您当前的位置: 首页 > 网页快照
Redlining lasers for nuclear fusion | Nature Photonics
The National Ignition Facility at Lawrence Livermore National Laboratory reported over 1.3 MJ output, representing 70% of both input laser energy and official ‘fusion ignition’. Operations manager, Bruno Van Wonterghem, delves into the optics and what to expect next. You have full access to this article via your institution. Download PDF Download PDF How did you become the NIF (National Ignition Facility) commissioning manager in 2001? . I was originally hired to commission the Beamlet laser project. At that time there was an inertial confinement fusion advisory committee looking at what would be required to get a recommendation to go forward with the NIF project. Part of that was a Nova technical contract, to demonstrate some physics and performance. The other part was demonstrating the laser technology that would be the basis for NIF. Credit: Lawrence Livermore National Laboratory Credit: Lawrence Livermore National Laboratory The Beamlet project was basically a one beamlet demonstration of the [192 beam] NIF laser architecture, using a single large multipath amplifier, square beams, new pulse shaping technology, very high fluence, and UV operation. We had a very short timeline of two years to demonstrate that we could generate 3.5 kJ, 351 nm light, in a 35 cm × 35 cm beam. That’s where I was thrown in. I started at the oscillator, I worked my way to the injection laser, to the main amplifier, and basically through the whole system. Did the NIF design change from the original plan? . There was a lot of work done to establish the mission needs and to get the support in congress. Very early on there was more than one plan. One option was upgrading the Nova laser, making it a 20 beam system. There was also a plan to build a larger system inside the Nova building. Lastly was a plan to build a new 192 beam system — at that time there was even a 240 beam design. Was the original mission energy production or more on fundamental physics? . At that time it was mostly about fundamental understanding and supporting the nuclear stockpile stewardship, but there was always an opening towards energy. Over time the approach has changed a little bit. What’s the difference between building the laser system for NIF compared to other lasers? . I’m used to working on lasers where you can get your hands on the beam and actually measure and adjust every component. In NIF you have no flexibility, no access; the beams are suspended 18 ft in the air inside a steel, vacuum enclosures. It’s a totally different approach that feels more like working with a satellite, because the lasers could well be up in space from my point of view, since you didn’t have access to it. That requires a complete change in mindset. Even Nova was built like a large laser laboratory; it was big but you could go and ‘stick your head in the beam’ and see what’s going on. We had to think about means of aligning beams that were totally different, using new tools; very procedural and very systems engineering based. At NIF we do careful metrology of optical components so that we do not have to adjust… the only components we adjust are the laser mirrors; they are corrected for every shot. We have developed a procedure that almost commissions the laser when you operate it every day. In the morning you start an alignment process for the beams and most of that process is now done automatically such that the lasers are properly pointed and centred. What problems came up during the commissioning on NIF? . NIF was amazingly easy [to commission]. There was a lot of engineering and a lot of work was done ahead of time, making sure that all of the components were well understood and tested. There were problems, but I wouldn’t say there were big problems; they were interesting problems. What type of optics problems were solved for NIF by the Beamlet project? . We recognized a number of problems on Beamlet that were dealt with in an engineering way on NIF. One of the major failures was learning how to ruin a spatial filter, causing the vacuum spatial filter to implode, then explode and ruin a whole laser in a nanosecond. I managed to do it not only once, but twice; I found two failure modes. One of them was related to damage on the lenses which was caused by contaminants in the vacuum absorbing into the colloidal coatings, which lowered the damage threshold. We figured out how to control those contaminants by using absorbing materials on the spatial filters, which we implemented on NIF. Every spatial filter has something like a litter box which collect organic contaminants. One of the mysteries that started in the Beamlet multipath amplifier project was that little beams showed up and you don’t know where on earth they came from. They didn’t cause damage but it turned out that in multipath amplifiers you can have reflections from lenses that illuminate the pinholes of a spatial filter and create little beams that are capable of propagating back and forth in the multipath amplifier and create these little annoying beams. I was able diagnose from the time