Reviews in Plasmonics

Reviews in Plasmonics
Chris D. Geddes Editor
Reviews in
Plasmonics
2016
Editor
Dr Chris D. Geddes, Professor
More information about this series at http://www.springer.com/series/7164
Chris D. Geddes
Editor
Reviews in Plasmonics 2016
123
Editor
Chris D. Geddes
Institute of Fluorescence
University of Maryland Baltimore County
Baltimore, MD, USA
ISSN 1555-7731 ISSN 1555-7677 (electronic)

ISBN 978-3-319-48080-0 ISBN 978-3-319-48081-7 (eBook)
DOI 10.1007/978-3-319-48081-7
Library of Congress Control Number: 2017930597
© Springer International Publishing AG 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, express or implied, with respect to the material contained herein or for any
errors or omissions that may have been made.
Printed on acid-free paper
This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
This is the third volume in the Plasmonics series,

Reviews in Plasmonics by Springer (http://www.
springer.com/series/7164). The first two volumes were
very well received by the Plasmonics community with
several notable reviews/commentaries of the volumes
and series as a whole.
In this 2016 volume, we are again delighted with the
broad and timely Plasmonics content from contributors
around the world. We subsequently thank the authors for
their very timely and exciting contributions again this
year. We hope you all will find this volume as useful as
the previous volumes.
In closing, I would like to thank both Tanja Koppejan
and Meran Owen at Springer for their help in compiling this volume and with the
broader series.
Professor Chris D. Geddes, BSc., Ph.D., CChem., CPhys, CSci., FInstP, FRSC
Director: The Institute of Fluorescence,
Professor of Chemistry and BioChemistry,
University of Maryland Baltimore County,
September 12, 2016, Baltimore, MD, USA
http://theinstituteoffluorescence.com/
http://www.chrisgeddes.com/
v
Contents
1 Plasmonic Nanowire Waveguide for Deep Subwavelength

Confinement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Huizhong Xu
2 Plasmon-Enhanced Fluorescence of Rare Earth Nanocrystals . . . . . . . . 15
Ya-Fang Zhang, Jia-Hong Wang, Ming Chen, Si-Jing Ding,
Zhi-Jun Luo, Yong-Jie Lin, Ya-Lan Wang, Xue-Feng Yu,
Zhong-Hua Hao, Li Zhou, and Qu-Quan Wang
3 Sensing Through Surface Plasmon Resonance Technique . . . . . . . . . . . . . 39
R.K. Gupta
4 Fractal Plasmonic Nanoantennae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Ravi Sadananda Hegde
5 Compact Slow-Light Enhaced Plasmonic Waveguide
Refractive Index Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Yin Huang, Changjun Min, and Georgios Veronis
6 Fabrication, Properties and Applications of Plasmene Nanosheet . . . . 109
Qianqian Shi, Dashen Dong, Kae Jye Si, and Wenlong Cheng
7 Experimental Observation of Melting of the Effective
Minkowski Spacetime in Cobalt-Based Ferrofluids . . . . . . . . . . . . . . . . . . . . 137
Igor I. Smolyaninov and Vera N. Smolyaninova
8 Surface Plasmon Resonance Based Fiber Optic Sensors
Utilizing Zinc Oxide Thin Films and Nanostructures . . . . . . . . . . . . . . . . . . 159
Banshi D. Gupta and Rana Tabassum
vii
viii Contents
9 Frontiers of Light Dynamics in Photonic Crystals . . . . . . . . . . . . . . . . . . . . . . 199

Renju Rajan, P. Ramesh Babu, and K. Senthilnathan
10 LSPR Biosensing: Recent Advances and Approaches . . . . . . . . . . . . . . . . . . 211
Pranveer Singh
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
Chapter 1
Plasmonic Nanowire Waveguide for Deep
Subwavelength Confinement
Huizhong Xu
Abstract This chapter discusses the basic principles and experimental imple-
mentation of plasmonic nanowire waveguides where surface plasmon modes play
an important role as a result of the careful choice of the waveguide materials
and its geometry. Two waveguide designs are given to demonstrate the concept:
simple dielectric nanowire waveguides and hybrid plasmonic waveguides. In the
case of simple nanowire waveguides, the dielectric constants of the core dielectric
and the cladding metal are matched to enable the presence of semi-propagating
modes in waveguides of extremely small size. As an example, numerical modeling,
fabrication, and optical characterization of ZnO nanowire waveguides are then
discussed. The operation of the hybrid plasmonic waveguide is based on hybrid
modes which can possess strong confinement of surface plasmon modes and long-
range propagation of cylindrical modes at the same time. The unique properties
of these modes are studied using numerical modeling and a scheme allowing the
efficient excitation of these modes is also given. Applications of these plasmonic
nanowire waveguides in areas such as near-filed scanning microscopy are also
briefly discussed.
Keywords Extraordinary transmission • Hybrid waveguide • Nanowire

waveguide • Near-field scanning microscopy • Subwavelength confinement •
Surface plasmons
1.1 Introduction
Waveguides capable of achieving high confinement with low loss are a key goal
in realizing plasmonic circuits and networks of high efficiency and throughput
[1–8]. These waveguides can find applications in a variety of fields such as
high-resolution imaging and sensing [9–11], subwavelength lithography [12], and
H. Xu ()
Department of Physics & Astronomy, San Francisco State University, 1600 Holloway Avenue,
94132, San Francisco, CA, USA
e-mail: huizhong@sfsu.edu
© Springer International Publishing AG 2017 1

C.D. Geddes (ed.), Reviews in Plasmonics 2016, Reviews in Plasmonics,
DOI 10.1007/978-3-319-48081-7_1
2 H. Xu
high-efficiency solar cells [13, 14]. Many approaches to achieving propagation of

extended range while maintaining nanoscale confinement have been developed in
the past decade or so. For example, novel waveguide devices where surface plasmon
polaritons (SPPs) play an important role in a chain of nanoparticles [15] and a
hybrid design [16, 17] have been studied and exhibit superior confinement and
propagation properties. Sub-wavelength apertures of various shapes [18–31] and
with dielectric filling [32–35] have also been studied as ways of achieving efficient
light transmission. Furthermore, in the case of dielectric waveguides, it has been
demonstrated that propagating modes can exist inside a waveguide of arbitrarily
small size when the properties of the dielectric core and the cladding metal
are matched [36]. In this chapter, the basic principles and potential applications
of plasmonic nanowire waveguides where the materials are carefully chosen to
resonantly excite SPPs will be discussed. This concept can be used in both simple
cylindrical plasmonic waveguides [37, 38] and hybrid plasmonic waveguides of
complex design [39] to demonstrate deep subwavelength confinement with superior
propagation properties.
1.2 Cylindrical Waveguide
A dielectric cylinder of radius a surrounded by a cladding metal of finite conduc-

tivity is schematically shown in Fig. 1.1. The z-direction propagating constant is
denoted by kz , and "1 , "2 , 1 and 2 are the complex relative permittivity and
permeability of the core and the cladding at the specified wavelength, respectively.
For this waveguide, the allowed modes are so-called hybrid modes (HE and EH),
which can be expressed as a linear combination of the well-known TE and TM
modes [40]. In the cylindrical coordinate system (, , z), Hz of the TE mode is
given by:
(
AJm .k1 / sin m exp .ikz z/ ; < a
Hz D .1/ ; (1.1)
BHm .ik2 / sin m exp .ikz z/ ; a
Fig. 1.1 Schematic of a

dielectric cylinder surrounded
by a cladding metal forming a
dielectric waveguide
1 Plasmonic Nanowire Waveguide for Deep Subwavelength Confinement 3
and Ez of the TM mode is given by:

(
CJm .k1 / cos m exp .ikz z/ ; < a
Ez D .1/ ; (1.2)
DHm .ik2 / cos m exp .ikz z/ ; a
where A, B, C and D are unknown coefficients to be determined, m is the azimuth

.1/
mode order, Jm and Hm are the mth order Bessel function and first-kind Hankel
function respectively. The two propagation constants are defined as:
q
k1 D k0 "1 1 kz2 =k02 (1.3a)
q
k2 D k0 kz2 =k02 "2 2 (1.3b)
where k0 D 2/ is the free space wave vector. The propagating constant kz can be
determined by matching the boundary conditions at D a, leading to the following
equation:
" .1/ 0
# " .1/ 0
#
1 Jm0 .k1 a/ 2 Hm .ik2 a/ "1 Jm0 .k1 a/ "2 Hm .ik2 a/

k1 Jm .k1 a/ ik2 Hm.1/ .ik2 a/ k1 Jm .k1 a/ ik2 Hm.1/ .ik2 a/
(1.4)
m2 k2 1 1 2
D 2 2z C :
k0 a k12 k22
In most cases, the relative permeability of the dielectric core and the cladding
metal can be taken to be one. We also focus on the first-order modes (m D 1), to
which incident plane waves can be efficiently coupled. It is convenient to introduce
the effective refractive index for a mode, defined as neff D kz /k0 . Therefore, the
original problem of finding modes becomes solving Eq. (1.4) for neff . To obtain
the dispersion relations for the various modes in the waveguide, Eq. (1.4) can be
solved numerically by assuming a Drude model for the dielectric constant of the
metal [35, 41, 42].
For a given incident wavelength, we are interested in determining the cutoff size
of a waveguide beyond which propagating modes no longer exist. To illustrate this,
we choose silver to be the cladding metal and the incident wavelength to be 488 nm,
and its refractive index ("Ag D 7.879 C 0.736i) is interpolated from Ref. [43]. We
then numerically compute the effective refractive index of the fundamental mode
HE11 as a function of the radius-to-wavelength ratio. Shown in Fig. 1.2 is a plot
of the real and imaginary parts of the effective refractive index of the fundamental
mode HE11 , denoted as n0eff and n00eff respectively, versus the radius-to-wavelength
ratio a/.
For a dielectric core made of air (see Fig. 1.2a), we found that n0eff (solid lines)
quickly drops for values of a/ around 0.20 where n00eff (dashed lines) quickly
increases, indicating the cutoff of the waveguide. The cutoff value of 0.20 for the
4 H. Xu
3
(a) real
imaginary
2.5
effective refractive index
1.5
0.5
0
0.03 0.08 0.13 0.18 0.23 0.28
radius/wavelength
3
(b) real
imaginary
2.5
1.5
0.5
0
0.03 0.08 0.13 0.18 0.23 0.28
radius/wavelength
Fig. 1.2 Real (solid line) and imaginary (dashed line) parts of the effective refractive index of the
fundamental mode HE11 as a function of the radius-to-wavelength ratio for (a) an air waveguide
and (b) a water waveguide in a silver film
radius-to-wavelength ratio is not far away from the theoretical value of 0.29 for the
TE11 mode for a cylindrical waveguide made of a perfect conductor. We also find
that the cutoff value can be reduced by filling the core with a dielectric material as
a result of the decrease of the wavelength inside the waveguide. For example, for a
dielectric core made of water (see Fig. 1.2b), we found that n0eff (solid lines) quickly
drops for values of a/ around 0.13 where n00eff (dashed lines) quickly increases,
indicating a cutoff radius-to-wavelength ratio of 0.13. Therefore it may be possible

