Electromagnetic Waves, Antennas, and RF: Boundary Waves - Reflection, Transmission, and Refraction
ElectromagneticWavesAntennasAndRFBoundaryWavesReflectionTransmissionAndRefraction
2026-10-09 23:04:05
2026-10-09 23:04:05
Topic
removing the boxedfor pdf rendering
electromagnetic boundary conditions
electric field reflection coefficient
electric field transmission coefficient
power reflection coefficient
power transmission coefficient
Snell's law
TE polarization
TM polarization
Brewster angle
critical angle
total internal reflection
standing-wave ratio
electromagnetic boundary conditions
reflection
transmission
refraction
intrinsic impedance
reflection coefficient
transmission coefficient
Fresnel equations
Snell law
TE polarization
TM polarization
Brewster angle
critical angle
total internal reflection
standing wave
RF interface
% this is the default PhysicsLibrary preamble. as your knowledge
% of TeX increases, you will probably want to edit this, but
% it should be fine as is for beginners.
% almost certainly you want these
\usepackage{amssymb}
\usepackage{amsmath}
\usepackage{amsfonts}
% used for TeXing text within eps files
%\usepackage{psfrag}
% need this for including graphics (\includegraphics).
% If you use \includegraphics{figure} with no file extension, PhysicsLibrary
% will prefer figure.png before falling back to figure.eps.
\usepackage{graphicx}
\DeclareGraphicsExtensions{.png,.jpg,.jpeg,.pdf,.eps}
% for neatly defining theorems and propositions
%\usepackage{amsthm}
% making logically defined graphics
%\usepackage{xypic}
% there are many more packages, add them here as you need them
% define commands here
\section*{Electromagnetic Waves, Antennas, and RF: Boundary Waves - Reflection, Transmission, and Refraction}
EM20 showed how a uniform electromagnetic wave propagates \emph{inside} a homogeneous material. Real propagation paths are rarely uniform forever. A wave eventually meets a boundary: air to glass, air to soil, free space to a radome, dielectric to conductor, or one layer of a circuit or antenna structure to another.
At that boundary, Maxwell's equations do not permit the fields to change arbitrarily. Their allowed discontinuities are fixed by the electromagnetic boundary conditions. Those conditions, together with the plane-wave relation between electric and magnetic fields, determine how much of the incident wave is reflected, how much is transmitted, whether the transmitted wave changes direction, and whether a phase reversal occurs \cite{Griffiths2017,Cheng1989,Ulaby2015,Balanis2012}.
The central chain developed in this article is
\begin{equation}
\boxed{
\text{Maxwell equations}
\longrightarrow
\text{boundary conditions}
\longrightarrow
\Gamma,\tau
\longrightarrow
R,T
\longrightarrow
\text{Snell and Fresnel laws}.
}
\end{equation}
\section{The interface geometry}
Let medium 1 occupy $z<0$ and medium 2 occupy $z>0$. The interface is the plane $z=0$, with unit normal
\begin{equation}
\hat{\mathbf n}=\hat{\mathbf z}
\end{equation}
pointing from medium 1 into medium 2.
Each medium has constitutive parameters
\begin{equation}
(\epsilon_1,\mu_1,\sigma_1),
\qquad
(\epsilon_2,\mu_2,\sigma_2),
\end{equation}
and therefore its own propagation constant and intrinsic impedance,
\begin{equation}
\gamma_m=\alpha_m+i\beta_m,
\qquad
\eta_m=\sqrt{\frac{i\omega\mu_m}{\sigma_m+i\omega\epsilon_m}},
\qquad m=1,2.
\end{equation}
For the first derivation we use lossless media, so $\eta_1$ and $\eta_2$ are real. The complex-impedance extension is introduced later.
\section{Boundary conditions come directly from Maxwell's equations}
The wave-interface formulas are not separate laws added to electromagnetism. They follow from the integral forms of Maxwell's equations by shrinking the integration surface or contour around the interface \cite{Griffiths2017,Jackson1999,Cheng1989}.
\subsection{Tangential electric field}
Faraday's law in integral form is
\begin{equation}
\oint_C \mathbf E\cdot d\boldsymbol\ell
=-\frac{d}{dt}\int_S\mathbf B\cdot d\mathbf a.
\end{equation}
Take a narrow rectangular contour that straddles the interface. As its height tends to zero, the enclosed area tends to zero, so the magnetic-flux term vanishes for finite fields. The two long sides remain, giving
\begin{equation}
\boxed{
\hat{\mathbf n}\times(\mathbf E_2-\mathbf E_1)=\mathbf 0.
