6 Conclusions

OK. What have we learnt?

• The spin polarization of the reflected electron beam from a bulk magnetic surface, in the model described, is this:
  

  $$\mathbf{P} = -\frac{4e^2\hbar{}m_eV_1\mathbf{B_1}}{12e^2m_e^2V_1^2+e^2\hbar^2B_1^2}+O(\{y_1,z_1\})\textrm{.}$$ (11)

  
  

• The extension of the model to multi-layer systems is likely to be a trivial, but time-consuming, mathematical task.
• Two other, similar models are planned, one of which differs from this by the use of a more ``first-principles'' treatment of the Zeeman energy, and the other by a fully classical treatment of the reflection process, and
• there is a strong case for retaining all three models, and using Bayesian statistics to compare them in the light of experimental data, rather than discarding all but the most ``first-principles'' model.

Thank you for listening. I'll show the slides of references gradually, while I field some questions.