This headline is deeply misleading. It is true that energy states are quantized. It does NOT follow that the transition between them is "instantaneous". The transition has always been known to be gradual, but the "gradualness" is not a smooth transition in the energy of the system, it's a smooth transition between being in one energy state to being in a superposition of two energy states to being entirely in the second energy state. That process plays out over (typically a very short but nonetheless non-zero) time. This has been known since the very beginning of QM.
What's news here is that this transition, which has always been predicted by theory, has been experimentally observed for the first time.
I have questions: you describe the change of states of particle as
"a smooth transition between being in one energy state to
being in a superposition of two energy states to being
entirely in the second energy state."
At the terminus of this transition, is the particle no longer in a superposition? IE there is now a 100% chance it's in the second state and 0% chance that it's in the original state?
If so, does that imply that the function of the particle's state (with respect to time) is discontinuous? Since there's a point at which it goes from being a superposition to exactly 0%.
The first answer is that a system is never 100% in any particular state because that would violate the uncertainty principle. When we speak of a system "definitely" being in a particular state that's an approximation/simplification. It actually means that the system is in a superposition of some range of states that for all intents and purposes we can treat as being the same, and the probability of it being in a state that cannot be treated the same for all intents and purposes is so close to zero we can ignore that.
The second answer is that whether or not a system is in a superposition depends on your point of view. A system can only ever be in a "single state" (according to the above approximation) with respect to some observable, and if it is in a single state with respect to that observable then it is necessarily in a superposition with respect to the complementary observable, e.g. a particle that is in a definite state with respect to position is necessarily in a superposition with respect to velocity.
So the whole process is everywhere and always continuous.
I was extremely confused by the proposition because yes, it's known that energy states are quantized but I've never heard of any popular literature before suggesting the transition itself was quantized. Intuitively it doesn't even make sense that it could be quantized.
I can't figure out what the antecedent of "this" is. But like I said, the fact that quantum jumps are not instantaneous is not and has never been controversial AFAIK.
UPDATE: It turns out that in the VERY early days of QM, some researchers (notably Niels Bohr) thought that quantum jumps were instantaneous. But it has been known and universally accepted that they are not for decades now.
While not a perfect analogy, if I understand this properly, essentially if electrons jumping states were pitching in an unregulated game of baseball, the chances a pitcher will toss the ball increases as time goes on and previously we thought the ball just appeared over the plate instantaneously when that random time occurred.
But instead, while you still don't know when exactly, you do know there will be a wind up for the pitch before the ball is thrown, and we can see this windup, and we can time things to hit the ball, or shoot the pitcher with a blowdart and keep the pitch from happening.
So if there are systems in certain configurations to which this property is applicable to, and if we needed to keep things in that configuration until a time of our choosing, for instance if a state change occurs it would have cascading effects, we can do so.
There doesn’t have to be one. The transitions are random and therefore not influenced by the recency of the last transition, so there’s no need for a ‘memory’.
Having said that, the random chance of a transition occurring must be a probability over time. So if there is a fundamental probability, we can ask what is the unit of time over which that probability is expressed? I’m not sure that question makes sense, but maybe it’s the time it takes for the transition to occur, if that is fixed for a given transition?
My question was about the probability governing the chance at a jump. Apparently, it is not constant and takes t as an input. (Otherwise how does it know that it is rising to 1)
Ok, I see, that makes sense. One thing I’d like to know is, does the frequency of transitions correlate to the time it takes for the transition to occur.
I’m not sure. There must be some factor that causes one quantum transition to occur more or less frequently than another. That factor isn’t really hidden because we can measure the frequency of transitions, but as far as I know, we don’t understand it’s nature. That factor, whatever it is, may determine the time it takes for the transition to occur but it’s not necessarily like a memory, but more like a trajectory. Clearly something in the quantum state is changing, so it is a system of some sort.
