Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz...

29
9/17/2019 1 Advanced Computation: Computational Electromagnetics Transfer Matrix Method (TMM) Outline One‐dimensional structures in electromagnetics Formulation of 44 matrix equation for 1D structures Solution in an LHI layer Transfer matrices for multilayer structures Stability of transfer matrices Formulation of 22 matrix equation for 1D structures Slide 2 1 2

Transcript of Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz...

Page 1: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

1

Advanced Computation:

Computational Electromagnetics

Transfer Matrix Method (TMM)

Outline

• One‐dimensional structures in electromagnetics

• Formulation of 44 matrix equation for 1D structures

• Solution in an LHI layer

• Transfer matrices for multilayer structures

• Stability of transfer matrices

• Formulation of 22 matrix equation for 1D structures

Slide 2

1

2

Page 2: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

2

Slide 3

One‐Dimensional Structures in Electromagnetics

1D Structures

Slide 4

Sometimes it is possible to describe a physical device using just one dimension.  Doing so dramatically reduces the numerical complexity of the problem and is ALWAYS GOOD PRACTICE.

z

x

y

Region IReflection Region

Region IITransmission Region

3

4

Page 3: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

3

3D  1D Using Homogenization

Slide 5

Many times it is possible to approximate a 3D device in one dimension.  It is very good practice to at least perform the initial simulations in 1D and only moving to 3D to verify the final design.

1 2 3 4 r

Physical Device Effective Medium Approximation in 1D

3D  1D Using Circuit‐Wave Equivalence

Slide 6

r, r,i i in

ii i

i

Z

5

6

Page 4: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

4

Slide 7

Formulation of 44 Matrix Equationfor 1D Structures

Starting Point

Slide 8

0 r

0 r

0 r

yzx

x zy

y xz

HHk E

y z

H Hk E

z x

H Hk E

x y

0 r

0 r

0 r

yzx

x zy

y xz

EEk H

y z

E Ek H

z xE E

k Hx y

Start with Maxwell’s equations in the following form.  Here, isotropic materials are assumedand the positive sign convention is used for waves.

0H j H Positive sign convention

7

8

Page 5: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

5

Calculation of the Wave Vector Components

Slide 9

The components kx and ky are determined by the incident wave and are continuous throughout the 1D device.  The kz component is different in each layer and calculated from the dispersion relation in that layer.

0 r,inc r,inc

0 r,inc r, c

,inc

,inc in

sin cos

sin sin

x

y

x

y

k k

k

k

k k

2 2 2, 0 r, r,z i i i x yk k k k

Layer #i

kx and ky Continuous Throughout Device

Slide 10z

xinck

kx

refk -kz,air

kx

kz,air 22 2

,air 0 airz xk k n k ,inc 0 ai

,inc a

r

0 ir

cos sin

cosz

x

k k n

k k n

kx

kz,11k

22 2,1 0 1z xk k n k

1n

2n

3n

kx

kz,22k

22 2,2 0 2z xk k n k

kx

kz,33k

22 2,3 0 3z xk k n k

kx

kz,airtrnk

22 2 2,trn 0 air ,airz x zk k n k k

9

10

Page 6: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

6

Waves in Homogeneous Media

Slide 11

0 0 0 0 y yx xz zjk y jk yjk x jk xjk z jk zjk r jk rE r E e E e e e H r H e H e e e

A wave propagating in a homogeneous layer is a plane wave.  It has the following mathematical form.

0 0 y yx xz zjk y jk yjk x jk xjk z jk z

x x xE r E e e e jk E e e e jk E r jkx x x

When derivatives of these solutions are calculated, we see that

0 0 y yx xz zjk y jk yjk x jk xjk z jk z

y y yE r E e e e jk E e e e jk E r jky y y

Note: e+jkz sign convention was used for propagation in +z direction.

