MIT at center of political power play over campus fusion reactor
bostonglobe.com
bostonglobe.com
Absolutely this is a key factor in why scientific grants have terrible yield vs private funding.
As a former fusion scientist, I saw this coming, and left the field to make my own fortune and direct it towards what I believed the best energy research was.
Name me one field of human endeavor without politics.
Tons of stuff. I'd rather politics axe "workings of the universe" than vaccinations and disease control.
The ISS is the most expensive structure ever built, after all.
Even vaccinations are used for other purposes than to keep people healthy. Clever, but likely to have a lot of adverse side-effects.
I think you're thinking that I'm saying something like "well, if we've never made it to the moon, it sounds preposterous to think we ever will." It sounds like you think developing a politics-less way of funding scientific research is a matter of working on something to produce a new system. Please let me know if that is in fact your perspective.
It is not mine. To me, the idea of not having politics be involved in science sounds as absurd as... well lets assume we're friends and imagine that you, I, and a mutual friend of ours planned to play lazor tag together. We go to pick her up and she says that she can't join us because she has food poisoning and can't come. As we are driving from her house to the lazor tag place, you say that biology shouldn't interfere with plans with your friends. Now, you just mean that it sucks that she has food poisoning and you wish that this sort of thing didn't happen occasionally. But your statement is slightly more absolute than that: biology, being fundamental to the human condition, is always involved in plans made among humans. Complex metabolic, biomechanical, and neurological processes are an inherent part of a lazor tag outing. I claim that politics is fundamental to the human condition.
I claim that much like food poisoning happens occasionally, so does pettiness and favoritism. That does not mean that it is pointless to keep one's meats either freezing or cooked. That does not make it pointless to use separate cutting boards for meat and vegetables. Healthier biology is possible to achieve and so is healthier politics. But it is still politics.
My saying "politics is always involved" may now seem pedantic, so here is why it is meaningful and important: If you adopt an attitude that attempts to simply avoid politics, you will likely avoid studying and paying attention to politics and group dynamics. On a pragmatic level, this will serve you poorly. But on a more idealistic level, this will serve your cause poorly because you will abandon the opportunity to create healthy politics. What does healthy politics look like? That is a subject for whole essays and whole lifetimes of study, but I suspect you can recognize a social system with healthy politics by its high degree of trust.
So, if by politics you mean "the act of people mutually deciding what to do with limited resources," then yes, you cannot eliminate politics from scientific funding. But that feels like a tautology to me.
If we use the colloquial definition of politics, namely "nepotism, influence peddling, backbiting over unrelated issues, earmarking, pork barreling, gerrymandering, etc." then I do believe we can and should strive to eliminate politics from scientific funding decisions.
Thanks for your detailed reply; I hope this clarifies each of our intended meanings.
But that isn't what the MIT fusion reactor is doing. We already know how fusion works; we already know what conditions we need to create to make a fusion reactor self-sustaining. What we don't yet know is how to create them here on Earth.
So the question isn't about fundamental knowledge; it's about engineering and economics. We've been trying to make tokamak fusion work for decades. The Alcator C-Mod experiment that's mentioned in these articles was put into place while I was at MIT in the mid-1980's, and people back then were saying we would have ignition, a self-sustaining controlled fusion reaction, sometime within the next decade. It's now been three decades and we still don't have ignition. Certainly the experiment has been upgraded and refined and has continued to yield useful data, but that's not the crucial question; the crucial question is, is it worth the taxpayer money that's been put into it? That money wasn't supposed to buy useful data, it was supposed to buy ignition.
To be clear, I'm not arguing that the funds for MIT's experiment should be redirected to ITER or to the other US sites mentioned in the article. The same reasoning that I just applied to the MIT experiment, could be applied to all of the fusion research that's been done since we started doing fusion research. We simply did not realize the magnitude of the engineering problem involved in achieving a controlled, self-sustaining fusion reaction without a Sun's mass worth of gravity to confine it.
Frankly, if ITER has the scaling laws of Tokamaks right, there won't ever be a working Tokamak smaller than a 10 story building, or producing less than about 1.5 GW. And q values can go up to maybe 10, not counting generator losses. To be honest : we don't want that.
Getting this refunded basically means that more money is going to various other fusion ideas, and that's much better than having all our eggs in one hugely expensive, massively unwieldy and "just 10 more years" project (just 10 more years for about 40 years now).