to get even smaller cutoff ratios when the core is filled with dielectric materials of
higher refractive index.
1.3 Plasmonic Nanowire Waveguide
As shown previously [36], when the following condition is satisfied:
"1 C "2 0; (1.5)
an approximate solution to Eq. (1.4) can be obtained in the limit of small radius-to-
wavelength ratio. Therefore, in order for an extremely small waveguide to support
a propagating mode, one needs to match the dielectric constants of the core and the
cladding such that they cancel out each other. This condition can be satisfied by a
waveguide made of a real cladding metal and a dielectric core of high refractive
index such that they cancel each other. Under the above condition, k1 simplifies
p
to k0 "1 "2 and is approximately zero. The small k1 implies that the fields of
this mode have a weak dependence on the radial coordinate, which suggests that
this mode is of plasmonic nature. In fact, Eq. (1.5) is the exact condition required
to resonantly excite SPPs at the corresponding planar metal-core interface in the
electrostatic limit [44]. We also note that for sufficiently small holes one needs to
take into account the nonlocal nature of the complex dielectric constant [35, 42, 45].
For the hole sizes discussed in this chapter, we expect such effects to be negligibly
small.
To provide an example of such plasmonic dielectric waveguides, we choose
silver and ZnO to be the materials for the cladding metal and the dielectric core
respectively. The dielectric constant of ZnO ("ZnO D 4.244) is obtained from Ref.
[46]. The effective refractive index of the fundamental mode HE11 is computed as a
function of the waveguide diameter for a fixed incident wavelength of 488 nm, and
the result is shown in Fig. 1.3. We can see that n0eff (solid lines) quickly drops while
n00eff (dashed lines) quickly increases for values of a/ around 0.05. Below this cutoff
ratio, the imaginary part n00eff is much greater than the real part n0eff , indicating that
the mode quickly attenuates and thus becomes evanescent. This demonstrates that
this waveguide may be able to allow the transmission of light at waveguide sizes
below /10.
In fact, we have used the finite element method to calculate the transmission of
488 nm light polarized in the x-direction through a ZnO waveguide in a 100-nm-
thick silver film situated on top of a fused silica substrate with water as the exit
medium [37]. As shown in Fig. 1.4, the transmission spectrum displays a resonance
peak at a diameter of 40 nm and the normalized transmission can exceed 40 %.
Using a single-mode model, we showed that this strong resonant transmission is a
result of both the small cutoff ratio and the excitation of SPPs at both end interfaces
of the waveguide, which is evident from the intensity profile near the entrance and
the exit as shown in Fig. 1.5.
6 H. Xu
3
real
imaginary
2.5
1.5
0.5
0
0.03 0.08 0.13 0.18 0.23 0.28
radius/wavelength
Fig. 1.3 Real (solid lines) and imaginary (dashed lines) parts of the effective refractive index of
the fundamental mode HE11 as a function of the radius-to-wavelength ratio for a ZnO waveguide
in a silver film
45
40
35
normalized transmission (%)
30
25
20
15
10
0
20 30 40 50 60 70 80 90 100
diameter (nm)
Fig. 1.4 Normalized transmission versus waveguide diameter for 488 nm linearly polarized light
incident on a ZnO waveguide in a 100-nm-thick silver film situated on a fused silica substrate. The
medium on the exit side is assumed to be water (Modified from Fig. 1.2 of Ref. [37])
Fig. 1.5 Electric field intensity distribution of 488 nm linearly polarized light transmitted through
a 40-nm-diameter ZnO waveguide in a 100-nm-thick silver film situated on a fused silica substrate.
The medium on the exit side is assumed to be water. False colors denote the square of the norm of
the electric field (normalized by the incident electric field) in a logarithmic scale
The strong resonant transmission of visible light through ZnO nanowire waveg-
uides was observed in recent experiments [38]. Using a hydrothermal method [47,
48], ZnO nanowires with diameters ranging from 30 nm to 60 nm are synthesized
on glass coverslips coated with a 10-nm-thick gold film. These nanowires are
then embedded in a 100-nm-thick silver film to form ZnO nanowire waveguides.
The fabricated devices are shown in Fig. 1.6. Optical transmission measurements
are then made by illuminating the nanowire waveguides from the silver side and
collecting the transmitted light from the glass side. A resonance in transmission at a
wavelength of 484 nm is observed for an average nanowire dimeter around 42 nm.
The transmission amplitude is in good agreement with finite element method sim-
ulations when the experimental conditions including the presence of the gold film
and the length and orientation of the nanowires are taken into consideration [38]. We
note that it is also possible to take advantage of Fabry-Perot resonances to enhance
transmission when the waveguide is operating in the propagating mode [33, 36].
This phenomenon of extraordinary transmission through plasmonic nanowire
waveguides in a real metal may be useful for implementing a nanoscopic near-field
8 H. Xu
Fig. 1.6 Scanning electron microscopy images of ZnO nanowires embedded in silver. (a) Plan
view; (b) cross-sectional view
probe with superb transmission even at aperture diameters of /20 [36]. It is then
possible to design a near-field scanning probe with the near-field intensity three
orders of magnitude greater than an air-filled aperture and a total end size of only
60 nm, which is considerably smaller than currently available near-field scanning
probes of more than 200 nm [49]. Such a small end size could be particularly
advantageous in cellular imaging applications where minimal disturbance to the
sample is desired [50].
1.4 Hybrid Plasmonic Waveguide
The plasmonic nanowire waveguide can achieve deep subwavelength confinement,

however it cannot offer long-range propagation as result of the noticeable
attenuation in the semi-propagating mode. To achieve excellent confinement and
Fig. 1.7 Schematic of a multi-mode hybrid waveguide (Modified from Fig. 1.1 of Ref. [39])
propagation at the same time, Oulton and coworkers [17] have proposed a hybrid
waveguide design for efficient guiding of light at telecommunications wavelength
(1550 nm). The hybrid waveguide geometry is made of a dielectric-metal interface
between SiO2 and Ag, and a GaAs cylinder placed inside the SiO2 half with a
nanometer-sized gap above the SiO2 /Ag interface. By taking advantage of the
strong confinement offered by the SPPs at the SiO2 /Ag interface and the superior
propagation of the fundamental mode supported by the GaAs cylinder, this hybrid
waveguide is able to simultaneously achieve subwavelength confinement and long-
range propagation.
A similar waveguide design has recently been proposed by Colanduoni et al.
[39] to achieve efficient guiding of 488 nm light with enhanced confinement and
propagation. The waveguide geometry is composed of a dielectric-metal interface
between water and silver, and a horizontal cylindrical ZnO nanowire situated inside
water with a nanometer-sized gap above the water-silver interface, as shown in
Fig. 1.7.
As a result of matching the dielectric constants of the ZnO cylinder and the silver
metal, this waveguide is able to support multiple modes of strong confinement and
long-range propagation. The intensity distribution profiles of the first two modes in
the cross section of the horizontal cylinder/dielectric/metal interface are shown in
Fig. 1.8. Both modes exhibit strong confinement in the gap, and for the second mode
fields of weaker intensity are also present inside the horizontal cylinder.
The effective indices of the two modes as a function of the cylinder diameter
for three different gap sizes are shown in Fig. 1.9. We see that the real part of the
effective index of the fundamental mode levels off as the diameter increases. For the
10 nm gap waveguides, the limiting value is around 2.06, the refractive index of ZnO
for 488 nm light, as expected. However, for the 5 nm and 2 nm gap waveguides, the
mode indices surpass the refractive index of ZnO by a considerable amount, which
can be understood as the result of matching the dielectric constant of Ag with that
of ZnO on the cylinder/dielectric/metal interface [39].
10 H. Xu
Fig. 1.8 Distribution of jEj across the cross-sectional area of the waveguide for (a) fundamental
mode and (b) second mode. The maximum value of jEj is chosen to be 1 V/m for all modes. The
gap size is 10 nm and the diameter of the horizontal waveguide is 400 nm (Modified from Fig. 1.2
of Ref. [39])
The confinement of these modes can be characterized by examining the full

