Beta Decay, why did they have to resort to Neutrinos?

[GregM]

TL;DR Summary

Beta Decay. Looking for explanation or data on why the 1911~1929AD nuclear energy pool hypothesis for beta decay was falsified; which caused the community to propose neutrinos.

I'm reviewing history of subatomic physics.
By 1931AD the nuclear physics community had decided to propose the neutrino because they couldn't explain beta decay without it.
Alpha and Gamma decays were more confined wrt the energy they would extract from the nucleus i.e. they had energy bands. By comparison Beta decay had a wide spectrum; the electron could be emitted from its nucleus with any one of a range of energies.
The initial guess from around 1911AD to 1929AD to explain this spectrum was simply that there was a pool of energy in the nucleus and the beta decay took a (pseudo) random amount of this energy; But by 1930AD this was ruled out.
I've not been able to find how they falsified this first hypothesis of beta decay. My best guess is they needed the neutrino to get spin conservation.
There are many analogies to the first beta decay hypothesis, such as a photon from a blackbody, or a water particle evaporating from a pool.
Can anyone say or link to what caused the community to abandon the nuclear energy pool hypothesis of beta decay and resort to neutrinos?
Also why was Bohr arguing that beta decay broke conservation of energy? ( should be the same answer as why they resorted to neutrinos )

 

[PeroK]

It was conservation of energy and momentum that was not obeyed by beta decay. That led to the hypothesis of an additional undetected particle to balance the energy-momentum equations.

 

[mfb]

The initial guess from around 1911AD to 1929AD to explain this spectrum was simply that there was a pool of energy in the nucleus and the beta decay took a (pseudo) random amount of this energy

This doesn't explain why alpha and gamma decay wouldn't do the same.

You can measure the energy spectrum of electrons and compare it to the prediction if there is an additional massless (or almost massless) particle emitted, and you get an excellent agreement.

 

[GregM]

It was conservation of energy and momentum that was not obeyed by beta decay. That led to the hypothesis of an additional undetected particle to balance the energy-momentum equations.

Yes I know already.
Perhaps I should reword my OP.
I'm interested in the work done from 1911 to 1930. It takes a college student 60 seconds to decide if energy or momentum equations do not add-up. Why did it take the community of leading physicists 19 years to do the same with beta-decay?
Your 15 words pretty much sum everything we know of that time period. I'm beginning to think no-one has gone back and studied the (mostly German) papers at the time to see what took them so long; while quantum physics and other sciences moved forward quickly over the same time period.

@mfb
>You can measure the energy spectrum of electrons and compare it to the prediction if there is an additional massless (or almost massless) particle emitted, and you get an excellent agreement.
No you can't. The energy and momentum of incoming neutrinos are unknown before the collision with the nucleus, therefore you can't make predictions. You can only watch and see the energy and momentum of the nulceus and electron after the collision, and then retrospectively deduce the energy and momentum of the incident neutrino.

 

[malawi_glenn]

The energy and momentum of incoming neutrinos are unknown before the collision with the nucleus, therefore u can't make predictions

Incoming neutrinos in beta decay?

It is a standard calculation to calculate the energy distribution of electrons in beta-decay, i.e. a "spectrum".

 

[GregM]

>Incoming neutrinos in beta decay?
kek. Yes. I know its convention to assume the neutrino is an outgoing product of a spontaneous beta decay. But by the symmetry in energy and momentum conservation, and that neutrinos are directly undetectable, there's no way of knowing if a beta decay was caused by an incident neutrino or if it was spontaneous and an outgoing neutrino filled the holes in the symmetry. Either way, energy, momentum and spin are conserved; there's no way of discerning if a neutrino went in or out.

 

[Vanadium 50]

Well, wrong theories are usually not taught in schools. At least that's the plan.

You'll probably need to go back to the papers of the day and read them, and they will likley not all be in English. I personally would be surprised if a theory where nuclei carried some hidden amount of extra energy had many backers. First, it's post-relarivity, and a difference in energy would be visible as difference in mass. Second, it completely messes up thermofynamics (c.f. Gibbs paradox), Third, there are decay chains (e.g. ⁹⁰Sr →⁹⁰Zr → ⁹⁰Y although this is not the best example) and there is no correlation between the energy of the first and second beta.

 

[malawi_glenn]

Neutrinos are detecable for sure, because they interact.

