WEBVTT 1 00:00:15.850 --> 00:00:19.649 Mark Kushner: We do. So that's incidental. 2 00:00:20.070 --> 00:00:21.650 Mark Kushner: But look at Ontario. 3 00:00:28.050 --> 00:00:30.830 Mark Kushner: Those numbers that they can speak. 4 00:00:42.740 --> 00:00:46.120 Mark Kushner: How's it going? 5 00:01:05.040 --> 00:01:06.340 Mark Kushner: Sometimes it really. 6 00:01:28.350 --> 00:01:30.580 Mark Kushner: Perfect. 7 00:01:34.910 --> 00:01:47.400 Mark Kushner: We're just about ready to get started now. 8 00:01:47.610 --> 00:01:55.530 Mark Kushner: So, it's my pleasure to introduce our speaker today. Matt Redwitt has come from. 9 00:01:55.610 --> 00:02:14.770 Mark Kushner: I'll tell you a little bit about him, his bio. He's an assistant professor of mechanical engineering at Stanford University. He received his BSc, MA, and PhD degrees from Princeton in Mechanical and aerospace engineering, and then from 2019 to 2022, 10 00:02:14.870 --> 00:02:21.779 Mark Kushner: And the Lawrence Fellow in the National Ignition Facility, Photon Science Directorate at Livermore National Laboratory. 11 00:02:21.920 --> 00:02:35.309 Mark Kushner: And, your research, applies high-power lasers to the development of optical diagnostics for fluids and plasmas, and the study of intense matter-light-matter interactions. 12 00:02:35.590 --> 00:02:41.349 Mark Kushner: And the construction of compact light and particle sources, and you combine it. 13 00:02:41.470 --> 00:02:53.209 Mark Kushner: adaptive high-repetition rate instruments and large-scale simulations to explore new regimes in food mechanics, thermodynamics, material science, and plasma physics. And we're… 14 00:02:53.210 --> 00:03:09.710 Mark Kushner: In addition to having him as a MIPSI speaker, he's also got his group across the road in Zeus doing an experiment right now, so he's rather busy. But, we have a couple of things for Matthew, because this year he is the Early Career Onboard Lecturer, so… 15 00:03:09.710 --> 00:03:15.430 Mark Kushner: And to celebrate that, you get this rather nice little plaque picture, and we also… 16 00:03:15.910 --> 00:03:36.819 Mark Kushner: Now, this can't be true anymore. But supposedly, there are fewer MIPSI mugs than there are Nobel Prizes in physics. One of the… Thank you very much. …Mipsi mugs and take back as well. So, thank you all so much for coming to speak with us. We're going to hear about plasma and gas optics for ultra-recents. 17 00:03:36.820 --> 00:03:39.820 Mark Kushner: lasers, but… Okay, thank you, Luis. 18 00:03:39.860 --> 00:03:44.069 Mark Kushner: Thank you all for, for being here this afternoon. 19 00:03:44.290 --> 00:03:56.249 Mark Kushner: It is a pleasure to be here and to see everything that you have going on here at Michigan in plasma Science. It's a wide range of impressive facilities across quite a few different research areas. 20 00:03:56.480 --> 00:04:06.709 Mark Kushner: Today, I'm going to be talking about research that's most relevant to one of those areas of research here at Michigan, and that's our development of 21 00:04:06.960 --> 00:04:15.500 Mark Kushner: Optics made of plasma and gases, so these are optics for… that can tolerate extremely high intensities and laser fluences. 22 00:04:15.610 --> 00:04:27.030 Mark Kushner: And what we'll discuss today is how we think about trying to use these optics to design new, more capable laser systems than what today's technology provides. 23 00:04:27.290 --> 00:04:43.919 Mark Kushner: So this is one of the areas that my group works on. There. We work in a few different areas, mostly related to laser technology and high-temperature plasma physics of various types, so some fundamental work in plasma physics. 24 00:04:43.930 --> 00:04:55.289 Mark Kushner: A lot of work on plasma and gas photonics, laser diagnostics, and relativistic interactions, as well as particle and light sources and some 25 00:04:55.290 --> 00:05:05.489 Mark Kushner: work on strong field interactions. But, today, we're going to focus on one of these topics, which is plasma and gas optics, or photonics, or… 26 00:05:05.580 --> 00:05:11.279 Mark Kushner: whatever nomenclature, I guess, I happen to feel like when I make the particular slide that we're on. 27 00:05:11.490 --> 00:05:26.870 Mark Kushner: So, I'd first like to introduce the members of my group, some of whom are here this week, who have done… some of them have done various parts of this work, so we have a handful of graduate and undergraduate students who work with us, and you can… 28 00:05:26.870 --> 00:05:33.140 Mark Kushner: see some of them having fun in the lab. They are all smiling under their masks, for sure. 29 00:05:33.250 --> 00:05:35.100 Mark Kushner: Okay, so… 30 00:05:35.410 --> 00:05:51.680 Mark Kushner: Today's outline, what we'll talk about to keep you oriented, I'm going to discuss a little bit of an overview of why we're interested in plasma optics and the kinds of capabilities that they deliver for high-intensity physics. 31 00:05:51.680 --> 00:06:02.749 Mark Kushner: And then we'll talk through our experimental and computational results on gas gratings, plasma gratings, and making holographic lenses with these mechanisms. 32 00:06:03.240 --> 00:06:11.780 Mark Kushner: So… Bat. When we talk about extreme lasers, lasers that deliver, 33 00:06:11.780 --> 00:06:35.370 Mark Kushner: very, high capabilities. We can divide those up into a few different categories. First, we have, what we might call high peak power laser systems. So these are typically, short, what we call short pulse lasers delivering femtosecond pulses with up to multiple petawatts, maybe 10 petawatts of peak power. We can use these for things like, 34 00:06:35.370 --> 00:06:50.109 Mark Kushner: Building compact particle accelerators for exploring strong field physics, and as sources, compact sources of secondary radiation. So, X-rays, terahertz, whatever part of the electromagnetic spectrum you want, in principle. 35 00:06:50.380 --> 00:07:08.650 Mark Kushner: We also work with high-energy lasers, so these are facilities like the National Ignition Facility, which can be up to megajoules of energy delivered in nanosecond pulses, and these are typically used for things like inertial confinement fusion, high energy density science, lab astro problems. 36 00:07:08.650 --> 00:07:22.529 Mark Kushner: And at Stanford, we also have an example of X-ray-free electron lasers, so big accelerators driving extremely bright X-ray sources for things like material science and ultra-fast science of various types. 37 00:07:22.610 --> 00:07:29.370 Mark Kushner: Now, the optics we're going to talk about today, at the moment, are mostly relevant to these high-power and high-energy systems. 38 00:07:29.370 --> 00:07:47.740 Mark Kushner: But what's important to keep in mind is that the architectures underlying these lasers are a little bit different, and the types of optics that are most useful for them are a little bit different, and you'll see as we go along that this drives us towards slightly different approaches to thinking about how we build the next generation of these systems. 39 00:07:47.740 --> 00:07:48.370 Mark Kushner: Excellent. 40 00:07:49.180 --> 00:07:50.140 Mark Kushner: So… 41 00:07:50.580 --> 00:08:05.629 Mark Kushner: Laser is a… laser's a pretty cool tool, but fundamentally, the capabilities of these laser systems is that we can use them to deliver a fairly moderate amount of energy for these high-power lasers. 42 00:08:05.700 --> 00:08:20.760 Mark Kushner: But we can deliver that energy in a very short amount of time, and a very small area. So we can take something 10 joules in everyday life, 10 joules is not a lot of energy, right? Think about a cup of coffee, 10 joules isn't necessarily going to get you that far. 43 00:08:20.760 --> 00:08:32.540 Mark Kushner: But delivered in tens of femtoseconds to micron spot sizes, we now have extremely high intensities, that allow us to do, 44 00:08:33.470 --> 00:08:38.109 Mark Kushner: a lot of very interesting science. So, the problem that we have to solve… 45 00:08:38.110 --> 00:08:56.969 Mark Kushner: if we are designing plasma optics, or optics for these systems, is can we use them to focus energy in space, and can we use them to compress energy in time? And those are the, sort of, the two most important optical things that we have to do as we think about building these systems. 46 00:08:56.970 --> 00:09:05.090 Mark Kushner: Now, in terms of focusing in space, you all kind of know what a lens is, and maybe know about OAPs and focusing mirrors. 