Podcast appearances and mentions of spencer klein

  • 14PODCASTS
  • 18EPISODES
  • 57mAVG DURATION
  • ?INFREQUENT EPISODES
  • May 30, 2023LATEST

POPULARITY

20172018201920202021202220232024


Best podcasts about spencer klein

Latest podcast episodes about spencer klein

MoFo Perspectives Podcast
Above Board: The Changing Landscape of Shareholder Activism

MoFo Perspectives Podcast

Play Episode Listen Later May 30, 2023 22:59


In this episode of the Above Board podcast, Morrison Foerster partner and host Dave Lynn discusses shareholder activism with the firm's global Mergers & Acquisitions Group co-chairs, Spencer Klein and Brandon Parris. They discuss the key shareholder activism developments that boards of directors should be considering in 2023, including the latest developments with proxy contests and M&A activism, the rise of the “occasional activist,” the use of ESG considerations in advancing campaigns and the recent implications of the SEC's universal proxy rule.

On the Dogwatch
22. The Golden Age of Seiko with Spencer at Klein Vintage Watch

On the Dogwatch

Play Episode Listen Later Feb 7, 2022 76:00


Today On the Dogwatch we turn our attention to the appreciation of the Seiko watch, and how much satisfaction one can get from taking them apart, fixing them, and putting them together again as new. We speak with Spencer Klein at Klein Vintage Watch and discuss his journey into the watch world from the moment he traded a t-shirt for a Girard-Perregaux, some of the history of the Seiko company during its golden age, and how Spencer's craft and business have developed into what it is today. Before we begin, our feature for today is a   dog named Sport who is a cross between a ShihTzu and Yorkshire Terrier–a Shorkie. This cross produces small and affectionate dogs that top out at about a dozen pounds, good for cuddling and laps, and for accommodating life in the big city, like Sport does in the Big Apple. Sport, like the Dogwatch, is a fan of the Michael J. Fox Foundation, and we encourage you to learn more about them and join in helping accelerate the next generation of treatments for Parkinson's disease.

Beyond the Legal Lens Podcast
Episode 20: Spencer Klein

Beyond the Legal Lens Podcast

Play Episode Listen Later Dec 21, 2021 16:41


It is a pleasure to have Spencer Klein on the show with me today.  Spencer is one of my oldest friends and we lived across the hall from each other starting at two years old.   We talk about growing up and the expectations we had for ourselves when it came to our careers.  Ironically, Spencer is the one that went into the movie business and I went into the legal field. His father is a lawyer and my dad has a newfound career as a background actor.   It has come full circle as having a podcast is an outlet for my video production and journalism endeavors.  We discuss our trajectories and Spencer talks about how he uses his business skills regularly as there is more to the movie industry than he realized.  Thanks so much for joining me today on this very special episode.  

ironically spencer klein
Cabaret de Cancel
27: Rachel Fedorchak and Sophia Don't Know They're In a Saw Episode

Cabaret de Cancel

Play Episode Listen Later Jul 16, 2021 64:57


I want to play a game... the rules are simple... listen to this podcast and enjoy it, but for every laugh comes a price. A price that only Spencer Klein has paid because to prepare for this Saw special he watched all nine movies. Sophia will also pay a price because she has to listen and react... that poor fool thought this episode was gonna be called "tits meet ocean" or something regular... hahahahahaha. Game Over! Also, Rachel is our amazing guest! Now listen!

game over spencer klein
Cabaret de Cancel
24: The Father's Day Special - Your Honor That Was a Fib

Cabaret de Cancel

Play Episode Listen Later Jun 20, 2021 60:11


Join Ross Klein and our mystery analyst as they help Spencer and Sophia navigate courtrooms, hospitals, con-artist uncles, and this crazy little thing called life. Although Father's day is important, Spencer catches his white whale in this episode and that should be a holiday as well. Featuring the musical stylings of Spencer Klein, Robert Schumann and Isaac Boll, this episode is NOT one to miss!

Cabaret de Cancel
6: Band Kids Are Real People

Cabaret de Cancel

Play Episode Listen Later Feb 5, 2021 62:00


Hey you! Hop on board and join insult comics, Jeff Parkinson and Sophia Stio, in the merciless, take-no-prisoners, nothing-held-back, comedy central style roast of Spencer Klein!

Doctority: Plastic Surgery
Episode 20: Medical College of Wisconsin (Ft. Dr. David Rivedal and Dr. Spencer Klein)

Doctority: Plastic Surgery

Play Episode Listen Later Oct 25, 2020 39:50


Learn about the six-year integrated plastic surgery residency program at the Medical College of Wisconsin, in Milwaukee, WI, with Dr. David Rivedal and Dr. Spencer Klein. Recorded in August 2020. Emails: David: drivedal@mcw.edu Spencer: sklein@mcw.edu

Two Broke Watch Snobs
Ep. #50 Apple Watch Series 3, Giveaway News, and Seiko Orange Monsters

Two Broke Watch Snobs

Play Episode Listen Later Oct 2, 2017 96:01


The Two Broke Watch Snobs finally announce their 3K follower giveaway! Listen in for details on what watch they’re giving away and keep watching the Instagram for rules on how to enter. Mike is still on the hunt for a new “dressy watch” and the guys have an impromptu discussion about the Seiko Orange Monster and the hysteria surrounding its cancellation. Oh, and it’s confirmed, a recent TBWS contributor is actually a Bond villain ;) Finally, Kaz and Mike share their thoughts about the Apple Watch Series 3. They review some noteworthy tech details and just can’t wrap their heads around Apple’s approach to marketing this damned thing. Show Notes Madison Time: http://www.madisontimewatches.com/ Wind Up NYC: http://www.windup.wornandwound.com/ Chelsea Market: http://chelseamarket.com/ Timex Navi Harbor: http://www.timex.com/navi-harbor/TW2R52800LG.html Nintendo Power: https://en.wikipedia.org/wiki/Nintendo_Power @alexvanslyke on Instagram: https://www.instagram.com/alexvanslyke/ Seiko SKX175: https://www.instagram.com/p/BUwjgGLFaTZ/?taken-by=twobrokewatchsnobs Seiko Z199 Jubilee: https://www.youtube.com/watch?v=VkWa-D__fnI&t=29s Gavox Avidiver: https://www.instagram.com/p/BXtMryKldrR/?taken-by=twobrokewatchsnobs Seiko SARB035:https://www.seiyajapan.com/products/s-sarb035 Seiko Orange Monster: https://www.longislandwatch.com/Seiko_Superior_SRP309_SRP309K_Orange_Monster_Watch_p/srp309k1.htm RandRob YouTube Channel: https://www.youtube.com/watch?v=Zzzo2wAkRf4 Seiko SKXA35: http://quartzimodo.com/seiko-skxa35-divers-200m-review/ Seiko SKX011: https://www.areatrend.com/us/seiko-mens-skx011-orange-rubber-automatic-watch-1681896676 Seiko Alpinist: https://www.youtube.com/watch?v=2peXxZmMKnI SeikoHolic Crystals: https://www.kleinvintagewatchrepair.com/shop/ Spencer Klein & Jonathan Koch's legacy: https://www.youtube.com/watch?v=zDRm2fgeLgA Apple Watch Series 3: https://www.apple.com/apple-watch-series-3/ Fitbit: https://www.fitbit.com/home PC Master Race: https://www.reddit.com/r/pcmasterrace/

Podcast Maxximum
Episode 5: Earliest Memories, Audio Failing, and PMX Has a Co-Host!

Podcast Maxximum

Play Episode Listen Later Sep 6, 2015 88:55


Wow, crazy episode this week on PMX! Maxx brings on Spencer Klein to co-host the show this week and naturally, we had some audio hiccups throughout, but managed to keep the show as easy to follow as possible. Maxx and Spencer talk about what game genres they are tired of, their earliest memories of video games, and what they think some contenders for game of the year are. All this and more on this week's episode of Podcast Maxximum! Follow Us On Twitter! Maxx: @maxxpmx Spencer: @Recneps_ PMX: @podcastmaxximum