It cannot be said that                   because the structure is not homogeneous in the z direction.

zz jk zjk

z

Reduction of Maxwell’s Equations to 1D

Slide 12

0 r

0 r

0 r

yy z x

xx z y

x y y x z

dHjk H k E

dz

dHjk H k E

dz

jk H jk H k E

0 r

0 r

0 r

yy z x

xx z y

x y y x z

dEjk E k H

dzdE

jk E k Hdz

jk E jk E k H

Given that

x yjk jkx y

Maxwell’s equations become

Note: z is the only independent variable left so its derivative is ordinary.

d

z dz

0 r

0 r

0 r

yzx

x zy

y xz

HHk E

y z

H Hk E

z x

H Hk E

x y

0 r

0 r

0 r

yzx

x zy

y xz

EEk H

y z

E Ek H

z xE E

k Hx y

11

12

Page 7: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

7

Normalize the Parameters

Slide 13

Normalize the coordinates (x, y, and z) and wave vector components (kx, ky, and kz) according to

0z k z

Using the normalized parameters, Maxwell’s equations become…

0 0 0

yx zx y z

kk kk k k

k k k

0 r

0 r

0 r

yy z x

xx z y

x y y x z

dHjk H k E

dz

dHjk H k E

dz

jk H jk H k E

0 r

0 r

0 r

yy z x

xx z y

x y y x z

dEjk E k H

dzdE

jk E k Hdz

jk E jk E k H

r

r

r

yy z x

xx z y

x y y x z

dHjk H E

dz

dHjk H E

dz

jk H jk H E

r

r

r

yy z x

xx z y

x y y x z

dEjk E H

dzdE

jk E Hdz

jk E jk E H

Solve for the Longitudinal Components Ez and Hz

Slide 14

r

r

r

yy z x

xx z y

x y y x z

dHjk H E

dz

dHjk H E

dz

jk H jk H E

r

r

r

yy z x

xx z y

x y y x z

dEjk E H

dzdE

jk E Hdz

jk E jk E H

Solve the third and sixth equations for the longitudinal field components Hz and Ez.

r

z x y y x

jH k E k E

r

z x y y x

jE k H k H

13

14

Page 8: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

8

r

r

r

yy z x

xx z y

z x y y x

dHjk H E

dz

dHjk H E

dzj

E k H k H

r

r

r

yy z x

xx z y

z x y y x

dEjk E H

dzdE

jk E Hdz

jH k E k E

Eliminate the Longitudinal Components

Slide 15

Eliminate the longitudinal field terms by substituting them back into the remaining equations.

2r r r

2r r r

yy x x y y x

xx y x y x y

dEk H k k H H

dzdE

k H k k H Hdz

2r r r

2r r r

yy x x y y x

xx y x y x y

dHk E k k E E

dz

dHk E k k E E

dz

Rearrange Maxwell’s Equations

Slide 16

Rearrange the terms and the order of the equations.

2

rr r

2

r r

2

rr r

2

rr r

x yx xx y

y y x yr x y

x yx xx y

y y x yx y

k kdE kH H

dz

dE k k kH H

dz

k kdH kE E

dz

dH k k kE E

dz

2r r r

2r r r

2r r r

2r r r

yy x x y y x

xx y x y x y

yy x x y y x

xx y x y x y

dEk H k k H H

dzdE

k H k k H Hdz

dHk E k k E E

dz

dHk E k k E E

dz

15

16

Page 9: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

9

Matrix Form of Maxwell’s Equations

Slide 17

The remaining four equations can be written in matrix form as

2

rr r

2

rr r

2

rr r

2

rr r

0 0

0 0

0 0

0 0

x y x

x xy x y

y y

x xx y x

y y

y x y

k k k

E Ek k k

E EdH Hdz k k kH H

k k k

2

rr r

2

rr r

2

rr r

2

rr r

x yx xx y

y y x yx y

x yx xx y

y y x yx y

k kdE kH H

dz

dE k k kH H

dz

k kdH kE E

dz

dH k k kE E

dz

BTW…for Fully Anisotropic Materials

Slide 18

2

2

ˆyz yz zy x y yz zx yz zyzx x

y x x yx yyzz zz zz zz zz zz zz zz

zy yxz zx xzxy x y x

zz zz zz zz zzy

x

y

k k kj k k jk

kE jk j k kE

Hz

H

2

2

x y xz zyxz zxx xy

zz zz zz

x y yz zx yz zy yz yz zyx zxyx yy y x x

zz zz zz zz zz zz zz zz

y x yxz zxxx

zz zz zz

k k

k k kj k k jk

k k k

x

y

x

y

xz zy zyxz zx xzxy y x y

zz zz zz zz zz

E

E

H

H

jk j k k

Note:  This is for the 𝑒 sign convention.