47 00:09:05.090 --> 00:09:13.440 Mark Kushner: Compressing in time is a little bit more complicated. The basis of this is, what's called shirt pulse amplification. So, 48 00:09:13.440 --> 00:09:31.109 Mark Kushner: wait, you have a center here that's named after one of these people. But the idea here is that if we have a low power source of femtosecond light, we can turn that into an extremely high-power source of femtosecond light by taking our small, low-power pulse. 49 00:09:31.110 --> 00:09:40.679 Mark Kushner: Stretching it out in time, so spreading that energy in time, running it through amplifiers to increase the energy with sort of moderate power levels. 50 00:09:40.680 --> 00:09:49.619 Mark Kushner: And then recompressing it, so that we now have a very short and very high power pulse with all of that energy from the amplifiers. 51 00:09:49.690 --> 00:10:03.209 Mark Kushner: Now, this, approach has been what has basically enabled all of the advances in high-peak power lasers from the 80s through today. It's the basis of lasers like Zeus and other systems. 52 00:10:03.530 --> 00:10:16.790 Mark Kushner: But the issue, comes from the fact that once we've compressed it, so as soon as we hit that final compression grading, we now are back at the full power in the laser. So that optic, which 53 00:10:16.840 --> 00:10:33.839 Mark Kushner: has fine micron-scale structure on it, has to handle petawatts or 10 petawatts of power, so it has to be big, it has to be expensive. We constantly run the risk of damaging it and, you know, killing the laser, basically. And so. 54 00:10:34.020 --> 00:10:47.649 Mark Kushner: What we are limited by, if we think about adding increased power to laser systems, is largely the fact that there's a finite amount of power that we can put on this final grating and anything downstream of it. 55 00:10:47.880 --> 00:10:52.019 Mark Kushner: That limits, the capability of our high-power laser system. 56 00:10:53.800 --> 00:10:54.890 Mark Kushner: So… 57 00:10:54.910 --> 00:11:19.840 Mark Kushner: To give you a sense of the scale here, got a laser pointer this time. So here's my laser pointer. You can see it's, like, that size, approximately. If you think about the sort of millijoule, kilohertz Thai sapphire systems that you'll find all over campus, they might have beam sizes that are sort of centimeter, subcentimeter in scale, and our moderate laser at Stanford, which delivers about 10 58 00:11:19.840 --> 00:11:20.930 Mark Kushner: terawatts. 59 00:11:20.930 --> 00:11:24.519 Mark Kushner: Is, a 4cm beam or so. 60 00:11:24.910 --> 00:11:32.319 Mark Kushner: So, 4 centimeters is pretty big for, like, compared to a laser pointer, but, manageable. 61 00:11:32.570 --> 00:11:51.449 Mark Kushner: Then, if we start to think about the beam diameters as we go up in size, here's our little Stanford laser here. We have, Bella, and I'm afraid I didn't look up what the most recent power you've gotten on Zeus is. It's, I know it's, it's getting up there, but something like this for these systems. 62 00:11:51.450 --> 00:12:02.319 Mark Kushner: So these optics are now starting to get fairly big, at the Eli scale, where we have to have meter-scale optics once we've done compression. 63 00:12:02.450 --> 00:12:07.739 Mark Kushner: So, this is still viable, if expensive, to get to 10 pedals once. 64 00:12:07.860 --> 00:12:18.610 Mark Kushner: But some of the science we'd like to do, to really probe the Schwinger limit, for example, requires beams that are not at the 10 petawatt level, but, say, an exawatt. 65 00:12:18.630 --> 00:12:37.330 Mark Kushner: So if we were to take today's technology and build an X-watt laser, we'd be stuck with something like this. And now we're talking about mirrors and gratings and optics that are 10 meters across, or bigger, and suddenly this isn't even a problem of money, but also technology capability. 66 00:12:37.330 --> 00:12:44.179 Mark Kushner: Even if you throw the small… the GDP of a small country at it, it's still going to be pretty challenging to put something like this together. 67 00:12:45.760 --> 00:12:54.669 Mark Kushner: What plasmas offer us is the possibility to deliver this kind of power in beams that are much smaller. 68 00:12:54.780 --> 00:13:05.620 Mark Kushner: We'll talk about why this is possible, the idea that we can use plasmas to deliver this extremely high power from footprints that are not quite so big. 69 00:13:06.120 --> 00:13:18.350 Mark Kushner: So, question for you, maybe, is how would you go about building a laser system at this scale? How would you think about building an X-Watt laser, or maybe of equal interest to us? 70 00:13:18.350 --> 00:13:37.830 Mark Kushner: How would we build petawatt lasers, the kinds of things that might enable future particle accelerators, at repetition rates much greater than a kilohertz? So if I wanted a, say, a 10 or 100 kilohertz petawatt laser, that's also going to be a problem with today's technology. And what are the kinds of approaches that we might take? 71 00:13:38.450 --> 00:13:57.499 Mark Kushner: So, the fundamental limitation is coming from the fact that if we put too much power on a solid-state optic, we'll damage it. We'll turn a beautiful piece of glass or a grating into something that looks like this. You can see this is a hole drilled through a piece of glass. Not what you want to do with a few hundred thousand dollar optic. 72 00:13:57.750 --> 00:14:02.349 Mark Kushner: So, there is a limited intensity that we can put on solid-state optics. 73 00:14:02.360 --> 00:14:19.270 Mark Kushner: And there's only so far that material science is going to take us, because at some light intensity, much above, say, 10 to the 13 watts per centimeter squared, we are going to ionize pretty much everything. So whether or not you want to have a plasma, you will have a plasma on your optics. 74 00:14:19.530 --> 00:14:32.149 Mark Kushner: So, the natural question then is, can we use plasmas to make the optics? Because in principle, if we can make optics with plasmas, then we're limited by relativistic effects rather than the formation of plasma. 75 00:14:32.150 --> 00:14:44.149 Mark Kushner: And we might be able to put anywhere from a thousand to a million times more intensity on each of our optics. That means everything can be smaller, and it's no longer so unreasonable to talk about extremely high peak powers. 76 00:14:45.660 --> 00:14:54.810 Mark Kushner: So that's if plasmas can work as optics. Now, obviously, there's a lot of work being done by that, hood, there. 77 00:14:55.030 --> 00:15:10.459 Mark Kushner: This basic fact that plasmas have a much higher damage threshold has motivated a lot of work on different types of plasma and gas optics. Some of these are in use, like the use of crossbeam energy transfer to rebalance energy inside NIF hall rooms. 78 00:15:10.460 --> 00:15:15.470 Mark Kushner: Some of these are only at the proof-of-principle level, polarizers and wave plates. 79 00:15:15.470 --> 00:15:30.430 Mark Kushner: Plasma mirrors are used in high-intensity beamlines, and plasma waveguides have enabled a lot of progress in laser-driven wavefield accelerators. Plasma amplifier has been somewhat less successful at revolutionizing laser design. 80 00:15:30.430 --> 00:15:37.539 Mark Kushner: But, there have been a bunch of approaches, but the issues we run into is that plasmas are pretty hard 81 00:15:37.570 --> 00:16:01.369 Mark Kushner: to do stuff with. They really do not like doing what you want them to do. You push them in one direction, they want to go in the other direction. We have lots of instabilities, sensitivity to things that are non-ideal, and kinetic effects are really complicated. So, if we're gonna deliver optics that can be used in systems, we need to be able to resolve some of this robustness issue, and that brings us to a few properties that we want. 82 00:16:01.390 --> 00:16:03.509 Mark Kushner: From a potential plasma optic. 83 00:16:03.670 --> 00:16:17.970 Mark Kushner: We'd like a high damage tolerance. Kind of, maybe obviously, if the plasma isn't giving us a high threshold for laser damage, we might as well switch back to glass. It's much easier to use glass and metal. If we don't have this, we don't really have any of those. 84 00:16:18.070 --> 00:16:35.670 Mark Kushner: Ideally, we'd like approaches that support high repetition rates, that we can run the laser many, many times, and this pushes us towards gas or liquid-based approaches for making the plasma, rather than solid targets, where it's difficult to refresh the surface at high repetition rate. 85 00:16:35.710 --> 00:16:44.849 Mark Kushner: We're going to need optical quality, which pushes us maybe towards laser-shaped plasmas, where we can do things on micro wavelength scales. 