Arbab said:
In Dirac theory the electron velocity is equal to the speed of light. Why should that appear? Why should we try to solve this problem outside quantum mechanics hypothesis? Should we look for an alternative theory?
When we treat Dirac’s theory as a
single-particle theory, the free Dirac particle undergoes a rapidly oscillating motion (
Zitterbewegung) around the classical trajectory. This “motion” is caused by the
interference between the positive and the negative energy components of the wave packet representing the free Dirac particle. However, the Zitterbewegung
vanishes if wave packets with
either positive
or negative energy are used. This shows that a relativistic single particle theory
is not possible, it can only be
approximately considered when the corresponding wave packets is
restricted to either positive or negative energy range.
If the above is not clear enough for you, let me know and I will do the math for you. Basically, paradoxical results are obtained if the “velocity operator” is calculated according to
[tex]\frac{d \hat{x}_{i}}{dt} = \frac{1}{i \hbar} [ \hat{x}_{i} , H_{D}] = c \alpha_{i} .[/tex]
This shows that
1) the absolute value of the electron velocity is equal to the speed of light (the
einenvalues of [itex]\mathbf{\alpha}[/itex] are [itex]\pm 1[/itex]),
2) the components of the velocity cannot be measured simultaneously (the [itex]\alpha_{i}[/itex]’s
don’t commute with each other).
Of course, this is just
rubbish because, according to Ehrenfest’s theorem, QM is meant to reproduce the classical results for the
average values. Indeed, we can show that the classical results are obtained
only if wave packets with only positive, or only negative, energy are used. To see that, first integrate the equation
[tex]\frac{d \alpha_{i}}{dt} = \frac{1}{i \hbar} [ \alpha_{i} , H_{D} ] = \frac{2i}{\hbar} (c \hat{p}_{i} - \alpha_{i}H_{D}) ,[/tex] then average over positive or negative energy wave packet to kill the Zitterbewegung and obtain the relativistic velocity
[tex]\vec{v} = c \langle \vec{\alpha} \rangle = \frac{c^{2} \vec{p}}{E_{p}} .[/tex]