Frankly, ITER is one of those huge failed projects that just won't admit failure for mostly political reasons. We all know about lots of them. Either it will fail directly because physics somehow prevents Tokamaks from being cost effective, or it will fail because it won't succeed by the time we need it to succeed. So far, the net result of the project is that a 6 story Tokamak can't work, due to plasma instabilities. So they're building a 10 story one.
The project itself has stupendous accomplishments. It has demonstrated almost unprecedented international scientific collaboration. The amount of money freely given to fund ITER is ridiculous. The amount of companies collaborating with academics on it is in the hundreds. It has so many governments invested in it ... It has so many committees and university boards invested in it it's hard to find a decent physics department that isn't invested in it ... but all of these are political achievements. The physics side of the project is finding physics not all that cooperative.
You should think about what ITER is trying to do as a way to apply massive force to reality until it bends to our will. It's not smart at all (of course the details of doing this are very intricate. There's a difference. Smart is finding a way to beat the calculation speed of a huge datacenter with a 1990's pentium when it comes to calculating digits of pi [1]. Intricate is building a huge datacenter. Both are great accomplishments, of course).
The problem with most of the other projects are facing is that they violate "holy cows" of physics in some way. Polywell physics require a very low-pressure non-thermalized gas, which has been demonstrated but violates thermodynamics theory. It's never going to get past more than one or two physics boards. Z-pinch is one of those tricks that's just too good to be true if it works (it does work to some extent of course). Like polywells, it's a huge risk, so getting physics professors to bet their careers on it is a non-starter. Laser fusion (and other forms of inertial confinement fusion) has similar "WTF" parts that will prevent their widespread acceptance. Because most of these things have multiple projects running, in practice this is about 12 projects.
All fusion projects, with 2 exceptions (one of which is strongly suspected to be a fraud), are happening inside America, funded by either the DoD or DoE. Each of them has a much lower chance of success than ITER, I would agree with that, but if they do succeed, the payoff will be much greater. If polywells work, for example, we should be able to build a 100 megawatt or so fusion reactor the size of a 60s TV set, that could operate in a building that needn't be bigger than a big house. You know, easily small and efficient enough to install on even medium sized ships. Hell, you could probably power planes with it. None of the projects even approach the size and inefficiency of ITER (meaning 10 stories, maximum achievable q value of 10 or lower). ITER should be shelved as "not good enough" and people should go back to the cafe napkin stage.
Google "aneutronic fusion". Granted, ITER will never support it, but that's just another reason to try other things. ITER makes everything radioactive, but aside from that being the easiest reaction there is no good reason for doing that.
As for the size of fusion reactors, go to your nearest (big) hospital. They have a football-sized one in the radiology department. The box it's in is fridge-sized. A Farnsworth-Hirsh fusor to be exact. It will be shielded, because it's using DT fusion. If you really knew what you were doing you could get it to fuse p+B11 and it would produce electrons, but it would defeat the point for the hospital, as it's neutron radiation is exactly why they have it.
Now granted this reactor has a q value of 0.02 or so. But if there were a smart way to get it to a q of 100 or so that would almost be your lightbulb example. Get it to work on p-B11 and ...
Edit: For the morbidly curious 1w / (3 x 10 -13)j/electron = 3.3e+12 electrons. 10M from the device you have a sphere with a surface area of 1256m^2.
http://miscpartsmanuals3.tpub.com/TM-55-315/TM-55-3150021.ht... 1 rem (neutrons/cm2) from 2.5 MeV neutrons = 3* 10^6 neutrons / cm^2. 100 rem received over a short time period are likely to cause acute radiation syndrome (ARS), possibly leading to death within weeks if left untreated. So, 3 * 10^6 * 100 * 100^2 * 1256 = ~4 * 10^ 15 electrons would kill ~1/2 the people standing within 30 feet in ~10 minutes. And far less time than that to start handing out cancers left and right. And at 10 feet that's ~60 seconds. (Assuming I did not mess up...)
PS: It's far safer to spread the same dose over longer periods of time. But, if your working near a useful fusion device it needs to be heavily shielded.
http://en.wikipedia.org/wiki/Nuclear_fusion#Criteria_and_can...