86 00:16:45.240 --> 00:16:53.450 Mark Kushner: And these optics, and this is one that's tricky and pushes us towards particular mechanisms, is that 87 00:16:54.070 --> 00:17:16.629 Mark Kushner: there are many ways that you can think about making a plasma that will affect light. There are only a few of those ways where you can use less energy to make the plasma than you can control with it. So we need our plasma optic to need a small laser to control a big laser, rather than the other way around, or we haven't really gained anything. We need a really big laser to make an optic for a small laser. 88 00:17:17.210 --> 00:17:34.450 Mark Kushner: Ideally, we're not relying too much on nonlinearities. We want a beam that comes out that has good… is still a beam, still a laser beam. Compatibility with femtosecond pulses is nice, which requires some kind of broad spectral response. 89 00:17:34.490 --> 00:17:47.229 Mark Kushner: And ultimately, they need to be robust and stable. They need to work not just once, but 10 times a second, or a thousand times a second, for, for years, ultimately, if we're going to do something like replace an accelerator. 90 00:17:48.610 --> 00:18:03.529 Mark Kushner: So, the approach that we're going to take are based on volumetric diffractive optics. So here, we're going to write periodic patterns of plasma in shapes, like… rating shapes like this, or lens shapes like this. 91 00:18:03.550 --> 00:18:15.970 Mark Kushner: And the reason that we have chosen to take this approach, so there are quite a few reasons, but one, this is a linear optical effect. It's basically a diffractive optic, there's no plasma nonlinearity here. 92 00:18:16.760 --> 00:18:33.400 Mark Kushner: Most usefully, the optical properties depend mostly on where we create the plasma, more strongly than how dense the plasma is. There is dependence on density, but most of the optical properties depend most strongly on where the fringes are, not 93 00:18:33.400 --> 00:18:44.349 Mark Kushner: How much plasma is created at a particular location. And if we're writing with lasers, we can control where these fringes are really well, even if plasma density is a little bit harder to get a handle on. 94 00:18:44.860 --> 00:18:54.729 Mark Kushner: And then if we're operating with transmissive optics, it turns out that we can do this with gas density plasmas. We don't need to go to solids, and that is another advantage. 95 00:18:55.140 --> 00:19:04.630 Mark Kushner: So, the idea here, we have diffractive optics, which are based on the fact that if we have a periodic structure, something like this. 96 00:19:04.630 --> 00:19:20.090 Mark Kushner: A light incident at the Bragg angle will be efficiently diffracted, and there is a relationship here between the angle of incidence, the wavelength of the beam that's being diffracted, and the grading period that we have to satisfy in order to have this efficient interaction. 97 00:19:20.090 --> 00:19:34.820 Mark Kushner: This is, you know, similar to X-ray diffraction from crystals or any of a variety of effects. Now, I'll note that these are volumetric gratings, which is a little bit different than the surface diffraction gratings that are more common for solid-state objects. 98 00:19:36.550 --> 00:19:37.430 Mark Kushner: So… 99 00:19:37.490 --> 00:19:56.349 Mark Kushner: The nice thing about volumetric gradings in the Bragg regime, which is the thick grading regime, loosely, is that, in theory, a grading operating in this regime can be 100% efficient. That unlike a surface grading, where we always get diffraction to other orders. 100 00:19:56.350 --> 00:20:15.059 Mark Kushner: For a thick volumetric grading, we can have 100% diffraction to a single order, so there's no theoretical limit on how efficient these optics can be. Now, obviously, there are going to be lots of practical limits, but it's nice to not have the theoretical constraints on the efficiencies we can get. 101 00:20:15.060 --> 00:20:15.920 Mark Kushner: Now… 102 00:20:15.990 --> 00:20:39.950 Mark Kushner: The efficiency of a transmission grading of this type goes with the sine squared of what's called the coupling coefficient times the grading thickness times the detuning parameter. We're generally going to be trying to operate in a regime where B is 1, so it's basically this coupling coefficient times D, where the coupling coefficient depends most critically 103 00:20:39.950 --> 00:20:46.629 Mark Kushner: on the amplitude of the index modulation that we can create. So how… how strong the grading is. 104 00:20:46.630 --> 00:21:11.220 Mark Kushner: And what happens is if you create a grating strong enough, it's a certain length, you'll get 100% diffraction efficiency here. If you keep making the grading thicker, the light will start to diffract back into the zeroth order, and for a plane wave, we'll just oscillate back like that as it travels through the grading. We want to operate at this point here, this first maximum point for a diffraction grading. 105 00:21:12.640 --> 00:21:21.069 Mark Kushner: Okay, so, that's great on the optical side. We need to think about how we actually make one of these gratings in the lab. 106 00:21:21.170 --> 00:21:36.199 Mark Kushner: So what we can do is take a pair of laser beams and cross them, and you perhaps all know, maybe, that if you cross a pair of laser beams, you'll get an interference pattern between them, kind of like a two-slit type experiment. 107 00:21:36.700 --> 00:21:44.920 Mark Kushner: The pattern of this interference will be related to the wavelength of the laser and the crossing angle, and you'll get an intensity profile that looks like this. 108 00:21:45.180 --> 00:21:56.499 Mark Kushner: So, if we have some mechanism that couples an intensity profile to plasma density, this intensity profile will map to a 109 00:21:56.500 --> 00:22:14.720 Mark Kushner: modulation of our plasma density, and because the index of refraction of a plasma depends on the density of the plasma, and this is for collisionless plasma, if we have a modulation of the plasma density, we will have a volumetric diffractive optic of this type. So two beams… 110 00:22:14.720 --> 00:22:29.650 Mark Kushner: write the grading, we get a pattern of plasma, and then a probe beam coming in, in principle, can diffract off it, and in theory, this diffraction efficiency can be 100%. Okay, so that's the basic starting point. 111 00:22:29.890 --> 00:22:38.550 Mark Kushner: This is… you can think of this as a simple hologram. So, basically, these gratings are the hologram of a plane wave. 112 00:22:38.550 --> 00:22:55.970 Mark Kushner: So, you can imagine that if we, instead of just crossing two beams, put some kind of pattern into one of the beams, we'll be a… we'll get a grating that is a hologram of that pattern, and then a probe beam coming in will have that pattern imparted onto it. So there's… 113 00:22:55.970 --> 00:23:15.670 Mark Kushner: The process generalizes. In principle, if we can do it for a grading, we can do it for other kinds of optics, anything that can be represented in this holographic way. And that leads us first to lenses, which are a very simple extension of gratings, where one of the beams is focusing, and we get curved fringes. 114 00:23:15.670 --> 00:23:22.889 Mark Kushner: And then the probe will focus, but also to more complicated things like transferring orbital angular momentum and other kinds of optics. 115 00:23:23.180 --> 00:23:23.980 Mark Kushner: Okay. 116 00:23:24.250 --> 00:23:25.200 Mark Kushner: Soap. 117 00:23:26.940 --> 00:23:36.319 Mark Kushner: You know, the fundamental question is how do you actually, what mechanisms can you use to, create the plasma or the density modulation? 118 00:23:36.610 --> 00:23:45.600 Mark Kushner: Now, many of you are working on various types of plasma physics. In principle, there are many different ways that you can structure plasma density with an electromagnetic wave. 119 00:23:45.720 --> 00:23:53.839 Mark Kushner: But there are only a few of them where you can use a weak wave to structure density in a way that will persist for a strong wave. 120 00:23:54.210 --> 00:24:19.059 Mark Kushner: One of them, the one that we've, one that we've used in the lab most heavily is based on ionization, where we use a pair of femtosecond beams, typically, cross so that the constructive fringes are above the ionization threshold, and we get the formation of plasma where we have high intensity, we have neutral gas where we have low intensity. These have different indices of refraction and create a great 121 00:24:19.060 --> 00:24:20.460 Mark Kushner: sitting like that. 122 00:24:20.460 --> 00:24:33.590 Mark Kushner: We can also do it ponderotively, so if we drive a grating, the electrons are expelled and pull the ions along with them. We'll get a fairly long-lived grating that can be used to diffract. 