So wait, you have a large bit of material placing pressure on a smaller piece of harder material, ad infinitum...
Won't that just leave you with a series of bisected samples, each harder than the last?
No, due to the cell geometry. The face contacting the softer material is large, and the face contacting the harder material is smaller. As force is constant (not pressure), you end up with less pressure on the weaker face, and more (though hopefully less than your intermediate material's inelastic deformation pressure) on the harder face.
This lets you apply huge pressures to a very small sample, between two diamond faces. My understanding is that they handle the edges by using a metal gasket, which is allowed to deform inelastically to transfer force to a side housing with more surface area (think "o-ring seal").
Diamond anvil cells were big news when they came out because they were so _small_. You could hold them in-hand or put them on a lab bench and apply pressure by turning screws, whereas past high-pressure machines had been huge monstrosities. And with the diamond anvils as windows, you can even to spectrographic measurements of samples as they're being compressed (though the diamond's absorption bands interfere, and the faces can warp under very high pressure). Very nifty gadgets.
If I understand you correctly then you are saying that if there is a particle at some point and the point is moving because of expansion then the particle will not move with the point (or with the same rate)
The particle moves with the point. This causes particles at sufficiently distant locations in the universe to be moving faster than light relative to each other.
This is still consistent with relativity. The usual way of explaining this is to say that the particles are standing still, and space itself is moving. I'm not sure that explanation is really meaningful, but it's handy as an introduction.
The upshot of relativity is that you only get FTL relative motion through movement of space itself (corresponding to certain types of curved spacetime), and that once something is moving FTL relative to you, you aren't going to see it again (it's fallen down a black hole or is past an expanding universe's horizon of observable space). Certain even more exotic geometries of spacetime let you see these things again, but it's an open question whether or not this can actually happen (these are the kinds of geometries you use for time travel).
If space is not deformed - i.e., you have flat space, with no extremely large masses - then you aren't able to cause particles to move FTL relative to you, no matter what reference frame you're observing from.
Actually, I don't see much demand for these "medium speed" controllers. For control applications, they're overkill most of the time, and for multimedia stuff, they're too slow/small.
I've been playing with Atmel's 8-bit line. What makes these chips nice is that they're fast enough to do a lot of things in software that would otherwise require dedicated hardware (PWM, audio input/output/processing), while still leaving enough cycles free to do the high-level control work. Atmel also has a habit of throwing everything including the kitchen sink as peripherals into the controllers, making them very versatile. Yet, you can clock them down and turn off peripherals you don't need in order to get the same kind of power consumption you'd get with a simpler chip, when needed.
From Atmel's point of view, this type of architecture makes sense - instead of 20 similar lines of microcontrollers with different peripherals, they have two or three (for different voltages, mainly).
From a widget designer's point of view, this saves on learning curve and equipment (become familiar with and buy equipment for one or two families of device instead of dozens), and gives them a chip they can use as all-purpose glue with only a modest hit over an application-specific solution.
In summary: Go Atmel:).
[As for 8 vs. 32 bits, the 8 bit family will likely always be lower power for digital functions, due to fewer nodes being switched per operation.]
You're right that the D and Li phases wouldn't last much longer afterward pp stopped, though it depends on the mass loss rate. But there would've been a fair amount of D and Li available.
I'm not convinced of this, for reasons mentioned in my previous post - as pp -> D e+ ve is the rate limiting step, the amount of D present at any given time would be miniscule (it's burned far more quickly than it's produced, so remaining D represents the tail end of the survival time distribution). Any that is left when pp fusion ceases would still be fusing at a rate far, far greater than pp fusion takes place. If the D stuck around for years after pp fusion stopped, I'd be surprised. I'd have to run numbers, but "days" seems more plausible.
I doubt there'd be much D outside the active part of the core, as a) it's only produced in the active part of the core, and b) as the active region shrank as the star lost mass, it would be surrounded by a region in which D could still burn - so the D left in the areas that were once core would be burned up almost immediately.
That's what I meant, though again it depends on the density profile of the object. The core of the object is likely too dense to let anything pass through it. We don't know what the core of Jupiter is made of - this object is something like 50 times denser.
We know what Jupiter's outer atmosphere is made of, and that's all we'd need to know for this thing, as after it hit the "red dwarf" stage, it would lose most of its internal structure. As far as I can see, all of the remaining material should be from the core of the original, larger star, so the atmosphere should represent a reasonable sample of what that original stellar core contained. If I'm overlooking something, by all means let me know.
Black dwarves are white dwarves that have crystallized.
Ok, that clears up the definition a bit. I'd argue that "brown dwarf in a new spectral class category" is the most reasonable classification, then.
I even seem to recall that there was a white dwarf at around 11 LY that we could study.
Sirius B. First white dwarf ever discovered.
I'm pretty sure that there was an _isolated_ white dwarf found somewhere around that distance, as well. Apparently they're common as dirt (much like the yellow dwarf stars they come from). Sirius A probably makes a more interesting neighbour, but it'll give us quite a bit of light pollution if we're trying to watch for sail reflectivity changes for the communications link.
That would let us image stellar atmospheres
We actually already can image stellar atmospheres. Only big stars, but stellar atmospheres nonetheless.
An intereferometric array on that scale would give us detailed images of anything magnetically active within a hundred light years. So, far _sharper_ pictures of the outer layers of stellar atmospheres:).
Arguably we can also get some idea of what's going on by looking at brightness variations and mapping them to resonance modes of the star, but a radio intereferometer with that kind of baseline can resolve features the size of the Moon at 100 light-years, and features the size of Lake Ontario at 10 light-years. We could pick out every flare and prominence and sunspot on each star in range.
What Drexler is postulating isn't a virus sized machine that eventually evolves into a termite and can infultrate a limited number of structures. He's postulating a machine that will be able to reduce almost anything to copies of itself so fast, with no barriers to slow it.
It doesn't have to be a macrostructure like a termite - single-celled organisms are capable of eating things too, just less efficiently (usually).
The barrier to digestion speed is available energy. An organism optimized solely for digesting organics and producing copies of itself could do so surprisingly quickly. Not in an eyeblink, but fast enough to be a significant problem. The thing that prevents this from happening in the natural world (at least more quickly than it currently does) is predation and competition. If you engineer something that's far enough off the beaten path that it has no natural predators and attacks a niche that there are no efficient competitors for, it could conceivably grow unchecked (until we decide to do something about it; the usual array of chemical agents should kill engineered bacteria-like nanomachines as easily as natural bacteria).
If you allow macrostructures of arbitrarily intelligent design (i.e., someone's engineered a plague, and it isn't evolving naturally), you can postulate that your layer of lichen- or mould-like digestive goo would form a tissue structure that allows rapid transfer of energy within itself. That would allow digestion in any given region to proceed far more quickly than it would otherwise, at the expense of digestion elsewhere. Doesn't provide much consolation if the thing being eaten is something important.