}
\end{equation}
Thus the tangential electric field is continuous across an ordinary interface.
\subsection{Tangential magnetic field}
Ampere-Maxwell law gives
\begin{equation}
\oint_C \mathbf H\cdot d\boldsymbol\ell
=I_{\text{free}}+\frac{d}{dt}\int_S\mathbf D\cdot d\mathbf a.
\end{equation}
If a free surface-current density $\mathbf K_s$ exists at the interface, the limiting contour gives
\begin{equation}
\boxed{
\hat{\mathbf n}\times(\mathbf H_2-\mathbf H_1)=\mathbf K_s.
}
\end{equation}
For the dielectric interfaces considered first,
\begin{equation}
\mathbf K_s=\mathbf 0,
\end{equation}
so the tangential magnetic field is also continuous.
\subsection{Normal electric flux density}
Gauss's law applied to a thin pillbox gives
\begin{equation}
\boxed{
\hat{\mathbf n}\cdot(\mathbf D_2-\mathbf D_1)=\rho_s,
}
\end{equation}
where $\rho_s$ is free surface-charge density.
\subsection{Normal magnetic flux density}
Gauss's law for magnetism similarly gives
\begin{equation}
\boxed{
\hat{\mathbf n}\cdot(\mathbf B_2-\mathbf B_1)=0.
}
\end{equation}
The normal component of $\mathbf B$ is therefore continuous because magnetic monopole surface charge does not appear in classical electromagnetism.
\begin{center}
\includegraphics{EM21_fig01_interface_boundary_conditions.png}
\vspace{0.4em}
\textit{Figure 1. Maxwell boundary conditions at an interface. A narrow loop isolates tangential-field conditions while a pillbox isolates normal-flux conditions.}
\end{center}
\section{Normal incidence: incident, reflected, and transmitted waves}
Consider a linearly polarized plane wave incident normally from medium 1. Choose
\begin{equation}
\mathbf E_i(z)=\hat{\mathbf x}E_{i0}e^{-i\beta_1 z}
\end{equation}
for the incident phasor. The reflected wave propagates in the $-z$ direction,
\begin{equation}
\mathbf E_r(z)=\hat{\mathbf x}E_{r0}e^{+i\beta_1 z},
\end{equation}
and the transmitted wave propagates in the $+z$ direction,
\begin{equation}
\mathbf E_t(z)=\hat{\mathbf x}E_{t0}e^{-i\beta_2 z}.
\end{equation}
For a uniform plane wave,
\begin{equation}
\mathbf H=\frac{1}{\eta}\hat{\mathbf k}\times\mathbf E.
\end{equation}
Therefore
\begin{equation}
\mathbf H_i=\hat{\mathbf y}\frac{E_{i0}}{\eta_1}e^{-i\beta_1z},
\end{equation}
\begin{equation}
\mathbf H_r=-\hat{\mathbf y}\frac{E_{r0}}{\eta_1}e^{+i\beta_1z},
\end{equation}
and
\begin{equation}
\mathbf H_t=\hat{\mathbf y}\frac{E_{t0}}{\eta_2}e^{-i\beta_2z}.
\end{equation}
The minus sign in $\mathbf H_r$ is essential: the reflected wave carries energy toward $-z$.
\section{Deriving the normal-incidence reflection coefficient}
At $z=0$, tangential $\mathbf E$ is continuous:
\begin{equation}
E_{i0}+E_{r0}=E_{t0}.
\end{equation}
Tangential $\mathbf H$ is also continuous when $\mathbf K_s=0$:
\begin{equation}
\frac{E_{i0}}{\eta_1}-\frac{E_{r0}}{\eta_1}
=\frac{E_{t0}}{\eta_2}.
\end{equation}
Define the electric-field reflection coefficient
\begin{equation}
\boxed{\Gamma\equiv\frac{E_{r0}}{E_{i0}}}
\end{equation}
and transmission coefficient
\begin{equation}
\boxed{\tau\equiv\frac{E_{t0}}{E_{i0}}.}
\end{equation}
The first boundary condition gives
\begin{equation}
1+\Gamma=\tau.
\end{equation}
The magnetic-field condition gives
\begin{equation}
\frac{1-\Gamma}{\eta_1}=\frac{\tau}{\eta_2}.
\end{equation}
Substitute $\tau=1+\Gamma$:
\begin{equation}
\eta_2(1-\Gamma)=\eta_1(1+\Gamma).