123 00:24:33.590 --> 00:24:46.909 Mark Kushner: It is important that the ions are coupled to the electrons. If it is a pure electron grading, it will be washed out by the probe beam that's as intense as the pump beam. What the ions allow us to do is, 124 00:24:46.910 --> 00:25:01.620 Mark Kushner: We can use relatively weak beams over a long period of time to move the ions, and then a short beam can be much more intense, and it basically does not have enough time to move the ions, so we can get a power ratio there. 125 00:25:01.860 --> 00:25:13.649 Mark Kushner: The other… the final approach is actually not a plasma approach, it's purely a gas phase, where we're going to heat up a gas very efficiently and create strong density modulations in a gas. 126 00:25:13.650 --> 00:25:21.940 Mark Kushner: And although, this gives us a somewhat weaker index modulation, it can be very useful for nanosecond duration beams. 127 00:25:22.110 --> 00:25:32.960 Mark Kushner: And typically, this is what we are examining most closely for high-energy nanosecond lasers, and these for high-power femtosecond lasers. 128 00:25:33.250 --> 00:25:43.370 Mark Kushner: Okay, so that's the basic introduction. Let's talk about how these things actually work in practice. So I'll start with gas gratings. 129 00:25:43.520 --> 00:26:01.969 Mark Kushner: The way that we do this is, we take a pair of ultraviolet beams and we cross them in a gas that contains a little bit of ozone. So ozone's a really good absorber of ultraviolet, it's why we don't get skin cancer every time we go outside, because it's up in the atmosphere and it absorbs sunlight very well. 130 00:26:02.020 --> 00:26:21.989 Mark Kushner: To give you a sense of how good an absorber of UV it is, if we take a frequency quadrupled YAG laser, so that's light at 266 nanometers, and we put it through 1 centimeter of atmospheric pressure gas that is 1% ozone, it will not come out the other side, okay? 131 00:26:22.100 --> 00:26:25.550 Mark Kushner: More than 99%, absorption. 132 00:26:26.010 --> 00:26:31.980 Mark Kushner: If that beam has a fair amount of energy in it, all of that energy has to go into the gas. 133 00:26:31.990 --> 00:26:42.250 Mark Kushner: And maybe you can guess, your expectation for what happens is if you deposit a bunch of energy in a gas, is it warms up. 134 00:26:42.250 --> 00:26:59.649 Mark Kushner: Okay, so, strictly speaking, what we're doing is we're photodissociating ozone. Most… a good fraction of the energy ends up as translational energy of these products. We're very efficiently coupling the UV energy into translational gas heating. 135 00:26:59.850 --> 00:27:15.919 Mark Kushner: The ideal gas law tells us that if we increase the temperature, at some point, that's going to lead to a reduction in density. So, this instantaneous, or close to instantaneous, increase in temperature 136 00:27:15.920 --> 00:27:23.380 Mark Kushner: Launches a pair of acoustic waves, as well as a non-propagating entropy wave. 137 00:27:23.450 --> 00:27:33.990 Mark Kushner: I think the animations are struggling a little bit over, over the connection, but it's basically an oscillation, a standing wave oscillation that is produced in the density. 138 00:27:33.990 --> 00:27:50.929 Mark Kushner: Now, what is useful for this over other ways of doing this is that this density modulation is extremely strong, so this can be up to something like 50% of atmospheric density, the amplitude of these oscillations. 139 00:27:51.050 --> 00:28:03.689 Mark Kushner: Period is, hundreds of nanoseconds, so it's very high frequency, very loud sound. It doesn't go very far, we don't hear it, you only hear the shock waves, later. 140 00:28:03.750 --> 00:28:11.900 Mark Kushner: But these are extremely strong density modulations, about as strong an index modulation as you can create in neutral gas. 141 00:28:13.050 --> 00:28:15.030 Mark Kushner: So, 142 00:28:15.350 --> 00:28:27.620 Mark Kushner: What that means is that if we send a probe beam through, an optic that's anywhere from a few millimeters to a centimeter or so thick, we can, in principle, diffract it off this density modulation. 143 00:28:27.770 --> 00:28:34.830 Mark Kushner: Fulfilling the same sort of brag, conditions that we had before. The scope of the… 144 00:28:35.010 --> 00:28:47.169 Mark Kushner: The fact that we get an index modulation around 10 to the minus 5 tells you that the length should be about 10 to the 5th, lambda, with, lambda of, 145 00:28:47.250 --> 00:28:55.450 Mark Kushner: one micron that gets you up to sort of the centimeter or so scale. So that's kind of how that works out. 146 00:28:58.020 --> 00:28:58.980 Mark Kushner: So… 147 00:28:59.060 --> 00:29:23.929 Mark Kushner: What this looks like experimentally, you can see that one of my students set this up, is we create, in the open laboratory in this case, a flow of ozone, the oxygen from which we produce the ozone from, and carbon dioxide in a co-flow geometry, and then put a couple of U-beam beams through it, and then put a probe beam at some other 148 00:29:23.930 --> 00:29:29.819 Mark Kushner: Wavelength, or at the same wavelength, through the grading, and ideally, it will all diffract off like this. 149 00:29:29.870 --> 00:29:42.449 Mark Kushner: Our imprint profiles are often not super good, it turns out not to be a huge problem, and we do, in fact, get these very strong fringe structures and a pretty nice grading structure in this… in this setup. 150 00:29:43.620 --> 00:29:44.890 Mark Kushner: So… 151 00:29:44.890 --> 00:30:08.599 Mark Kushner: We can measure what the response of the gas is. So, we hit the gas, and we'll see that the diffraction efficiency oscillates, as we expect from theory, as these acoustic waves propagate past each other. So it's a standing wave oscillation, anywhere from 10 nanoseconds to maybe hundreds of nanoseconds in period, which depends on the 152 00:30:08.600 --> 00:30:21.160 Mark Kushner: the speed of sound and the period of the grading. It… depending on how strongly we drive it, we can get some other features in there. But we have some control over it, and it… 153 00:30:21.250 --> 00:30:29.909 Mark Kushner: In fact, behaves kind of how we would expect from a theoretical perspective, which is always a positive sign on trying to build something. 154 00:30:33.030 --> 00:30:45.190 Mark Kushner: The… so, we can measure the diffraction efficiency as a measure of how much energy we inject into these gratings. So, it turns out that to get good diffraction efficiency. 155 00:30:45.190 --> 00:30:55.340 Mark Kushner: We're going to need to add something like a couple hundred millijoules per centimeter squared. So make a centimeter optic. We'll need a couple hundred millijoules. 156 00:30:55.340 --> 00:31:08.599 Mark Kushner: of nanosecond duration UV light. So, that's sort of… I mean, it's… I'm not going to say it's a tiny amount of energy, but it's not an enormous amount of energy compared to what we think we can control with these optics. 157 00:31:08.750 --> 00:31:26.359 Mark Kushner: So there's some interesting dynamics here. We don't need to go into all of them, but you can see here is evidence that if we drive the grading strong enough, we actually get diffraction back to the zeroth order of the beams. So we are able to reach the peak of the efficiency curve. 158 00:31:26.750 --> 00:31:38.919 Mark Kushner: And what that looks like in practice is that we can take a beam going in one direction, and we can get it to go in an entirely different direction, just by reflecting it off about a cubic centimeter of gas. 159 00:31:38.990 --> 00:31:57.050 Mark Kushner: So, these are experimental measurements of a beam somewhat downstream, so you can imagine that there's a screen sitting here. With the gas grating off, this is what the beam looks like at this position. As soon as we turn the grating on, it instead diffracts like this. 160 00:31:57.050 --> 00:32:20.000 Mark Kushner: I guess these should be mirror-imaged compared to this, and it goes in this direction over here, with what we have measured to be, in this case, up to 99% efficiency. So, 99% of the energy that came into the grating goes off in this other direction. It's still a beam, which is important, and it does maintain the properties of the instant beam. So, this is a gas behaving like an optic. 161 00:32:20.000 --> 00:32:22.180 Mark Kushner: At least once, in this case. 162 00:32:25.770 --> 00:32:43.389 Mark Kushner: These gratings have the nice property that they will work at different wavelengths. So, as long as you hit the grating at the right angle for the wavelength you're interested in, you can use them at 532, or 1064, or with femtosecond pulses at 800 nanometers. 