They are harder to initiate than the D-T reaction, which is why they're not the ones being tried first for fusion-generated electricity; but they're perfectly good fusion reactions which have been observed in experiments.
You do if you're using fuel for which there are no possible paths that result in neutrons. For example, He-3, proton-lithium-6, or proton-boron-11.
Still, if your going to pretend we can do anything harder harder than DT or perhaps DD your might as well go for p+p as it's fairly 'clean' and the fuel is plentiful.
You're right that there will be some fuel particles that escape confinement and react with the chamber walls; but proper confinement makes this effect very small, much smaller than the numbers you were quoting for radiation exposure, which are based on using fuels that produce neutrons as reaction products. Neutron's can't be confined in a plasma because they're uncharged, so they immediately escape and hit the chamber walls. Protons, and other fuel particles, don't have that problem since they're charged and can be confined (if they couldn't be you wouldn't be able to make the reactor work at all).
Which gets back to my first point you can have low nitron fusion but if your generating useful amount of power your going to be makeing significant amounts of neutrons simply because there so deadly.
Do you have a reference for this? As I understand it, there aren't significant amounts of protons in our current plasma experiments to begin with, and at the densities we use in those experiments, the cross section for p-p reactions is way too small for them to appear.
> if your generating useful amount of power your going to be makeing significant amounts of neutrons
For current and foreseeable reactors, I agree; but I don't think this is a valid blanket statement about every possible type of fusion reactor that could ever be built, even when our technology has advanced well beyond where it is now.
As to the long term potential I don't think we can rule it out in the longer term, just that when people talk about fusion without neutrons they mean low levels or don't actually know what there talking about.
And I'm asking if you have seen any actual evidence that it happens. I have not, and the information I have seen, which I mentioned, leads me to believe that it should not have happened in any fusion experiments we've done to date. That's why I asked you for a reference.
> the sun is 1/10th of ITER's goal temperature so your well in the range for PP fusion
No, the ITER is not "in the range for PP fusion", because temperature is not the only requirement. You also need sufficient density. The density in the Sun's core is many orders of magnitude larger than the density of plasma in ITER or any other Earthbound fusion experiment. That has a huge effect on the PP reaction cross section.
As to PP fusion from what read. Without a large enough plasma the beta more common proton emission path >99.99% ends up costing more energy than you gain from beta-plus decay <0.01%. "The least stable is 5He, with a half-life of 7.6×10−22 seconds, although it is possible that 2He has an even shorter half-life" http://en.wikipedia.org/wiki/Isotopes_of_helium#Helium-2_.28...
As to PP fusion from what read. Without a large enough plasma the beta more common proton emission path >99.99% ends up costing more energy than you gain from beta-plus decay <0.01%. "The least stable is 5He, with a half-life of 7.6×10−22 seconds, although it is possible that 2He has an even shorter half-life" http://en.wikipedia.org/wiki/Isotopes_of_helium#Helium-2_.28...
How much less common? Again, have you seen any actual evidence that P-P fusion events actually happen in actual Earth-bound plasmas? Because the numbers I've seen indicate that at Earth-bound plasma densities, such events are so unlikely that we should not expect to have observed any.
PS: And thanks for this, it's good to be called out on something like this. I tossed out the PP comment without thinking though simple contamination is a far larger source of high energy neutrons.
It looks to me like the comment about the rate being too small to measure in the lab applies to the PP fusion into deuterium; that's the reaction referred to in the statement you quote. The positron that's produced is not a "beta-plus decay of deuterium"; it's a product of the PP fusion into deuterium, part of the same overall reaction.
They said that fusion is not a certain number of years away, but around 80 billion dollars global funding away.
Here is the full QA
http://hardware.slashdot.org/story/12/04/11/0435231/mit-fusi...
However, it's not clear that we need to know all that to achieve controlled fusion, and the fusion reactors that are being pushed for global funding are not being pushed as improving our basic science about plasma physics. They are being pushed as engineering efforts that will achieve ignition. As I said in another post upthread, we know what conditions we need to achieve for ignition; we just don't know how to engineer a device to achieve them. That's what ITER and these other devices are supposed to help us figure out.
"Because of science - not religion or politics - even people like you and me can have possessions that only a hundred years ago kings would have gone to war to own. Scientific method should not be take lightly.