width half maximum (FWHM) of the field distribution along the lateral direction
in the middle of the gap. The width increases with the diameter of the horizontal
cylinder and the gap size. For a gap size of 2 nm, the FWHM for the fundamental
mode can be as low as 30 nm for a cylinder diameter of 120 nm. While for the
second mode, the FWHM is about 50 nm for a cylinder diameter of 400 nm. To
better characterize the confinement, we have adopted the same method used by
Oulton and coworkers [17] to calculate the effective mode area. The total energy
density is first integrated over the cross-sectional area of the waveguide then divided
by the maximum energy density to obtain the effective mode area, which is then
normalized by the diffraction limit œ2 /4. Since in certain applications molecules do
not exist inside the dielectric waveguide thus cannot be excited by the fields inside
2.8
Gap 2 nm, 1st mode Gap 2 nm, 2nd mode
2.6 Gap 5 nm, 1st mode Gap 5 nm, 2nd mode
Gap 10 nm, 1st mode Gap 10 nm, 2nd mode
2.4
2.2
Mode Index
1.8
1.6
1.4
1.2
120 160 200 240 280 320 360 400
Diameter (nm)
Fig. 1.9 Effective mode indices of fundamental mode (solid lines) and second mode (dashed lines)
as a function of the horizontal waveguide diameter for three different gap sizes: 2, 5, and 10 nm
(Modified from Fig. 1.3 of Ref. [39])
the dielectric waveguide, we can exclude the contribution of the fields inside the
dielectric waveguide in calculating the effective mode area. The results are shown
in Fig. 1.10. For a gap size of 2 nm and a 400-nm-diameter waveguide, we found
this effective mode area for the second mode is only 1/50 of the diffraction limit
œ2 /4, which is in the deep subwavelength range.
We note that these modes also have long propagation length, defined as the
distance before the amplitude of the mode drops by a factor of 1/e, even for
smaller gap sizes and diameters (i.e. higher confinement). In general, the second
mode exhibits increased propagation length for certain reduced gap size and
diameter combinations. For example, for a gap size of 2 nm and a 400-nm-diameter
waveguide, the propagation length of the second mode is more than 20 m while
that of the fundamental mode is only a few wavelengths [39].
For this multi-mode hybrid waveguide, it is found that it is possible to excite
these modes by using a vertical ZnO nanowire waveguide placed inside the silver
film directly beneath the horizontal ZnO cylinder [39]. In general, the fundamental
mode always has the largest excitation amplitude, while the higher-order modes
have smaller amplitudes. Due to the short propagation distance but high amplitude
of the fundamental mode, it dominates the fields near the excitation spot. However
the second mode starts to dominate the fields further away from the spot as a result
of its slow attenuation, which makes it very attractive in confined light guiding
applications. With the second mode, decent field strength with a confinement of
50 nm can be obtained at a distance of 40œ from the excitation aperture.
12 H. Xu
0.1
1st mode
2nd mode
0.08
Normalized Mode Area
0.06
0.04
0.02
0
220 240 280 320 360 400
Diameter (nm)
Fig. 1.10 Effective mode area normalized by the diffraction limit œ2 /4 as a function of the
horizontal waveguide diameter for fundamental mode (squares) and second mode (circles). The
gap size is 2 nm. To calculate the effective mode area, the total energy density is first integrated
over the cross-sectional area of the waveguide then divided by the maximum energy density. The
calculation is done with the contribution of the fields inside the dielectric waveguide excluded
(Modified from Fig. 1.5 of Ref. [39])
1.5 Conclusions
In this chapter, two plasmonic nanowire waveguide based approaches have been
discussed to realize efficient guiding of light on the nanoscale. In the simple
plasmonic nanowire waveguide approach, efficient guiding is achieved by utilizing
the surface plasmon mode existing at the cylindrical boundary between the dielectric
wire and the metal cladding, resulting in deep subwavelength confinement and
relatively short propagation distance of about one wavelength. In the second
approach of a plasmonic hybrid waveguide, hybrid modes – combination of the
SPPs mode at the planar metal-dielectric interface and cylindrical modes of the
dielectric cylinder above the metal-dielectric interface – are used to achieve deep
subwavelength confinement and propagation distance of tens of wavelength. Due to
their nanoscopic confinement, these waveguides may find applications in a variety of
fields such as near-field scanning optical microscopy, single-molecule spectroscopy,
nano-lithography, and integrated plasmonic circuits.
References
1. Barnes WL, Dereux A, Ebbesen TW (2003) Surface plasmon subwavelength optics. Nature
424:824–830
2. Maier SA (2007) Plasmonics: fundamentals and applications. Springer, New York
3. Grandidier J, des Francs GC, Massenot S, Bouhelier A, Markey L, Weeber J-C, Finot C, Dereux
A (2009) Gain-assisted propagation in a plasmonic waveguide at telecom wavelength. Nano
Lett 9:2935–2939
4. Krasavin AV, Zayats AV (2010) Silicon-based plasmonic waveguides. Opt Express 18:11791–
11799
5. Schuller JA, Barnard ES, Cai W, Jun YC, White JS, Brongersma ML (2010) Plasmonics for
extreme light concentration and manipulation. Nat Mater 9:193–204
6. Stockman MI (2011) Nanoplasmonics: past, present, and glimpse into future. Opt Express
19:22029–22106
7. Lindquist NC, Nagpal P, McPeak KM, Norris DJ, Oh SH (2012) Engineering metallic
nanostructures for plasmonics and nanophotonics. Rep Prog Phys 75:036501
8. Sorger VJ, Oulton RF, Ma RM, Zhang X (2012) Toward integrated plasmonic circuits. MRS
Bull 37:728–738
9. Zhang X, Liu Z (2008) Superlenses to overcome the diffraction limit. Nat Mater 7:435–441
10. Anker JN, Hall WP, Lyandres O, Shah NC, Zhao J, Van Duyne RP (2008) Biosensing with
plasmonic nanosensors. Nat Mater 7:442–453
11. Yan R, Gargas D, Yang P (2009) Nanowire photonics. Nat Photonics 3:569–576
12. Luo X, Ishihara T (2004) Subwavelength photolithography based on surface-plasmon polariton
resonance. Opt Express 12:3055–3065
13. Catchpole KR, Polman A (2008) Plasmonic solar cells. Opt Express 16:21793–21800
14. Atwater HA, Polman A (2010) Plasmonics for improved photovoltaic devices. Nat Mater
9:205–213
15. Maier SA, Kik PG, Atwater HA, Meltzer S, Harel E, Koel BE, Requicha AA (2003) Local
detection of electromagnetic energy transport below the diffraction limit in metal nanoparticle
plasmon waveguides. Nat Mater 2:229–232
16. Liu Z, Steele JM, Srituravanich W, Pikus Y, Sun C, Zhang X (2005) Focusing surface plasmons
with a plasmonic lens. Nano Lett 5:1726–1729
17. Oulton RF, Sorger VJ, Genov DA, Pile DFP, Zhang X (2008) A hybrid plasmonic waveguide
for subwavelength confinement and long-range propagation. Nat Photonics 2:496–500
18. Takahara J, Yamagishi S, Taki H, Morimoto A, Kobayashi T (1997) Guiding of a one-
dimensional optical beam with nanometer diameter. Opt Lett 22:475–477
19. Naber A, Molenda D, Fischer UC, Maas H-J, Höppener C, Lu N, Fuchs H (2002) Enhanced
light confinement in a near-field optical probe with a triangular aperture. Phys Rev Lett
89:210801
20. Shi X, Hesselink L, Thornton RL (2003) Ultrahigh light transmission through a C-shaped
nanoaperture. Opt Lett 28:1320–1322
21. Tanaka K, Tanaka M (2004) Optimized computer-aided design of I-shaped subwavelength
aperture for high intensity and small spot size. Opt Commun 233:231–244
22. Baida FI, Poujet Y, Guidal B, Van Labeke D (2005) New design for enhanced transmission and
polarization control through near-field optical microscopy probes. Opt Commun 256:190–195
23. Jin EX, Xu X (2005) Obtaining super resolution light spot using surface plasmon assisted sharp
ridge nanoaperture. Appl Phys Lett 86:111106
24. Janunts NA, Baghdasaryan KS, Nerkararyan KV, Hecht B (2005) Excitation and superfocusing
of surface plasmon polaritons on a silver-coated optical fiber tip. Opt Commun 253:118–124
25. Baida FI, Belkhir A, Labeke DV, Lamrous O (2006) Subwavelength metallic coaxial waveg-
uides in the optical range: Role of the plasmonic modes. Phys Rev B 74:205419
26. Ding W, Andrews SR, Maier SA (2007) Internal excitation and superfocusing of surface
plasmon polaritons on a silver-coated optical fiber tip. Phys Rev A 75:063822
14 H. Xu
27. Gramotnev DK, Pile DFP, Vogel MW, Zhang X (2007) Local electric field enhancement during
nanofocusing of plasmons by a tapered gap. Phys Rev B 75:035431
28. Choi H, Pile DFP, Nam S, Bartal G, Zhang X (2009) Compressing surface plasmons for nano-
scale optical focusing. Opt Express 17:7519–7524
29. de Waele R, Burgos SP, Polman A, Atwater HA (2009) Plasmon dispersion in coaxial
waveguides from single-cavity optical transmission measurements. Nano Lett 9:2832–2837
30. Banzer P, Kindler J, Quabis S, Peschel U, Leuchs G (2010) Extraordinary transmission through
a single coaxial aperture in a thin metal film. Opt Express 18:10896–10904
31. Liu H-L, Wang N, Liu Y-H, Wu X-J (2010) Influence of vertex angles on the extraordinary
optical transmission characteristics of triangular and bow-tie-shaped apertures. Solid State
Commun 150:1822–1826
32. García de Abajo FJ (2002) Light transmission through a single cylindrical hole in a metallic
film. Opt Express 10:1475–1484
33. Olkkonen J, Kataja K, Howe DG (2005) Light transmission through a high index dielectric-
filled sub-wavelength hole in a metal film. Opt Express 13:6980–6989
34. García-Vidal FJ, Moreno E, Porto JA, Martín-Moreno L (2005) Transmission of light through
a single rectangular hole. Phys Rev Lett 95:103901
35. Shin H, Catrysse PB, Fan S (2005) Effect of the plasmonic dispersion relation on the
transmission properties of subwavelength cylindrical holes. Phys Rev B 72:085436
36. Xu H, Zhu P, Craighead HG, Webb WW (2009) Resonantly enhanced transmission of light
through subwavelength apertures with dielectric filling. Opt Commun 282:1467–1471
37. Velasco MG, Cassidy P, Xu H (2011) Extraordinary transmission of evanescent modes through
a dielectric-filled nanowaveguide. Opt Commun 284:4805–4809
38. Garcia VL, Velasco MG, Mamer SB, Singh KR, Hossain NK, He G, Sadoqi M, Xu H (2012)
Resonant transmission of light through ZnO nanowaveguides in a silver film. Appl Phys Lett
101:081113
39. Colanduoni J, Nikolov D, Xu H (2016) Multi-mode Hybrid Plasmonic Waveguides with
Enhanced Confinement and Propagation. Plasmonics 11:763–769
40. Kong JA (1986) Eleectromagnetic Wave Theory. Wiley, New York
41. Pfeiffer CA, Economou EN, Ngai KL (1974) Surface polaritons in a circularly cylindrical
interface: Surface plasmons. Phys Rev B 10:3038–3051
42. Novotny L, Hafner C (1994) Light propagation in a cylindrical waveguide with a complex,
metallic, dielectric function. Phys Rev E 50:4094–4106
43. Palik ED (ed) (1998) Handbook of optical constants of solids. Academic, New York
44. Raether H (1988) Surface Plasmons on Smooth and Rough Surface and on Gratings. Springer,
Berlin
45. Aers GC, Boardman AD, Paranjape BV (1980) Non-radiative surface plasma-polariton modes
of inhomogeneous metal circular cylinders. J Phys F Metal Phys 10:53–65
46. Bass M (1994) Handbook of Optics, vol 2, 2nd edn. McGraw-Hill, New York
47. Greene LE, Law M, Tan DH, Montano M, Goldberger J, Somorjai G, Yang P (2005) General
Route to Vertical ZnO Nanowire Arrays Using Textured ZnO Seeds. Nano Lett 5:1231–1236
48. Lopez OE, Tucker AL, Singh KR, Mamer SB, Sadoqi M, Xu H (2014) Synthesis of zinc oxide
nanowires on seeded and unseeded gold substrates: Role of seed nucleation and precursor
concentration. Superlattice Microst 75:358–370
49. Dunn RC (1999) Near-field scanning optical microscopy. Chem Rev 99:2891–2928
50. Pollack L, Webb WW (2010) Complex molecular dynamics in the spotlight. Nat Biotechnol
28:564–565
Chapter 2
Plasmon-Enhanced Fluorescence of Rare Earth
Nanocrystals
Ya-Fang Zhang*, Jia-Hong Wang*, Ming Chen, Si-Jing Ding, Zhi-Jun Luo,
Yong-Jie Lin, Ya-Lan Wang, Xue-Feng Yu, Zhong-Hua Hao, Li Zhou,
and Qu-Quan Wang
Abstract Rare earth (RE) nanocrystals (NCs) exhibit featured fluorescence prop-
erties and are widely used in many applications of biomedical labeling, display,
sensing, light-emitting devices, etc. Plasmon-enhanced and -manipulated fluores-
cence of RE NCs usually shows different behaviors compared with organic dye
molecules and semiconductor NCs. In this Chapter, we firstly present the enhanced
upconversion fluorescence of RE fluoride NCs by the dopant-controlled synthesis.
Then the preparation of plasmon/RE core/shell NCs by wet-chemical growth
method is introduced. Finally, we demonstrate the plasmon-enhanced upconversion
fluorescence of a single RE NC and by plasmonic array nanostructures.
*Equally contributed
Y.-F. Zhang • J.-H. Wang • M. Chen • S.-J. Ding • Z.-J. Luo • Z.-H. Hao () • L. Zhou ()
Key Laboratory of Artificial Micro- and Nano-structures of the Ministry of Education,
School of Physics and Technology, Wuhan University, 430072, Wuhan, P. R. China
e-mail: zhhao@whu.edu.cn; zhouli@whu.edu.cn
Y.-J. Lin
The Institute for Advanced Studies, Wuhan University, 430072, Wuhan, P. R. China
Y.-L. Wang
Microwave Photonics Laboratory, Wuhan Electronic Information Institute, 430019,
Wuhan, P. R. China
X.-F. Yu
Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences,
518055, Shenzhen, P. R. China
Q.-Q. Wang ()
Key Laboratory of Artificial Micro- and Nano-structures of the Ministry of Education,
School of Physics and Technology, Wuhan University, 430072, Wuhan, P. R. China
The Institute for Advanced Studies, Wuhan University, 430072, Wuhan, P. R. China
e-mail: qqwang@whu.edu.cn

