Surface Plasmon Resonance. Magneto-optical optical enhancement and other possibilities Applied Science Department The College of William and Mary
Plasmonics Recently surface plasmons have attracted significant attention for a variety of exciting applications (e.g. metamaterials, cloaking, etc.) May 2008 - Surface-Plasmon circuitry T. W. Ebbesen, C. Genet, S. I. Bozhevolnyi
Outline Introduction: Surface Plasmon resonance based sensors Magneto-plasmonic sensors Au-Co-Au Trilayers Gratings Au-Co nanocomposites What is next? Conclusions
1. Surface plasmon resonance for biosensing
Surface Plasmon Resonance When light strikes a conducting thin film it is possible to excite a surface plasmon polariton i.e. charge oscillations in the metal that lead to evanescent surface electromagnetic waves propagating along a metal/dielectric interface. For the surface plasmon resonance to be excited, the incident light wave vector must match the surface plasmon resonance momentum. This is possible when: The surface plasmon resonance is highly confined at the interface, and therefore is very sensitive to the dielectric optical properties.
How can SPR be excited? ħk SP > ħk 0
Application: How is SPR used in bio-sensing? A glass slide with a thin gold coating is chemically modified to be able to bind to specific bio-agents. The slide is mounted onto a prism. Light passes through the prism and slide, reflects off the gold and passes back through the prism to a detector Changes in reflectivity versus angle or wavelength give a signal that is proportional to the volume of bio-agent bound near the Au surface.
Reflectivity Fundamentals When surface plasmon resonance is excited, it radiates light backwards. The electromagnetic field is highly enhanced at the metal/ dielectric surface interface 1.0 0.5 prism Au film 0.0 0 200 400 Au thickness (A) 600 800 60 70 80 90 Incidence angle (deg) The Au films thickness can be optimized to achieve full extinction in the reflected beam-> this is the optimum excitation condition for surface plasmon resonance
2. Magneto-optical optical effects and surface plasmon resonance
Fundamentals Since the electromagnetic field is strongly enhanced inside the Au film when the surface plasmon resonance is excited, the introduction of a magnetic film can cause strong enhancement of its magneto-optical activity. Au Au Co C. Hermann, PRB 63, 235422 (2001)
Magneto-optical optical Kerr effect The light that is reflected from a magnetized surface can change in both polarization and reflectivity. This results from the off-diagonal components of the dielectric tensor ε. MOKE can be further categorized by the direction of the magnetization vector with respect to the reflecting surface and the plane of incidence.
Transverse MOKE When the magnetization is perpendicular to the plane of incidence and parallel to the surface it is said to be in the transverse configuration. In this geometry, the MOKE effect results in a change in reflectivity that is proportional to the component of magnetization that is perpendicular to the plane of incidence and parallel to the surface. Further, the surface plasmon is also affected: J. B. González-Díaz et al., PRB 76, 153402 (2007).
System studied Au (3 nm) Co (2.5-6 nm) Au (20 nm) Au-Co-Au tri-layer samples were grown on glass with DC sputtering. Accurate control of the growth rate allowed precise control of the layers thickness. Au and Co thickness were designed to achieve: Optimum excitation of the surface plasmon resonance Maximum enhancement of the MO activity.
Preparation Preparation: sputtering deposition Sputtering System High purity and thickness control! Base pressure 10-9 Torr (UHV). Rheed, Quadrupole in-situ. Substrate temperature:rt- 700ºC. 6 magnetron sputtering guns Gas: Ar, etc... Deposition rates (P Ar =5.10-3 Torr) Au 0.32 Å/s Co 0.066 Å/s Cr 0.13 Å/s Ni 0.12 Å/s Ag 1.03 Å/s
Characterization solenoid H prism R R pp pp H detector HeNe laser
Custom SPR station
Custom SPR station
Au-Co-Au Au (20 nm) Au (3 nm) Co (2-10 nm) glass 1.0 46 0.5-2.350-2.039 0.0-1.728-1.417 R R -0.5-1.0-1.5 pp p p-2.0 30 35 40 45 50 Co thickness (A) 55 60 44 45 46 Angle (deg) Angle (deg) 45 44 30 35 40 45 50 55 60 Co thickness (A) R R -1.106-0.7950-0.4840-0.1730 0.1380 0.4490 0.7600 pp pp
Deviation of the optical constants from bulk values 3.2 n 2.8 nm n 3.5 nm n 5nm n 6 nm n Bulk 5 n 2.4 4 k 1.6 3 2.1 2.8 Energy (ev) 2.0 2.5 3.0
Results (I) The thickness of all the metallic layers was designed to achieve full extinction of the reflected intensity. 0.4 6 nm Co 5 nm Co 4 nm Co 2.8 nm Co 2. 5 nm Co 0.02 R (0) 0.2 R (0) 0.01 0.00 0.0 42 44 46-0.01 44.1 44.4 44.7 incidence angle (deg) incidence angle (deg)
Results (II) Transverse Kerr magneto-optical signal With no prism (i.e. the Au surface plasmon is not excited) With prism (plasmon excited) 4 R (H)-R (0) ( ) 0.6 0.4 0.2 6 nm Co 5 nm Co 4 nm Co 3 nm Co 2.5 nm Co R (H)-R (0) ( ) 3 2 1 0-1 6 nm Co 5 nm Co 4 nm Co 3 nm Co 2.5 nm Co 40 45 50 Incidence angle (deg) -2 40 45 50 incidence angle (deg) The measured signals are normalized with respect to the incident excitation. When the surface plasmon is excited we observe ~ one order magnitude enhancement in the transverse magneto-optical Kerr signal.
