noc19-ee21 Lecture 14-Uniform Plane Waves-I

IIT Kanpur July 201
20 followers
Dec 3, 2023
Uniform plane wave, electric flux density, isotropic material, homogeneous isotropic medium, oscilloscope, Laplacian, Cartesian coordinate system.
Hello and welcome to NPTEL MOOC on Electromagnetic Waves in Guided and Free Space or Wireless
Media.
In this module, which is module number 14, we will look at one-dimensional uniform plane
waves.
To begin with, we recall the following Maxwell's equations, which we have described in the
previous module.
We have del cross E is equal to minus del B by del t, right, where we have already seen
that electric field E is actually a function of both space, which is represented by the
position vector r as well as time t.
Similarly, all the field quantities are typically function and time.
Of course, if the fields are independent of time or if the fields, you know, are not varying
with respect to time, then you get what is called as electrostatic fields.
And in those fields, the fields will be functions only of the space coordinates, which is represented
by the position vector r.
In this module and in our course, we will assume that we are not dealing with electrostatic
case.
Of course, that is how you will actually get the waves to be generated and propagated.
The other equation that will be of interest for us is del cross H is equal to J, which
is the conduction current density, which is the result of the current that is because
of the free charge carriers that are, you know, moving around in the particular material,
which we consider, as well as the displacement current density, which is given by del D by
del t.
As before, the magnetic field H is also a function of space and time.
So is the displacement or the flux density, electric flux density, which is also a function
of space and time.
The other two equations that are of interest that we already have seen is del dot D equals
rho V and del dot B is equal to 0.
And we have also seen the relationships between D and B. D is basically epsilon 0 and in a
medium which is described to be linear, homogeneous, okay, and isotropic material for which vacuum
or free space or air is an example, sorry, this is linear, homogeneous, isotropic medium,
okay.
So for such a medium, we can characterize D and E by a number called as relative permittivity
epsilon r, where epsilon 0 is the absolute permittivity or the permittivity of the free
space, okay.
This epsilon r is a number at least for the lossless case, so we can even add lossless
as an additional constraint.
In that case, epsilon r is a number which is greater than 1 usually, so that we can
describe D and E by a simple proportionality relationships, okay.
Similarly, B in this course will always be equal to mu 0 times H. In a magnetic material,
you can also introduce what is called as magnetic permeability, but we are not going to do that
because we will consider exclusively only those medium in which the medium is non-magnetic,
okay.
So, we have these 4 equations with us, these are the Maxwell's equations in differential
form and using these equations, let us see if it is possible to describe the wave propagation,
okay.
Now, by what we actually mean, what do we mean by a wave?
We have seen waves on a transmission line, right.
So we had a transmission line which extended to either infinite length in the ideal scenario
or in a practical case, it extended to a finite length.
Nevertheless, what we found was that the voltage at any position, right, the voltage difference
between these 2 conductors at any position is actually or any position and time is in
the form of a propagating wave.
So, V of zt was any function assuming that I am only looking at the forward propagating
wave was any function which had this argument t-z by V, where V is the velocity of propagation
of this voltage, right.
For the simple case of a sinusoid, this V plus was a function which is like sin or cosine
and the argument of this one was t-z by V, instead of t-z by V, we also wrote the argument
as omega t-beta z and then we related omega and beta to the velocity of propagation V,
right.
So, omega by beta was actually equal to V and what we meant by wave, so if you actually
hook up an oscilloscope here, okay, so this is an oscilloscope and then you hook up another
oscilloscope at a certain distance which is greater than, which is at a distance further
away from the initial position, so you can call this as plane z1 and another this plane
as z2 and then you looked out what kind of a waveform would be displayed assuming a general
arbitrary V plus kind of a function.
If this was the wave that you saw at z1, the corresponding wave that you would have seen
or the corresponding voltage waveform that you would have seen would be displaced by
a distance that is actually proportional to these 2 or in terms of time, it would be whatever
the distance that has been that is the difference between these 2 planes divided by the velocity
that you had would be the amount of delay which we will call as say dt by which this
particular pulse would be delayed, right.
So, it is conceivable that if you actually take a you know, very very long transmission
line and then start hooking up imaginary oscilloscopes at every point, okay, we have considered lossless
transmission line and then you will start noting down the voltages, a pattern begins
to emerge, what is the pattern?