I think you're asking two somewhat entwined but not-same questions: Consider the case of a mushroom that is actually mycelium and fungi network underground. The actual mushroom top that has spores is at the very very end of the lifecycle. The mushroom fungal network of mycelium actually spans months, years, or even decades before the final "fruiting" part of the mycelium takes place. It's likely that these quantum leaps are much like the fruiting event of a long-time-growing networked-body of sorts. In which case, frequency of fruiting and time it takes to occur would be different for different types of mushroom, or different types of plants, and therefore it's likely that there are probably different "types" of quantic networks that are direct precursors to quantum jerks that only appear instantaneous when your equipment is frame-rated. This, of course, only incurs more questions, like what the heck is this pre-leap quantum body like, and how can back can we actually trace a quantum event, and if everything is broken down into quantum events how can we meaningfully measure development and duration of something seemingly induratable? It'd be like trying to trace back fully grown corn to the seed. Of course, it's likely covered in kernels that will inspire and directly create other corn-phenomena, but the original kernel cracking open is information long gone. Quantum entanglement therefore actually means you're inextricably linking two quantum streams, not quantum events. Using that sort of process, it may be possible to find out where quanta are "produced" if I may use such coarse terminology that harkens back to the natural plowshare of the most simple kind of field.
“God does not play dice with the universe; He plays an ineffable game of His own devising, which might be compared, from the perspective of any of the other players [i.e. everybody], to being involved in an obscure and complex variant of poker in a pitch-dark room, with blank cards, for infinite stakes, with a Dealer who won't tell you the rules, and who smiles all the time.”
— Neil Gaiman and Terry Pratchett, Good Omens: The Nice and Accurate Prophecies of Agnes Nutter, Witch.
As a quantum physicist: This might be an interesting experiment, but as far as the theory goes, this is exactly what is to be expected from standard quantum mechanics.
Quantum leaps being instantaneous would be a (possibly common) misconception.
The leap in a quantum leap is describing the notion of a discrete jump in measurement outcomes.
Okay, so you're a quantum physicist and you can answer my question!
My understanding of QM is that a quantum's system's state is suddenly and discontinuously changed by a measurement.
My understanding of this article is a bit confused, and I think that there are two possible things we could be seeing:
1. Quantum collapses of superpositions actually do take time ("This Changes Everything")
2. This particular quantum system is not actually being measured, but is oscillating in superposition in some odd way. ("Just a Particular System")
What's done in such an experiment is that we initialize a quantum state in a particular state (that this is possible is not actually obvious, but let's assume it's true) and then we make a lot of repeated measurements after different amounts of times.
So we're not talking about a single system, but instead about statistics about a set of measurements with systems that have been set up in the same initial state. (Each system only being measured once after a certain time after it's being set up).
The system is being measured, collapses DO take time, but this is not news to people who study certain branches of quantum information. The experiment confirming it though is a breakthrough.
I guess I’m kind of confused on what the discovery is; we have a quantum system evolving unitarily from one state at t_0 to another state at t, and the probability of measuring the system in one of two discrete states changes continuously as well (despite the actual measurement outcome being discrete). I thought this has been known for a long time, implied by the time dependent Schrodinger equation, so I didn’t quite catch from the Quanta article what is new, mathematically. Can anyone clarify?
> I didn’t quite catch from the Quanta article what is new, mathematically
Nothing is new. This experiment is simply confirming a prediction of standard QM. All the talk about "quantum jumps" and how something has supposedly changed about the way we understand them is just pop science reporters misunderstanding the actual science.
If you're a lay person like me and you want to make sense of this, read "Einstein’s Unfinished Revolution: The Search for What Lies Beyond the Quantum" by Lee Smolin. It was just published recently, and it anticipates this result!
A continuous jump means that the system travels coherently through the superposition of |0>+|1> through the jump, and other such a|0>+b|1> states for a^2+b^2=1. It isn't in 0, and then suddenly 1.
It’s a process of transition between two distinct states. Say a bit flips from 0 to 1; there is a short time period during which that bit is physically “in transition”, but it would be incorrect to say that the bit has a value of, say, 0.5 during the transition.
Also remember that quantum mechanics is a way to make statistical predictions of outcomes of certain experiments; it does not claim to explain what is actually happening underneath.
I strongly disagree with your second paragraph. Quantum theories do, we think, correspond to what is actually happening underneath. They might not be perfect correspondences, in that our theories are incomplete or approximate, in the same way that Newtonian gravity is an approximation of General Relativity in the weak-gravity regime.