Results (III) Combining the enhancement of the MO effect and the extinction of the reflected beam, a remarkable enhancement of the relative fielddependent variation of the reflectivity is obtained Rpp Rpp ( Ms ) Rpp (0) = R R (0) pp pp R (H)-R (0)/ R (0) (% ) 150 100 50 0-50 6 nm Co 5 nm Co 4 nm Co 2.8 nm Co 2.5 nm Co R (H)-R (0)/ R (0) (% ) 10 5 0 6 nm Co 5 nm Co 4 nm Co 2.8 nm Co 2.5 nm Co -100 42 44 46 48 42 44 46 incidence angle (deg) incidence angle (deg)
Measurements in water solutions Reflectivity (R) Field-dependent dependent R/R air water n=1.33 water+glycerine n=1.3375 water+glycerine n=1.3475 1.5 1.0 air water n=1.33 water+glycerine n=1.3375 water+glycerine n=1.3475 R pp R pp /R pp 0.5 0.0 40 45 50 55 60 65 70 75 incidence angle (deg) -0.5 40 45 50 55 60 65 70 75 incidence angle (deg) Angle shift
Sensitivity: Typical metric to compare sensors Sensitivity R pp R pp = = % RIU n d 1 170,800 % RIU -1 in water 280,000 % RIU -1 in air 10000 1000 100 MO-SPR (Cr-Co-Cr-Au) 19,100 % RIU -1 Sensitivity (% RIU -1 ) 10 1 water air air 0 20 40 60 80 100 Co Thickness (Angs.) B. Sepúlveda et al. Opt. Letters 31, 1085 (2006). SPR 3,900 % RIU -1 J. Homola et al. Sens. and Act. 54, 3 (1999).
The Physics: Plasmon Excitation, Electric Field depth dependence and Magneto-optical optical enhancement Grown 3 samples with the Co film placed in three different positions Co (2.8 nm)-au (23 nm) Au(11.5 nm)-co (2.8 nm)-au (11.5 nm) Au(20 nm)-co (2.8 nm)-au (3 nm)
12 Rpp 101308 Co 2.8nm-Au23nm Rpp 101508 Au 11.5nm-Co 2.8nm-Au 11.5nm c Rpp 101408 Au 20nm-Co 2.8nm-Au 3nm Rpp 101308 Co 2.8nm-Au23nm 6 0 40 45 50 theta Reflectivity. We note that the position of the minimum changes because the conditions to excite the plasmon have changed.
Experimental magneto-optical data DRpp 101308 Co 2.8nm-Au23nm 20000 10000 0-10000 DRpp 101308 Co 2.8nm-Au23nm DRpp 101508 Au 11.5nm-Co 2.8nm-Au 11.5nm DRpp 101408 Au 20nm-Co 2.8nm-Au 3nm 40 45 50 DRpp/Rpp 101308 Co 2.8nm-Au23nm 2 1 0-1 -2-3 DRpp/Rpp 101308 Co 2.8nm-Au23nm DRpp/Rpp 101508 Au 11.5nm-Co 2.8nm-Au 11.5nm DRpp/Rpp 101408 Au 20nm-Co 2.8nm-Au 3nm 40 45 50 theta theta R pp R pp /R pp
6 Derivative Rpp Co2.8-Au23 Derivative Rpp Au 11.5-Co2.8-Au11.5 Derivate Au20-Co2.8-Au3 Derivative Rpp Co2.8-Au23 4 2 0-2 -4 40 45 50 theta The derivative of R pp : does not evolve in the same manner as the experimental R pp -> the changes observed in R pp, i.e. the magnetooptical response, are not related to modification of the plasmon excitation in the samples.