Spectrum
Klein and Stezelberger

Spectrum

Play Episode Listen Later Jul 13, 2012 29:59


Physicist Spencer Klein and Electrics Engineer Thorsten Stezelberger, both at Lawrenc Berkeley National Lab, describe the Neutrino Astronomical project IceCube, which was recently completed in Antarctica. They also go on to discuss proposed project Arianna.TranscriptsSpeaker 1: Spectrum's next [inaudible]. Welcome to spectrum [00:00:30] the science and technology show on k a l x Berkeley, a biweekly 30 minute program bringing you interviews featuring bay area scientists and technologists as well as a calendar of local events and news. Speaker 2: Good afternoon. I'm Brad Swift, the host of today's show, Rick Karnofsky and I interview Spencer Klein and Torsten Stessel Berger about the neutrino astronomy project. Ice Cube. Spencer Klein is a senior scientist and group leader at Lawrence Berkeley National Lab. [00:01:00] He's a member of the ice cube research team and the Ariana planning group. Thorsten Stetso Berger is an electronics engineer at Lawrence Berkeley National Lab. He too is part of the ice cube project and the Ariana team. They join us today to talk about the ice cube project and how it is helping to better define neutrinos. Spencer Klein and Thorsten setser Berger. Welcome to spectrum. Speaker 3: Thank you. Thank you. Can you talk to us a little bit about neutrinos? [00:01:30] Well, neutrinos are subatomic particles which are notable because they barely interact at all. In fact, most of them can go through the earth without interacting. This makes them an interesting subject for astrophysics because you can use them to probe places like the interior of stars where otherwise nothing else can get out and are most of them neutrinos from those sources. There's a wide range of neutrino energies that are studied. Some of the lowest energy neutrinos are solar neutrinos which [00:02:00] come from the interior of our sun. As you move up to higher energies, they come from different sources. We think a lot of the more energetic ones come from supernovas, which is when stars explode, they will produce an initial burst of neutrinos of moderate energy and then over the next thousand years or so, they will produce higher energy neutrinos as ejected spans, producing a cloud filled with shock fronts and you're particularly interested in those high energy. Speaker 3: Yes, ice cube is designed to study those neutrinos and also [00:02:30] neutrinos from even more energetic neutrinos where we don't really know where they come from. There are two theories. One is that they come from objects called active Galactic Nuclei. These are galaxies which have a super massive black hole at their center and they're rejecting a jet of particles perpendicular, more along their axis. And this jet is believed to also be a site to accelerate protons and other cosmic rays to very high energies. The other possible source of ultra energy neutrinos [00:03:00] are gamma ray bursts, which are when two black holes collide or a black hole collides with a neutron star. And if the neutrinos don't interact or interact so rarely and weekly with matter, how do we actually detect them? Well, the simple answer is you need a very large detector. Ice Cube is one cubic kilometer in volume and that's big enough that we think we should be able to detect neutrinos from these astrophysical sources. Speaker 3: The other project we work on, Ariana is even bigger. It's [00:03:30] proposed, but it's proposed to have about a hundred cubic kilometers of volume. And so you have an enormous detector to detect a few events and once you detect them, how can you tell where they came from? Well, with ice cube we can get the incoming direction of the neutrinos to within about a degree. So what we do is we look for neutrinos. Most of what we see out of these background atmospheric neutrinos which are produced when cosmic rays interact in the earth's atmosphere. But on top [00:04:00] of that we look for a cluster of neutrinos coming from a specific direction. That would be a clear sign of a neutrino source, which would be, you know, and then we can look in that direction and see what interesting sources lie. That way we can also look for extremely energetic neutrinos which are unlikely to be these atmospheric neutrinos. Speaker 3: And how is it that you measure that energy? What happens is a neutrino will come in and occasionally interact in the Antarctic. Ice should mention that ice cube is located at the South Pole where [00:04:30] there's 28 hundreds of meters of ice on top of the rock below. Occasionally in Neutrino will come in and interact in the ice and if it's something called a type of neutrino called the [inaudible] Neutrino, most of its energy will go into a subatomic particle called the Meuron. Meuron is interesting because it's electrically charged. As it goes through the ice, it will give off light, something we call Toronto radiation. So we've instrumented this cubic kilometer of ice with over 5,000 optical [00:05:00] modules, which are basically optical sensors. And so we record the amount and arrival times of the light at these optical sensors. And from that we can determine the neutrino direction to about within a degree. Speaker 3: And we can also get an estimate of the energy. Um, essentially is the on is more energetic. It will also produce other electrically charged particles as it travels. Those will give off more light. And so the light output is proportional to the neutrino energy. So you're taking an advantage of the fact that there's [00:05:30] a lot of ice in Antarctica and also that it's very big. Are there other reasons to do it at the South Pole? Well, the other critical component about the ice is that it has to be very clear, shouldn't scatter light and it shouldn't absorb light. And in fact the light can travel up to 200 meters through the ice before being absorbed. This is important because that means we can have a relatively sparse array. You know, we have only 5,000 sensors spread over a cubic kilometer. That's only if the light can travel long distances through the ice. [00:06:00] And do you have to take into account that the ice in the Antarctic is not perfectly clean? Yes. When we reconstruct the neutrino directions, we use this sophisticated maximum likelihood fitter. Essentially we try all sorts of different Milan directions and see which one is the most likely. And that takes into account the optical properties of the ace and includes how they vary with depth. There are some dust layers in the ice where the absorption length is much shorter and some places, [00:06:30] well most of the ice where it's much better. Speaker 4: Our guests on spectrum today are Spencer Klein and Thorsten Stetson Burger from Lawrence Berkeley national lab. They are part of a physics project named Ice Cube. In the next segment they talk about working at the South Pole. This is KALX Berkeley. Speaker 3: Can you compare the two experiments, both ice Cuban on a little bit? Well, ice cube is designed [00:07:00] for sort of moderate energy neutrinos, but for the really energetic neutrinos are, they are rare enough so that a one cubic kilometer detector just isn't big enough. And so for that you need something bigger and it's hard to imagine how you could scale the optical techniques that ice cube uses to larger detectors. So that's why we looked for a new technique in it. Here I should say we, the royal, we either many people, many places in the world looking at different versions. And so what we've chosen is looking [00:07:30] for radio [inaudible] off the mission. You know, we have this interaction in the ice. Some of the time. If it's an electron Neutrino, it produces a compact shower of particles. That shower will have more negatively charged particles than positively charged. Speaker 3: And so it will emit radio waves, you know, at frequencies up to about a Gigahertz coherently, which means that the radio emission strength depends on the square of the neutrino energies. So when you go to very high neutrino energies, this is a preferred technique. Radio waves can [00:08:00] travel between 300 meters and a kilometer in the ice, which means you can get by with a much sparser array. So you can instrument a hundred cubic kilometers with a reasonable number of detectors. When Ariane is developed, it will get to access higher energies. Will it still didn't detect some of the moderately high energies that ice cube is currently reaching? No, and there's no overlap because of the coherence and just not sensitive. I mean, ice cube will occasionally see these much higher energy neutrinos, [00:08:30] but it's just not big enough to see very many of them. Uh, you commented on, or you mentioned the size of the collaboration. Speaker 3: Can you sort of speak about how big these projects are? Sure. Ice Cube has got about 250 scientists in it from the u s Europe, Barbados, Japan, and New Zealand. Oh yeah. And plus one person from Australia now. And that's a well established, you know, it's a large experiment. Arianna is just getting going. It's got, I'll say less than a dozen [00:09:00] people in it. Mostly from UC Irvine and some involvement from LDL. How many years have you had experience with your sensors in the field then? That's kind of a complicated question and that the idea of doing neutrino astronomy in the Antarctic ice has been around for more than 20 years. The first efforts to actually put sensors in the ice, we're in the early 1990s these used very simple sensors. We just had a photo multiplier tube, essentially a very sensitive [00:09:30] optical detector, and they sent their signals to the surface. There are no complicated electronics in the ice. Speaker 3: The first Amanda effort in fact failed because the sensors were near the surface where the light was scattering very rapidly. Turns out the upper kilometer of ice is filled with little air bubbles, but then as you get down in depth, there's enough pressure to squeeze these bubbles out of existence. And so you go from very cloudy ice like what you see if you look in the center of an ice cube and then you go deeper [00:10:00] and you end up with this incredibly clear ice. So the first efforts were in this cloudy ice. Then in the second half of the 1990s Amanda was deployed in the deep highs. This is much smaller than ice cube in many respects. The predecessor, of course, the problem with Amanda was this transmission to the surface. It worked but it was very, very touchy and it wasn't something you could scale to the ice cube size. So one where people got together and came up with these digital optical modules where all of the digitizing electronics [00:10:30] is actually in the module. We also made a lot of other changes and improvements to come up with a detector that would be really robust and then we deployed the first ice cube string in 2005 and continued and then the last string was deployed at the end of 2010 Speaker 5: so basically from the scientific point or engineering point of view, we're learning about the detector. We got data from the first strain. It was not very useful for take neutrino science but you can learn to understand [00:11:00] the detector, learn how the electronics behaves, if there is a problem, change code to get different data. Speaker 3: When we did see some new is in that run and there's this one beautiful event where we saw this [inaudible] from a neutrino just moving straight up the string. I think it hit 51 out of the 60 optical sensors. So we're basically tracking it for 800 meters. It was just a beautiful that Speaker 5: what is the lifelight down there? The food, the day to day, [00:11:30] we've never been there in the winter time, so I can only talk about a summer and in the summer you're there for something specific like drilling or deploying a, so to summertime keeps you pretty busy and you do your stuff and then afterwards you hang out a little bit to wind down. And sometimes with some folks playing pool or ping pong or watching movies or just reading something and then time [00:12:00] again for the sleep or sleeping. And the next day for drawing for example, we had three shifts. And so that kept you pretty, pretty busy. One season when I was thrilling there I was on what we call the graveyard shift. Starting from 11 to I think eight in the morning. I saw and yeah, it was daylight. You don't notice it except you always get dinner for breakfast and scrambled eggs and potatoes for dinner. Speaker 3: The new station at the South Pole is really very nice and I would [00:12:30] say quite comfortable, good recreational facilities. I mean, and I would say the food was excellent, really quite impressive and you get to hang out with a bunch of international scientists that are down there. How collegial isn't, it Speaker 5: depends a little bit on the work. Like when I was rolling on night shift, we mostly got to hang out with people running the station. That was fairly collegial. Speaker 3: There's actually not very many scientists at the South Pole. In the summer there were about 250 [00:13:00] people there and maybe 20 of them were scientists. Most of them were people dealing with logistics. These are people, you know, heavy equipment operators. Fuel Lees would get the fuel off of the plane, cooks people, and even then can building the station wasn't quite done yet. The drillers will lodge wide variety of occupations but not all that many scientists. How close are the experiments to the station? Speaker 5: They are quite a few experiments [00:13:30] based in the station. Ice Cube is a kilometer away about probably Speaker 3: Lamotta and a half to the, to the ice cube lab, which is where the surface electronics is located. Speaker 5: So it's pretty close walking distance called walk. But it depends. I mean I don't mind the calls or it was a nice walk but they have like ice cube, uh, drilling. We are like lunch break also. It's [00:14:00] a little bit far to walk kilometer out or even throughout depending where you drill. So we had a car to drive back and forth to the station to eat lunch. Otherwise you are out for too long. Speaker 3: Yeah, they give you a really good equipment and so it's amazing how plaza you can be about walking around when it's 40 below, outside. Speaker 5: Especially if you do physical work outside as part of drilling also. It's amazing how much of that cold weather Ikea you actually take off because you just [00:14:30] do staff