Bell's inequality places a strong constraint on what sort of physical theories can explain quantum phenomena. If you believe in locality (the universe has no global variables, and information propagates outwards through local interactions), then the wavefunction is a real thing, and the actual state of the universe.
The formalisms of quantum mechanics are widely accepted, but as you know, there is considerable disagreement about what may actually be happening underneath:
Yes, there is disagreement about what is happening underneath, but whatever your interpretation, it is highly constrained by Bell's theorem. Quantum mechanics has to be more than just a statistical description of some underlying deterministic process, unless you are willing to throw out locality.
I'm highly partial to the Many Worlds Interpretation, as I think it's the only interpretation that takes quantum mechanics seriously. If you assume that quantum mechanics describes both the system you're studying and the measurement apparatus (including the scientist taking the measurements, and the rest of the universe), then you're led inexorably to the Many Worlds Interpretation.
So this is what I do not understand about Bell's Theorem: it seems to dismiss in the assumptions the possibility of determinism (I don't understand where the super comes from, as there is no distinction) and then goes on to conclude "Hey, there must be fundamental randomness". Didn't we just assume the conclusion?
It does not assume from the outset that there is no determinism. It sets bounds on the types of results that you can get with deterministic local variables. Those bounds are violated in the real world. Hence, the universe is either quantum mechanical, or there exist hidden nonlocal variables that give the illusion of quantum behavior.
There is a way to escape the inference of superluminal speeds and spooky action at a distance. But it involves absolute determinism in the universe, the complete absence of free will. Suppose the world is super-deterministic, with not just inanimate nature running on behind-the-scenes clockwork, but with our behavior, including our belief that we are free to choose to do one experiment rather than another, absolutely predetermined, including the "decision" by the experimenter to carry out one set of measurements rather than another, the difficulty disappears. There is no need for a faster than light signal to tell particle A what measurement has been carried out on particle B, because the universe, including particle A, already "knows" what that measurement, and its outcome, will be.
I think it’s useful to be clear about the dividing line between demonstrated scientific results and our scientific intuitions, even when those intuitions are driven by observed patterns that have been reliable in the past. Intuitions are excellent drivers for formulating new theories and experiments, but it is epistemically dangerous to conflate beliefs and knowledge in our minds.
> Suppose the world is super-deterministic, with not just inanimate nature running on behind-the-scenes clockwork, but with our behavior, including our belief that we are free to choose to do one experiment rather than another, absolutely predetermined, including the "decision" by the experimenter to carry out one set of measurements rather than another, the difficulty disappears.
You see, I think Bell is kind of obfuscating here: this is just normal determinism and it seems to me like Lagrange would have been perfectly fine with this. Causality only, even for the atoms that happen to reside in a human brain.
Edit: what do you mean with your last paragraph? Because I interpret it like this: the belief that interferes is that we have free will and can choose and basically change the past (counterfactuals). But I have a hunch that you mean it like: QM is weird, be careful.
My understanding is that the “super” in superdeterminism is just making clear that we are considering the experimenter completely determined as well, but yes, they are essentially the same thing.
Last paragraph I mean what you interpret — our everyday experience of “free will” may blind us to certain possibilities that we have no strong evidence for or against, just as our everyday experience of a classical world makes QM seem “weird” and unintuitive, even when the experimental evidence is well-established.
There’s a particular curiosity when the possibility of a world without free will calls into question the extent of the power of science itself. Are there other techniques that can provide satisfying arguments about the potential nature of reality in a world where we cannot rely on the power of experiment to reveal this nature?
I think it does call into question the power of science, but also: assuming non-determinism does not give us this power back! If materials (atoms / humans) can arbitrarily disregard the laws of nature by making choices (however that would be implemented), what does an experiment even mean?
Edit:
This is a relevant theorem, that nicely complements Bell:
"it would be incorrect to say that the bit has a value of, say, 0.5 during the transition."
In what sense? It has a <z>=0, meaning there's a statistically 50%/50% chance of measuring 1 or 0 but the system is in principle coherent.
