What does random mean wrt nuclear decay?

[mesogen]

What does "random" mean wrt nuclear decay?

From what I understand, the process of nuclear decay proceeds at a very predictable rate. Given a lump of say, U-235, half of all the nuclei in the lump will have decayed after 700 my.

There is no way, though, to determine which nuclei in the lump will decay at any specific time, so this is considered random.

But what actually triggers the decay?

Apparently, the only reason the process is deemed random is because we don't have a way of probing the nucleus to look for some signal that will tell us that the nucleus is about to decay.

If we were able to isolate and monitor a single U-235 atom, and somehow magically probe the nucleus to see all the inner goings-on, then what chain of events do you think we would look for to be able to say that the nucleus was on the verge of decay?

I guess what I'm asking is: "Is the decay of a nucleus called 'random' simply because we are incapable of determining when it will decay, or is it called 'random' because there is no actual cause to the effect?"

 

[DecayProduct]

I'm no particle physicist, but nuclei decay because of the weak nuclear force. An exchange of particles takes place within a nuclei to cause a neutron to decay into a proton, and result in beta emission. The mass of the exchanged particles is very large, so they are very short lived. When they are generated cannot be known due to uncertainty, and so quantum mechanically, it is a random event. There is a lot more to it, but I'm not even qualified to make this simple post, let alne get into the guts of it all.

And even if we could go about probing it, seems like the very act of probing changes the overall out come, so again, we can't look in on it in progress.

 

[mgb_phys]

I guess what I'm asking is: "Is the decay of a nucleus called 'random' simply because we are incapable of determining when it will decay, or is it called 'random' because there is no actual cause to the effect?"

Thats the $64,000 question.
Quantum theory says it's totally random, many people eg Einstein refused to accept this, hence the famous 'God doesn't play dice' quote.
Logically you can always say there is some hidden mechanism we can't yet detect - simply because it is impossible to prove a negative - but most physicists would believe that the decay is random and QM is just a different stranger world to our everyday experience.

 

[Tac-Tics]

Is the decay of a nucleus called 'random' simply because we are incapable of determining when it will decay, or is it called 'random' because there is no actual cause to the effect?

It's random because it's quantum mechanical in nature. The inner workings are, in principle, undiscoverable. Whatever the underlying cause is, it is in violation of the uncertainty principle to determine the outcome without influencing it.

 

[QuantumPion]

Nuclear decay is not "chaotic, but too difficult to model", like the weather. It's actually random, because the decay is based on quantum mechanical effects. Whether the alpha particle manages to quantum tunnel out of the nucleus is not deterministic.

 

[mesogen]

And even if we could go about probing it, seems like the very act of probing changes the overall out come, so again, we can't look in on it in progress.

Yes, but I was hoping for a magical probing. :)

It's random because it's quantum mechanical in nature.

I guess I'm like Einstein and the thought of a causeless event just doesn't sit right with me.

Nuclear decay is not "chaotic, but too difficult to model", like the weather. It's actually random, because the decay is based on quantum mechanical effects. Whether the alpha particle manages to quantum tunnel out of the nucleus is not deterministic.

I guess I didn't realize that tunneling was considered a random event. And, I guess I didn't think of nuclear decay as a tunneling phenomenon.

I'm loosely acquainted with electron tunneling, as in describing conduction through (around?) a dielectric molecule. People will cover electrodes with self-assembled monolayers of something like hexane and then attach redox-active molecules to the hexane molecules, and then do electrochemistry. The hexane layer is insulating, but it's only a molecule thick, so it still conducts through electron tunneling. The conductivity is measurable and repeatable, so this raises a question that also applies to nuclear decay.

If each individual event is totally random, not just stochastic and therefore difficult to predict, how can they statistically add up to a predictable rate?

 

[mgb_phys]

If each individual event is totally random, not just stochastic and therefore difficult to predict, how can they statistically add up to a predictable rate?

The probability of tunneling is predicatable, it depends on macrcopic quantities such as energy levels etc.
So the overall rate of decay of a large sample is stochastic, it's just which atom rolls a six next that you can't determine.

 

[mesogen]

The probability of tunneling is predicatable, it depends on macrcopic quantities such as energy levels etc.
So the overall rate of decay of a large sample is stochastic, it's just which atom rolls a six next that you can't determine.

That's kind of what I'm asking. How is this possible?

Each individual event is totally random, like for real. But, when you put them in a big group, the rate is predictable.

This makes no sense to me.

 

[mgb_phys]

Imagine shaking a big box of dice, what each dice (or die if you are being pedantic) will show is completely random - but the average number showing a 1,2...6 etc if you throw a box of 1000 dice is predicatable.

similarly if you decided all the 6s had decayed and removed them, then throw all the dice again - and remove the new 6s, you will get exactly the same decay curve as with atoms

 

[Tac-Tics]

I guess I'm like Einstein and the thought of a causeless event just doesn't sit right with me.

Causelessness dosen't sit right because it's nonsense. If two systems are identical, their evolutions will be identical because that's what identical means. But physicists use "identical" to mean something much less strict... "identical as far as I can tell" or "identical as far as theory would suggest possible to arrange."

The limitations imposed by quantum mechanics says nothing about the lack of an internal mechanism. It simply states that tests of a mechanism is beyond the scope of science. It enters the realm of philosophy, and suddenly, you lost the attention of the physicists. Thus, almost everyone just say "it's totally random and weird" and gets on with their lives.

If each individual event is totally random, not just stochastic and therefore difficult to predict, how can they statistically add up to a predictable rate?