and you warm up. Speaker 4: [inaudible] you are listening to spectrum on KALX Berkeley coming up, our guests, Spencer Klein and Torsten Stotzel Burger detail, the ice cube data analysis process, Speaker 3: the ongoing maintenance of Ice Cube Sarah Plan for its lifetime Speaker 5: for the stuff [00:15:00] in the eyes, it's really hard to replace that. You cannot easily drill down and take them out. They are plans, uh, to keep the surface electronics, especially the computers update them as lower power hardware becomes available. Otherwise I'm not aware of preventive maintenance. You could do with like on a car. Yeah. Speaker 3: I have to say the engineers did a great job on ice cube. About 98% of the optical modules are working. Most of the failures were infant [00:15:30] mortality. They did not survive the deployment when we've only had a handful of optical modules fail after deployment and all the evidence is we'll be able to keep running it as long as it's interesting. And is there a point in which it's no longer interesting in terms of how many sensors are still active? I think we'll reach the point where the data is less interesting before we run out of sensors now. Okay. You know, we might be losing one or two sensors a year. In fact, we're still at the point where [00:16:00] due to various software improvements, including in the firmware and the optical modules, each year's run has more sensors than the previous years. Even if we only had 90% of them working, that would be plenty. Speaker 3: And you know, that's probably a hundred years from now. What do we have guests on to speak about the LHC at certain they were talking about the gigantic amounts of data that they generate and how surprisingly long it takes for scientists to analyze that data to actually get a hold [00:16:30] of data from the detector. And you're generating very large amounts of data. And furthermore, it's in Antarctica. So how much turnaround time is there? Well, the Antarctica doesn't add very much time. We typically get data in the north within a few days or a week after it's taken. There is a bit of a lag and try and take this time to understand how to analyze the data. For example, now we're working on, for the most part, the data that was taken in 2010 and [00:17:00] you know, hope to have that out soon probably for summer conferences. But understanding how to best analyze the data is not trivial. For example, this measurement of the mule on energies, very dependent on a lot of assumptions about the ice and so we have ways to do it now, but we're far from the optimal method Speaker 5: and keep in mind that detector built, it's just finished. So before you always added in a little bit more. So each year the data looked different because you've got more sensors in the data. Speaker 3: [00:17:30] Let's say for things where turnaround is important. For example, dimension, these gamma ray bursts, there's where this happens when a bunch of satellites see a burst of x-rays or gamma rays coming from somewhere in the sky. They can tell us when it happened and give us an estimate of the direction. We can have an and I would say not quite real time, but you know that we could have turned around if a couple of weeks. We also measure the rates in each of the detectors. This is the way to look for low energy neutrinos from a [00:18:00] supernova that is essentially done in real time. If the detector sees an increase, then somebody will get an email alert essentially immediately. If we got one that looked like a Supernova, we could turn that around very quickly. So are the algorithms that you're using for this longer term analysis improving? Speaker 3: Yes. They're much more sophisticated than they were two years ago. I'd say we're gradually approaching and I'm ask some Todrick set of algorithm, but we're still quite a ways [00:18:30] to go. We're still learning a lot of things. You know, this is very different from any other experiment that's been done. Normally experiments if the LHC, if they are tracking a charged particle, they measure points along the track. In our case, the light is admitted at the trend off angle. About 41 degrees. So the data points we see are anywhere from a few meters to a hundred meters from the track. And because of the scattering of light, it's a not so obvious how to find [00:19:00] the optimum track and it's, you know, it's very dependent on a lot of assumptions and we're still working on that. And we have methods that work well. As I said, you know, we can get an angular resolution of better than a degree in some cases, but there's still probably some room to be gotten there. Speaker 5: And then also, I mean I'm not involved in the science, but I hear people have new ideas how to look at a data. So that's still evolving too. Speaker 3: Yeah. Like you know, one analysis that people are working on, but we don't have yet would [00:19:30] be a speculative search where you're looking for a pair of event, a pair of neo-cons going upward through the detector in the same direction at the same time, which would quite possibly be a signal of some sort of new physics. And it's certainly an interesting typology to look for, but we're not there yet. And are there different teams looking at the same data to try to find different results and broaden the search so to speak? Uh, yes. We have seven or eight different physics working [00:20:00] groups in each of those groups is concentrating on a different type of physics or a different class of physics. For example, one group is looking for point sources, you know, hotspots in the sky. Second Group is looking at atmospheric and diffuse neutrinos trying to measure the energy spectrum of the neutrinos. Speaker 3: We do see both the atmospheric and also looking for an additional component. There's a group doing cosmic ray physics. There's a group looking for exotic physics. These are things like these pairs [00:20:30] of upward going particles. Also looking for other oddities such as magnetic monopoles. There's a group that's looking for neutrinos that might be produced from weakly interacting. Massive particles, IAA, dark matter, but there's a group that's monitoring the rates of the detector. This scalers looking for Supernova and oh, there's also a group looking for talented Trinos, which is the this very distinctive topology town. Neutrinos are sort of the third flavor of neutrinos and those are [00:21:00] mostly only produced by extraterrestrial sources and they look very distinctively. You would look for case where you see two clusters of energy and the detector separated by a few hundred meters. Speaker 5: Looking at what's next, what would be the sort of ideal laboratory? If you want something that's very big, obviously Antarctica is a great challenge. Can you do neutrino detection in space for instance? [inaudible] Speaker 3: hmm, that's an interesting question. There are people who [00:21:30] are talking about that and the main application is trying to look for these cosmic gray air showers. The best experiments to study high energy, cosmic gray air showers are these things called air shower arrays, which are an array of detectors. Um, the largest one is something called the OJ Observatory in Argentina. It covers about 3000 square kilometers with an array of detectors on kind of a one and a half kilometer grid. And that's about as largest surface detector as you could imagine. Building the alternative [00:22:00] technology is look for something called air fluorescents. When the showers go through the air, they light it up. Particularly the nitrogen is excited and in that kind of like a fluorescent tube. So you see this burst of light as the shower travels through the atmosphere. O J in addition to the surface detectors has these cameras called flies eyes that look for this fluorescence, but it's limited in scale. And people have proposed building experiments that would sit on satellites or a space station [00:22:30] and look down and look at these showers from above. They could cover a much larger area. They could also look for showers from upward going particles, I. E. Neutrino interactions. But at this point that's all pretty speculative. Speaker 5: And when's your next trip to Antarctica? Uh, that's all depending on funding. I would like to go again and hopefully soon. I think I'm cautiously optimistic. We'll be able to go again this year. Hmm. Spencer in Thorsten. Thanks for joining us. Thank you. Thank you. Speaker 4: [00:23:00] [inaudible] regular feature of spectrum is to mention a few of the science and technology events that are happening locally over the next few weeks. Lisa Katovich joins me for that Speaker 6: calendar. The August general meeting of the East Bay Astronomical Society is Saturday, July 14th at the Chabot space and science centers, Dellums [00:23:30] building 10,000 Skyline Boulevard in Oakland. Ezra Bahrani is the evening Speaker. The title of his talk is UFOs, the proof, the physics and why they're here. The meeting starts at 7:30 PM Speaker 2: join Nobel laureates and social and environmental justice advocates at the towns and Tay Gore third annual seminar for Science and technology on behalf of the peoples of Bengali and the Himalayan basins, the subject, the global water crisis [00:24:00] prevention and solution. Saturday, July 21st 1:30 PM to 7:30 PM the event is jointly sponsored by UC Berkeley's department of Public Health and the international institute of the Bengali and Himalayan basins. Guest Speakers include three Nobel laureates, Charles h towns, Burton Richter and Douglas Ashur off. Also presenting our Francis towns advocate for social justice, Dr. Rush, Gosh [00:24:30] and Sterling Brunel. The event will be held in one 45 Dwinelle hall on the UC Berkeley campus. That's Saturday, July 21st 1:30 PM to 7:30 PM for more details, contact the UC Berkeley School of Public Health, Speaker 6: the next science at cal lectures on July 21st the talk will be given by Dr Jeffrey Silverman and it's entitled exploding stars, Dark Energy, and the runaway universe. Dr Silverman has been a guest [00:25:00] on spectrum. His research has been in the study of Super Novi. His lecture will focus on how the study of supernovae led to the recent discovery that the universe is expanding, likely due to a repulsive and mysterious dark energy. It was these observations that were recently awarded the 2011 Nobel Prize in physics. The lecture is July 21st at 11:00 AM and the genetics and plant biology building room 100 Speaker 2: next to news stories. Speaker 6: 3000 species [00:25:30] of mosquitoes are responsible for malaria, dengue, a fever, yellow fever, West Nile virus, and cephalitis and many more diseases. In Burkina Faso alone, residents can expect 200 bytes a day. Rapid resistance to pesticides on the part of malaria mosquitoes has prompted researchers all over the globe to deploy novel strategies against this and other diseases. Targeting Dengue. A fever has an advantage over malaria as only one species. Eighties [00:26:00] Egypt die is responsible for spreading it versus the 20 species responsible for spreading malaria. A British biotechnology company called Oxitec has developed a method to modify the genetic structure of the male eighties Aegypti mosquito transforming it into a mutant capable of destroying its own species. In 2010 they announced impressive preliminary results of the first known test of 3 million free flying transgenic mosquitoes engineered [00:26:30] to start a population crash after infiltrating wild disease spreading eighties a Gyp dye swarms on Cayman Island. Speaker 6: Oxitec has recently applied to the FDA for approval of its mosquito in the u s with Key West under consideration as a future test site in 2009 key west suffered its first dengate outbreak in 73 years. Australian researchers are testing and mosquito intended to fight dengue, a fever bypassing the disruptive Wolbachia bacteria to other mosquitoes, a very [00:27:00] different approach than transgenic genes funded largely by the bill and Melinda Gates Foundation. The project has shown that the Wolbachia strain not only shortens the life of a mosquito, but also reduces the amount of virus it develops. Releases in Queensland, Australia last year showed that Wolbachia could spread through a wild population quickly and future test sites are under consideration. In Vietnam. Speaker 2: The UC Berkeley News Center reports a prototype network being installed by chemists at the University of California. Berkeley [00:27:30] will employ 40 sensors spread over a 27 square mile grid. The information the network will provide could be used to monitor local carbon dioxide emissions to check on the effectiveness of carbon reduction strategies now mandated by the state, but hard to verify built and installed by project leader Professor Ron Cohen and graduate student Virginia Tighe and their lab colleagues. The shoe box size sensors will continuously measure carbon dioxide, carbon monoxide, [00:28:00] nitrogen dioxide, and ozone levels as well as temperature, pressure and humidity streaming. The information live to the web through the site. beacon.berkeley.edu the sensor network dubbed Beacon stretches from the East Bay regional parks on the east to interstate eight 80 on the west from El Surrito on the north nearly to San Leandro on the south encompassing open space as well as heavily traffic areas. [00:28:30] Most of the sensors are being mounted on the roofs of local schools in order to get students interested in the connection between carbon dioxide emissions and climate change. The UC Berkeley researchers work with Oakland's Chabot space and science center to create middle school and high school activities using live sensor data stream through the web as part of the students energy and climate science curriculum. The beacon network is a pilot program funded by the National Science Foundation to determine what information can be learned [00:29:00] from a densely spaced network Speaker 1: [inaudible].Speaker 2: The music heard during the show is from most done at David's album, folk and acoustics made available through a creative Commons license 3.0 attribution. Speaker 1: Thank you for listening to spectrum. If you have comments about the show, please send them to us via email. Our email address [00:29:30] is spectrum dot kalx@yahoo.com join us in two weeks at this same time. [inaudible]. Hosted on Acast. See acast.com/privacy for more information.