\end{equation}
Collect the $\Gamma$ terms:
\begin{equation}
\eta_2-\eta_1=\Gamma(\eta_2+\eta_1).
\end{equation}
Hence
\begin{equation}
\boxed{
\Gamma=\frac{\eta_2-\eta_1}{\eta_2+\eta_1}.
}
\end{equation}
Then
\begin{equation}
\boxed{
\tau=1+\Gamma
=\frac{2\eta_2}{\eta_1+\eta_2}.
}
\end{equation}
These two compact expressions are the normal-incidence Fresnel amplitude coefficients written in impedance form.
\section{What the sign of the reflection coefficient means}
If
\begin{equation}
\eta_2<\eta_1,
\end{equation}
then
\begin{equation}
\Gamma<0.
\end{equation}
The reflected electric field is then reversed in sign relative to the incident electric field at the boundary. For a sinusoid, this is a $180^\circ$ phase reversal.
If
\begin{equation}
\eta_2>\eta_1,
\end{equation}
then
\begin{equation}
\Gamma>0,
\end{equation}
and there is no such sign reversal.
If
\begin{equation}
\boxed{\eta_2=\eta_1,}
\end{equation}
then
\begin{equation}
\boxed{\Gamma=0.}
\end{equation}
The interface is impedance matched and no plane-wave reflection occurs at normal incidence.
\begin{center}
\includegraphics{EM21_fig02_normal_incidence.png}
\vspace{0.4em}
\textit{Figure 2. At normal incidence the reflected and transmitted amplitudes are fixed by continuity of tangential $E$ and $H$. Reflection occurs when the intrinsic impedances differ.}
\end{center}
\section{Power reflection and transmission coefficients}
The time-averaged power density of a sinusoidal plane wave in a lossless medium is
\begin{equation}
\langle S\rangle=\frac{|E_0|^2}{2\eta}.
\end{equation}
Define the reflected power fraction
\begin{equation}
R\equiv\frac{\langle S_r\rangle}{\langle S_i\rangle}.
\end{equation}
Because the incident and reflected waves are in the same medium,
\begin{equation}
\boxed{R=|\Gamma|^2.}
\end{equation}
For the transmitted power fraction,
\begin{equation}
T\equiv\frac{\langle S_t\rangle}{\langle S_i\rangle},
\end{equation}
so
\begin{equation}
T
=\frac{|E_{t0}|^2/(2\eta_2)}{|E_{i0}|^2/(2\eta_1)}.
\end{equation}
Therefore
\begin{equation}
\boxed{
T=\frac{\eta_1}{\eta_2}|\tau|^2.
}
\end{equation}
For two lossless media with no absorption at the boundary,
\begin{equation}
\boxed{R+T=1.}
\end{equation}
This is simply energy conservation applied to the interface.
\section{Example: air into a dielectric with relative permittivity 4}
Let medium 1 be air, approximated as free space,
\begin{equation}
\eta_1=\eta_0,
\end{equation}
and let medium 2 be a nonmagnetic lossless dielectric with
\begin{equation}
\epsilon_{r2}=4,
\qquad
\mu_{r2}=1.
\end{equation}
Its intrinsic impedance is
\begin{equation}
\eta_2=\sqrt{\frac{\mu_0}{4\epsilon_0}}=\frac{\eta_0}{2}.
\end{equation}
Thus
\begin{equation}
\Gamma
=\frac{\eta_0/2-\eta_0}{\eta_0/2+\eta_0}
=-\frac{1}{3},
\end{equation}
so
\begin{equation}
\boxed{\Gamma=-0.333.}
\end{equation}
The reflected electric field has one-third the incident amplitude and is phase reversed.
The transmitted electric-field coefficient is
\begin{equation}
\tau=1+\Gamma=\frac{2}{3}.
\end{equation}
The reflected power fraction is
\begin{equation}
\boxed{R=\frac{1}{9}\approx0.111.}
\end{equation}
The transmitted power fraction is
\begin{equation}
T
=\frac{\eta_0}{\eta_0/2}
\left(\frac{2}{3}\right)^2
=\frac{8}{9},
\end{equation}
so
\begin{equation}
\boxed{T\approx0.889.}
\end{equation}
Notice the important distinction: $\tau=2/3$ is a \emph{field-amplitude} ratio, not a power ratio.
\section{Reflection from a lossy medium}
The same normal-incidence amplitude formula remains valid when medium 2 has complex intrinsic impedance:
\begin{equation}
\boxed{
\Gamma=\frac{\eta_2-\eta_1}{\eta_2+\eta_1}.
