7/23/2019 Tech Background and Tutorials
1/515
OptiBPMTechnical Background and Tutorials
Waveguide Optics Modeling Software System
Version 12.1for Windows XP, Vista, Windows 7
7/23/2019 Tech Background and Tutorials
2/515
7/23/2019 Tech Background and Tutorials
3/515
OptiBPMTechnical Background and TutorialsWaveguide Optics Modeling Software Systems
Copyrigh t 2013 OptiwaveAll rights reserved.
All OptiBPM documents, including this one, and the information contained therein, is copyright material.
No part of this document may be reproduced, stored in a retrieval system, or transmitted in any form or by any means whatsoever,including recording, photocopying, or faxing, without prior written approval of Optiwave.
DisclaimerOptiwave makes no representation or warranty with respect to the adequacy of this documentation or the programs which itdescribes for any particular purpose or with respect to its adequacy to produce any particular result. In no event shall Optiwave, itsemployees, its contractors or the authors of this documentation be liable for special, direct, indirect or consequential damages,losses, costs, charges, claims, demands, or claim for lost profits, fees or expenses of any nature or kind.
7/23/2019 Tech Background and Tutorials
4/515
Contact Information
Technical Support
Tel (613) 224-4700 E-mail [email protected]
Fax (613) 224-4706 URL www.optiwave.com
General Enquiries
Tel (613) 224-4700 ext.0 E-mail [email protected] (613) 224-4706 URL www.optiwave.com
Sales
Tel (613) 224-4700 ext.249 E-mail [email protected]
Fax (613) 224-4706 URL www.optiwave.com
7/23/2019 Tech Background and Tutorials
5/515
Table of contents
Technical Background..............................................................................................11
Notes: .....................................................................................................................................12
Beam Propagation Method (BPM) ...........................................................................13
Slowly Varying Envelope Approx imation ...........................................................................13
Differential Equations of BPM..............................................................................................15
Semi-Vector and Scalar BPM ...............................................................................................19
Crank-Nicholson Method and Scheme Parameter .............................................................20
ADI ..........................................................................................................................................22
Boundary Condi tions for the Beam Propagation Method .................................................23
Perfectly Matched Layer (PML) ............................................................................................26
Wide Angle Beam Propagation Method ..............................................................................30
Wide-Angle Beam Propagation Method based on Pad Approximant Operators ..........48
References .............................................................................................................................65
Conformal Mapping Regions ...................................................................................67
References .............................................................................................................................69
Notes: .....................................................................................................................................70
Diffusion in Lithium Niobate ....................................................................................71
7/23/2019 Tech Background and Tutorials
6/515
Titanium Diffusion in Lithium Niobate ................................................................................71
Magnesium Diffusion in Lithium Niobate............................................................................73
Proton Exchange Process in Lithium Niobate ...................................................................74
Electroopt ic Effect ....................................................................................................77
Index elli psoid .......................................................................................................................78
References .............................................................................................................................80
Scatter ing Data..........................................................................................................81
Introduct ion ...........................................................................................................................81
Modeling of the Opt ical Components - Survey of Methods ..............................................81
Circui t Complexi ty - Class if ication ......................................................................................83
The Tool - Scattering Data Approach ..................................................................................86
Implementation with OptiSystem.........................................................................................90
Solu tions ................................................................................................................................90
References: ..........................................................................................................................104
Non-linear BPM algorithm ......................................................................................105
Appendix A: Technical background and manual calculat ion method ...............107
Vectoral Beam Propagation for Anisotropic Waveguides ..................................117
Introduct ion .........................................................................................................................117
Mathematical Formulations................................................................................................118
Vectoral Modal Analysis for Anisotropic Waveguide..........................................129
Introduct ion .........................................................................................................................129
Appendix I ............................................................................................................................131
Appendix II ...........................................................................................................................137
References ...........................................................................................................................138
7/23/2019 Tech Background and Tutorials
7/515
Fiber Mode Solvers .................................................................................................139
Introduct ion .........................................................................................................................139
Real-valued formulation .....................................................................................................140
Debye Potential ...................................................................................................................140
Separation of Variab les ......................................................................................................142
Solu tion of the linear system .............................................................................................143
Dispersion equat ion ............................................................................................................144
LP Modes .............................................................................................................................146
References ...........................................................................................................................148
Finite Difference Mode Solver................................................................................149
Introduct ion .........................................................................................................................149
Magnetic Formulation .........................................................................................................150
Magnetic Finite Difference Equat ions ...............................................................................151
Impli cit ly Restarted Arnold i Method (IRAM).....................................................................152
Transparent Boundary Cond it ion (TBC) ...........................................................................158
References ...........................................................................................................................161
Tutorials ...................................................................................................................165
Lesson 1: Getting Started ......................................................................................167
Install ing OptiBPM ..............................................................................................................167
Hardware and software requirements ...............................................................................167
Protecti on key......................................................................................................................167
OptiBPM directory ...............................................................................................................168
Installation ...........................................................................................................................168
Installing OptiBPM on Windows XP, Vista or Windows 7 ......................................168
Introduction to OptiBPM.....................................................................................................169
7/23/2019 Tech Background and Tutorials
8/515
What is OptiBPM? ...............................................................................................................169
Introduction to optical waveguides ...................................................................................172
What are optical waveguides? ...........................................................................................172
Quick start............................................................................................................................173
Starting OptiBPM ...................................................................................................173
GUI main parts .....................................................................................................................174
How to use OptiBPM ...........................................................................................................176
Defining materials ..................................................................................................176
Defining 2D and 3D channel profiles......................................................................181
Defining the layout settings ....................................................................................185
Creating a basic project .........................................................................................189
Inserting the input plane.........................................................................................190
Running the simulation...........................................................................................195
Lesson 2: Create a simple MMI coupler................................................................207
Defining materials for the MMI coupler ..................................................................207
Defining the layout settings ....................................................................................209
Creating an MMI coupler........................................................................................211
Inserting the input plane.........................................................................................216
Running the simulation...........................................................................................221
Viewing the simulation results in OptiBPM_Analyzer.............................................225
Lesson 3: Create a single-bend device.................................................................227
Defining materials for the single-bend device ........................................................227
Defining the layout settings ....................................................................................229
Creating an arc waveguide ....................................................................................230
Inserting the input plane.........................................................................................234
Selecting output data files......................................................................................239
Running the simulation...........................................................................................241
Viewing the simulation results in OptiBPM_Analyzer.............................................245
Notes: ...................................................................................................................................248
7/23/2019 Tech Background and Tutorials
9/515
Lesson 4: Create an MMI star coupler ..................................................................249
Defining materials for the MMI star coupler ...........................................................250
Defining the layout settings ....................................................................................250
Creating an MMI star coupler.................................................................................251
Inserting the input plane.........................................................................................251
Running the simulation...........................................................................................253
Viewing the maxima...............................................................................................256
Drawing the output waveguide...............................................................................260
Assigning a path to the output waveguide .............................................................261
Viewing the simulation results in OptiBPM_Analyzer.............................................264
Adding output waveguides and viewing the new simulation results.......................265
Viewing the new simulation results in OptiBPM_Analyzer .....................................267
Notes: ...................................................................................................................................269
Lesson 5: Wavelength scripting with VB Script...................................................271
Defining materials for the waveguide .....................................................................271
Defining the layout settings ....................................................................................272
Creating the waveguides........................................................................................272
Modifying the input plane .......................................................................................273
Assigning paths to the waveguides........................................................................274
Running the simulation...........................................................................................276
Viewing the simulation results in OptiBPM_Designer ............................................278
Using VB Scripting for the simulation.....................................................................280
Viewing the simulation results in OptiBPM_Analyzer.............................................284
Notes ....................................................................................................................................287
Lesson 6: Design a 3dB coupler using VB Script................................................289
Defining the materials for the 3dB coupler .............................................................289
Defining the layout settings ....................................................................................292
Drawing and positioning the waveguides...............................................................295
Generating the layout script ...................................................................................302