That the whatever the function that may be there, you know V plus of t – z by V that
would be progressively delayed as you keep moving along the transmission line, so this
is in fact the behaviour of a propagating wave, okay, we have also seen a different
kind of a wave that is called as a standing wave in which you would have a forward propagating
voltage hitting upon some discontinuity for example, that could be a load whose characteristic
impedance sorry, load whose impedance would be different from the characteristic impedance
of the transmission line and it would generate a reflected wave, right.
So, these 2 waves would you know combine together and it would not be moving so much but at
any particular position if you were to stand, you would actually see the amplitude to be
changing, right but these are essentially what is called as standing waves which we
will not concern consider now, what we are interested is the progressive waves or the
waves which are propagating, I arbitrarily assume that the propagation is along z direction
for the transmission line case, we will continue to make that assumption although there is
nothing in space that would tell that my z direction should coincide with your z direction
but what is important is that if you pick a particular direction, in that direction
there should be a propagation type of a behaviour, right.
So, you imagine putting up oscilloscopes and somehow being able to look at electric fields
or magnetic fields, these electric and magnetic fields should exhibit a behaviour which is
given mathematically by this function V plus of t – z by V, so that is essentially what
we mean by a wave, more ordinarily you may have done lot of experiments you know, you
take a string, tie it up on to one end and then you actually start doing this you know,
moving behaviour of the other end, the free end of the string and then you would actually
visualise that the string is actually going up and down and there is some sort of a waviness
into that string, right.
So, there are these different types of waves for example, seismic waves are the waves that
are generated because of the plate movement you know, I am not an expert but those are
essentially also type of a waves because they would also move, right, so any of this
phenomenon which has this function of t – z by V kind of a behaviour would qualify for
a wave, please remember that it is not only some function of t – z by V, it could be
t plus z by V, it could be t minus x by V or it could be any general direction that
the wave could be propagating, okay.
However, coming back to this electromagnetic waves, what makes it very different from the
other kinds of waves is that these electromagnetic waves can travel in vacuum as well because
there are no material charges pushing this wave, right, so in contrast to a sound wave
which requires the particles to be pushed and pulled sort of a elastic motion that you
would actually see, there is nothing like that that is required for an electromagnetic
wave.
So, a moving electric field would generate a moving magnetic field or rather a time varying
electric field would generate a time varying magnetic field which in turn would generate
a time varying electric field and magnetic field and these couplings can go on in create
a propagating wave as we will shortly see.
So, we have to understand that these waves when they propagate in free space, they actually
have a different velocity whereas, if the waves propagate in a material for example,
light waves moving in a small slab of glass would have a different velocity, right.
In most cases, that velocity is governed by epsilon r, in some rather very specific cases,
the velocity is also governed by other characteristics when epsilon r itself becomes complex, okay
but that story is for something later.
So, we will concentrate on the simple scenario where we are going to consider a medium to
be linear, homogeneous, isotropic as well as lossless, okay and for this medium, the
characteristic of the media is given by specifying epsilon r as well as mu0, epsilon0 anyway
is already defined, okay.
This is decidedly non-magnetic medium, so you can even attach a minus or a hyphen and
say Nm where Nm would stand for a non-magnetic medium, okay.
Our starting point at least from the math perspective would be to go back to this equation
del cross E is equal to – del B by del t, okay and then replace B with H that I can
do because I already know what is the relationship between B and H and I also know that mu0 is
a constant, so I can pull that out of the differentiation and therefore, partial derivative
and then I have – mu0 del H by del t.
What I do now is to take the curl of the equation again, okay, so I am going to take the curl
of this first equation which is Faraday's law and what do I get?
I will get – mu0 del cross del H by del t, okay, right.
Now without going into lot of mathematical justification, I will simply interchange this
operation of curl with partial derivative, okay.
I am allowed to do this under no certain special conditions which you can read about
in any math textbook but when I do that, what do I have?
I have – mu0 del by del t del cross H, okay but I know what is del cross H, there is another
equation which tells me that del cross H is equal to, so you can fill up that equation
which would be J plus del D by del t, okay.
