Revista Mexicana de Fsica

ISSN: 0035-001X
rmf@ciencias.unam.mx
Sociedad Mexicana de Fsica A.C.
Mxico

Backes, T.D.; Citrin, D.S.

Out-of-plane plasmon-polariton modes in two-dimensional nanoparticle arrays
Revista Mexicana de Fsica, vol. 56, nm. 2, 2010, pp. 56-58
Sociedad Mexicana de Fsica A.C.
Distrito Federal, Mxico

Available in: http://www.redalyc.org/articulo.oa?id=57030352010

How to cite
Complete issue
Scientific Information System
More information about this article Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal
Journal's homepage in redalyc.org Non-profit academic project, developed under the open access initiative
CONTRIBUTED TALKS REVISTA MEXICANA DE FISICA S 56 (2) 5658 JUNIO 2010

Out-of-plane plasmon-polariton modes in two-dimensional nanoparticle arrays

T.D. Backes
School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta,
Georgia 30332-0250,
e-mail: backes@gatech.edu
D.S. Citrin
Unite Mixte Internationale 2958 Georgia Tech-CNRS, Georgia Tech Lorraine,
Metz Technopole, 2 rue Marconi, 57070 Metz, France,
e-mail: david.citrin@ece.gatech.edu
Recibido el de julio de 2008; aceptado el de octubre de 2008

Plasmon-polaritons in a square, two-dimensional lattice of noncontacting nanoparticles are studied using the coupled dipole model. The
approach considers diffractive losses out of the plane and also for the vector nature of the electromagnetic field, and is thus a three-dimensional
theory of the two-dimensional nanoparticle array. We consider the out-of-plane polarization, where long range coupling is present, and
present the case where the plasmon energy crosses the light cone near the M-point in the Brillouin Zone. In this case, only a restricted set of
nonradiative modes are present.

Keywords: Surface plasmon; nanofabrication; ultrasmall.

Se estudian los plasmares-polaritones en un arreglo cuadrado de partculas, sin tocarse unas con otras, usando el modelo del dipolo acoplado.
La aproximacion considera perdidas difractivas fuera del plano y por la naturaleza del vector del campo electromagnetico, siendo un modelo
tridimensional de un arreglo tridimensional de nanopartculas. Consideramos la polarizacion fuera del plano, donde se presenta acoplamiento
de largo alcance y presentamos el caso donde la energa del plasmon cruza el cono de luz cerca del punto-M de la zona de Brillouin. En este
caso solo esta presente un conjunto restringido de modos no radiactivos.

Descriptores: Plasmon superficial; nanofabricacion; ultrasmall.

PACS: 73.20.Mf; 42.25.Fx; 87.80.Cc

1. Introduction as oscillating point dipole and coupling within the array is

accounted for by the the retarded dipole field. More specifi-
Arrays on noncontacting noble metal nanoparticles (NP) have cally, we apply this technique to show that square arrays can
garnered a great amount of recent interest. The Plasmon po- be suitably designed such that all SPPs are radiative with the
lariton (PP), a hybrid plasmonic optical mode, has the po- exception of a small region in the Brillouin Zone near the
tential to allow for a number of applications, such as inter- M-point.
connects, detectors, and lenses. In this letter, we examine
the square 2D nanoparticle array (NPA) in particular. These
types of arrays are of interest due to the defined surface plas- 2. Calculations
mon (SP) resonances that occur in nanoparticles, in compar-
The dipolar Greens function (GF) G (q, ) for excitation
ison to the broad resonances of metal sheets.
wavevector q and mode , can be used to describe the PPs.
Some studies of 2D arrays have appeared in the litera- This treatment captures both radiative PPs [12], the fields
ture. Sakoda [3] simulated using the finite-difference time-
domain method the dispersion curves of a square arrays of
metal cylinders. The finite-element method was used in [4]
to explore negative refraction in arrays. In Ref. 5, the multi-
ple multipole method was used to analyze triangular lattices
in addition to the square lattice. In all of these studies, the
arrays were of composed infinite cylinders, and out-of-plane
losses were not accounted for.
This paper examines the out-of plane-Plasmon-polariton
modes in square-lattice NPAs using a coupled-dipole model.
This model follows the work done with respect to 1D NP
chains [13, 14]. We treat a NPA, with the NP diameter and
the center-to-center inter-NP spacing d. The dipole approx-
imation has been shown to be accurate except in the case of
NPs that are nearly in contact [17]. The NPs are treated F IGURE 1. Diagram of the nanoparticle array.
 OUT-OF-PLANE PLASMON-POLARITON MODES IN TWO-DIMENSIONAL NANOPARTICLE ARRAYS 57

scattered by the NPA, and nonradiative surface polaritons, the

nearfield standing waves. The PPs are given by the singular-
ities of G (q, ),
G1
(q, )qq0
!
2 2p ip
= 0 (q, ) qq0 , (1)
2p
with p and p the the single-NP surface-plasmon reso-
nance energy and nonradiative damping, respectively. 0
is the radiative decay of a single NP SP, and the re-
tarded dipole-dipole coupling is contained within the radia-
tive self-energy (SE)
0

