2
\$\begingroup\$

I am a material engineer and have a question about a formula derivation regarding microwave absorption properties. I really cannot figure it out after days of trying. This should be simple for a specialist.

In this attached paper, how could one derive Eq(10) based on Eq (8) and (9)? Is k_2 in Eq (8) a complex number? enter image description here Source: https://www.researchgate.net/profile/Sture-Petersson/publication/230997780_Semi-analytic_model_of_the_permeable_base_transistor/links/02e7e530bd23818ff0000000/Semi-analytic-model-of-the-permeable-base-transistor.pdf#page=168

\$\endgroup\$

1 Answer 1

3
\$\begingroup\$

This is about very thin metal layers, much thinner than the skin depth, so it should indeed be obvious that an incoming wave will see as impedance the metal sheet resistance parallel to the wave impedance of the medium behind it. And this parallel impedance is what Eq. 10 is showing. But the authors derive it by beginning with a more general analysis.

As for the question about \$k_2\$, certainly it is a complex number. Since \$k_2\$ is the propagation number in region 2, which is metal, Eq (6) applies, so it is a real number times \$(1+j)\$. Something similar also applies to \$\eta_\text{met}\$, the wave impedance in the metal. More importantly, this \$\eta_\text{met}\$ is very low, because the metal as a conductor has a very low (and lossy) wave impedance. We further need the fact that the sheet resistance is: $$ R_\text{s}=\frac{1}{\sigma\,d} $$

I don't see in your picture how the authors defined \$\eta_\text{met}\$, but it should be somewhere and it should be similar to Eq. (6) for \$k_2\$. If you divide those two quantities the simple result is: $$ \frac{\eta_\text{met}}{k_2} =\frac{-j}{\sigma} \tag{1} $$

Now we are ready to tackle Eq. 8: $$\begin{align} Z_{1-2}(d) = \eta_\text{met}\ \frac{\eta_0-j\,\eta_\text{met}\tan(k_2d)} {\eta_\text{met}-j\,\eta_0\tan(k_2d) } \approx \eta_\text{met}\ \frac{\eta_0-j\,\eta_\text{met}\,k_2d} {\eta_\text{met}-j\,\eta_0\ k_2d} \\[10pt] = \ \frac{\eta_0\,\eta_\text{met}/(k_2d)-j\,\eta_\text{met}^2} {\eta_\text{met}/(k_2d)-j\,\eta_0 } = \ \frac{-\eta_0\,j/(\sigma d)-j\,\eta_\text{met}^2} {-j/(\sigma d)-j\,\eta_0 } \end{align}$$ where the second line used result \$(1)\$ from above and the first line uses the approximation of Eq. (7) in the paper. Now neglect the term with \$\eta_\text{met}^2\$ because it is small, and use the definition of \$R_s\$: $$\begin{align} Z_{1-2}(d) &= \ \frac{-\eta_0\,j/(\sigma d)-j\,\eta_\text{met}^2} {-j/(\sigma d)-j\,\eta_0 } \approx \ \frac{-\eta_0\,j/(\sigma d)} {-j/(\sigma d)-j\,\eta_0 } = \ \frac{\eta_0\, R_\text{s}} {\eta_0+ R_\text{s} } \end{align}$$

\$\endgroup\$
2
  • \$\begingroup\$ Thanks $Dave Tweed, I did not know that we needed backslashes for the inline math! Why is that different from physics or math SE? \$\endgroup\$ Commented Apr 26 at 6:53
  • 1
    \$\begingroup\$ Mainly because we also frequently use $ in regular text -- more than in those other groups, I guess -- so here we have to escape it with `\` in order to invoke MathJax inline. \$\endgroup\$
    – Dave Tweed
    Commented Apr 26 at 15:14

Not the answer you're looking for? Browse other questions tagged or ask your own question.