Simulations (transfer matrix formalism) 0.0020 DRpp Co2.8-Au23 DRpp Au11.5-Co2.8-Au11.5 DRpp Au20-Co2.8-Au3 6 DRppRpp Co2.8-Au23 DRppRpp Au11.5-Co2.8-Au11.5 DRppRpp Au20-Co2.8-Au3 DRpp Co2.8-Au23 0.0015 0.0010 0.0005 0.0000 DRppRpp Co2.8-Au23 4 2 0-2 -0.0005 40 45 50-4 40 45 50 A A
Electric Field Air interface Au20-Co2.8-Au3 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 Air interface Au20-Co2.8-Au3 Co (34 A from air) Au20-Co2.8-Au3 Glas interface Cortes Au20-Co2.8-Au3 40 42 44 46 48 50 angle (deg) Air interface Cortes Au11.5Co2.8Au11.5 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 Air interface Cortes Au11.5Co2.8Au11.5 angle (deg) Co (244 A from air) Cortes Au11.5Co2.8Au11.5 Glas interface Cortes Au11.5Co2.8Au11.5 40 42 44 46 48 50 angle (deg) Air interface Cortes Co2.8Au23 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 Air interface Cortes Co2.8Au23 Co (244 A from air) Cortes Co2.8Au23 Glas interface Cortes Co2.8Au23 40 42 44 46 48 50 angle (deg)
Electromagnetic field in the middle of the Co film in the 3 samples 1.6 Co (244 A from air) Cortes Co2.8Au23 Co (244 A from air) Cortes Au11.5Co2.8Au11.5 Co (34 A from air) Au20-Co2.8-Au3 E 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 40 42 44 46 48 50 angle (deg) Thus, the field enhancement due to SPP excitation enhances also the MOKE
Simulations polar Rotation 70 60 50 40 30 20 10 0-10 rotacion Au20Co2.8Au3 rotacion Au115-Co28-Au115 rotacion Co2.8Au23 40 45 50 ellipticity 49 42 35 28 21 14 7 0-7 -14-21 -28 elipticidad Au20Co2.8Au3 elipticidad Au115-Co28-Au115 elipticidad Co2.8Au23 40 45 50 angle (deg) angle (deg) Our simulations also indicate dramatic enhancement of the polar Kerr rotation and ellipticity. We will investigate this experimentally.
Adhesion issues: Cr-Au-Co-Au In order to explore the material for a possible bio-sensing application there are additional concerns. Au (3 nm) Co (2.8 nm) Au (9.5-10.5 nm) Cr (3 nm) Adhesion of Au on glass is poor. Tri-layers are degraded when exposed to a water flux. Cr has been extensively used to improve the adhesion of Au on glass, but it is a highly absorptive metal and therefore it broadens the surface plasmon resonance peak. At present time the common belief has been that the introduction of Cr layers decreases the sensitivity of these kind of sensors. We have demonstrated that this is not true.
Results: Cr-Au-Co-Au The thickness of the layers was once more designed to achieve the full extinction of the reflected intensity Rpp Cr 3 nm-au 10.5 nm-co 2.8 nm-au 3 nm normalized Rpp Cr 3 nm-au 9.5 nm-co 2.8 nm-au 3 nm normalized Rpp Cr 3 nm-au 9 nm-co 2.8 nm-au 3 nm normalized 0.2 1.0 R(0) 0.0 0.8 0.6 Rpp Cr 3 nm-au 10.5 nm-co 2.8 nm-au 3 nm Rpp Cr 3 nm-au 9.5 nm-co 2.8 nm-au 3 nm Rpp Cr 3 nm-au 9 nm-co 2.8 nm-au 3 nm 45 48 incidence angle (deg) R(0) 0.4 0.2 0.0 39 42 45 48 51 54 57 60 63 incidence angle (deg)
Results: Cr-Au-Co-Au Transverse Kerr magneto-optical signal 4 R (H)-R (0) ( ) 2 0 Rpp Cr 3 nm-au 10.5 nm-co 2.8 nm-au 3 nm Rpp Cr 3 nm-au 9.5 nm-co 2.8 nm-au 3 nm Rpp Cr 3 nm-au 9 nm-co 2.8 nm-au 3 nm 42 45 48 R (H)-R (0) ( ) 3 2 1 0-1 -2 40 45 50 incidence angle (deg) 6 nm Co 5 nm Co 4 nm Co 3 nm Co 2.5 nm Co Recall transverse Kerr effect Without Cr buffer layer incidence angle (deg) A small decrease in the normalized signal is observed due to increased absorption in the Cr buffer layer
Results: Cr-Au-Co-Au Yet, combining the enhancement of the MO effect and the extinction of the reflected beam, again a remarkable enhancement of the relative variation of the reflectivity is obtained. 250 DRpp/Rpp Cr 3 nm-au 10.5 nm-co 2.8 nm-au 3 nm DRpp/Rpp Cr 3 nm-au 9.5 nm-co 2.8 nm-au 3 nm DRpp/Rpp Cr 3 nm-au 9 nm-co 2.8 nm-au 3 nm 200 R (H)-R (0)/ R (0) (% ) 150 100 50 R (H)-R (0)/ R (0) (% ) 6 5 4 3 2 1 0 DRpp/Rpp Cr 3 nm-au 10.5 nm-co 2.8 nm-au 3 nm DRpp/Rpp Cr 3 nm-au 9.5 nm-co 2.8 nm-au 3 nm DRpp/Rpp Cr 3 nm-au 9 nm-co 2.8 nm-au 3 nm 0-1 40 45 50 55 60 incidence angle (deg) -2 40 45 50 55 incidence angle (deg)