X
(q, ) = h,j; ()ei(qx hd+qy jd) ,
h,j=
F IGURE 2. Z-polarized Plasmon-polariton real dispersion
which is simply the retarded dipole-dipole energy per NP (the Re(q, k) as functions of q for a Au =50-nm square-lattice
prime indicating that the (h, j) = (0, 0) term is not included). nanoparticle array with interparticle center-to-center spacing
We divide the solution to G1 (q, ) = 0 into real and imag-
d = 75 nm embedded in a semiconductor with complex refractive
inary components E (q) and i (q) and use the pole ap- index n = 3.5 0.05i.
proximations
E (q) p + Re (q, p )
and
(q) Im (q, p )/~.
The SE is the dynamic dipole-dipole energy associated
with NPs (h, j) and (h0 , j 0 ),
(h,j)(h0 ,j 0 ) () = E(h,j),(h0 ,j 0 ) p(h0 ,j 0 )
with E(h,j),(h0 ,j 0 ) the oscillating [as exp(it)] electric field
associated with the SP of NP (h, j) at the position of NP
(h0 , j 0 ) and p(h0 ,j 0 ) is the dipole moment of the SP of NP
(h0 , j 0 ). A schematic diagram of the NPA is shown in Fig.
1. Based on the approach of Refs. [13, 14, 18, 19], the nonva-
nishing SE for the mode transverse to the plane of the array
is
0
X 1 F IGURE 3. Z-polarized Plasmon-polariton radiative loss
(q, ) = n0 3 (h2 + j 2 )3/2 Im(q, k) as functions of q for a Au =50-nm square-
h,j=
k
lattice nanoparticle array with interparticle center-to-center spacing
d = 75 nm embedded in a semiconductor with complex refractive
i 1
2 2 index n = 3.5 0.05i.
k (h + j 2 ) k(h2 + j 2 )1/2
2
+j 2 )1/2 i(qx h+qy j)
eik(h e , (2)
where 0 = 3 p2 /(40 c3 ), ~ = , 0 is the electric con- 3. Results
stant, k = d1/2 /c is the dimensionless optical wavevec-
tor in the background medium, q = qd is the dimensionless We turn our focus on the complex SE. We assume reason-
excitation wavevector, is the dielectric constant,
which in- able values for interparticle spacing d = 75 nm and nanopar-
cludes the gain of the medium, and n = is the complex ticle size = 50-nm. We consider Au NPs since the radia-
index of refraction. It should be noted that the out-of-plane tive decay time is on the order of the nonradiative dephas-
mode SE is completely decoupled from modes that exist in ing time [22]. The embedding medium is chosen to have
the plane of the NPA, and the PPs are purely transverse. In Re n = 3.5, with complex dielectric constant = n2 , and a
order to compute the SE in dimensionless form, we divide the small amount of loss is added for numerical stability. We
SE by 0 such that (q, k) = 01 (q, ). approximate the shift of p due to the embedding medium:

