http://lcls.slac.stanford.edu Blog posts from the people behind LCLS. en-us Copyright 2009 SLAC National Accelerator Laboratory http://lcls.slac.stanford.edu/images/lclsLogo.gif http://lcls.slac.stanford.edu http://lcls.slac.stanford.edu/Article.aspx?article_id=100
We're also focusing on the designs for the experimental hutches for the Far Experimental Hall with a completion date expected in fall of this year. Plans are underway to hold a workshop to focus attention on the sixth hutch science program: materials under extreme conditions. The dates will be set shortly, with planning focusing on end of March and beginning of April. Look for the announcement on the LCLS web site.

The other area of attention is the data acquisition (DAQ), storage and on-line/off-line analysis. The initial LCLS operations will be AMO and the DAQ is progressing well, but we need to be ready for the users to have on-line data visualization to insure success for our early science.

All in all, it's starting to feel like drinking from a fire hose and I don't expect it to let up any time soon.
]]> Sun, 08 Feb 2009 01:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=100 http://lcls.slac.stanford.edu/Article.aspx?article_id=95
The betting pool is decided: Two shots. The winnner: Paul Emma, the man on the mouse. (Emma heads up the group responsible for commissioning the electron portion of the LCLS.) Two clicks. The control room erupted in applause.

"I knew these guys were good, but... [laughs]," said John Galayda (director of LCLS construction). "This is story book." "The man knows his machine," said Jim Turner (project leader for today's shift in the control room).

The first try made it only a quarter of the way down, much head shaking ensued. But with the second click, score. Truly an amazing feat. Very small needle, and the team managed to thread it in two tries. ]]> Sun, 14 Dec 2008 00:40:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=95 http://lcls.slac.stanford.edu/Article.aspx?article_id=94
Once the go ahead was given, operators turned the beam on to pulse once a second. That's when the steering began. A graph showing the beam position monitors populated with a progression of spikes down the line as tweaks in steering magnets honed the beam's focus toward its target. After nearly a five-hour delay, getting the beam to the TD-UND took all of 10 minutes.

Now the radiation physics group will make measurements of the facility by hand, which will take a few hours. It's possible that the beam will continue on and enter the undulator hall tonight.

Erickson said that while the operators were wrangling the beam he counted 29 heads surrounding the console. It only took minutes for operators to get control of the beam, but it was an intense few minutes. Reminded me of scenes from the bridge of a starship in a science fiction movie, as if instead of electrons being guided through magnetic fields, it was us threading tensely through an asteroid field.]]> Sun, 14 Dec 2008 00:10:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=94 http://lcls.slac.stanford.edu/Article.aspx?article_id=93
Also: turns out the betting pool relates specifically to getting the beam through the undulator hall. Since they haven't even gotten to the beginning of it yet, the first shot doesn't count. No wonder there's such uncertainty: the vacuum chamber in the UH is barely bigger than a drinking straw, and it's three football fields long. Alignment of the beam is no small feat.]]> Sat, 13 Dec 2008 22:50:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=93 http://lcls.slac.stanford.edu/Article.aspx?article_id=91
There's a pool running among the physicists on how many shots it will take to reach the beam dump. The man with the mouse button, Paul Emma, bet 2 shots; the others range from 1 to a 100. Daniel Ratner bet 9 x Pi, which I think might capture the spirit best, considering the number of unknowns involved. Not sure whether the first shot counts, considering the shot didn't make out of the sections already commissioned. Good news for Paul, who basically gets a do-over now, although 2 shots seems more than a little optimistic.]]> Sat, 13 Dec 2008 18:18:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=91 http://lcls.slac.stanford.edu/Article.aspx?article_id=90
9:10 am At the moment a few members of the radiation physics department is conducting a sweep of the facility to ensure no unauthorized persons remain in the hall. Seems to be taking some time… but it's a big place. Part of today's tests include intentionally mis-steering the beam at low energy to test the beam-loss monitoring system. This safety checkout will create an important benchmark for understanding what higher energy beams would do. But it also intensifies and already stringent set of safety guidelines.

Security vigilance near the exclusion zones is about as serious as it gets. Any unauthorized person within 20 feet of the beam transport hall would shut down the whole project up to the level of the Department of Energy.