Spectrum
Klein and Stezelberger

Spectrum

Play Episode Listen Later Jul 13, 2012 29:59


Physicist Spencer Klein and Electrics Engineer Thorsten Stezelberger, both at Lawrenc Berkeley National Lab, describe the Neutrino Astronomical project IceCube, which was recently completed in Antarctica. They also go on to discuss proposed project Arianna.TranscriptsSpeaker 1: Spectrum's next [inaudible]. Welcome to spectrum [00:00:30] the science and technology show on k a l x Berkeley, a biweekly 30 minute program bringing you interviews featuring bay area scientists and technologists as well as a calendar of local events and news. Speaker 2: Good afternoon. I'm Brad Swift, the host of today's show, Rick Karnofsky and I interview Spencer Klein and Torsten Stessel Berger about the neutrino astronomy project. Ice Cube. Spencer Klein is a senior scientist and group leader at Lawrence Berkeley National Lab. [00:01:00] He's a member of the ice cube research team and the Ariana planning group. Thorsten Stetso Berger is an electronics engineer at Lawrence Berkeley National Lab. He too is part of the ice cube project and the Ariana team. They join us today to talk about the ice cube project and how it is helping to better define neutrinos. Spencer Klein and Thorsten setser Berger. Welcome to spectrum. Speaker 3: Thank you. Thank you. Can you talk to us a little bit about neutrinos? [00:01:30] Well, neutrinos are subatomic particles which are notable because they barely interact at all. In fact, most of them can go through the earth without interacting. This makes them an interesting subject for astrophysics because you can use them to probe places like the interior of stars where otherwise nothing else can get out and are most of them neutrinos from those sources. There's a wide range of neutrino energies that are studied. Some of the lowest energy neutrinos are solar neutrinos which [00:02:00] come from the interior of our sun. As you move up to higher energies, they come from different sources. We think a lot of the more energetic ones come from supernovas, which is when stars explode, they will produce an initial burst of neutrinos of moderate energy and then over the next thousand years or so, they will produce higher energy neutrinos as ejected spans, producing a cloud filled with shock fronts and you're particularly interested in those high energy. Speaker 3: Yes, ice cube is designed to study those neutrinos and also [00:02:30] neutrinos from even more energetic neutrinos where we don't really know where they come from. There are two theories. One is that they come from objects called active Galactic Nuclei. These are galaxies which have a super massive black hole at their center and they're rejecting a jet of particles perpendicular, more along their axis. And this jet is believed to also be a site to accelerate protons and other cosmic rays to very high energies. The other possible source of ultra energy neutrinos [00:03:00] are gamma ray bursts, which are when two black holes collide or a black hole collides with a neutron star. And if the neutrinos don't interact or interact so rarely and weekly with matter, how do we actually detect them? Well, the simple answer is you need a very large detector. Ice Cube is one cubic kilometer in volume and that's big enough that we think we should be able to detect neutrinos from these astrophysical sources. Speaker 3: The other project we work on, Ariana is even bigger. It's [00:03:30] proposed, but it's proposed to have about a hundred cubic kilometers of volume. And so you have an enormous detector to detect a few events and once you detect them, how can you tell where they came from? Well, with ice cube we can get the incoming direction of the neutrinos to within about a degree. So what we do is we look for neutrinos. Most of what we see out of these background atmospheric neutrinos which are produced when cosmic rays interact in the earth's atmosphere. But on top [00:04:00] of that we look for a cluster of neutrinos coming from a specific direction. That would be a clear sign of a neutrino source, which would be, you know, and then we can look in that direction and see what interesting sources lie. That way we can also look for extremely energetic neutrinos which are unlikely to be these atmospheric neutrinos. Speaker 3: And how is it that you measure that energy? What happens is a neutrino will come in and occasionally interact in the Antarctic. Ice should mention that ice cube is located at the South Pole where [00:04:30] there's 28 hundreds of meters of ice on top of the rock below. Occasionally in Neutrino will come in and interact in the ice and if it's something called a type of neutrino called the [inaudible] Neutrino, most of its energy will go into a subatomic particle called the Meuron. Meuron is interesting because it's electrically charged. As it goes through the ice, it will give off light, something we call Toronto radiation. So we've instrumented this cubic kilometer of ice with over 5,000 optical [00:05:00] modules, which are basically optical sensors. And so we record the amount and arrival times of the light at these optical sensors. And from that we can determine the neutrino direction to about within a degree. Speaker 3: And we can also get an estimate of the energy. Um, essentially is the on is more energetic. It will also produce other electrically charged particles as it travels. Those will give off more light. And so the light output is proportional to the neutrino energy. So you're taking an advantage of the fact that there's [00:05:30] a lot of ice in Antarctica and also that it's very big. Are there other reasons to do it at the South Pole? Well, the other critical component about the ice is that it has to be very clear, shouldn't scatter light and it shouldn't absorb light. And in fact the light can travel up to 200 meters through the ice before being absorbed. This is important because that means we can have a relatively sparse array. You know, we have only 5,000 sensors spread over a cubic kilometer. That's only if the light can travel long distances through the ice. [00:06:00] And do you have to take into account that the ice in the Antarctic is not perfectly clean? Yes. When we reconstruct the neutrino directions, we use this sophisticated maximum likelihood fitter. Essentially we try all sorts of different Milan directions and see which one is the most likely. And that takes into account the optical properties of the ace and includes how they vary with depth. There are some dust layers in the ice where the absorption length is much shorter and some places, [00:06:30] well most of the ice where it's much better. Speaker 4: Our guests on spectrum today are Spencer Klein and Thorsten Stetson Burger from Lawrence Berkeley national lab. They are part of a physics project named Ice Cube. In the next segment they talk about working at the South Pole. This is KALX Berkeley. Speaker 3: Can you compare the two experiments, both ice Cuban on a little bit? Well, ice cube is designed [00:07:00] for sort of moderate energy neutrinos, but for the really energetic neutrinos are, they are rare enough so that a one cubic kilometer detector just isn't big enough. And so for that you need something bigger and it's hard to imagine how you could scale the optical techniques that ice cube uses to larger detectors. So that's why we looked for a new technique in it. Here I should say we, the royal, we either many people, many places in the world looking at different versions. And so what we've chosen is looking [00:07:30] for radio [inaudible] off the mission. You know, we have this interaction in the ice. Some of the time. If it's an electron Neutrino, it produces a compact shower of particles. That shower will have more negatively charged particles than positively charged. Speaker 3: And so it will emit radio waves, you know, at frequencies up to about a Gigahertz coherently, which means that the radio emission strength depends on the square of the neutrino energies. So when you go to very high neutrino energies, this is a preferred technique. Radio waves can [00:08:00] travel between 300 meters and a kilometer in the ice, which means you can get by with a much sparser array. So you can instrument a hundred cubic kilometers with a reasonable number of detectors. When Ariane is developed, it will get to access higher energies. Will it still didn't detect some of the moderately high energies that ice cube is currently reaching? No, and there's no overlap because of the coherence and just not sensitive. I mean, ice cube will occasionally see these much higher energy neutrinos, [00:08:30] but it's just not big enough to see very many of them. Uh, you commented on, or you mentioned the size of the collaboration. Speaker 3: Can you sort of speak about how big these projects are? Sure. Ice Cube has got about 250 scientists in it from the u s Europe, Barbados, Japan, and New Zealand. Oh yeah. And plus one person from Australia now. And that's a well established, you know, it's a large experiment. Arianna is just getting going. It's got, I'll say less than a dozen [00:09:00] people in it. Mostly from UC Irvine and some involvement from LDL. How many years have you had experience with your sensors in the field then? That's kind of a complicated question and that the idea of doing neutrino astronomy in the Antarctic ice has been around for more than 20 years. The first efforts to actually put sensors in the ice, we're in the early 1990s these used very simple sensors. We just had a photo multiplier tube, essentially a very sensitive [00:09:30] optical detector, and they sent their signals to the surface. There are no complicated electronics in the ice. Speaker 3: The first Amanda effort in fact failed because the sensors were near the surface where the light was scattering very rapidly. Turns out the upper kilometer of ice is filled with little air bubbles, but then as you get down in depth, there's enough pressure to squeeze these bubbles out of existence. And so you go from very cloudy ice like what you see if you look in the center of an ice cube and then you go deeper [00:10:00] and you end up with this incredibly clear ice. So the first efforts were in this cloudy ice. Then in the second half of the 1990s Amanda was deployed in the deep highs. This is much smaller than ice cube in many respects. The predecessor, of course, the problem with Amanda was this transmission to the surface. It worked but it was very, very touchy and it wasn't something you could scale to the ice cube size. So one where people got together and came up with these digital optical modules where all of the digitizing electronics [00:10:30] is actually in the module. We also made a lot of other changes and improvements to come up with a detector that would be really robust and then we deployed the first ice cube string in 2005 and continued and then the last string was deployed at the end of 2010 Speaker 5: so basically from the scientific point or engineering point of view, we're learning about the detector. We got data from the first strain. It was not very useful for take neutrino science but you can learn to understand [00:11:00] the detector, learn how the electronics behaves, if there is a problem, change code to get different data. Speaker 3: When we did see some new is in that run and there's this one beautiful event where we saw this [inaudible] from a neutrino just moving straight up the string. I think it hit 51 out of the 60 optical sensors. So we're basically tracking it for 800 meters. It was just a beautiful that Speaker 5: what is the lifelight down there? The food, the day to day, [00:11:30] we've never been there in the winter time, so I can only talk about a summer and in the summer you're there for something specific like drilling or deploying a, so to summertime keeps you pretty busy and you do your stuff and then afterwards you hang out a little bit to wind down. And sometimes with some folks playing pool or ping pong or watching movies or just reading something and then time [00:12:00] again for the sleep or sleeping. And the next day for drawing for example, we had three shifts. And so that kept you pretty, pretty busy. One season when I was thrilling there I was on what we call the graveyard shift. Starting from 11 to I think eight in the morning. I saw and yeah, it was daylight. You don't notice it except you always get dinner for breakfast and scrambled eggs and potatoes for dinner. Speaker 3: The new station at the South Pole is really very nice and I would [00:12:30] say quite comfortable, good recreational facilities. I mean, and I would say the food was excellent, really quite impressive and you get to hang out with a bunch of international scientists that are down there. How collegial isn't, it Speaker 5: depends a little bit on the work. Like when I was rolling on night shift, we mostly got to hang out with people running the station. That was fairly collegial. Speaker 3: There's actually not very many scientists at the South Pole. In the summer there were about 250 [00:13:00] people there and maybe 20 of them were scientists. Most of them were people dealing with logistics. These are people, you know, heavy equipment operators. Fuel Lees would get the fuel off of the plane, cooks people, and even then can building the station wasn't quite done yet. The drillers will lodge wide variety of occupations but not all that many scientists. How close are the experiments to the station? Speaker 5: They are quite a few experiments [00:13:30] based in the station. Ice Cube is a kilometer away about probably Speaker 3: Lamotta and a half to the, to the ice cube lab, which is where the surface electronics is located. Speaker 5: So it's pretty close walking distance called walk. But it depends. I mean I don't mind the calls or it was a nice walk but they have like ice cube, uh, drilling. We are like lunch break also. It's [00:14:00] a little bit far to walk kilometer out or even throughout depending where you drill. So we had a car to drive back and forth to the station to eat lunch. Otherwise you are out for too long. Speaker 3: Yeah, they give you a really good equipment and so it's amazing how plaza you can be about walking around when it's 40 below, outside. Speaker 5: Especially if you do physical work outside as part of drilling also. It's amazing how much of that cold weather Ikea you actually take off because you just [00:14:30] do staff and you warm up. Speaker 4: [inaudible] you are listening to spectrum on KALX Berkeley coming up, our guests, Spencer Klein and Torsten Stotzel Burger detail, the ice cube data analysis process, Speaker 3: the ongoing maintenance of Ice Cube Sarah Plan for its lifetime Speaker 5: for the stuff [00:15:00] in the eyes, it's really hard to replace that. You cannot easily drill down and take them out. They are plans, uh, to keep the surface electronics, especially the computers update them as lower power hardware becomes available. Otherwise I'm not aware of preventive maintenance. You could do with like on a car. Yeah. Speaker 3: I have to say the engineers did a great job on ice cube. About 98% of the optical modules are working. Most of the failures were infant [00:15:30] mortality. They did not survive the deployment when we've only had a handful of optical modules fail after deployment and all the evidence is we'll be able to keep running it as long as it's interesting. And is there a point in which it's no longer interesting in terms of how many sensors are still active? I think we'll reach the point where the data is less interesting before we run out of sensors now. Okay. You know, we might be losing one or two sensors a year. In fact, we're still at the point where [00:16:00] due to various software improvements, including in the firmware and the optical modules, each year's run has more sensors than the previous years. Even if we only had 90% of them working, that would be plenty. Speaker 3: And you know, that's probably a hundred years from now. What do we have guests on to speak about the LHC at certain they were talking about the gigantic amounts of data that they generate and how surprisingly long it takes for scientists to analyze that data to actually get a hold [00:16:30] of data from the detector. And you're generating very large amounts of data. And furthermore, it's in Antarctica. So how much turnaround time is there? Well, the Antarctica doesn't add very much time. We typically get data in the north within a few days or a week after it's taken. There is a bit of a lag and try and take this time to understand how to analyze the data. For example, now we're working on, for the most part, the data that was taken in 2010 and [00:17:00] you know, hope to have that out soon probably for summer conferences. But understanding how to best analyze the data is not trivial. For example, this measurement of the mule on energies, very dependent on a lot of assumptions about the ice and so we have ways to do it now, but we're far from the optimal method Speaker 5: and keep in mind that detector built, it's just finished. So before you always added in a little bit more. So each year the data looked different because you've got more sensors in the data. Speaker 3: [00:17:30] Let's say for things where turnaround is important. For example, dimension, these gamma ray bursts, there's where this happens when a bunch of satellites see a burst of x-rays or gamma rays coming from somewhere in the sky. They can tell us when it happened and give us an estimate of the direction. We can have an and I would say not quite real time, but you know that we could have turned around if a couple of weeks. We also measure the rates in each of the detectors. This is the way to look for low energy neutrinos from a [00:18:00] supernova that is essentially done in real time. If the detector sees an increase, then somebody will get an email alert essentially immediately. If we got one that looked like a Supernova, we could turn that around very quickly. So are the algorithms that you're using for this longer term analysis improving? Speaker 3: Yes. They're much more sophisticated than they were two years ago. I'd say we're gradually approaching and I'm ask some Todrick set of algorithm, but we're still quite a ways [00:18:30] to go. We're still learning a lot of things. You know, this is very different from any other experiment that's been done. Normally experiments if the LHC, if they are tracking a charged particle, they measure points along the track. In our case, the light is admitted at the trend off angle. About 41 degrees. So the data points we see are anywhere from a few meters to a hundred meters from the track. And because of the scattering of light, it's a not so obvious how to find [00:19:00] the optimum track and it's, you know, it's very dependent on a lot of assumptions and we're still working on that. And we have methods that work well. As I said, you know, we can get an angular resolution of better than a degree in some cases, but there's still probably some room to be gotten there. Speaker 5: And then also, I mean I'm not involved in the science, but I hear people have new ideas how to look at a data. So that's still evolving too. Speaker 3: Yeah. Like you know, one analysis that people are working on, but we don't have yet would [00:19:30] be a speculative search where you're looking for a pair of event, a pair of neo-cons going upward through the detector in the same direction at the same time, which would quite possibly be a signal of some sort of new physics. And it's certainly an interesting typology to look for, but we're not there yet. And are there different teams looking at the same data to try to find different results and broaden the search so to speak? Uh, yes. We have seven or eight different physics working [00:20:00] groups in each of those groups is concentrating on a different type of physics or a different class of physics. For example, one group is looking for point sources, you know, hotspots in the sky. Second Group is looking at atmospheric and diffuse neutrinos trying to measure the energy spectrum of the neutrinos. Speaker 3: We do see both the atmospheric and also looking for an additional component. There's a group doing cosmic ray physics. There's a group looking for exotic physics. These are things like these pairs [00:20:30] of upward going particles. Also looking for other oddities such as magnetic monopoles. There's a group that's looking for neutrinos that might be produced from weakly interacting. Massive particles, IAA, dark matter, but there's a group that's monitoring the rates of the detector. This scalers looking for Supernova and oh, there's also a group looking for talented Trinos, which is the this very distinctive topology town. Neutrinos are sort of the third flavor of neutrinos and those are [00:21:00] mostly only produced by extraterrestrial sources and they look very distinctively. You would look for case where you see two clusters of energy and the detector separated by a few hundred meters. Speaker 5: Looking at what's next, what would be the sort of ideal laboratory? If you want something that's very big, obviously Antarctica is a great challenge. Can you do neutrino detection in space for instance? [inaudible] Speaker 3: hmm, that's an interesting question. There are people who [00:21:30] are talking about that and the main application is trying to look for these cosmic gray air showers. The best experiments to study high energy, cosmic gray air showers are these things called air shower arrays, which are an array of detectors. Um, the largest one is something called the OJ Observatory in Argentina. It covers about 3000 square kilometers with an array of detectors on kind of a one and a half kilometer grid. And that's about as largest surface detector as you could imagine. Building the alternative [00:22:00] technology is look for something called air fluorescents. When the showers go through the air, they light it up. Particularly the nitrogen is excited and in that kind of like a fluorescent tube. So you see this burst of light as the shower travels through the atmosphere. O J in addition to the surface detectors has these cameras called flies eyes that look for this fluorescence, but it's limited in scale. And people have proposed building experiments that would sit on satellites or a space station [00:22:30] and look down and look at these showers from above. They could cover a much larger area. They could also look for showers from upward going particles, I. E. Neutrino interactions. But at this point that's all pretty speculative. Speaker 5: And when's your next trip to Antarctica? Uh, that's all depending on funding. I would like to go again and hopefully soon. I think I'm cautiously optimistic. We'll be able to go again this year. Hmm. Spencer in Thorsten. Thanks for joining us. Thank you. Thank you. Speaker 4: [00:23:00] [inaudible] regular feature of spectrum is to mention a few of the science and technology events that are happening locally over the next few weeks. Lisa Katovich joins me for that Speaker 6: calendar. The August general meeting of the East Bay Astronomical Society is Saturday, July 14th at the Chabot space and science centers, Dellums [00:23:30] building 10,000 Skyline Boulevard in Oakland. Ezra Bahrani is the evening Speaker. The title of his talk is UFOs, the proof, the physics and why they're here. The meeting starts at 7:30 PM Speaker 2: join Nobel laureates and social and environmental justice advocates at the towns and Tay Gore third annual seminar for Science and technology on behalf of the peoples of Bengali and the Himalayan basins, the subject, the global water crisis [00:24:00] prevention and solution. Saturday, July 21st 1:30 PM to 7:30 PM the event is jointly sponsored by UC Berkeley's department of Public Health and the international institute of the Bengali and Himalayan basins. Guest Speakers include three Nobel laureates, Charles h towns, Burton Richter and Douglas Ashur off. Also presenting our Francis towns advocate for social justice, Dr. Rush, Gosh [00:24:30] and Sterling Brunel. The event will be held in one 45 Dwinelle hall on the UC Berkeley campus. That's Saturday, July 21st 1:30 PM to 7:30 PM for more details, contact the UC Berkeley School of Public Health, Speaker 6: the next science at cal lectures on July 21st the talk will be given by Dr Jeffrey Silverman and it's entitled exploding stars, Dark Energy, and the runaway universe. Dr Silverman has been a guest [00:25:00] on spectrum. His research has been in the study of Super Novi. His lecture will focus on how the study of supernovae led to the recent discovery that the universe is expanding, likely due to a repulsive and mysterious dark energy. It was these observations that were recently awarded the 2011 Nobel Prize in physics. The lecture is July 21st at 11:00 AM and the genetics and plant biology building room 100 Speaker 2: next to news stories. Speaker 6: 3000 species [00:25:30] of mosquitoes are responsible for malaria, dengue, a fever, yellow fever, West Nile virus, and cephalitis and many more diseases. In Burkina Faso alone, residents can expect 200 bytes a day. Rapid resistance to pesticides on the part of malaria mosquitoes has prompted researchers all over the globe to deploy novel strategies against this and other diseases. Targeting Dengue. A fever has an advantage over malaria as only one species. Eighties [00:26:00] Egypt die is responsible for spreading it versus the 20 species responsible for spreading malaria. A British biotechnology company called Oxitec has developed a method to modify the genetic structure of the male eighties Aegypti mosquito transforming it into a mutant capable of destroying its own species. In 2010 they announced impressive preliminary results of the first known test of 3 million free flying transgenic mosquitoes engineered [00:26:30] to start a population crash after infiltrating wild disease spreading eighties a Gyp dye swarms on Cayman Island. Speaker 6: Oxitec has recently applied to the FDA for approval of its mosquito in the u s with Key West under consideration as a future test site in 2009 key west suffered its first dengate outbreak in 73 years. Australian researchers are testing and mosquito intended to fight dengue, a fever bypassing the disruptive Wolbachia bacteria to other mosquitoes, a very [00:27:00] different approach than transgenic genes funded largely by the bill and Melinda Gates Foundation. The project has shown that the Wolbachia strain not only shortens the life of a mosquito, but also reduces the amount of virus it develops. Releases in Queensland, Australia last year showed that Wolbachia could spread through a wild population quickly and future test sites are under consideration. In Vietnam. Speaker 2: The UC Berkeley News Center reports a prototype network being installed by chemists at the University of California. Berkeley [00:27:30] will employ 40 sensors spread over a 27 square mile grid. The information the network will provide could be used to monitor local carbon dioxide emissions to check on the effectiveness of carbon reduction strategies now mandated by the state, but hard to verify built and installed by project leader Professor Ron Cohen and graduate student Virginia Tighe and their lab colleagues. The shoe box size sensors will continuously measure carbon dioxide, carbon monoxide, [00:28:00] nitrogen dioxide, and ozone levels as well as temperature, pressure and humidity streaming. The information live to the web through the site. beacon.berkeley.edu the sensor network dubbed Beacon stretches from the East Bay regional parks on the east to interstate eight 80 on the west from El Surrito on the north nearly to San Leandro on the south encompassing open space as well as heavily traffic areas. [00:28:30] Most of the sensors are being mounted on the roofs of local schools in order to get students interested in the connection between carbon dioxide emissions and climate change. The UC Berkeley researchers work with Oakland's Chabot space and science center to create middle school and high school activities using live sensor data stream through the web as part of the students energy and climate science curriculum. The beacon network is a pilot program funded by the National Science Foundation to determine what information can be learned [00:29:00] from a densely spaced network Speaker 1: [inaudible].Speaker 2: The music heard during the show is from most done at David's album, folk and acoustics made available through a creative Commons license 3.0 attribution. Speaker 1: Thank you for listening to spectrum. If you have comments about the show, please send them to us via email. Our email address [00:29:30] is spectrum dot kalx@yahoo.com join us in two weeks at this same time. [inaudible]. See acast.com/privacy for privacy and opt-out information.

