hig he r o rde r me tho ds fo r simula ting l ig ht pro
play

Hig he r-o rde r Me tho ds fo r Simula ting L ig ht Pro pa g a tio - PowerPoint PPT Presentation

Kur t Busch, Univer sit t Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Hig he r-o rde r Me tho ds fo r Simula ting L ig ht Pro pa g a tio n a nd L ig ht-Ma tte r I nte ra c tio n in Na no -Pho to nic Syste ms K urt Busc h I


  1. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Hig he r-o rde r Me tho ds fo r Simula ting L ig ht Pro pa g a tio n a nd L ig ht-Ma tte r I nte ra c tio n in Na no -Pho to nic Syste ms K urt Busc h I nstitut für T he o re tisc he F e stkö rpe rphysik Unive rsitä t K a rlsruhe , 76128 K a rlsruhe , Ge rma ny DF G-Ce nte r fo r F unc tio na l Na no struc ture s a nd K a rlsruhe Sc ho o l o f Optic s & Pho to nic s phot onics.t fp.uni-kar lsr uhe.de

  2. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Ac kno wle dg me nts L a sha T ke she la shvili Mic ha e l K ö nig Je ns Nie g e ma nn Ja n Gie se le r Ma rtin Po to tsc hnig * K a i Sta nnig e l phot onics.t fp.uni-kar lsr uhe.de

  3. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Mo tiva tio n L ine a r, no nline a r a nd q ua ntum o ptic a l pro b le ms in na no -pho to nic syste ms invo lve multiple time a nd le ng th sc a le s T his re q uire s a c c ura te , sta b le , a nd e ffic ie nt so lve rs fo r line a r a nd no nline a r Ma xwe ll’ s e q ua tio n a nd So lito n c o llisio n in a c o llisio n in a So lito n c o uple d syste ms fib e r Bra g g g ra ting fib e r Bra g g g ra ting phot onics.t fp.uni-kar lsr uhe.de

  4. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Mo tiva tio n: Sta nda rd Appro a c he s F inite -E le me nt-Me tho d F DT D-Me tho d Disc re tiza tio n o n unstruc ture d g rids Disc re tiza tio n o n Ye e -g rid 2 nd o rde r in spa c e a nd time Hig he r-o rd e r in spa c e F re q ue nc y-d o ma in pre fe rre d E ffic ie nt a nd e a sy to imple me nt phot onics.t fp.uni-kar lsr uhe.de

  5. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Mo tiva tio n: Do no t trust Co mpute rs I phot onics.t fp.uni-kar lsr uhe.de

  6. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Mo tiva tio n: Do no t trust Co mpute rs I I phot onics.t fp.uni-kar lsr uhe.de

  7. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Mo tiva tio n: Do no t trust Co mpute rs I I I phot onics.t fp.uni-kar lsr uhe.de

  8. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Mo tiva tio n: Do no t trust Co mpute rs I V Silicon rods in air Two-level atom (initially excited) phot onics.t fp.uni-kar lsr uhe.de

  9. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Outline T he K rylo v-Sub spa c e / Disc o ntinuo us Ga le rkin Appro a c h – Ho w the me tho d wo rks a nd pe rfo rms – Adva nc e d spa tia l disc re tiza tio n E xte nsio n to No nline a r & Co uple d Syste ms – L a wso n-T ra nsfo rma tio n a nd Ro se nb ruc k-Wa nne r so lve rs – Pe rfo rma nc e E xa mple s a nd Applic a tio ns – Spo nta ne o us e missio n in pho to nic c rysta ls – Pla smo nic struc ture s phot onics.t fp.uni-kar lsr uhe.de

  10. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Outline T he K rylo v-Sub spa c e / Disc o ntinuo us Ga le rkin Appro a c h – Ho w the me tho d wo rks a nd pe rfo rms – Adva nc e d spa tia l disc re tiza tio n E xte nsio n to No nline a r & Co uple d Syste ms – L a wso n-T ra nsfo rma tio n a nd Ro se nb ruc k-Wa nne r so lve rs – Pe rfo rma nc e E xa mple s a nd Applic a tio ns – Spo nta ne o us e missio n in pho to nic c rysta ls – Pla smo nic struc ture s phot onics.t fp.uni-kar lsr uhe.de

  11. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de T he K rylo v-Sub spa c e Me tho d Ma xwe ll e q ua tio ns in Sc hrö ding e r fo rm A fo rma l so lutio n o f is g ive n b y phot onics.t fp.uni-kar lsr uhe.de

  12. � Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de T he K rylo v-Sub spa c e Me tho d Disc re tiza tio n o f a nd (e .g . o n a Ye e -Grid ) Ve ry la rg e b ut spa rse ma trix Ma trix-Ve c to r-Pro d uc ts a re fe a sa b le We do no t re q uire the full ma trix , o nly its a c tio n o n a ve c to r: We do no t wa nt a ny re stric tio ns o n the pro pe rtie s o f the ma trix (suc h a s ske w-symme try e tc .) phot onics.t fp.uni-kar lsr uhe.de