9:15 am The authorization has been signed, and the head of accelerator systems has officially unlocked the circuit (and of course, the key is on bright red chain) to enable withdrawal of the beam stoppers. As soon as the security sweep is complete, the button gets pushed. Could be any minute -- word is the last step is putting security tape on some stairs near the research yard.]]> Sat, 13 Dec 2008 16:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=90 http://lcls.slac.stanford.edu/Article.aspx?article_id=83
The heater laser beam actually comes from part of the UV drive laser system of the injector gun. A small piece of the IR beam that is not converted in UV (like a waste beam) is re-used for the heater. But as with the UV drive laser, it starts upstairs in the laser bay and needs to be transported 30 feet down into the injector vault. The laser group, and in particular Sasha Gilevich and Alan Miahnahri, were in charge during this fall downtime to bring the laser heater beam parallel to the heater chicane where there is a very special undulator. As you can see on the picture, the heater chicane is at the very end of the injector vault. (Not a very comfortable place!) The ceiling is low; it is fairly crowded with pipes, waveguides, cable trays, etc. But even in this "cave" type conditions Alan and Sasha were able to get the beam there safely.

The other challenge was to have the laser beam focused close to the middle of the undulator chamber and perfectly parallel to the chamber axis (defined by the heater undulator chicane). The beam has to travel almost 30 meters from the output of the laser to the heater undulator chamber. There, its size has to be less than 200 micrometers and it has to be perfectly aligned. If you touch just a tiny bit one actuator of a mirror upstairs in the laser bay, because of the lever arm, the beam is gone! That's why we have a beam steering stabilization feedback loop. It uses CCD cameras that watch the beam at some strategic places and the loop keeps it always in the same spot by controlling mirror actuators. Hopefully the pointing stability of the laser is good enough (around 20 micrometers) so that the loop is only running to compensate for some slow drifts of the beam direction due to temperature changes for example. So as long as nobody bumps a mirror, the laser heater beam should be always there to help the LCLS electron beam to run at its highest performance.
]]> Mon, 17 Nov 2008 19:10:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=83 http://lcls.slac.stanford.edu/Article.aspx?article_id=82
Well, over here at LCLS it’s more like "If you build it, they will come." "It" in our case is not a baseball field but rather an X-ray laser and "they" are the scientists that will use the LCLS to make profound discoveries. However, we don’t have to make a blind leap of faith like Mr. Kinsella did in building his baseball field in the middle of nowhere (no offense to anyone from Iowa). We are building "it" and "they" are certain to come. I can promise you that because LCLS most certainly would not be funded if we didn't have scientists beating on our doors waiting for the LCLS to turn on. Our only regret may be that we don't have enough available beam time to satisfy everyone.

Anyway, the key to the quote in either case is that you have to finish building "it." You can't play baseball on or partially built field, and you certainly can't do experiments with a partially built laser. I came to this realization, which when you think about it it's not much of a profound revelation, at the joint SSRL/LCLS Users' meeting held a few weeks ago. For those of you unfamiliar with such gatherings, users' meetings are annual events where management and staff of scientific user facilities present various topics (facility statistics, funding outlook, future upgrades and enhancements) to the user community. Additionally, there are sessions where users present their scientific research that they've accomplished using the facility. I was a bit disappointed at the overall attendance of this year's meeting, but I expect this will change for next year's meeting once the LCLS is operating. I'll let you know if my prediction is correct in a year!
]]> Mon, 17 Nov 2008 18:30:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=82 http://lcls.slac.stanford.edu/Article.aspx?article_id=81
The ESRF is looking to upgrade its facility through new optics, detectors and improvements to the accelerator, but when taken in light of the LCLS turn on next summer, the excitement just doesn't feel the same. Recent commissioning results for the LCLS electron beam has boosted anticipation as the LCLS team now has the baseline beam quality that would produce gigawatt levels of 1.5 Å X-rays from an ideal undulator. And these pulses are not just one here one there, but over days.