Disassembly of inorganic materials is harder for a nanobot based on carbon biology, so I'm doubtful it would happen any time soon. If inorganic nanobots were created, they'd be at a competitive disadvantage, because their chemistry is lower-energy than that of carbon nanobots, so I think "green goo" is a more plausible scenario on earth than "grey goo". Unless you invoke the "complex macrostructures" argument and let the grey goo build power plants, but that's a far more difficult type of goo to design.
And in genetic engineering, yeah, we can make things that are competative, since all the challenging design problems have been solved. But we're no where near the capability of making something so competative it would reduce the surface of the earth to an ooze made up of just the organism.
We're not too far off. How hard would it be to tweak lichen or mould strains to be more aggressive, and secrete toxins that would make it harder for other life to survive in the same area? Nobody (reasonable) is postulating goo-grade nanotech tomorrow. "Within 50 years" is the most optimistic estimate I'd call plausible.
But when someone is out there preaching "If we're not careful a scientist somewhere will accidently turn earth into cybertron" and he's the one being listened to, it's a problem.
At minimum, you can postulate that artificial nanoconstructs can do anything biological constructs can - because biology nicely demonstrates that it's possible to build devices with the performance characteristics of bacteria, and at worst we can tweak nature's existing designs.
Building-eating viruses? We already have dry rot and termites. Goo that reshapes the world in its image? We already have the biosphere, and it did exactly that to a primordial Earth of barren rock, empty oceans, and CO+hydrocarbon atmosphere. I'm skeptical of us building nanoconstructs with performance capabilities much _better_ than biologicals, for general-purpose constructs at least - after all, life has had a long time to opimize its designs. However, handwaving the problem away doesn't work either, because at minimum we can build constructs that are _competitive_. We even have blueprints and working examples laid out for us.
Deuterium and tritium are byproducts of the pp chain - the star constantly regenerates them.
Fair enough. However, these should have been burned very quickly, as when pp fusion stopped, the star would have been *well* within the envelope for rapid DD fusion, and probably a lot of lithium fusion paths as well. This would have occurred while the star was still in pp fusion mode, too; the p + p reaction is the rate limiting step (requires a Weak transformation, which is extremely unlikely compared to Strong interactions). So little, if any, D and Li should have been available at the time pp fusion ceased.
Spectrography might be able to detect the relative ratios of deuterium, etc. in the object, but not likely - its core is still quite dense, and there's little light being generated from there.
It will be able to pick up absorption lines, as this thing does have an atmosphere. The material that's presently the atmosphere should mostly be material from what was the core of the original star, though the core's original fine structure would have vanished when the object became a red dwarf with deep convective mixing.
This object certainly is not a black dwarf - it's not a degenerate object, for one. It's also certainly not a brown dwarf - compositionally, it's totally different.
This depends on which definitions for the objects you accept. If a black dwarf is defined as an object composed mostly of electron-degenerate matter that's below a certain temperature, it's certainly true. Ditto if it's defined as a non-fusing star depleted of hydrogen that's below a certain temperature. If it's defined as a "star that is no longer burning", though, this would qualify. Depends on what we want the category to include (right now, it's "white dwarf husks" by default because that's the only sample in the category that exists).
As for brown dwarfs, while there are classes within them, the definitions that I've heard have ranged from "sub-stellar object large enough to have sustained deuterium fusion" to "any sub-stellar object that condensed directly from a nebula, as opposed to a protoplanetary disk", with no clear dividing line between brown dwarfs and gas giant planets that formed in isolation. I'd argue that the definition is certainly flexible enough to include a stripped stellar core in the category, albeit in its own spectral class.
What amazes me is that it's 300 light years away. 300 ly! That's virtually in our backyard! Jeez, if someone ever develops faster-than-light travel, this would be one of my first stops for astrophysics. In all seriousness, though, it'd be on my top list for observations when the next class of optical telescopes develop.
I'm still waiting for us to launch clusters of radio telescopes for multi-AU baseline radio interferometry. That would let us image stellar atmospheres and planetary magnetospheres, and tell us one heck of a lot about stars and planets in our local neighbourhood.
As for interstellar probes, launching a very small sailcraft for flybys to every star within 10 LY should be do-able. Communications back is the bottleneck, and that can be done by waving the mirror if necessary (extremely slow data transmission, but we'll have years to listen to it). I even seem to recall that there was a white dwarf at around 11 LY that we could study.
I'm not optimistic about faster than light drives being available any time soon.
Though it may have lost its hydrogen and helium burning capeability I would hypothesise that the thing is now an L or T dwarf that is to say it might be Duterium or Lithium burning,
This seems unlikely, as both D and Li burn a lot more readily than p, if I understand correctly. Thus, the star should have used these up very early in its life. If it started life as something larger than a red dwarf, you could argue that there would be deuterium and lithium in its outer layers that wouldn't have mixed with the core material, but a) the outer layers were mostly what was stripped off by the companion star, and b) the star would have passed through a red dwarf stage as it lost mass, resulting in more thorough mixing during that time period.
So, I'm not sure it's a good bet to say that there would still be D or Li left. What do the spectrographs say, for this object?
or its spectral profile might be very dusty or contain methane. In otherwords we might have just seen an L or T dwarf being made but I highly doubt this is a new class of star.
I'm not sure "star" is the correct term any more, as there's no fusion happening (in all likelihood). A few classifications I can think of:
Stellar remnant. Pretty broad category, so probably not specific enough. Also tends to refer to things like planetary nebula and not stars (we haven't seen anything star-like that's been around long enough to cool down past "white dwarf" levels).
Black dwarf. It's a stellar core that can no longer sustain fusion. But this term usually refers to the (as yet unobserved) cooled ashes of a burned-out stellar core (cold white dwarf).
Brown dwarf. It's a sub-stellar mass that's still massive enough that it probably could sustain deuterium fusion, if it had any deuterium to fuse. That probably makes it a brown dwarf on a technicality, even though it's of a bizzare spectral type compared to other brown dwarfs (as you point out).
MACHO. This is another category that's probably too broad to be useful. If it's stripped to below the point where deuterium fusion can occur, but is not a planet (i.e. condensed from a nebula directly as opposed to from another star's protoplanetary disk), it probably counts as a MAssive Compact Halo Object, on a technicality.
I'm voting for "brown dwarf" or "black dwarf", but those are still on technicalities.
Actually, I was thinking of just doing a correlation on a large number of images of a small section of the sky. I sure wasn't thinking about trying coherent detection on the light from a star.
In that case, "baseline" isn't relevant; you just have to keep pointing in the same direction and sum successive images. This is how the deep field images were produced, among other things.
Compare number of people who would pay for a ride on SpaceShipOne vs. number of people who would pay for something more practical - say getting you and two bags to Hawaii in 1.5 hours. Imagine a SSO like design big enough for 20 people and second stage and launched at 45 degrees instead of vertical. Any rocket scientists in here to calculate what a range of something like that might be?
