Nanoplasmonics Enhancement of optical processes is severely limited by the metal loss

Transcription

1 Nanoplasmonics Enhancement of optical processes is severely limited by the metal loss Surface plasmon polaritons Why do we need metals in nanophotonics? How does the metal loss limit the nanoplasmonic enhancement? Can one compensate the SPP loss with gain?

2 Can light flow down on the surface like a water drop? Water drop gravity Surface tension liquid/substrate SPP (surface plasmon-polariton) Light drop Electromagnetic force TM pol. EM frequency/metal

3 Surface plasmons (Gary Wiederrecht, Purdue University) Definitions: collective excitation of the free electrons in a metal Can be excited by light: photon-electron coupling (polariton) Thin metal films or metal nanoparticles Bound to the interface (exponentially decaying along the normal) Longitudinal surface wave in metal films Propagates along the interface anywhere from a few microns to several millimeters (long range plasmon) or can be extremely confined in nanostructures (localized plasmon) Surface plasmon-polaritons (SPP) = SPs + photons

4 Near-field profile of SPPs L SP 1/e 1 SP Metal D SP Air Why do we need metals in photonics? How does the propagation loss (Ohmic damping) limit nanoplasmonic enhancement? : SPP-LEDs, SPP-solar cells, SPP-sensors Can one compensate the SPP loss with gain? : SPP nanocavity lasers

5 Why metals? Plasmonics: the next chip-scale technology -limit Plasmonics e-limit photons Plasmonic devices, therefore, might interface naturally with similar speed photonic devices and similar size electronic components. For these reasons, plasmonics may well serve as the missing link between the two device technologies that currently have a difficult time communicating. By increasing the synergy between these technologies, plasmonics may be able to unleash the full potential of nanoscale functionality and become the next wave of chip-scale technology. SPPs

6 Metals can support a guided mode with x-ray wavelength at optical frequencies Dispersion relation of SPP: k SPP c m d m d 36 nm X-ray wavelengths at optical frequencies SP resonance when ( ) m SP d p 1 d SiO 2 Ag Therefore we need a metal with negative permittivity 2 p m( ) 1 2 d i

7 Metals Silver Re[()] Permittivity () Dispersion relation Propagation length J & C Exp Drude model E [ev] SP Ag/air light line air SP Ag/glass light line glass 3 6 [nm] Gold Re[] E [ev] J & C Exp Drude model E [ev] SP Au/air light line air kx [um -1 ] SP Au/glass light line glass L [um] [nm] E [ev] kx [um -1 ] L [um] Copper Re [] J & C Exp Drude model E [ev] SP Cu/air light line air SP Cu/glass light line glass 3 6 [nm] E [ev] kx [um -1 ] L [um]

8 Fundamental parameters of SPP modes 1. Propagation and Absorption Indices: complex effective metal / index 1+ n k k n i SP SP SP SP metal propagation index absorption index 2. Propagation length: L SP 2 SP ( : wavelength) 3. Effective number of oscillations (ohmic Q-factor): Q L n L ( / n ) SP SP SP SP SP SP SP 1 for a good resonance mode 4. Mode size (range of evanescent field tail ): DSP 2 2 n 1 SP

9 Ag, Au, Al, Cu, Pt, Pd Aluminum looks good for plasmonics in visible range, better than gold in some senses!

10 Guiding and focusing of SPPs using metal tips D. E. Chang, A. S. Sørensen, P. R. Hemmer, and M. D. Lukin, Strong coupling of single emitters to surface plasmons, PR B 76,3542 (27) M. I. Stockman, Nanofocusing of Optical Energy in Tapered Plasmonic Waveguides, Phys. Rev. Lett. 93, (24)

11 Dispersion relation of metal nanotips For a thin, nanoscale-radius wire y m x d k nk For, the phase velocity v c/ n( z) p and the group velocity v c d n d The time to reach the point R = (or z = ) g / ( )/ n SPP R Intensity Energy density

12 Why are the velocities so slow? Dielectric Metal Total Energy Flux : Medium 1: Medium 2: p p p f p ˆ 1 dzw1( z) x 4 f f p ˆ 2 dzw 2( z) x 4 f ( f / k ) Always opposite to each other by the surface confinement condition

13 Apllications Permittivity of a metal ( ) 1 m 2 2 p p i / 2 2 p Dispersion relations k SPP d m c d m 1/2

14 Type-A : low k Type-A - Low frequency region (IR) - Weak field-confinement H. Won, APL 88, 1111 (26). - Most of energy is guided in clad - Low propagation loss clad sensitive (sensor) applications SPP waveguides applications

15 Type-A: Plasmonic waveguides for interconnections Jung (ETRI), 4 Gbit/s light signal transmission on a long-range SPP waveguide, APL, PTL, 27. Tx Drive IC TIA & Pre amp IC LR-SPP waveguide SMA VCSEL array PD array SMA Rx 14 nm-thick, 2.5 m-wide gold stripes Gb/s World best Loss (db) = 131 nm Waveguide length (cm).6 db/cm : World best record in propagation loss. (Previous world record : 3.2 db/cm by Berini, 26)