In either way, you have to assume the existence of a very light mass and very weakly interacting particle.
Then you work out what predictions such theory would pose. And as I wrote, it is a standard calculation to derive the energy spectrum of the electron in beta decay - which you can compare with experimental data (excellent agreement). Such spectra would only depend on the mass of the nuclei involved.

If you propose "nah there are no neutrinos that is being emitted in beta-decay, the electrons are ejected due to an incomming flux of neutrinos with unknown energy distribution..." - then how are you gonna explain that it is not the other way around? Like what I did in the previous paragraph. Then you would need to have different energy distributions of the incomming neutrino fluxes, depending on what your mother nuclei is. And voila you have an epicycle theory in some sense...

 

[GregM]

...First, it's post-relarivity, and a difference in energy would be visible as difference in mass. Second, it completely messes up thermofynamics (c.f. Gibbs paradox), Third, there are decay chains (e.g. ⁹⁰Sr →⁹⁰Zr → ⁹⁰Y although this is not the best example) and there is no correlation between the energy of the first and second beta.[/sup][/sup][/sup]

>Mass-Energy conservation
That might explain it. The pool of energy in the nucleus hypothesis for beta decay would predict daughter nuclei with slight variations in mass, i.e. the mass-energy not taken by the outgoing electron. It would have taken a lot of experiments and refinements to discern such a tiny discrepancy, enough to keep experimenters busy for 19 years.
Your decay chains logic also makes sense.

Thanks, that's given me something to look into.

 

[Dr.AbeNikIanEdL]

The initial guess from around 1911AD to 1929AD to explain this spectrum was simply that there was a pool of energy in the nucleus and the beta decay took a (pseudo) random amount of this energy; But by 1930AD this was ruled out.

I don't think this was the prevailing theory, or at least not the only one. I found Ellis, Wouter 1927 and Meitner, Orthmann (1930, in German). In their introduction they both describe the option that the electrons are emitted with a sharp energy and then loose energy through secondary effects (i.e. by further emitting/decaying*). The papers themselves are experimental papers that rule this out by measuring the total energy emitted in the decays (and are carried away by particles reacting with their calorimeters). I find no mention of a 'magical' energy reservoir in the nucleus or something like that in either of the papers.

* At least that's what I understand, the language is rather outdated. The second paper also appears to cite a concrete proposal by Meitner for such a secondary effect that I could not track down however. The first paper is even more agnostic about the nature of those effects.

 

[GregM]

I don't think this was the prevailing theory, or at least not the only one. I found Ellis, Wouter 1927 and Meitner, Orthmann (1930, in German). In their introduction they both describe the option that the electrons are emitted with a sharp energy and then loose energy through secondary effects (i.e. by further emitting/decaying*). The papers themselves are experimental papers that rule this out by measuring the total energy emitted in the decays (and are carried away by particles reacting with their calorimeters). I find no mention of a 'magical' energy reservoir in the nucleus or something like that in either of the papers.

* At least that's what I understand, the language is rather outdated. The second paper also appears to cite a concrete proposal by Meitner for such a secondary effect that I could not track down however. The first paper is even more agnostic about the nature of those effects.

As I said, the info we've got on that period of nuclear research seems incomplete. A review of google scholar papers for beta decay spectrum theory and experiments from 1911 to 1929 gives scant or obtuse results. I think it would be a good project for a science history scholar with access to a better archive to work on.

a 'magical' energy reservoir in the nucleus

? Nuclei were known to have energy states, and would emit gamma rays as they dropped from a higher to a lower. Nothing magical about that. Wasn't a radical stretch to suggest beta decay could do something similar to gamma decay, particularly since there was no other obvious explanation for the beta decay spectra. Only odd thing was the continuous aspect rather than usual quantized, perhaps that's what you meant by a magical nuclear energy reserve rather the the non-magical nuclear energy reserve used by gamma rays.

 

[mfb]

@mfb
>You can measure the energy spectrum of electrons and compare it to the prediction if there is an additional massless (or almost massless) particle emitted, and you get an excellent agreement.
No you can't. The energy and momentum of incoming neutrinos are unknown before the collision with the nucleus, therefore you can't make predictions. You can only watch and see the energy and momentum of the nulceus and electron after the collision, and then retrospectively deduce the energy and momentum of the incident neutrino.

