New study on quantum tunneling

[nec209]

Here is a new study on quantum tunneling

 

Physicists watch quantum particles tunnel through solid barriers. Here's what they found.

https://www.livescience.com/quantum-tunneling-observed-and-measured.html

 

[hoola]

since they used a magnetic field to tunnel through, couldn't that be reason for the seemingly very slow transition rate? Like a magnet dropped down a copper tube, the fall is slowed down due to the back reactions.

[swansont]

2 hours ago, hoola said:

since they used a magnetic field to tunnel through, couldn't that be reason for the seemingly very slow transition rate? Like a magnet dropped down a copper tube, the fall is slowed down due to the back reactions.

What is the analogue of the copper tube? What would be exerting this “back-reaction”?

It was atoms and a magnetic barrier.

 

[drumbo]

The entire study is flawed. They did not account for Heisenberg's uncertainty principle which states that the more precisely the position of some particle is determined, the less precisely its momentum can be predicted from initial conditions, and vice versa. By localizing a pseudo-magnetic field inside the barrier, they interfered with the spin precession of the atoms as a clock to measure the time that they require to cross the classically forbidden region.

[swansont]

8 minutes ago, drumbo said:

The entire study is flawed. They did not account for Heisenberg's uncertainty principle which states that the more precisely the position of some particle is determined, the less precisely its momentum can be predicted from initial conditions, and vice versa. By localizing a pseudo-magnetic field inside the barrier, they interfered with the spin precession of the atoms as a clock to measure the time that they require to cross the classically forbidden region.

What is a pseudo-magnetic field?

How is this localized?

How does localization affect spin precession? What is the commutation relation involved, that would allow one to invoke the HUP?

[nec209]

The Heisenberg's uncertainty is saying if you are trying to measure something on quantum level the measurement process itself interferes with the object you are measuring and  increasing the error in the measurement. 
 

[Strange]

7 minutes ago, nec209 said:

The Heisenberg's uncertainty is saying if you are trying to measure something on quantum level the measurement process itself interferes with the object you are measuring and  increasing the error in the measurement. 

It doesn't say that at all.

You are describing the observer effect: https://en.wikipedia.org/wiki/Observer_effect_(physics)

[swansont]

As Strange said, no. The HUP applies to limitations on measurements involving observables that don’t commute.

[nec209]

On 8/9/2020 at 5:42 PM, Strange said:

It doesn't say that at all.

You are describing the observer effect: https://en.wikipedia.org/wiki/Observer_effect_(physics)

I think it's that if you try to measure the location of an electron, it will change to a different location. To measure the location you need photons but the photon will bounce off the electron and the electron will be in a different location and  so the measurement is not correct anymore.

[Strange]

4 minutes ago, nec209 said:

I think it's that if you try to measure the location of an electron, it will change to a different location. To measure the location you need photons but the photon will bounce off the electron and the electron will be in a different location and  so the measurement is not correct anymore.

That is the observer effect, not the HUP.

[joigus]

I concur with Strange. HUP has to do with mean square deviations from average value, not measurement effects. You can apply as a bound to preparations (non-disruptive, or filtering measurements).

This has been discussed elsewhere in the forums.

There are implications of non-commutativity and 2nd-kind measurements (disruptive measurements), but it's more subtle, and I'd rather not go into that. Too many different cases involved.

Even preparations are more subtle. Classical quantum mechanics books get it wrong, on account of being too simplistic. It's generally discussed that you can prepare an electron with an arbitrarily accurate state of p_x, p_y, p_z (because in the framework of the theory they're commuting). That's a theoretical fiction. You can collimate electron beams with a selected value of, say p_z, but you cannot guarantee that p_x, p_y are exactly zero, if nothing else, for the very simple experimental reason that you must make your beams go through diffraction windows in the perpendicular direction in order to filter them. I think that lies at the core of why people like M. Berry and others are finding a richer structure in electron beams than the naive academic picture that they're plane waves. Among other things, they can generally encapsulate orbital angular momentum.

There are parallel developments in optics (Laguerre-Gaussian, Hermite-Gaussian beams...).

I'm not 100 % sure that what I'm saying is totally watertight as to electrons. I'm particularly interested to read @Strange and @swansont's take on this.

Sorry if these comments stray too off topic.

Edited August 11, 2020 by joigus

[swansont]

11 hours ago, nec209 said:

I think it's that if you try to measure the location of an electron, it will change to a different location. To measure the location you need photons but the photon will bounce off the electron and the electron will be in a different location and  so the measurement is not correct anymore.

That was Heisenberg’s original argument, but it’s incorrect.

The uncertainty is inherent, owing to the wave nature of QM. The variables are Fourier transforms of each other, and the uncertainty drops out of the math.

 

10 hours ago, joigus said:

 

Even preparations are more subtle. Classical quantum mechanics books get it wrong, on account of being too simplistic. It's generally discussed that you can prepare an electron with an arbitrarily accurate state of p_x, p_y, p_z (because in the framework of the theory they're commuting). That's a theoretical fiction. You can collimate electron beams with a selected value of, say p_z, but you cannot guarantee that p_x, p_y are exactly zero,

Experiments I’ve seen use polarization states for photons, or spin projection for electrons. Easier to prepare, I would imagine.