Hints of a new particle at LHC

[PB666]

http://www.nature.com/news/lhc-sees-hint-of-boson-heavier-than-higgs-1.19036

The particle produces 2 @  375 GeV photons when it decays. Suggesting it has 7500 gEV of energy. 

Not a higgs either

Edited December 16, 2015 by PB666
Fixed several typos and ommisions

[Yasmy]

Well, it's not known yet if it's a Higgs or not.  There could be a single Higgs particle, or there could be four all with different masses.

Also, it's GeV, not gEV.

Edited December 16, 2015 by Yasmy

[PB666]

Four higgs particles is not going to create a pair of photons, that can be ruled out, that out of the way this paricle is 10 times the size of the previous higgs.

Second i don't recall anything about the previous higgs indicating it made photon pairs as it decayed. 

 

So its different enough not to call it a higgs, if it makes you happy i can call it sighg, having all the letters but not "higgs"

[Findthepin1]

20 minutes ago, PB666 said:

this paricle is 10 times the size of the previous higgs.

welcome to the McDonalds Boson

[Frida Space]

Interesting discovery. Will have to follow it up with more results.

Edited December 16, 2015 by Frida Space

[Frida Space]

10 hours ago, PB666 said:

The particle produces 2 @ 750 GeV photons when it decays. Suggesting it has 1500 gEV of energy. 

However the article refers to a "750 GeV boson" and "pairs of photons of 750 GeV combined". This means that the photon pair has 750 GeV of energy, but each individual photon "only" has 375 GeV, meaning indeed that the boson has 750 GeV.

This is confirmed when the article states that the possible boson "would be about 4 times more massive than the next heaviest particle discovered so far, the top quark". The top quark has a mass of 173 GeV/c^2, which means that the 750 GeV boson would be indeed about 4 times more massive.

So it's a 750 GeV boson, not a 1500 GeV one.

[Steel]

1 hour ago, Frida Space said:

Interesting discovery. Will have to follow it up with more results.

Not a discovery yet. The results are only ~2.5 and ~3.5 sigma confidence from the two experiment that got this data, so not a discovery until its 5 sigma, which will happen probably next year provided this isn't a statistical anomaly.

[Frida Space]

3 hours ago, Steel said:

Not a discovery yet. The results are only ~2.5 and ~3.5 sigma confidence from the two experiment that got this data, so not a discovery until its 5 sigma, which will happen probably next year provided this isn't a statistical anomaly.

Yes, I meant "discovery" as "finding/result". Sorry for my poor English

Edited December 16, 2015 by Frida Space

[PB666]

7 hours ago, Frida Space said:

However the article refers to a "750 GeV boson" and "pairs of photons of 750 GeV combined". This means that the photon pair has 750 GeV of energy, but each individual photon "only" has 375 GeV, meaning indeed that the boson has 750 GeV.

This is confirmed when the article states that the possible boson "would be about 4 times more massive than the next heaviest particle discovered so far, the top quark". The top quark has a mass of 173 GeV/c^2, which means that the 750 GeV boson would be indeed about 4 times more massive.

So it's a 750 GeV boson, not a 1500 GeV one.

Interesting because on the plot they show online they also have a set of points at 1500, not as many as at 750. Do they have another massive particle in the pipeline?

I should say looks over 1000, may not be 1500.

Edited December 17, 2015 by PB666

[Yasmy]

While the Higgs preferentially decays into a b/anti-b pair or W or Z boson pairs, it does decay to two photons via an intermediate loop. Of course it can't decay directly into two photons, or photons would be massive, like the photon's electroweak partners: the W and Z bosons.

A second Higgs is considered by many to be the most likely thing to see beyond the Standard Model particles. This news report does nothing to change that. Other options include a whole zoo of things, such as a massive axion, a Kaluza-Klein graviton (not to be confused with, or to replace, a normal graviton), some supersymmetric partner to the Standard Model particles, or some completely new, unheard of particle.

Edited December 16, 2015 by Yasmy

[K^2]

10 hours ago, PB666 said:

Four higgs particles is not going to create a pair of photons, that can be ruled out, that out of the way this paricle is 10 times the size of the previous higgs.

Second i don't recall anything about the previous higgs indicating it made photon pairs as it decayed.

