Non-relativistic quantum mechanics

Date: 2015-10-07; view: 376.

The above discussed transformation of the Kepler problem can now be used to calculate the energy levels of the hydrogen atom as was described in [15]. To this purpose one introduces holomorphic coordinates

(5.1)

so that the Hamiltonian H₄ in (4.19) can be written as:

(5.2)

where k = e². Introducing then holomorphic coordinates for left and right moving quanta

(5.3)

the Hamiltonian (5.2) turns into

(5.4)

One can now quantize this system with the Moyal product. The four-dimensional Moyal star product transforms under KS-transformation and the above transformations into

(5.5)

The energy levels can then be obtained by the -genvalue equation

(5.6)

where is the product of four Wigner functions of the one-dimensional harmonic oscillator given in (2.6). Eq. (5.6) gives then

e2= ω(nR12+nL12+nR34+nL34+2).

(5.7)

To get the energy levels of the hydrogen atom one has to impose the constraint

(5.8)

which for the energy levels corresponds to n_R12-n_L12+n_R34-n_L34=0 or n_R12+n_R34=n_L12+n_L34≡n-1. Putting this and into (5.7) one gets the well-known energy levels of the hydrogen atom

(5.9)

Geometric algebra in a fermionic star product formulation can be used in general to describe quantum mechanics, when combined in a straightforward way with the bosonic star product formalism. In classical mechanics described with geometric algebra and the Clifford star product the fermionic part of the underlying superanalysis was deformed and the basis vectors played only a mathematical role by generating the structures of vector analysis. Going over to quantum mechanics means that also the scalar coefficients of superanalysis have to be multiplied by a deformed product, namely the bosonic Moyal star product. This leads then to a deformed version of geometric algebra and describing geometric algebra in terms of star products allows to combine the Clifford star product and the Moyal product into one star product, which should be called Moyal–Clifford product. The Clifford product on the phase space that described the structures of classical Hamilton mechanics was given by (4.20). In quantum mechanics one needs now a product with which general multivector functions on the phase space are multiplied. These multivector functions are the observables of the theory and as such can only be multivectors in the space basis vectors σ_r. So one has to go over from the Clifford product (4.20) to the Clifford product (3.14), which can be done by implementing constraints that identify the corresponding basis vectors [6]. The Moyal–Clifford product for a single particle system is then

(5.10)

To see the consequences of the additional Moyal deformation in geometric algebra one can for example consider the Moyal–Clifford product of two vectors in d = 2 dimensions. The generalization of (3.15) can be written as

a MCb=(a1 Mb2-a2 Mb1)σ1σ2+a1 Mb1+a2 Mb2.

(5.11)

Under the Moyal product the coefficients in general do not commute if they are functions of q_n and p_n. This means that the Moyal–Clifford product of the same vectors a _MCa is in general not a scalar, but has also a bivector part. It is this additional bivector part, which appears only for ≠0, that constitutes the spin as a physical observable. This can be seen if one considers the minimal substituted Hamiltonian which is in the formalism of deformed geometric algebra given by:

(5.12)

(5.13)

The first three terms describe the Landau problem of a charged particle in a magnetic field which can be solved in the star product formalism as described in [16] or [7]. The other three terms that describe the interaction of the spin and the magnetic field appear only because of the Moyal product. If the magnetic field points in σ₃-direction the vector potential is given by and only the first Moyal-commutator contributes:

(5.14)

where and σ³=-iσ₁σ₂ is a real quaternion, which is constructed according to Figs. (3.13) and (2.14). The difference between this calculation and the conventional approach is that in the conventional formalism the Clifford structure is introduced by putting in Pauli matrices by hand in (5.12). The Pauli matrices describe the spin and lead analogously to the additional term H_S, this is known as the Feynman trick [17]. In geometric algebra the Clifford structures do not have to be added, they are just the basis vectors that already exist in classical mechanics, but become apparent as physical objects in the quantum case. It is then straightforward to calculate the -eigenfunctions of H_S which turn out to be the spin Wigner functions described in Section 1 [7].

One should note that the Moyal–Clifford product is a product for functions on the phase space, which play the role of observables. As seen above these observables are in general multivectors, where the terms of higher grade are described by the space basis vectors σ_n and not by the phase space basis vectors η_n and ρ_n, because the latter are not observable quantities. Nevertheless the basis vectors of phase space can be considered to play an indirect role in the expression (2.1) of the Moyal product, because the imaginary structure can be interpreted as a two blade on phase space. If the phase space is just two dimensional there is only one candidate for the imaginary structure, namely the symplectic volume form j = ηρ. That the i in the Moyal product has to be an unit area bivector can be seen from the integral representation of the Moyal product [18]:

(5.15)

where is the area of the triangle spanned by the vectors , and . So i plays here the role of the unit area bivector in phase space. The two-dimensional Moyal product can then be written with the gradient _x = η∂_q + ρ∂_p as:

(5.16)

so that the correspondence principle has the form

(5.17)

One should also note the similarity to the fermionic star product of two vectors a = a₁η + a₂ρ and b = b₁η + b₂ρ:

(5.18)

where A (a, b) is the area of the triangle spanned by the vectors a and b.