The key word here is spectrum. You cannot make an accurate prediction for an individual decay, but you can predict the overall distribution, i.e. the energy spectrum. You compare that to predictions for an additional unmeasured outgoing particle and you find a great agreement. There is no other mechanism that would produce the same spectrum without absurd assumptions.

 

[Dr.AbeNikIanEdL]

Nuclei were known to have energy states, and would emit gamma rays as they dropped from a higher to a lower. Nothing magical about that. Wasn't a radical stretch to suggest beta decay could do something similar to gamma decay, particularly since there was no other obvious explanation for the beta decay spectra. Only odd thing was the continuous aspect rather than usual quantized, perhaps that's what you meant by a magical nuclear energy reserve rather the the non-magical nuclear energy reserve used by gamma rays.

It would be magical in that it would not affect nuclei either before or after the decay. Anyway, my point was that asking “How did they rule out an ‘additional energy reservoir’ to get to ‘new particle’?” might be misguided since apparently that’s not what they tried to rule out, at least not immediately before concluding there should be a new particle. In which paper did who suggest this additional energy?

 

[vela]

It takes a college student 60 seconds to decide if energy or momentum equations do not add-up. Why did it take the community of leading physicists 19 years to do the same with beta-decay?

I don't think it took physicists nineteen years to decide energy and momentum didn't add up. It took time to come up with a good explanation for why that seemed to be happening. Unlike today, physicists were reluctant to add a completely new particle to the mix just to explain beta decay. They instead looked for a more mundane explanation with the physics they already had on hand.

 

[Vanadium 50]

There are two things I don't like anout the premise. One is that it's pretty easy to critticize the people of the time for not going faster. After all, it's the very next page in the textbook! why so long between Maxwell and Einstein, or Einstein SR and Eistein GR, or Dalton and Mendeleev, or.... (Why did it take the detective so long to figure that the butler did it anyway?) Maybe we're just smarter than they were,

But I doubt it.

Possibly related is that during the thumb-twiddling period, there was a major war going on followed by the collapse of four major empires. And after that a major world plague. Is it possible that people were also a little distracted.

 

[hutchphd]

I also understand that those Neutrinos and their antimatter cousins a reticent to appear in public either coming or going. Although I feel vaguely unclean whenever I do it I am compelled to quote Donald Rumsfeld :

Reports that say that something hasn't happened are always interesting to me, because as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns—the ones we don't know we don't know. And if one looks throughout the history of our country and other free countries, it is the latter category that tends to be the difficult ones.[1]

The OP seems to not recognize this.

 

[GregM]

It would be magical in that it would not affect nuclei either before or after the decay.

Of course. Same if that had been the case with other nuclear rays. An internal nuclear energy reserve for gamma rays was hypothesized: it was found. An internal nuclear energy reserve for alpha rays was hypothesized: it was found.

In which paper did who suggest this additional energy?
I don't know. In my attempts to research this subject I've only found a cloud of speculation and vague references from commentators, while actual papers with beta decay energy source hypotheses are sparse at best.
Most vague references are from historians and commentators on Pauli's 1930 neutrino hypothesis, who say things like 'during the 1920s every possible source of beta decay energy was investigated and not found, therefore theorists were forced to reluctantly take the unprecedented step of conjecturing an effectively undetectable particle'
So far the only concrete failed hypothesis we've got is the one you've referenced: the continuous spectra comes from the emitted electron's post-emission interaction with its environment ( best candidate were the atomic electrons ).

 

[GregM]

@mfb @hutchphd @Vanadium 50 @vela
Thanks for your comments. I take it that you, like me, haven't found a list of papers from 1911~1929AD on hypotheses for beta decay energy sources.

 

[Vanadium 50]

Haven't even looked.

 

[vanhees71]

I've just googled, because indeed there seems to be little about the history of $\beta$ decay from the nuclear-physics perspective, while you find a lot from the HEP point of view. From the theory side that's of course mostly starting with the neutrino hypothesis and Fermi's subesequent development of a quantum field theory of the weak interaction, followed by work on parity violation in the mid 50ies (Lee, Yang, Gell-Mann, Feynman,...) + experiments (Wu, Ledermann,...) and finally the development of quantum flavor dynamics as part of the Standard Model (Ward, Glashow, Salam, Weinberg, Higgs,...).