Not much more than a couple times its maximum altitude. Say 300-500 km tops without much more delta-v. To get sub-orbital hops of thousands of kilometres takes very nearly as much delta-v as reaching orbit does (and so is much more difficult than SS1's jaunt).
For air-launched craft, and even gentle glide recovery like SS1 uses, you end up with a couple of hours of "going up" and "going down" time outside of your sub-orbital skip, as well. This considerably eats at your trip time savings (though you could probably do it in 2-3 hours total). The only way around this is to use a conventional launch (requiring a bigger rocket to get around atmosphere problems), and a heat-shield-and-parachute re-entry.
Still an interesting thought experiment, but people have been talking about fast sub-orbital transport for decades, and have yet to make a convincing enough commercial case for it for someone to finance it (heck, even the Concorde is being retired, and it's much cheaper to run).
Suppose that you wanted to detect a very faint object. You could aim your telescope at a given point in the sky for a couple of hours each night. You could integrate the image over a six month period. That should give you a baseline of 186 million miles never mind a paltry couple of thousand miles.
The catch is that you have to know both your position to sub-wavelength precision and the current time to within a fraction of a wavelength period in order to make measurements over that time and distance range.
For 1e9 Hz signals, this means knowing the position of the earth to within a centimetre or two, and knowing the time to about one part in 1e17 (the best atomic clocks I've heard of are 1e14-1e15, and 1e13-1e14 is probably the best you can actually get your hand on). We _might_ know Earth's position that accurately, but I'm not sure (ask an astronomer). Also, the source being studied has to be emitting coherent light at a stable frequency over the same time period for interferometry-after-the-fact like this to work (whereas it just has to be stable for about a twentieth of a second for earth-based radio interferometry).
So, using Earth's orbit as a baseline and integrating over very long time periods doesn't work for most radio sources (it might work for an extremely stable lower-frequency source). Pulsars might be predictable enough, if you apply known models to compensate for spin-down over the observation period. This would let you get a better angular fix on them than you would be able to do by other means.
What I'm waiting for is for a constellation of sun-orbiting radio telescopes to be built with a multi-AU baseline. Sources would have to be stable for hours, but you could get really interesting images of the storms on nearby stars and the magnetospheres of the planets around those stars by this technique. In practice, you'd probably have a cluster of telescopes in each of the orbits, so that you could do interferometry of more rapidly-changing events with a shorter baseline.
But here, we have a distance of several thousand kilometers, so the signals are digitalized and put together at the computer. This is difficult because it's really hard to synchronise the time -atom clocks are not precise enough. Hence the synchronisation is done "so that it fits best", not using any precise clock. (I don't think this is any easier to do, kudos to the scientists at arecibo and VLBI!)
While I agree that best-fit time tweaking to known point-sources is a good calibration technique and should be used, I'd have thought that atmoic clocks are more than precise enough. Typical radio telescope frequencies are on the order of 1e9 Hz, if I understand correctly, while you can easily get atomic clocks accurate to one part in 1e11 (and the record last I'd heard was 1e13 to 1e14). This means that as long as you can re-synchronize them accurately every few seconds to every few hours, you can trust the timestamps.
Synchronizing the clocks is itself a very tricky problem, but you can use this calibration technique to do it, or you can string fiber with known propagation time between all of your telescopes (or a subset of them) and do it the hard way. Similarly you can synchronize via satellite bounce, as long as you know the position of the telescopes and the satellite extremely accurately beforehand (within a few centimetres or better).
In any case, there are two things to focus on. One is the energy density and the other is the speed at which it is
imparted. If you pump the energy in slow enough, then the atoms can dissapate it faster than it comes in. At some point, you toss in the energy fast enough that a portion of it accumulates -- hopefully enough to blow apart the atom (if that's what you're intending to do). I dunno the actual accumulated energy necessary to rip apart a nucleus. I never got that far in my physics courses.
I am aware of most of the issues involving laser pumped ICF; my understanding is that an ultrashort pulse avoids some of the mechanisms that scatter incoming light for longer pulses (on long timescales, your laser pulse converts the pellet to highly-reflective plasma, which loses most of your energy; this part of the motivation behind particle-beam ICF).
My question was about the original poster's statements about inducing fission. I can think of a couple of ways it _might_ happen, but would be interested in learning about the real ones. The statement about pulse power and binding energy was a nonsequiter.
Are the `Trusted Computing' Frequently Asked Questions a good start for you?
I've been reading the TC FAQ, and I still don't understand how this is supposed to do something useful.
It works to prevent tampering by doing security checks against hardware-stored data while in a privileged operating mode, but the whole point of the latest slew of security problems is that unprivileged software can gain access to privileged operating modes. So, this won't do a whole lot to protect you from malware, as was one of its (many) claims.
I'd also expect cracked bios flashes to appear within months of a TC implementation that significantly hindered unlicensed software use. Not to mention cracked versions of the software that didn't handshake with the TC routines. Encryption of software to prevent cracking has been around for years, and has been ineffective for years - you just have to snag unencrypted images of the code and data you're interested in from memory. All of these cracking approaches have countermeasures that can be taken against them, but at this point you're trusting OS and application manufacturers to design software robustly and with keen foresight. I'm skeptical of this occurring in the near future.
There's also the problem of the hardware hashing making the machine non-upgradeable, and the problem of the machine requiring an active 'net connetion for applications to authenticate with their central servers, and the problem of "mod chipping" (removing the TC chip and replacing it with a compromised version).
In summary, I don't think that TC will work for its nominally intended purposes (securing machines against malicious attacks, and ensuring that software and media are used only as licensed). I'm kind of curious as to whether the proponents of TC realize this (and just want to alter licensing schemes for Joe Average), or not (and think it will work).
Must - read - comment - carefully. One character apart, but it's a very important character. He was talking about using the terawatt laser to initiate (spontaneous) fission not fusion. -- I.E. the laser generates enough energy density to rip apart the nucleus of a atom (not smash two of them together)
I'd understood this, but brain faulted and written about fusion as well.
My comment re. photon energy holds, though. I don't really see how photons with energy that low would even _interact_ in a meaningful way with the nucleus, as the energy level spacings are all (or almost all) in the hundreds-of-keV-and-up range. This is why I'd asked about other mechanisms.
At an energy of around 3milliwatts, which using the formula Power = Energy * Time gives a power of about 3TW. This kind of power laser gives focussed energies greater than 10^20 W/cm^2. This actually means that you can use a single laser of this power to initiate fission, because the energy of bound nuclei is less than this!
Can you give me more information on exactly how an ultrashort pulse is better for initiating fusion than a slower pulse? My understanding was that it was plasma temperature and density that mattered, which depends only on the energy deposited and how symmetrical you can get the pellet implosion for an ICF scheme. I'd thought that to act on the nuclei directly you needed individual photons with energy comparable to the binding energy, as opposed to just an instantaneous power at such-and-such a level (much as how you need a photon of a given energy to see the photoelectric effect, with arbitrary quantities of lower-energy photons not doing much).