Astronomy (Audio)
Extreme Science

Astronomy (Audio)

Play Episode Listen Later May 21, 2012 89:30


Learn why Spencer Klein goes to the ends of the Earth to search for ghostly neutrinos in Antarctica. From Chernobyl to Central Asia, Tamas Torok travels the globe to study microbial diversity in extreme environments. Andrew Minor uses the world's most advanced electron microscopes to explore materials at ultrahigh stresses and in harsh environments. And microbes that talk to computers? Caroline Ajo-Franklin is pioneering cellular-electrical connections that could help transform sunlight into fuel. Series: "Lawrence Berkeley National Laboratory " [Science] [Show ID: 23652]

Astronomy (Video)
Extreme Science

Astronomy (Video)

Play Episode Listen Later May 21, 2012 89:30


Learn why Spencer Klein goes to the ends of the Earth to search for ghostly neutrinos in Antarctica. From Chernobyl to Central Asia, Tamas Torok travels the globe to study microbial diversity in extreme environments. Andrew Minor uses the world's most advanced electron microscopes to explore materials at ultrahigh stresses and in harsh environments. And microbes that talk to computers? Caroline Ajo-Franklin is pioneering cellular-electrical connections that could help transform sunlight into fuel. Series: "Lawrence Berkeley National Laboratory " [Science] [Show ID: 23652]

Lawrence Berkeley National Lab (Video)

Learn why Spencer Klein goes to the ends of the Earth to search for ghostly neutrinos in Antarctica. From Chernobyl to Central Asia, Tamas Torok travels the globe to study microbial diversity in extreme environments. Andrew Minor uses the world's most advanced electron microscopes to explore materials at ultrahigh stresses and in harsh environments. And microbes that talk to computers? Caroline Ajo-Franklin is pioneering cellular-electrical connections that could help transform sunlight into fuel. Series: "Lawrence Berkeley National Laboratory " [Science] [Show ID: 23652]

Lawrence Berkeley National Lab (Audio)

Learn why Spencer Klein goes to the ends of the Earth to search for ghostly neutrinos in Antarctica. From Chernobyl to Central Asia, Tamas Torok travels the globe to study microbial diversity in extreme environments. Andrew Minor uses the world's most advanced electron microscopes to explore materials at ultrahigh stresses and in harsh environments. And microbes that talk to computers? Caroline Ajo-Franklin is pioneering cellular-electrical connections that could help transform sunlight into fuel. Series: "Lawrence Berkeley National Laboratory " [Science] [Show ID: 23652]