}
\end{equation}
Now $\Gamma$ is generally complex. Its magnitude gives the reflected field-amplitude fraction and its argument gives the reflection phase shift:
\begin{equation}
\Gamma=|\Gamma|e^{i\phi_\Gamma}.
\end{equation}
The transmitted wave also attenuates after entering medium 2 because
\begin{equation}
E_t(z)\propto e^{-\alpha_2 z}.
\end{equation}
Thus two physically distinct effects must be separated:
\begin{equation}
\boxed{
\text{boundary reflection}
\quad\text{versus}\quad
\text{propagation loss after transmission}.
}
\end{equation}
For a general lossy medium at normal incidence, the time-average power flux should be computed from
\begin{equation}
\langle\mathbf S\rangle
=\frac{1}{2}\operatorname{Re}
\left(\mathbf E\times\mathbf H^*\right),
\end{equation}
rather than by blindly using the lossless expression $|E|^2/(2\eta)$ with a complex $\eta$.
\section{Why refraction occurs at oblique incidence}
Now let the incident wave strike the boundary at angle $\theta_i$ measured from the normal. The reflected and transmitted waves make angles $\theta_r$ and $\theta_t$.
At every point along the interface, the boundary conditions must hold for all time. Therefore the incident, reflected, and transmitted waves must have the same phase variation parallel to the interface. If the interface lies along $x$, phase matching requires
\begin{equation}
\boxed{
\beta_1\sin\theta_i
=\beta_1\sin\theta_r
=\beta_2\sin\theta_t.
}
\end{equation}
The first equality gives
\begin{equation}
\boxed{\theta_r=\theta_i.}
\end{equation}
The second gives
\begin{equation}
\beta_1\sin\theta_i
=\beta_2\sin\theta_t.
\end{equation}
For lossless isotropic media,
\begin{equation}
\beta=\frac{n\omega}{c},
\end{equation}
so
\begin{equation}
\boxed{
n_1\sin\theta_i=n_2\sin\theta_t.
}
\end{equation}
This is Snell's law, obtained directly from electromagnetic phase continuity along the boundary.
\begin{center}
\includegraphics{EM21_fig03_oblique_refraction.png}
\vspace{0.4em}
\textit{Figure 3. Oblique incidence requires equal tangential phase variation along the interface. This gives $\theta_r=\theta_i$ and Snell's law.}
\end{center}
\section{TE and TM polarizations at an interface}
At oblique incidence, polarization matters because the electric and magnetic fields project differently onto the interface.
\subsection{TE polarization}
For transverse-electric (TE, or $s$) polarization, the electric field is perpendicular to the plane of incidence. Applying the tangential $E$ and $H$ boundary conditions gives
\begin{equation}
\boxed{
\Gamma_{\text{TE}}
=\frac{\eta_2\cos\theta_i-\eta_1\cos\theta_t}
{\eta_2\cos\theta_i+\eta_1\cos\theta_t}.
}
\end{equation}
The corresponding electric-field transmission coefficient is
\begin{equation}
\boxed{
\tau_{\text{TE}}
=\frac{2\eta_2\cos\theta_i}
{\eta_2\cos\theta_i+\eta_1\cos\theta_t}.
}
\end{equation}
\subsection{TM polarization}
For transverse-magnetic (TM, or $p$) polarization, the magnetic field is perpendicular to the plane of incidence. The electric-field reflection coefficient is
\begin{equation}
\boxed{
\Gamma_{\text{TM}}
=\frac{\eta_2\cos\theta_t-\eta_1\cos\theta_i}
{\eta_2\cos\theta_t+\eta_1\cos\theta_i}.
}
\end{equation}
The electric-field transmission coefficient is
\begin{equation}
\boxed{
\tau_{\text{TM}}
=\frac{2\eta_2\cos\theta_i}
{\eta_2\cos\theta_t+\eta_1\cos\theta_i}.
}
\end{equation}
At normal incidence,
\begin{equation}
\theta_i=\theta_t=0,
\end{equation}
and both TE and TM reflection coefficients reduce to
\begin{equation}
\Gamma=\frac{\eta_2-\eta_1}{\eta_2+\eta_1}.
\end{equation}
\section{Power coefficients at oblique incidence}
For lossless media, the normal component of average power flow is proportional to
\begin{equation}
\frac{|E_0|^2}{2\eta}\cos\theta.