Is it an electric field gradient effect due to the very high amplitude that does this?
So any chance one of these things being used in some sort of propultion system. Sounds kinda like the the ion drives used on deep space one. (or tie fighters?)
An ion-based propulsion system works best when using different methods, for a couple of reasons. One is that both the laser system driving the plasma accelerator, and the plasma accelerator itself, are pretty inefficient. Other forms of acceleration tend to be more efficient. A wikipedia article summarizing several of these is here. To make a long story short, most of these work by directly applying force to the particles, either using DC electric fields (ion drive), AC electric fields (cyclotron drive component of a VASIMR drive, or some combination of electric and magnetic fields (plasma thruster, hall effect thruster, pulsed inductive thruster).
The second reason is that any kind of relativistic particle beam gives you the same power/thrust ratio a photon drive would. So, you'd be better off just firing the laser aft and using photons as reaction mass (or using a maser or some other more efficient light source, or even just heating up a big sheet of carbon).
In principle, you could use either the ultra-short-pulse lasers or particle beams derived from a plasma accelerator as initiators for inertial confinement fusion, which could be used in a fusion drive. However, particle beams used for fusion initiation tend to be much lower energy (a high-energy beam would just pass through the frozen pellet without depositing most of its energy).
Still nifty devices; just not directly applicable as space drives.
Compact accelerators could be a huge, paradign shifting win for fusion power.
Um, no. Like most particle accelerators, these are horribly inefficient. There are far better ways to dump energy into fusion plasma or frozen pellets than this, even for the particle-beam ICF schemes.
You're using an extremely inefficient chirped-pulse T3 laser to drive an extremely inefficient (but very compact!) particle accelerator.
What I want to see is some of these babies aimed at giant solar sails which provide accelration to a spaceship...
These lasers produce very high power, but for an extremely short time. Typical pulse energy is on the order of tens or hundreds of joules, so your space ship won't be moving very fast.
Ultra-short-pulse lasers are used to investigate chemical reactions, and the exchange of energy between lasers and plasma (useful to understand how to get inertial confinement fusion working properly). They're also handy for creating the kind of plasma disturbance needed for laser wakefield acceleration.
For driving a solar sail, you'd want a very large array of continuous-wave lasers, phase-locked to provide an effective aperture size of hundreds of kilometres (so that you can stay focused on the sail at a distance of light-days to light-years, depending on whether you're going for a flyby or a Forward-style decelleration scheme). If the aperture's any smaller, divergence causes most of your beam to be wasted on empty space.
Individual lasers in this system have to be powerful, but not "petawatt" powerful. You're limited by the amount of light your sail reflector can safely handle, and by the heat sinking your laser requires.
Particle accelerators use electric fields to accelerate charged particles to high speed. Normally you literally have a set of electrodes producing the field. This makes for a bulky device, because you're limited in how close together electrodes can be that have a given voltage difference (due to hardware constraints).
Laser wakefield acceleration is one of a family of acceleration schemes that work by making a disturbance in a plasma, and using the ripples in that to accelerate charged particles. These ripples can be thought of as being similar to sound waves, but because plasma consists of charged particles, you get voltage differences between the peaks and troughs. Laser wakefield acceleration (and beat-wave acceleration and particle wakefield acceleration and so forth) use these voltage differences to accelerate particles (the usual analogy is to say that the accelerated particles are "surfing" on the slope of a moving wave, picking up speed the whole time).
The advantage of plasma accelerator schemes is that the voltage gradients are much steeper than in conventional accelerators (large voltage change in a very short distance). This means that as long as you can keep them behaving nicely, you can use a plasma accelerator that fits on a tabletop to produce particle energies you'd otherwise need a huge linear accelerator to generate.
Unlike conventional accelerators, there's no easy way to chain plasma accelerators together to get arbitrary energies. This is being worked on. They're also working on using better lasers to create larger plasma disturbances and get single-stage accelerators to work better; that's the focus of this article.
Now, the Star Trek terms:
monochromatic beam
This means that all of the accelerated particles wind up at more or less the same energy, instead of being at different energies. The analogy is with monochromatic light (all photons at the same energy, and hence colour).
relativistic electron
Electron accelerated to energies much higher than its rest mass. For an electron, this means they're above the 1 MeV range. For a proton, it would be above the GeV range. For relativistic heavy ions (e.g. the ones in the RHIC device), it's the TeV range.
At ultrarelativistic speeds, particles are travelling almost exactly at the speed of light, which makes accelerator design a bit easier.
plasma channel
The area in the plasma where the laser has passed, and conditions are right for acceleration. This is a cylindrical channel, usually.
accelerating gradient
How much voltage changes with distance. This determines the acceleration felt by the particles you're driving, which tells you how big a device you need to reach a given energy or what particle energy you can expect to get out of a given device.
GeV per meter
Units in which acceleration gradient is measured.
compact staged particle accelerator
A particle accelerator that's small (this is the advantage of plasma accelerators), and that use multiple stages to reach higher energies than any single acceleration stage could. This is tricky to do with plasma accelerators, but not impossible. Very handy if you can get it working reliably for hundreds of stages.
next generation petawatt power
These are the ultra-short-pulse, high-energy lasers you may have been hearing about. Right now, you can get off the shelf systems that dump a few joules of energy into a pulse less than a picosecond long. Power during the pulse is in the terawatt range (which is why these are called "T3 / Table-Top Terawatt" lasers). Having a short, sharp pulse instead of a long, drawn-out one makes laser wakefield acceleration work better. The next generation of ultra-short-pulse laser delivers higher power in an even shorter time. The goal is to get petawatt power du
Re:Unknown Error In The Submission
on
Nuclear Batteries
·
· Score: 1
Thanks. By weaponized, I meant as the "dirty" component of a dirty bomb, since it's plain they aren't the right material for nuclear fission.
I'm kind of puzzled by the popularity of "dirty bomb" threats with the media of late, as it's very difficult to actually make that big a mess with them (due to scarcity of materials, unless you're carting out a wheelbarrel-load of spent fuel rods, in which case you're dead before you can do anything). There are quite a number of very interesting agents that could be tipped into a city's water supply that would do far worse damage before they were detected.
Hopefully, the people in charge of security realize this too, and are just keeping quiet.
But much of what I've read since I posted this leads me to agree with you that the pollution problems are the most worrisome.
It's kind of ironic, actually - I'm arguing both sides of the waste problem at once in this section:). I'm trying to convince some people that it exists, while trying to convince one very zealous individual that it really isn't that much worse than the chemical waste problem.
It'll be interesting to see what actually ends up getting done with these power sources. The article raised a number of interesting possibilities.
Slashdot Mirror
So wait, you have a large bit of material placing pressure on a smaller piece of harder material, ad infinitum...