Inserting and editing the input plane ......................................................................303
Running the simulation...........................................................................................306
7/23/2019 Tech Background and Tutorials
10/515
Modifying the layout script......................................................................................309
Viewing simulation results in OptiBPM_Analyzer...................................................312
Lesson 7: Applying predefined dif fus ion processes...........................................315
Creating a linear waveguide formed by the Titanium Diffusion in Lithium Niobate 315
Defining the layout settings ....................................................................................318
Designing the waveguide.......................................................................................319
Setting the simulation parameters..........................................................................321
Running the simulation...........................................................................................323
Creating a buried waveguide formed by Titanium Diffusion in Lithium Niobate.....325
Saving the template under a new name.................................................................325
Adding a new profile...............................................................................................325
Creating the top linear waveguide..........................................................................326
Notes ....................................................................................................................................329
Lesson 8: 3D OptiMode solver - COST project Waveguide ................................331
Defin ing the materials for the 3D Mode solver .................................................................331
Defin ing the waveguide ......................................................................................................334
Parameterize the Problem ..................................................................................................336
Add Substrate Layer ...........................................................................................................336
Setting Simulati on Parameters ..........................................................................................338
Modi fy ing the Film Thickness ............................................................................................340
Automat icall y modifying a parameter by scr ip t ...............................................................340
Lesson 9: Create a chip-to-fiber but t coupler ......................................................343
Defining materials and waveguides for the chip-to-fiber butt coupler ....................345
Defining the layout settings ....................................................................................346
Creating a chip-to-fiber butt coupler.......................................................................347
Editing the Input plane ...........................................................................................349
Setting the simulation parameters..........................................................................352
Running the simulation...........................................................................................352
7/23/2019 Tech Background and Tutorials
11/515
References ...........................................................................................................................355
Notes: ...................................................................................................................................356
Lesson 10: Electro-Optic Modulator ....................................................................357
Define Dielectric Materials .....................................................................................359
...............................................................................................................................361
Define Electrode Materials .....................................................................................361
Define Profiles........................................................................................................363
Draw the waveguide...............................................................................................366
Draw the electrodes ...............................................................................................372
Electrostatic Simulation..........................................................................................376
Electro-Optic Simulation.........................................................................................380
References ...........................................................................................................................383
Lesson 11: Integrated optical circuit simulation using OptiBPM and OptiSystem Scattering data export .........................................................................................385
Theoretical Background .....................................................................................................385
Part I OptiBPM.................................................................................................................387
Generating Scattering Data script..........................................................................390
Exporting scattering data .......................................................................................394Creating the arms...................................................................................................395
Part II OptiSystem...........................................................................................................397
Loading the *.s file in OptiSystem ..........................................................................397
Loading the files for both Arms ..............................................................................401
Completing the layout in OptiSystem.....................................................................403
Connecting the components ..................................................................................405
Running the calculation..........................................................................................407
Creating the graph to view the results....................................................................407
Reference:............................................................................................................................409
Notes ....................................................................................................................................410
Lesson 12: Scan the refractive index (RI).............................................................411
7/23/2019 Tech Background and Tutorials
12/515
Defining the materials and the channel..................................................................411
Defining the layout settings ....................................................................................414
Drawing the Linear Waveguide..............................................................................417
Inserting the Input Plane ........................................................................................418
Creating the script..................................................................................................419
Running the simulation...........................................................................................421
Viewing the simulation results in OptiBPM_Analyzer.............................................423
Lesson 13: Applying User-defined dif fus ion profiles..........................................425
Developing the user defined prof ile ..................................................................................426
Defining the materials ............................................................................................426
Creating the reference profile.................................................................................427
Defining the layout settings ....................................................................................427
User defined prof ile.............................................................................................................428
Defining variables...................................................................................................430
Defining functions...................................................................................................430
Defining the user defined profile ............................................................................431
Checking the user defined profile against the reference profile .............................432
Notes ....................................................................................................................................434
Lesson 14: Mach-Zehnder Interferometer switch ................................................435
CAD design of the circuit ...................................................................................................436
Defining the materials ............................................................................................436
Creating the Ti-diffused profile...............................................................................436
Defining the Wafer .................................................................................................438
Creating the device ................................................................................................439
Checking the RI profile of the x-y slice...................................................................442
Defining the electrode region .................................................................................443
Defining the input plane and simulation parameters.............................................. 447
Running the simulation...........................................................................................448
Creating a script.....................................................................................................450
Lesson 15: Step index fiber modes .......................................................................453
7/23/2019 Tech Background and Tutorials
13/515
Step index fiber ...................................................................................................................454
Defining the materials ............................................................................................455
Creating the SMF-28 fiber profile...........................................................................455
Defining the layout settings ....................................................................................457
Creating the linear waveguide................................................................................458
Setting the simulation parameters..........................................................................460
Viewing the refractive index distribution (X-Y cut)..................................................461
Calculating the mode .............................................................................................463
Lesson 16: Graded index fiber modes ..................................................................467
Graded Index - Core fiber ..................................................................................................467
Creating the material..............................................................................................470
Defining user variables...........................................................................................470
Defining user functions...........................................................................................471
Defining the User defined profile............................................................................473
Defining the layout settings ....................................................................................475
Creating the linear waveguide fiber........................................................................476
Setting the simulation parameters..........................................................................477
Viewing the refractive index distribution (X-Y cut)..................................................477
Calculating the mode .............................................................................................479
Notes ....................................................................................................................................482
Lesson 17: Modal analysis of an anisotropic buried waveguide .......................483
Defining the materials ............................................................................................484
Creating the profile.................................................................................................486
Defining the layout settings ....................................................................................488
Creating the linear waveguide................................................................................490
Setting the simulation parameters..........................................................................491
Viewing the elements of the permittivity (XY cut)...................................................493
Calculating the mode .............................................................................................494
Computing the magnetic field.................................................................................500
References: ..........................................................................................................................500
7/23/2019 Tech Background and Tutorials
14/515
Lesson 18: BPM analysis of an anisotropic bur ied waveguide..........................501
Before you start th is lesson ...............................................................................................501
Setting the simulat ion parameters ....................................................................................502
Viewing the elements of the permi tti vity (XY cut) ............................................................503
Setting the field in the input plane.....................................................................................505
References: ..........................................................................................................................509
7/23/2019 Tech Background and Tutorials
15/515
11
Technical Background
The following section contains information on the following topics:
Beam Propagation Method (BPM)
Conformal Mapping Regions
Diffusion in Lithium Niobate
Scattering Data
Non-linear BPM algorithm
Vectoral Beam Propagation for Anisotropic Waveguides
Vectoral Modal Analysis for Anisotropic Waveguide
7/23/2019 Tech Background and Tutorials
16/515
Notes:
7/23/2019 Tech Background and Tutorials
17/515
BEAMPROPAGATIONMETHOD(BPM)
13
Beam Propagation Method (BPM)
The finite difference beam propagating method (BPM) is one of the most powerfultechniques to investigate linear and nonlinear lightwave propagation phenomena inaxially varying waveguides such as curvilinear directional couplers, branching andcombining waveguides, S-shaped bent waveguides, and tapered waveguides. BPMis also quite important for the analysis of ultra short light pulse propagation in opticalfibers [9].
Like the Finite Difference Time Domain method, finite difference BPM solvesMaxwell's equations by using finite differences in place of partial derivatives. In thissense BPM is computational intensive, and able to accurately model a very widerange of devices. It differs from a full and direct solution of the equations as found inthe Finite Difference Time Domain Method in two ways. The first is that BPM is doneentirely in the frequency domain, and as such only weak non-linearities can be
modelled. The second is in the use of a slowly varying envelope approximation in theparaxial direction. In BPM, it is assumed the device has an optical axis, and that mostof the light travels in this direction, or at least approximately in this direction (paraxialapproximation). In OptiBPM, as with the majority of literature on the subject, this axisis taken to be , the third space co-ordinate. Many practical optical devices arenaturally aligned close to a single direction, so once this is associated with the axis,the slowly varying approximation can be applied.