DOI 10.1007/978-3-319-48081-7_2
16 Y.-F. Zhang et al.
Keywords Rare earth nanocrystals • Metallic nanocrystals • Core-shell

nanostructures • Surface plasmon resonance • Upconversion fluorescence
2.1 Introduction
The collective oscillations of conduction electrons in metallic nanostructures,

termed as surface plasmon resonance (SPR), could be applied to amplify the signals
of optical transition, such as fluorescence. The plasmon-enhanced fluorescence
of organic dye molecules, semiconductor nanocrystals (NCs) and rare earth (RE)
ions is widely studied is recent decades. Meanwhile, the underlying enhancement
mechanisms are well investigated. The main effects discussed for fluorescence
enhancement are the Förster resonance energy transfer (FRET) and the Purcell
effect [1–8]. The plasmon resonances of metallic nanoparticles provide the large
extinction cross-section and local field, which are expected to contribute to the
fluorescence enhancement. However, the competition between enhancement and
quenching of fluorescence is always occurred in the discussion. Many factors should
be considered for optimizing the enhancement, including: (1) the spectral overlap
between the plasmon and the fluorophore’s absorption or emission; (2) the distance
between the plasmonic nanoparticles and the fluorophores; (3) the intensity of
plasmon-induced local field as well as the contribution of absorption and scattering
of plasmons [1, 9–14].
RE materials possess the characteristic properties of fluorescence and attract
broad attention [15–18]. RE elements are composed of the 15 lanthanides (from
lanthanum to lutetium), with scandium and yttrium. Compared with the organic
dye molecules and semiconductor NCs, the fluorescence induced by the intra-4f
or 4f5d transitions of RE ions exhibits many advantages of extremely narrow
emission peak, large Stokes shift (up to 500 nm), long emission lifetimes (up to
milliseconds), high photochemical stability, reduced photobleaching, etc. However,
due to the forbidden 4f transitions, the direct excitation of many RE ions has
a relatively low efficiency [19, 20]. The doping RE ions into host lattices is
popularly used in preparing high-efficient luminescent RE NCs. The high absorption
coefficient of host materials as well as the energy transfer between host materials
and doped RE ions and/or the energy transfer between co-doped RE ions could
produce desirable luminescence RE materials. The RE NCs possess the emis-
sion wavelength coving ultraviolet-visible-near infrared (UV-VIS-NIR) region, and
exhibit the photoluminescence quantum yield (QY) approaching nearly 100 %.
Although showing the high quantum efficiency, RE NCs suffer from low
absorption cross-section [21–23]. Therefore, the plasmon-enhanced fluorescence
has been proposed to improve the emission properties [9, 20, 24–30]. The recent
progress of plasmon-enhanced fluorescence of RE NCs is focused on the plasmon-
enhanced upconversion fluorescence, [24, 26, 29–36] the interaction with the
periodical and waveguiding plasmonic nanostructures, [37–41] the observation
of single-particle fluorescence enhancement, [29, 42–45] the tip-enhanced effect,
[46, 47] the distance-dependent enhancement, [26, 34] and the plasmon-enhanced
2 Plasmon-Enhanced Fluorescence of Rare Earth Nanocrystals 17
energy transfer [30, 48]. In this Chapter, we illustrate current progress in research
of improving fluorescence properties of RE NCs, including dopant-controlled
synthesis of RE fluoride NCs for enhancing upconversion fluorescence, preparation
of plasmon/RE core/shell NCs by wet-chemical growth method for optimizing
plasmon-RE interaction, plasmon-enhanced fluorescence in a single upconversion
RE NC, and plasmon-enhanced fluorescence in upconversion RE NCs by coupling
with plasmonic periodic nanostructures.
2.2 Dopant-Controlled Synthesis of Upconversion Rare

Earth Nanocrystals for Improved Fluorescence
Upconversion emission materials of RE NCs are proposed as a promising probe

for biological labeling and imaging because of their the excellent light penetration
depth and weak autofluorescence background under NIR excitation [19, 20, 49–53].
For example, the RE-doped fluoride compounds, especially NaYF4 , have attracted
intense attention because they have low phonon energy, and are regarded as excellent
host materials for both downconversion and upconversion processes [49, 50, 54–
59]. Hexagonal NaYF4 NCs are the most efficient upconversion fluorophores and
provide over an order of magnitude stronger fluorescence than the NaYF4 cubic
form [56, 60–64]. The crystal phase should be fine controlled in the preparing of
RE-doped NaYF4 NCs with desirable luminescence properties. The water-soluble
NaYF4 NCs are usually cubic phase [52, 65] or a mixture of cubic and hexagonal
phases [66].
Yu et al. found in the synthesis of water-soluble cubic-phase NaYF4 NCs, when
the Y3C ions (ionic radius r D 0.89 Å) are replaced by Gd3C (r D 0.94 Å) and Nd3C
(r D 1.00 Å) ions, the synthesized NaGdF4 and NaNdF4 NCs are hexagonal phase
[67]. The RE doping effect on the cubic-to-hexagonal phase transition of NaYF4
NCs is further investigated. As the transmission electron microscope (TEM) images
and X-ray diffraction (XRD) patterns shown in Fig. 2.1a–c, the undoped NaYF4
NCs are cubic nanospheres with diameters of 50 nm. When La3C (or Gd3C ) was
used as dopant with a doping concentration of 10 %, the synthesized NaYF4 :10 %
La NCs became hexagonal long NRs with diameters of 150 nm and lengths of
1100 nm. Their regular hexagonal cross-section can be observed in the inset SEM
image.
In general, the hexagonal phase NaREF4 is more thermodynamically stable than
the cubic form, and the cubic-to-hexagonal phase transition is a disorder-to-order
character which requires sufficient free energy to overcome the activation barrier.
High temperature and prolonged reaction time have often been used to provide
enough energy to overcome the energy barrier between the cubic and hexagonal
NaREF4 [68–70]. Yan’s group revealed that the energy barriers for hexagonal
NaREF4 is lower for the RE ions with larger ionic radius [70]. It is reasonable that
the doping of RE elements with larger ionic radius (such as Gd3C , Nd3C and La3C )
could drive the cubic-to-hexagonal phase transition. Further experimental results
Fig. 2.1 (a, b) TEM images of undoped NaYF4 and NaYF4 :10 % La NCs. The inset in (b) shows
the cross sections of NaYF4 :10 % La NRs. (c) XRD spectra of undoped NaYF4 , NaYF4 :10 %
Gd, and NaYF4 :10 % La nanocrystals. The cubic NaYF4 (JCPDS: 77-2042) and hexagonal
NaYF4 (JCPDS: 28-1192) are shown. (d) Upconversion emission spectra of NaYF4 :(Yb, Er)/La,
NaYF4 :(Yb, Er)/Ce NRs, and NaYF4 :(Yb, Er) nanospheres under 975 nm CW laser excitation. (e)
In vivo multicolor upconversion fluorescence imaging of the leg in a mouse after being injected
intramuscularly with 100 L of 1.0 mg/mL NaYF4 :(Yb, Er)/La NRs, under defocused 975 nm CW
laser excitation. The right column shows the emission spectra at corresponding regions (Adapted
with permission from Ref. [67])
show that the RE elements could be divided into three groups according to their
increasing ionic radius: Group I (r D 0.850.92 Å): Lu, Yb, Tm, Er, Ho, Dy, and Tb;
Group II (r D 0.941.01 Å): Gd, Eu, Sm, Nd, and Pr; Group III (r D 1.031.06 Å):
Ce and La. The RE dopants in Group I show almost no influence on the phase and
shape of the NaYF4 host. In contrast, the RE dopants in Groups II and III could
induce the formation of hexagonal NaYF4 :10 % RE NRs. The dopants of Ce and La
in group III could produce long NRs with the length up to microns and the aspect
ratio over 7.
For showing the improved fluorescence and the biological imaging applica-
tion, the NaYF4 :(Yb, Er)/La (Y:Yb:Er:La D 70:20:5:5) and NaYF4 :(Yb, Er)/Ce
(Y:Yb:Er:Ce D 70:20:5:5) NCs have been prepared, in which the doped Yb, Er,
and Ce/La ions act as sensitizers, emitters, and phase controllers, respectively. As
shown in Fig. 2.1d, the hexagonal NaYF4 :(Yb, Er)/La and NaYF4 :(Yb, Er)/Ce
NRs dispersed in water exhibit much stronger upconversion fluorescence than
that of the cubic NaYF4 :(Yb, Er) nanospheres with the same concentration. The
hexagonal NaYF4 :(Yb, Er)/La NRs have been employed as the probes for deep
tissue multicolor imaging due to their suitable particle size and greatly improved
upconversion fluorescence (Fig. 2.1e).
Fig. 2.2 (a–d) TEM images of NaYF4 :Yb/Er (18/2 mol %) products obtained after heating for 2 h
in the absence of Gd3C dopant ions (a, b) and in the presence of 30 mol % Gd3C dopant ions (c, d),
respectively. The upper inset in (b) shows the Fourier-transform diffractogram of a small nanocube
in (b) with cubic phase. The lower inset in (b) shows the selected area electron diffraction pattern
of a hexagonal-phase NR in (a). The inset in (d) shows the Fourier-transform diffractogram of a NR
shown in (c), indicating the hexagonal phase. (e) Room temperature upconversion emission spectra
of cyclohexane solutions comprising NaYF4 :Yb/Er/Gd (18/2/5 mol %), NaYF4 :Yb/Tm/Er/Gd
(20/0.2/0.1/5 mol %), NaYF4 :Yb/Tm/Er/Gd (20/0.2/0.05/5 mol %), and NaYF4 :Yb/Tm/Gd
(20/0.2/5 mol %) nanoparticles, respectively. (f) Luminescent image of a 3D structure generated
in the PDMS sample composed of NaYF4 :Yb/Er/Gd (18/2/5 mol %) nanoparticles via computer-
controlled NIR laser scanning (Adapted with permission from Ref. [71])
Wang et. al. demonstrated that the size and crystal phase of NaYF4 nanocrystals
could be simultaneously controlled through lanthanide doping in oil-phase reaction
condition [71]. As shown in Fig. 2.2a–d, the NaYF4 :Yb/Er (18/2 mol %) products
obtained after heating for only 2 h is a mixture of two distinct particle morphologies:
the small cubic-phase nanocubes and large hexagonal-phase NRs. However, in the
presence of Gd3C dopant ions (30 mol %), the cubic-to-hexagonal phase conversion
is clearly complete under same reaction conditions, and the products are hexagonal-
phase NRs. The effect of reaction time and doping concentration were carefully
investigated. The first-principles calculations were also performed to confirm that
the influence of lanthanide doping on crystal phase and size arises from a strong
dependence on the size and dipole polarizability of the substitutional dopant ion.
The small-sized NaYF4 upconversion nanoparticles that emit tunable visible colors
have been readily synthesized by rational doping of Yb3C , Er3C , Gd3C and Tm3C
ions at different concentrations (Fig. 2.2e). These nanoparticles can readily be
incorporated in polydimethylsiloxane (PDMS) monoliths to construct volumetric
three-dimensional (3D) displays (Fig. 2.2f).
Fig. 2.3 (a–d) Room temperature upconversion emission spectra of NaYF4 :Yb/Er (18/2 mol %)
(a), NaYF4 :Yb/Tm (20/0.2 mol %) (b), NaYF4 :Yb/Er (25–60/2 mol %) (c), and NaYF4 :Yb/Tm/Er
(20/0.2/0.2–1.5 mol %) (d) particles in ethanol solutions (10 mM). The spectra in (c) and (d) are
normalized to Er3C 660 nm and Tm3C 480 nm emissions, respectively. (e–n) Luminescent photos
showing corresponding colloidal solutions of NaYF4 :Yb/Tm (20/0.2 mol %) (e), NaYF4 :Yb/Tm/Er
(20/0.2/0.2–1.5 mol %) (f–j), and NaYF4 :Yb/Er (18–60/2 mol %) (k–n) [52]. (o) X-ray powder
diffraction patterns of NaYF4 : Yb/Er (18/2 mol %) obtained after heating for 8 h at 200 ı C in
the presence of 0, 3, 5 and 30 mol % Mn2C dopant ions, respectively. (p) Room temperature
upconversion emission spectra of NaYF4 : Yb/Er NCs with 0 and 30 mol % Mn2C dopant ions,
respectively [72] (Adapted with permission from Refs. [52, 72])
Paramagnetic Mn2C ions have also been found to tune the crystal phase as well
as manipulate the cooperative energy transfer and adjust the fluorescence spectrum
of (Yb3C , Er3C ) co-doped NaYF4 nanocrystals [72]. Generally, a broad range of
color output by single wavelength excitation at 980 nm could be achieved by
doping varied amounts of Yb3C , Tm3C , and Er3C ions into a single host source of
NaYF4 (Fig. 2.3a–n) [52]. It is generally believed that the NIR spectral range (700–
1100 nm) and the red region (600–700 nm) are referred to as the “optical window”
of the biological tissues, where the light scatting, absorbance and autofluorescence
of tissue are minimum in view of the lack of efficient endogenous absorbers. Tian
et al. observed the hexagonal to cubic phase transformation of NaYF4 nanocrystals
through doping Mn2C ions (Fig. 2.3o). The Mn-doped NaYF4 :Yb/Er nanocrystals
show a suppressed green emission and an enhanced red emission (Fig. 2.3p), which
exhibits the potential advantages for in vivo bio-imaging [72].
2.3 Synthesis of Rare Earth Nanocrystal Shells on Plasmonic