Won't that just leave you with a series of bisected samples, each harder than the last?
No, due to the cell geometry. The face contacting the softer material is large, and the face contacting the harder material is smaller. As force is constant (not pressure), you end up with less pressure on the weaker face, and more (though hopefully less than your intermediate material's inelastic deformation pressure) on the harder face.
This lets you apply huge pressures to a very small sample, between two diamond faces. My understanding is that they handle the edges by using a metal gasket, which is allowed to deform inelastically to transfer force to a side housing with more surface area (think "o-ring seal").
Diamond anvil cells were big news when they came out because they were so _small_. You could hold them in-hand or put them on a lab bench and apply pressure by turning screws, whereas past high-pressure machines had been huge monstrosities. And with the diamond anvils as windows, you can even to spectrographic measurements of samples as they're being compressed (though the diamond's absorption bands interfere, and the faces can warp under very high pressure). Very nifty gadgets.
If I understand you correctly then you are saying that if there is a particle at some point and the point is moving because of expansion then the particle will not move with the point (or with the same rate)
The particle moves with the point. This causes particles at sufficiently distant locations in the universe to be moving faster than light relative to each other.
This is still consistent with relativity. The usual way of explaining this is to say that the particles are standing still, and space itself is moving. I'm not sure that explanation is really meaningful, but it's handy as an introduction.
The upshot of relativity is that you only get FTL relative motion through movement of space itself (corresponding to certain types of curved spacetime), and that once something is moving FTL relative to you, you aren't going to see it again (it's fallen down a black hole or is past an expanding universe's horizon of observable space). Certain even more exotic geometries of spacetime let you see these things again, but it's an open question whether or not this can actually happen (these are the kinds of geometries you use for time travel).
If space is not deformed - i.e., you have flat space, with no extremely large masses - then you aren't able to cause particles to move FTL relative to you, no matter what reference frame you're observing from.
I hope this clears things up a bit.
Actually, I don't see much demand for these "medium speed" controllers. For control applications, they're overkill most of the time, and for multimedia stuff, they're too slow/small.
:).
I've been playing with Atmel's 8-bit line. What makes these chips nice is that they're fast enough to do a lot of things in software that would otherwise require dedicated hardware (PWM, audio input/output/processing), while still leaving enough cycles free to do the high-level control work. Atmel also has a habit of throwing everything including the kitchen sink as peripherals into the controllers, making them very versatile. Yet, you can clock them down and turn off peripherals you don't need in order to get the same kind of power consumption you'd get with a simpler chip, when needed.
From Atmel's point of view, this type of architecture makes sense - instead of 20 similar lines of microcontrollers with different peripherals, they have two or three (for different voltages, mainly).
From a widget designer's point of view, this saves on learning curve and equipment (become familiar with and buy equipment for one or two families of device instead of dozens), and gives them a chip they can use as all-purpose glue with only a modest hit over an application-specific solution.
In summary: Go Atmel
[As for 8 vs. 32 bits, the 8 bit family will likely always be lower power for digital functions, due to fewer nodes being switched per operation.]
You're right that the D and Li phases wouldn't last much longer afterward pp stopped, though it depends on the mass loss rate. But there would've been a fair amount of D and Li available.
:).
I'm not convinced of this, for reasons mentioned in my previous post - as pp -> D e+ ve is the rate limiting step, the amount of D present at any given time would be miniscule (it's burned far more quickly than it's produced, so remaining D represents the tail end of the survival time distribution). Any that is left when pp fusion ceases would still be fusing at a rate far, far greater than pp fusion takes place. If the D stuck around for years after pp fusion stopped, I'd be surprised. I'd have to run numbers, but "days" seems more plausible.
I doubt there'd be much D outside the active part of the core, as a) it's only produced in the active part of the core, and b) as the active region shrank as the star lost mass, it would be surrounded by a region in which D could still burn - so the D left in the areas that were once core would be burned up almost immediately.
That's what I meant, though again it depends on the density profile of the object. The core of the object is likely too dense to let anything pass through it. We don't know what the core of Jupiter is made of - this object is something like 50 times denser.
We know what Jupiter's outer atmosphere is made of, and that's all we'd need to know for this thing, as after it hit the "red dwarf" stage, it would lose most of its internal structure. As far as I can see, all of the remaining material should be from the core of the original, larger star, so the atmosphere should represent a reasonable sample of what that original stellar core contained. If I'm overlooking something, by all means let me know.
Black dwarves are white dwarves that have crystallized.
Ok, that clears up the definition a bit. I'd argue that "brown dwarf in a new spectral class category" is the most reasonable classification, then.
I even seem to recall that there was a white dwarf at around 11 LY that we could study.
Sirius B. First white dwarf ever discovered.
I'm pretty sure that there was an _isolated_ white dwarf found somewhere around that distance, as well. Apparently they're common as dirt (much like the yellow dwarf stars they come from). Sirius A probably makes a more interesting neighbour, but it'll give us quite a bit of light pollution if we're trying to watch for sail reflectivity changes for the communications link.
That would let us image stellar atmospheres
We actually already can image stellar atmospheres. Only big stars, but stellar atmospheres nonetheless.
An intereferometric array on that scale would give us detailed images of anything magnetically active within a hundred light years. So, far _sharper_ pictures of the outer layers of stellar atmospheres
Arguably we can also get some idea of what's going on by looking at brightness variations and mapping them to resonance modes of the star, but a radio intereferometer with that kind of baseline can resolve features the size of the Moon at 100 light-years, and features the size of Lake Ontario at 10 light-years. We could pick out every flare and prominence and sunspot on each star in range.
What Drexler is postulating isn't a virus sized machine that eventually evolves into a termite and can infultrate a limited number of structures. He's postulating a machine that will be able to reduce almost anything to copies of itself so fast, with no barriers to slow it.
It doesn't have to be a macrostructure like a termite - single-celled organisms are capable of eating things too, just less efficiently (usually).
The barrier to digestion speed is available energy. An organism optimized solely for digesting organics and producing copies of itself could do so surprisingly quickly. Not in an eyeblink, but fast enough to be a significant problem. The thing that prevents this from happening in the natural world (at least more quickly than it currently does) is predation and competition. If you engineer something that's far enough off the beaten path that it has no natural predators and attacks a niche that there are no efficient competitors for, it could conceivably grow unchecked (until we decide to do something about it; the usual array of chemical agents should kill engineered bacteria-like nanomachines as easily as natural bacteria).
If you allow macrostructures of arbitrarily intelligent design (i.e., someone's engineered a plague, and it isn't evolving naturally), you can postulate that your layer of lichen- or mould-like digestive goo would form a tissue structure that allows rapid transfer of energy within itself. That would allow digestion in any given region to proceed far more quickly than it would otherwise, at the expense of digestion elsewhere. Doesn't provide much consolation if the thing being eaten is something important.