At the same time the accelerator team is investigating new operating points. Most recently, with low charge (20 picocoulombs) they have produced electron bunches that would produce 2 femtosecond pulses at 1.5 Å with ~1 x 10¹¹ photons. So every time we focus on capabilities and plans for first X-ray experiments our horizons are changed and experimental challenges increased: timing an external pump laser to an X-ray pulse of these parameters now must be on the femtosecond time scale, and the optical pump laser would ideally be, at most, a few cycles. We could start to dream of looking now at both atom and electron dynamics. These ideas further challenge us as we are getting close to procuring transform-limited pulses at 1.5 nm and with X-ray optics extending to 0.15 nm. My only problem is keeping pace with the potential and keep focused on the things we have proposed up to now.

All this brings into focus the need to develop the X-ray optical tool kit to manipulate the pulses, to dream of coherent control in the hard X-ray region, and to conceive of experiments that would utilize a transform-limited pulse at 1.5 Å. This also brings me to my fear; we will not be able to keep pace with the accelerator developments as we start our experimental program and stay focused on getting results rather than constantly looking to the future. I can’t wait for the chance to see the first X-rays next summer and start the adventure of a the revolutionary science LCLS will bring.
]]> Mon, 03 Nov 2008 17:30:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=81 http://lcls.slac.stanford.edu/Article.aspx?article_id=80
I blogged about beamline 13 yesterday, with a link to a more detailed article on it's instrumentation. Basically it's now the dedicated soft X-ray beamline at SSRL, with one instrument (13.3) serving as a test bed for a technique to be used at the LCLS—high resolution imaging of microscopic samples, using coherent X-rays.

Part of the problem with using X-rays for imaging is focusing. Traditional lenses don't work because X-rays are so penetrating they go straight through most materials and can't be made to bend easily. (Although special Fresnel lenses called "zone plates" work for certain applications.) Andreas Scherz and colleagues are now working on "lensless" imaging techniques that have a lot in common with holography.

Henry Chapman (DESY, Hamburg) gave a talk on imaging biomolecules that I found most interesting. A major concern with imaging biomolecules (such as proteins or viruses) is getting them into a beam of X-rays. The preferred route is to stream them one by one (if possible) in a vacuum, so that there is nothing extra for the X-rays to interact with and cause noise in the image.

The cool part to me is the technology behind making jets of biomolecules. Apparently this is not something that's been done to much extent, and an entirely new technology is developing around how to inject objects of that tiny size in a regular stream. Even more difficult to imagine, a few groups are looking at how to make jets of *living* cells… in a recent conversation with Janos Hajdu, he told me about researchers who successfully fired a stream of living zebra fish embryos at incredible velocities, in the hundreds of meters per second, and then caught them, all without damaging them. The embryos went on to hatch normally.

Advances in sample delivery technologies is good news for biologists, clearing the way for studies of living structures at super high resolution. The LCLS beam, however, will vaporize any sample instantaneously. The technical hurdle of capturing enough information about a sample in that sliver of time before it vaporizes has been solved for longer wavelengths of X-rays, which bodes well for honing the technique using the hard X-rays of the LCLS.

(Even if we do procure a fish-safe fish gun, I don't think SLAC should worry about building a hatchery.)
]]> Sat, 18 Oct 2008 21:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=80 http://lcls.slac.stanford.edu/Article.aspx?article_id=79
I spent more time on the SSRL side of the party today, with special attention to the afternoon sessions describing the suite of new beamlines. Several are being built, with beamline 13 cranking into full swing, 4 taking shape, and 14 still in the planning stages. I wrote in detail about 13 here.

It's surely obvious to the average scientist, but I am always struck by the ingenuity of some of these experimental devices, each intended to bring X-rays onto some specially prepared sample and then to measure what comes out. Simple in principle, but often mindbendingly complex in practice. And more to the point, the "what comes out" part is what makes lightsource science such a Swiss-army-knife of a tool.

The "scanning transmission x-ray microscope" (STXM) instrument at 13.1, for example, sends in X-rays, and gets out X-ray photons, electrons, and fluoresced photons. All three of those can be collected and characterized. Plus, the incoming X-ray beam is focused into a tiny spot just a few microns across, and can be scanned around a sample. This adds up to a super high resolution map of the chemistry or molecular structure of materials.