16 Double-electrode metal waveguides : S-band, Y-branch Song, Long-range surface -plasmon--polaritons on asymmetric double-electrode structures, APL, 28. D metal strip SPP mode d3 w 2 d1 D core metal slab d3 cladding S-band metal strip metal slab Y-branch

17 Type-A : SPP sensors SPR (surface plasmon resonance) on metal-coated prism Metal SPP waveguide 35 3 Output signal Intensity (uw) Reference arm Sensing arm Refractive index of water

18 LRSPP waveguides sensors with a -fluidic channel Bragg grating -fluidic channel cladding (b) cladding (c) 45 nm cladding cladding (d) (e) (f)

19 But, dielectric GMR sensors are also excellent. Fast - instant results Outstanding accuracy cross referenced data High sensitivity detection of small molecules to large bacteria High resolution sharp detection peaks, high signal to noise Mass producible high density formats Reflectance (1) Baseline (2) After analyte binds Sensor element Initial market applications in drug discovery and proteomics: Antigen-antibody assays, peptides and cell-based assays, DNA arrays Wavelength ()

20 Type-B : middle k Type-B - Visible-light frequency region - Coupling of localized field and propagation field Nano-hole - Moderated field enhancement Sensors, display applications Extraordinary transmission of light

21 Type-B: Color filters using EOT T (%) T (db) FDTD Analytic model Hz Wavelength (nm) h air MC D 1 metal D 2 n = 1.75 R, A (%) R, A (%) FDTD Analytic model D 2 Glass-side R 1 A 1 R 2 A 2 Air-side incidence incidence Wavelength (nm)

22 Type-C : high k Ag p-gan Type-C QW n-gan - UV frequency region - Strong field confinement - Very-low group velocity Nano-focusing, Nano-lithography SP-enhanced LEDs, solar cells Light emission QW

23 Type-C : SP Nano Lithography

24 Type-C : SP solar cells Plasmonic structures can offer at least three ways of reducing the physical thickness of the photovoltaic absorber layers while keeping their optical thickness constant. H. Atwater ans A. Polman, Plasmonics for improved photovotaic devices, Nature Materials, 9, March 21. metal nanoparticles at the surface of the solar cell Light is preferentially scattered and trapped into the semiconductor thin film metal nanoparticles embedded in the semiconductor. Light trapping by the excitation of localized surface plasmons corrugated metallic film on the back surface of a thin photovoltaic absorber layer Light trapping by the excitation of surface plasmon polaritons at the metal/semiconductor interface.

25 Why do we need metal? How does the propagation loss limit the nanoplasmonic enhancement? : SPP-LEDs, SPP-solar cells, SPP-sensors Can one compensate the SPP loss with gain? : SPP nano-cavity lasers L SP 1/e 1 SP D SP metal air

26 Consider a Type-C application: SPP-assisted LEDs Conventional LED R E R nr R SP enhanced LED SP E R E R nr SP R R R SP SP

27 Purcell enhancement factor of surface plasmon-polaritons SE Rate : R f i ( ) ( ) p 2 E Dipole moment of the radiating source Photon DOS (Density of States) Electric field strength of half photon (vacuum fluctuation) Purcell Factor : F P R RSP 1 1 ( ksp / k) 1 R 2 L / ( v / c ) SP We need a slow and tightly confined mode

28 Purcell factor F P 1 k SP / k SP V / c' k U SP E ( a) SP 2 MQW P-GaN separation (a) The gap separation less than 2 nm is required for F P > 1.

29 But, the metal loss is severe in such a small separation. Fraction D A C B B Distance [nm] A A : Lossy surface wave mode B : Surface plasmon mode C : Direct radiation mode D : Balance At a = 2 nm, A : 25.1 % B : 55.9 % C : 19 % High nonradiative recombination rate due to the metal loss

30 The SP approach was started for organic LEDs ITO glass (anode) Organic molecules Cathode & Mirror SPP quenching (~4%) Strongly coupled to SPPs Main issue: SPP Radiation coupling Nanostructures on metal mirror Metallic thin film SPP2 SPP1 SPP band gap ( ~ / ) k SPP Direct coupling ( / k SPP ) SPP cross-coupling ( /[ k k ]) SPP1 SPP2

31 Cross-Coupled vs Coupled SPP (1) (2) (3) (4)

32 SPP Enhanced PL of InGaAs QW Most cited paper Un-processed (a) Half-processed (b) Fully-processed (c) 48nm period (2 nd order coupling) (d) 25nm period (1 st order coupling) (16nm gap)

