The black circles are the experimental data. That's what they really saw. (Actually, these measures are very indirect and noisy, so this is the data after a lot of processing, but let's assume that this is standard incontrovertible processing.)
Each measurement has an experimental error, that is marked by the + cross in each circle. So imagine that each circle is a bigger fuzzy ellipse that is 7 times wider and 3 times taller than the circles.
If they assume that there is no new particle, in their simulation they would get the dashed curve, that is always decreasing. It's dashed in the original graph and painted in gray in the "improved" version.
The problem is that the experimental data in the ellipses are too far from the theoretical curve of the simulation without a new particle, so something strange is happening. (A new particle, experimental error, other ...)
They assume that the difference is caused by a new unknown particle, with some properties. In particular, they have to guess the mass of the particle.
If they guess that the particle is very heavy, for example 17.6MeV, they get a curve that looks like an "U", it's dot-dashed in the original and painted in blue in the "improved" graph. It coincides with the "no particle" curve when the angle is less than 130°, but then it starts increasing. The new curve doesn't look like the experimental ellipses at all, so they have to keep trying.
If they guess that the particle is very light, for example 15.6MeV, they get a curve that is also decreasing, it's dotted in the original graph and painted in yellow in the "improved" graph. It coincides with the "no particle" curve when the angle is less than 110°, but then it is bigger. But comparing it with the experimental data, it is too big for 110°-130° and too small for 130°-160°. The new curve also doesn't look like the experimental ellipses at all, so they have to keep trying.
If they guess that the mass of the particle is 16.6MeV, they get a curve with a bump, it's solid in the original graph and painted in red in the "improved" graph. It coincides with the "no particle" curve when the angle is less than 110°, it's slighter bigger at 120° and has an interesting maximum near 140°. The new curve pass through all but the last ellipse, so it looks like a good fit. So the conclusion is that they probably have a new unknown particle with a mass of 16.6MeV!!!
Actually, they (obviously) tried with more possible masses, not only this three values. And compare the simulation with the data using statistics, not the drawings. The details are in figure 5. With 16.6MeV they got the better agreement with the experimental data.
The particles have other properties besides mass. Some of them can be calculated with theoretically, for example the charge. Others, like spin, have to be tested using simulations, because using other values produces different looking curves.
Obviously, the complete paper has more details, but I hope this is a good starting point to start reading it.
1. They see a bump (aka "resonance") which can generically be modeled as a new particle/state (whether it's a dark force particle, or a nuclear state or atomic state or whatever).
2. (Small correction to the above post: the mass of the resonance is 16.6 MeV not GeV) Atomic physics operates at the eV scale, and the mass of a Be nucleus is roughly 10 GeV (10^9 eV). So, given that this bump fits the 16.6 MeV (10^7 eV) scale which between the other two scales, this might either be some detailed nuclear physics (eg: some Berellium resonance) or an honest-to-god new particle.
3. Since the angular separation is roughly 135 degrees, it seems reasonable that the intermediate state is moving fast ("boosted") [1]. That means that it cannot be as heavy as a whole nucleus -- therefore, if this was a Berellium resonance then the angle between the electron-positron pair would leave back-to-back (180 degree separation). That is what pushes the interpretation towards a new particle/force.
Good explanation from you two, but let me ask: how much does the article states that this is a new "fundamental force" as opposed to simply some new particle working within the known forces?
From what I understood the angle would imply that this force interacts with a given range but not outside of this range (so it might be strong nuclear? hidden dimensions? new forces?)
In quantum field theories the forces are the same as new bosons. (The exact relationship is a bit hard to explain, since particles enter the picture as a really cool trick to solve PDEs, not as a fundamental entity of the theory.)
Looking quickly through the two papers mentioned [0, 1], there is a (statistically) very solid result that something is going on with 8Be, which is only very barely a nucleus and is probably better thought of as a bound state of two alpha particles. In [0] there is a experimental observation of some kind of effect which can be explained by a 17 MeV boson. [1] looks at other experiments and finds a way around current experimental bounds by building a "protophobic" model, that is a boson that does not interact with protons. Interestingly they find that in this model they can explain the anomalous magnetic moment of the muon, if the parameters are somewhat close to already excluded limits. (Fig. 1 bottom panel in [1]) So the next generation of experiments should rule out or discover the X boson. (And actually there are many experiments which seem to have a shot at this.)