17

18

Page 10: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

10

Slide 19

Solution in an LHI Layer

Matrix Differential Equation

Slide 20

Maxwell’s equations can now be written as a single matrix differential equation.

d

dz

ψΩψ 0

2

rr r

2

rr r

2

rr r

2

rr r

0 0

0 0

0 0

0 0

x y x

y x yx

y

x x y x

y

y x y

k k k

k k kE z

E zz

H z k k kH z

k k k

ψ Ω

19

20

Page 11: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

11

Solution of the Differential Equation (1 of 3)

Slide 21

The matrix differential equation is

d

dz

ψΩψ 0

This is actually a set of four coupled differential equations.

0zz e Ωψ ψ

This is easy to write, but how is the 

exponential of a matrix 𝑒𝛀 calculated?

2

rr r

2

rr r

2

rr r

2

rr r

x yx xx y

y y x yx y

x yx xx y

y y x yx y

k kdE kH H

dz

dE k k kH H

dz

k kdH kE E

dz

dH k k kE E

dz

The system of four equations can be solved as a single matrix equation as follows.

1

2

3

4

0

0

0

0

zx x

zy y

zx x

zy y

E z e E

E z e E

H z e H

H z e H

Functions of Matrices (1 of 2)

Slide 22

It is sometimes necessary to evaluate the function of a matrix.

?f A

It is NOT correct to calculate the function applied to every element in the matrix A individually.  A different technique must be used.

11 12 1

21 22 2

1 2

N

N

M M MN

f A f A f A

f A f A f Af

f A f A f A

A

This is more of an array operation than a matrix operation so it is incorrect to perform on a matrix.

21

22

Page 12: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

12

Functions of Matrices (2 of 2)

Slide 23

To calculate f(A) correctly, first calculate the eigen‐vectors and eigen‐values of the matrix A.

eigen-vector matrix of

eigen-value matrix of

V AA

D A

Given the eigen‐vector matrix V and the eigen‐value matrix D, the function of the matrix is evaluated as

1f f A V D Vf(D) is very easy to evaluate because D is a diagonal matrix so the function only has to be performed individually on the diagonal elements.

1

2

0 0

0 0

0

0 0 0 M

D

D

D

D

1

2

0 0

0 0

0

0 0 0 M

f D

f Df

f D

D

[V,D] = eig(A);

Solution of the Differential Equation (1 of 2)

Slide 24

So far, the following matrix differential equation had a general solution of

0zdz e

dz

Ωψ

Ωψ 0 ψ ψ

It is now possible to evaluate the matrix exponential using the eigen‐values and eigen‐vectors of the matrix .

eigen-vector matrix

eigen-value matrix

λ

1z ze e Ω λW W

1

2

3

4

0 0 0

0 0 0

0 0 0

0 0 0

z

zz

z

z

e

ee

e

e

λ

23

24

Page 13: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

13

Solution of the Differential Equation (2 of 2)

Slide 25

The solution to the matrix differential equation is therefore

d

dz

ψΩψ 0 1 0ze λW W ψ

The final solution is then

zdz e

dz

λψ

Ωψ 0 ψ W c

c

0zz e Ωψ ψ

The unknown initial values (0) can be combined with W-1 because that product just leads to another column vector of unknown constants.

Interpretation of the Solution

Slide 26

zez λWψ c

(z’) – Overall solution which is the sum of all the modes at plane z’.

W – Square matrix who’s column vectors describe the “modes” that can exist in the material.  These are essentially pictures of the modes which quantify the relative amplitudes of Ex, Ey, Hx, and Hy. 

ez’ – Diagonal matrix describing how the modes propagate.  This includes accumulation of phase as well as decaying (loss) or growing (gain) amplitude.

c – Column vector containing the amplitude coefficient of each of the modes.  This quantifies how much power is in each mode.

25

26

Page 14: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

14

Getting a Feel for the Numbers (1 of 2)

Slide 27

For a layer with r = 9.0 and r = 1.0 (i.e. n = 3.0) and a wave at normal incidence, the matrix will be

0 0 0 1

0 0 1 0

0 9 0 0

9 0 0 0

Ω

This matrix has the following eigen‐vectors and eigen‐values.