0 60 40 20 100 80 Sensitivity R pp R pp Sensitivity = = % RIU n d 1 703,000 % RIU -1 in air 280,000 % RIU -1 in air SPR 3,900 % RIU -1 J. Homola et al. Sens. and Act. 54, 3 (1999). Co Thickness (Angs.) 10000 Sensitivity (% RIU -1 ) 1000 100 10 1 Au-Co-Au Cr-Au-Co-Au 120
Detection limit 703,000 % RIU -1 in air n min =1.42 x10-7 RIU 280,000 % RIU -1 in air 170,800 % RIU -1 in water MO-SPR (Cr-Co-Cr-Au) 19,100 % RIU -1 B. Sepúlveda et al. Opt. Letters 31, 1085 (2006). n min =3.57 x 10-7 RIU n min =5.85 x 10-7 RIU n min = 5 x 10-6 RIU SPR 3,900 % RIU -1 J. Homola et al. Sens. and Act. 54, 3 (1999). n min = 5 x 10-5 RIU
Conclusions A large enhancement of the magneto-optical response of Au-Co-Au trilayers with and without Cr buffer layer was obtained when the surface plasmon resonance was excited. Layer thickness was designed to achieve maximum extinction of the reflected beam. Combining both effects, a remarkable enhancement of the relative change in reflectivity ( Rpp/Rpp) was obtained. This feature can significantly improve the detection limit in sensors based on surface plasmon resonance.
WORK IN PROGRESS We have achieved field modulated enhanced SPR in trilayered Au-Co Co-Au samples and also with trilayers grown on a Cr buffer layer. We are now testing these sensors in liquids. We are also investigating the use of diffraction gratings nano-patterned on the sensor surface to couple the light to the surface plasmons.. This approach can eliminate constrains on the thickness of the films deposited and the kind of substrate used.
Diffraction gratings and plasmons
Nano-patterning We have explored e-beam lithography to nanopattern magneto-plasmonic materials with two goals in mind: Use diffraction gratings for photons-plasmons coupling Explore localized enhancement of the electromagnetic field to further enhance the magneto-optical activity
Au film with e-beam patterned grating 7.6 inside outside 7.2 800 nm 6.8 6.4 Rpp 6.0 5.6 5.2 4.8 0 9 18 27 36 45 54 63 72 81 theta
Au-Co-Au trilayer and Nano-patterned grating on top Reflectivity (arb. units) grating continuous film 0 10 20 30 40 50 60 70 80 θ incidence angle (deg.) R/R Relative change in reflectivity 0.6 0.5 0.4 0.3 0.2 0.1 0.0 grating continuous film 0 20 40 60 80 θ incidence angle (deg.)
Au-Co nanocomposites Sputtering codeposition of Au and Co Alternative solution for the adherence issue Decrease Co Absorption Easier to prepare Au 95% - Co 5 % Au 40% - Co 60 % Deposition temperature RT, 300 C, 600 C
Au-Co nanocomposites: morphology 50 nm thick films 600 C 300 C Heigth (nm) 100 50 0 600 C 300 C RT 0 500 1000 Profile (nm) RT Good adhesion to glass.
300 C samples 10% Co-90% Au 30% Co-70% Au 50% Co-50% Au Ms (emu/cc) 500 0-500 -5 0 5-5 0 5-5 0 5 H (koe) H in plane H perpendicular Ms (emu/cc) 600 400 200 0 0 10 20 30 40 50 Co Concentration (% ) In plane magnetic anisotropy Magnetic moment scales with Co concentration
Au-Co nanocomposites: MO-SPR Measurements in air Au 20%- Co 80 % grown @ 300 C 0.06 R(0) R(H)-R(0) / R(0) 0.04 0.02 30 40 50 60 incidence angle (deg) 0.00 30 40 50 60 incidence angle (deg)
The future Investigate the metal-insulator transition in VO 2 films grown on glass. The films will be excited with IR laser radiation following Cavalleri s work (PRL, 2001) We will then investigate the effect of this MI transition on plasmonic structures deposited/patterned on VO 2 films.