Google came up with

C. Jensen, Controversy and Consensus: Nuclear beta decay 1911-1934, Springer (2000)

That seems to be a pretty comprehensive study on the early history.

Interestingly there's also a book chapter written by C. S. Wu in 1959:

C. S. Wu, History of beta decay in: O. R. Firsch et al (ed), Beiträge zur Physik und Chemie des 20. Jahrhunderts, Lise Meitner, Otto Hahn und Max von Laue zum 80. Geburtstag, Springer (1959)

Despite the German title of the book the contribution by Wu is in English.

 

[GregM]

C. Jensen, Controversy and Consensus: Nuclear beta decay 1911-1934, Springer (2000)

wow, nice find. I want to read that book. The secrets of 1920s of beta decay research are inside.
$66 before tax though.

 

[malawi_glenn]

I want to read that book. The secrets of 1920s of beta decay research are inside.
$66 before tax though.

There is a thing called a library ;)

Springer also sometimes (like 3-4 times / year) have 50% discount offers, I suggest that you register an account.

 

[Vanadium 50]

There is a thing called a library

That's just crazy talk!

 

[Vanadium 50]

And while I am kvetching, looking at the title - "resort to neutrinos"? That is the correct answer.

Let me make two other points:
(1) In this period, QM was still being developed. There was no theory, even a bad one, of the weak interaction. Expecting them to have sorted out the weak interaction decades before sorting out QFT seems a bit...ungenerous. Of course, we can still complain that they didn't write down QM on Wednesday, QFT on Thursday and then took a long weekend.
(2) The beta spectrum does not look like phase space,. It's more peaked at the high side than that (because of the W propagator). If someone were to have compared what was observed to a naive 3-body phase space, it wouldn't have matched.

 

[malawi_glenn]

The beta spectrum does not look like phase space,. It's more peaked at the high side than that (because of the W propagator). If someone were to have compared what was observed to a naive 3-body phase space, it wouldn't have matched

It matched good enough in the 1930's at least to call Fermi's theory a "success". Sure, there was some details left, which was "solved" a couple of decades later.

 

[Vanadium 50]

Yes, but Fermi's theory was not just three body phase space: it added an energy dependence in the Ferm constanr (which we know now is a recast mass of the W boson),

 

[ohwilleke]

Neutrinos are detectable for sure, because they interact.

They are now. They weren't back then because they didn't have the instrumentation to do it.

Even now, the percentage of neutrinos produced by a process that are actually affirmatively detected in a neutrino detector is very far from 100%.

 

[malawi_glenn]

Even now, the percentage of neutrinos produced by a process that are actually affirmatively detected in a neutrino detector is very far from 100%.

It's even very far from 1% :) hehe

 

[vanhees71]

And while I am kvetching, looking at the title - "resort to neutrinos"? That is the correct answer.

Let me make two other points:
(1) In this period, QM was still being developed. There was no theory, even a bad one, of the weak interaction. Expecting them to have sorted out the weak interaction decades before sorting out QFT seems a bit...ungenerous. Of course, we can still complain that they didn't write down QM on Wednesday, QFT on Thursday and then took a long weekend.

It was not even known that there are weak interactions at all. They had some idea that there are electrons and protons within the nucleus. The neutron discovery in 1932 resolved this.

(2) The beta spectrum does not look like phase space,. It's more peaked at the high side than that (because of the W propagator). If someone were to have compared what was observed to a naive 3-body phase space, it wouldn't have matched.

It looks like from a 3-body decay, which is why Pauli's letter to the "radioactive ladies and gentlemen" was not a wild guess!

 

[Vanadium 50]

It looks like from a 3-body decay

But not 3-body phase space. You need the $q^5/M_W^4$ to match data,

 

[Vanadium 50]

It matched good enough in the 1930's at least to call Fermi's theory a "success".

Fermi's theory did not predict a phase-space like spectrum (see above), Indeed, one of the successes of the theory was that it got the spectrum right.

 

[vanhees71]

But not 3-body phase space. You need the $q^5/M_W^4$ to match data,

Sure, but that was explained theoretically only later by Fermi's weak-interaction model, using Pauli's hypothesis about the particle we call neutrino today (in 1930 Pauli called it neutron, but then they named the neutron neutron, which was discovered in 1932; so finally the name neutrino was given to Pauli's hypothetical particle).