163 00:32:43.390 --> 00:33:01.899 Mark Kushner: You can even use them at 266, which ozone does absorb really strongly, because there's a limited amount of energy that ozone can absorb. We want to use these for high-energy lasers. They have way more energy than ozone can absorb, so the amount of ozone absorption in ozone 164 00:33:01.900 --> 00:33:19.979 Mark Kushner: can be totally negligible for the probe beam coming through, and there isn't enough time for the gas to hydrodynamically respond to that energy deposition. So, the fact that ozone absorbs at this wavelength turns out to not be a problem at all. Maybe… maybe surprisingly, at least surprisingly to me. 165 00:33:20.510 --> 00:33:22.290 Mark Kushner: So… 166 00:33:22.590 --> 00:33:32.150 Mark Kushner: I said that we would talk about stability and robustness, so obviously we have to start thinking about what limits our ability to do this reproducibly. 167 00:33:32.150 --> 00:33:42.780 Mark Kushner: So, what you're looking at here is a measure of diffraction efficiency over a couple of hours of operation, with every individual shot collected at 10 Hz. 168 00:33:42.830 --> 00:33:45.159 Mark Kushner: And so… 169 00:33:45.310 --> 00:33:57.210 Mark Kushner: most of the time, we're getting efficiencies above 95%. We do have some, drops in efficiency, and this can actually, 170 00:33:58.140 --> 00:34:12.529 Mark Kushner: Maybe I'll say, considering how unstable the lasers that we are using to drive this grading are, this is shockingly good stability. So it turns out that the diffraction efficiency is actually slightly better than the stability of the driving lasers. 171 00:34:12.530 --> 00:34:23.379 Mark Kushner: Because we're sitting up at this peak here, and so small variations of energy, do not cause, cause slightly smaller variations in diffraction efficiency. 172 00:34:24.469 --> 00:34:36.810 Mark Kushner: So, even though we've got a gas flow that's moving around a little bit, we've got pump lasers that are not stable at the fundamental, and certainly are not stable when they've been quadrupled to 4 omega, 173 00:34:37.000 --> 00:34:48.019 Mark Kushner: The performance is not really so bad, and certainly we think that with decent pump lasers, this can be made to be extremely, extremely stable. 174 00:34:48.530 --> 00:34:53.550 Mark Kushner: So, I'll talk a little bit about applications. One of the… 175 00:34:53.670 --> 00:35:07.219 Mark Kushner: areas that we've gotten support to pursue these optics for is a general plasma problem, which is how might you build in an inertial fusion energy power plant? And so, NIF, which is pictured, operates with a yield of… 176 00:35:07.370 --> 00:35:11.400 Mark Kushner: You know, maybe up to… up to 10 megajoules, 177 00:35:11.540 --> 00:35:14.719 Mark Kushner: Optimistically, at one shot a day. 178 00:35:15.000 --> 00:35:34.340 Mark Kushner: An IFE plant would have to do something like 100 megajoules per shot at 10Hz, that kind of order. And so suddenly, we have a huge issue with what happens to these optics that face the interaction. So NIF puts a bunch of debris shields on it, and at a couple hundred shots a year, that more or less works. 179 00:35:34.380 --> 00:35:42.020 Mark Kushner: If you're doing a couple hundred shots in less than a minute, replacing debris shields is no longer a viable solution. 180 00:35:42.020 --> 00:35:56.009 Mark Kushner: Anything that we think about making glass optics from is going to degrade pretty quickly when exposed to the neutron flux that will be present around this target chamber. There's debris, shrapnel, X-rays. 181 00:35:56.050 --> 00:36:03.069 Mark Kushner: Backscattered laser light, any one of which would be an enormous problem for these target-facing optics. 182 00:36:03.410 --> 00:36:16.730 Mark Kushner: So, the advantage of a gas-based approach is that it allows us to remove all solid-state optics from line of sight of the target. So we can do non-line-of-sight focusing of a beam 183 00:36:16.730 --> 00:36:31.410 Mark Kushner: Onto a target, and in principle, that means that optics can be shielded, and the only thing exposed to neutrons is the gas, which is constantly renewed and is not sensitive to either cumulative damage or neutron exposure. 184 00:36:33.010 --> 00:36:33.840 Mark Kushner: So… 185 00:36:33.940 --> 00:36:42.819 Mark Kushner: That's, enough on the basics of gas gratings. Let me run through what we have on plasma ratings. So… 186 00:36:42.830 --> 00:36:46.270 Mark Kushner: We're gonna take a slightly different approach here. As I… 187 00:36:46.280 --> 00:36:58.299 Mark Kushner: Explained briefly before the ideas that we cross two femtosecond beams, we get ionization in the bright fringes, we get this modulated… these layers of plasma and neutral gas. 188 00:36:58.300 --> 00:37:07.539 Mark Kushner: They have different indices of refraction, and so a beam will diffract off them. The idea being that we're reproducing this, 189 00:37:07.540 --> 00:37:21.349 Mark Kushner: with something that looks like this with a much higher damage threshold. So this is a picture of the fluorescence you get from the plasma grating. You can see the structure there. Not super resolved, but it is definitely modulated. 190 00:37:21.510 --> 00:37:42.910 Mark Kushner: So, first way you might try to do this, or the way that this was originally done, was to do it in laboratory air, and just cross two beams. And you can… these are the trails that you get in air, the fluorescence from the plasma that's formed. You can see the two pump beams and the probe beam coming through. They make an X. 191 00:37:42.910 --> 00:37:51.510 Mark Kushner: It's pretty easy to set this, or relatively easy to set this up, but it is really hard to get any kind of diffraction efficiency. 192 00:37:51.510 --> 00:37:57.609 Mark Kushner: And basically, what you're trying to do is diffract off a grating that looks a little bit like a diamond shape. 193 00:37:57.610 --> 00:38:16.279 Mark Kushner: It turns out that that is neither a transmission grading nor reflection grading. Even in theory, it has really bad optical properties and is not going to give you what you want. You can sort of get 10% efficiency if you try really hard, but typically it's going to give you 1% or much less. So… 194 00:38:16.520 --> 00:38:38.019 Mark Kushner: Instead, we started exploring a series of geometries in gas cells and gas jets, where we get very clean front and back surfaces to the grating. So this allows us to create actual transmission gratings, and not just sort of a modulated plasma mess in the intersection of many different lasers. 195 00:38:38.020 --> 00:38:56.420 Mark Kushner: And as soon as we move to this kind of geometry, we start getting diffracted and undiffracted beams that look like this. I… I guess I didn't show the air-diffracted beams, but typically, light would spray everywhere in all sorts of crazy crescent shapes, and definitely not beam-shaped. 196 00:38:56.420 --> 00:39:03.059 Mark Kushner: So, sort of, start to get diffraction into beams from these kinds of geometries. 197 00:39:03.260 --> 00:39:09.470 Mark Kushner: It's tough to get enough plasma density to do this very well. 198 00:39:09.470 --> 00:39:34.459 Mark Kushner: So, we first tried this with a 4-micron probe. So, maybe you remember from your plasma physics class, you go to longer wavelength, the critical density goes down, so we can get closer to critical density more easily with a longer wavelength probe. And in this kind of geometry, we were able to get reasonably good diffraction. Now, for this particular experiment, this represents about 199 00:39:34.460 --> 00:39:39.320 Mark Kushner: 60% of the diffract… of the instant energy. So certainly in the… 200 00:39:39.590 --> 00:39:57.000 Mark Kushner: you know, non-trivial efficiency regime for optics, and maybe if you were going to use one of these on a beamline, this would be acceptable. Not maybe where we would like to get ultimately, but plasma mirrors only operate at about… or a double plasma mirror is only going to be about 60% efficient, so… 201 00:39:57.000 --> 00:40:01.300 Mark Kushner: You know, getting to the… to the efficiencies that are maybe relevant. 202 00:40:01.300 --> 00:40:05.460 Mark Kushner: It is much harder to do this at 800 nanometers. 203 00:40:05.760 --> 00:40:09.009 Mark Kushner: But we can. So, 204 00:40:09.180 --> 00:40:26.559 Mark Kushner: This is a interferometry image of a plasma grating created for an 800 nanometer beam. It's a little tricky to do interferometry of structures that are this small compared to the wavelength of the interferometry light. 