0.32 0.32 0 0

0 0 0.32 0.32

0 0 0.95 0.95

0.95 0.95 0 0

j j

j j

W

3.0 0 0 0

0 3.0 0 0

0 0 3.0 0

0 0 0 3.0

j

j

j

j

λ

Getting a Feel for the Numbers (2 of 2)

Slide 28

It can be observed that the modes occur as either an Ex‐Hy or an Ey‐Hx pair.  This is consistent with plane waves.  Due to the normalization, they are 90° out of phase.  A sign difference indicates forward and backward waves.  Only the relative amplitude difference between E and H is important here.

The refractive index is known (n = 3.0), so the eigen‐values are consistent with what may be expected.  The signs correspond to forward and backward waves.

0.32 0.32 0 0

0 0 0.32 0.32

0 0 0.95 0.95

0.95 0.95 0 0

j j

j j

W

3.0 0 0 0

0 3.0 0 0

0 0 3.0 0

0 0 0 3.0

j

j

j

j

λ

inccos

inc

r r

cos

3

jn zze e

jn

n

r0

r

r

r

1

3

E

H

E

H

The modes in W only contain information about the relative amplitudes of the field components.

The numbers in  describe how the modes accumulate phase in the z direction.  This is essentially just the complex refractive index of the material.

27

28

Page 15: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

15

Mode 4

Mode 3

Mode 2

Mode 1

Visualizing the Modes

Slide 29

0.32 0.32 0 0

0 0 0.32 0.32

0 0 0.95 0.95

0.95 0.95 0 0

j j

j j

W

3.0 0 0 0

0 3.0 0 0

0 0 3.0 0

0 0 0 3.0

j

j

j

j

λ

Mode 1

0.95

-j0.32

Mode 2

0.95

j0.32

Mode 3

0.95

j0.32 Mode 4

0.95

-j0.32

Slide 30

Transfer Matrices for Multilayer Structures

29

30

Page 16: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

16

Geometry of an Intermediate Layer

Slide 31

Layer i Layer i+1Layer i-1

0iψ

1 0 1i ik L ψ

iL

0i ik Lψ

1 0iψ

1iL 1iL

1icic1ic

i izψ

is a local z‐coordinate inside the ith layer that starts at zero at the layer’s left side.

iz0

iz

Field Relations

Slide 32

Field inside the ith layer:

,

,

,

,

i i

x i i

y i i zi i i i

x i i

y i i

E z

E zz e

H z

H z

λψ W c

Boundary conditions at the first interface:

Boundary conditions at the second interface:

1 0 1

1 0 1

1 1

0

i i

i i i

k Li i i i

k L

e

λ

ψ ψ

W c Wc

0

0 1

1 1

0

i i

i i i

k Li i i i

k L

e

λ

ψ ψ

W c W c

Must include k0 in the exponential to normalize Li-1 because the parameter i-1 expects to multiply a normalized coordinate.

Note: Must equate the field  on either side of the interfaces and not the mode coefficients c.

31

32

Page 17: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

17

The Transfer Matrix

Slide 33

The transfer matrix Ti of the ith layer is defined as:

Start with the boundary condition equation from the second interface and rearrange terms…

1i i i c T c

iT0 01

1 1 1 1 i i i ik L k Li i i i i i i ie e

λ λW c W c c W W c

011

i ik Li i ie

λT W W…then read off the transfer matrix.

The Transfer Matrix Method

Slide 34

The transfer matrix method (TMM) consists of working through the device one layer at a time and calculating an overall (global) transfer matrix.

1T 2T 3T 4T 5T

global 5 4 3 2 1 T T T T T T This is standard matrix multiplication.

Reflection Region

Transmission Region

The order of multiplication may seem backwards here, but it is not.  Recall the definition of the transfer matrix to have this make sense.

33

34

Page 18: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

18

The Global Transfer Matrix

Slide 35

The transfer matrix so far is not yet the “true” global transfer matrix because it does not connect the reflection region to the transmission region.  It only connects the amplitude coefficients of Layer 1 to the amplitude coefficients in the transmission region.  This is a result of how we defined the transfer matrix.

ref ref 1 1W c W c

Solving this for c1 yields

trn global 1c T c

The global transfer matrix must connect the amplitude coefficients in the reflection region to the amplitude coefficients in the transmission region. Boundary conditions at the first interface require

The global transfer matrix is derived by substituting this result into the first equation.

