2

The most attractive description of the double slit experiment for me is that in Beltrametti and Cassinelli's book.$^{[1]}$ The essence of their description is the following. enter image description here

Beltrametti-Cassinelli description of the double-slit experiment - They somehow treat the $x$ position classically and suppose that the time of the collision to $S_2$ is independent of the position of the collision. They also adopt a reference frame in which the particle has no velocity along the x-axis. So the state space for the one-dimensional configuration space (i.e., along the y-axis) is $\mathfrak L^2(\mathbb R)$.

They take the state of the particle in the absence of the screens as a density operator $D$ and calculate the density operator $D'$ in the presence of screen $S_1$ as the density operator of the conditional probability in state $D$ of finding the particle in the interval $E$ at time instant $\tau$ with respect to the condition of finding it in $F_1\cup F_2$ at time instant $0$, that is, they take $$p_{D'}(\pi_E)=p_D(\pi_{E}(\tau)|\pi_{F_1\cup F_2}(0))\tag 1$$ where $\pi_E(\tau)$ is the projection belonging to the statement "at time instant $\tau$, the particle is located in $E$" and $\pi_{F_1\cup F_2}$ to "at time instant $0$, the particle is located in $F_1\cup F_2$". They use the Heisenberg picture, so that $$\pi(t)=U_t^{-1}\pi(0)U_t\tag 2$$ with the evolution operator $U_t$, and the state $D$ is independent of time. The conditional probability is defined as $$p_D(A|B)=\frac{\mathrm{tr}(DBAB)}{\mathrm{tr}(DB)}\tag 3,$$ that is, $$D'=\frac{BDB}{\mathrm{tr}(DB)}\tag {3'}.$$ At a point they take $D$ a pure state, that is, $D=D^\psi:=\psi\otimes\psi$ with a $\psi\in\mathcal H$ having unit norm). The story ends with $$p_D(\pi_E(\tau)|\pi_{F_1\cup F_2}(0)) = \int_E |U_r(C_1\psi_1(y)+C_2\psi_2(y))|^2 dy \tag 4$$ where $\displaystyle \psi_i=\frac{\chi_{F_i}\psi}{\|\chi_{F_i}\psi\|}$ and $\displaystyle C_i=\frac{\|\chi_{F_i}\psi\|}{\|\chi_{F_1\cup F_2\|}}$ and $\chi_F$ is the characteristic function of the set $F$.

Question - The Aharonov-Bohm experiment is a slightly modified version of the double slit. Is this Beltrametti-Cassinelli kind of description applicable to the Aharonov-Bohm experiment too? Has this ever been attempted? If yes and not, then why not? Wouldn't be it more simple than working with path integrals?

enter image description here


$^{[1]}$ Enrico G. Beltrametti and Gianni Cassinelli - The Logic of Quantum Mechanics, Addison-Wesley, 1981, pp. 283 - 285

mma
  • 644
  • Quanta don't have positions and they don't have velocities. I do not understand what you are trying to do here. The entire argument doesn't match any physically workable definition of quantum that I have ever seen. – FlatterMann Jan 19 '24 at 10:47
  • 2
    @FlatterMann This is an approximation. As written above, x coordinate is treated classically while y by quantum. mechanics. It's not my idea, see the reference given. I note that this kind of notion of conditional probability isn't clear to me, but it works here. – mma Jan 19 '24 at 11:23
  • There is no approximation in which quanta have position and velocity. Moreover, the double slit is not even a good example for quantum mechanics. It's no even a unitary process. – FlatterMann Jan 19 '24 at 11:52
  • x coordinate is classical. No quanta. Only y coordinate is quanta. It has a probability distribution in a given state. – mma Jan 19 '24 at 11:54
  • Classical corpuscles don't have diffraction functions. Only quanta of energy have those... but quanta don't have location. A single quantum doesn't have a probability distribution and a state, either. Only the quantum mechanical ensemble has those. It seems to me that there is something rather strange going on in this argument that mixes concepts that are completely unrelated. – FlatterMann Jan 19 '24 at 11:58
  • In my opinion, double slit is the simplest example in quantum mechanics. And yes, it is a unitary process. This process is responsible for spreading the probability after the slits from the slits to the whole y axis. – mma Jan 19 '24 at 12:00
  • Let's take this to chat, if you want to. Technically the double slit is not even quantum mechanics. Planck's constant is not in it and we aren't analyzing multi-quantum correlations like in the case of entanglement experiments, either. The double slit diffraction function is perfectly classical. Young had an explanation for it in 1801, a hundred years before quantum mechanics was even conceived of. And like I said, the scattering of a real double slit is not even unitary, i.e. the formalism doesn't even apply to it without ad-hoc modifications. – FlatterMann Jan 19 '24 at 12:03
  • This looks like an extraordinarily complicated way to write something very simple. Sure, one can always inflate the size of notation as much as desired, but why? – knzhou Feb 03 '24 at 06:17

0 Answers0