Tauc-Lorentz dispersion model
modelname
Tauc-Lorentz|TLImaginary part of complex dielectric function is described by four parameters \(E_\mathrm{g}, N_1, E_1, B_1\). The TL model can be calculated by five different parametrizations, see attributes
CC
, JM
, DF1
, ASF
and DF2
.
In the case that no attribute is chosen, the another four parameters are generated, i.e.
\(f_{\rm JM}\), \(f_{\rm CC}\), \(f_{\rm DF1}\), \(f_{\rm DF2}\).
Dielectric function is then calculated as linear combination of the all version of Tauc-Lorent model:
$$
\begin{array}{l}
\hat\varepsilon(E) = 1 + f_{\rm DF2} \ \hat\chi_{\rm DF2}(E) + (1-f_{\rm DF2}) \\
\times \Bigg( f_{\rm CC} \ \hat\chi_{\rm CC}(E) + (1-f_{\rm CC}) \Big( f_{\rm DF1} \ \hat\chi_{\rm DF1}(E) + (1-f_{\rm DF1})
\big( f_{\rm JM} \hat\chi_{\rm JM}(E) + (1-f_{\rm JM}) \, \hat\chi_{\rm ASF}(E) \big)\Big) \Bigg)
\end{array}
$$
where \(\hat\chi\) are corresponding susceptibilities.
attributes
-
CC
- Campi-Corriasso version of TL model: $$ \begin{array}{lcl} \displaystyle \varepsilon_\mathrm{i} = \frac{2}{\pi} \frac{N_1 B_1 (E-E_\mathrm{g})^2}{E(((E-E_\mathrm{g})^2-(E_1-E_\mathrm{g})^2)^2+B_1^2 (E-E_\mathrm{g})^2)} & \mathrm{for} & E > E_\mathrm{g} \\ \varepsilon_\mathrm{i} = 0 & \mathrm{for} & E \leq E_\mathrm{g} \end{array} $$ Real part of dielectric function is expressed analyticlally using the Kramers-Kronig relation. -
JM
- Jellison-Modine version of TL model: $$ \begin{array}{lcl} \displaystyle \varepsilon_\mathrm{i} = \frac{N_1(E-E_\mathrm{g})^2}{{\cal C}_\mathrm{N} E ((E^2-E_1^2)^2+B_1^2 E^2)} & \mathrm{for} & E > E_\mathrm{g} \\ \varepsilon_\mathrm{i} = 0 & \mathrm{for} & E \leq E_\mathrm{g} \end{array} $$ where \(C_\mathrm{N}\) is normalization constant. Real part of dielectric function is expressed analyticlally using the Kramers-Kronig relation. -
DF1
- Our first version of TL model: $$ \begin{array}{lcl} \displaystyle \varepsilon_\mathrm{i} = \frac{N_1 (E - E_\mathrm{g})^2}{{\cal C}_\mathrm{N}E^3((E-\sqrt{E_1^2-B_1^2/4})^2+B_1^2/4)} & \mathrm{for} & E > E_\mathrm{g} \\ \varepsilon_\mathrm{i} = 0 & \mathrm{for} & E \leq E_\mathrm{g} \end{array} $$ where \(C_\mathrm{N}\) is normalization constant. Real part of dielectric function is expressed analyticlally using the Kramers-Kronig relation. -
ASF
- A.S. Ferlauto at al. version of TL model: $$ \begin{array}{lcl} \displaystyle \varepsilon_\mathrm{i} = \frac{N_1 \, E \, (E-E_\mathrm{g})^2}{{\cal C}_\mathrm{N} ((E-E_\mathrm{g})^2 + E_\mathrm{p}^2) \, ((E^2-E_1^2)^2+B_1^2 E^2)} & \mathrm{for} & E > E_\mathrm{g} \\ \varepsilon_\mathrm{i} = 0 & \mathrm{for} & E \leq E_\mathrm{g} \end{array} $$ where \(C_\mathrm{N}\) is normalization constant. This approximation have one extra parameter \(E_\mathrm{p}\), which influences transition strength below and above peak energy. Real part of dielectric function is expressed analyticlally using the Kramers-Kronig relation. -
DF2
- Our second version of TL model: $$ \begin{array}{lcl} \displaystyle \varepsilon_\mathrm{i} = \frac{N_1 (E - E_\mathrm{g})^2}{{\cal C}_\mathrm{N}E((E-E_\mathrm{g})^2 + E_\mathrm{p}^2)((E-\sqrt{E_1^2-B_1^2/4})^2+B_1^2/4)} & \mathrm{for} & E > E_\mathrm{g} \\ \varepsilon_\mathrm{i} = 0 & \mathrm{for} & E \leq E_\mathrm{g} \end{array} $$ where \(C_\mathrm{N}\) is normalization constant. Real part of dielectric function is expressed analyticlally using the Kramers-Kronig relation. -
CJF
- Linear combination of CC, JM and DF1 version of TL model: $$ \hat\varepsilon(E) = 1 + f_{\rm CC} \ \hat\chi_{\rm CC}(E) + (1-f_{\rm CC}) \Big( f_{\rm JM} \ \hat\chi_{\rm JM}(E) + (1-f_{\rm JM}) \, \hat\chi_{\rm DF1}(E) \Big) . $$ -
number_of_terms
- Adds terms and corresponding parameters \(N_2, E_2, B_2, \dots\).
Example
media: f = TL s = TL:CC
newAD2> par Egf = 1 fixed (0,inf) eV Eqf = 1 fixed (0,inf) eV fCCf = 0 fixed [0,1] fJMf = 0 fixed [0,1] fDF1f = 0 fixed [0,1] fDF2f = 0 fixed [0,1] N1f = 0 fixed [0,inf) eV2 E1f = 3 fixed (1e-4,inf) eV B1f = 1 fixed (1e-4,inf) eV Egs = 1 fixed (0,inf) eV N1s = 0 fixed [0,inf) eV2 E1s = 3 fixed (1e-4,inf) eV B1s = 1 fixed (1e-4,inf) eV newAD2>