Disassembly of inorganic materials is harder for a nanobot based on carbon biology, so I'm doubtful it would happen any time soon. If inorganic nanobots were created, they'd be at a competitive disadvantage, because their chemistry is lower-energy than that of carbon nanobots, so I think "green goo" is a more plausible scenario on earth than "grey goo". Unless you invoke the "complex macrostructures" argument and let the grey goo build power plants, but that's a far more difficult type of goo to design.
And in genetic engineering, yeah, we can make things that are competative, since all the challenging design problems have been solved. But we're no where near the capability of making something so competative it would reduce the surface of the earth to an ooze made up of just the organism.
We're not too far off. How hard would it be to tweak lichen or mould strains to be more aggressive, and secrete toxins that would make it harder for other life to survive in the same area? Nobody (reasonable) is postulating goo-grade nanotech tomorrow. "Within 50 years" is the most optimistic estimate I'd call plausible.
CO+hydrocarbon atmosphere
CO_2_ plus hydrocarbon atmosphere, I know...
But when someone is out there preaching "If we're not careful a scientist somewhere will accidently turn earth into cybertron" and he's the one being listened to, it's a problem.
At minimum, you can postulate that artificial nanoconstructs can do anything biological constructs can - because biology nicely demonstrates that it's possible to build devices with the performance characteristics of bacteria, and at worst we can tweak nature's existing designs.
Building-eating viruses? We already have dry rot and termites. Goo that reshapes the world in its image? We already have the biosphere, and it did exactly that to a primordial Earth of barren rock, empty oceans, and CO+hydrocarbon atmosphere. I'm skeptical of us building nanoconstructs with performance capabilities much _better_ than biologicals, for general-purpose constructs at least - after all, life has had a long time to opimize its designs. However, handwaving the problem away doesn't work either, because at minimum we can build constructs that are _competitive_. We even have blueprints and working examples laid out for us.
If it was possible to make a Nanobot that could turn everything into grey goo, wouldnt everything already be grey goo?
It's been pointed out that it already is.
It's called the biosphere.
Deuterium and tritium are byproducts of the pp chain - the star constantly regenerates them.
Fair enough. However, these should have been burned very quickly, as when pp fusion stopped, the star would have been *well* within the envelope for rapid DD fusion, and probably a lot of lithium fusion paths as well. This would have occurred while the star was still in pp fusion mode, too; the p + p reaction is the rate limiting step (requires a Weak transformation, which is extremely unlikely compared to Strong interactions). So little, if any, D and Li should have been available at the time pp fusion ceased.
Spectrography might be able to detect the relative ratios of deuterium, etc. in the object, but not likely - its core is still quite dense, and there's little light being generated from there.
It will be able to pick up absorption lines, as this thing does have an atmosphere. The material that's presently the atmosphere should mostly be material from what was the core of the original star, though the core's original fine structure would have vanished when the object became a red dwarf with deep convective mixing.
This object certainly is not a black dwarf - it's not a degenerate object, for one. It's also certainly not a brown dwarf - compositionally, it's totally different.
This depends on which definitions for the objects you accept. If a black dwarf is defined as an object composed mostly of electron-degenerate matter that's below a certain temperature, it's certainly true. Ditto if it's defined as a non-fusing star depleted of hydrogen that's below a certain temperature. If it's defined as a "star that is no longer burning", though, this would qualify. Depends on what we want the category to include (right now, it's "white dwarf husks" by default because that's the only sample in the category that exists).
As for brown dwarfs, while there are classes within them, the definitions that I've heard have ranged from "sub-stellar object large enough to have sustained deuterium fusion" to "any sub-stellar object that condensed directly from a nebula, as opposed to a protoplanetary disk", with no clear dividing line between brown dwarfs and gas giant planets that formed in isolation. I'd argue that the definition is certainly flexible enough to include a stripped stellar core in the category, albeit in its own spectral class.
What amazes me is that it's 300 light years away. 300 ly! That's virtually in our backyard! Jeez, if someone ever develops faster-than-light travel, this would be one of my first stops for astrophysics. In all seriousness, though, it'd be on my top list for observations when the next class of optical telescopes develop.
I'm still waiting for us to launch clusters of radio telescopes for multi-AU baseline radio interferometry. That would let us image stellar atmospheres and planetary magnetospheres, and tell us one heck of a lot about stars and planets in our local neighbourhood.
As for interstellar probes, launching a very small sailcraft for flybys to every star within 10 LY should be do-able. Communications back is the bottleneck, and that can be done by waving the mirror if necessary (extremely slow data transmission, but we'll have years to listen to it). I even seem to recall that there was a white dwarf at around 11 LY that we could study.
I'm not optimistic about faster than light drives being available any time soon.
That meters and yards were the exact same thing.
It's close enough that it makes a wonderful way of explaining to someone who doesn't know what a metre is, how big this thing is.
Where's the problem?
This seems unlikely, as both D and Li burn a lot more readily than p, if I understand correctly. Thus, the star should have used these up very early in its life. If it started life as something larger than a red dwarf, you could argue that there would be deuterium and lithium in its outer layers that wouldn't have mixed with the core material, but a) the outer layers were mostly what was stripped off by the companion star, and b) the star would have passed through a red dwarf stage as it lost mass, resulting in more thorough mixing during that time period.
So, I'm not sure it's a good bet to say that there would still be D or Li left. What do the spectrographs say, for this object?
or its spectral profile might be very dusty or contain methane. In otherwords we might have just seen an L or T dwarf being made but I highly doubt this is a new class of star.
I'm not sure "star" is the correct term any more, as there's no fusion happening (in all likelihood). A few classifications I can think of:
Pretty broad category, so probably not specific enough. Also tends to refer to things like planetary nebula and not stars (we haven't seen anything star-like that's been around long enough to cool down past "white dwarf" levels).
It's a stellar core that can no longer sustain fusion. But this term usually refers to the (as yet unobserved) cooled ashes of a burned-out stellar core (cold white dwarf).
It's a sub-stellar mass that's still massive enough that it probably could sustain deuterium fusion, if it had any deuterium to fuse. That probably makes it a brown dwarf on a technicality, even though it's of a bizzare spectral type compared to other brown dwarfs (as you point out).
This is another category that's probably too broad to be useful. If it's stripped to below the point where deuterium fusion can occur, but is not a planet (i.e. condensed from a nebula directly as opposed to from another star's protoplanetary disk), it probably counts as a MAssive Compact Halo Object, on a technicality.
I'm voting for "brown dwarf" or "black dwarf", but those are still on technicalities.
Actually, I was thinking of just doing a correlation on a large number of images of a small section of the sky. I sure wasn't thinking about trying coherent detection on the light from a star.
In that case, "baseline" isn't relevant; you just have to keep pointing in the same direction and sum successive images. This is how the deep field images were produced, among other things.
Compare number of people who would pay for a ride on SpaceShipOne vs. number of people who would pay for something more practical - say getting you and two bags to Hawaii in 1.5 hours.