Slowly Varying Envelope Approximation
Suppose is an electric or magnetic component of the optical electromagnetic field.This component is a periodic (harmonic) function of position, it changes most rapidlyalong the optical axis, , and has a period that is on the order of the optical
wavelength. The slowly varying approximation involves replacing the quickly varyingcomponent, , with a slowly varying one,
where , and is known as the Reference Index. If the light is travellingmostly parallel to the axis (paraxial approximation), and is monochromatic(wavelength ), then it should be possible to select a reference index whichmakes a slowly varying function in all three directions, , , and . If is slowlyvarying, the requirements on the mesh to represent derivatives by finite differencesare relaxed. It is possible to choose fewer mesh points to improve the speed of thecalculation without compromising the accuracy too much. This is the main reason whyBPM can do accurate calculations of light propagation using step sizes many times
larger than the optical wavelength, especially in the direction.
Equation 1can also be interpreted to give some indication about what will be requiredfrom the mesh, it must be sufficiently fine to approximate spatial derivatives ofaccurately by using finite differences. Since varies much more slowly than ,significant savings in calculation time can be obtained. On the other hand, if theproblem cannot be reduced so simply, then the will need to vary more quickly, and
(1)
zz
z
x y z, ,( ) x y z, ,( ) jkn0z( )exp=
k 2 = n0z
n0 x y z
z
7/23/2019 Tech Background and Tutorials
18/515
BEAMPROPAGATIONMETHOD(BPM)
14
the changes must be taken up by making the mesh finer (smaller , , and shorterpropagation step ). For many practical waveguides, the refractive index contrastin the transverse plane ( plane) is small, which means the variation of the field
in the transverse plane is slow, compared to the wavelength. On the other hand,sometimes there are large index contrasts in the transverse plane, and a finer mesh
in , , or both axes might be needed. Another case in which the effectiveness ofEquation 1is compromised is when some of the light deviates from the direction ofthe axis. This is another case in which faster variation in is unavoidable, andsometimes a finer mesh or Wide Angle methods are necessary.
In any case, it is recommended to experiment with several different meshes appliedto the same problem, and to compare the results. Usually, the same results are foundin all cases, but failing this, the results can be analyzed for trends. Eventually asuitable range of mesh parameters can be found which gives consistent results. InOptiBPM, it is very easy to apply many different meshes to the same problem andcompare the results.
Sometimes a single value of reference index will not reduce the variation everywhere,
but will follow the optical field well in one region of the device. Another region mayfollow a different Reference Index. For example, in a star coupler, a waveguideconnects to a free propagation region. The reference index for the waveguide is thewaveguide's modal index, but the reference index for the coupler would be the slabwaveguide's modal index. The best solution in this case is to define one referenceindex for one region and a different one for another region. This is easily done inOptiBPM.
x yz
x y
x y
z
7/23/2019 Tech Background and Tutorials
19/515
BEAMPROPAGATIONMETHOD(BPM)
15
Differential Equations of BPM
In this section we show the derivation for the differential equations found in BPM. Ofcourse, more complete accounts of this material can be found elsewhere (seeReferences [1]- [8]). This section exists for the convenience of the user of OptiBPM,
to define important terms and illustrate the nature of the different levels ofapproximation.
The derivation starts with Maxwell's equations for electromagnetic fields in acontinuous medium in the frequency domain. For optical materials it is usuallypermittivity which differs from the vacuum level. The permeability is unchanged, andso the electromagnetic field is influenced by the device geometry because therefractive index is a function of position. The refractive index distributionenters Maxwell's equations as a coefficient
and no charges are present
The magnetic field can be eliminated by taking the curl of Equation 2and usingEquation 3.
where .
By applying the vector identity
Equation 6becomes
BPM has a bias towards the axis, so it is natural to treat the component of bothand differently than the transverse components, and .
(2)
(3)
(4)
(5)
(6)
(7)
(8)
n x y z, ,( )
E j0H=
H j0n2
x y z, ,( )E=
n2E( ) 0=
H 0 .=
E k2n2E=
k 00=
( . ) 2= .
2E k2n2E+ E( ).=
z zE x y
7/23/2019 Tech Background and Tutorials
20/515
BEAMPROPAGATIONMETHOD(BPM)
16
Writing
the transverse components of Equation 8can be written
The divergence Equation 4can be used to eliminate the longitudinal term above.
Equations (9)and (10)can be used to separate the transverse and longitudinal termsin Equation 4
If the refractive index is changing slowly along , then the second term can beneglected. Then the second term on the right hand side of Equation 11can bereplaced with the divergence term of Equation 12.
Using the chain rule on the second divergence term on the right hand side gives aterm that cancels the first divergence term, leaving a single term on the right handside. Also, on the left side, the derivative can be written explicitly
(9)
(10)
(11)
(12)
(13)
(14)
E Et zEz+=
t zz-----+=
2Et k2n
2Et+ t t Et
Ezz--------+
.=
t n2Et( )
n2
z--------Ez n
2Ezz--------+ + 0 .=
z
2Et k2n
2Et+ t t Et
1
n2
----t n( 2 Et)=
z
t2
Et2Etz2---------- k
2
n2
Et+ + t1
n2---- ( tn
2
) Et .=
7/23/2019 Tech Background and Tutorials
21/515
BEAMPROPAGATIONMETHOD(BPM)
17
At this point the slowly varying envelope approximation is used, the field componentsExandEyare varying quickly in the direction, but are probably varying slowly inand . Therefore the substitution
is likely to result in a function which is slowly varying in all directions,compared to the wavelength of the light. Of course, a suitable value of ReferenceIndex needs to be used, and often OptiBPM can find by itself an appropriatenumber. However, sometimes OptiBPM cannot find a good number, and theReference Index needs to be adjusted manually. It is easy to see if the ReferenceIndex has been set correctly after the simulation is finished. A good value will resultin slow variation of the phase of the solution with respect to .
Substituting Equation 15into Equation 14
If the reference index has been set correctly then the first term will be much smallerthan the second, and it is safe to neglect the first term. The remaining equation is offirst order in . It is customary to collect the transverse differential operators on theright hand side,
where the components of the operator are
(15)
(16)
(17)
(18)
z xy
Et x y z, ,( ) et x y z, ,( ) jn 0kz( )exp=etx y z, ,( )
n0
et x y z, ,( ) z
2etz2--------- 2jkn
0
etz------- k
2
n( 2
n02
)et t2
et t1
n2---- tn
2
( ) et+ + + 0 .=
z
2jkn0z-----
ex
ey
Pxx
Pyx
Pxy
Pyy
ex
ey=
P
Pxx
x-----
1
n2
----x-----n
2. 2
y2-------+= k
2n( 2 n0
2 )+
Pxy
x-----
1
n2
----y-----n
2. 2
xy------------=
Pyx
y-----
1
n
2----
x-----n
2.=2
yx------------
Pyy
y-----
1
n2
----y-----n
2. 2
x2-------+= k
2n( 2 n0
2 ).+
7/23/2019 Tech Background and Tutorials
22/515
7/23/2019 Tech Background and Tutorials
23/515
BEAMPROPAGATIONMETHOD(BPM)
19
Semi-Vector and Scalar BPM
The above system of equations for BPM is called the Full-Vector form, as it includesboth transverse components of the field. Often it is not necessary to have both fieldcomponents in the simulation. If it is known that the device does not change the
polarization of light, then it is sufficient to model one polarization at a time (or onlymodel the expected polarization). In these cases the contribution of to and viceversa can be ignored, and the governing equation is
for Semi-Vector TE and
for Semi-Vector TM. In Semi-Vector BPM, one can notice a difference in thepropagation of light depending on whether the initial light is polarized parallel orperpendicular to the device substrate. However, once polarized horizontally, forexample, the model does not allow the horizontal light to transfer into a verticalpolarization.