While the LCLS is certain to break new scientific ground, it is but a single avenue in the onward march of lightsource science. As a high "peak brightness" machine, the LCLS can throw a hugely powerful pulse of X-rays onto a sample in an unimaginably short amount of time. But, as high "average brightness" machines, synchrotron lightsources provide a (nearly) continuous stream of very bright X-rays that can be scanned over a sample, or made to gently "tickle" a sample without damaging it.

SLAC and SSRL are eager to secure their scientific future along both fronts, with early stage plans in the works to build another, larger "ultimate" synchroton at the lab. Free electron lasers like the LCLS will surely proliferate in the coming years. But in terms of number of users and experiments, the future of lightsource science, because of the huge variety of experiments available using a synchrotron, will always lie much more heavily on the synchrotron side of things.*


*The four operational lightsources under DOE collectively had 8,500 users in 2007.


]]> Sat, 18 Oct 2008 01:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=79 http://lcls.slac.stanford.edu/Article.aspx?article_id=75
However, as Kung went on to point out, the sun is indeed still shining. Lightsource science, she said, stands at a revolutionary dividing line, similar in scope to the discovery of electricity. Revolutionary science is required to bring about real change in how we think about energy, and the existing lightsource facilities under BES (four of the six in the US) are playing a major role in that revolution. In the 20th century, she reminded us, we learned how to observe matter on the atomic scale. Today, we're learning how to control it.

SLAC director Persis Drell took an even longer view in the meeting's keynote, delivered this afternoon to a crowd of about 150. These are exciting times, she said, in the realm of atomic and molecular physics, and the LCLS will certainly open new doors in the control of matter on the scale of the ultra fast and ultra small. Among the myriad factors involved, success will depend in part on how the culture of collaboration within the lightsource community evolves. Drell said the particle physics community learned a similar lesson long ago, that the best results came when accelerator physicists and experimentalists worked closely together.

That kind of association is not the traditional model for the lightsource community. But that's likely to change, I learned in a subsequent conversation with Keith Hodgson (SLAC Associate Lab Director for Photon Science). He said making the LCLS beam work is as much an art as science for accelerator physicists, and users will surely come to depend on that connection.

SSRL Director Jo Stohr made an interesting point earlier in the day that bodes well for the evolution of such a relationship. SLAC is actively developing plans for a new office complex intended to centralize SLAC's accelerator physicists under one roof, combining the varied talents drawn from both SSRL and LCLS, as well as from the long legacy of PEP. Getting those physicists talking and mingling and sitting together, he said, will surely pay dividends.

]]> Fri, 17 Oct 2008 01:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=75 http://lcls.slac.stanford.edu/Article.aspx?article_id=74
The pace of managing the electron beam down the accelerator is outstripping many of the early estimates, according to accelerator physicist Paul Emma. One of the biggest technical hurdles has been figuring out how to stabilize the drive laser so that every pulse is near to identical, 24 hours, 7 days a week. (Lasers, apparently, have a reputation for stability issues.) But, problem solved—the timing jitter issue has been conquered by a factor of 3 over previous best estimates, and stability hovers at 99%.

As for the scientific program and the nascent user organization, a call for proposals earlier this year resulted in 28 full experimental proposals involving 219 scientists. Nineteen of those proposals were for the Atomic, Molecular and Optical (AMO) instrument, which will be the first instrument online in 2009. The other instruments will come online over the following three years.

Still, even with proposals rolling in and interest and momentum building, much work remains in sorting through the who's and when's of the science program startup. The priority at first, of course, will be high-impact experiments that are less technically demanding. Fitting for a machine with such promise, but about which so little is yet known. However, according to Jochen Schneider, who leads the experimental program, the LCLS science calendar already looks rigorous, with 500 hours of experimental time scheduled by the end of next year. And when the ball really gets rolling in 2010, that number jumps to 4,000.

Hopefully the palpable optimism in the meeting hall today will carry through. The first instruments haven't even been built yet, but even that is changing rapidly. John Bozek, scientist in charge of the AMO instrument, confirmed that the first pieces of the instrument are now on order. Assembly looks to begin early next year, and will be ready for the first science experiments in less than a year.
]]> Thu, 16 Oct 2008 00:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=74 http://lcls.slac.stanford.edu/Article.aspx?article_id=73
This is the second year for LCLS and SSRL to hold their users' meeting jointly. SSRL of course has been at this for a number of years, but the LCLS is just getting off the ground and thus has not yet cultivated a very large user base. The exciting part for me is how scientists I talk to regard the LCLS collaboration as something completely new, a hybrid of sorts with regard to traditional modes of scientific collaboration.