Imagine a SSO like design big enough for 20 people and second stage and launched at 45 degrees instead of vertical. Any rocket scientists in here to calculate what a range of something like that might be?
Not much more than a couple times its maximum altitude. Say 300-500 km tops without much more delta-v. To get sub-orbital hops of thousands of kilometres takes very nearly as much delta-v as reaching orbit does (and so is much more difficult than SS1's jaunt).
For air-launched craft, and even gentle glide recovery like SS1 uses, you end up with a couple of hours of "going up" and "going down" time outside of your sub-orbital skip, as well. This considerably eats at your trip time savings (though you could probably do it in 2-3 hours total). The only way around this is to use a conventional launch (requiring a bigger rocket to get around atmosphere problems), and a heat-shield-and-parachute re-entry.
Still an interesting thought experiment, but people have been talking about fast sub-orbital transport for decades, and have yet to make a convincing enough commercial case for it for someone to finance it (heck, even the Concorde is being retired, and it's much cheaper to run).
Suppose that you wanted to detect a very faint object. You could aim your telescope at a given point in the sky for a couple of hours each night. You could integrate the image over a six month period. That should give you a baseline of 186 million miles never mind a paltry couple of thousand miles.
The catch is that you have to know both your position to sub-wavelength precision and the current time to within a fraction of a wavelength period in order to make measurements over that time and distance range.
For 1e9 Hz signals, this means knowing the position of the earth to within a centimetre or two, and knowing the time to about one part in 1e17 (the best atomic clocks I've heard of are 1e14-1e15, and 1e13-1e14 is probably the best you can actually get your hand on). We _might_ know Earth's position that accurately, but I'm not sure (ask an astronomer). Also, the source being studied has to be emitting coherent light at a stable frequency over the same time period for interferometry-after-the-fact like this to work (whereas it just has to be stable for about a twentieth of a second for earth-based radio interferometry).
So, using Earth's orbit as a baseline and integrating over very long time periods doesn't work for most radio sources (it might work for an extremely stable lower-frequency source). Pulsars might be predictable enough, if you apply known models to compensate for spin-down over the observation period. This would let you get a better angular fix on them than you would be able to do by other means.
What I'm waiting for is for a constellation of sun-orbiting radio telescopes to be built with a multi-AU baseline. Sources would have to be stable for hours, but you could get really interesting images of the storms on nearby stars and the magnetospheres of the planets around those stars by this technique. In practice, you'd probably have a cluster of telescopes in each of the orbits, so that you could do interferometry of more rapidly-changing events with a shorter baseline.
But here, we have a distance of several thousand kilometers, so the signals are digitalized and put together at the computer. This is difficult because it's really hard to synchronise the time -atom clocks are not precise enough. Hence the synchronisation is done "so that it fits best", not using any precise clock. (I don't think this is any easier to do, kudos to the scientists at arecibo and VLBI!)
While I agree that best-fit time tweaking to known point-sources is a good calibration technique and should be used, I'd have thought that atmoic clocks are more than precise enough. Typical radio telescope frequencies are on the order of 1e9 Hz, if I understand correctly, while you can easily get atomic clocks accurate to one part in 1e11 (and the record last I'd heard was 1e13 to 1e14). This means that as long as you can re-synchronize them accurately every few seconds to every few hours, you can trust the timestamps.
Synchronizing the clocks is itself a very tricky problem, but you can use this calibration technique to do it, or you can string fiber with known propagation time between all of your telescopes (or a subset of them) and do it the hard way. Similarly you can synchronize via satellite bounce, as long as you know the position of the telescopes and the satellite extremely accurately beforehand (within a few centimetres or better).
In any case, there are two things to focus on. One is the energy density and the other is the speed at which it is
imparted. If you pump the energy in slow enough, then the atoms can dissapate it faster than it comes in. At some point, you toss in the energy fast enough that a portion of it accumulates -- hopefully enough to blow apart the atom (if that's what you're intending to do). I dunno the actual accumulated energy necessary to rip apart a nucleus. I never got that far in my physics courses.
I am aware of most of the issues involving laser pumped ICF; my understanding is that an ultrashort pulse avoids some of the mechanisms that scatter incoming light for longer pulses (on long timescales, your laser pulse converts the pellet to highly-reflective plasma, which loses most of your energy; this part of the motivation behind particle-beam ICF).
My question was about the original poster's statements about inducing fission. I can think of a couple of ways it _might_ happen, but would be interested in learning about the real ones. The statement about pulse power and binding energy was a nonsequiter.
Are the `Trusted Computing' Frequently Asked Questions a good start for you?
I've been reading the TC FAQ, and I still don't understand how this is supposed to do something useful.
It works to prevent tampering by doing security checks against hardware-stored data while in a privileged operating mode, but the whole point of the latest slew of security problems is that unprivileged software can gain access to privileged operating modes. So, this won't do a whole lot to protect you from malware, as was one of its (many) claims.
I'd also expect cracked bios flashes to appear within months of a TC implementation that significantly hindered unlicensed software use. Not to mention cracked versions of the software that didn't handshake with the TC routines. Encryption of software to prevent cracking has been around for years, and has been ineffective for years - you just have to snag unencrypted images of the code and data you're interested in from memory. All of these cracking approaches have countermeasures that can be taken against them, but at this point you're trusting OS and application manufacturers to design software robustly and with keen foresight. I'm skeptical of this occurring in the near future.
There's also the problem of the hardware hashing making the machine non-upgradeable, and the problem of the machine requiring an active 'net connetion for applications to authenticate with their central servers, and the problem of "mod chipping" (removing the TC chip and replacing it with a compromised version).
In summary, I don't think that TC will work for its nominally intended purposes (securing machines against malicious attacks, and ensuring that software and media are used only as licensed). I'm kind of curious as to whether the proponents of TC realize this (and just want to alter licensing schemes for Joe Average), or not (and think it will work).
Must - read - comment - carefully. One character apart, but it's a very important character. He was talking about using the terawatt laser to initiate (spontaneous) fission not fusion. -- I.E. the laser generates enough energy density to rip apart the nucleus of a atom (not smash two of them together)
I'd understood this, but brain faulted and written about fusion as well.
My comment re. photon energy holds, though. I don't really see how photons with energy that low would even _interact_ in a meaningful way with the nucleus, as the energy level spacings are all (or almost all) in the hundreds-of-keV-and-up range. This is why I'd asked about other mechanisms.
At an energy of around 3milliwatts, which using the formula Power = Energy * Time gives a power of about 3TW. This kind of power laser gives focussed energies greater than 10^20 W/cm^2. This actually means that you can use a single laser of this power to initiate fission, because the energy of bound nuclei is less than this!