Let us consider the first term, I have del J by del t, J is because of the conduction
current, so which requires that I have no free charges plus or minus whatever the type
of charges that are possible and these charges have to physically move in order to constitute
the J field or the conduction current density field, okay.
But my medium is a complete insulator, there is no free charges anywhere, the medium also
extends all the way to infinity everywhere that you can think of, the medium extends
all the way to infinity and it is only specified by the parameters epsilon 0, mu 0 and epsilon
r, okay and epsilon r is also real quantity, so clearly there is no site of any free charges
and therefore, there should not be any conduction current density, right.
So this del J by del t term readily goes to 0 and you can eliminate it and you can now
consider the second term which is del square by del t square, why did it become second
partial derivative?
Because there is a del by del t here which goes on to another del by del t, therefore
this becomes a second partial derivative with respect to D, okay or rather of the quantity
D, right but we already know that D can be written as epsilon 0, epsilon r both of which
are assumed to be constants, so you can pull them out of the partial derivative and then
write here as del square E by del t square.
Now, let us complete the left hand and the right hand side equations after this simplification,
so del cross del cross E which is still unknown, we do not know what exactly to make out of
this quantity, okay but the right hand side at least is now simplified, you will get – mu
0 epsilon 0 epsilon r del square E by del t square, at least this is some sort of an
okay thing, right on the right hand side, I have a function of E alone, right and on
the left hand side presumably, I have a function of E alone, right because this is a curl operation
on E that depends only on the electric field components, taking the curl of E would also
depend only on the electric field components.
So, on both sides, I have an electric field component and an electric field component
or rather functions of electric field and function of electric field, okay now, what
do we do about the left hand side, okay, I am going to erase the equation right here
because I want to preserve this equation that we have already written and discuss the meaning
of the left hand side term, okay.
What is the term?
You have a curl of curl of E, right now, one can actually go to specific coordinate system
that you are talking about for example, in the Cartesian coordinate system, this del
cross E could be written as x hat, y hat, z hat, these are the unit vectors along the
coordinate system which is given by x, y and z, x, y and z are 3 mutually perpendicular
lines or axis and any point on this one can be specified by giving the 3 points x, y and
z or the corresponding vector OP or the position vector P can be given by this particular quantity
x, x hat, y, y hat plus z, z hat, this is something that you already know, so curl of
E in this coordinate system would be x hat, y hat, z hat, del by del x, del by del y,
del by del z and please remember that electric field E is also a vector, so it will be E
x, E y and E z and each of these E x, E y and E z themselves are a functions of the
position vector and in the coordinate system that we considered, the position vector can
be specified by giving the components of the position vector which are x, y and z, right,
so you have x, y and z, t that would be that is each E x, E y and E z will be a function
how and what we still do not know but it would be a function of x, y, z and t, right, so
this is what you would have for the curl of E and after you have evaluated this curl
of E, you will actually have the curl expressions, you can put them back into this expression
of this curl and evaluate the complete left hand side, okay.
But there is a small shortcut for us which makes use of the vector identities, okay,
what is the shortcut that means, the shortcut is actually the rule which says that del cross
del cross E is actually gradient of divergence of E – del square E, okay and this del square
is called as the vector or simply sometimes called as the Laplacian, okay and of course,
you have del dot E and gradient of this one, does it make sense, you know in terms of the
vectors, yes, curl of electric field will result in a vector, taking the curl of a vector
will result in a vector which is fine, this del square is a scalar operation but because
it operates on a vector, this is sometimes called as vector Laplacian and this del square
of E results in a vector, good, del dot E will result in a scalar but taking the gradient
of a scalar will get back the vector, so this equation makes sense at least.