The traditional lightsource community is vast, and for the most part user groups operate quite separately from the light source facility itself. As long as things are moving smoothly, there's not a lot of need for the user to worry about the mechanics of accelerator operation.

The LCLS is a different beast. Users and accelerator physicists will work much more closely to tailor the machine for each experiment. And being the first light source of its kind, the experiments themselves require new ways of thinking.

Over the next few days I hope to explore the nature of this new beast, with a look at the current science of traditional light sources, and how the LCLS is changing the way light source scientists work together.]]> Wed, 15 Oct 2008 15:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=73 http://lcls.slac.stanford.edu/Article.aspx?article_id=64
I, and the rest of the LUSI team, can now breathe easy. However, the review season will come again and all of the preparations will begin anew. Nonetheless, I'm going to really enjoy the time I have now to dig into the details of the instrument design and try to think of some creative ways to do things. Hopefully this will translate into some interesting blog posts!

]]> Mon, 15 Sep 2008 19:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=64 http://lcls.slac.stanford.edu/Article.aspx?article_id=63 We've got at SLAC many people that are in love with the "little queen," the nickname in France for the bicycle (la petite reine). And some push that love to the top of competition.

Within the LCLS group, bike fans have been riding together during lunch. We have a mailing list, "the SLAC noon riders," where whoever wants to ride e-mails the list to find a partner for a loop. The surroundings of Stanford are a dreamland for bikers!


I prefer climbing, and there are plenty of nice climbs here; but one of my colleagues, Tim Montagne, is a "tracky" (a competitive track cyclist). The track ring is his domain, but he rides with us from time to time to build up endurance.

In early September I went to see Tim race for the US Masters National at the San Jose velodrome. It was a long preparation for him since early in the spring. (Like for the LCLS X-FEL project, before using the beam for real experiments you have to go through very long commissioning runs and fine tune all parameters!) Tim and his two team members were registered in the +40 age group Team Sprint race. Those guys were freaking fast! Like bullets they did three laps around the ring: Tim was leading the first lap from a standing start, then his two partners did the second lap and the last racer Bobby finished the last lap alone in 1:07.19. That was the best time among the other competitors, winning them the Gold medal!

Great victory for the LCLS somehow... I watched the Pursuit race too. It's like a team time trial but on the track. Four bikers follow each other in a straight line as much as they can and as close as possible (like the width of the tires) to fight against air resistance. Pretty nice, and a good example too of team spirit.


]]> Fri, 12 Sep 2008 19:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=63 http://lcls.slac.stanford.edu/Article.aspx?article_id=59
I must admit, I often feel the same way. Watching the news is generally a very negative experience. The topics these days are typically limited to a declining economy, war in Iraq, inflation, crime, and various other depressing issues. I think people would be receptive to hearing about positive ways their tax dollars are being spent. The LCLS is a good example and we have a chance to take the initiative and spread the word about this exciting machine. However, we must do a much better job than what we’ve done thus far.

As an example, we can take a look at the Large Hadron Collider at CERN. People know about the big machine that crosses the Swiss-French border. It’s actually so popular that the now famous author of The DaVinci Code, Dan Brown, has written a New York Times best-selling follow up novel that uses the LHC as a major scene called Angels and Demons. The book fascinated the public so much that a movie adaptation staring Tom Hanks is expected to be in theaters in the spring of 2009.

Why is the LHC so popular? Is it simply the enormous price tag of that experiment that draws attention? I don’t think so. Take a stop in the physics section of your local bookstore and do a quick survey of the books in that aisle. What you’ll notice is that most of the books are about the origin of the universe, supersymmetry, string theory, etc. These seem to be the topics that resonate with the public. I can see where this stems from since this type of research aims at answering many of the basic questions we’ve all asked at some point: Where did we come from? How did the universe begin? The LHC was constructed to improve our understanding of these questions.

However, I doubt that the experimental evidence of theoretically predicted particles will have role in the problems facing our globe today.