My take on this, it is probably a parametrization of something we do not understand in terms of a particle. This something may be a experimental effect or a theoretical lack of understanding or insufficient methods for calculation, or a new particle. The energy range is sort of interesting since it is close to the energy range of hadron spectroscopy, a field which is littered with 2-sigma 'tantalizing hints' of new physics. To end this, I believe it is a temptation to speculate (read I will claim I resisted the temptation until "told you so" is an appropriate response), that a heavy charged particle, cough di-photon excess, could be involved in creating the different couplings to u and d quarks.
Yeah, I found the lack of discussion about possible causes by modifying atomic physics rather than go directly to new particles disappointing. Maybe there is a good reason to trust that standard theory cannot be modified to introduce bumps in the angular distribution, but I think it really should be mentioned in that case.
Honestly, this gives a very hep-th impression: "An anomaly! we solve it by adding an X-particle with properties that magically explain the anomaly. Hurray!"
Its actually in hep-ph though, which I guess shows that I'm out of touch : / Anyway, exploit hunters have there uses too I guess.
Aha! I'm currently at a conference where one of the authors of [1] gave a talk [2] about this and there was an interesting follow-up discussion. Never expected it to be on HN already!
Some thoughts from the thick of things:
1. It's important to remember that nuclear physics backgrounds for such an experiment are hard to model (so that one can subtract them out). The field is littered with "anomalies" coming from systematics and backgrounds. This is why physicists are very cautious with this claim. The authors of the theoretical model claim that it should be easily testable (interesting signals in several different experiments) over the next 2-3 years, so that's one reason to be optimistic about knowing where this stands.
2. This model goes against the grain of most of our theories for how the baryonic matter (essentially nuclei of all kinds) could be created in the universe! If you believe in this newly proposed model as truly another force of nature, then one might need a new paradigm for generating all the matter we see.
There are some theoretically intriguing aspects of the model, so I'm looking out for the follow-up paper by the authors (sometime in the next few weeks?) dealing with several of these aspects.
--
Edited to add a link to the slides and elaborating on point #2 below
For more advanced readers (along with the caveat that I don't fully understand the details of their model):
We understand very well the production of nuclei soon after the big-bang [3] and there's no room to modify that theory. So, any more fundamental model will have to produce baryons before that era of the universe. The Sakharov conditions [4] for producing baryons require a breaking of Baryon number (you can't have a net excess of baryons over anti-baryons unless you break the symmetry between the two). Usual models produce baryon number by using the fact that U(1) baryon number is anomalous in the presence of electroweak sphalerons [5]. However, in the newly proposed theory U(1) B-L [6] is gauged and hence cannot be anomalous (thereby invalidating the conventional paradigm for generating baryons). Maybe a spontaneous breaking of that symmetry happens early enough in the universe to not mess with the theory of nucleosynthesis but I'm not aware of work along such a paradigm.
Can you expand a bit on the 2. point? (Probably there are some implications for bariosynthesis, but I don't have a good feeling how they should look like.)
Were you expecting a virtuous "Build it and they will come" thing to happen in the community? I'd imagine a ton of reports coming out all the time, not everything of value bubbles to the top right away.
My point is not that it wasn't picked up right away -- it's that the discover only became news when Americans got hold of it. More, the reporting is more about the American confirmation and interpretation than the Hungarian discovery!
No. Considering the amount of time it took for the original research to begin with and for any subsequent attempt to replicate, the fact that 'americans' were cited as the replicators is superfluous.
If I understand this correctly, the search for another fundamental force is attempting to explain dark matter using the standard model? Are there any pools of thought that try to explain the universe without dark matter?
This may not be the right thread to ask this in but it's right enough I suppose.
What exactly does it take for a thing to count as a fundamental force?
I loosely associate forces with bosonic fields -- photons for EM, gluons for strong, and something confusing for weak. But then no one seems to want to call the Higgs field a "force".
This is an good question which I'm sure there is a good answer for, but which I haven't heard stated by particle physicists in a clean way that is widely accessible to other physicists. I believe the distinction generally hinges on whether the boson is a gauge boson, associated with some symmetry; this does not apply to the Higgs. But I don't really get why that makes it worthy of the intuitive label "force".
Here's one incomplete answer:
> Q: Why isn't Higgs coupling considered a fifth fundamental force?
> A: The Higgs exchange between matter particles can certainly be called a force. Whether it can be viewed a fundamental force, is a matter of taste. But there is one important distinction between the force due to the Higgs exchange and the usual fundamental interactions. The strong and electroweak interactions are described as gauge interactions. It means that they are not put by hand but arise automatically when you require that the matter fields be invariant under local certain internal transformations (phase rotations, color transformations, etc.). In contrast to that, the "Higgs force" is put into the model by hand, as its presence is not driven by any symmetry consideration. You can equally nicely consider a model with zero Yukawa coupling.