This means that half way through the jump, if you rotated it 90 degrees, you'd measure a _definite_ z value of -1 or 1, or you'd measure a bit value of 0 or 1, depending on whether you twisted it clockwise or counterclockwise.
In the sense that it ignores that coherence. Any state on the disc of the Bloch sphere's equator has <z> = 0, and 50:50 chance of measuring 1 or 0. If you want to assign a real value of 0.5 to a bit, then the only one that makes sense is the center, completely decoherent mixture.
With your second paragraph firmly in the back of my mind:
In the sister comment thread, two photon excitation seems to indicate that this intermediate state (0.5 in your coin example) is not invalid. Can you help me understand better?
Edit: Also, I think your coin "in transition" implicitly assumes that the system has more degrees of freedom than 1 bit (You can only flip a coin in 3 dimensions).
This is a common misconception. Quantum jump are jumps between two stable levels... there's nothing preventing "jumps" to unstable states, the particle just doesn't stay long in that specific state (although it can stay there long enough to be useful as in the case of two-photon absorption [0] and microscopy [1]).
If I understand it correctly, what this seems to imply is that it's not so much a (discontinuous) jump but rather a continuous(-ish) transition from one state to another.
Tangentially related request: I’m frequently very impressed with how many HN users seem to have a pretty solid grasp of physics (or maybe the right term is quantum mechanics?) and I’d love to be able to follow along but my science education more or else ended with high school. Could anyone recommend a good resource for someone with zero knowledge of this space to get very basic foundation?
It's not easy and you'd need to find problems to work, but http://www.feynmanlectures.caltech.edu is one source some people swear by. Others say it's too hard for a beginner text. I think both have a point, but it's free to see for yourself.
It's about gravity, not quantum field theory, but it's still very sound, very intuitive and explains the principles of the LIGO gravity wave detector. Also contains a bit of speculation about the culture and psychology of scientific creativity which I thought was great.
If you find a similar book on quantum phenomena, please post it!
Understanding Physics by Isaac Asimov is a great book that discusses Newtonian mechanics, thermodynamics, electromagnetism, and atomic physics. It presents it by discussing the history of the discoveries and experiments that advanced our understanding of the physics. The book is from 1966, so there have been numerous advances since, but it does go into a lot of detail of subatomic particles and radioactive decay. It is written to be understood by lay people and doesn't go very deep into the mathematics, which helps make it easier to understand.
The more I read about quantum mechanics the less i understand, and I'm absolutely unable to get into a proper learning path because it requires mathematics beyond my level and for which I'm not able to develop a taste on my own.
Of course I'm not interested in doing calculations but to appreciate quantum physics you have to know what the formalism behind are about and physicists are unable to explain it in simple terms for reasons I think I make out but can't properly formulate.
As an alternative path, Quantum Models of Cognition and Decision [1], may offer a less steep learning curve for the fact "you are the quantum system" and as such get to have actual experience with phenomena discussed in this book. To clear up the new-age vibe introduced in the last sentence, I think studying the maths through a phenomenon whose ambiguity is not questioned as a metaphysical abyss but is accepted as just being here in its mundane simplicity (semantic ambiguity in daily language use, that kind of thing) alleviates a lot of trouble in grasping what the maths mean in a physics course. Also the book is written for people coming from the fields related to psychology so it's a lot more approachable.
The math behind Quantum Mechanics is surprisingly simple, mostly just the linear algebra you learned in High School with some fun Greek symbols thrown in. (I'm convinced mathematicians just can't help themselves with tossing in more Greek letters just so the papers look more impressive)
In any event check out "The Mathematics of Quantum Mechanics" by Martin Laforest [1]. Free PDF online, totally readable and easy to follow with typical High School math background.
From that link: “But what about when we want to describe physical quantities that have continuous values, such as the position of a particle along a line? In this case, we need a vector space of infinite and continuous dimension. Turns out that it’s possible to define a Hilbert space on the set of continuous functions, e.g., f (x) = x3 + 2x + 1. This is referred to as “wave mechanics” and we won’t cover it in this book.”
