Separation of x and t in -Et+px And RMS Motion versus Quantum Fluctuations

 Traditionally, a quantum bound state is said to approach classical behaviour for large n (from En=energy). The problem is that peak and valley spatial density regions still exist although an envelope passing roughly through the peaks of Wn(x)Wn(x) (where Wn is the wavefunction) matches C/v(x) where v(x) is the classical spatial density. This is proportional to delta t the time a particle spends within constant sized delta x pieces of a bound path. Alternatively, a wave packet is frequently used, but its width tends to increase in time.

     In a previous note (1), we argued that one may write the free particle classical action (relativistic or nonrelativistic) as A= -Et+px with t and x being independent. This leads to dA/dx partial = p and dA/dt partial = -E. Creating eigenvalue equations  -i d/dx partial exp(iA) = p exp(iA) and id/dt partial exp(iA) = Eexp(iA), one may see that exp(-iEt) is periodic in two dimensions in time and exp(ipx) in space. 

  In this note we examine whether the separation of x and t suggests two schemes for the spatial region. The first scheme is linked to exp(-iEt) where E is average energy at each x. In such a case E=pp/2m (for a free particle) at each x and x=p/m t. exp(ipx) on the other hand demonstrates fluctuations in two dimensions within this region which would not normally be seen in the classical realm due to lack of resolution. 

    If one considers a bound system, these two schemes become clearer.  exp(-En t) is associated with an energy En equal to  KEclassical(x) + V(x). On the other hand there is the quantum momentum distribution Wn(x)=Sum over p a(p)exp(ipx) such that:  

    [Sum over p  pp/2m a(p)exp(ipx)]/Wn(x) + V(x) = En  ((A))

As a result, the momentum distribution Wn(x) is linked to En at each x through ((A)). The two dimensional waves exp(ipx) interfere to create an rms v which is classical. This classical motion exists for all n. Wn(x)Wn(x) on the other hand demonstrates quantum fluctuations, but these include also an rms probability of C/v(x) where v(x) is the classical velocity. For large n, C/v(x) becomes the envelope of Wn(x)Wn(x) and so quantum fluctuations are within the envelope. For lower n we argue that C/v(x) must be “removed” from Wn(x)Wn(x) to see the full quantum fluctuations, but that C/v(x) still applies to the classical rms motion which is present and associated with exp(-iEnt). In other words we use neither wavepackets or the idea that classical mechanics is involved only for large n. Classical mechanics is present for all n in ((A)).