Physicists will typically say "friction force", or "van der Waals force", or "centrifugal force" etc. when we talk about the interactions of ordinary things. These are all caused by the four fundamental forces (or they are ficticious, like the centrifugal force).
And yes, the non-fundamental forces (like friction) are complex interactions of fundamental forces. For friction it's the sum of electric forces between atoms. The important point is that an atomistic model of friction gives you an essentially unsolvable problem, while the effective macroscopic model of friction is high school stuff.
Unsolvable? Rubbish. I was under the impression with regards to the 4 forces that the only mystery was the mathematical relationship bewteen electricity and gravity. I'm sure it is not as big a mystery as popular science would have us believe.
When I say unsolvable, I mean that you can't solve for the speed of a brick sliding down an inlined plane by considering the electrostatic interactions between the ~10^17 pairs of interacting atoms per square centimetre of contact surface. It's just unsolvable in practice.
But the effective formula that the magnitude of the friction force equals a material constant times the magnitude of the normal force from the plane onto the brick gives you a very simple model (solvable with pen and paper).
Modified gravity theories might be second in popularity to dark matter. What should be added is that it is a very distant second place. Vast majority of researchers currently favor dark matter. Observations like those from the Bullet cluster are hard to explain otherwise.
I don't see how this really suggests a new fundamental force, but certainly it's known that everything about nuclear physics and radioactive decay are not fully understood. Hopefully this is a signpost to some of those answers.
The Higgs boson represent an exception, and is the result of quite a complicated mechanism. And if the new particle were a Higgs-like boson it would be even more surprising.
Except for the Higgs boson, the usual picture is that matter particles are fermions and force mediators are bosons:
One interpretation I've seen is that there are 4 Higgs bosons, three of which are "swallowed" by 3 of the the weak-force bosons which and become the W+ W- and Z. The remaining 4th Higgs boson (the 4th degree of freedom of the Higgs field) is what we can see, together with the 4th weak-force boson, the photon, which remains massless.
So I think it's reasonable to consider that the Higgs boson belongs to the weak force.
I'm not a physicist and I barely understand the ELI5 version of this stuff, but no, I do not think that saying that the Higgs 'belongs' to the weak force is correct as it is a proper excitation of a separate fundamental field.
IIRC, the Higgs field is a scalar field (same field strength at all points of space/time), whereas the other fundamental forces are vector fields (field strength varies with space/time).
Is this correct? And if so, does this fact matter in deciding whether to say Higgs == fundamental force?
It depends on the point of space-time, but at each point of space-time it is just a scalar value, unlike a vector for a vector field. Moreover in physics we classify fields by their Lorentz transformation properties/representation. So when we say scalar field we actually mean that is transforms in the "(0,0) representation" of the Lorentz group (although then it's not yet specified whether one means scalar or pseudoscalar).
If it were the same at all space time points it would be a constant, and rather boring (more like a cosmological constant).
AFAIK, in general relativity, the gravitational field is not a vector field but a tensor field.
EDIT: btw, you got the scalar/vector definition wrong.
A scalar field is one where each point in space can be labeled by a single scalar (not the same vale for all points), for example a temperature reading at different locations.
A vector field assign a distinct vector (i.e. magnitude and direction) to each point in space, for example if you take a compass around the globe you can assign to each location the direction the compass arrow is pointing.
A tensor field is of course assign a tensor to every point in space. I only have rough understanding of what a tensor is, as a first approximation it can be understood as a matrix with special properties.
These types of signals are a very common path that people who look for new forces follow. The phase space to search for deviations in the standard model is very big, so when things like this pop up they offer better avenues of exploration than throwing darts.
Is it a new force? It looks like the type of particle that would explain it hasn't been entirely ruled out from all the global data we have. If other planned experiments that will be sensitive to the same properties of the proposed particle and aren't dependent on nuclear decays do see it, then there you go. If they don't then they don't.
Highly unlikely, but exciting nevertheless. A 17-MeV anomaly is rather large. It would be nice if it did shed insight into dark matter / energy, but the connection does seem to be a stretch.
I mean, it could. But I haven't seen any mention that the conditions of this experiment match the interstellar conditions that indicate Dark Matter (which is usually the angular velocity of galaxies not matching the theoretical prediction).