On the other hand, for structures with a very low index contrast, it won't make muchdifference in the order of application of operators like and . In this case, theoperators commute, and both P
xxand P
yyreduce to the scalar operator:
(19)
(20)
(21)
ey ex
2jkn0exz------- Pxx ex=
2jkn0eyz------- Pyy ey=
x n2
P 2
x2-------
2
y2------- k
2n
2n0
2( ).+ +=
7/23/2019 Tech Background and Tutorials
24/515
7/23/2019 Tech Background and Tutorials
25/515
BEAMPROPAGATIONMETHOD(BPM)
21
often realizes the most stable method. In Equation 24, it is possible to show (beforethe introduction of lossy boundary conditions) that the norm of the operator on the leftis the same as the operator on the right. Therefore the application of many stepsshould lead to no change in the norm of the solution vector . This guarantees thestability of the method, at least in the sense of conservation of energy.
et
7/23/2019 Tech Background and Tutorials
26/515
7/23/2019 Tech Background and Tutorials
27/515
7/23/2019 Tech Background and Tutorials
28/515
7/23/2019 Tech Background and Tutorials
29/515
7/23/2019 Tech Background and Tutorials
30/515
7/23/2019 Tech Background and Tutorials
31/515
BEAMPROPAGATIONMETHOD(BPM)
27
conductivities of PML, respectively. The modified differential operator used inEquation 33and Equation 34is defined as
with
where , and are the unit vectors in the , ,and directions, respectively,and the values of are summarized in Table 1.
Figure 2 Optical waveguide sur rounded by PML
(38)
(39)
(40)
'
' 't z'z+=
't xsx
x----- ysy
y-----+=
'z sz-----=
x y, z x y zsx
7/23/2019 Tech Background and Tutorials
32/515
7/23/2019 Tech Background and Tutorials
33/515
BEAMPROPAGATIONMETHOD(BPM)
29
The parameter is written as
where are the free-space wavelength, the PML thickness, theposition of the PML surface, and the theoretical reflection coefficient, respectively.Here, the PML is terminated with the perfect electric or magnetic conductor for TE orTM mode, respectively. Usually, a parabolic is assumed for the conductivity, .
(45)
s
s
1 in non PML region
1 j 1+( )
4dn---------------------
d---
1
Rt---- ln in PML region
=
2c d x0 , , ,=
m 2=
7/23/2019 Tech Background and Tutorials
34/515
7/23/2019 Tech Background and Tutorials
35/515
7/23/2019 Tech Background and Tutorials
36/515
7/23/2019 Tech Background and Tutorials
37/515
7/23/2019 Tech Background and Tutorials
38/515
7/23/2019 Tech Background and Tutorials
39/515
BEAMPROPAGATIONMETHOD(BPM)
35
By substituting Equation 66, Equation 67, and Equation 68into Equation 64, we get:
Thus, using FD scheme into Equation 50for TM modes, we get:
with
(69)
(70)
(71)
(72)
(73)
(74)
rs----
x-----
1
sr------
Hyx
--------- 4
x2
--------r i( )s i( )-----------
Hy i 1( )s i 1( ) s i( )r i( ) r i 1( )+ +-----------------------------------------------------------------------_
1
s x h+( )rx h+( )-----------------------------------------
+1
s x h( )rx h( )---------------------------------------- Hy x( )
Hy x 2h )+(s x h+( )r x h+( )-----------------------------------------
+
+Hy i 1+( )
s i( ) s i 1+( )+( ) r i( ) + r i 1+( )( )----------------------------------------------------------------------------------.
2k0nre fHyz
--------- 2Hy
z2----------- PHy+=
PHyB
x2
--------Hy i 1( ) C
x2
--------Hy i( ) F
x2
--------Hy i 1+( )+ +
B 4r i( )
s i( )--------------
1
s i 1( ) s i( )+( ) r i( ) r i 1( )+( )-------------------------------------------------------------------------------=
C 4r i( )
s i( )--------------
1
s i( ) s i 1+( )+( ) r i( ) + r i 1+( )( )----------------------------------------------------------------------------------=
+1
s i( ) s i 1( )+( ) ( r i( ) r i 1( ) )+---------------------------------------------------------------------------------- + k0
2r i( ) nre f
2( )
F 4r i( )
s i( )--------------
1
s i( ) s i 1+( )+( ) r i( ) + r i 1+( )( )---------------------------------------------------------------------------------- .=
7/23/2019 Tech Background and Tutorials
40/515
7/23/2019 Tech Background and Tutorials
41/515
7/23/2019 Tech Background and Tutorials
42/515
7/23/2019 Tech Background and Tutorials
43/515
7/23/2019 Tech Background and Tutorials
44/515
7/23/2019 Tech Background and Tutorials
45/515
BEAMPROPAGATIONMETHOD(BPM)
41
a) Discretization of the weak di fferential equation
The discretization of Equation 87to a linear set of equation is done by introducing anexpansion for and then making appropriate choices for the weighting functions
. We chose the linear representation
where are the unknown coefficients of the expansion. When , wedefine linear elements. In this case, the basis functions for element are defined as
Figure 5 Nodal expansion function for ethfunctions considering linear approximation
The basis functions have unit magnitude at one node and vanish at all others withlinear variation between the nodes. When ,we have quadratic elements thatare also known as second order elements. Each element has three nodes, one of twoendpoints, and the third is usually placed at the center of the element. Within eachelement, the unknown function is approximated as a quadratic function
(88)
(89)
(90)
(91)
x( )Wm x( )
jex( ) Nj
ex( )j
e
j 1=
eo
=
jex( ) eo 2=
e
N1ex( )
x2e
x
x2e
x1e
---------------=
N2e
x( ) x x1
e
x2e
x1e
---------------.=
eo 3=
ex( ) ae bex cex2.+ +=
7/23/2019 Tech Background and Tutorials
46/515
BEAMPROPAGATIONMETHOD(BPM)
42
Enforcing Equation 91at the three nodes of the element yields:
Solving for ae, be, and ceand substituting them back into Equation 91, we obtain
where the interpolation or expansion functions are given by
(92)
(93)
(94)
1e
ae
bex1
ec
ex1
e( )2
+ +=
2e
ae
bex2
ec
ex2
e( )2
+ +=
3e
ae
bex3
ec
ex3
e( )2.+ +=
e
x( ) Njex( )j
e
j 1=
eo 3=
=
N1e x x 2
e( ) x x 3e( )
x1e
x 2e
( ) x( 1e
x3e )
-----------------------------------------=
N2e x x 1
e( ) x x 3e( )
x2e
x 1e
( ) x( 2e
x3e )
-----------------------------------------=
7/23/2019 Tech Background and Tutorials
47/515
BEAMPROPAGATIONMETHOD(BPM)
43
Figure 6 The nodal expansion function for ethfunctions considering quadratic approximation
(95)N3e x x 1
e( ) x x 2e( )
x3e
x 1e
( ) x( 3e
x2e )
-----------------------------------------.=
7/23/2019 Tech Background and Tutorials
48/515
BEAMPROPAGATIONMETHOD(BPM)
44
When the expansion Equation 92is substituted into Equation 87we get
The terms in the brackets are due to contributions from endpoints of the domain andtheir evaluation is subject to the specific boundary conditions. This equation nowexplicitly shows how the boundary conditions enter into the construction of the linearsystem. Hereon, we will refer to their contributions as endpoints since we have not yetspecified the type of boundary condition to be imposed.
We are now ready to make different choices for the weighting function to generate asystem of linear equations for the solution of . As stated earlier, this step is alsoreferred to as testing and Galerkin's method is usually employed in the finite elementmethod. Specifically, we choose and for each of these testing orweighting functions a single linear equation is generated.