Spectrum
Joe Cordaro

Spectrum

Play Episode Listen Later Apr 20, 2012 30:00


Joe Cordaro is a principle member of the technical staff at Sandia National Laboratories in Livermore. He is a research chemist who received his PhD in chemistry from UC Berkeley. He talks with us about his work in concentrated solar power systems.TranscriptSpeaker 1: Spectrum's next Speaker 2: [inaudible].Speaker 1: Welcome to spectrum the science and technology show on k a l [00:00:30] x Berkeley, a biweekly 30 minute program bringing you interviews featuring bay area scientists and technologists as well as a calendar of local events and news [inaudible]. Speaker 3: Good afternoon. My name is Brad Swift. I'm joined today by spectrum contributors. Rick Karnofsky and Lisa [inaudible]. Rick and I interviewed Joe Carderock, a principal member of the technical staff at Sandia national laboratories in Livermore. He is a research chemist. [00:01:00] Joe received his phd in chemistry from UC Berkeley. He talks with us about his work in concentrated solar power systems. Joe, welcome to spectrum. Thank you. Rick. Can you explain to us a little bit about concerted solar power? Sure. I'd be happy to. People have looked at using mirrors to focus light to do exactly what we are now doing in the 21st century since the mid 17 and 18 hundreds. There's a few reports that people using mirrors to focus [00:01:30] sunlight to heat up water in a boiler to generate steam for creating a pump for irrigation. And there's also been a report of a printing press that was powered off of steam that was generated using mirrors to focus light to once again heat up a boiler. Speaker 3: Um, that all happened in the 19th early 20th century. But from about the early 1920s until the 1970s not a lot of work went into looking at concentrated solar power to make electricity. Primarily that was because at the same [00:02:00] time that research to make solar electricity from sunlight was taking off, oil was discovered and that became much cheaper and economical than it was to invest in technology to look at concentrated solar power. So concentrated solar power is a method by using in mirrors to focus the sun's rays onto a type of central receiver in order to boil water, to turn a turbine to generate electricity. So it's really a complicated way to boil water just to make electricity, but it works [00:02:30] and it only uses the sun. Is this sort of input for energy? Yeah, it's actually pretty amazing that we, that we don't use this more often because there is no emission from it. Speaker 3: There's no greenhouse gases, there's no radioactive material and it's mostly made using commodity parts that can almost 70% be made in the United States. So there's three main architectures for concentrated solar power. There's the sterling engine, there's parabolic trough systems and then a central receiver tower [00:03:00] vista. Then engines are maybe the most efficient type of concentrated solar power, but they also have the most moving parts and a reliability is somewhat low right now. Their module, so you can add one and then another and another and another and increase your field side to base on demand. You can also just stick one in your backyard if you had the money to buy it and uh, didn't mind the thumping noise at the sterling pump makes so they're a little loud. The most employed type of concentrated solar [00:03:30] power facility right now is a parabolic trough system. And in a parabolic trough system you have a field of mirrors that are focused on a central tube that runs through the parabolic trough. Speaker 3: And this tube is about three inches in diameter. And inside the tube is a working fluid and it's usually a silicon based oil. And the silicon based oil is used because the uh, operating temperature for that is around zero degrees Celsius up to 450. If you're in the desert, you typically have cold winter nights, [00:04:00] so you need to have a flu that doesn't solidify at nighttime in the wintertime. And so zeros are pretty good, that lower limit, but the a heat transfer fluid and based on silicon is slightly expensive. And how does that upper limit established? How hot can these things really go? So the upper limit would be the thermal stability of working fluid and the upper stability is just dependent on the chemical nature of the fluid. So the bond strengths of the actual carbon oxygen and silicon bonds within the heat transfer fluid. Speaker 3: But as far as the amount [00:04:30] of heat energy that can be sort of harvested, that's going to be dependent on the thermal heat capacity of the fluid times the actual density times the uh, flow rate. So the more heat you can store per volume per time will give you a more energy out at the end of the day. But all of that is gonna be dependent on factors like your thermal conductivity between the two betters holding the heat transfer fluid, and then also the heat exchangers that are down the line when you convert from a silicon [00:05:00] oil heat to steam heat. So there's a lot of limiting factors that control your efficiency of these things and a lot of losses. Also third type of concentrated solar power facility called the central receiver tower. And in those systems you have one tower that could maybe be 50 to a a hundred meters above the ground and that tower surrounded by field of mirrors and those mirrors are flat. Speaker 3: I also call them heliostats and those mirrors track the sun and then reflect the sun's rays onto the central receiver tower. And [00:05:30] the essential receiver tower has a molten salt inside of it and the temperature that usually goes up to about 550 degrees Celsius. And the reason why we're using molten salt is because you can get a higher operating temperature. Then you count the silicon fluid and this molten salt heats up to its operating temperature, which has been pumped only a short distance to a heat exchanger, which then boils water to turn a turbine to make electricity. Speaker 2: This is spectrum on k a l x [00:06:00] Berkeley. We are talking with Joe Cordaro of Sandia national laboratories about concentrated solar power. Speaker 3: And Are we limited at all about where we would deploy a concentrated solar power plants or are these all going to end up in the deserts of Arizona or so one of the main limitations for concentrated solar powers that you need to have good sunlight, you need to do need to have many, many days of sunlight [00:06:30] per year with a high intensity. So putting a concentrated solar power field up in northern Europe or the northeast of the United States doesn't always make sense economically. It's a much better to put it in the desert in California or Arizona or New Mexico or Utah or in Africa. So the key being cloud free, cloud free with a lower latitudes. And how prevalent are concentrated solar power plants right now? Well, [00:07:00] they're building them pretty rapidly, but I think the total percentage of the electricity we get in the United States, it's probably less than 1%, but they're building these plants in California and Arizona, especially essential receiver towers. Speaker 3: There's a big plant being built in Ivanpah, which is outside of Barstow. There's a couple of being built outside Las Vegas and Phoenix. They're building them in Morocco. They're building them in Italy. There's quite a few in Spain and there's some in France. Israel is building them. The amount of electricity [00:07:30] coming from these plants is uh, increasing, but it's still nothing compared to coal or natural gas. So essentially receiver towers are being explored a lot more because they have the potential for higher efficiency because you can go to higher temperature. So the carnow efficiency basically says that the higher difference in temperature between your hot and cold for doing work gives you the higher efficiency. So if you can increase your high operating temperature to five, six, seven, 800 degrees Celsius, but keep [00:08:00] your low operating temperature is still above the boiling point of water, you'll have a much more efficient cycle. Speaker 3: So if you're limited by our heat transfer fluid, thermal stability of 450 degrees, then you're uh, overall fishing in the plant will be limited. So a lot of the work that the Department of energy is doing to try to improve the efficiencies of these systems is to look at higher operating temperatures. But with higher operating temperatures comes also a materials compatibility issues. And then also higher losses. So as you go to higher temperature, you not only get better [00:08:30] efficiency for your carnow efficiency, but you also get higher radiative losses. So you actually start to lose more heat throughout your whole system. And your materials become more difficult to match. And Costco, Costco really high. And why is that? Well, materials are becoming a big issue. There's not a lot of industries that currently use high temperature materials that except the nuclear industry. So if you want to do large scale industrial power plants, you really [00:09:00] want to stick with commodity items that can be purchased cheaply. Speaker 3: Otherwise the costs are too expensive. So there's a lot of analysis that goes into try to decide if I increase my temperature by just 200 degrees or even a hundred degrees, is the efficiency gain worth the cost? So one of the big issues with these costs and material selection are the corrosion issues with your heat transfer fluid. So if you have a fluid that's operating at 700 or 800 degrees Celsius, you start to have incompatible [00:09:30] materials between your heat transfer fluid and the actual material of the pipe is made out of, I don't know, most of these salt baths, very simple sort of two ion component systems like this. Well the only actual molten salt used in the fields now are based off of sodium, potassium nitrate and nitrite mixture. So there are four components, two to four components, and they're pretty simple. But they do have reactive properties with a lot of alloys. Speaker 3: So there are still some [00:10:00] corrosion issues, especially when you get above 550 degrees. So there's the longterm stability of the molten salt bath or the molten salt storage tank, or the molten salt pipes that have to be considered because it's a 30 year plant that leave expected design. So most power plants are built with the idea that it's going to have a 30 year lifetime. So you have to figure out what's gonna happen over 30 years. And the rate of a simple chemical reaction usually doubles with every 10 degrees increase in temperature. So if you have a simple first order [00:10:30] reaction, like the decomposition of a Moan Salt, and you increase the temperature by 10 degrees, you can expect your rate to double. And so that starts to really matter. If you're looking at something that's going to be a 30 year lifetime, Speaker 2: you were listening to spectrum on k a l x Berkeley. Brad swift and Rick Karnofsky are talking with Joe Cordura about concentrated solar power and [inaudible]. Speaker 3: [00:11:00] So how intense is the beam once all these mirrors reflected into the molten salts? The central receiver tower like I described, has a large receiving window that maybe 10 by 10 meters and it's a target area that's painted black in order to absorb as much sunlight as possible from maybe a hundred, maybe 200 or maybe a thousand mirrors in the field, and they're focusing the sun's energy onto the central target in order to [00:11:30] get a really, really high temperature so that you can heat up some working heat transfer fluid. There's a way that a lot of the engineer's describe the intensity is it by the number of sons that are being focused onto that area and you're focusing all of those mirrors on a central spot, but you can get up to 3000 suns mean focused onto a single spot. 3000 suns is quite a high amount of energy and also very high temperature and there have been reports of birds that have flown [00:12:00] in the path of the sun. It's hot enough that they've burst into a little ball of fire and then fallen down into a fiery death below. Fortunately, it's only a few birds every once in a while, but that's how hot it does get in front of those receivers. You get nowhere that high of intensity and a parabolic trough system because you only have one large curved, mere focusing the sunlight onto a tube rather than hundreds of mirrors all focusing onto a central receiver. Speaker 3: [00:12:30] Can you explain more about how you store the, is it the heat you're storing? Are you, what are you storing actually, so one of the biggest advantage of concentrated solar power is the ability to store thermal heat. When you use the sun to generate electricity, you're depending on the sun's sunlight to be consistent on the race to be consistent. And if a cloud goes in front of the sun and you're generate electricity using photovoltaics, your power drops to zero until the cloud moves [00:13:00] out of the sky. At nighttime, you can't generate any electricity either cause you don't have any sun. If you look at the peak demand time for electricity in the United States, it tracks with the date, time sun, which is good. But then it also continues into the evening until six seven eight o'clock at night when everyone comes home at night and turns on their washer and dryer turns on their television and it turns on their dishwasher. Speaker 3: If you don't have any electricity on the grid available, then you're going to have a big problem. Coal and nuclear power plants can just generate electricity 24 hours a day without any problem. So [00:13:30] concentrated solar power offers the ability to do that as well through what we call thermal storage. So if you have a huge field of parabolic troughs that are heating up a heat transfer fluid to a high temperature, you can then take this fluid and store it into a large tank. And this hot fluid is going to stay hot for eight 1220 hours to pay on how big you build that tank. So now if you have a hot tank that's storing all of this heat, you can draw heat from that tank rather than drawing it from the field. [00:14:00] So you can decouple the power generation cycle from the actual solar sunlight. Speaker 3: So the tank is kept at a high temperature and constantly being recharged by the sun. But if the sun disappears, you have a reserve of fluid that's still hot that you can use to generate electricity by boiling water. And the size of that tank is dependent on how many hours of storage you want. So people will make these tanks based off of an eight hour storage cycle or a 10 hour or 12 hour [00:14:30] storage time. So typically they're made up of an eight hour storage time because no one needs a lot of electricity at four, five in the morning, and then the sun comes back up again and you can start your whole plant back up. And that makes it much easier to tie into the grid and much easier to distribute electricity to the population. So what we call a dispatchable electricity generation. That's a big advantage for concentrated solar power compared to wind or photovoltaics and what [00:15:00] happens to the system if the outage is longer so you don't just have to worry about nights they have to worry about clouds or dust storms or, so there's a lot of potential backups that can be engineered into a system. Speaker 3: One of them being gas powered burners just put in line to boil water to power the system in reverse basically. So if there was a really big problem where you had no sunlight for a week, could potentially use natural [00:15:30] gas burners to boil water but cycle it in reverse and so then the water goes and operates as a heat transfer fluid actually warm up the salt again. Fortunately historical data I think shows that that just is not a big risk. I mean you wouldn't build a plant in the northeast where you actually could have a week of cloud cover and cold rainy weather. You'd build a plant in the desert and a week of no sun doesn't happen. There's been plants that have been in operation for 30 years [00:16:00] in the desert in California, and there's historical data that is available to kind of map out where in the world you would build these plants. Speaker 3: That goes back many, many, many years and the Department of Energy has collected this data, specifically the national renewable energy lab. Our enrol in Colorado has a lot of this data and industry and the national labs work strongly together to try to figure out where the best places to build these plants that have not only the highest solar [00:16:30] radiation, but also the lowest environmental impact when you build a plant because despite it being a zero emitter of greenhouse gases, there are environmental issues related to water usage and also endangered species and the Atlantan usage. Pretty big. Yeah, they can be quite large. So there are some land issues that are associated with building a system in the middle of the desert. There's also issues about how do you get the electricity to where consumers actually [00:17:00] live. If you build a power plant in the middle of the desert but everyone lives a couple hundred miles away or thousands of miles away, how do you actually get the electricity to more populated areas? And this is an issue Europe is dealing with because they want to build power plant in North Africa and then have electricity ship to continental Europe somehow. So it's another topic, but they're looking at ways to make high voltage DC transmission lines from northern Europe down into Africa. So you can actually distribute the electricity from where it's generated. Speaker 2: [inaudible]Speaker 3: [00:17:30] Joe Cornaro is our guest. The show is spectrum. The station is k a l x Berkeley. The topic is concentrated solar power. Speaker 3: And what are some of the other open research questions that are out there besides the materials compatibility issues that you, some of the other areas are looking at. How do you actually set up a field of mirrors that maybe [00:18:00] 50 acres big and then get everyone in those mirrors to actually align properly without making it an incredibly expensive task. So all of these mirrors have to track the sun at the same angle and you have to figure out how can you put all these mirrors on some type of mechanical platform that moves to track the sun and then direct the sunlight efficiently. Cause just a small error in one of the mirrors can really change your beam and decrease your efficiency quite significantly. [00:18:30] You also have to think about what happens when a big wind storm comes around in the desert and you have 70 mile an hour winds. Speaker 3: Now all the mirrors have to be stowed, turned pretty much horizontal so that they don't get blown away. Then you have to worry about the sand that comes by and and polishes. The mirrors are unpolished as them heres so there's a lot of technology goes into the coatings figuring out new pumps, valves and fittings when you're running at 800 degrees. So you can pump a fluid at 500 degrees. We have commercial equipment to do [00:19:00] that, but using that equipment at 700 or 800 degrees hasn't been tested. So manufacturers will make things that they say possibly will work at 800 but it's not actually been tested at 800 and then we don't even have sensors to measure things that 800 on a large scale like this to measure what kinds of things? A viscosity is a big one. So we want to know how fast a fluid is flowing through a pipe so we can calculate how much heat is coming out. Speaker 3: So we know how much steam we're going to generate and try [00:19:30] to measure viscosity at 800 degrees hasn't been done either. So we have active programs to look at making new sensors for viscosity. Some of the other issues, I'm trying to get more efficient steam cycles. Actually there are commercially available turbines to make steam for the uh, colon, natural gas industry that have been around for 50 75 years and they work really well up to a certain temperature. But if you can go higher with your heat transfer fluid, then you want to go higher with your turbine as well. And then [00:20:00] using steam no longer as efficient. And so people are looking at other types of cycles that don't use water anymore to make steam, but they're using super critical CO2 or helium or some other type of gas for what we call air brain cycles. Speaker 3: And those could operate up to 1200 degrees and Japan has actually looked at those for quite awhile, but America has been pretty scared of looking at a 1200 degree high pressure systems. As far as the risk. Yeah, as far as the risk goes, it is a little bit more dangerous [00:20:30] when you have 1200 degrees and high high pressure systems, but the efficiency could be a lot higher. So all of this is still open for optimization. All of it requires inputs from systems engineers to finance people to determine the cost, whether it's worth it down to scientists, to the Terman stability and compatibility of parts to the last thing you want to do is build a big field and then have to replace a huge [00:21:00] portion of it in three years because you have something break that'll make the entire project economically a nonstarter. So the risks have to be reduced to save as much as possible. Speaker 4: Joe, how was it? Did you became involved in concentrated solar power? Speaker 3: After I got to Sandia national labs, I began working in the concentrated solar power research project because I was a chemist in looking at materials, compatibility issues and also stability issues of heat transfer fluids and while it doesn't sound like the most sexy [00:21:30] area of chemistry to be in formulating new salts and looking at high temperature materials, I really, really enjoy it because it is actually being built is actually real science being turned into engineering projects that is actually being deployed throughout the world to solve our problems and to make us energy independent. So unlike a lot of academic research that I did in school, concentrated solar power is real. It's been done and it's been put to use and that makes me incredibly [00:22:00] excited about being part of that project. Joe Codero, thanks for joining us. Thank you for having me. Speaker 2: Regular feature of spectrum is to mention a few of the science and technology events happening in the bay area over the next few weeks. Rick and Lisa, join me for the calendar. Speaker 5: UC Berkeley's Institute of East Asian Studies [00:22:30] will hold a symposium titled Towards Longterm Sustainability in response to the Fukushima nuclear disaster. It takes place today and tomorrow and it starts soon, one 30 to five 30 today, so you better hurry up and get over there, but if you can't make it today, tomorrow will feature three Speakers, all of whom have been actively involved in analyzing the Fukushima nuclear plant accident, its historical context, and the sociopolitical actions taken by the various stakeholders. The symposium [00:23:00] will situate the causes and the consequences of the disaster in the context of a longterm sustainable future. For more information, go to the website, I. E. A s@berkeley.edu Speaker 4: cal day is tomorrow, Saturday, April 21st the Berkeley campus, the museums, the botanical garden are open to the public. There are a wide variety of presentations and facilities you can tour for details, go to the website, cal day.berkeley.edu Speaker 5: [00:23:30] on June 5th, 2012 Venus will transit or pass directly in front of the sun. A transit like this is so rare. No human alive today. We'll witness it again. The next one will not be until 2117 get ready. This event by going to the transit of Venus Planetarium program at the Lawrence Hall of science this Saturday on cow day at 3:00 PM learn why transits are so rare, how studying transits taught us exactly how big our solar system is [00:24:00] and how they may be the key to discovering other earths and other star systems. Then come back on June 5th and observed the actual transit of Venus at the Lawrence Hall of Science. The hall will have several solar telescopes for viewing the eclipse safely on the main plaza. Most of us are aware of the obesity epidemic facing the United States, but is it simply a matter of calories in, calories out on Thursday, May 3rd from 1210 to 1:00 PM in the auditorium of the Berkeley Art Museum, [00:24:30] you CSF neuroendocrinologist Robert Lustig will present the lecture health, Darwin Diet disease and dollars. He will examine some of the more controversial dietary factors contributing to the obesity epidemic, the role that these obesogens potentially play in our evolution toward an unhealthy nation. And possible solutions for turning this trend around. You must register for this event. Go to u h S. Dot. berkeley.edu Speaker 6: [00:25:00] on Saturday April 28th at 1:30 PM the Commonwealth Club and the Youth Science Initiative. Host the research group lead for Pixar and our guest on spectrum two weeks from today, Tony rose. Senator, the admission is $20 Commonwealth Club members get in for 12 Speaker 6: and is $7 for students 18 and under. The talk will be at the Los Altos High School Eagle Theater, two zero one almond avenue in Los Altos. Tony will discuss how math [00:25:30] is central to Pixar film production process and also the young makers program. That's the topic of our interview. In the next episode of spectrum, students are teamed up with adult mentors to design and build ambitious projects for the maker fair for tickets and more information, visit www.commonwealthclub.org another feature is spectrum guest Maggie Court. Baker will also be giving a lecture soon. Maggie is the science editor of Boeing, boeing.net and we'll be discussing her recent book before the lights go [00:26:00] out, conquering the energy crisis before it conquers us. She'll put the fun back in the infrastructure and described the surprising ways our electric system evolved, what we can and can't do about the energy crisis now and what the future might hold. This is the spring seminar for the Berkeley Science Review and will take place in the lead caching building room. Three four five on Wednesday May 2nd at 6:00 PM yeah, RSVP At B e r c. Dot. berkeley.edu [00:26:30] pseudo room, a newly forming East Bay hackerspace is having a free kickoff and fundraiser on Friday May 4th at 7:00 PM at Tech Liminal two six eight 14th street in downtown Oakland. Okay. Pseudo room is a collaborative community of tech developers, citizen scientists, activists and artists. Mitch Altman, cofounder of Noisebridge. We'll discuss hackerspaces for more information, visit s u d o room.org [00:27:00] now the news Speaker 5: significant declines are expected in the number of emperor penguins over the next century due to earlier spring warming around Antarctica. A new study in the April 13th edition of Science Daily reports that an international team of scientists using satellite mapping technology reveals there are twice as many emperor penguins in Antarctica than previously thought. Using a technique known as pan sharpening to increase the resolution of the satellite imagery. They were able to differentiate between birds, [00:27:30] eye shadow and Penguin Guano. In the first comprehensive census of a species taken from space 595,000 birds were counted almost double the previous estimates. Speaker 6: The origin of cosmic grays has long been and remains a mystery. The ice cube collaboration in which Berkeley lab is a crucial contributor published in an article in the April 18th issue of nature on their exhaustive search for a high energy neutrinos that would likely be produced if the violent extra galactic [00:28:00] explosions known as Gamma Ray bursts are a source of ultra high energy cosmic rays. They I know events they have correspondents to these bursts when they would predict to see at least 8.4 events that correspond to some of the 215 gamma ray bursts detected from two periods in 2008 and 2009 there are other popular models for the origin of cosmic rays including active galactic nuclei. The Ice Cube Neutrino telescope encompasses a cubic kilometer of ice under [00:28:30] the South Pole and has over 5,000 digital optical modules that track the direction and energy of speeding yuan's which are created when you Trina is collide with Adam's in the ice. On a later episode of spectrum, you'll hear from Spencer Klein and Thorsten Settle Berger about this experiment. Visit ice cube dot [inaudible] w I s c.edu for more information, Speaker 2: thanks to Rick Kaneski [00:29:00] and Lisa cabbage for help producing show music heard during the show is by Lasagna David from his album, folk and acoustic made available through creative Commons attribution license 3.0 spectrum shows are now available online at iTunes university. Go to itunes.berkeley.edu thank you for listening to spectrum. If you have comments about the show, please send [inaudible] [00:29:30] email address is spectrum dot [inaudible] dot com join us in two weeks. Same time. [inaudible]. See acast.com/privacy for privacy and opt-out information.