history of these pulses where they originated in the system. Then we could use filtering lenses to make sure those beams were outside of the field-of-view of the amplifier aperture. Another point that we studied on Beamlet was how much power we could push through the system before the beam starts to self-modulate, break-up and damage components. That helped us evaluate how good our optics for NIF needed to be; these are extremely critical parameters for NIF because we used optics that had new types of coatings, finishes and treatments in order to handle the high fluences. From our point of view the optical components are the critical component that enable NIF to run. How is adaptive optics crucial to NIF? . In the beamlet experiment we had non-uniform gain; the wavefront aberrations were horrible. The far-field looked like a butterfly and the near-field was sloping nearly 50% because of gain variations. There was amplitude modulation in the beam. We implemented compensation techniques on Beamlet, and later on NIF, with modern electro-optical technologies. For example to do spatial beam shifting… we know that the amplifiers are not perfect, so what we did is adjust the beam shape that we inject to compensate for that, and to make sure that at the output of the laser — where the energy is extremely high — that we have a very flat beam profile, so that we can extract the maximum possible energy. When we compensated like this we obtained a nearly perfect spatial profile. The technique is similar to what is used to compensate for atmospheric variations with telescopes; there was a guide star developed using a deformable mirror, about 10 cm x 10 cm, that was later moved to the Keck telescope. We did this for wavefront aberrations and then we did the same thing for phase compensation so we were able to create a laser that had almost perfect spatial, phase and temporal profiles. These techniques made it possible to create a laser that can really run at the limit of what the materials can handle, which is how NIF was designed; NIF was designed to run at the red line. Was the recent >1.3 MJ result enabled by optics, or by other factors like the target capsule? . It’s a combination of factors…on the final shot the change that we made was reducing the aperture of the hohlraum to reduce radiation losses and couple more energy into the capsule. Then, for a given laser pulse we could lower the power and extend the pulse a little, which is important because in the time after the laser pulse stops, the capsule starts to cool down. There was previously a belief that this time doesn’t matter because the force is so large on such a small capsule, but it turns out there is more physics related to that final phase. It is important to keep pushing on the capsule and I believe it made a large difference on that last shot. We don’t know for sure because we still have to do a number of experiments to disentangle the various effects, because at the same time the capsule was almost perfect through a number of developments. When might NIF reach fusion ignition? . I think that it could happen within 6 months to a year. We are now in a regime where small changes may have a big impact on the outcome. What makes me so optimistic is the fact that we are not optimized on so many difference areas, from the laser to the capsule we have room almost everywhere — to improve the quality of the implosion, the symmetry, the fuel density, and the energy we can couple in — to reach that additional 30% we need in order to reach the definition of ignition of the National Academy of Science, which is when the fusion energy equals the laser drive. I’m actually pretty optimistic; our typical cycle is on the order of two years if you look at our different campaigns but the last stage we achieved in 6 months…we have enough drive to achieve ignition, and probably even higher yields. Will laser driven fusion be viable as an energy source and when might that be possible? . A significant leap is required, from driver and target technology, to how we actually capture the neutrons. It doesn’t feel like it’s impossible now, but we may be many decades away from demonstrating a real power plant. I don’t see a fundamental reason why it can’t be done. We shouldn’t under estimate the challenge though, at every level; on the driver side, demonstrating lasers with 80% efficiency and high repetition rate, to the target technology, and survival in that extremely harsh environment. Once we demonstrate ignition, the impetus will be strong; the National Academy of Sciences has made a recommendation to, following ignition, start an ‘honest effort’ [on energy production applications]. The Department of Energy is already preparing; the Office of Science has planned a workshop for next year. People are starting to lay the groundwork and prepare. For our mission as a nuclear stockpile steward, it provides incredible new capability to support the current life extension program and nuclear modernization efforts. For example in terms of being able to test neutron survivability of systems, and also providing insight into a system that the