The basic research that will be performed at the LCLS may very well alter our understanding of many processes relevant to energy applications. This is now a forefront topic of global relevance that will only grow. It’s time to start selling our science and our facility and we now have a unique opportunity in time, a time where this is a topic on everyone’s minds, that we must take advantage of. I look forward to the day when the average citizen will recognize the LCLS. Does anyone have Dan Brown’s number?
]]> Thu, 28 Aug 2008 19:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=59 http://lcls.slac.stanford.edu/Article.aspx?article_id=53
Measuring beam power without touching the beam itself is a well understood technique at synchrotron labs. But the LCLS beam will carry so much more power than any other machine—about a one trillion watts per pulse, delivered in a quadrillionth of a second—that we don’t even really know the physics of how the beam will behave around traditional sensors. So the only way to know is to measure the beam power full-on and directly, and see what registers on the traditional sensors.

What’s amazing to me, though, is that the new sensor itself is not destroyed by this abuse. Friedrich says that was part of the engineering challenge—to develop a sensor using just the right combination of materials so that we get the information we want without destroying the detection equipment in the process. Although a variety of sophisticated materials exist that could do the job, the team chose silicon (because it's easy to get) to absorb the beam's power, and a material called manganite that turns the heat generated in the silicon into a signal.

Although custom instrumentation for physics experiments is nothing new, I think this case in particular illustrates exactly how “beyond the envelope” the LCLS really is.
]]> Fri, 01 Aug 2008 19:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=53 http://lcls.slac.stanford.edu/Article.aspx?article_id=41
Paul Emma, who heads the accelerator team continues to dazzle us with results that exceed our expectations. The ability to "time" a laser to the X-ray beam is one of the most significant challenges for experiments where we want to excite a sample with an optical laser and then probe it with the LCLS X-ray beam to look for changes in atomic positions in a myriad of different systems. The LCLS team just demonstrated that with respect to a "clock," the timing signal that one can use for the excitation laser we see "jitter," or uncertainty, in the relative arrival time of the electron beam that will create the X-rays of just under 50 femtoseconds. That may not sound like much, in fact isn’t very much and that’s just the point. If the experimenters can time the laser to this same clock with similar accuracy, then we can start to do experiments with time resolution of the same order as the X-ray pulse duration. That’s a big deal.


To further boost my anticipation and the reality of progress we just completed a workshop for prospective experimenters for the first hard X-ray instrument planned to come on line by June-July of 2010. A few weeks prior to that, we hosted a workshop for experimenters interested in using the first instrument that will be available—the atomic, molecular and optical (AMO) science instrument, ready in July 2009. The buzz at the AMO workshop was palpable, but its only a year away. More amazing was the atmosphere for the X-ray pump-probe workshop. A group of 60 or so scientists from around the world were truly excited about the possibility to do first experiments now just two years away. There is little doubt this excitement will only build. This enthusiasm is critical for the instrument scientists who are slogging away to build the best for the user community as it supports their mission and recognizes their efforts.

]]> Mon, 23 Jun 2008 19:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=41 http://lcls.slac.stanford.edu/Article.aspx?article_id=40

At that moment of the LCLS commissioning, the RF gun is running with a drive laser that has a super-Gaussian profile (also called top hat) in time and in space. It has been found that it could be optimum to get the needed emittance for the electron beam at the output of the injector with uniform slices inside the bunch. During the commissioning while running the machine at a charge of 250 pC, the target value for the laser temporal profile is a 6.5 ps wide super-Gaussian with a rise/fall time of 1 ps. The transverse mode has to be a homogenous flat top beam that has an adjustable diameter of 1.2 mm on the gun cathode.

Unfortunately for us (well, actually it is the fun part of the laser!), our classic Titanium:Sapphire laser system has the usual Gaussian profile in time and in space. For this reason, our challenge is to “shape” the laser beam so it can be used for the gun cathode.


It was first planned to convert the UV laser output beam that has a Gaussian spatial shape to a flat top by using a commercial “beam shaper” and imaging the shaper output to the photocathode with adjustable magnification. But this shaper had very high requirements to the quality of the input beam and to the accuracy of alignment. So it did not provide better beam quality than simply clipping the beam with a hard-edge aperture after a three lenses adjustable zoom telescope. This simple method has been finally chosen and even we have several apertures with different increasing diameters set on a remote controlled wheel so that the laser beam size on the cathode can be changed only by rotating this wheel. Like the barrel of a gun!