So, now with that let us also look at another equation, we had del dot D equal to rho V,
right, now we said that the medium is infinite and all that there are no free charges, if there
are no free charges, no conduction current density J, there is also no you know free
charges rho V itself, right, so del dot D equal to rho V simply becomes equal to 0 because there
are no free volume charges or free charges. Now, D is related to epsilon 0, epsilon r and
E or rather related to E via epsilon 0 and epsilon r which when you put them here and
realise that this epsilon 0, epsilon r is a constant, okay which can be pulled out side
the divergence operation, you will see that epsilon 0, epsilon r del dot E equal to 0,
the only way you can have this equation to be valid is when either epsilon r equal to 0 but
we have ruled out that possibility or this del dot E itself equal to 0 which we will readily accept,
okay. So, I have in the vector Laplacian del of del dot E – del square E,
del dot E equal to 0 under this particular medium, okay, for a different medium this may not be true,
in fact as we will see for waveguides, this equation is not true in general, okay but
luckily for us, we are dealing with this kind of a medium which is LHILNM medium and for which
this left hand side can simply be written as – del square E, now the right hand side also
has a minus sign, so I am going to remove the minus sign from both left as well as the right
hand side terms and then rearrange this mu0 epsilon 0, epsilon r and call it as 1 by up square
and del square E by del t square, okay, please note that this equation is true for E which is
you know itself consists of Ex, Ey and Ez components, okay and in the rectangular Cartesian
coordinate system, you can take this equation which we have written separately into 3 scalar
equations, okay, what we mean by that is I can have this equation decomposed into x,
y and z terms as del square Ex which is still a function of x, y, z and t to be equal to 1
by up square del square Ex by del t square, Ex of course is a function of x, y, z and t and of
course, you can now write 2 similar equations for Ey and Ez, okay and what is this up?
Up of course is now given by 1 by square root of mu0 epsilon 0 epsilon r actually
corresponds to the phase velocity, the meaning of phase will come back later on,
we will come to that later on but it is essentially the velocity with which the
wave is actually moving, okay of course, we have not established that this is exactly the wave
solution, we will do so shortly but in anticipation of the fact that we are dealing with waves,
I just call this as up to be the phase velocity, okay. As I have told you, you can write down a
similar equation for Ey and Ez and everything Ex, Ey and Ez will all be functions of x, y,
z and t and it will satisfy a similar equation. I will give you a short exercise to show that
not only the electric field satisfies this equation, even the magnetic field you know,
H would also satisfy the same equation, okay, it would be given by 1 by up square del square H by
del t square, the development of this equation is very simple, you start off with del cross H given
by the right hand side which you now can fill up, take the curl of curl of H, okay and you will get
the right hand side and show that the left hand side reduces to only – del square H and on the
right hand side, you will have – 1 by up square del square H by del t square, cancel off the minus
and then you will get this equation, so I encourage you to do this exercise just to get you know,
a kind of a mathematical you know, hold on the mathematical identities that we have used in
deriving this equation. We consider first only this equation, okay to talk about the further
development, what I have here on the left hand side is this Laplacian. Now, I have already made
my choice of coordinate system to be rectangular Cartesian coordinate system, in that coordinate
system, this del square can be written as del square by del x square plus del square by del y
square plus del square by del z square, okay. So, let us write down del square by del x square
plus del square by del y square plus del square by del z square, this entire thing you know acting
on Ex which itself is a function of all these 4 variables being equal to 1 by up square del
square Ex which is a function again of these 4 coordinates or rather 4 variables times del t
square, okay. What you observe is that the left hand side is a function which is only of space
changing like the derivatives of z, x and y tell you how the space derivatives of the electric
field component Ex would be there and on the right hand side, you have a time derivative,
second time derivative and this type of an equation where on the left hand side,
you would have seen the space derivative and on the right hand side, you would have seen the time
derivative is something that you would have seen in the transmission line.
So, if you recall the transmission line equations for the voltage, there we had del square V by del
z square equals del square V by del t square of course, you still had this 1 by up square kind of
an equation there, okay except that there up was actually equal to 1 by square root Lc and this
was the case for the transmission line which was lossless, right and it had a uniform cross section.
So, you have seen this equation, so this equation you know is going to give you a wave, right. So,
when you choose mathematically the function V, then the equation can be satisfied, the general
equation that would satisfy would be either V plus of t minus z by up or it would be V sorry,
V plus of t minus z by up or it would be V minus of t plus z by up, you have already seen that
this equation is going to give you one dimensional wave which is basically to say that it is a wave
which is propagating either along the plus z direction or along the minus z direction but
it is going to be a wave. Now, if you compare this equation with the previous equation for
the electric field that we have written, Ex component that we have written, you will see
that in addition to this del square by del z square terms, you also have 2 additional terms,
one is del square by del x square and then you have del square by del y square.