Nanostructures
Fluorescence enhancement and manipulation of RE NCs by plasmonic interaction

have been widely reported in metal/RE hetero-nanostructures. Many reported prod-
ucts of metal/RE hybrids show a random adhesion of metallic nanoparticles with
RE nanocrystals and have inhomogeneous shape and poor dispersion. Core/shell
nanostructures have attracted large amount of attentions for the excellent physical
properties and the promising applications in various fields [73–83]. The pre-grown
noble metal core could provide designed shapes, sizes, and plasmonic properties
[78, 84–89]. The core/shell configuration would maximize the interfacial area
and facilitate the interaction between core and shell materials [90, 91]. However,
growing RE NCs shells on metallic cores and forming well-controlled morphology
are still a challenge, because there have large lattice mismatch and interfacial energy
between these two materials.
2.3.1 Synthesis of Metal/Oxide Core/Shell Nanocrystals
There have been a number of reports describing the fabricaion of a hybrid core-shell
nanostructure of RE oxide nanoshells and noble metal nanocores [36, 92–96]. Silica
shells are often used to assist the growth of RE oxide onto metallic cores. As shown
in Fig. 2.4a, b, Ag/SiO2 /Y2 O3 :Er [36] and Ag/SiO2 /Lu2 O3 :Gd/Yb/Er [97] core/shell
nanostructures were synthesized. The hybrids exbhited enhanced upconversion
luminescence and potential capibilty for biomedical application. Furthermore,
Ag/Y2 O3 :Er core/shell hollow nanospheres were achieved when the SiO2 shells
were etched [36]. Zhang et al. synthesized Pt/CeO2 nanocomosite with core-shell
or yolk-shell structure in an aqueous phase through a transformation from cerium
hydroxide to cerium dioxide (Fig. 2.4c) [98]. Meanwhile, Wang’s group developed a
general approach for the preparation of unprecedented (plasmonic metal nanocrystal
core)/(CeO2 shell) nanostructures through heterogeneous nucleation and growth, in
which ethylenediaminetetraacetic acid (EDTA) molecules were chosed to control
the nucleation and help the growth of CeO2 shells onto Au nanospheres (Fig.
2.4d) and Au nanorods (AuNRs) (Fig. 2.4e) [91]. Recently, our group fabricated
a novel AuNR/Nd2 O3 yolk/shell nanostructure with uniform and well-controlled
morphology, as shown in Fig. 2.4f.
2.3.2 Synthesis of Au NR/Vanadate Core/Shell Nanocrystals
AuNR is a typical anisotropic metal NC with tunable longitudinal SPR in a wide

range from the VIS to NIR region [99, 100]. Because of the stable and high-purity
Fig. 2.4 (a) TEM images of the Ag/SiO2 /Y2 O3 :Er nanostructures [36]. (b) TEM images of
Ag/SiO2 /Lu2 O3 :Gd/Yb/Er nanoparticles [97]. (c) TEM images of Pt/CeO2 nanocomposites with
yolk/shell structure [98]. (d, e) TEM images of Au/Ce2 O3 (d) and AuNR/Ce2 O3 (e) core/shell
nanocomposite [91] (Adapted with permission from Refs. [36, 91, 97, 98]) (f) TEM images of
AuNR/Nd2 O3 yolk/shell nanoparticles (unpublished)
emission spectra, rare-earth vanadate has been widely-used as fluorescent [101–

103]. However, there is a big obstacle of the crystal lattice mismatch in the shell
growth. Meanwhile, the growth of rare-earth compound is always require violent
reaction, which would destroy the pre-grown AuNRs [104].
Wang et al. designed a three-step method to grow rare-earth vanadates onto
AuNRs (Fig. 2.5a) to combine the plasmonic and fluorescence properties [105].
The cetyltrimethylammonium bromide (CTAB) capped AuNR (CTAB-AuNR) was
firstly treated by sodium oleate (NaOL) for the surfactant exchange by oleate. Owing
to the electrostatic force and chemical chelating caused by the carboxylic acid group
in oleate, [106] the oleate modified AuNR (oleate-AuNR) would adsorb the rare-
earth cantions. In the next step, with the hydrolyzation of hexamethylene tetramine
(HMT), rare-earth oxide were nucleated, grown, ripened, and finally format com-
plete oxide shell (Fig. 2.5b). Then, the gold/rare-earth-oxide (Au/Gd2 O3 :Eu NRs)
were transferred into the aqueous solution of Na3 VO4 , and an 120 ı C hydrothermal
treatment was taken to achieve gold/rare-earth-vanadate (Au/GdVO4 :Eu NRs). The
morphology of product obtained though this in situ anion exchange is characterized
by TEM (Fig. 2.5c). In the large-scale TEM image, the vanadate shell seems like
a pile of closely arranged rod-like crystals with the same direction. In the HRTEM
(Fig. 2.5d), the lattice fringes spacing 0.36 nm is corresponding to the (200) plane of
GdVO4 . Due to the vanadate was transformation from the oxide, it is supposed that
Fig. 2.5 (a) Schematic illustration of the Au/GdVO4 :Eu core/shell NRs formation. (b) TEM
image of Au/Gd2 O3 :Eu NRs. (c) TEM image of Au/GdVO4 :Eu NRs. (d) HRTEM images of typical
Au/GdVO4 :Eu NRs. (e) HRTEM images corresponding to the regions marked by the white square
in (d). (f) Absorption spectra of Au/GdVO4 :Eu NRs compared with original CTAB-AuNRs. (g)
Fluorescence emission spectra of Au/GdVO4 :Eu NRs, Au/GdVO4 :Sm NRs and Au/GdVO4 :Dy
NRs (Adapted with permission from Ref. [105])
the growth orientation of GdVO4 :Eu is critically depend on the original Gd2 O3 :Eu
shell. In the initial formation of oxide shell, the tiny crystal seeds are generated on
the surface of AuNR, which maybe induce the random orientations of the crystal
domains in Gd2 O3 :Eu. As a result, the vanadate (GdVO4 :Eu) array orientation
against the AuNRs is completely random (Fig. 2.5e).
The Au/GdVO4 :Eu NRs core/shell nanostructure has well integrated the plas-
monic property of AuNR and the fluorescence property of GdVO4 :Eu. The absorp-
tion spectra of the hetero-nanostructures is shown in Fig. 2.5f, the longitudinal SPR
band of AuNR is well maintained and red-shifted 24 nm after the vanadate coating,
which suggest that this strategy has successfully avoid the damage risk of the
AuNRs and well maintained its plasmonic property. Under UV light irradiation, the
aqueous solution of the Au/GdVO4 :Eu NRs emit bright red fluorescence (Fig. 2.5g),
the quantum yield was measured to be as high as 17 %. Because the main emission
wavelength of the Eu3C emission located at 617 nm is exactly located in the valley
region between the transverse and longitudinal SPR bands of AuNRs, the self-
absorption or scattering can also be minimized. Furthermore, bright Sm3C and Dy3C
fluorescence can also be obtained by changing the dopant ions (Fig. 2.5g).
2.3.3 Synthesis of Au Nanoparticle/Vanadate Core/Shell