The temporal shaping is made through a little device called with the acronym A.O.P.D.F. It means Acousto Optic Programmable Dispersive Filter. It is a commercial product called the “Dazzler”. The funny anecdote is that “Dazzler” is a Marvel Comics superheroine associated with the X-Men. She is a mutant with the ability to convert sound vibrations into light and energy beams.

And that’s pretty much what happens inside the TeO2 crystal of this A.O.P.D.F. An acoustic wave is applied on it with a transductor (like in a microphone) that modulates its index of refraction acting then like a dispersive grating. So when the laser goes through it, it is diffracted with a wavelength dependency. By programming the RF wave, you can filter and shape the amplitude of the laser spectrum. The Dazzler is installed inside the laser system before the amplifiers. We have programmed our Dazzler to apply a super-Gaussian filter onto the spectrum amplitude. Spectrum is linked to time via the so called Fourier time/frequency relation. So the temporal envelope of the laser will be close to a super-Gaussian too.
]]> Fri, 13 Jun 2008 19:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=40 http://lcls.slac.stanford.edu/Article.aspx?article_id=27 Stir crazy: A mental condition experienced by prisoners

Let me say this right off the bat to avoid any inkling of confusion: I really like my job. I enjoy coming to work everyday. It's probably a safe assumption to say the majority of people in the world can’t make that claim. However, as exciting as this project is to work on it's been difficult adjusting to the time scales of such a large project. My experience as a graduate student was a great blend of desk work (simulations, analysis, article writing) and hands on work (experiments, laser tuning, machine shop, etc…). It's not healthy to have too much of either.

We’ve been in the planning stages of the XPP instrument for almost 2 years now and there’s at least another year before any hardware will be delivered. I get a "divide by zero" error on my calculator when I compute the ratio I described above. There’s no hands on work for me to do and no hope for any in the near future. Well, I guess "near" is relative but you get my point. This is torture to an experimental physicist. It's driving me stir crazy. My only solace is that the instrument is going to be amazing when it all comes together.
]]> Fri, 25 Apr 2008 19:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=27 http://lcls.slac.stanford.edu/Article.aspx?article_id=24
I thought I would tell you what a free electron laser means from the perspective of a user awaiting first light. Up until now, the state-of-the-art X-ray sources in the wavelength regime around 1 Angstrom (0.1 nanometers) have been the electron storage rings with undulators that produce extremely bright spontaneous synchrotron radiation. These sources provide tremendous capability to determine where the atoms are in a myriad of materials ranging from simple molecules to large bio-molecules with many thousands of atoms.

Knowing the atomic structure is fundamental to developing our understanding of the properties of materials, but if one could know how the atoms move by making molecular movies we would gain invaluable insights. To make a molecular movie we need a movie camera with the ability to see the atoms and to take the individual stills that will make up our movie on the time scale of the atomic motion.

Synchrotron radiation from electron storage rings give us the ability to see the atoms, but the exposure time (the X-ray pulse duration) is too long. Here is where a FEL comes into its own. The LCLS will create bursts of X-rays lasting only 100 femtoseconds. That is a time short enough to "freeze" the atomic motion and allow us to take a series of stills of the atomic positions during a chemical reaction, for example, and create the atomic movie. The making of these movies will be among the first experiments that we perform at the LCLS.

It all sounds simple, but before we take our "stills" we will need to build the equipment to do the experiments. The Department of Energy has funded a project, the LCLS Ultrafast Science Instrument Project (LUSI) to build three X-ray instruments for the LCLS. As part of each instrument, we will also need diagnostics to tell us the properties of each X-ray burst or pulse because such pulses vary each time. The diagnostics in fact are the most challenging task in the construction of the experiments.

There is a team of eager young instrument scientists focused on building the instruments, the diagnostics and most importantly pushing to see that first X-ray pulse and collect that first data set. We have started on this, but the excitement will build over the next few years as the day approaches for first light to the first LUSI instrument.
]]> Sat, 19 Apr 2008 19:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=24 http://lcls.slac.stanford.edu/Article.aspx?article_id=18
Eventually, as progress on the LCLS gathers steam, the LUSI group will be integrated into the LCLS project as a whole. But in order to design and build a machine of this magnitude and scope required early budgetary considerations that segregated the experiments into a category to themselves.