I just finished "Now: The Physics of Time" by Robert Muller. I have mixed feelings about the book, especially the chapters devoted to the author's interpretation of philosophy. I did enjoy hearing a history/overview of modern physics from someone in the field though, and it was a very approachable book. He was very clear about the open questions in quantum mechanics instead of hand-waving them away, which I appreciated.
I'm sure other books like Stephen Hawking's "A Brief history of Time" would be a good starting place too, but I can't speak to that one personally yet.
I wrote a comment one level above where I recommended "Einstein’s Unfinished Revolution: The Search for What Lies Beyond the Quantum" by Lee Smolin, and I thought I'd toss that in the ring here.
MIT OCW used to host their undergrad introduction to Physics lectures by Prof Walter Lewin which are pretty great. It was pulled down in 2014 but you'd definitely be able to find copies on Youtube.
"Natura Facit Saltus or No" was something Schrodinger wrote in 1952! (does nature jump or no?) Of course it's a continuous process, how else would water flow? As they mention in the article, here's the link to Schrodinger's paper "ARE THERE QUANTUM LEAPS?" https://academic.oup.com/bjps/article-abstract/III/11/233/14...
> "The strategy reveals that quantum measurement is not about the physical perturbation induced by the probe but about what you know (and what you leave unknown) as a result. “Absence of an event can bring as much information as its presence,” said Devoret."
If a quantum system is going to transition between two states that have different energy, would we not expect it to take a time specified by the Heisenberg Uncertainty Principle? If the change in energy is delta E, would we not expect the transition to take delta T = h bar/delta E?
> would we not expect it to take a time specified by the Heisenberg Uncertainty Principle?
There isn't actually an energy-time version of the uncertainty principle, at least not the simple one you're assuming here, although many pop science presentations talk as if there is. A good article discussing this is here:
For a quantum system transitioning between states, the probability of transition in general will vary as a function of time; how it varies depends on the specific state of the system. There is no general rule that relates the expected transition time to the change in energy. (Note also that not all transitions are between energy eigenstates.)
No, the time to complete one transition is the inverse of Rabi frequency, which is actually determined by strength of external field and nature of the transition (its dipole moment matrix elements), see e.g. https://en.wikipedia.org/wiki/Rabi_frequency
If I am understanding the article correctly, there is no hidden variable implied. Quantum physics claims changes are random, with nothing precise and determinate underneath that causes this.
Think of the example of flipping a coin. It's random, but for each flip if you had exact data for the movement of the coin as it left your hand, you would presumably be able to predict if it would land heads or tails. Quantum physics claims that for events at the quantum level, there is nothing definite underneath that determines what will happen, it is just fundamental randomness. The hidden variable idea is that there is something definite underneath, we just haven't discovered it yet.
What the experiment seems to have found is only that the probability of an event occurring changes smoothly over time from 0 to 1, not that there is some underlying exact cause for what the probability is at a given point in time.
They detected a quiet period of slower transitions before the spontaneous drop to the middle state... this implies that "something" happens prior to the transition.
They were able to detect this "something", and even prevent the state change.
In my mind, that "something" represents a hidden state or something like it within the synthetic atom.
The trouble is there are some strong arguments against that idea, so quantum physicists tend to assume that such "somethings" just don't exist. Yes, that is very counter-intuitive, but that is one reason quantum physics is so weird.
Amazing. What is the energy of the system while it's in transition? I guess they're arguing superposition of 1 and 0, continuously sliding to a higher likelihood with time. I mean, eventually a photon is emitted, and that can't be continuous. I guess I'll have to read the preprint.
By which frame of reference? I never understood the language around "instantaneous". Isn't simultaneity relative? So that where one frame of reference says two events are simultaneous, there or others that say they are not?
The old quantum theory was developed in non-relativistic setting, so this was not a concern. But you are right, relativity complicates lots of things in quantum theory, including the idea of "instantaneous" quantum jumps. In relativity, if some event is to be universally instantaneous, then it has to happen at a single point of space. Which is possible with point particles, but then you get the problem how those point particles can find each other to interact at a single point so often as measured cross sections indicate... perhaps they are not exactly points, but waves, but then we can't have instantaneous events, the event has to happen to the wave in big region of space where simultaneity is relative.