11 1 ref ref

c W W c

1trn global 1 ref ref

1global global 1 ref

c T W W c

T T W W1

global 5 4 3 2 1 1 ref T T T T T T W W

Slide 36

Stability of Transfer Matrices

35

36

Page 19: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

19

The Multi‐Layer Problem

Slide 37

The diagram below is focused on an arbitrary layer in a stack of multiple layers.  Now consider the wave solutions in this ith layer.

Wave Solutions in ith Layer

Slide 38

Recall that the wave number 𝑘 can be purely real (pure oscillation), purely imaginary (pure exponential decay), or complex (decay + oscillation).

k k jk

k jk

k kPure oscillation

Pure decay

Decaying oscillation

37

38

Page 20: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

20

Backward Waves in ith Layer

Slide 39

Due to reflections at the interfaces, there will also be backward traveling waves in each of the layers.  These can also have wave vectors that are real, imaginary or complex, so they can oscillate, decay/grow, or both.

All Waves are Treated as Forward Waves

Slide 40

The pure transfer matrix method treats all waves as if they are forward propagating.  Decaying fields associated with backward waves become exponentially growing fields and quickly become numerically unstable.

,

,

,

,

i i

x i i

y i i zi i i i

x i i

y i i

E z

E zz e

H z

H z

λψ W c

39

40

Page 21: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

21

TMM is Inherently Unstable

Slide 41

The wave solution was

x

y z

x

y

E z

E zz e

H z

H z

λψ W c

This treats all power as forward propagating.

It is known that backward waves exist.  It is also known that decaying fields exist when a wave is evanescent or propagating in a lossy material.

When backward waves are decaying and treated as forward propagating waves, they grow exponentially.  This leads to numerically instability.

The TMM is inherently an unstable method because it treats everything as forward propagating.

The Fix

Slide 42

All power in the waves is being treated as forward propagating because it power was distinguished as forward and backward waves.

Clearly, the first part of the fix is to identify forward and backward propagating waves.

This can be accomplished by calculating the Poynting vector associated with the modes and looking at the sign of the z component.  Be careful!  A normalized magnetic field is being used.

0 0

0

1

z x y y x

y xz x y

z x y y x

E H

E H E H

H HE E

j j

E H E Hj

0.32 0.32 0 0

0 0 0.32 0.32

0 0 0.95 0.95

0.95 0.95 0 0

i i

i i

W

41

42

Page 22: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

22

3.0 0 0 0

0 3.0 0 0

0 0 3.0 0

0 0 0 3.0

i

i

i

i

λ

Rearrange Eigen Modes

Slide 43

Now that it is known which eigen‐modes are forward and backward propagating, they can be rearranged to group them together.

0.32 0.32 0 0

0 0 0.32 0.32

0 0 0.95 0.95

0.95 0.95 0 0

i i

i i

W

3.0 0 0 0

0 3.0 0 0

0 0 3.0 0

0 0 0 3.0

i

i

i

i

λ

0.32 0 0.32 0

0 0.32 0 0.32

0 0.95 0 0.95

0.95 0 0.95 0

i i

i i

W

Also need to adjust the vertical positions of the eigen‐values so that ’ remains a diagonal matrix.

rearrange modes

New Interpretation of the Matrices

Slide 44

3.0 0 0 0

0 3.0 0 0

0 0 3.0 0

0 0 0 3.0

i

i

i

i

λ

0.32 0 0.32 0

0 0.32 0 0.32

0 0.95 0 0.95

0.95 0 0.95 0

i i

i i

W

x

y

x

y

E

E

H

H

E E

H H

zz

z

ee

e

λλ

λ

W WW

W W

0

0

The matrices are now partitioned into forward and backward propagating elements.

x

y

x

y

E

E

H

H

3.0 0 3.0 0

0 3.0 0 3.0

i i

i i

λ λNote: For anisotropic materials, all the eigen‐vectors and eigen‐values are in general unique.

43

44

Page 23: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

23

Revised Solution to Differential Equation

Slide 45

The matrix differential equation and its original solution was

zdz e

dz

λψ

Ωψ 0 ψ W c

After distinguishing between forward and backward propagating waves and grouping them in the matrices, the solution can be written as

z

E E

zH H

ez

e

λ

λ

0W W cψ

W W c0

There are now separate mode coefficients c+ and c- for forward and backward propagating modes, respectively.