Makes sense, actually. First, when you consider the complexity of the LCLS, something that's never been built before, just designing the laser alone is a monumental challenge. Add to that the challenge of then using its beam, which will be unlike anything available to scientists before, and you have what amounts to two very different kinds of projects, eventually coming together to achieve one goal: X-ray science of a caliber that so far has only been theorized.

The AMO instrument is different in that it always has been a project distinctly under the umbrella of the LCLS, and not LUSI. I suppose the initial funding entities found the prospect of building a laser without an experimental target a bit lopsided, thus the AMO. Looked at that way, if you buy yourself an LCLS, you get at least one experiment to play with. LUSI is the expansion pack. ]]> Tue, 15 Apr 2008 19:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=18 http://lcls.slac.stanford.edu/Article.aspx?article_id=14
1) To be frank, I’m just not a very good writer. It’s a painfully slow process for me to put my thoughts down on paper. I think part of my problem is that I dissect every single sentence I write down. Unfortunately for me, this level of scrutiny has no impact on the quality of my work.

2) Is my life really that interesting that people will want to take the time out their day to read about it?

3) I’m just not a talker. My wife gets on my case about this all the time. You see, talkers like to be around other talkers. That way they don’t have to worry about the possibility of engaging in a lopsided conversation, which I guess can be uncomfortable but I wouldn’t know first hand. Likewise, non-talkers can be around non-talkers without any problems either. However, things can turn bad when you mix a talker with a non-talker. This is the situation with my wife and me. She’s a talker, I’m a non-talker. My silence can be misinterpreted sometimes and this can cause some problems. This phenomena isn’t just limited to her though. A co-worker that I car pool with to and from SLAC nearly everyday from San Francisco has similar issues with me. It’s not unusual for him to wave his hand in front of my face to see if I’m alive. So, here’s my issue - how can someone who doesn’t have much to say maintain an online diary for the next few years?

So these are the things that if been thinking about. In the end, I decided that it’s a good idea for me to do this blogging thing. Maybe it can help me fix problems 1 and 3. There’s not much I can do about problem 2. Hopefully the excitement of the LCLS project will help with that. Anyway, without further ado, welcome to my world. Hopefully my writing improves with time. 
]]> Sat, 15 Mar 2008 22:15:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=14 http://lcls.slac.stanford.edu/Article.aspx?article_id=19
During the few weeks I spent at SLAC installing the drive laser for the LCLS injector before getting this position, I realized how exciting the LCLS project is and how fun it would be to live in the Bay Area. Besides the work that shows great perspectives, there are a lot of different things you can do here. Things that may bring over your relatives and friends to visit you without hesitation: good weather, a lot of great outdoor activities, San Francisco is kind of a European city with good restaurants, good Californian wines (crucial for French people), etc.

As soon as I arrived at SLAC, everybody was very helpful to get me installed in the US (including of course my supervisor, many thanks Bill). Indeed, some things are running different here compared to France even though Americans and Europeans could be closer to each other than Europeans and Japanese (that’s an example) could possibly be. And so far, after almost one year I don’t regret at all my decision.

I was running a complete different day to day rhythm in Paris. I was spending two hours in stressful traffic to go to work. Here I am biking to work (well I am lucky to have found something reasonable nearby with the help of the SLAC housing service). I am even going for a bike ride during lunch breaks with my colleagues (call us the SLAC noon riders). Tremendous motivation while riding to the top of Alpine road is like an image of how people are reactive and enthusiastic about being involved in the construction of the LCLS X-FEL beam!

In less than one year, the laser group with the help of the Controls group installed a lot of devices in the laser beam line (steering stabilization loops, remote controlled diagnostics…) to make it as reliable and easy to use as possible during the LCLS commissioning. Who could imagine a 98% uptime for the laser during 2007 commissioning before its installation in 2006. Like on your bike, you need good legs for this!
]]> Wed, 20 Feb 2008 20:00:00 GMT http://lcls.slac.stanford.edu/Article.aspx?article_id=19