20.01.2015 Views

Feynman Path Integral Formulation

Feynman Path Integral Formulation

Feynman Path Integral Formulation

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

6.14 Lattice Fermions, Tetrads and Spin Rotations 211<br />

Without loss of generality we can fix the average edge length to be equal to one,<br />

〈l〉 = 1, which then requires λ 0 = 5 2<br />

+ σ. In order for the model to be meaningful,<br />

the measure parameter is constrained by σ > −5/2, i.e. the measure over the edges<br />

cannot be too singular.<br />

6.14 Lattice Fermions, Tetrads and Spin Rotations<br />

On a simplicial manifold spinor fields ψ s and ¯ψ s are most naturally placed at the center<br />

of each d-simplex s. In the following we will restrict our discussion for simplicity<br />

to the four-dimensional case, and largely follow the original discussion given in<br />

(Drummond, 1986). As in the continuum (see for example Veltman, 1975), the construction<br />

of a suitable lattice action requires the introduction of the Lorentz group<br />

and its associated tetrad fields e a μ(s) within each simplex labeled by s.<br />

Within each simplex one can choose a representation of the Dirac gamma matrices,<br />

denoted here by γ μ (s), such that in the local coordinate basis<br />

{γ μ (s),γ ν (s)} = 2g μν (s) . (6.161)<br />

These in turn are related to the ordinary Dirac gamma matrices γ a , which obey<br />

{<br />

γ a ,γ b} = 2η ab (6.162)<br />

by<br />

γ μ (s) =e μ a (s)γ a , (6.163)<br />

so that within each simplex the tetrads e a μ(s) satisfy the usual relation<br />

e μ a (s) e ν b (s) ηab = g μν (s) . (6.164)<br />

In general the tetrads are not fixed uniquely within a simplex, being invariant under<br />

the local Lorentz transformations discussed earlier in Sect. 6.4.<br />

In the continuum the action for a massless spinor field is given by<br />

∫<br />

I = dx √ g ¯ψ(x)γ μ D μ ψ(x) , (6.165)<br />

where D μ = ∂ μ + 1 2 ω μabσ ab is the spinorial covariant derivative containing the spin<br />

connection ω μab . It will be convenient to first consider only two neighboring simplices<br />

s 1 and s 2 , covered by a common coordinate system x μ . When the two tetrads<br />

in s 1 and s 2 are made to coincide, one can then use a common set of gamma matrices<br />

γ μ within both simplices. Then in the absence of torsion the covariant derivative D μ<br />

in Eq. (6.165) reduces to just an ordinary derivative. The fermion field ψ(x) within<br />

the two simplices can then be suitably interpolated, by writing for example

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!