(96)
k02
qs nre f2
( )Wm x( )Nie
x( )
p
s---
dWm x( )dx------------------
dNiex( )
dx----------------- xie
dx1
e
x2e
i 1=
eo
e 1=
Ne
+ psNiex( )Wm x( )dx
2ie
z2---------- 2jk0nre f i
epsNi
ex( )Wm x( ) x
ie
z--------d
x1e
x2e
j 1=
eo
x
1e
x2e
j 1=
eo
pWm x( )d
dx------
x xf
=
pWm x( )d
dx------
x x0=
= 0 .
n{ }
Wm x( ) Njex( )=
7/23/2019 Tech Background and Tutorials
49/515
7/23/2019 Tech Background and Tutorials
50/515
7/23/2019 Tech Background and Tutorials
51/515
BEAMPROPAGATIONMETHOD(BPM)
47
Here .
b) Boundary Condition
The endpoint contributions appear only when or and vanish when theNeumann
or Dirichlet conditions are imposed.
To set the TBC as defined in Transparent Boundary Condition on page 23, wedefine:
Here,
where the real parts of and must be restricted to be positive to ensure onlyradiation outflow.
For Dirichlet, Neumann, and TBC boundary conditions, we set . For PML, weintroduce the parameter as defined by Equation 45.
(105)
(106)
(107)
(108)
I2Ni
ed
xd--------
Nje
d
xd-------- x
1
3le
------7 8 1
8 16 8
1 8 7
.=d
e
=
le x2e
x1e
=
e 1= e m=
d
dx------
x 0=
d
dx------
x xf
=
0==
x 0= x xf
==
ge
jk1 ife 1=
jkfp ife Ne=
0 otherwise.
=
k1 i, j 1
x2 i, x1 i,--------------------
1 i,
2 i,-------
ln=
k1 i, j 1
xf i, xf 1 i,------------------------
f i,f i,-------
ln=
k1 i, km i,
s 1=s
7/23/2019 Tech Background and Tutorials
52/515
BEAMPROPAGATIONMETHOD(BPM)
48
Wide-Angle Beam Propagation Method based on Pad Approximant
Operators
Here, we follow the Pad approximant approach to get the wide angle beam
propagation [31].It is worth it to point out that expansion via Pad is more accurate than Taylorexpansion for the same order of terms. When Pad is employed, larger angles, higherindex contrast and more complex mode interference can be analyzed in both guidedwave and free space as the Pad order increases. See below the guidelines for usingthe Pad technique:
We may formally rewrite Equation 50in the form
Equation 109can be reduced to the following expression:
where and are both polynomials of the operator .
Differentiating Equation 110based on FD scheme, we get
The parameter is introduced to control the finite difference scheme. The range ofis between 0and 1. We can have standard implicit scheme with , explicitscheme with and Crank-Nicolson scheme with . For Crank-Nicolson scheme the propagating field does not diverge as the beam propagates.
The numerator of the term on the right-hand side of Equation 111can be factorized as
where the coefficients can be obtained by solving the algebraicequations
(109)
(110)
(111)
(112)
(113)
z------ j
P 2k0 nre f1 j 2k0 nre f( ) z( )+-----------------------------------------------------.=
z------ j
N
D----,=
N D P
l 1+ D j z 1 ( )N
D j zN+----------------------------------------
l.=
0= 1= 0.5=
AnPN
BnPN 1
CnPN 2 + + +( ) 1 cnP+( ) 1 c2P+( ) 1 c1P+( )=
c1 c2 cn, , ,
D j z 1 ( )N( ) An
PN
Bn
PN 1
Cn
PN 2 + + +( ) 0 .= =
7/23/2019 Tech Background and Tutorials
53/515
BEAMPROPAGATIONMETHOD(BPM)
49
The denominator of the term on the right-hand side of Equation 111can be factorizedas
where the coefficients can be obtained by solving the algebraic equation
Here,
Thus, the unknown field at is related to the known field asfollows
or
In general, determination of the 's and 's requires the one-time solution of an th-order complex algebraic equation.
The explicit expressions for the corresponding Wide-angle (WA) orders using Padapproximant operator are shown below.
The term
in Equation 109can be taken into account quite effectively by using the recursion
relation:
(114)
(115)
(116)
(117)
(118)
AdPN
BdPN 1
CdPN 2 + +( ) 1 dnP+( ) 1 d2P+( ) 1 d1P+( )=
d1 d2 dn, ,
D j zN+ di Pi
i 0=
N
0 .= =
c0 d0 P 0 1= = =
l 1+
z z+ lat z
l 1+ AnP
NBnP
N 1CnP
N 2 + + +
AdPN
BdPN 1
CdPN 2 + + +
------------------------------------------------------------------------l
=
l 1+ 1 cnP+( ) 1 c2P+( ) 1 c1P+( )
1 dnP+( ) 1 d2P+( ) 1 d1P+( )----------------------------------------------------------------------------
l.=
c d n
2z2--------
z-----
n
j P 2k0nre f
1 j 2 k0nre f( ) z( )n 1+------------------------------------------------------------- .=
7/23/2019 Tech Background and Tutorials
54/515
BEAMPROPAGATIONMETHOD(BPM)
50
If Equation 118is now used to replace the derivative in the denominator ofEquation 118we can get the following explicit expressions for several wide-angle(WA) orders.
Fresnel Approximation (Pad 0thorder)
For paraxial approximant, we get the following equation:
here ,
From Equation 119, we get
and therefore
where
(119)
(120)
(121)
(122)
(123)
(124)
(125)
z
z-----
0
j P a
1 j
a---
z-----
1
+
------------------------ jP
a---= =
a 2k0nre f=z-----
1
0=
D 1,=
N P a=
D j z 1 ( )N 1 j z 1 ( )P
a--- 1 AnP+= =
D j zN+ 1 j zP
a---+ 1 AnP+= =
An j z 1 ( )
a------------------------=
Ad j za------.=
7/23/2019 Tech Background and Tutorials
55/515
7/23/2019 Tech Background and Tutorials
56/515
7/23/2019 Tech Background and Tutorials
57/515
7/23/2019 Tech Background and Tutorials
58/515
BEAMPROPAGATIONMETHOD(BPM)
54
Wide Angle (WA), Pad(2,2)
Using recurrence formula Equation 118for Pad(2,2) we get:
Here .
Using Equation 144into Equation 143, we get:
From Equation 145, we get:
Thus,
(143)
(144)
(145)
(146)
(147)
(148)
z----- 3j
P a
1 j a( ) z( ) 2+--------------------------------------------=
z-----
2
j
P
a---
P2
a3
-----+
1 2P
a2
-------+
---------------- .=
a 2k0nre f=
z-----
3
j
P
a---
2
a3
----P2
+
I 3P
a2
------- P
2
a4
-----+ +
---------------------------.=
N P
a---
2
a3
----P2
+=
D 1 3P
a2
------- P
2
a4
-----.+ +=
D j z 1 ( )N 1 3a
2---- j z
1 ( )a
----------------- P+=
+1
a4
---- 2 z1
a3
----------------------j P2 1= BnP AnP
2+ +
7/23/2019 Tech Background and Tutorials
59/515
BEAMPROPAGATIONMETHOD(BPM)
55
and,
where:
Thus, the unknown field is related to the known field asfollows
Multistep Method
In order to solve Equation 154, we applied the multistep method developed byHadley.
First, we rewrite Equation 154as
(149)
(150)
(151)
(152)
(153)
(154)
(155)
D j zN 1 3
a2
---- j z
a---+
P + +=+
1
a4
---- 2 z
a3
-------------j+ P2 1= BdP AdP
2+ +
An1
a4
---- 2 z 1 ( )
a3
---------------------------j ,=
Bn 3a
2---- z 1 ( ) a
------------------------j ,=
Ad1
a4
---- 2 z
a3
-------------j+ ,=
Bd3
a2
---- z
a
----------j.+=
l 1+
atz z+ latz
l 1+ 1 BnP AnP
2+ +
1 BdP AdP2
+ +-------------------------------------
l 1 c2P+( ) 1 c1P+( )1 d2P+( ) 1 d1P+( )
----------------------------------------------l.= =
1 d2P+
1 c2P+------------------
l 1+ 1 c1P+
1 d1P+------------------
l.=
7/23/2019 Tech Background and Tutorials
60/515
BEAMPROPAGATIONMETHOD(BPM)
56
Then, defining the field as
We rewrite Equation 154as
Since is known, we can obtain by solving:
If we consider FEM, we get,
Thus, we can rewrite Equation 158as
here
Using , we rewrite Equation 156as
If we are using FEM, we get,
Solving Equation 162or Equation 162, we can obtain the unknown field . It isapparent from the form of Equation 154that an th-order Pad propagator may bedecomposed into an -step algorithm for which the th partial step takes the form
when .