Can you give me more information on exactly how an ultrashort pulse is better for initiating fusion than a slower pulse? My understanding was that it was plasma temperature and density that mattered, which depends only on the energy deposited and how symmetrical you can get the pellet implosion for an ICF scheme. I'd thought that to act on the nuclei directly you needed individual photons with energy comparable to the binding energy, as opposed to just an instantaneous power at such-and-such a level (much as how you need a photon of a given energy to see the photoelectric effect, with arbitrary quantities of lower-energy photons not doing much).
Is it an electric field gradient effect due to the very high amplitude that does this?
So any chance one of these things being used in some sort of propultion system. Sounds kinda like the the ion drives used on deep space one. (or tie fighters?)
An ion-based propulsion system works best when using different methods, for a couple of reasons. One is that both the laser system driving the plasma accelerator, and the plasma accelerator itself, are pretty inefficient. Other forms of acceleration tend to be more efficient. A wikipedia article summarizing several of these is here. To make a long story short, most of these work by directly applying force to the particles, either using DC electric fields (ion drive), AC electric fields (cyclotron drive component of a VASIMR drive, or some combination of electric and magnetic fields (plasma thruster, hall effect thruster, pulsed inductive thruster).
The second reason is that any kind of relativistic particle beam gives you the same power/thrust ratio a photon drive would. So, you'd be better off just firing the laser aft and using photons as reaction mass (or using a maser or some other more efficient light source, or even just heating up a big sheet of carbon).
In principle, you could use either the ultra-short-pulse lasers or particle beams derived from a plasma accelerator as initiators for inertial confinement fusion, which could be used in a fusion drive. However, particle beams used for fusion initiation tend to be much lower energy (a high-energy beam would just pass through the frozen pellet without depositing most of its energy).
Still nifty devices; just not directly applicable as space drives.
Compact accelerators could be a huge, paradign shifting win for fusion power.
Um, no. Like most particle accelerators, these are horribly inefficient. There are far better ways to dump energy into fusion plasma or frozen pellets than this, even for the particle-beam ICF schemes.
You're using an extremely inefficient chirped-pulse T3 laser to drive an extremely inefficient (but very compact!) particle accelerator.
(Yes, I know IHBT again...)
Science an be good sometimes, but don't we have enough already?
So we shouldn't be working on accelerators that let us perform anti-cancer therapy more easily?
(I know, IHBT; this is just an exceptionally silly troll.)
What I want to see is some of these babies aimed at giant solar sails which provide accelration to a spaceship ...
These lasers produce very high power, but for an extremely short time. Typical pulse energy is on the order of tens or hundreds of joules, so your space ship won't be moving very fast.
Ultra-short-pulse lasers are used to investigate chemical reactions, and the exchange of energy between lasers and plasma (useful to understand how to get inertial confinement fusion working properly). They're also handy for creating the kind of plasma disturbance needed for laser wakefield acceleration.
For driving a solar sail, you'd want a very large array of continuous-wave lasers, phase-locked to provide an effective aperture size of hundreds of kilometres (so that you can stay focused on the sail at a distance of light-days to light-years, depending on whether you're going for a flyby or a Forward-style decelleration scheme). If the aperture's any smaller, divergence causes most of your beam to be wasted on empty space.
Individual lasers in this system have to be powerful, but not "petawatt" powerful. You're limited by the amount of light your sail reflector can safely handle, and by the heat sinking your laser requires.
Particle accelerators use electric fields to accelerate charged particles to high speed. Normally you literally have a set of electrodes producing the field. This makes for a bulky device, because you're limited in how close together electrodes can be that have a given voltage difference (due to hardware constraints).
Laser wakefield acceleration is one of a family of acceleration schemes that work by making a disturbance in a plasma, and using the ripples in that to accelerate charged particles. These ripples can be thought of as being similar to sound waves, but because plasma consists of charged particles, you get voltage differences between the peaks and troughs. Laser wakefield acceleration (and beat-wave acceleration and particle wakefield acceleration and so forth) use these voltage differences to accelerate particles (the usual analogy is to say that the accelerated particles are "surfing" on the slope of a moving wave, picking up speed the whole time).
The advantage of plasma accelerator schemes is that the voltage gradients are much steeper than in conventional accelerators (large voltage change in a very short distance). This means that as long as you can keep them behaving nicely, you can use a plasma accelerator that fits on a tabletop to produce particle energies you'd otherwise need a huge linear accelerator to generate.
Unlike conventional accelerators, there's no easy way to chain plasma accelerators together to get arbitrary energies. This is being worked on. They're also working on using better lasers to create larger plasma disturbances and get single-stage accelerators to work better; that's the focus of this article.
Now, the Star Trek terms:
This means that all of the accelerated particles wind up at more or less the same energy, instead of being at different energies. The analogy is with monochromatic light (all photons at the same energy, and hence colour).
Electron accelerated to energies much higher than its rest mass. For an electron, this means they're above the 1 MeV range. For a proton, it would be above the GeV range. For relativistic heavy ions (e.g. the ones in the RHIC device), it's the TeV range.
At ultrarelativistic speeds, particles are travelling almost exactly at the speed of light, which makes accelerator design a bit easier.
The area in the plasma where the laser has passed, and conditions are right for acceleration. This is a cylindrical channel, usually.
How much voltage changes with distance. This determines the acceleration felt by the particles you're driving, which tells you how big a device you need to reach a given energy or what particle energy you can expect to get out of a given device.
Units in which acceleration gradient is measured.
A particle accelerator that's small (this is the advantage of plasma accelerators), and that use multiple stages to reach higher energies than any single acceleration stage could. This is tricky to do with plasma accelerators, but not impossible. Very handy if you can get it working reliably for hundreds of stages.
These are the ultra-short-pulse, high-energy lasers you may have been hearing about. Right now, you can get off the shelf systems that dump a few joules of energy into a pulse less than a picosecond long. Power during the pulse is in the terawatt range (which is why these are called "T3 / Table-Top Terawatt" lasers). Having a short, sharp pulse instead of a long, drawn-out one makes laser wakefield acceleration work better. The next generation of ultra-short-pulse laser delivers higher power in an even shorter time. The goal is to get petawatt power du
Thanks. By weaponized, I meant as the "dirty" component of a dirty bomb, since it's plain they aren't the right material for nuclear fission.
:). I'm trying to convince some people that it exists, while trying to convince one very zealous individual that it really isn't that much worse than the chemical waste problem.
I'm kind of puzzled by the popularity of "dirty bomb" threats with the media of late, as it's very difficult to actually make that big a mess with them (due to scarcity of materials, unless you're carting out a wheelbarrel-load of spent fuel rods, in which case you're dead before you can do anything). There are quite a number of very interesting agents that could be tipped into a city's water supply that would do far worse damage before they were detected.
Hopefully, the people in charge of security realize this too, and are just keeping quiet.
But much of what I've read since I posted this leads me to agree with you that the pollution problems are the most worrisome.
It's kind of ironic, actually - I'm arguing both sides of the waste problem at once in this section
It'll be interesting to see what actually ends up getting done with these power sources. The article raised a number of interesting possibilities.