There's smarter people than me working on dark matter, but one thing to consider is that dark matter may well be simply that -- matter that we're unable to see for inane reasons such as being on a plane of vision that we're unable to see, similar to the unified model of an active galactic nucleus: http://web.physics.ucsb.edu/~ski/skipicture-1.html.
Again, exciting, but too soon to really say anything. Everyone should be suspect of the claims, it's just par for the course in physics.
Observations of accelerated particles' decay follow relativity, they decay with slower frequency relative to the observer. If dark matter collisions were the cause we shouldn't be observing that.
My favorite quote was “Perhaps we are seeing our first glimpse into physics beyond the visible Universe”. This sounds very likely. Considering that the 'bump' occurred at a certain angle and only then. This reminds me of the science that purportedly exists when combing specific minerals with specific elements. God how I would love to be a mad scientist.
Ok, so not a physics expert so this might sound stupid, but could this mean that radioactive decay is not truly random, just that we can't currently model it? If so, that could have some interesting implications for crypto systems that use radioactive decay as a random seed...
No. It just means that there are some more events that may occur and may change the probabilities of some events slightly. That is, where before we assumed that decay only occurs through events A and B, with A chosen 45% of the time and B chosen 55% of the time, we now know (assuming all these findings are true) that there are actually B' and B'', chosen 0.001% and 54.999% of the time respectively, which result in the same outcome particle-wise but with take different paths to get there, resulting in different particle properties (momentum, mostly).
For systems using radioactive decay as seed, this should have very little effect, as the involved previously-unknown effects are extremely small and the properties influenced by them are unlikely to be used as a source of randomness anyway.
No, this doesn't mean that at all. There are lots of other (accountable) structures in energy/angle correlation spectra like this, but it doesn't change the quantum mechanical aspects of it and the randomness in time distributions.
There is such a thing as radioactive decay. It's a very widely studied and replicated phenomenon, and is used in many practical applications (e.g. smoke detectors). It's about as close to objective fact as any scientific result can get, and that will not change.
Perhaps you mean that radioactive decay may have a causal mechanism involving dark matter? Possibly. For such a model to be taken seriously, it would need to agree with our (presumably very numerous and precise) existing observations, as well as either predict the outcome of some new, falsifiable experiment; and/or simplify/unify some disparate existing results into a common framework.
Even if such a model manages to overcome all of those filters, radioactive decay would still be a very real phenomenon. For example, discovering that light is an electromagnetic wave doesn't at all imply that "there's no such thing as light"; discovering that gravity is curvature of spacetime doesn't mean "there's no such thing as gravity"; etc. Like radioactive decay, those are phenomena which we have very direct evidence for, and which theories and models must try to explain.
In comparison, some theories/models postulate the existence of entities which we have no direct evidence for, such as phlogiston, luminiferous aether, dark matter, etc. The only evidence for these is the success of the model that predicts them; that evidence may be undermined if a better model is discovered which doesn't require these entities (e.g. the kinetic theory of gases for phlogiston and special relativity for luminiferous aether).
Hence, if we discover a better model for predicting flat galactic rotation curves, gravitational lensing in excess of the visible matter, etc. which doesn't require an exotic form of matter, then we might claim "there's no such thing as dark matter".
However, even if that were to happen, we certainly couldn't claim that "there's no such thing as flat galactic rotation curves", "there's no such thing as gravitational lensing in excess of the visible matter", etc. since, like radioactive decay, those are phenomena we have observed.
If a theory disagrees with observed phenomena, or claims "there's no such thing" as the observed phenomena, then the theory is wrong. That's precisely what empirical falsification is about. We've observed the phenomena, so that theory cannot describe the world we live in.
You're taking it way too literally, I think it was clear what I meant. I meant if the decays of atoms could be the result of interactions with the particles of dark matter, rather than spontaneous decay caused by random quantum fluctuations. If it's true, we should be able to observe differing rates of decay depending on the concentration of dark matter in the area.
You wrote "What if there's no such a thing as <something we have studied and understood in detail for decades> and we've been observing interactions with <something we hypothesise might exist but we know next to nothing about> all along?"
Why did you jump to the conclusion that the well understood thing should be replaced by the new and unknown thing?
It doesn't seem to me like a big leap of logic to go from something being shown to be influenced by a so far unknown particle and the same thing being the result of interactions with this particle.
A photon would produce a different shape of the electron spectrum. (It would not produce a bump at 16 MeV, but some extra contribution to the entire shape, since it is massless.)
