power around has been the Baghavad Gita. I think this is something I could become interested in.
Sunday, August 28, 2011
Back to the mess
I have to finish a paper for a conference by Wednesday night. It is on the measurements relating to Touschek lifetime and momentum acceptance. There's an analysis of data to be done, and general writing and preparation of the paper. I really don't want to work on it.
Why don't I want to do it? Its late, I know. A paper needs to be written, and doing an analysis at the last minute can derail me from the process of fixing figures, adding references, and putting a clear narrative together. But its also that I've gotten myself out of thinking of this topic, and am wary of going back. It was my compromise. I will not leave accelerator physics entirely. I will do some work in this field, but move outward at the same time. But the topic is a mess for me. It is a personal mess in that my own files and documents and the relevant equations are not in such clear order. And a general historical mess in that it relates to the topic of dynamic aperture and sextupole optimization which is an unsolved problem. That question of dynamic aperture and stability has been the piece that I have slowly worked on, and tried to lay out a personal groundwork, so I don't feel so lost working in that area. Maybe this is a reflection of the fact that I didn't really finish this process.
So, my own angle on Touschek lifetime and measurement that I would like to get across is that the measurements are a diagnostic for the various lattice optimizations. There is both vertical emittance reduction goals, and increase of momentum acceptance via sextupole optimization. Stating clearly what these mean, and having measurements to ground discussion and results puts this other more nebulous "accelerator physics" activity onto a ground that relates to the goal of the machine- production of stable, long lived synchrotron radiation.
Why don't I want to do it? Its late, I know. A paper needs to be written, and doing an analysis at the last minute can derail me from the process of fixing figures, adding references, and putting a clear narrative together. But its also that I've gotten myself out of thinking of this topic, and am wary of going back. It was my compromise. I will not leave accelerator physics entirely. I will do some work in this field, but move outward at the same time. But the topic is a mess for me. It is a personal mess in that my own files and documents and the relevant equations are not in such clear order. And a general historical mess in that it relates to the topic of dynamic aperture and sextupole optimization which is an unsolved problem. That question of dynamic aperture and stability has been the piece that I have slowly worked on, and tried to lay out a personal groundwork, so I don't feel so lost working in that area. Maybe this is a reflection of the fact that I didn't really finish this process.
So, my own angle on Touschek lifetime and measurement that I would like to get across is that the measurements are a diagnostic for the various lattice optimizations. There is both vertical emittance reduction goals, and increase of momentum acceptance via sextupole optimization. Stating clearly what these mean, and having measurements to ground discussion and results puts this other more nebulous "accelerator physics" activity onto a ground that relates to the goal of the machine- production of stable, long lived synchrotron radiation.
Subscribe to:
Posts (Atom)