205 00:40:26.560 --> 00:40:37.769 Mark Kushner: So we're a little… we're right at the edge of the resolution with what we can do with a setup like this. But you can see the fringe structures and the approximate, plasma densities associated with this. 206 00:40:37.940 --> 00:40:47.710 Mark Kushner: If we change the angles of instance, we'll get a change in the periods of these grading fringes. They do more or less behave, like we'd expect. 207 00:40:47.710 --> 00:40:58.299 Mark Kushner: These were pretty early ones, they're not very clean. This looks a little bit nicer, at least. We do improve as we go along and get more energy into these pulses. 208 00:40:58.370 --> 00:41:01.130 Mark Kushner: But the gradings kind of look like what we expect. 209 00:41:01.460 --> 00:41:07.670 Mark Kushner: And so now, at 800 nanometers, we are able to diffract into beans. 210 00:41:07.700 --> 00:41:20.309 Mark Kushner: We can do it on average at sort of the 35% to 40% efficiency, or we have done it. In these experiments, we do get single-shot efficiencies that are much higher. 211 00:41:20.310 --> 00:41:29.200 Mark Kushner: Which tells us that it is… it is possible to create a grating that is at much higher efficiency if everything happens to come out of the laser right. 212 00:41:29.200 --> 00:41:48.100 Mark Kushner: But these ionization gratings are much more sensitive to fluctuations in the laser than the neutral gas gratings. We're now using ionization, which is an extremely nonlinear process, rather than linear absorption, which is a linear process, and so small changes in intensities now have 213 00:41:48.100 --> 00:41:55.660 Mark Kushner: large effects on the grading that will form, and we think… see things bouncing around a lot more. Now, 214 00:41:56.000 --> 00:41:58.180 Mark Kushner: Some of this, 215 00:41:58.550 --> 00:42:08.250 Mark Kushner: Instability, we expect to get better as we go to higher laser energies. Our campaign on Zeus will maybe tell us whether or not that's true. 216 00:42:08.590 --> 00:42:14.249 Mark Kushner: So some of them are related to the limited amount of energy that we have available to make these gratings. 217 00:42:14.360 --> 00:42:22.930 Mark Kushner: But this is also at 10Hz, so it's sort of naturally well-suited to high repetition rate operation. There are no particular issues there. 218 00:42:23.720 --> 00:42:34.120 Mark Kushner: It also gives us very nice beams. So, these are beams that we measured coming off these plasma gratings, a reasonably smooth near-field profile. 219 00:42:34.120 --> 00:42:50.690 Mark Kushner: We were able to focus it to a spot of similar quality as what we put onto the grating, and we were able to put a pulse through the grating without substantially distorting its temporal profile. So we got out a pulse of basically the same shape as the one that we put in, in this case. 220 00:42:50.690 --> 00:43:04.749 Mark Kushner: So, they have good spatial and temporal quality, the diffraction angles follow analytic predictions, and they are… for those of you who work with high-power lasers, they have this other really nice property. 221 00:43:04.750 --> 00:43:15.530 Mark Kushner: Which is that, the plasma grating turns on very fast. Ionization is a fast process. It turns on within 100 femtoseconds or so. 222 00:43:15.530 --> 00:43:28.050 Mark Kushner: Which means we go from not having a grading to having a grading in a short amount of time. And so what you can think of this as is a very fast optical switch. It switches from off to on extremely quickly. 223 00:43:28.230 --> 00:43:40.279 Mark Kushner: Now, the advantage of this over something like a plasma mirror is that before we create the grading, there is no preferential direction associated with where the light will eventually go. 224 00:43:40.300 --> 00:43:55.539 Mark Kushner: A plasma mirror is a mirror surface, and so before you turn it into a plasma, it does send some light in the specular direction for that mirror. You can reduce that by putting an AR coating on it, but it still has a surface pointed in that direction. 225 00:43:55.800 --> 00:44:12.489 Mark Kushner: Here, before we create the grating, we just have a gas jet. There's no, there's nothing special about the… what will ultimately be the diffracted beam direction. So we get, basically no light in the direction of diffraction before we turn the grating on. 226 00:44:12.640 --> 00:44:29.049 Mark Kushner: This makes it very good at improving the contrast of a high-power laser beam. So here, we took a deliberately spoiled laser pulse, and we were able to improve the contrast by 5 orders of magnitude, up to about, 227 00:44:29.170 --> 00:44:41.340 Mark Kushner: well, we could only measure about half a picosecond before the pulse arrived, effectively entirely removing the pedestal of our laser. So, an efficient and high repetition rate 228 00:44:41.340 --> 00:44:52.280 Mark Kushner: you know, each one of these, this is a… these are single-shot measurements taken in a trace. It's quite hard to do this with a plasma mirror, but able to improve substantially the… 229 00:44:52.510 --> 00:44:54.430 Mark Kushner: Quality of this laser pass. 230 00:44:55.270 --> 00:44:56.090 Mark Kushner: So… 231 00:44:57.070 --> 00:45:05.099 Mark Kushner: I said I'd talk… I said we'd talk about why these are useful for building lasers, so a reminder that this is what shirt pulse amplification looks like. 232 00:45:06.580 --> 00:45:16.919 Mark Kushner: The first place to start is to think about whether we can replace this optic, because even replacing this final grading will give us some improvements in system capabilities. 233 00:45:17.270 --> 00:45:20.310 Mark Kushner: So… 234 00:45:20.820 --> 00:45:38.699 Mark Kushner: Volume transmission gratings, these diffractive gratings, are dispersive. If we send in a beam that has different wavelengths in it, they'll all diffract at slightly different angles. Now, this dispersive effect gets stronger the larger the angle that we're coming in at. 235 00:45:38.700 --> 00:45:43.299 Mark Kushner: And it gets weaker and goes to zero if we're incident at a very small angle. 236 00:45:43.520 --> 00:46:01.570 Mark Kushner: Now, the issue we face with plasma gratings is that they have some kind of spectral bandwidth. So, the angle of instance for which they're efficient depends on wavelength, so if your wavelength gets too far away from that angle, you won't diffract from the grading anymore. They have a limited bandwidth of light they can diffract. 237 00:46:01.800 --> 00:46:22.009 Mark Kushner: As you go to large angles of instance, that spectral bandwidth gets very narrow, so you cannot diffract very many wavelengths. The ones you are diffracting are well spread out, but you can't diffract very many of them. As you go to smaller angles, you can diffract all the wavelengths, but they basically don't change their angle very much. 238 00:46:22.010 --> 00:46:23.160 Mark Kushner: So… 239 00:46:23.160 --> 00:46:39.559 Mark Kushner: We have this issue where in order to build a good compressor, we need a reasonable amount of dispersion. They do need to go in different directions, otherwise our grading will have to be… or our compressor will have to be very long in order for the colors to have traveled different distances. 240 00:46:39.760 --> 00:46:58.510 Mark Kushner: But if we don't have very much plasma density to work with, we're going to be limited in the spectral bandwidth that we have. So, we want to get plasma density as high as possible to make this easier, but we have this fundamental trade-off between spectral bandwidth and dispersion in this type of optic. 241 00:46:59.870 --> 00:47:03.740 Mark Kushner: What this leads us to initially… 242 00:47:03.770 --> 00:47:09.949 Mark Kushner: Is an architecture that looks like this, where we're only trying to do, 243 00:47:09.970 --> 00:47:16.259 Mark Kushner: part of the compression with our plasma compressor. So, we couple a traditional 244 00:47:16.260 --> 00:47:22.679 Mark Kushner: Compressor that produces pulses that are, say, tens of picoseconds long, rather than femtoseconds, so… 245 00:47:22.680 --> 00:47:36.969 Mark Kushner: You know, a 10 picosecond pulse with the same energy as a 10 femtosecond pulse will be a thousand times lower intensity. And then do the compression from tens of picoseconds down to tens of femtoseconds in the plasma stage. 246 00:47:37.230 --> 00:47:56.440 Mark Kushner: we're able to, measure the dispersion of our gradings. They actually, agree relatively well with, what we'd expect, theoretically. And this is measurements that student Victor took of, the 247 00:47:56.550 --> 00:48:06.799 Mark Kushner: Direction that light goes in after it leaves the grating as a function of its wavelength compared to our theoretical expectation of this dispersion curve. 