Spectrum
Joe Cordaro

Spectrum

Play Episode Listen Later Apr 20, 2012 30:00


Joe Cordaro is a principle member of the technical staff at Sandia National Laboratories in Livermore. He is a research chemist who received his PhD in chemistry from UC Berkeley. He talks with us about his work in concentrated solar power systems.TranscriptSpeaker 1: Spectrum's next Speaker 2: [inaudible].Speaker 1: Welcome to spectrum the science and technology show on k a l [00:00:30] x Berkeley, a biweekly 30 minute program bringing you interviews featuring bay area scientists and technologists as well as a calendar of local events and news [inaudible]. Speaker 3: Good afternoon. My name is Brad Swift. I'm joined today by spectrum contributors. Rick Karnofsky and Lisa [inaudible]. Rick and I interviewed Joe Carderock, a principal member of the technical staff at Sandia national laboratories in Livermore. He is a research chemist. [00:01:00] Joe received his phd in chemistry from UC Berkeley. He talks with us about his work in concentrated solar power systems. Joe, welcome to spectrum. Thank you. Rick. Can you explain to us a little bit about concerted solar power? Sure. I'd be happy to. People have looked at using mirrors to focus light to do exactly what we are now doing in the 21st century since the mid 17 and 18 hundreds. There's a few reports that people using mirrors to focus [00:01:30] sunlight to heat up water in a boiler to generate steam for creating a pump for irrigation. And there's also been a report of a printing press that was powered off of steam that was generated using mirrors to focus light to once again heat up a boiler. Speaker 3: Um, that all happened in the 19th early 20th century. But from about the early 1920s until the 1970s not a lot of work went into looking at concentrated solar power to make electricity. Primarily that was because at the same [00:02:00] time that research to make solar electricity from sunlight was taking off, oil was discovered and that became much cheaper and economical than it was to invest in technology to look at concentrated solar power. So concentrated solar power is a method by using in mirrors to focus the sun's rays onto a type of central receiver in order to boil water, to turn a turbine to generate electricity. So it's really a complicated way to boil water just to make electricity, but it works [00:02:30] and it only uses the sun. Is this sort of input for energy? Yeah, it's actually pretty amazing that we, that we don't use this more often because there is no emission from it. Speaker 3: There's no greenhouse gases, there's no radioactive material and it's mostly made using commodity parts that can almost 70% be made in the United States. So there's three main architectures for concentrated solar power. There's the sterling engine, there's parabolic trough systems and then a central receiver tower [00:03:00] vista. Then engines are maybe the most efficient type of concentrated solar power, but they also have the most moving parts and a reliability is somewhat low right now. Their module, so you can add one and then another and another and another and increase your field side to base on demand. You can also just stick one in your backyard if you had the money to buy it and uh, didn't mind the thumping noise at the sterling pump makes so they're a little loud. The most employed type of concentrated solar [00:03:30] power facility right now is a parabolic trough system. And in a parabolic trough system you have a field of mirrors that are focused on a central tube that runs through the parabolic trough. Speaker 3: And this tube is about three inches in diameter. And inside the tube is a working fluid and it's usually a silicon based oil. And the silicon based oil is used because the uh, operating temperature for that is around zero degrees Celsius up to 450. If you're in the desert, you typically have cold winter nights, [00:04:00] so you need to have a flu that doesn't solidify at nighttime in the wintertime. And so zeros are pretty good, that lower limit, but the a heat transfer fluid and based on silicon is slightly expensive. And how does that upper limit established? How hot can these things really go? So the upper limit would be the thermal stability of working fluid and the upper stability is just dependent on the chemical nature of the fluid. So the bond strengths of the actual carbon oxygen and silicon bonds within the heat transfer fluid. Speaker 3: But as far as the amount [00:04:30] of heat energy that can be sort of harvested, that's going to be dependent on the thermal heat capacity of the fluid times the actual density times the uh, flow rate. So the more heat you can store per volume per time will give you a more energy out at the end of the day. But all of that is gonna be dependent on factors like your thermal conductivity between the two betters holding the heat transfer fluid, and then also the heat exchangers that are down the line when you convert from a silicon [00:05:00] oil heat to steam heat. So there's a lot of limiting factors that control your efficiency of these things and a lot of losses. Also third type of concentrated solar power facility called the central receiver tower. And in those systems you have one tower that could maybe be 50 to a a hundred meters above the ground and that tower surrounded by field of mirrors and those mirrors are flat. Speaker 3: I also call them heliostats and those mirrors track the sun and then reflect the sun's rays onto the central receiver tower. And [00:05:30] the essential receiver tower has a molten salt inside of it and the temperature that usually goes up to about 550 degrees Celsius. And the reason why we're using molten salt is because you can get a higher operating temperature. Then you count the silicon fluid and this molten salt heats up to its operating temperature, which has been pumped only a short distance to a heat exchanger, which then boils water to turn a turbine to make electricity. Speaker 2: This is spectrum on k a l x [00:06:00] Berkeley. We are talking with Joe Cordaro of Sandia national laboratories about concentrated solar power. Speaker 3: And Are we limited at all about where we would deploy a concentrated solar power plants or are these all going to end up in the deserts of Arizona or so one of the main limitations for concentrated solar powers that you need to have good sunlight, you need to do need to have many, many days of sunlight [00:06:30] per year with a high intensity. So putting a concentrated solar power field up in northern Europe or the northeast of the United States doesn't always make sense economically. It's a much better to put it in the desert in California or Arizona or New Mexico or Utah or in Africa. So the key being cloud free, cloud free with a lower latitudes. And how prevalent are concentrated solar power plants right now? Well, [00:07:00] they're building them pretty rapidly, but I think the total percentage of the electricity we get in the United States, it's probably less than 1%, but they're building these plants in California and Arizona, especially essential receiver towers. Speaker 3: There's a big plant being built in Ivanpah, which is outside of Barstow. There's a couple of being built outside Las Vegas and Phoenix. They're building them in Morocco. They're building them in Italy. There's quite a few in Spain and there's some in France. Israel is building them. The amount of electricity [00:07:30] coming from these plants is uh, increasing, but it's still nothing compared to coal or natural gas. So essentially receiver towers are being explored a lot more because they have the potential for higher efficiency because you can go to higher temperature. So the carnow efficiency basically says that the higher difference in temperature between your hot and cold for doing work gives you the higher efficiency. So if you can increase your high operating temperature to five, six, seven, 800 degrees Celsius, but keep [00:08:00] your low operating temperature is still above the boiling point of water, you'll have a much more efficient cycle. Speaker 3: So if you're limited by our heat transfer fluid, thermal stability of 450 degrees, then you're uh, overall fishing in the plant will be limited. So a lot of the work that the Department of energy is doing to try to improve the efficiencies of these systems is to look at higher operating temperatures. But with higher operating temperatures comes also a materials compatibility issues. And then also higher losses. So as you go to higher temperature, you not only get better [00:08:30] efficiency for your carnow efficiency, but you also get higher radiative losses. So you actually start to lose more heat throughout your whole system. And your materials become more difficult to match. And Costco, Costco really high. And why is that? Well, materials are becoming a big issue. There's not a lot of industries that currently use high temperature materials that except the nuclear industry. So if you want to do large scale industrial power plants, you really [00:09:00] want to stick with commodity items that can be purchased cheaply. Speaker 3: Otherwise the costs are too expensive. So there's a lot of analysis that goes into try to decide if I increase my temperature by just 200 degrees or even a hundred degrees, is the efficiency gain worth the cost? So one of the big issues with these costs and material selection are the corrosion issues with your heat transfer fluid. So if you have a fluid that's operating at 700 or 800 degrees Celsius, you start to have incompatible [00:09:30] materials between your heat transfer fluid and the actual material of the pipe is made out of, I don't know, most of these salt baths, very simple sort of two ion component systems like this. Well the only actual molten salt used in the fields now are based off of sodium, potassium nitrate and nitrite mixture. So there are four components, two to four components, and they're pretty simple. But they do have reactive properties with a lot of alloys. Speaker 3: So there are still some [00:10:00] corrosion issues, especially when you get above 550 degrees. So there's the longterm stability of the molten salt bath or the molten salt storage tank, or the molten salt pipes that have to be considered because it's a 30 year plant that leave expected design. So most power plants are built with the idea that it's going to have a 30 year lifetime. So you have to figure out what's gonna happen over 30 years. And the rate of a simple chemical reaction usually doubles with every 10 degrees increase in temperature. So if you have a simple first order [00:10:30] reaction, like the decomposition of a Moan Salt, and you increase the temperature by 10 degrees, you can expect your rate to double. And so that starts to really matter. If you're looking at something that's going to be a 30 year lifetime, Speaker 2: you were listening to spectrum on k a l x Berkeley. Brad swift and Rick Karnofsky are talking with Joe Cordura about concentrated solar power and [inaudible]. Speaker 3: [00:11:00] So how intense is the beam once all these mirrors reflected into the molten salts? The central receiver tower like I described, has a large receiving window that maybe 10 by 10 meters and it's a target area that's painted black in order to absorb as much sunlight as possible from maybe a hundred, maybe 200 or maybe a thousand mirrors in the field, and they're focusing the sun's energy onto the central target in order to [00:11:30] get a really, really high temperature so that you can heat up some working heat transfer fluid. There's a way that a lot of the engineer's describe the intensity is it by the number of sons that are being focused onto that area and you're focusing all of those mirrors on a central spot, but you can get up to 3000 suns mean focused onto a single spot. 3000 suns is quite a high amount of energy and also very high temperature and there have been reports of birds that have flown [00:12:00] in the path of the sun. It's hot enough that they've burst into a little ball of fire and then fallen down into a fiery death below. Fortunately, it's only a few birds every once in a while, but that's how hot it does get in front of those receivers. You get nowhere that high of intensity and a parabolic trough system because you only have one large curved, mere focusing the sunlight onto a tube rather than hundreds of mirrors all focusing onto a central receiver. Speaker 3: [00:12:30] Can you explain more about how you store the, is it the heat you're storing? Are you, what are you storing actually, so one of the biggest advantage of concentrated solar power is the ability to store thermal heat. When you use the sun to generate electricity, you're depending on the sun's sunlight to be consistent on the race to be consistent. And if a cloud goes in front of the sun and you're generate electricity using photovoltaics, your power drops to zero until the cloud moves [00:13:00] out of the sky. At nighttime, you can't generate any electricity either cause you don't have any sun. If you look at the peak demand time for electricity in the United States, it tracks with the date, time sun, which is good. But then it also continues into the evening until six seven eight o'clock at night when everyone comes home at night and turns on their washer and dryer turns on their television and it turns on their dishwasher. Speaker 3: If you don't have any electricity on the grid available, then you're going to have a big problem. Coal and nuclear power plants can just generate electricity 24 hours a day without any problem. So [00:13:30] concentrated solar power offers the ability to do that as well through what we call thermal storage. So if you have a huge field of parabolic troughs that are heating up a heat transfer fluid to a high temperature, you can then take this fluid and store it into a large tank. And this hot fluid is going to stay hot for eight 1220 hours to pay on how big you build that tank. So now if you have a hot tank that's storing all of this heat, you can draw heat from that tank rather than drawing it from the field. [00:14:00] So you can decouple the power generation cycle from the actual solar sunlight. Speaker 3: So the tank is kept at a high temperature and constantly being recharged by the sun. But if the sun disappears, you have a reserve of fluid that's still hot that you can use to generate electricity by boiling water. And the size of that tank is dependent on how many hours of storage you want. So people will make these tanks based off of an eight hour storage cycle or a 10 hour or 12 hour [00:14:30] storage time. So