I'm not sure on the specifics of the problem you're working on, but orderly patterns can easily emerge from random data. The central limit theorem, where a bell curve appears from the average of sufficiently many perfectly noisy inputs, is a perfect example of this kind of behavior.

 

[mesogen]

Imagine shaking a big box of dice, what each dice (or die if you are being pedantic) will show is completely random - but the average number showing a 1,2...6 etc if you throw a box of 1000 dice is predictable.

similarly if you decided all the 6s had decayed and removed them, then throw all the dice again - and remove the new 6s, you will get exactly the same decay curve as with atoms

Is it simply because there is a roughly a 1/6 chance there will be a 6 rolled, so if you have enough dice in one place roughly 1/6 of them will turn out to be a 6?

But it is possible that no 6s at all will turn up. And just as likely that all 6s will turn up. Both outcomes are just as likely (not very).

And as you begin to have fewer and fewer dice, the predictability goes way down.

Is this also the case with nuclear decay?

Is there a small but real chance that all the nuclei will decay at once? Or more or less than the probability dictates?

Also, if you have a small sample, say a hundred atoms, does the predictable rate go out the window?

 

[mgb_phys]

Is this also the case with nuclear decay?

Yes, half-life is only an average quantity for a large collection of atoms, once you get down to a handfull of atoms the decay rate gets noisy.

 

[Tac-Tics]

Is there a small but real chance that all the nuclei will decay at once? Or more or less than the probability dictates?

Anything's possible, but most things are astronomically unlikely.

 

[DecayProduct]

I don't think it is so much that there is no cause, but that we just can never witness it in action. The very fact that the event occurs is reason enough to know that there was a cause for it. We can infer it after the fact, but never watch it in progress.

Given enough random events, they will settle themselves into a range of predictability. To me, that is more puzzling than any quantum effect. Like dropping marbles over a peg board and watching the bell curve appear. That is just astonishing. Given enough radioactive atoms, there is some non-zero chance that they will all decay at the same time. There is also some non-zero chance that none of them will ever decay. I don't really know, but my gut feeling is that I'd expect it to really be just about in the center of these two extremes.

I also tend to side with Einstein about QM and randomness. After seeing that QM made real testable predictions, he felt that while it may hold some water, it was still just the cover for a deeper reality, where things began to make "sense" again.

 

[hamster143]

The question you're asking is invalid. Your core assumption is that there's a unique "you", in other words, that your future is unique and predetermined, and that is not in agreement with modern understanding of physics.

Imagine a tree. Consider the set of tips of all branches on the tree, and consider its evolution. As the tree grows, branches sometimes split, and their tips multiply. What began as a unique point (the tip of the initial shoot) ends up a set whose members have many distinct histories.

If you identify yourself with the tip of the initial shoot and follow its growth, at every point your history is unique, but your future is uncertain and has many outcomes. The question "which branch corresponds to the original tip" can't be answered, just as the question "when will this atom of uranium decay".

In physical terms, if we start with the universe where this atom of uranium is still present and evolve it for a year, it will become a superposition of many different universes with different recorded decay times. In each of those universes there will be a person wondering "why did this atom decay at that specific moment?".

You could say that the history of decays you observed is sort of an "address" of your physical state in the multiverse.

 

[ZacharyFino]

the superpositions are only probability densities, as far as what is observable to us (exceptions) the atom actually decays at one actual time (measuring it is another thing) and the multiple addressees are more like probability blueprints than a "physical" state in the universe.

 

[Tac-Tics]

The question you're asking is invalid. Your core assumption is that there's a unique "you", in other words, that your future is unique and predetermined, and that is not in agreement with modern understanding of physics.

This sounds like the Many Worlds Interpretation. I never saw the appeal of it. It seems like an interpretation for people with little mathematical imagination. "The universe isn't random, it just does both at once! and then... mumble mumble something about wave collapse..." It's just a tool for physicists to sweep the philosophy under the rug.

 

[hamster143]

This sounds like the Many Worlds Interpretation. I never saw the appeal of it. It seems like an interpretation for people with little mathematical imagination. "The universe isn't random, it just does both at once! and then... mumble mumble something about wave collapse..." It's just a tool for physicists to sweep the philosophy under the rug.

There's no need to worry about wave function collapse. You measured something to be X, therefore your brain is entangled with the state in which the observable is equal to X. Quantum physics teaches us that uranium nucleus evolves into the state, which is a superposition of decays at times t=t1, t=t2 ... etc. Therefore, the outcome of "the experiment" is a superposition of states "an observer who saw the decay at t=t1", "an observer who saw the decay at t=t2", etc. And none of these states/observers would interact with each other or even be aware of each other, because quantum physics as we know it is perfectly linear.

The fact that an observer lacks an unique, deterministic future is a deep philosophical insight.

 

[jtbell]

I guess what I'm asking is: "Is the decay of a nucleus called 'random' simply because we are incapable of determining when it will decay, or is it called 'random' because there is no actual cause to the effect?"

There is no generally accepted answer to this question. The standard mathematical formulation of QM simply does not address the question, "when, exactly, will this particular nucleus decay?" It only allows us to calculate probabilities.

Proposed answers fall in the realm of interpretations of QM, about which there seems to be little general agreement among physicists who study such issues seriously. Look in the Quantum Physics forum here, and you'll find that probably half the threads turn into arguments about Copenhagen versus Bohm versus many-worlds, etc., or about Bell's Theorem and the significance of experiments that test it.