Slide 46

Formulation of 22 Matrix Equationfor 1D Structures

45

46

Page 24: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

24

Recall Derivation Up to 44

Slide 47

0 r

0 r

0 r

0 r

0 r

0 r

yzx

x zy

y xz

yzx

x zy

y xz

EEk H

y z

E Ek H

z xE E

k Hx y

HHk E

y z

H Hk E

z x

H Hk E

x y

Start with Maxwell’s equations from 

Lecture 2.

Assume LHI.

0 r

0 r

0 r

0 r

0 r

0 r

yy z x

xx z y

x y y x z

yy z x

xx z y

x y y x z

dEjk E k H

dzdE

jk E k Hdz

jk E jk E k H

dHjk H k E

dz

dHjk H k E

dz

jk H jk H k E

Assume device is infinite and uniform in x and y directions.

x yjk jkx y

r

r

r

r

r

r

yy z x

xx z y

x y y x z

yy z x

xx z y

x y y x z

dEjk E H

dzdE

jk E Hdz

jk E jk E H

dHjk H E

dz

dHjk H E

dz

jk H jk H E

Normalize z and wave vectors kx, ky, 

and kz.

0

0 0 0

yx zx y z

z k z

kk kk k k

k k k

2

rr r

2

rr r

2

rr r

2

rr r

x yx xx y

y y x yx y

x yx xx y

y y x yx y

k kdE kH H

dz

dE k k kH H

dz

k kdH kE E

dz

dH k k kE E

dz

Eliminate longitudinal 

components Ez and Hz by substitution.

Derivation of Two 22 Matrix Equations

Slide 48

2

rr r

2

rr r

x yx xx y

y y x yx y

k kdE kH H

dz

dE k k kH H

dz

2

rr r

2

rr r

x yx xx y

y y x yx y

k kdH kE E

dz

dH k k kE E

dz

The four equations can be written as two separate matrix equations instead.

2r r

2r r r

1x x y x x

y yy x y

E k k k HdE Hdz k k k

2r r

2r r r

1 x y x xx

yy y x y

k k k EHdEHdz k k k

Note: These equations are valid regardless of the sign convention because there is always a k multiplying another k and erasing the sign.

47

48

Page 25: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

25

Standard “PQ” Form

Slide 49

The two matrix equations more compactly by defining auxiliary P and Q matrices.

2r r

2r r r

1x x y x x

y yy x y

E k k k HdE Hdz k k k

2r r

2r r r

1 x y x xx

yy y x y

k k k EHdEHdz k k k

x x

y y

E HdE Hdz

P

xx

yy

EHdEHdz

Q

2r r

2r r r

1 x y x

y x y

k k k

k k k

P

2r r

2r r r

1 x y x

y x y

k k k

k k k

Q

Note: This same “PQ” form will be seen again again for other methods like MoL, RCWA, and waveguide analysis.  TMM, MoL, and RCWA are all implemented the same after P and Q are calculated.

Matrix Wave Equation

Slide 50

The two staring matrix equations are

To derive a matrix wave equation, first differentiate Eq. (1) with respect to z’.

x x

y y

E HdE Hdz

P

xx

yy

EHdEHdz

QEq. (1) Eq. (2)

22

2

2

0

0x x

y y

E EdE Edz

Ω

Ω PQ

Second, substitute Eq. (2) into this result.

2

2

x xx x

y yy y

E EH Hd d d d dE EH Hdz dz dz dz dz

P P

2

2

x x

y y

E EdE Edz

P Q

49

50

Page 26: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

26

Numerical Solution (1 of 3)

Slide 51

The system of equations to be solved is

22 2

2

0

0x x

y y

E EdE Edz

Ω Ω PQ

This has the general solution of

x z z

y

E ze e

E z

Ω Ωa a proportionality constant of forward wave

proportionality constant of backward wave

a

a

No mode sorting!   Here, a second‐order differential equation is solved so the modes we are all propagating in a single direction.  They must be explicitly written twice to account for forward and backward waves and thus they are automatically distinguished.  In the 4×4 approach, a first‐order differential equation was solved that lumped forward and backward modes together.