The run time for an th-order propagator is obviously times the paraxial run time.Therefore, for Pad (2,2) we follow the steps:
(156)
(157)
(158)
(159)
(160)
(161)
(162)
(163)
(164)
l 1 2+
l 1 2+ 1 d2P+
1 c2P+------------------
l 1+.=
l 1 2+ 1 c1P+
1 d1P+------------------
l.=
l
l 1 2+
1 d1P+( )l 1 2+
1 c1P+( )l.=
P Me
[ ] 1
Ke
[ ] k02
nre f2
Me
[ ]( ).=
Me[ ] d1 L
e[ ]+( )l 1 2+ Me[ ] c1 Le[ ]+( )l=
Le[ ] Ke[ ] k0
2nre f
2M
e[ ].=
l 1 2+
1 d2P+( )l 1+ 1 c2P+( )l 1 2+ .=
Me[ ] d2 L
e[ ]+ )l 1+ Me[ ] c2 Le[ ]+( )l 1 2+ .=
l 1+
nn i
l i n+ 1 ciP+
1 diP+
----------------- l i 1( ) n+
=
i 1 2 n, , ,=
n n
7/23/2019 Tech Background and Tutorials
61/515
7/23/2019 Tech Background and Tutorials
62/515
7/23/2019 Tech Background and Tutorials
63/515
BEAMPROPAGATIONMETHOD(BPM)
59
Thus, the unknown field is related to the known field asfollows
Multistep method
In order to solve Equation 183, we use the Multistep method, that is, the unknown fieldcan be obtained from the known field by successively solving Equation 164
for . Therefore for Pad(3,3) we follow the steps:
1 Using , compute considering the linear system:
In case of using FEM solve,
2 Using , compute considering the linear system:
In case of using FEM compute,
(180)
(181)
(182)
(183)
(184)
(185)
(186)
(187)
Ad1
a6
---- 3j z
a5
---------------+=
Bd6
a4
---- 4j z
a3
---------------+=
Cd5
a2
---- j za
----------+ .=
l 1+
atz z+ latz
l 1+ 1 CnP BnP
2AnP
3+ + +
1 CdP BdP2
AdP3
+ + +-------------------------------------------------------
l 1 c3P+( ) 1 c2P+( ) 1 c1P+( )1 d3P+( ) 1 d2P+( ) 1 d1P+( )
----------------------------------------------------------------------l.= =
l 1+
l
i 1 2 3, ,=
l
l 1 3+
1 d1P+( )l 1 3+
1 c1P+( )l.=
Me[ ] d1 L
e[ ]+( )l 1 3+ Me[ ] c1 Le[ ]+( )l.=
l 1 3+
l 2 3+
1 d2P+( )l 2 3+
1 c2P+( )l 1 3+
.=
Me[ ] d2 L
e[ ]+( )l 2 3+ Me[ ] c2 Le[ ]+( )l 1 3+ .=
7/23/2019 Tech Background and Tutorials
64/515
BEAMPROPAGATIONMETHOD(BPM)
60
3 Finally, using , compute the following linear system to obtain theunknown field :
In case of using FEM compute,
(188)
(189)
l 2 3+
l 1+
atz z+
1 d3P+( )l 1+
1 c3P+( )l 2 3+
.=
Me[ ] d3 L
e[ ]+( )l 1+ Me[ ] c3 Le[ ]+( )l 2 3+ .=
7/23/2019 Tech Background and Tutorials
65/515
7/23/2019 Tech Background and Tutorials
66/515
BEAMPROPAGATIONMETHOD(BPM)
62
Thus,
and
here
(195)
(196)
(197)
(198)
(199)
(200)
D j z 1 ( )N 1 7a
2---- j
z 1 ( )a
------------------------ P+=
+15
a4
------ j6 z 1 ( )
a3
--------------------------- P2+ 10
a6
------ j10 z 1 ( )
a5
------------------------------ P3
+1
a8
---- j4 z 1 ( )
a7
--------------------------- P4=1 DnP CnP
2BnP
3AnP
4+ + + +
D j zN 1 7
a2
---- j za
----------+ P 15
a4
------ j6 z
a3
-------------+ P2+ +=+
+10
a6
------ j10 z
a5
----------------+ P3 1
a8
---- j4 z
a7
-------------+ P4+
= 1 DdP CdP2
BdP3
AdP4
+ + + +
An1
a8
---- j4 z 1 ( )
a7
---------------------------=
Bn10
a6
------ 10j z 1 ( )
a5
--------------------------------=
Cn15
a4
------ 6j z 1 ( )
a3
-----------------------------=
Dn7
a2
---- j z 1 ( )a
--------------------------.=
7/23/2019 Tech Background and Tutorials
67/515
BEAMPROPAGATIONMETHOD(BPM)
63
Thus, the unknown field is related to the unknown field asfollows:
Multistep Method
In order to solve Equation 201, we use the multistep method, that is, the unknown fieldis obtained with the following the steps:
1 Compute solving the linear system:
In case of using FEM compute,
2 Compute solving the linear system:
In case of using FEM compute,
3 Compute solving the linear system:
In case of using FEM compute,
(201)
(202)
(203)
(204)
(205)
(206)
(207)
l 1+
atz z+ latz
l 1+ 1 DdP CdP
2BdP
3AdP
4+ + + +
1 DnP CnP
2
BnP
3
AnP
4
+ + + +
--------------------------------------------------------------------------l
=
=1 c4P+( ) 1 c3P+( ) 1 c2P+( ) 1 c1P+( )1 d4P+( ) 1 d3P+( ) 1 d2P+( ) 1 d1P+( )
----------------------------------------------------------------------------------------------l.
l 1+
l 1 4+
1 d1P+( )l 1 4+ 1 c1P+( )l.=
Me[ ] d1 L
e[ ]+( )l 1 4+ Me[ ] c1 Le[ ]+( )l.=
l 1 2+
1 d2P+( )l 1 2+
1 c2P+( )l 1 4+
.=
Me[ ] d2 L
e[ ]+( )l 1 2+ Me[ ] c2 Le[ ]+( )l 1 4+ .=
l 3 4+
1 d3P+( )l 3 4+
1 c3P+( )l 1 2+
.=
Me[ ] d3 L
e[ ]+( )l 3 4+ Me[ ] c3 Le[ ]+( )l 1 2+ .=
7/23/2019 Tech Background and Tutorials
68/515
BEAMPROPAGATIONMETHOD(BPM)
64
4 Finally, knowing , compute the known field solvinglinear system:
In case of using FEM compute,
The advantage of the multistep is that the matrix equation to be solved in each stepis the same size as the Fresnel equation and for 2D problems is tridiagonal when weconsider Finite Difference method (FD) or Finite Element Method (FEM) consideringlinear element.
(208)
(209)
l 3 4+
l 1+
atz zl+
1 d4P+( )l 1+
1 c4P+( )l 3 4+
.=
Me[ ] d4 L
e[ ]+( )l 1+ Me[ ] c4 Le[ ]+( )l 3 4+=
7/23/2019 Tech Background and Tutorials
69/515
7/23/2019 Tech Background and Tutorials
70/515
7/23/2019 Tech Background and Tutorials
71/515
7/23/2019 Tech Background and Tutorials
72/515
7/23/2019 Tech Background and Tutorials
73/515
7/23/2019 Tech Background and Tutorials
74/515
7/23/2019 Tech Background and Tutorials
75/515
7/23/2019 Tech Background and Tutorials
76/515
DIFFUSIONINLITHIUMNIOBATE
72
The diffusion-induced index change is described by the product of the dispersionfactor and the distribution function [1],
where the dispersion factor is different for the ordinary and extraordinary cases( in m)
In turn, the diffusion induced distribution function is a function of the distributionconstant , the dopant concentration profile and the distribution powerfactor ,
where the distribution constants
and
and the distribution power factors
and
are different for the ordinary and extraordinary cases. They are fitted from publishedexperimental data [1].