Nanocrystals
Because of large lattice mismatch and interfacial energy, it is difficult to form

core/shell metal/rare-earth hetero-nanostructures with well-defined morphology
[36, 107]. Many methods have been introduced to break down this barrier [81,
108, 109]. Recently, utilizing specific chelating agent to form molecule-metal cation
complex to assist nucleation and crystal growth are reported [91, 110]. Chen et
al. reported a convenient and facile method to prepare Au/GdVO4 :Eu core/shell
nanoparticles, in which GdVO4 :Eu shells were directly grown onto the pre-prepared
gold nanoparticles and the final product is flower-shaped as well as uniform [111]. In
the schematic illustration (Fig. 2.6a), the organic additive of citrate ions (Cit3- ) have
been introduced into the reaction and the Cit3- plays a critical role in the primary
growth procedure of GdVO4 :Eu shell on the Au nanoparticles. A classical method
with some modifications has been employed to produce the Au nanoparticles
stabilized by the Cit3- . Meanwhile, the Cit3- is a strong chelating agent with four
binding sites for metal ions, so it can be bound with RE ions [101, 112]. It reacts with
RE3C to form stable RE3C -Cit3- complex through strong coordination interaction on
the Au surface. Then under hydrothermal conditions, the stable RE3C -Cit3- complex
interacts with VO4 3- and an anion-exchange reaction between VO4 3- and Cit3- would
take place in the high temperature and pressure. Consequently, the vanadate nuclei
are generated and bind on the Au surface. Therefore, in the whole synthesis process,
the Cit3- has assisted to overcome the huge obstacle induced by the large crystalline
mismatch and interfacial energy.
The morphology of Au/GdVO4 :Eu nanoparticles has been observed. The nanos-
tructures are flower-like and well-arranged. From the SEM image (Fig. 2.6b) and
TEM image (Fig. 2.6c), the Au/GdVO4 :Eu nanoparticles are uniform as well as
monodispersed and the vanadate shells are well-defined. Furthermore, the thickness
of rare-earth shells can be adjusted via controlling the addition of rare-earth
precursor during the reaction. The shell thickness of Au/GdVO4 :Eu nanoparticles
increases when the addition of 0.01 M Gd(NO3 )3 containing 5 % Eu(NO3 )3
increases. The SPR band of the products is red-shifted from 542 nm to 562 nm
simultaneously (Fig. 2.6d). The corresponding fluorescence intensity have also
become stronger as the shells become thicker (Fig. 2.6e). Besides that, while
adjusting the concentration of Au nanoparticles in the reaction, similar results can
be obtained.
Fig. 2.6 (a) Schematic illustration of the primary growth of Au/GdVO4 :Eu core/shell heterostruc-
tures. (b) SEM image and (c) TEM image of Au/GdVO4 :Eu core/shell flower-like nanoparticles.
(d) Absorption and (e) fluorescence spectra of Au/GdVO4 :Eu nanoparticles synthesized with
varied amount of 0.1 M Gd(NO3 )3 aqueous solution (Adapted with permission from Ref. [111])
It is observed that the crystallinity of rare-earth vanadate shells are improved and
the shells become thicker as well as more compact in the reaction of different time.
As a result, the corresponding samples represent plasmon shift and fluorescence
enhancement. On the other hand, the shells thickness can be adjusted and controlled
by the amount of rare-earth precursor as well as Au nanoparticles. It is obvious
that the SPR band and fluorescence intensity of final products could be adjusted
in consequence. Taken together, the plasmonic and fluorescence properties of
Au/GdVO4 :Eu nanoparticles are determined by the thickness of vanadate shells.
However, the plasmon interaction between the core and the shell could also affect
the optical properties of the samples, while the specific details are needed to be
further studied.
2.4 Plasmon-Enhanced Upconversion Fluorescence

of a Single Rare Earth Nanocrystal
Plasmonic resonances can enhance the upconversion emission of the NCs by two
different ways referred as absorption matching and emission matching. In order
to uncover the microscopic mechanisms of the plasmonic enhancement effect,
several research works have been carried out on the plasmon coupling upconversion
emission at single particle level [29, 42, 44, 45, 113].
The controlled coupling of single NaYF4 NC to single gold nanosphere was
demonstrated by S. Schietinger et al. for the first time [29]. They revealed that both
the excitation and the emission process in the upconversion NCs are influenced by
the coupling to SPR. They first brought a 60 nm gold nanosphere in the vicinity of
a 30 nm NaYF4:Yb3C /Er3C NC with AFM in contact mode, as shown in Fig. 2.7a.
The axis defined by the centers of the gold sphere and NC was aligned along the
polarization axis. The red emission increases by a factor of 2.7, while the green
Fig. 2.7 Enhanced upconversion PL of NaYF4 :Yb3C /Er3C NCs by plasmonic nanostructures at
single level. (a) AFM image of a gold nanoshpere being put near the upconversion NC. (b) The
upconversion PL spectra of the NaYF4 :Yb3C /Er3C NC with (blue curve) and without (violet curve)
the gold nanosphere [29]. (c) SEM image of the heterodimer consisting of a single NaYF4 :Yb3C ,
Er3C NC and a single AuNR. (d) Polarization dependent upconversion PL of the single rod-
phosphor heterodimer. (e) Theoretical simulation of local electromagnetic field distribution around
the heterodimer NC [42] (Adapted with permission from Refs. [29, 42])
emission is 4.8 times stronger than the bare upconversion NC, shown in Fig. 2.7b.
They also investigated the coupling between a single upconversion NC sandwiched
between two gold spheres. The emission intensity and dynamics changed with the
distance between the upconversion NC and the two gold nanosphperes.
Kagan’s group synthesized the NC heterodimers consisting of a single upcon-
version NaYF4 :Yb3C ,Er3C and a single AuNR, as shown in Fig. 2.7c [42]. By
tuning the longitudinal SPR of AuNRs to match the 977 nm excitation wavelength
of NaYF4 :Yb3C ,Er3C and by spatially localizing the fluorophores in the intense
near-fields surrounding the rod tips, they achieved several-fold upconversion lumi-
nescence enhancements. The enhancement effects exhibited a strong dependence on
the excitation light’s polarization relative to the AuNR axis, as shown in Fig. 2.7d.
The polarization sensitivity was more pronounced for the low-power excitation, and
greater enhancement was observed at lower excitation power densities due to the
nonlinear behavior of the upconversion process. They also performed theoretical
simulations using COMSOL Multiphysics software of the polarization-dependent
photoluminescence (PL) of the fluorophores with and without the AuNRs. Figure
2.7e clearly shows that the longitudinal polarization results in significantly enhanced
local electromagnetic field near the rod tip where the fluorophore is located, while
for transverse polarization case, no enhancement is observed.
Y. L. Wang et al. studied the coupling of upconversion fluorescent nanoparticle
NaYF4 :Yb3C /Er3C to silver nanoplatelets at the single particle level and made
a quantitative comparison between experimental results and numerical modeling
[114]. The plasmonic resonance wavelength of the proposed Ag nanoplatelets could
be widely tuned from green to NIR wavelengths with different lateral sizes [115,
116]. Figure 2.8b shows the upconversion luminescent spectra of a single NC with
(red line) and without (black line) Ag nanoplatelets, with the plasmonic resonance
matching the absorption line (980 nm). A 22-fold enhancement of PL due to the
local field enhancement by the Ag nanoplatelets was observed. While under the
emission matching case, with SPR wavelength of Ag nanoplatelets being tuned to
656 nm, only five-fold enhancement was achieved. Numerical simulation of the
field enhancement for the upconversion nanocrystal surrounded by Ag nanoplatelets
were carried out and the results were shown in Fig. 2.8d–f. The good quantitative
comparison between theory and experiment is made due to the actual geometric
relation control between the emitter and the plasmonic enhancers at the single
emitter level.
2.5 Fluorescence Tunability of Rare Earth Upconversion

Nanocrystals by Coupling with Plasmonic Nano Arrays
Plasmonic array is also an effective method to achieve upconversion enhancement.