Numerical Solution (2 of 3)

Slide 52

Recall that

1 1 z z z ze e e e Ω λ Ω λW W W W

So the overall solution can now be written as

1 1x z z

y

E ze e

E z

λ λW W a W W a

2

2 2

Eigen-vector matrix of

Eigen-value matrix of

W Ω

λ Ω

21

1

222

2 NN

zz

zzz

zz

e e

eee

ee

λ

1f f A V D V

This relation is used to calculate the matrix exponentials.

51

52

Page 27: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

27

Numerical Solution (3 of 3)

Slide 53

So the overall solution can now be written as

1 1x z z

y

E ze e

E z

λ λ

cc

W W a W W a

The column vectors a+ and a‐ are proportionality constants that have not yet been determined.

The eigen‐vector matrix W multiplies a+ and a‐ to give another column vector of undetermined constants.

To simplify the math, we combine these products into new column vectors labeled c+ and c‐ .

x z z

y

E ze e

E z

λ λW c W c

Solution for the Magnetic Field (1 of 2)

Slide 54

Since the electric and magnetic fields are coupled and not independent, it should be possible to compute V from W.   First, differentiate the above solution with respect to z’.

The magnetic field has a solution of the same form, but will have its own eigen‐vector matrix V to describe its modes.

x z z

y

H ze e

H z

λ λV c V c

x z z

y

H zde e

H zdz

λ λVλ c Vλ c

We are free to choose any sign we wish because it can be accounted for in c-.  We put the minus sign in the solution here so that both terms in the differentiated equation will be positive.  You will see soon why this is desired.

53

54

Page 28: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

28

Solution for the Magnetic Field (2 of 2)

Slide 55

x z z

y

H zde e

H zdz

λ λVλ c Vλ c

From the previous slide,

xx

yy

E zH zdE zH zdz

Q

Recall

x z z

y

E ze e

E z

λ λW c W cand

Combining these results leads to

z z z z

z z

e e e e

e e

λ λ λ λ

λ λ

Vλ c Vλ c Q W c W c

QW c QW c

Comparing the terms on the left and right sides of this equation shows that

1 Vλ QW V QWλ

Combined Solution for E and H

Slide 56

Electric Field Solution

x z z

y

E ze e

E z

λ λW c W c

amplitude coefficients of forward wave

amplitude coefficients of backward wave

eigen-vector matrix

diagonal eigen-value matrix

c

c

W

λ

Combined Solution

1 x z z

y

H ze e

H z

λ λV c V c V QWλ

x

zy

zx

y

E z

E z ez

H z e

H z

λ

λ

W W 0 cψ

V V 0 c

Magnetic Field Solution

Does this equation look familiar?

This is the same equation obtained for the 44 approach after the modes were sorted.

55

56

Page 29: Transfer Matrix Method (TMM) · 2019. 9. 17. · Slide21 The matrix differential equation is d dz ψ Ωψ 0 This is actually aset of four coupled differential equations. ψψ ze Ωz

9/17/2019

29

Two Paths to Combined Solution

Slide 57

0 r

0 r

E k H

H k E

Maxwell’s Equations Field Solution

2

2

ˆyz yz zy x y yz zx yz zyzx x

y x x yx yyzz zz zz zz zz zz zz zz

zy yxz zx xzxy x y

zz zz zz zz zzy

x

y

k k kj k k jk k

kE jk j k kE

Hz

H

2

2

x y xz zyxz zxxx xy

zz zz zz

x y yz zx yz zy yz yz zyx zxyx yy y x x

zz zz zz zz zz zz zz zz

y x yxz zxxx

zz zz z

k k

k k kj k k jk

k k k

x

y

x

y

xz zy zyxz zx xzxy y x y

z zz zz zz zz zz

E

E

H

H

jk j k k

2r r

2r r r

2r r

2r r r

1

1

x y x

y x y

x y x

y x y

k k k

k k k

k k k

k k k

P

Q

4×4 Matrix Sort Eigen‐Modes

PQ Method

No sorting!  

Isotropic or diagonally anisotropic

Anisotropic

E E

H H

zz

z

ee

e

λλ

λ

W WW

W W

0

0

x

y

x

y

E

E

H

H

zE E

zH H

z

z

ez

e

ez

e

λ

λ

λ

λ

0W W cψ

V V c0

W W 0 cψ

V V 0 c

57