The concentration profile can be derived following the classical diffusion theory[2], [3]. The profile has a bell-shaped form
In the lateral direction, that is horizontal to the crystal surface, the profile ischaracterized by the combination of error functions. In the crystal in-depth direction,that is vertical to the crystal surface, it has a Gaussian shape. The profile's parametersinclude the profile constant , the dopant stripe width before diffusion , thehorizontal (lateral) diffusion length , and the vertical (in depth) diffusion length .
(4)
(5)
(6)
(7)
di ( ) hi x y,( )
ni x y, ,( ) di ( )hi x y,( ),= i o e.,=
do ( ) 0.67
2
2
0.13---------------------= de ( )
0.8392
2
0.0645---------------------------=
Fi c x y,( )i
hi x y,( ) Fic x y,( )[ ]
i= i o e,=
Fo 1.3 10 25
cm3=
Fe 1.2 10 23
cm3=
o 0.55=
e 1.00=
c x y,( ) c0 erf w
2D
H
---------- 1 2x
w
------+ erf w
2D
H
---------- 1 2x
w
------ +
y2
DV2
------ exp=
c0 wDH DV
7/23/2019 Tech Background and Tutorials
77/515
7/23/2019 Tech Background and Tutorials
78/515
7/23/2019 Tech Background and Tutorials
79/515
7/23/2019 Tech Background and Tutorials
80/515
DIFFUSIONINLITHIUMNIOBATE
76
The distribution function in the vertical direction is defined by the hyper-Gaussiandependence [4], [5]
where is the power of the hyper-Gaussian distr ibuti on, usually ranging between10 and 23, and is the diffusion depth after annealingor effective guide depth.The diffusion depth as well as the maximum refractive index difference afterannealing might have much different values then the ones before annealing. Itis up to the user to define these parameters which can be case dependent. As anexample, we quote empirical relationships to calculate and for annealing at200 and 400C taken from [4], [5]:
where is the initial PE depth, is the initial change of the refractive indexafter proton exchange and are the fitting parameters (Table 3). Therefractive index distribution is strongly dependent on the post-exchange time andthe post-exchange temperature .
In a case when the predefined profiles are found not sufficient, user has the option todefine his own refractive index profiles by using the User Function Profiles in Profile
designer.
Table 2
Table 3
(16)
(17)
Taken from [6] Toluic Acid Benzoic Acid
Temperature range [oC] 109-263 122-249
Diffusion constant [ m2/hr] 7.02 107 7.36 109
Activation energy [kJ/mol] 75.58 94
[oC] b [hrc] c [] p [hrq] q []
300 0.7031 0.0754 0.2325 0.1033
400 1.2884 0.7577 0.3749 0.6574
f y( ) y DV( )a
[ ]exp=
aDV
DVnes
nes DV
DV
DV0( )--------- 1 bta
c
= 1
nes
nes0( )----------- pta
q
=
V
0( )nes
0( )b c p q, , ,
taTa
D0
Q
Ta
7/23/2019 Tech Background and Tutorials
81/515
7/23/2019 Tech Background and Tutorials
82/515
7/23/2019 Tech Background and Tutorials
83/515
7/23/2019 Tech Background and Tutorials
84/515
ELECTROOPTICEFFECT
80
References
[1] E. Strake, G.P. Bava, and I. Montrosset: Modes of channel waveguides: A novel quasi-analyticaltechnique in comparison with scalar finite-element method, J. Light. Technol.6, (1988): 1126-1135.
[2] J. Crank: The mathematics of diffusion. Oxford University Press, New York, (1975).
[3] G.B. Hocker and W.K. Burns: Mode dispersion in diffused channel waveguides by the effectiveindex method,Appl. Opt. 16, (1977): 113-118.
[4] J. Nikolopoulos and G.L. Yip: Accurate modeling of the index profile in annealed proton-exchanged LiNbO3, Proc. SPIE1583, Integrated Optical Circuits, (1991): 71-82.
[5] J. Nikolopoulos and G.L. Yip: Theoretical modeling and characterization of annealed proton-exchanged planar waveguides in z-cut LiNbo3, J. Light. Technol.9, (1991): 864-870.
[6] E.Y.B. Pun, K.K. Loi, and P.S. Chung: Experimental studies of proton-exchanged waveguidesin Lithium Niobate using toluic acid, Proc. SPIE1583, Integrated Optical Circuits, (1991): 64-70.
[7] J.L. Jackel: Proton exchange: past, present, and future, Proc. SPIE1583, Integrated Optical
Circuits, (1991): 54-63.
[8] H. Jin, M. Belanger, and Z. Jakubczyk: General analysis of electrodes in integrated-optics andelectrooptic devices, J. Quant. Electron.27, (1991): 243-251.
[9] H. Jin, R. Vahldieck, M. Belanger, and Z. Jakubczyk: A mode projecting method for the quasi-static analysis of electrooptic device electrodes considering finite metallization thickness andanisotropic substrate, J. Quant. Electron.27, (1991): 2306-2314.
[10] R. Syms and J. Cozens: Optical guided waves and devices, Section 3.7, London, McGraw-Hill,(1992).
[11] A. Yariv: Optical Electronics in Modern Communications. 5th edition, Oxford University Press,(1996).
7/23/2019 Tech Background and Tutorials
85/515
7/23/2019 Tech Background and Tutorials
86/515
7/23/2019 Tech Background and Tutorials
87/515
7/23/2019 Tech Background and Tutorials
88/515
SCATTERINGDATA
84
Figure 3 Basic four -channel Mach-Zehnder multi /demultiplexer
The device can be fully analyzed by the BPM, however we can show it is notnecessary. The distances among the input ports (named with the respect to
multiplexed wavelengths) are typically 0.25 mm and the device is several millimeterslong, the circuit area is quite large. Note, the speed of the BPM simulator isapproximately proportional to the layout area. Further, we can see that the layout is,say, somehow coarse, i.e. we can declare some places where the field exists in theevanescent form and the field is negligible. Running the simulation through theseplaces is inefficient, however we cannot exclude these dead places from therectangular simulation window. Our new approach proposes an elegant solution (seethe pertinent paragraph below).
Multidirectional BPM device
The second category is very close to the previous one. The structure on the figurebelow is another arrangement of the simple MZI. There are two crucial items on this
example. First, the path difference is relatively big, what was designed by a circularloop instead of a primitive arc as done in the first example. The second item appearingrather crucial when BPM would be considered as the simulation tool, is - the input andoutput ports are mutually in perpendicular directions. As the final remark, the so-called dead space again occupies the majority of space in the simulation window(see Figure 4).
7/23/2019 Tech Background and Tutorials
89/515
SCATTERINGDATA
85
Figure 4 Multidirectional BPM device
Devices consisting of the combination of BPM & gratings (Add/drop)
In this category, we shall study the devices, where a grating is present, BPM isunsuitable (see Figure 5). Of course, as it was mentioned we can use CMT, i.e. theOptiGrating product. At this point, we may recognize the final item of our new method.
To analyze some advanced structures by means of several independent techniques,we will have to connect all the particular results using some common way. Moreover,we will need a common environment to do so.
Concerning the functionality of the circuit, two identical Bragg gratings are designedto have the reflection maximum, say, at . For example, the principalparameters of the sine groove grating are as follows: .