typically they're made up of an eight hour storage time because no one needs a lot of electricity at four, five in the morning, and then the sun comes back up again and you can start your whole plant back up. And that makes it much easier to tie into the grid and much easier to distribute electricity to the population. So what we call a dispatchable electricity generation. That's a big advantage for concentrated solar power compared to wind or photovoltaics and what [00:15:00] happens to the system if the outage is longer so you don't just have to worry about nights they have to worry about clouds or dust storms or, so there's a lot of potential backups that can be engineered into a system. Speaker 3: One of them being gas powered burners just put in line to boil water to power the system in reverse basically. So if there was a really big problem where you had no sunlight for a week, could potentially use natural [00:15:30] gas burners to boil water but cycle it in reverse and so then the water goes and operates as a heat transfer fluid actually warm up the salt again. Fortunately historical data I think shows that that just is not a big risk. I mean you wouldn't build a plant in the northeast where you actually could have a week of cloud cover and cold rainy weather. You'd build a plant in the desert and a week of no sun doesn't happen. There's been plants that have been in operation for 30 years [00:16:00] in the desert in California, and there's historical data that is available to kind of map out where in the world you would build these plants. Speaker 3: That goes back many, many, many years and the Department of Energy has collected this data, specifically the national renewable energy lab. Our enrol in Colorado has a lot of this data and industry and the national labs work strongly together to try to figure out where the best places to build these plants that have not only the highest solar [00:16:30] radiation, but also the lowest environmental impact when you build a plant because despite it being a zero emitter of greenhouse gases, there are environmental issues related to water usage and also endangered species and the Atlantan usage. Pretty big. Yeah, they can be quite large. So there are some land issues that are associated with building a system in the middle of the desert. There's also issues about how do you get the electricity to where consumers actually [00:17:00] live. If you build a power plant in the middle of the desert but everyone lives a couple hundred miles away or thousands of miles away, how do you actually get the electricity to more populated areas? And this is an issue Europe is dealing with because they want to build power plant in North Africa and then have electricity ship to continental Europe somehow. So it's another topic, but they're looking at ways to make high voltage DC transmission lines from northern Europe down into Africa. So you can actually distribute the electricity from where it's generated. Speaker 2: [inaudible]Speaker 3: [00:17:30] Joe Cornaro is our guest. The show is spectrum. The station is k a l x Berkeley. The topic is concentrated solar power. Speaker 3: And what are some of the other open research questions that are out there besides the materials compatibility issues that you, some of the other areas are looking at. How do you actually set up a field of mirrors that maybe [00:18:00] 50 acres big and then get everyone in those mirrors to actually align properly without making it an incredibly expensive task. So all of these mirrors have to track the sun at the same angle and you have to figure out how can you put all these mirrors on some type of mechanical platform that moves to track the sun and then direct the sunlight efficiently. Cause just a small error in one of the mirrors can really change your beam and decrease your efficiency quite significantly. [00:18:30] You also have to think about what happens when a big wind storm comes around in the desert and you have 70 mile an hour winds. Speaker 3: Now all the mirrors have to be stowed, turned pretty much horizontal so that they don't get blown away. Then you have to worry about the sand that comes by and and polishes. The mirrors are unpolished as them heres so there's a lot of technology goes into the coatings figuring out new pumps, valves and fittings when you're running at 800 degrees. So you can pump a fluid at 500 degrees. We have commercial equipment to do [00:19:00] that, but using that equipment at 700 or 800 degrees hasn't been tested. So manufacturers will make things that they say possibly will work at 800 but it's not actually been tested at 800 and then we don't even have sensors to measure things that 800 on a large scale like this to measure what kinds of things? A viscosity is a big one. So we want to know how fast a fluid is flowing through a pipe so we can calculate how much heat is coming out. Speaker 3: So we know how much steam we're going to generate and try [00:19:30] to measure viscosity at 800 degrees hasn't been done either. So we have active programs to look at making new sensors for viscosity. Some of the other issues, I'm trying to get more efficient steam cycles. Actually there are commercially available turbines to make steam for the uh, colon, natural gas industry that have been around for 50 75 years and they work really well up to a certain temperature. But if you can go higher with your heat transfer fluid, then you want to go higher with your turbine as well. And then [00:20:00] using steam no longer as efficient. And so people are looking at other types of cycles that don't use water anymore to make steam, but they're using super critical CO2 or helium or some other type of gas for what we call air brain cycles. Speaker 3: And those could operate up to 1200 degrees and Japan has actually looked at those for quite awhile, but America has been pretty scared of looking at a 1200 degree high pressure systems. As far as the risk. Yeah, as far as the risk goes, it is a little bit more dangerous [00:20:30] when you have 1200 degrees and high high pressure systems, but the efficiency could be a lot higher. So all of this is still open for optimization. All of it requires inputs from systems engineers to finance people to determine the cost, whether it's worth it down to scientists, to the Terman stability and compatibility of parts to the last thing you want to do is build a big field and then have to replace a huge [00:21:00] portion of it in three years because you have something break that'll make the entire project economically a nonstarter. So the risks have to be reduced to save as much as possible. Speaker 4: Joe, how was it? Did you became involved in concentrated solar power? Speaker 3: After I got to Sandia national labs, I began working in the concentrated solar power research project because I was a chemist in looking at materials, compatibility issues and also stability issues of heat transfer fluids and while it doesn't sound like the most sexy [00:21:30] area of chemistry to be in formulating new salts and looking at high temperature materials, I really, really enjoy it because it is actually being built is actually real science being turned into engineering projects that is actually being deployed throughout the world to solve our problems and to make us energy independent. So unlike a lot of academic research that I did in school, concentrated solar power is real. It's been done and it's been put to use and that makes me incredibly [00:22:00] excited about being part of that project. Joe Codero, thanks for joining us. Thank you for having me. Speaker 2: Regular feature of spectrum is to mention a few of the science and technology events happening in the bay area over the next few weeks. Rick and Lisa, join me for the calendar. Speaker 5: UC Berkeley's Institute of East Asian Studies [00:22:30] will hold a symposium titled Towards Longterm Sustainability in response to the Fukushima nuclear disaster. It takes place today and tomorrow and it starts soon, one 30 to five 30 today, so you better hurry up and get over there, but if you can't make it today, tomorrow will feature three Speakers, all of whom have been actively involved in analyzing the Fukushima nuclear plant accident, its historical context, and the sociopolitical actions taken by the various stakeholders. The symposium [00:23:00] will situate the causes and the consequences of the disaster in the context of a longterm sustainable future. For more information, go to the website, I. E. A s@berkeley.edu Speaker 4: cal day is tomorrow, Saturday, April 21st the Berkeley campus, the museums, the botanical garden are open to the public. There are a wide variety of presentations and facilities you can tour for details, go to the website, cal day.berkeley.edu Speaker 5: [00:23:30] on June 5th, 2012 Venus will transit or pass directly in front of the sun. A transit like this is so rare. No human alive today. We'll witness it again. The next one will not be until 2117 get ready. This event by going to the transit of Venus Planetarium program at the Lawrence Hall of science this Saturday on cow day at 3:00 PM learn why transits are so rare, how studying transits taught us exactly how big our solar system is [00:24:00] and how they may be the key to discovering other earths and other star systems. Then come back on June 5th and observed the actual transit of Venus at the Lawrence Hall of Science. The hall will have several solar telescopes for viewing the eclipse safely on the main plaza. Most of us are aware of the obesity epidemic facing the United States, but is it simply a matter of calories in, calories out on Thursday, May 3rd from 1210 to 1:00 PM in the auditorium of the Berkeley Art Museum, [00:24:30] you CSF neuroendocrinologist Robert Lustig will present the lecture health, Darwin Diet disease and dollars. He will examine some of the more controversial dietary factors contributing to the obesity epidemic, the role that these obesogens potentially play in our evolution toward an unhealthy nation. And possible solutions for turning this trend around. You must register for this event. Go to u h S. Dot. berkeley.edu Speaker 6: [00:25:00] on Saturday April 28th at 1:30 PM the Commonwealth Club and the Youth Science Initiative. Host the research group lead for Pixar and our guest on spectrum two weeks from today, Tony rose. Senator, the admission is $20 Commonwealth Club members get in for 12 Speaker 6: and is $7 for students 18 and under. The talk will be at the Los Altos High School Eagle Theater, two zero one almond avenue in Los Altos. Tony will discuss how math [00:25:30] is central to Pixar film production process and also the young makers program. That's the topic of our interview. In the next episode of spectrum, students are teamed up with adult mentors to design and build ambitious projects for the maker fair for tickets and more information, visit www.commonwealthclub.org another feature is spectrum guest Maggie Court. Baker will also be giving a lecture soon. Maggie is the science editor of Boeing, boeing.net and we'll be discussing her recent book before the lights go [00:26:00] out, conquering the energy crisis before it conquers us. She'll put the fun back in the infrastructure and described the surprising ways our electric system evolved, what we can and can't do about the energy crisis now and what the future might hold. This is the spring seminar for the Berkeley Science Review and will take place in the lead caching building room. Three four five on Wednesday May 2nd at 6:00 PM yeah, RSVP At B e r c. Dot. berkeley.edu [00:26:30] pseudo room, a newly forming East Bay hackerspace is having a free kickoff and fundraiser on Friday May 4th at 7:00 PM at Tech Liminal two six eight 14th street in downtown Oakland. Okay. Pseudo room is a collaborative community of tech developers, citizen scientists, activists and artists. Mitch Altman, cofounder of Noisebridge. We'll discuss hackerspaces for more information, visit s u d o room.org [00:27:00] now the news Speaker 5: significant declines are expected in the number of emperor penguins over the next century due to earlier spring warming around Antarctica. A new study in the April 13th edition of Science Daily reports that an international team of scientists using satellite mapping technology reveals there are twice as many emperor penguins in Antarctica than previously thought. Using a technique known as pan sharpening to increase the resolution of the satellite imagery. They were able to differentiate between birds, [00:27:30] eye shadow and Penguin Guano. In the first comprehensive census of a species taken from space 595,000 birds were counted almost double the previous estimates. Speaker 6: The origin of cosmic grays has long been and remains a mystery. The ice cube collaboration in which Berkeley lab is a crucial contributor published in an article in the April 18th issue of nature on their exhaustive search for a high energy neutrinos that would likely be produced if the violent extra galactic [00:28:00] explosions known as Gamma Ray bursts are a source of ultra high energy cosmic rays. They I know events they have correspondents to these bursts when they would predict to see at least 8.4 events that correspond to some of the 215 gamma ray bursts detected from two periods in 2008 and 2009 there are other popular models for the origin of cosmic rays including active galactic nuclei. The Ice Cube Neutrino telescope encompasses a cubic kilometer of ice under [00:28:30] the South Pole and has over 5,000 digital optical modules that track the direction and energy of speeding yuan's which are created when you Trina is collide with Adam's in the ice. On a later episode of spectrum, you'll hear from Spencer Klein and Thorsten Settle Berger about this experiment. Visit ice cube dot [inaudible] w I s c.edu for more information, Speaker 2: thanks to Rick Kaneski [00:29:00] and Lisa cabbage for help producing show music heard during the show is by Lasagna David from his album, folk and acoustic made available through creative Commons attribution license 3.0 spectrum shows are now available online at iTunes university. Go to itunes.berkeley.edu thank you for listening to spectrum. If you have comments about the show, please send [inaudible] [00:29:30] email address is spectrum dot [inaudible] dot com join us in two weeks. Same time. [inaudible]. Hosted on Acast. See acast.com/privacy for more information.

UC Davis Particle Physics Seminars
Results from IceCube

UC Davis Particle Physics Seminars

Play Episode Listen Later Apr 12, 2012 30:46


Spencer Klein reviews results from the IceCube experiment.