next up previous
Next: Case Study: Mach-Zender Interferometer Up: Module 3: Nonlinear and Previous: Case Study: Isolators and

Nonlinear Optics

When laser beams are passed through dielectric materials, the high intensity tex2html_wrap_inline1271 field can induce a nonlinear relation between tex2html_wrap_inline1317 and tex2html_wrap_inline1271 . In practice, devices exploiting such nonlinear behavior also involve anisotropic behavior. For simplicity, we assume the material is isotropic, and concentrate on the nonlinearities.

We first illustrate a model for examining the nonlinearity, the Lorentz equation. We examine the microscopic effect of an electric field tex2html_wrap_inline1271 on an atom in the material. Let r denote the displacement of the electron cloud due to the electric field; that is, the induced dipole moment is p=-er where e is the charge of the electron. Now, tex2html_wrap_inline1271 is a macroscopic quantity, and it is not valid to use this value on an atomic scale (the neighboring atoms, for example, influence the local behavior); let us represent the ``local'' electric field by tex2html_wrap_inline1911 , where g is a constant ``fudge factor''.

Now, if we perturb the electron cloud, there will be a tendency for it to return to equilibrium (the Coulombic attraction to the nucleus, for example, induces a restoring force); let us quantify this restoring force by tex2html_wrap_inline1915 . That is, if we think of a simple harmonic oscillator, tex2html_wrap_inline1915 is the frequency of oscillation. There is a damping constant tex2html_wrap_inline1919 , which represents resistance to perturbation. Then the differential equation relating r to tex2html_wrap_inline1923 is:

  equation408

where tex2html_wrap_inline1697 is the charge-to-mass ratio of the electron.

The nonlinear behavior of the system is modeled by the introduction of the term tex2html_wrap_inline1927 in (4.1). The parameter tex2html_wrap_inline1929 is a constant, and is very small in most materials. We can obtain a power series solution:

  equation416

and hence the total displacement can be represented as the sum of displacements tex2html_wrap_inline1931 proportional to the tex2html_wrap_inline1933 power of E. The corresponding polarization density is tex2html_wrap_inline1937 where N is the volume density of electrons. Hence:

  equation420

where:

  equation423

For the isotropic case, the coefficients tex2html_wrap_inline1941 are scalars. In most materials, the nonlinear terms ( tex2html_wrap_inline1943 and higher order terms) are anisotropic, represented by tensors (matrices). In many applications, the electric field is comprised of fields at a few, discrete frequencies, tex2html_wrap_inline1945 . For example, there may be a bias field ( tex2html_wrap_inline1947 ) and a laser beam tex2html_wrap_inline1949 present in the material. In this case, tex2html_wrap_inline1951 will be comprised of frequencies at tex2html_wrap_inline1915 and its harmonics ( tex2html_wrap_inline1955 , etc.), as well as a bias term.

The second order susceptibility coefficients are called Pockels coefficients, and the corresponding effect is called the Pockels or electro-optic effect. The third order coefficients are called Kerr coefficients, and the corresponding effect is called the Kerr effect. The Pockels coefficients are very small, several orders of magnitude smaller than the linear dielectric constant tex2html_wrap_inline1463 , and the Kerr coefficients are even smaller. In particular, the Kerr effect is generally relevant or noticeable only in materials which exhibit special symmetries (called inversion symmetry) so the Pockels effect is identically 0. We will examine the Pockels effect, though.




next up previous
Next: Case Study: Mach-Zender Interferometer Up: Module 3: Nonlinear and Previous: Case Study: Isolators and

Prof. F. Fontaine
Thu May 9 15:59:44 EDT 1996