\end{equation}
Therefore, for either TE or TM polarization,
\begin{equation}
\boxed{R=|\Gamma|^2}
\end{equation}
and
\begin{equation}
\boxed{
T
=\frac{\eta_1}{\eta_2}
\frac{\cos\theta_t}{\cos\theta_i}
|\tau|^2.
}
\end{equation}
For lossless interfaces,
\begin{equation}
R+T=1.
\end{equation}
The cosine factor appears because the relevant conserved quantity is power crossing the interface, not power along the slanted propagation direction.
\section{Brewster angle}
A special feature appears for TM polarization. At one incidence angle the numerator of $\Gamma_{\text{TM}}$ can vanish:
\begin{equation}
\Gamma_{\text{TM}}=0.
\end{equation}
For nonmagnetic lossless media, this occurs at the Brewster angle
\begin{equation}
\boxed{
\tan\theta_B=\frac{n_2}{n_1}.
}
\end{equation}
At $\theta_i=\theta_B$, an ideal TM-polarized plane wave has zero reflected amplitude.
For air to glass with
\begin{equation}
n_1=1,
\qquad
n_2=1.5,
\end{equation}
\begin{equation}
\theta_B=\tan^{-1}(1.5)\approx56.3^\circ.
\end{equation}
This effect is polarization selective: the TE reflection coefficient does not vanish at the same angle.
\section{Critical angle and total internal reflection}
Suppose the wave travels from a higher-index medium into a lower-index medium,
\begin{equation}
n_1>n_2.
\end{equation}
Snell's law requires
\begin{equation}
\sin\theta_t=\frac{n_1}{n_2}\sin\theta_i.
\end{equation}
At a sufficiently large $\theta_i$, the right side would exceed 1 if interpreted as an ordinary propagating transmitted angle. The limiting incidence angle is the critical angle,
\begin{equation}
\boxed{
\theta_c=\sin^{-1}\left(\frac{n_2}{n_1}\right).
}
\end{equation}
For
\begin{equation}
\theta_i>\theta_c,
\end{equation}
there is no propagating transmitted plane wave carrying power away from the boundary in the usual refracted direction. Instead, the field in medium 2 is evanescent in the normal direction while the reflection magnitude becomes
\begin{equation}
\boxed{|\Gamma|=1.}
\end{equation}
The phase of $\Gamma$ remains nontrivial even though all time-average normal power is reflected. This phenomenon is total internal reflection.
\section{Fresnel reflectance versus angle}
For air-to-glass incidence, $n_1=1$ and $n_2=1.5$. The TE and TM reflection coefficients behave differently as the incidence angle grows. The TM reflectance falls to zero at the Brewster angle, while the TE reflectance rises monotonically toward unity near grazing incidence.
\begin{center}
\includegraphics{EM21_fig04_fresnel_reflectance.png}
\vspace{0.4em}
\textit{Figure 4. Fresnel power reflectance for an air-to-glass interface. TM reflection vanishes at the Brewster angle, while both polarizations approach total reflection near grazing incidence.}
\end{center}
\section{Standing-wave structure on the incident side}
When a reflected wave is present, the total electric field in medium 1 is
\begin{equation}
E_1(z)=E_{i0}e^{-i\beta_1z}+E_{r0}e^{+i\beta_1z}.
\end{equation}
Using
\begin{equation}
E_{r0}=\Gamma E_{i0},
\end{equation}
we obtain
\begin{equation}
\boxed{
E_1(z)=E_{i0}
\left(e^{-i\beta_1z}+\Gamma e^{+i\beta_1z}\right).
}
\end{equation}
The reflected and incident fields interfere, producing spatial maxima and minima. For a lossless medium and fixed $|\Gamma|<1$,
\begin{equation}
|E|_{\max}=|E_{i0}|(1+|\Gamma|),
\end{equation}
\begin{equation}
|E|_{\min}=|E_{i0}|(1-|\Gamma|).
\end{equation}
Their ratio is
\begin{equation}
\boxed{
\mathrm{SWR}
=\frac{1+|\Gamma|}{1-|\Gamma|}.
}
\end{equation}
This same mathematical structure later reappears in transmission-line voltage standing-wave ratio (VSWR). The transmission-line result is not a coincidence: it is another manifestation of forward and backward waves meeting an impedance discontinuity.
\section{Several important limiting cases}
\subsection{Matched boundary}
If
\begin{equation}
\eta_1=\eta_2,
\end{equation}
then at normal incidence
\begin{equation}
\Gamma=0,
\qquad
R=0.