Simultaneity is only relative when events A and B are far enough apart that it would have been impossible for light to get from one to the other between the two events. When there is a single location, it's objective whether something has zero or nonzero duration.
And while observers disagree on their personal measurements of duration, they will always agree about what a clock sitting at the location will measure.
“Another text saying that the founding fathers of quantum mechanics were not only wrong but idiots, as some current geniuses revealed. In reality, it's the other way around, of course. [...] I don't think it makes sense for me to discuss the paper and Ball's summary more deeply. One would have to correct every sentence that is wrong or at least misleading – which is basically every sentence both in Ball's text as well as the text in Nature.“
Does this imply anything about the Many Worlds Interpretation [1]? I am not a physicist (IANAP?), but if states change continuously vs discretely... should that not discount MWI?
well, obviously. This was always pretty clear to anyone who put some serious thought into it. Did anyone really think electrons just magically jump from state to state? There is a real physical process there that takes time.
>well, obviously. This was always pretty clear to anyone who put some serious thought into it. Did anyone really think electrons just magically jump from state to state?
Only prominent QM scientists. Amateur pundits always knew that they don't magically jump.
There is a famous story about a guy trying to convince Niels Bohr in 60's that quantum transitions in different molecules close to each other can coherently influence each other and result in amplified emission of coherent radiation. Bohr utters something in the sense that is impossible, quantum transitions are random, photons are not correlated, so the molecules can never cooperate. The guy didn't take the prominent QM scientist too seriously, went on and built the device. He was the famous Charles Townes, one of discoverers of the maser/laser effect. Of course, he knew the transitions aren't instantaneous but take time and mutual interaction of molecules and mirrors can synchronize them.
So is the moral of the story that everybody (or close) that doubts some consensus in physics is a misunderstood genius?
There's also 1000 times more stories of crackpots "knowing" everything, from how to do cold fusion, to perpetual machines, to why Relativity or QM is wrong, etc. They even have the diagrams and math to show you they're right.
And yes, some of them even "knew" things verified later - if you have random unsupported opinions some of those will also be legit.
Unless the parent had some actual proof for their insight before the verification, the phrase "This was always pretty clear to anyone who put some serious thought into it" (as if physicists who didn't regard this didn't) is as good as someone saying the same about a coin toss ("hey, it turned out to be heads, anyone could see that").
No, the story is interesting because it shows the authorities on the subject do get things wrong, and because Townes and many other important scientists knew that the simple idea of instantaneous quantum jumps due to Bohr and Heisenberg and maybe Pauli (I think these three were one the most prominent proponents) wasn't that well secured by the general quantum theory and by the experiments.
The moral is that Bohr just was a guy who lucked into being in the right time and place to contribute to quantum mechanics but that does not make him an otherwise exceptionally insightful into it. Actually, this also is kind of the case with Maxwell. He managed to put together the all-important theory about electromagnetic waves while having in his mind some weird mechanical image of the vacuum that makes one wonder what this guy was actually thinking. And I have to add to this that I consider Maxwell to be a much greater physicist than Bohr. It could well be that being the discoverer of something leads to some kind of intellectual myopia and that further generations are needed to look more sensibly at what was actually discovered.
Yes, that is a very good point. The 'quantum leap' thing comes from the origin of quantum mechanics when things were not well understood and a struggle for understanding was going on and crude concepts were introduced. The more correct theory for atomic decays has been known to be quantum electrodynamics for a long time now. Since an atom is a thing that extends over some space the Feynman diagrams that contribute to it have multiple vertices and the whole thing clearly should be extended over space and time. It is nice to have experimental confirmation of this but it really is quite the opposite of unexpected.
I have made some simulations of quantum 'jumps'. Although the veracity of the simulation could be debated, it's pretty clear what's happening when you see it in action.
Some day i'll write it up in a blog post and put the videos on youtube.
I mean, the paper just describes an experiment that confirmed the theory that's from the 90s. So it's definitely not a surprise, since this behaviour was expected...
What's news here is that this transition, which has always been predicted by theory, has been experimentally observed for the first time.