Figure 5 Mach-Zehnder interferometer
0 1550nm=L 5mm , n 0.0007==
7/23/2019 Tech Background and Tutorials
90/515
7/23/2019 Tech Background and Tutorials
91/515
7/23/2019 Tech Background and Tutorials
92/515
SCATTERINGDATA
88
reduced form of it, hence it is called S-data matrixthrough the text). The four-portcoupler is then characterized by the following relation:
Equation 1can be easily described in an oral form. The element is responsiblefor a response from the first input waveguide into the first output waveguide. Similarly,the matrix element relates to the response in the second output waveguide whenthe light was introduced into the first input waveguide. The meaning of the other matrixelements is defined in analogous way. Note, the elements are complex numbers andthey consist of real and imaginary parts. Equation 1has to be repeatedly determinedfor each wavelength run. The relations among the input and output ports takes thegeneral form (for one particular wavelength)
We use the following relations to obtain the desired S-data elements. We aresearching for a response on a unitary input modal field (unitary means the unitamplitude and zero phase additional advance of the modal filed). We have to use thefollowing approximation, because it would be somehow cumbersome to transport acomplete information carrying the output field distribution yielded from the BPMsimulation. Concerning the relative power presence at the output waveguide, we thus
compare the actual final BPM field obtained after the simulation at each outputwaveguide with that one being theoretically present if the single straight waveguidelight propagation considered for the pertinent output waveguide. This can be easilydone by well-known power overlap integral
where one field is the BPM one, while the second is the modal field distribution of
output waveguide (the asterisk means the complex conjugated field). Thedenominator, consisting of the product of two integrals, ensures the normalization ofthe overlap integral from 0 to 1. determines the integration region.
(1)
(2)
(3)
b1
b2
S11S12
S21S22
a1
a2=
S11
S12
b1b2
bM
S11 S12 S1NS21 S22 S2N SM1SM2 SMN
a1a2
aN
=
P
E1E2 xd
2
E12
x E22
xd
d
---------------------------------------- ,=
7/23/2019 Tech Background and Tutorials
93/515
SCATTERINGDATA
89
The response with respect to the cumulated phase delay is then given by (both realand imaginary part)
where , is the reference index, is the propagation length. Finally,is an additional angle related to the BPM propagation (this can be observed from
the analyzer). Sometimes in one BPM calculation more than one reference index willbe used. There may be regions having different reference indices. If more then oneregion appears in the layout, we have to characterize these regions separately byparticular lengths and reference indices and the phase delay term in Equation 4must be replaced accordingly as shown schematically in Equation 5
with being the number of regions in the layout. In other words Equation 4simplyadds the phase shift due to the optical length of the wafer. Since the responses weresought for the unit input modal fields, the collected values of for each combination(input/output/wavelength) created according to Equation 2the S-data matrixelements. The exported file is a text file with an '.s ' extension. The first line containstwo integer values: the number of inputs and the number of outputs. All thesubsequent lines represent an individual scattering data matrix for a single
wavelength. They contain the wavelength in microns and all the complex values of theamplitude transmission between each input and output port. The total number ofelements in each row is the number of the inputs times the number of the outputstimes 2 (real and imaginary parts) plus one (wavelength). These lines appear suchthat the wavelength is in increasing order.
Central wavelength approximation
We can, however, employ a more approximate solution which is much faster tocompute. Concerning the wavelength interval, we are mostly dealing with a very smallvicinity of some central wavelength, say. To save more simulation time, we can runthe simulation just for this wavelength to obtain the one set of matrix elements. Let uscall the matrix (the *.sfile will then be just one line with the headings). In most
practical applications, the weighted power distribution coming from the power overlapintegral in an output port varies by a negligible amount. The phase change may becrucial, on the other hand. We can then approximately derive the phase change withrespect to the wavelength. We can approximately write in the vicinity of a centralwavelength
(4)
(5)
(6)
b P j kn0L wg+( )[ ],exp=
k 2 = n0 Lwg
kn0L
n0
L nr
Lr,
r
r
bSij
Sc
c
S k( ) Sc i kc k( )n0L{ },exp
7/23/2019 Tech Background and Tutorials
94/515
SCATTERINGDATA
90
with , while where is the actual wavelength. The valuesare the S-data obtained after the ordinary simulation for the central wavelength. In
conclusion, we may need just one wavelength simulation to describe the deviceoptical response. The limitation of the central wavelength approximation is obvious.
The accuracy is decreased with increasing propagation length as well as with thebroader wavelength interval. The approximation will be also suitable for lower valuesof the reference refractive index.
Implementation with OptiSystem
We are now ready to embark into the main part of the paper. Namely, we have alreadyshown we could divide an advanced circuit into smaller parts that are suitable for theBPM analysis. We have also the S-data tool to provide the effective mathematicaldescription of all those particular sub-elements. The last part missing in our accountis to find an environment connecting the sub-elements into the entire circuit. To reachour goal, we used OptiSystem, which provides a very flexible environment forschematic designing and simulation of advanced optical communication systems.
Any complex circuit can be assumed to be an optical system, and OptiSystem can bevery effectively used as the motherboard in the presented study.
The practical implementation is trivial from a customer point of view. At this point, keepin mind that we have the sub-elements characterized in the form of small files.OptiSystem introduces a general component OptiBPM Component NxMthat canload these small files easily.
The component on the left is shown with its default value of 8x8input/output ports. Apparently, we do not need much additionaldescription saying that the loaded component by the *.s filereads firstly the headings, i.e. the number of both ports andautomatically adapts the icon to have the correct number of
ports (visually can be seen in the solved examples below). TheOptiSystem environment also follows the unidirectional of theBPM, this can be seen from the icon by the orientation of the
port arrows (from left to right). We are now able to fully exploit a number of variousOptiSystem facilities to analyze a given circuit. This brings an immense flexibility intothe circuit simulations. Moreover, we can decrease the simulation time comparing thenew approach with a BPM run. We will demonstrate these improvements by solvingthe four examples previously mentioned.
Solutions
The concluding part of our discourse will show both the complexity and the flexibilityof the presented modeling improvements. The OptiSystem schematic layout brings anew vision of the optical circuit behavior. Any advanced photonic circuit may befurther studied as a set consisting of separated fundamental components. As we willalso see, the use of the OptiSystem platform extends the simulation possibilitiesdramatically, because of a possible entering of various photonic and even electronicsimulation tools.
kc 2 c= k 2 = Sc
7/23/2019 Tech Background and Tutorials
95/515
SCATTERINGDATA
91
Four Channel Mach-Zehnder Multi/Demultiplexer
The realistic layout of the mux/demux example in OptiBPM looks as shown inFigure 8(note the size ratio of 1:10). What first attracts our attention is the presenceof a number of relatively primitive, i.e. straight and curved, waveguides. The layout is
very coarse compared to the entire wafer area. Moreover, the light interactionsactually occur on just a small fraction of the layout. In addition, all six four-portcouplers are identical (see Figure 9).
Figure 8 Mach-Zehnder MUX/DEMUX example in OptiBPMSize ratio 1:10
Let us mark the couplers as shown in Figure 9. At this point we encounter the keypoint of the approach - its division criteria with the respect to the sub-circuitfunctionality. We have to split the circuit into the particular cells. The cells must be selfexisting fully functional units, although the splitting of the whole device must notdisturb its functionality in any way when later composed together by means of theOptiSystem.
7/23/2019 Tech Background and Tutorials
96/515
7/23/2019 Tech Background and Tutorials
97/515
SCATTERINGDATA
93
First the most accurate way is to simulate them by BPM similarly to the coupler - letus call it total analysis with respect to this approach. However, the detailed study isnot needed for all waveguides. We would prefer to do so for the waveguides playinga role between the couplers, since we are primarily interested in the phase differencebetween the arms. The waveguides connecting the ports with the core of the device
(i.e. the outermost waveguides) are simple isolated waveguides engaged in modalpropagation in the fundamental mode. After some preliminary inspection we can evenneglect the losses or we can prescribe them analytically. In other words, a user'sexperience and intuitive estimat