\end{equation}
No reflection occurs even though other material properties may differ.
\subsection{Perfect electric conductor}
For an ideal perfect electric conductor, the tangential electric field at the surface must vanish. The reflected field therefore cancels the incident tangential electric field at the boundary:
\begin{equation}
E_{r0}=-E_{i0}.
\end{equation}
Hence
\begin{equation}
\boxed{\Gamma_{\text{PEC}}=-1,}
\qquad
\boxed{R=1.}
\end{equation}
This is complete reflection with a $180^\circ$ electric-field phase reversal.
\subsection{Very good but finite conductor}
A good conductor has a small complex intrinsic impedance compared with free space. Therefore
\begin{equation}
\Gamma\approx-1,
\end{equation}
but not exactly. A small transmitted field enters the conductor and then decays over the skin depth developed in EM20.
\section{The boundary-wave workflow}
A reliable interface calculation can be organized as follows:
\begin{enumerate}
\item determine $\epsilon$, $\mu$, and $\sigma$ in each medium;
\item compute $\gamma$ and $\eta$ as needed;
\item identify the incidence angle and polarization;
\item enforce tangential phase matching to obtain reflection and refraction angles;
\item apply the appropriate TE or TM Fresnel coefficient;
\item convert field-amplitude coefficients into power coefficients with the correct impedance and angle factors;
\item if a transmitted medium is lossy, propagate the transmitted field with $e^{-\alpha z}$ after the interface.
\end{enumerate}
This workflow keeps three distinct physical mechanisms separate:
\begin{equation}
\boxed{
\text{reflection at the boundary},
\qquad
\text{refraction at the boundary},
\qquad
\text{attenuation inside the material}.
}
\end{equation}
\section{Connection to RF systems and antennas}
Boundary-wave physics appears throughout RF engineering. Examples include radomes, dielectric antenna substrates, soil and seawater propagation, building penetration, microwave windows, coaxial and waveguide discontinuities, and the transition from an antenna's near environment into free space.
The reflection coefficient also foreshadows impedance matching in circuit and transmission-line language. At normal incidence,
\begin{equation}
\Gamma=\frac{\eta_2-\eta_1}{\eta_2+\eta_1}
\end{equation}
has exactly the same algebraic form as the voltage reflection coefficient at a transmission-line load,
\begin{equation}
\Gamma_L=\frac{Z_L-Z_0}{Z_L+Z_0}.
\end{equation}
The common structure arises because both problems enforce continuity relations between paired wave variables and their impedance ratios.
\section{What EM21 adds to the series}
EM20 described propagation within one uniform medium. EM21 adds the physics of crossing between media. The key results are
\begin{equation}
\boxed{
\Gamma=\frac{\eta_2-\eta_1}{\eta_2+\eta_1},
\qquad
\tau=\frac{2\eta_2}{\eta_1+\eta_2}
}
\end{equation}
for normal incidence,
\begin{equation}
\boxed{R=|\Gamma|^2,}
\end{equation}
\begin{equation}
\boxed{n_1\sin\theta_i=n_2\sin\theta_t,}
\end{equation}
and the TE/TM Fresnel coefficients for oblique incidence.
These results establish the mathematical foundation for multilayer propagation, transmission lines, standing waves, impedance matching, waveguides, radomes, and antenna-to-medium coupling. The natural companion is an EM21E1 problem set devoted to boundary-condition derivations, Fresnel coefficients, Brewster angle, total internal reflection, and power bookkeeping.
\section*{References}
\begin{thebibliography}{99}
\bibitem{Griffiths2017}
D. J. Griffiths,
\emph{Introduction to Electrodynamics}, 4th ed.,
Cambridge University Press, 2017.
\bibitem{Jackson1999}
J. D. Jackson,
\emph{Classical Electrodynamics}, 3rd ed.,
Wiley, 1999.
\bibitem{Cheng1989}
D. K. Cheng,
\emph{Field and Wave Electromagnetics}, 2nd ed.,
Addison-Wesley, 1989.
\bibitem{Ulaby2015}
F. T. Ulaby and U. Ravaioli,
\emph{Fundamentals of Applied Electromagnetics}, 7th ed.,
Pearson, 2015.
\bibitem{Balanis2012}
C. A. Balanis,
\emph{Advanced Engineering Electromagnetics}, 2nd ed.,
Wiley, 2012.
\bibitem{Pozar2012}
D. M. Pozar,
\emph{Microwave Engineering}, 4th ed.,
Wiley, 2012.
\end{thebibliography}