Some of each. It's hard to blame the media for thinking it could be important in light of the press conferences and press releases. The last section of this article gives some details.
I'll wager a pint (or suitable non-alcoholic equivalent) on it :) Win-win, since if I 'lose' I buy you a pint but something truly wonderful will have been discovered :)
People are not downvoting the comment because they believe that the headline is not clickbait, and the article is not nonsense. They are downvoting because the comment adds nothing that has not already been discussed many times already in other contexts. Headlines are often click-baity, now let's move on please.
1.)The given audience in any form is always changing. He might have given new information to someone who hasn't seen it before(young,new english speaker, ect)
2.)We are talking about SCIENCE and whats great about SCIENCE is that to do it well requires SKEPTICISM. The article itself even says that the results haven't been confirmed by a different specialized lab. I would be floored if we discovered another fundamental force but like all things we should ANALYZE all points of a subject.
> 2.)We are talking about SCIENCE and whats great about SCIENCE is that to do it well requires SKEPTICISM. The article itself even says that the results haven't been confirmed by a different specialized lab. I would be floored if we discovered another fundamental force but like all things we should ANALYZE all points of a subject.
I think you're being too generous with regard to the OP's comment. That comment appeared to be mostly a throw-away attempt to reference Bettridge's law, rather than offering any substantiate thoughts on the potential discovery or its likelihood of being a spurious result.
Bringing up the Betteridge thing became a cliché years ago. It was interesting when pg originally posted it, quickly became tedious, and now is merely mechanical. It's also not as true as people think.
Yes there are bad titles out there, though this article's is light years from the worst. That is no reason to bog HN threads down with predictable gunk. Anything predictable is boring.
i hypothesize that the votes aren't about right-ness or wrong-ness, rather it may be more akin to "yes, we know, we've seen the betteridge link in every other post". it's also not clear that the article is "clickbait nonsense".
In my opinion, it is unlikely to be a new force, but WELL worth pursuing to confirm the results and explore further because of that small chance that it revolutionizes physics in important ways.
The original paper is https://arxiv.org/abs/1504.01527 The important graph is figure 4 in page 4. I made a colored version: http://imgur.com/Mya66KZ
The black circles are the experimental data. That's what they really saw. (Actually, these measures are very indirect and noisy, so this is the data after a lot of processing, but let's assume that this is standard incontrovertible processing.)
Each measurement has an experimental error, that is marked by the + cross in each circle. So imagine that each circle is a bigger fuzzy ellipse that is 7 times wider and 3 times taller than the circles.
If they assume that there is no new particle, in their simulation they would get the dashed curve, that is always decreasing. It's dashed in the original graph and painted in gray in the "improved" version.
The problem is that the experimental data in the ellipses are too far from the theoretical curve of the simulation without a new particle, so something strange is happening. (A new particle, experimental error, other ...)
They assume that the difference is caused by a new unknown particle, with some properties. In particular, they have to guess the mass of the particle.
If they guess that the particle is very heavy, for example 17.6MeV, they get a curve that looks like an "U", it's dot-dashed in the original and painted in blue in the "improved" graph. It coincides with the "no particle" curve when the angle is less than 130°, but then it starts increasing. The new curve doesn't look like the experimental ellipses at all, so they have to keep trying.
If they guess that the particle is very light, for example 15.6MeV, they get a curve that is also decreasing, it's dotted in the original graph and painted in yellow in the "improved" graph. It coincides with the "no particle" curve when the angle is less than 110°, but then it is bigger. But comparing it with the experimental data, it is too big for 110°-130° and too small for 130°-160°. The new curve also doesn't look like the experimental ellipses at all, so they have to keep trying.
If they guess that the mass of the particle is 16.6MeV, they get a curve with a bump, it's solid in the original graph and painted in red in the "improved" graph. It coincides with the "no particle" curve when the angle is less than 110°, it's slighter bigger at 120° and has an interesting maximum near 140°. The new curve pass through all but the last ellipse, so it looks like a good fit. So the conclusion is that they probably have a new unknown particle with a mass of 16.6MeV!!!
Actually, they (obviously) tried with more possible masses, not only this three values. And compare the simulation with the data using statistics, not the drawings. The details are in figure 5. With 16.6MeV they got the better agreement with the experimental data.
The particles have other properties besides mass. Some of them can be calculated with theoretically, for example the charge. Others, like spin, have to be tested using simulations, because using other values produces different looking curves.
Obviously, the complete paper has more details, but I hope this is a good starting point to start reading it.
EDIT: Fixed "GeV" -> "MeV"