  13. � Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de T he K rylo v-Sub spa c e Me tho d Build up the Krylov Subspa c e Ortho -no rma lize the b a sis b y Arno ldi-me tho d Ortho g o na l Ba sis Ob ta in pro je c tio n o f o nto he numb e r o f b a sis ve c to rs c a n b e sma ll (m ~ 10) T phot onics.t fp.uni-kar lsr uhe.de

  14. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de T he K rylo v-Sub spa c e Me tho d T he ke y a ppro xima tio n the n is Wo rks fo r a rb itra ry ma tric e s T he a c c ura c y o f the me tho d is a t le a st Me mo ry usa g e : ( m +1)/ 2 re la tive to F DT D J. Nie g e ma nn, L . T ke she la shvili, a nd K . Busc h, J. Co mput. T he o r. Na no sc i. 4 , 627 (2007) phot onics.t fp.uni-kar lsr uhe.de

  15. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Co mpa riso n o f Pe rfo rma nc e (1D) T he me tho d a llo ws muc h la rg e r time ste ps Krylov (m=4) Krylov (m=8) Krylov (m=16) Krylov (m=32) FDTD phot onics.t fp.uni-kar lsr uhe.de

  16. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Co mpa riso n o f Pe rfo rma nc e (2D) I n a 2D syste m, the e ffe c t is e ve n mo re pro no unc e d ) Krylov (m=16) 2 ) Krylov (m=8) FDTD 4 3 = = m m ( ( v o v l y o r K l y r K phot onics.t fp.uni-kar lsr uhe.de

  17. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de I mpo rta nt Add -Ons - Via ADE s Dispe rsive Ma te ria ls – Drude -, L o re ntz-, De b ye -Mo de l – Se llma ie r-type Mo de ls So urc e s Ope n Syste ms: Co mple x fre q ue nc y shifte d PML s phot onics.t fp.uni-kar lsr uhe.de

  18. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Ma te ria l Dispe rsio n via ADE s All typic a l a na lytic dispe rsio n re la tio ns (Drude , L o re ntz, De b ye ) c a n b e imple me nte d via ADE s. E xpe rime nta l dispe rsio n fitte d b y c o mb ine d (multiple ) L o re ntz- o r Drude -te rms. E xa mple : (Sing le L o re ntz-te rm) phot onics.t fp.uni-kar lsr uhe.de

  19. � � Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Adva nc e d Spa tia l Disc re tiza tio n With the K rylo v-sub spa c e me tho d , a c c ura c y o f time -inte g ra tio n c a n b e c hose n a rbitra rily Pro b le m: E rro r fro m the spa tia l disc re tiza tio n is limiting the to ta l a c c ura c y Hig he r o rde r ste nc ils Still o nly 2nd o rde r in the pre se nc e o f b o unda rie s Po ssib le so lutio n: Ad a ptive g rid re fine me nt a ro und b o unda rie s phot onics.t fp.uni-kar lsr uhe.de

  20. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Unstruc ture d Grid in 1D Ada pt the g rid so the po int de nsity is hig he r a ro und the ma te ria l b o unda rie s n=1 n=1.5 phot onics.t fp.uni-kar lsr uhe.de

  21. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Unstruc ture d Grid Pe rfo rma nc e (1D) 4 th -o rde r ste nc il a nd a da ptive g rids: 4 th o rde r is ma inta ine d in the pre se nc e o f ma te ria l b o unda rie s FDTD N=399 Krylov (m=8) Krylov (m=32) Krylov (m=64) Krylov (m=16) Krylov (m=4) N=799 N=1599 N=3199 K . Busc h e t a l., physic a sta tus so lidi (b ) 244 , 3479 (2007) phot onics.t fp.uni-kar lsr uhe.de

  22. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Unstruc ture d Grids in 2D/ 3D Disc o ntinuo us Ga le rkin finite e le me nt te c hniq ue (b o rro we d fro m hydro dyna mic s) phot onics.t fp.uni-kar lsr uhe.de

  23. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Re sults o n Unstruc ture d Grids (2D) phot onics.t fp.uni-kar lsr uhe.de

  24. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de Outline T he K rylo v-Sub spa c e / Disc o ntinuo us Ga le rkin Appro a c h – Ho w the me tho d wo rks a nd pe rfo rms – Adva nc e d spa tia l disc re tiza tio n E xte nsio n to No nline a r & Co uple d Syste ms – L a wso n-T ra nsfo rma tio n a nd Ro se nb ruc k-Wa nne r so lve rs – Pe rfo rma nc e E xa mple s a nd Applic a tio ns – Spo nta ne o us e missio n in pho to nic c rysta ls – Pla smo nic struc ture s phot onics.t fp.uni-kar lsr uhe.de

  25. � � Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de E xte nsio n to No nline a r Syste ms T he me tho d c a n b e e xte nde d to no nline a r syste ms L a wso n-T ra nsfo rma tio n: H is the line a r pa rt o f the no nline a r syste m Ro se nb ruc k-Wa nne r so lve rs: H is the Ja c o b ia n o f the no nline a r syste m phot onics.t fp.uni-kar lsr uhe.de

  26. Kur t Busch, Univer sit ät Kar lsr uhe, kur t @t fp.physik.uni-kar lsr uhe.de E xte nsio n to No nline a r Syste ms L a wso n-T ra nsfo rma tio n phot onics.t fp.uni-kar lsr uhe.de

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend


More recommend