DoE is looking at — a very high yield facility that will support hundreds of MJ, or even up to a GJ, of yield. I think NIF will be able to go a long way towards establishing what it will take to develop such a facility. That is the other part of our specific mission that drives us. The ITER Tokamak nuclear fusion reactor being built in southern France has explicit energy production goals right? . ITER is really being built as a power plant demonstration with the goal of heating the plasma with tens of MW and obtaining several hundred MW, which is quite different to NIF which was developed to investigate thermonuclear fusion in the laboratory. ITER is several years away from starting to be commissioned and then about 10 years away from when we can conduct experiments. It will be the biggest magnetic based fusion facility that has ever been built. NIF and ITER are both at the edge of what is now possible in engineering and technology. ITER will use magnetic confinement for a continuous plasma in which fusion takes place, but NIF uses inertial confinement to generate bursts of energy from a series of discrete fusion reactions. Is one approach better than the other? . I don’t think so. They both have advantages and disadvantages, but there isn’t anything that you specifically say puts one in front of the other. They both have serious challenges and they both have benefits. I can’t say that one would be better; they are totally different. In magnetic confinement fusion you confine the plasma for a very long time — seconds to minutes — as opposed to inertial confinement fusion where the whole event takes place in a few billionths of a second and you gain energy by repeating the process. With each approach you have to deal with scaling-up, instabilities and, ultimately, controlling a plasma. How do you deal with running at the damage limit of optical components? . NIF is one of the only lasers that runs above the damage threshold for the UV optics. By design, we damage optics. We developed a system that is capable of monitoring every optical component in the final optics and detect any incipient damage that is created as we operate the laser. The first thing we do once we detect damage on an optic is use a spatial light modulator (SLM) to put a ‘black patch’ over that position of the optic, blocking the beam, which stops the damaged spots from growing. We can block up to about 40 spots on an individual optic; when we have too many spots the optic goes to a mitigation facility where a CO 2 laser melts the damaged spots. We put that optic back into the system, until this recycling occurs a number of times, and then we have to discard the optic. NIF really runs on the ‘red line’ doesn’t it? . That is true. We need to run in a place where nonlinear growth of the modulation in the beams escapes us, and starts to become so large it starts to self-focus and beings to basically drill holes in the final optics. To control that — and figure out how we can get even more power out of the lasers — is really important as we try to go up to yield levels of 2.2 MJ and 2.6 MJ, and maybe a little bit over 500 TW. So we continue to research exactly where the ‘red limits’ are. We are also looking at means to protect from back scatter from the target; in certain conditions the plasma on the hohlraum can send light back into the beamlines which can cause damagemachine safety is a continuous concern for us. We review every experiment to make sure we don’t take undue risks, so we can continue to do new experiments with acceptable machine risks. What is the lifetime of NIF? . We have a sustainment project to deal with wear and tear in the facility. We have seen contamination in the amplifiers, and there are target chamber improvements we can make to protect our final optics. This will help us make sure that we get the support to extend the lifetime of NIF. We are close to saying that NIF is the facility that will reach ignition, and even intermediate yields, with the current technology, as long as we make sure that the facility is maintained and make moderate investments to improve the capability, to improve the pulse shaping, to make small upgrades in in power and energy, to provide more diagnostics to help gain insight in what it takes to get to this more robust ignition. For the next 10–15 years NIF will be the only game in town. Author information . Affiliations . Nature Photonics David F. P. Pile Authors David F. P. Pile View author publications You can also search for this author in PubMed ? Google Scholar Corresponding author . Correspondence to David F. P. Pile . Rights and permissions . Reprints and Permissions About this article . Cite this article . Pile, D.F.P. Redlining lasers for nuclear fusion. Nat. Photon. 15, 863–865 (2021). https://doi.org/10.1038/s41566-021-00917-5 Download citation Published : 29 November 2021 Issue Date : December 2021 DOI : https://doi.org/10.1038/s41566-021-00917-5 Share this article . Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative .
From:
监测目标主题     
(1)  
(1)  
(1)  
(1)  
系统抽取主题     
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)