If you can remove these 2 terms, right then the equation would be identical to whatever the
voltage equation on a transmission line is, right. Let us mathematically take it,
okay. We will worry about how to generate this later on in some other you know module, okay.
However, nothing compels us to stop taking this del square by del x square and del square by del y
square terms to be 0, meaning that what I am assuming is that Ex is not a function of x,
it is not a function of y, it is simply a function only of z and t, okay. As I have told you,
I still have not you know exactly told you, right that how we are going to make this x and y
dependence go away but take it mathematically that you can always do this, right.
And when you do this, you know assume that Ex is going to be just a function of z and t,
then you will land up in a very interesting scenario saying that you have del square Ex
by del z square where Ex is now a function of z and t to be equal to 1 by Up square del square
Ex by del t square, Ex is a function of z and t and any such solution on Ex which would be in
the form of Ex of z and t which would be in the form of some E plus of t minus z by Up or E minus
t plus z by Up are potentially the solutions for this equation, okay.
The story is not complete yet because by following the same logic, I can show that even Ey will
satisfy the same equation and then Ez will also satisfy the same equation, okay but we will later
on see that I cannot have all 3 of them satisfying this type of an equation because there is another
constraint called del dot E equal to 0 which forces something else to happen and we are going
to consider that one in the next, sorry that we are going to consider that one shortly but for
now we know that Ex satisfies this equation and you can also put down Ey which satisfies
the same equation. Now, let us come back to this H case, the equation we had was del square H
equal to 1 by Up square del square you know H by del t square is what we had.
Similarly, you will have Hx and Hy, okay. You will have all these equations satisfying very
similar wave equation and therefore have the forms to be similar. Now, if you go back to
the transmission line analogy and then think of a certain transmission line which is propagating
along z or rather transmission line which is lying along z and then choose one of the axis
to be the x axis, right and then say the potential difference is going to be because of the Ex
component, then the voltage V of zt on the transmission line is analogous to the voltage
or rather the component Ex which is propagating as a function of z and t as well.
So, in fact mentally you can imagine that there is a transmission line which is associated with
the potential difference V which is analogous to Ex component, okay. So, Ex propagating along z can
be associated with the transmission line, yeah uniform lossless transmission line. But now you
may ask, well the transmission line not only has the voltage, it also has the current. So,
what shall we do about the current component or the corresponding component of the current there?
Well, we can show that by writing down this del cross E equations that there is a very natural
pairing of Ex and Hy, okay. You can treat this Ex as the voltage V and Hx as the current I on
the transmission line. So, this one pair Ex and Hy both propagating along z, you know with a
given velocity Up can be thought of as having a transmission line or analogous to a transmission
line along the z axis. Now, this is not the whole story because you can find similarly Ey and –Hx,
okay. I just put –Hx for a reason that will come out later on in the other module.
But I will now go ahead with that one. Ey and –Hx will also be associated with the same
transmission line z, okay. So, these pairings Ex Hy both travelling along the, I mean both being
the components of the waves which is propagating along the z axis as well as this Ey and –Hx both
can be associated mentally and formally one can show that it is true that they can be associated
with a uniform lossless transmission line, okay. Now, what about the Ez component? Well,
we have this condition that del dot E equal to 0 and if you go back to the Cartesian coordinate
system, this translates to del Ex by del x plus del Ey by del y plus del Ez by del z, okay.
This would be equal to 0, okay. However, what we are going to assume,
what we have already assumed is that Ex is not a function of x, Ey is not a function of y. So,
that leaves us only with Ez, del Ez by del z equal to 0 which you know the simplest solution
for this case would be to make Ez itself equal to 0, okay. So, now our wave for the electric
field would have only components Ex and Ey which both will be functions of z and t.
Similarly, you will have Hx and Hy, okay because del dot H is also equal to 0 and
these 4 nonzero components together constitute a wave which is propagating along the z direction
but because we have paired them like Ex and Hy, Ey and –Hx, you can treat these 2 as 2
subcomponents, okay and you know think of this as one type of a wave and the other one as another
type of a wave. So, one can be thought of as the x polarized wave, the other can be thought of as
the y polarized wave, both polarized waves propagating along the z axis. We will continue
the discussion in the next module. Thank you very much.