Rev. Mex. Fs. S 56 (2) (2010) 5658

58 T.D. BACKES AND D.S. CITRIN
(p )air
p = , [23]
( 23 + 13 Re n)
and define kp as the optical wavevector evaluated at = p .
The SP energy in air is taken to be (p )air = 2 eV [2022,24].
Figure 2 shows the dispersion Re(q, k), and Fig. 3
shows the radiative loss Im(q, k) as functions of q for the
Z-mode, which is purely transverse. In this case, the photon
dispersion has undergone zone folding almost everywhere in
the BZ once it crosses the SP dispersion, except in a small
region close to the M-point. We see that Re(q, k) exhibits
overall negative phase velocity with an additional feature at
the crossing with the light cone q = k. This appears as a sharp
decline beyond which the dispersion is mostly constant. The
radiative width in a small region close to the M-point is such
that the SP-like SPPs are nonradiative; elsewhere in the BZ,
1. M. Quinten, A. Leitner, J.R. Krenn, and F.R. Aussenegg, Opt.
Lett. 23 (1998) 1331.
2. A. Alu and N. Engheta, Phys. Rev. B 74 (2006) 205436.
3. K. Sakoda, Optical properties of photonic crystals (Springer,
Berlin, 2001).
4. M. Davanco, Y. Urzhumov, and G. Shvets, Opt. Express 15
(2007) 9681.
5. E. Moreno, D. Erni, and C. Hafner, Phys. Rev. B, 65 (2002)
155120.
6. S.I. Bozhevolnyi, J. Erland, K. Leosson, P.M.W. Skovgaard,
and J.M. Hvam, Phys. Rev. Lett. 86 (2001) 3008 .
7. S.I. Bozhevolnyi, V.S. Volkov, , K. Leosson, and A. Boltasseva,
Appl. Phys. Lett. 79 (2001) 1076.
8. F.I. Baida, D. van Labeke, Y. Pagani, G. Al Naboulsi, and G.M.
Al Naboulsi, J. Microscopy 213 (2002) 144.
9. P.T. Worthing and W.L. Barnes, Proc. SPIE 4655 (2002) 132.
10. A. Taleb, V. Russier, A. Courty, M.P. Pileni, Phys. Rev. B 59
(1999) 13350.
11. S. Zou and G.C. Schatz, J. Chem. Phys. 121 (2004) 12606 .
12. V.A. Markel, J. Mod. Opt. 40 2281 (1993).
13. D.S. Citrin, Nano Lett. 4 1561 (2004).
14. D.S. Citrin, Nano Lett. 5 (2005) 985 .
15. A.F. Koenderink and A. Polman, Phys. Rev. B 74 (2006) 33402.
16. A.F. Koenderink, R. de Waele, J.C. Prangsma, and A. Polman,
Plasmon dispersion in metal nanoparticle chains from angle-
resolved scattering, arXiv:070.2731v1 [cond-mat.other].
Rev. Mex. Fs. S 56 (2) (2010) 5658
radiative decay is prevalent. At this particular interparticle
spacing, beam forming can occur.
4. Conclusion
We have presented a theory for the out-of-plane polarized
PPs in NPAs with square symmetry. We have examined a
region where the light cone has undergone zone folding near
the X-point, but not in a small region near the M-point. This
q dependence allows for the restriction of nonradiative dark
modes that do not have radiative losses to a small region in
the Brillouin Zone.
Acknowledgments
This work was supported in part by the National Science
Foundation grant NSF DMR-0305524, and by NASA grant
NNX07-AR54H.
17. S.Y. Park and D.G. Stroud, Phys. Rev. B 69 (2004) 125418.
18. D.S. Citrin, Opt. Lett. 31 98 (2006).
19. D.S. Citrin, J. Opt. Soc. Amer. B 22 (2005) 1763.
20. S.A. Maier, P.G. Kik, and H.A. Atwater, Appl. Phys. Lett. 81
1714 (2002).
21. S.A. Maier, M.L. Brongersma, P.G. Kik, and H.A. Atwater,
Phys. Rev. B 65 193408 (2002).
22. C. Sonnichsen et al., Phys. Rev. Lett. 88 (2002) 77402.
23. K.L. Kelly, E. Coronado, L.L. Zhao, and G.C. Schatz, J. Phys.
Chem. 107, (2003) 668.
24. T. Klar, M. Perner, S. Grosse, G. von Plessen, W. Spirkl, and J.
Feldmann, Phys. Rev. Lett., 80 (1998) 4249.
25. T.V. Shahbazyan, I.E. Perakis, and J.-Y. Bigot, Phys. Rev. Lett.
81 (1998) 3120.
26. T.V. Shahbazyan and I.E. Perakis, Phys. Rev. B 60 (1999) 9090.
27. L.A. Coldren and S.W. Corzine, Diode Lasers and Photonic
Integrated Circuits, (Wiley, New York, 1995).
28. D.S. Citrin, Opt. Lett. 20, (1995) 901.
29. D.S. Citrin, Phys. Rev. B 48 (1993) 2535.
30. D.S. Citrin, Phys. Rev. Lett. 69 (1992) 3393.
31. V.M. Agranovich and O.A. Dubovskii, Pisma Zh. Eksp. Teor.
Fiz. 3 (1966) 345 [JETP Lett. 3 (1966) 223].
32. F. Tassone and F. Bassani, Nuovo Cimento 14D (1992) 1241.
33. S. Jorda, Solid State Commun. 87 (1993) 439.