33 SP propagation length PL SPs 1 2 k k m d c m d 3 2 ( ) 2 m m 2 Nanopatterning Propagation Length of SPs [nm] Surface Plasmon on the Ag/GaN Interface Wavelength of Photon [nm] Green LEDs might be possible. Frequency (2c/m) = sp, 2sp, 3sp, 53nm sp~14 nm SP-dispersion on Ag/GaN 46nm sp~7 nm In-plane Wavevector (2 /m)

34 Grating on p-gan Little damage to p-gan Enlarged surface area for low contact resistance

35 Wafer-scale fabrication of ~ 1 nm patterns laser (266 nm) Shutter Rotation stage Pinhole Objective Lens Wafer holder with rotator Mirror z y x Aperture Photoresist Mirror x y z NANO EGGBOX rotator 6/1

36 EL measurement of the fabricated SPP-LED

37 An Optimistic Estimation of SP-enhanced LEDs At green (53 nm) with a 1 st order grating MQW 5 nm 1 nm grating depth FDTD calculation 2 nm 2.3 times more Photons generated 6 nm 14 nm 1 nm grating period 18 nm Good directionality Surface plasmon Photons escaped % 34.1% within 2 o after escape 1/(2n 2 ) = 8 % Wavelength (nm) (Bare-chip LED with 8 % extraction) ( Optimized LED with 5 % extraction) (82 % / 8 %) x 2.3 ~ 24 times Brighter (82 % / 5 %) x 2.3 ~ 4 times Brighter

38 Extraction efficiency of highly confined surface plasmon-polaritons to far-field radiation: an upper limit J. Yoon, et. al, Opt Exp, Vol. 16, 1269 (28). F SP F E R E E ( 1 E ) F Rtotal ( 1 ) FP SP SP 1 SP SP SP SP SP P E SP even if F to get FSP 1 P We need an efficient grating coupler Silver Λ = nm.

39 But, for a reasonable F P < 1: F SP R E E ( 1 E ) F Rtotal ( 1 ) FP 1 SP SP SP SP SP P Jacob B. Khurgin et. al, J. Opt. Soc. Am. B, Vol. 24, 1968 (27). F SP, opt SP ( max, p, opt ) ( 1 cosmax ) For 1-dim. Grating structure Substantial improvement can be achieved only for very inefficient emitters with < 1 %.

40 Figures of merit: Enhancement by SPPs Jacob B. Khurgin et. al, J. Opt. Soc. Am. B, Vol. 24, 1968 (27). SPP enhancement of spontaneous radiation is most noticeable only if the original radiative efficiency of the emitter is very small, far less than 1 %. For this reason it does not appear that SPP offers any advantage for LEDs. The only possibly exception is Si emitters whose original radiative efficiency is very low. The main application of SPP enhancement should remain as improving the efficiency of weak photoluminescence and nonlinear processes Nano resonators with gain?

41 Why do we need metal? How does the propagation loss limit nanoplasmonic enhancement? Can one compensate the SPP loss with gain? SPP nano-cavity lasers L SP 1/e 1 SP D SP metal air

42 M.A. Noginov, et al, Demonstration of a spaser-based nanolaser, Nature, 46, August 29. R. Oulton, et al, Plasmon lasers at deep subwavelength scale, Nature, 461, October 29. (Jacob B. Khurgin) Rate equation for the number of electrons in the SPP mode dn dt SPP gn ( 1) n SPP SPP When the modal gain g is close to compensating the loss, 14 ( SPP / )~ ~1 / dn dt s i e A inj J 2 ~ ~ 1 for -scale A/ cm for (5 nm) -area!!!!!!

43 Challenges of SPPs Ekmel Ozbay, Science, vol.311, pp (13 Jan. 26). Some of the challenges that face plasmonics research in the coming years are.. (i) (ii) demonstrate optical frequency subwavelength metallic wired circuits with a propagation loss that is comparable to conventional optical waveguides; Too lossy. develop highly efficient plasmonic organic and inorganic LEDs with tunable radiation properties; Not efficient. (iii) achieve active control of plasmonic signals by implementing electro-optic, all-optical, and piezoelectric modulation and gain mechanisms to plasmonic structures; Not enough gain. (iv) demonstrate 2D plasmonic optical components, including lenses and grating couplers, that can couple single mode fiber directly to plasmonic circuits; Long-range SPPs only on Ag or Au, not Cu or Al. (v) develop deep subwavelength plasmonic nanolithography over large surfaces.?? 2 nd and 3 rd Nonlinearity (SHG, SERS), Nano antennas (guiding and focusing), Sensors,.. Ultrafast processing by the enhanced SE rate.

44 How about another polaritons: SMPs, instead of SPPs? Surface Magnetic(magneton)-Polaritons Surface Magnetic Polariton (SMP) H Normalized energy flow densities N S N S Coupling to TE-pol. light J. Yoon, et. al, Opt Exp, Vol. 13, 417 (25).