The walls of the ivory tower of science collapsed when bureaucrats realized that there were jobs to be had and money to be made in the administration and promotion of science. Governments began making big investments just prior to World War II...
Science was going to determine the balance of power in the postwar world. Governments went into the science business big time.
Scientists became administrators of programs that had a mission. Probably the most important scientific development of the twentieth century is that economics replaced curiosity as the driving force behind research...
James Buchanan noted thirty years ago - and he is still correct - that as a rule, there is no vested interest in seeing a fair evaluation of a public scientific issue.."
-From Kary Mullis, the Nobel Prize in Chemistry winner (and the genius inventor of PCR) in an excellent essay in his book "Dancing Naked in the Mind Field".
http://www.psfc.mit.edu/research/alcator/pubs/index.htm
There is a sunk cost effect where it must cost a billion to build another, but only a couple million to keep it running, so once you throw it out, you better be sure you'll never find it useful again because another experiment would have a capital cost of about a century of labor/maintenance cost. Something that big and cool would seem quite capable of generating enough "stuff" to be worthwhile.
Considering that there are private companies willing to invest in a project to mine asteroids, I think this is too pessimistic.
Also, you are implicitly assuming that governments are better at long time horizon projects than private companies. I've gotta disagree with that. A government's time horizon is the next election cycle. Yes, projects like MIT's fusion reactor can continue to be funded for decades, but in order to keep that funding coming, the people working on the project have to spend a significant amount of time every year or two convincing the government that the project is worth continuing to fund. We're only hearing about it now because they weren't able to do that this time around.
A private company that is willing to make a long time horizon investment is not going to have the same overhead required to keep the funding going, simply because there are fewer masters to answer to.
(To be truly pessimistic...) Nor will NASA or any other governmental agency ever make use of asteroid resources. That'd be too useful. Instead they'll spend 400 years doing political stunts and in-space handshakes. In 400 years time, we'll all have uploaded ourselves into The Cloud and gone exploring simulated galaxies. The real moon will be empty, save for some smiley faces carved in it from Earth with lasers.
(To address the actual topic at hand, I think long-term research isn't getting done on either side.)
We'll see. The companies are being funded by people who have enough personal wealth to do it without having to answer to public company shareholders or directors; so they don't have to show revenue or profit the way a publicly traded company would.
If you, or anyone, can come up with an example of a company actually doing that, not just saying they want to, then I'd definitely reconsider.
I'm guessing that you're talking about Planetary Resources with the asteroid mining thing? I didn't see much about hard numbers from them, but I'm doubtful that they will ever spend tens of billions of dollars before getting a penny in revenue.
I can believe that private companies are good at making longer-term investment in technology that directly benefits a profitable line of business that they're in. Stuff like Intel doing research towards smaller component sizes on ICs. But who's going to invest long-term in fusion power? There's no company with a revenue stream related to it that I know of. The costs will probably break a hundred billions dollars before anybody gets plants on the grid that actually generate any revenue at all. I'm just not seeing it. If it were a realistic solution, why isn't it already happening?
http://beta.slashdot.org/story/167399
I'm not sure exactly when it was published, since their new Beta thing hides dates older than 2y ago, but it's great for showing why the MIT project, and all of the other fusion projects, deserve to get a lot more money than they're getting now.
Aside from that, comparing it to ITER is a bit silly since ITER won't be operating for another eight years. We shouldn't just shut down research while we wait.
I see that you noted this, as well. I hope Slashdot will reverse this. Some older posts are still very useful and relevant; however, knowing their context including timing is an important part of this.
Slashdot used to cater to "technical people", i.e. people who can cope with dated material that is older than e.g. "last week", and that is still specific -- some of us still remember calendars more than a year old and what was happening at specific times of those years.
I hope that is not changing. Such a change is not the kind of "social" that I want...
--
P.S. From the non-beta version of the site:
> Posted by Soulskill on Wednesday April 11, 2012 @10:52AM
I can only imagine how horrifying it is to have your thesis work mired in these political struggles.
It's not surprising they they have so much trouble getting time on the machine. They barely have the funding to run it. During the budget problems last year they went a whole year without running it at all.
1. http://www.forbes.com/sites/jamesconca/2012/06/10/energys-de...
My question - will this type of political play create enemies? I can't imagine that the politicians who were overruled were happy, and they can make MIT pay with 1000 small cuts.