nextuppreviouscontents
Next:Quantum BBGKY equations Up:Quantum equations of information Previous:Equation of motion of

Density of states, quantum Liouville equation

To capture the notion of a distribution of pure states which have no phase relationship between them consider a collection of states tex2html_wrap_inline14719 each occurring at time t with probability density tex2html_wrap_inline14747. In order to simplify the discussion in the following calculations, let tex2html_wrap_inline14749tex2html_wrap_inline14751, and write the distribution function in conjugate coordinates as tex2html_wrap_inline14753. Note that tex2html_wrap_inline12887 is positive semidefinite, and nonzero only on the surface tex2html_wrap_inline14757. In anticipation of a quantum Liouville equation, define a flow on the space tex2html_wrap_inline14759 by tex2html_wrap_inline14761. In a manner analogous to the derivation of the classical Liouville equation of section 5 we may derive the quantum Liouville equation in terms of the conjugate coordinates. To be technically precise, we must go through the algebra using the coordinates tex2html_wrap_inline14763, and the result of doing this is
eqnarray2920
This is the precise analogy of the classical Liouville equation and, except for the factor tex2html_wrap_inline14777 and the complex coordinates, it has the same form as that equation.

Given tex2html_wrap_inline12887, the expected value of an operator Q is given by
equation2946
where we note that we depend on the fact that the probability density tex2html_wrap_inline12887 is nonzero only on the normalized surface of tex2html_wrap_inline14719's to connect tex2html_wrap_inline14719 and tex2html_wrap_inline14801. Equation 7.5 leads immediately to the coordinate-independent or invariant representation of the expectation value by (note that the operator tex2html_wrap_inline14803, where tex2html_wrap_inline14805)
eqnarray2968
where
equation3199
is the density operator.

We move from the equation of motion of the probability density of states to the equation of motion of the density operator. Take the equation of motion of the probability density of states, equation 7.4, multiply it on the left by tex2html_wrap_inline14719, multiply it on the right by tex2html_wrap_inline14853, and then integrate over tex2html_wrap_inline14719 and tex2html_wrap_inline14801. The term involving the partial time derivative of the density of states is trivial to find, but the other terms involve more effort to work out. The partial time derivative term is
equation3011
while the other terms are found using the identity
equation3023
and integrating by parts, noting that tex2html_wrap_inline12887 is nonzero only on the normalized surface of tex2html_wrap_inline14719's. Doing the algebra and substituting yields the coordinate-independent equation of motion of the density operator
equation3045
where tex2html_wrap_inline14883 is the operator commutator. (It is perhaps easiest to demonstrate this result by showing that the ijth elements of the operators are equal for all i,j). This relationship can also be derived by doing a direct differentiation with respect to time of tex2html_wrap_inline14889 as defined in equation 7.9 on the one hand, and then on the other, noting that equation 7.4 shows that the probability density at a point moving with the Schrodinger flow is constant, transforming to time zero coordinates, and doing the time derivative on that form. The derivation done here shows how the coordinate-dependent and coordinate-independent descriptions of the motion interrelate, demonstrates how to begin with the coordinate independent quantum Liouville equation and then derive the coordinate independent quantum Liouville equation, and prepares the reader for the quantum reduced density equations of motion, directly analogous to the classical BBGKY equations of section 5.2.


nextuppreviouscontents
Next:Quantum BBGKY equations Up:Quantum equations of information Previous:Equation of motion of
David Wolf

Tue Mar 25 08:11:49 CST 1997