No, the claim is that there is a heavier Higgs Boson which has a decay mode into two photons. It's entirely plausible given general Higgs mechanism. Whether there is a single Higgs Boson or several different ones, potentially with different masses, depends on the exact symmetry responsible for Higgs Mechanism, and there are several that are compatible with current observations.

Could always be a completely new field with its own symmetries, of course.

P.S. I'm tempted to say that it's the supersymmetry pair, but I doubt a lot of people would appreciate that humor.

Edited December 16, 2015 by K^2

[PB666]

1 hour ago, K^2 said:

No, the claim is that there is a heavier Higgs Boson which has a decay mode into two photons. It's entirely plausible given general Higgs mechanism. Whether there is a single Higgs Boson or several different ones, potentially with different masses, depends on the exact symmetry responsible for Higgs Mechanism, and there are several that are compatible with current observations.

Could always be a completely new field with its own symmetries, of course.

P.S. I'm tempted to say that it's the supersymmetry pair, but I doubt a lot of people would appreciate that humor.

The problem is likelihood, what is the likelihood that you could have a higgs just capable of generating two photons. Yes there are likely larger higgs, but this one is an odd fit without seeing all kinds of other products. I notice that some articles mention higgs, but it is unclear were they are talking about higg-like bosons, which could be anything, its just like saying a w-boson is a z-like boson. Biased really, though other here-named options like super sym and 5th dim gravitons are also not choice other. 

Placeholder particle name until more characterization is done. 

[PB666]

7 hours ago, Steel said:

Not a discovery yet. The results are only ~2.5 and ~3.5 sigma confidence from the two experiment that got this data, so not a discovery until its 5 sigma, which will happen probably next year provided this isn't a statistical anomaly.

Its real, combining the results at least suggests so. 5 sigma is a bit overkill

 

[K^2]

4 hours ago, PB666 said:

The problem is likelihood, what is the likelihood that you could have a higgs just capable of generating two photons.

Two photons is an expected and observed mode of decay for Higgs. The branching fraction just happens to be somewhat low. Keep in mind that pairs of photons just happens to be the thing they are detecting in this experiment. pp collisions create huge showers of all kinds of junk. We might not be able to detect WW pairs. They are short-lived and interact with stuff in the shower, making coincidences harder to detect. Photon-photon, on the other hand, is very easy to pick up on coincidence counters.

Here is a picture from Wikipedia.

As you can see, at 750GeV we see something like 10^-7 of decay events going to photon-photon.

[PB666]

6 hours ago, K^2 said:

Two photons is an expected and observed mode of decay for Higgs. The branching fraction just happens to be somewhat low. Keep in mind that pairs of photons just happens to be the thing they are detecting in this experiment. pp collisions create huge showers of all kinds of junk. We might not be able to detect WW pairs. They are short-lived and interact with stuff in the shower, making coincidences harder to detect. Photon-photon, on the other hand, is very easy to pick up on coincidence counters.

Here is a picture from Wikipedia.

As you can see, at 750GeV we see something like 10^-7 of decay events going to photon-photon.

What i said, unlikely, 

Think about what you are saying: collisions, subset collisions  with large cross-sectional overlap, subset collisions that produce gamma, subset produced gamma that is detected. To get a sigma of 2.5 to 3.5 you would need to have something like 10^10 collisions.

The photons themselves carry more info than hv, the also carry directional and polarity information which should be interpretable.

Edited December 17, 2015 by PB666

[K^2]

26 minutes ago, PB666 said:

The photons themselves carry more info than hv, the also carry directional and polarity information which should be interpretable.

These are pair coincidence counts. They tell us that particle that decayed produced no massive product in the same decay and that its total angular momentum was zero. This is consistent with lepton-antilepton bound state, a bunch of mesons, and yes, the Higgs boson. There is no "directional" or "polarity" information gained from such events.

And sure, it'd take billions of collision events. Which is basically what LHC is designed to produce.

Edited December 17, 2015 by K^2

[Steel]

11 hours ago, PB666 said:

Its real, combining the results at least suggests so. 5 sigma is a bit overkill

 

5 Sigma is the required confidence for a discovery, theres still a somewhat reasonable chance right now that this is a statistical fluke

Edited December 17, 2015 by Steel

[PB666]

10 hours ago, K^2 said:

 

gnore the quote above, the new gui is sometimes very glitchy, i quotedd steel's post, the gui quote k2's posted and then posted it before i had a chance to edit it.