Several works have been done to tune the upconverted fluorescence and the resonant
energy transfer process by using different types of periodical metallic nanostructures
such as disk-coupled dots-on-pillar antenna array, [117] Au nanohole array, [38]
Fig. 2.8 (a) SEM image of the composite nanosystem: one upconversion NC surrounded by five
Ag nanoplatelets. (b, c) PL upconversion spectra of single nanocrystals with and without Ag
nanoplatelets, when the LSPR is tuned to resonate at 980 nm (b) and 656 nm (c). (d) The 3D
structure used in the simulation, top and cross view. (e, f) Field distributions around the emitter
without (e) and with Ag nanoplatelets (f) (Adapted with permission from Ref. [114])
gold pyramid array, [30] Au nanopillar array, [118] Au nanocavity array, [41] AuNR
monolayer supercrystals,[37] Ag nanorod (AgNR) array, [119] Ag grating films,
[40] waveguide grating substrate [39] and so on.
W. Zhang et al. investigated the enhancment of the upconversion luminescence of
NaYF4 :Yb3C /Er3C NCs by using a 3D plasmonic nanoantenna architecture: disk-
coupled dots-on-pillar antenna array (D2PA). By optimizing the D2PA structure,
they obtained a 310-fold upconversion luminescence enhancement uniformly over a
large area (Fig. 2.9a–c) [117]. M. Saboktakin et al. used arrays of subwavelength
holes (nanoholes) in Au films as templates to localize and enhance the lumi-
nescence of hexagonal phase NaYF4 :Yb3C /Er3C and got about 35 folds resonant
enhancements of the luminescence intensity, with a precise design of its periodicity
being resonant with the excitation wavelength of the upconversion (Fig. 2.9d–f)
[38]. Q. Sun et al. utilized resonant surface plasmon polaritons waves to enhance
upconversion PL in doped-lanthanide NCs. They demonstrated that SPP waves
could increase the rate of resonant energy transfer from Yb3C to Er3C ions by
sixfold. They observed high enhancement in upconversion photoluminescence,
especially at the gold pyramid tip. While for green upconversion PL, strong
quenching were observed on the flat gold metal, see Fig. 2.9g–i [30].
Wang et al. studied the upconversion luminescence of Mn2C -doped (Yb3C ,
3C
Er )/NaYF4 NCs with a specific higher-Q plasmonic cavity composed of a AgNR
array [119]. They investigate strong cooperative energy transfer in Mn2C -doped and
bare (Yb3C , Er3C )/NaYF4 NCs by using a strongly-confined plasmonic cavity. The
Fig. 2.9 Tuning the upconversion luminescence of the NaYF4 :Yb3C /Er3C nanocrystals by
coupling with metallic nanostructure arrays. (a) SEM image of D2PA. (b) SEM top view images
of D2PA substrate (upper) and reference substrate with upconversion NCs on their top (lower). (c)
Enhanced upconversion luminescence on the optimized D2PA substrate [117]. (d) SEM images of
nanoholes in Au film. Inset: Intensity enhancement inside an aperture in the nanohole array at its
resonance normalized with respect to a similar array made in glass. (e) Upconversion luminescence
spectra of NaYF4 :Yb3C /Er3C NCs in a Au nanohole array (red curve) and in a glass nanohole
array (blue curve). (f) Emission enhancement factor as a function of wavelength [38]. (g) SEM
images of gold phyramid array. (h) Schematic of the energy transfer, upconversion, and quenching
processes on the top and bottom of the gold pyramid substrate. (i) Upconversion emission spectra
NaYF4:Yb3C /3 % Er3C NCs on the gold pyramid (red curve), flat gold (green curve), and glass
substrate (blue curve), respectively [30] (Adapted with permission from Refs. [30, 38, 117])
AgNRs were deposited on AAO template. Mn2C :(Yb3C , Er3C )/NaYF4 nanocrystals
were spin coated on the barrier layer side of the AAO template (Fig. 2.10a).
The SPR of the AgNRs was tuned to be resonant with the emission peak of the
upconversion NCs. The upconversion PL was greatly enhanced by the plasmon
resonance (Fig. 2.10b). The PL intensity of the three emission peaks around
520, 540, and 655 nm increases with the excitation power but at different rates
Fig. 2.10 Optical structure of the Mn2C : (Yb3C , Er3C )/NaYF4 nanocrystals coupling with
plasmonic nanocavity. (a) The nanostructure of the sample. (b) Tunable upconversion PL spectra
of Mn2C : (Yb3C , Er3C )/NaYF4 nanocrystals coupled to plasmonic nanocavity. (c) The PL spectra
dependence on the excitation power of the Mn2C : (Yb3C , Er3C )/NaYF4 on AgNRs cavity. (d) The
PL intensity dependence of the three peaks (520 nm, 540 nm, 655 nm) on the excitation power. (e)
The PL intensity ratio of I655 /I540 and (f) I540 /I520 as a function of the excitation power, the brown
and dark green lines are the results without the AgNRs (Adapted with permission from Ref. [119])
(Fig. 2.10c–d). The intensity ratio of the red emission to that of the green emission
changing with the excitation power are shown in Fig. 2.10e–f. These ratio changes
are attributed to field enhancement and energy transfer induced by the AgNRs
cavity. Multiple energy transfer channels are introduced due to doping of Mn2C ions,
and the local electric field induced by the plasmonic AgNR nanocavity enhances the
energy transfer process in Mn2C doped (Yb3C , Er3C )/NaYF4 nanocrystals.
2.6 Conclusions
RE nanocrystals possess characteristic fluorescence properties and are expected to

be a promising luminescent materials in many applications. The studying the way
to improve the fluorescence properties of RE nanocrystals is very important and
attracts wide attention. The ion doping could control the crystal phase and size of RE
nanocrystals, thus enhances the upconversion fluorescence intensity and alters the
emission spectra. Another feasible way to enhance and manipulate the emission is
applying surface plasmon resonance. Recently, many wet-chemical growth methods
are developed for preparation the plasmon/RE core/shell nanocrystals with uniform
morphology and controlled size. In this well-controlled core/shell nanocrystals,
the interaction between plasmonic core and RE shell could be finely adjusted for
pursuing excellent fluorescence properties. The observation of plasmon-enhanced
upconversion fluorescence in a single particle has been demonstrated. In this
case, the distance between plasmon and RE NC could be fine controlled and
accurately modeled for revealing enhancement factor. Plasmonic array is also an
effective method to achieve upconversion enhancement and tailoring the resonant
energy transfer effect. RE NCs with enhanced fluorescence properties exhibit great
potential in many applications.
Acknowledgments This work was supported in part by the National Program on Key Science
Research of China and the National Science Foundation of China.
References
1. Li M, Cushing SK, Wu N (2015) Plasmon-enhanced optical sensors: a review. Analyst

140:386–406
2. Anger P, Bharadwaj P, Novotny L (2006) Enhancement and quenching of single-molecule
fluorescence. Phys Rev Lett 96:113002
3. Bardhan R, Grady NK, Cole JR, Joshi A, Halas NJ (2009) Fluorescence enhancement by Au
nanostructures: nanoshells and nanorods. ACS Nano 3:744–752
4. Clapp AR, Medintz IL, Mauro JM, Fisher BR, Bawendi MG, Mattoussi H (2004) Fluo-
rescence resonance energy transfer between quantum dot donors and dye-labeled protein
acceptors. J Am Chem Soc 126:301–310
5. Ganesh N, Zhang W, Mathias PC, Chow E, Soares J, Malyarchuk V, Smith AD, Cunningham
BT (2007) Enhanced fluorescence emission from quantum dots on a photonic crystal surface.
Nat Nanotechnol 2:515–520
6. Tam F, Goodrich GP, Johnson BR, Halas NJ (2007) Plasmonic enhancement of molecular
fluorescence. Nano Lett 7:496–501
7. Tovmachenko OG, Graf C, Van Den Heuvel DJ, Van Blaaderen A, Gerritsen HC
(2006) Fluorescence enhancement by metal-core/silica-shell nanoparticles. Adv Mater 18:
91–95
8. Zhang J, Fu Y, Lakowicz JR (2007) Enhanced Förster Resonance Energy Transfer (FRET) on
a Single Metal particle. J Phys Chem C 111:50–56
9. Dong J, Zhang Z, Zheng H, Sun M (2015) Recent progress on plasmon-enhanced fluores-
cence. Nanophotonics 4:472–490
10. Bauch M, Toma K, Toma M, Zhang Q, Dostalek J (2014) Plasmon-enhanced fluorescence

biosensors: a review. Plasmonics 9:781–799
11. Chen Y, Munechika K, Ginger DS (2007) Dependence of fluorescence intensity on the
spectral overlap between fluorophores and plasmon resonant single silver nanoparticles. Nano
Lett 7:690–696
12. Mock JJ, Hill RT, Degiron A, Zauscher S, Chilkoti A, Smith DR (2008) Distance-dependent
plasmon resonant coupling between a gold nanoparticle and gold film. Nano Lett 8:2245–
2252
13. Pompa P, Martiradonna L, Della Torre A, Della Sala F, Manna L, De Vittorio M, Calabi F,
Cingolani R, Rinaldi R (2006) Metal-enhanced fluorescence of colloidal nanocrystals with
nanoscale control. Nat Nanotechnol 1:126–130
14. Ray K, Badugu R, Lakowicz JR (2007) Polyelectrolyte layer-by-layer assembly to con-
trol the distance between fluorophores and plasmonic nanostructures. Chem Mater 19:
5902–5909
15. Judd B (1962) Optical absorption intensities of rare-earth ions. Phys Rev 127:750
16. Meiser F, Cortez C, Caruso F (2004) Biofunctionalization of fluorescent rare-earth-doped
lanthanum phosphate colloidal nanoparticles. Angew Chem Int Ed 43:5954–5957
17. Ofelt G (1962) Intensities of crystal spectra of rare-earth ions. J Chem Phys 37:511–520
18. Schuetz P, Caruso F (2002) Electrostatically assembled fluorescent thin films of rare-earth-
doped lanthanum phosphate nanoparticles. Chem Mater 14:4509–4516
19. Liu C, Hou Y, Gao M (2014) Are rare-earth nanoparticles suitable for in vivo applications?
Adv Mater 26:6922–6932
20. Gai S, Li C, Yang P, Lin J (2014) Recent progress in rare earth micro/nanocrystals:
soft chemical synthesis, luminescent properties, and biomedical applications. Chem Rev
114:2343–2389
21. Jacobs RR, Krupke WF, Weber MJ (1978) Measurement of excited-state-absorption loss for
Ce3C in Y3Al5O12 and implications for tunable 5d! 4f rare-earth lasers. Appl Phys Lett
33:410–412
22. Mortier M, Auzel F (1999) Rare-earth doped transparent glass-ceramics with high cross-
sections. J Non-Cryst Solids 256:361–365
23. Walsh BM, Barnes NP, Di Bartolo B (1998) Branching ratios, cross sections, and radiative
lifetimes of rare earth ions in solids: application to Tm3C and Ho3C ions in LiYF4. J Appl
Phys 83:2772–2787
24. Park W, Lu D, Ahn S (2015) Plasmon enhancement of luminescence upconversion. Chem
Soc Rev 44:2940–2962
25. Derom S, Berthelot A, Pillonnet A, Benamara O, Jurdyc A, Girard C, Des Francs GC
(2013) Metal enhanced fluorescence in rare earth doped plasmonic core–shell nanoparticles.
Nanotechnology 24:495704
26. Feng AL, You ML, Tian L, Singamaneni S, Liu M, Duan Z, Lu TJ, Xu F, Lin M (2015)
Distance-dependent plasmon-enhanced fluorescence of upconversion nanoparticles using
polyelectrolyte multilayers as tunable spacers. Sci Rep 5:7779
27. Lee SM, Choi KC (2010) Enhanced emission from BaMgAl10O17:Eu2C by localized
surface plasmon resonance of silver particles. Opt Express 18:12144–12152
28. Liu F, Aldea G, Nunzi J-M (2010) Metal plasmon enhanced europium complex luminescence.
J Lumin 130:56–59
29. Schietinger S, Aichele T, Wang H-Q, Nann T, Benson O (2009) Plasmon-enhanced upconver-
sion in single NaYF4:Yb3C/Er3C codoped nanocrystals. Nano Lett 10:134–138
30. Sun Q-C, Mundoor H, Ribot JC, Singh V, Smalyukh II, Nagpal P (2013) Plasmon-
enhanced energy transfer for improved upconversion of infrared radiation in doped-lanthanide
nanocrystals. Nano Lett 14:101–106
31. Lee K-T, Park J-H, Kwon SJ, Kwon H-K, Kyhm J, Kwak K-W, Jang HS, Kim SY, Han JS,
Lee S-H (2015) Simultaneous enhancement of upconversion and downshifting luminescence
via plasmonic structure. Nano Lett 15:2491–2497
32. Li Z, Chen S, Li J, Liu Q, Sun Z, Wang Z, Huang S (2012) Plasmon-enhanced upconversion