248 00:48:06.930 --> 00:48:20.119 Mark Kushner: So this tells us that these gratings are dispersing light, and they're doing it more or less linearly in kind of the directions that we'd expect. So in order to build a compressor. 249 00:48:20.120 --> 00:48:28.089 Mark Kushner: We have to turn this process around and put in light that has angular dispersion, and use the grading to remove that angular dispersion. 250 00:48:28.090 --> 00:48:34.869 Mark Kushner: So it's easier to do it this way, more useful to do it the other way around, which is what we are working towards. 251 00:48:35.580 --> 00:48:49.099 Mark Kushner: So, to give you a sense of, operating parameters, what we need to do this, or this… we can write down an equation that relates 252 00:48:49.210 --> 00:49:02.519 Mark Kushner: The amount of distance we need between our gratings to the pulse duration that we want to compress, and the bandwidth of our laser, and the diffraction angle of the grating. 253 00:49:02.770 --> 00:49:09.810 Mark Kushner: Which is basically saying that if we have a certain amount of bandwidth, 254 00:49:09.980 --> 00:49:19.739 Mark Kushner: We can, go up to these red, or we can be up to the red line in terms of plasma density modulation we can achieve. 255 00:49:19.800 --> 00:49:38.689 Mark Kushner: And the black lines will tell us how long the compressor needs to be for each picosecond of pulse duration. So, here, this is basically saying that if we only have 10 to the minus 3, then to compress a 10 picosecond pulse, we'll need a compressor 3 meters long. 256 00:49:38.690 --> 00:49:40.620 Mark Kushner: Which is kind of reasonable. 257 00:49:40.620 --> 00:49:47.360 Mark Kushner: not reasonable if you want to compress a nanosecond pulse. And this is what drives us towards doing, 258 00:49:47.360 --> 00:50:08.339 Mark Kushner: say, 10 to 100 scale picosecond scale compression, but maybe most importantly, it's what's setting the constraint on the magnitude of the index modulation, and therefore the plasma density we need to achieve. That's why we need to get up to 10 to the minus 3 to 10 to the minus 2, 259 00:50:08.360 --> 00:50:17.259 Mark Kushner: Relative index modulations, corresponding to, say, 10 to the 19 or so plasma densities. 260 00:50:17.480 --> 00:50:25.820 Mark Kushner: Okay, that's maybe a little… Detailed, and we can skip to some more fun stuff than details on compressors. 261 00:50:26.120 --> 00:50:43.629 Mark Kushner: In the final few minutes I have here today, I'm going to discuss the few results that we have on holographic lenses. So, the idea is that we can take two pump beams. If instead of just crossing them at an angle, they have different focusing. 262 00:50:43.630 --> 00:50:55.479 Mark Kushner: Here, in a collinear geometry, we're going to end up with an interference pattern that looks like a zone plate, so our plasma mechanism is going to produce rings of plasma rather than linear gratings. 263 00:50:55.550 --> 00:51:01.980 Mark Kushner: If we've done that properly, and we send the probe beam through, we can either focus or collimate with an optic like this. 264 00:51:02.360 --> 00:51:13.980 Mark Kushner: So, we are working on producing these in both plasmas and gases. We've, gases were a little easier to do first, so that's where… 265 00:51:14.210 --> 00:51:30.530 Mark Kushner: student… student Dave started with a geometry that looks like this. It turns out that an off-axis configuration is both easier to implement and more efficient than an on-axis configuration, because we… 266 00:51:30.550 --> 00:51:42.230 Mark Kushner: Basically, we don't have as much variation in period across the grading, and that turns out to be a positive. The idea here is that we have two pump beams that have slightly different focal lengths. 267 00:51:42.230 --> 00:51:52.700 Mark Kushner: They'll create grating fringes that look like this, and a probe beam coming off it should be focused. So, undiffracted beam will continue straight, diffracted beam will be focused to a spot that looks like this. 268 00:51:52.870 --> 00:51:56.470 Mark Kushner: And so Dave set up an experiment where he measured this. 269 00:51:57.310 --> 00:52:13.290 Mark Kushner: And so what this is, is it's a camera sitting in each of these beams and scanning through to get what the beam looks like as a function of Z. The undiffracted beam is near-collimated, so it travels… or the instant beam is near-collimated. It travels in a straight line with the grading off. 270 00:52:13.560 --> 00:52:28.970 Mark Kushner: As soon as we turn the grading on, if we do this scan of what the beam profile looks like, we get something that looks like this. So now the beam is focused to a focal spot. It turns out to be a pretty nice focal spot at a position a little bit downstream. 271 00:52:28.980 --> 00:52:34.720 Mark Kushner: So these are all experimental measurements. We went from a beam that looks like this to one that looks like this. 272 00:52:37.900 --> 00:52:57.540 Mark Kushner: We can, by changing the focal lengths of the pump beams, tunably adjust the focal length of the lens. So, here the instant beam probe beam isn't changing, we're just changing the pump beams, and we can go from a variety of focusing configurations to here, a defocusing lens configuration. 273 00:52:57.540 --> 00:53:02.690 Mark Kushner: And make the beam spread out as if it were a concave lens rather than a convex one. 274 00:53:03.260 --> 00:53:11.250 Mark Kushner: And again, the positions of these focal spots are pretty close to our theoretical expectations for where they should be. 275 00:53:13.940 --> 00:53:20.309 Mark Kushner: We can also run this in a collimating geometry, so if we bring a beam in that's tightly focusing. 276 00:53:20.310 --> 00:53:40.899 Mark Kushner: And you can imagine that this might be useful if, for example, you want a very long, a large F-number experiment, and you have limited lab space, maybe less of an issue for you here, but certainly for labs like ours, lab space is not infinite, and you might want a 50-meter focal length and have a 5-meter lab. 277 00:53:41.190 --> 00:53:54.150 Mark Kushner: With these lenses, we can stick them much closer to focus than a conventional optic, and turn something that looks like this into a collimated beam that looks like this. So, this is, 278 00:53:54.150 --> 00:54:13.700 Mark Kushner: something like a couple hundred millijoule diffracted beam that we brought to a collimated beam of about a millimeter in diameter. So, at least none of the optics we have particularly survive in this environment, but the gas optic does it quite nicely, with reasonably nice focal spots. 279 00:54:14.340 --> 00:54:29.300 Mark Kushner: So, these optics have pretty good pointing stability. I think this is kind of a video of the beam coming off one of these optics. This is pretty similar to what the pointing stability of the beam coming onto the optic is. 280 00:54:29.310 --> 00:54:41.840 Mark Kushner: It's actually, in one axis, slightly better due to the limited angular bandwidth of the optic. These are the focal spots moving around in terms of, 281 00:54:41.930 --> 00:54:53.810 Mark Kushner: in this case, fractions of the spot size. So, here, the pointing stability is about a tenth the spot diameter, which is basically what it is in the instant. 282 00:54:55.250 --> 00:54:56.250 Mark Kushner: Okay. 283 00:54:56.410 --> 00:55:20.259 Mark Kushner: And then maybe finally, I'll just comment that we can also do this with femtosecond pulses. So there are some different damage threshold limits. For nanosecond pulses, our expectation is that these gas lenses, or experimental measurements, indicate that they can take about a kilojoule per centimeter squared, so that's a lot more than the 100 millijoules per square centimeter that it took to write them. 284 00:55:20.410 --> 00:55:26.209 Mark Kushner: And for, femtosecond pulses, we can sort of reach ionization-relevant intensities. 285 00:55:26.320 --> 00:55:29.990 Mark Kushner: Okay, so… With that… 286 00:55:30.080 --> 00:55:38.939 Mark Kushner: I think I've reached the end of my time. I'd like to acknowledge the variety of people who have contributed to various aspects of this work. 287 00:55:38.950 --> 00:55:54.310 Mark Kushner: Although I didn't talk about what we're working on with Alec today, there are parts of this that are related to it. A few of my team members will be here, hanging around for the next few weeks at Zeus, and you're welcome to go, 288 00:55:54.310 --> 00:56:02.189 Mark Kushner: ask them any questions that I don't answer to your satisfaction, and as well as the funding agencies that supported this work. 289 00:56:02.530 --> 00:56:08.550 Mark Kushner: So with that, I think I will leave up this summary, and I'd be happy to take any questions. 