In response to steel.  The reality is that it is more complicated, 5 sigma is used in certain fields because certain causes of variation are either difficult or impossible to control for. 2 sigma in genetics used to be frequently wrong, for example for studying genetic association with low penetrance disease because authors were sampling several genes and publishing each paper separately some authors even failed to report on genes that gave negative results.  When bonnferonnis corection was applied to these statistics they corrected (coorected threshold was frequently in the 10^-8 to 10^-10).  But when authors looked at the expected versus observed low probabilty associations they found that the threshhold accepted H0 when it was probably not true and these actually outnumber rejected H0 by several magnitudes, subsequent studies proved. 

The basic problem is there is no good way to correct for percieved variation other than discovering the source of that variation and crossmultiplying it out. (I hate this gui, really it just goes and starts doing what it wants, im saying this here because i went back to edit part, but the gui would not let me scroll down, i had to close the kb gui, try to scroll down and reapply kb). In this instance when you have no idea how big the boson has to be, which is probably better than using single sided if you were simply asking is the a particle out there bigger than the biggest  particle (~170 GeV for a fermion).

1 sigma splits +/-16%

2 sigma splits +/-2.3%

3 sigma splits +/- 0.13%

4 sigma splits +/- 0.0032%

5 sigma splits +/- 0.000049%

So basically if you need a 1 in 3.5 million chance of being wrong.....

A sigma of 2.5 is wrong 0.62% of the time

A sigma of 3.5 is wrong 0.032% of the time, without considering other sources of variation, the odds of two standalone experiments being wrong based on these two in any good meta analysis would place the p value below the 5s threshold. .

However this is not the data, the problem is that we do not know how many ecperiments have been done, and whether we should include the results of past experiments that were inconclusive, because of things like marginal power. Then for each of these we have to ask the level of dependency or relevancy on the question being asked, for example is there no evidence of a particle between 170GeV and 999Gev. That needs to be taken intonthe stats. 

 

Edited December 17, 2015 by PB666

[PB666]

On December 17, 2015 at 5:55:48 AM, K^2 said:

These are pair coincidence counts. They tell us that particle that decayed produced no massive product in the same decay and that its total angular momentum was zero. This is consistent with lepton-antilepton bound state, a bunch of mesons, and yes, the Higgs boson. There is no "directional" or "polarity" information gained from such events.

And sure, it'd take billions of collision events. Which is basically what LHC is designed to produce.

Wait, the graph shows the paths of the photons, is this contrived?

I still find it hard to believe that they have captured 10 billion collisions in 2 years in the TeV range. 

Edited December 18, 2015 by PB666

[PB666]

http://www.bbc.com/news/science-environment-36703721

Quote

scientists detected more photon (light) particles being produced than expected - the aforementioned "bump". More precisely, they saw an excess of photon pairs with a combined mass of 750 Gigaelectronvolts (GeV). -http://www.bbc.com/news/science-environment-36703721

"The big reason that people are excited about this bump is that both experiments (Atlas and CMS) saw a hint in roughly the same place. But even this is not completely unlikely." - same article.

OK, based on what has been observed.

- Statistics may be weakening despite the coincident, so there is also a possibility its something else, not 750 GeV

- If it does exists is a boson of some type, possibly a higgs variant (spin 0) possibly as graviton (spin 2).

- Not a supersymmetry particle, lacks the characteristics required to be a higgsino or gluino.

 

[PB666]

OK, maybe not.

http://www.bbc.com/news/science-environment-36976777

 

[Frida Space]

1 hour ago, PB666 said:

OK, maybe not.

http://www.bbc.com/news/science-environment-36976777

 

The beauty of science

[PB666]

The perils of statistics

[wumpus]

On 12/16/2015 at 0:39 PM, Steel said:

Not a discovery yet. The results are only ~2.5 and ~3.5 sigma confidence from the two experiment that got this data, so not a discovery until its 5 sigma, which will happen probably next year provided this isn't a statistical anomaly.

~3.5 means p<.05 and ready for publication in sciences less deluged with data (and mostly less picky and not willing to admit the size of the data) than particle physics.  Not sure how many sciences get the shear quantity and quality (after lots of calculation and filtering) of particle physics.