290 00:56:18.690 --> 00:56:25.620 Mark Kushner: I've got a horrible thought that I've actually asked you this before, but, is there any possibility of using, 291 00:56:25.950 --> 00:56:34.540 Mark Kushner: very high-power ultrasound for shaping the biometric gas. Yeah, so there was a… there was a paper on that a couple years ago, 292 00:56:34.690 --> 00:56:44.510 Mark Kushner: To give you a sense, ultrasound… the ultrasound that they were able to achieve was, those are modulations of about 10 to the minus 7. 293 00:56:44.620 --> 00:56:54.149 Mark Kushner: So they basically had to go through 7 times through a centimeter… 7 centimeter long optic to get something like 40% efficiency. 294 00:56:54.340 --> 00:57:05.369 Mark Kushner: That means that the… the bandwidths are extremely small. You… I would argue that you can't really do anything useful with an optic that weak, probably. It's, 295 00:57:05.900 --> 00:57:12.060 Mark Kushner: It's very limiting on the amount that you can turn a beam and the amount of spectral bandwidth that you can get. 296 00:57:12.100 --> 00:57:29.200 Mark Kushner: The acoustic waves here are something like the equivalent of 180 decibels, you know, so that's… that's really loud for sound, and that would be hard to do with transducers. I'm not sure that's at all possible to do with transducers. 297 00:57:29.200 --> 00:57:39.640 Mark Kushner: They are, at this frequency, they're very heavily damped, so they're not gonna propagate very far. We're writing them in place, but they're not gonna… 298 00:57:39.640 --> 00:57:49.320 Mark Kushner: you wouldn't be able to propagate them across a centimeter scale beam diameter. So, I guess the answer is, in principle, the same optics apply. 299 00:57:49.340 --> 00:57:56.839 Mark Kushner: But I don't see a pathway to getting the index modulations that are required for the applications that we're thinking about. 300 00:58:00.790 --> 00:58:13.580 Mark Kushner: Yeah, so it seems like some of these techniques are working, like, surprisingly well so, so soon, and I'm just curious, has anybody, purchased this technology, using it in an application yet, or is it still just all R&D? 301 00:58:13.580 --> 00:58:24.050 Mark Kushner: It's all R&D. The first demonstration of the gas diffraction gratings was from a group in Japan in 2020, so it's pretty recent. 302 00:58:24.050 --> 00:58:30.470 Mark Kushner: stuff. Certainly, there's… Interest, maybe. But… 303 00:58:30.640 --> 00:58:47.959 Mark Kushner: as much as we're pushing towards, like, stability and robustness, this is still very much like a laboratory tabletop experiment. You need an ozone generator to manage gas flows. It's not… it's not like a piece of glass that I can just stick in an experiment. There's… 304 00:58:47.960 --> 00:58:52.600 Mark Kushner: some infrastructure that goes along with it. So… 305 00:58:54.380 --> 00:59:04.329 Mark Kushner: think it's reasonable to think about in the future, but it's not, like, trivial to set up. Yeah, I guess, related to that, though, like, let's say you could get, you know, 90… 306 00:59:04.480 --> 00:59:13.830 Mark Kushner: 9% efficient and really reliable. Has there been any studies on, like, how much impact that could have on, say, like, a future IFE driver or something? 307 00:59:14.050 --> 00:59:19.379 Mark Kushner: I certainly, if you look at the designs that are out there, 308 00:59:19.690 --> 00:59:26.239 Mark Kushner: There'll be speculative stuff on, like, if we have optics of this capability, what does it mean? 309 00:59:26.750 --> 00:59:35.070 Mark Kushner: Some, some companies certainly are pursuing designs that have very small aperture beams, coming through. 310 00:59:35.200 --> 00:59:40.720 Mark Kushner: So I… you probably have to ask around, 311 00:59:41.270 --> 00:59:51.750 Mark Kushner: Companies on exactly how much they've thought about, you know, what their solution is, versus hope that there will be a solution by the time it comes around to actually building the reactor. 312 00:59:51.960 --> 00:59:53.520 Mark Kushner: The advantage… 313 00:59:53.680 --> 01:00:09.639 Mark Kushner: here, or one of the advantages, is because the damage threshold appears to be kilojoules per centimeter squared, compared to, say, 10 joules for conventional optics, you could take something like a NIF beam, which is… 314 01:00:09.870 --> 01:00:26.629 Mark Kushner: 40 centimeters by 40 centimeters, and the optics required to diffract that using this approach were only 3 centimeters by 3 centimeters. So now, instead of a solid angle this that you're taking out of a chamber, you have a much smaller 315 01:00:27.090 --> 01:00:46.970 Mark Kushner: port requirement. You can use more of your reactor wall to capture neutrons, rather than reserve for instant lasers. So I… I do think that some approach like this is necessary, for an IFE power plant, whether or not it is currently built in to the… or whether or not 316 01:00:47.220 --> 01:01:05.519 Mark Kushner: reactor designs have gotten to the stage to actually consider the problems, associated with final optics in a very serious way. Yeah, I mean, growing glass that big, right? Optics… well, I'm not… So for glass, it's okay. I'm just wondering if there's, like, a big cost savings of, like, a… 317 01:01:05.840 --> 01:01:18.449 Mark Kushner: I… so I'd argue that it's mostly coming from the fact that glass won't survive. Like, if you only had to make a meter-scale glass lens once, it's expensive, but not prohibitively, so… 318 01:01:18.760 --> 01:01:31.540 Mark Kushner: But, you know, you expose it to neutrons in an IFE environment for a week, it's not going to be glass anymore, or not useful glass, anyway. So that… that's kind of more of the issue, rather than… because… 319 01:01:32.030 --> 01:01:43.130 Mark Kushner: It's a little bit different for gratings, which… those are expensive, but, like, a glass lens is a little bit more manageable than sort of a 10-meter scale grating for a… 320 01:01:43.840 --> 01:01:45.989 Mark Kushner: A high peak power system. 321 01:01:51.430 --> 01:01:58.339 Mark Kushner: Yeah, so for ionization grading, you need kind of longer pulse duration, whether 322 01:01:59.720 --> 01:02:06.580 Mark Kushner: picosecond, maybe 10 picoseconds for ions to move. So, you need a separate compressor for this, or what? 323 01:02:06.580 --> 01:02:22.559 Mark Kushner: So we… for these gradings, we're not relying on the ions moving. So we create the electrons locally, and basically we're relying on the fact that they don't move before we use them, because we're creating them in some places and not in other places, and we just leave it like that. 324 01:02:22.630 --> 01:02:26.500 Mark Kushner: For the ponder-motive gratings, where the ions do move. 325 01:02:26.550 --> 01:02:43.090 Mark Kushner: That requires a longer pulse beam to drive it. We… it's beyond our local experimental capability. In the ideal case there, you're looking at pulses that are picoseconds long to drive the gratings, at least computationally. 326 01:02:43.280 --> 01:02:54.040 Mark Kushner: you know, there's maybe a gap between computation and experiment, but you would be looking at longer pulses to create the grading, and so that would be a separate beam. I… 327 01:02:54.040 --> 01:03:05.099 Mark Kushner: I would sort of view that in the long run, if these were built into the system, you would not be doing what we are currently doing in Zeus, which is, you know, picking off parts of the beam after the compressor. 328 01:03:05.100 --> 01:03:12.899 Mark Kushner: You would upstream, further upstream somewhere, be picking off dedicated beam lines to make these, to make these optics. 329 01:03:13.090 --> 01:03:20.129 Mark Kushner: That it, you know, would be a separate, dedicated part of the laser system that was not related to the science driver. 330 01:03:20.750 --> 01:03:30.090 Mark Kushner: But obviously, we can't, reconfigure a large-scale facility doing that without any evidence that it's actually going to work. 331 01:03:33.040 --> 01:03:35.569 Mark Kushner: What do you think the shortest wavelength 332 01:03:35.710 --> 01:03:38.159 Mark Kushner: You'll pass barometers to work for. 333 01:03:39.850 --> 01:03:52.659 Mark Kushner: Well, it's a good question. We have been trying them at very short wavelengths, but at some point, you run into issues with absorption. And so in the EUV and XUV, 334 01:03:53.050 --> 01:04:02.319 Mark Kushner: In principle, the physics still holds, but you've got to select the right material and wavelengths to work with. 335 01:04:08.400 --> 01:04:10.990 Mark Kushner: Well, let's thank Matthew again. 336 01:04:15.320 --> 01:04:19.960 Mark Kushner: Yes, this week and next week, so… 337 01:04:28.640 --> 01:04:35.919 Mark Kushner: We have on-campus words, or… Yes, yes, we do.