fluorescence in NaYF4:Yb/Er/Gd nanorods coated with Au nanoparticles or nanoshells. J
Appl Phys 111:014310
33. Liu N, Qin W, Qin G, Jiang T, Zhao D (2011) Highly plasmon-enhanced upconver-
sion emissions from Au@“-NaYF4:Yb,Tm hybrid nanostructures. Chem Commun 47:
7671–7673
34. Saboktakin M, Ye X, Oh SJ, Hong S-H, Fafarman AT, Chettiar UK, Engheta N, Murray
CB, Kagan CR (2012) Metal-enhanced upconversion luminescence tunable through metal
nanoparticle–nanophosphor separation. ACS Nano 6:8758–8766
35. Yuan P, Lee YH, Gnanasammandhan MK, Guan Z, Zhang Y, Xu Q-H (2012) Plasmon
enhanced upconversion luminescence of NaYF4:Yb, Er@SiO2@Ag core–shell nanocompos-
ites for cell imaging. Nanoscale 4:5132–5137
36. Zhang F, Braun GB, Shi Y, Zhang Y, Sun X, Reich NO, Zhao D, Stucky G (2010) Fabri-
cation of Ag@SiO2@Y2O3:Er nanostructures for bioimaging: tuning of the upconversion
fluorescence with silver nanoparticles. J Am Chem Soc 132:2850–2851
37. Yin Z, Zhou D, Xu W, Cui S, Chen X, Wang H, Xu S, Song H (2016) Plasmon-Enhanced
upconversion luminescence on vertically aligned gold nanorod monolayer supercrystals. ACS
Appl Mater Interfaces 8:11667–11674
38. Saboktakin M, Ye X, Chettiar UK, Engheta N, Murray CB, Kagan CR (2013) Plasmonic
enhancement of nanophosphor upconversion luminescence in Au nanohole arrays. ACS Nano
7:7186–7192
39. Lin JH, Liou HY, Wang C-D, Tseng C-Y, Lee C-T, Ting C-C, Kan H-C, Hsu CC (2015)
Giant Enhancement of Upconversion Fluorescence of NaYF4:Yb3C,Tm3C Nanocrystals
with Resonant Waveguide Grating Substrate. ACS Photon 2:530–536
40. Xu W, Song H, Chen X, Wang H, Cui S, Zhou D, Zhou P, Xu S (2015) Upconversion
luminescence enhancement of Yb3C, Nd3C sensitized NaYF4 core–shell nanocrystals on
Ag grating films. Chem Commun 51:1502–1505
41. Fisher J, Zhao B, Lin C, Berry M, May PS, Smith S (2015) Spectroscopic imaging and power
dependence of near-infrared to visible upconversion luminescence from NaYF4:Yb3C,Er3C
nanoparticles on nanocavity arrays. J Phys Chem C 119:24976–24982
42. Greybush NJ, Saboktakin M, Ye X, Della Giovampaola C, Oh SJ, Berry NE, Engheta
N, Murray CB, Kagan CR (2014) Plasmon-enhanced upconversion luminescence in single
nanophosphor–nanorod heterodimers formed through template-assisted self-assembly. ACS
Nano 8:9482–9491
43. Yamamoto K, Fujii M, Sowa S, Imakita K, Aoki K (2015) Upconversion luminescence of
rare-earth-doped Y2O3 nanoparticle with metal nano-cap. J Phys Chem C 119:1175–1179
44. Piatkowski D, Hartmann N, Macabelli T, Nyk M, Mackowski S, Hartschuh A (2015)
Silver nanowires as receiving-radiating nanoantennas in plasmon-enhanced up-conversion
processes. Nanoscale 7:1479–1484
45. Green K, Wirth J, Lim SF (2016) Optical investigation of gold shell enhanced 25 nm diameter
upconverted fluorescence emission. Nanotechnology 27:135201
46. Mauser N, Piatkowski D, Mancabelli T, Nyk M, Mackowski S, Hartschuh A (2015) Tip
enhancement of upconversion photoluminescence from rare earth ion doped nanocrystals.
ACS Nano 9:3617–3626
47. Chen G, Ding C, Wu E, Wu B, Chen P, Ci X, Liu Y, Qiu J, Zeng H (2015) Tip-enhanced
upconversion luminescence in Yb3C–Er3C codoped NaYF4 nanocrystals. J Phys Chem C
119:22604–22610
48. Lu D, Cho SK, Ahn S, Brun L, Summers CJ, Park W (2014) Plasmon enhancement
mechanism for the upconversion processes in NaYF4:Yb3C,Er3C nanoparticles: Maxwell
versus Förster. ACS Nano 8:7780–7792
49. Boyer J-C, Vetrone F, Cuccia LA, Capobianco JA (2006) Synthesis of colloidal upconverting
NaYF4 nanocrystals doped with Er3C, Yb3C and Tm3C, Yb3C via thermal decomposition
of lanthanide trifluoroacetate precursors. J Am Chem Soc 128:7444–7445
50. Heer S, Kömpe K, Güdel HU, Haase M (2004) Highly efficient multicolour upconversion
emission in transparent colloids of lanthanide-doped NaYF4 nanocrystals. Adv Mater
16:2102–2105
51. Heer S, Lehmann O, Haase M, Güdel HU (2003) Blue, green, and red upconversion emission
from lanthanide-doped LuPO4 and YbPO4 nanocrystals in a transparent colloidal solution.
Angew Chem Int Ed 42:3179–3182
52. Wang F, Liu X (2008) Upconversion multicolor fine-tuning: visible to near-infrared emission
from lanthanide-doped NaYF4 nanoparticles. J Am Chem Soc 130:5642–5643
53. Wang J, Wang F, Wang C, Liu Z, Liu X (2011) Single-band upconversion emission in
lanthanide-doped KMnF3 nanocrystals. Angew Chem Int Ed 50:10369–10372
54. Ehlert O, Thomann R, Darbandi M, Nann T (2008) A four-color colloidal multiplexing
nanoparticle system. ACS Nano 2:120–124
55. Zhang F, Wan Y, Yu T, Zhang F, Shi Y, Xie S, Li Y, Xu L, Tu B, Zhao D (2007)
Uniform nanostructured arrays of sodium rare-earth fluorides for highly efficient multicolor
upconversion luminescence. Angew Chem Int Ed 46:7976–7979
56. Gao L, Ge X, Chai Z, Xu G, Wang X, Wang C (2009) Shape-controlled synthesis of
octahedral ’-NaYF4 and its rare earth doped submicrometer particles in acetic acid. Nano Res
2:565–574
57. Boyer J-C, Van Veggel FC (2010) Absolute quantum yield measurements of colloidal
NaYF4:Er3C,Yb3C upconverting nanoparticles. Nanoscale 2:1417–1419
58. Mai H-X, Zhang Y-W, Sun L-D, Yan C-H (2007) Highly efficient multicolor up-conversion
emissions and their mechanisms of monodisperse NaYF4:Yb, Er core and core/shell-
structured nanocrystals. J Phys Chem C 111:13721–13729
59. Yi G, Lu H, Zhao S, Ge Y, Yang W, Chen D, Guo L-H (2004) Synthesis, characterization, and
biological application of size-controlled nanocrystalline NaYF4:Yb, Er infrared-to-visible up-
conversion phosphors. Nano Lett 4:2191–2196
60. Krämer KW, Biner D, Frei G, Güdel HU, Hehlen MP, Lüthi SR (2004) Hexagonal sodium
yttrium fluoride based green and blue emitting upconversion phosphors. Chem Mater
16:1244–1251
61. Schäfer H, Ptacek P, Eickmeier H, Haase M (2009) Synthesis of hexagonal Yb3C, Er3C-
doped NaYF4 nanocrystals at low temperature. Adv Funct Mater 19:3091–3097
62. Yi GS, Chow GM (2006) Synthesis of hexagonal-phase NaYF4: Yb, Er and NaYF4:Yb,Tm
nanocrystals with efficient up-conversion fluorescence. Adv Funct Mater 16:2324–2329
63. Zeng JH, Su J, Li ZH, Yan RX, Li YD (2005) Synthesis and upconversion luminescence of
hexagonal-Phase NaYF4:Yb3C, Er3C phosphors of controlled size and morphology. Adv
Mater 17:2119–2123
64. Zhang Y, Yang L, Xu C, Zhong J, Sun C (2010) Sensitized deep-ultraviolet up-conversion
emissions of Gd3C via Tm3C and Yb3C in hexagonal NaYF4 nanorods. Appl Phys B
98:243–247
65. Li Z, Zhang Y (2006) Monodisperse silica-coated polyvinylpyrrolidone/NaYF4 nanocrystals
with multicolor upconversion fluorescence emission. Angew Chem Int Ed 45:7732–7735
66. Wang F, Chatterjee DK, Li Z, Zhang Y, Fan X, Wang M (2006) Synthesis of polyethylen-
imine/NaYF4 nanoparticles with upconversion fluorescence. Nanotechnology 17:5786
67. Yu X, Li M, Xie M, Chen L, Li Y, Wang Q (2010) Dopant-controlled synthesis of water-
soluble hexagonal NaYF4 nanorods with efficient upconversion fluorescence for multicolor
bioimaging. Nano Res 3:51–60
68. Liang X, Wang X, Zhuang J, Peng Q, Li Y (2007) Synthesis of NaYF4 nanocrystals with
predictable phase and shape. Adv Funct Mater 17:2757–2765
69. Li C, Yang J, Quan Z, Yang P, Kong D, Lin J (2007) Different microstructures of “-NaYF4
fabricated by hydrothermal process: effects of pH values and fluoride sources. Chem Mater
19:4933–4942
70. Mai H-X, Zhang Y-W, Si R, Yan Z-G, Sun L-D, You L-P, Yan C-H (2006) High-quality
sodium rare-earth fluoride nanocrystals: controlled synthesis and optical properties. J Am
Chem Soc 128:6426–6436
71. Wang F, Han Y, Lim CS, Lu Y, Wang J, Xu J, Chen H, Zhang C, Hong M, Liu X (2010)
Simultaneous phase and size control of upconversion nanocrystals through lanthanide doping.
Nature 463:1061–1065
72. Tian G, Gu Z, Zhou L, Yin W, Liu X, Yan L, Jin S, Ren W, Xing G, Li S (2012) Mn2C
dopant-controlled synthesis of NaYF4:Yb/Er upconversion nanoparticles for in vivo imaging
and drug delivery. Adv Mater 24:1226–1231
73. Zeng J, Huang J, Liu C, Wu CH, Lin Y, Wang X, Zhang S, Hou J, Xia Y (2010) Gold-based
hybrid nanocrystals through heterogeneous nucleation and growth. Adv Mater 22:1936–1940
74. Li P, Wei Z, Wu T, Peng Q, Li Y (2011) Au ZnO hybrid nanopyramids and their
photocatalytic properties. J Am Chem Soc 133:5660–5663
75. Seh ZW, Liu S, Zhang SY, Bharathi M, Ramanarayan H, Low M, Shah KW, Zhang YW,
Han MY (2011) Anisotropic growth of titania onto various gold nanostructures: synthesis,
theoretical understanding, and optimization for catalysis. Angew Chem Int Ed 50:10140–
10143
76. Liu X, Lee C, Law W-C, Zhu D, Liu M, Jeon M, Kim J, Prasad PN, Kim C, Swihart
MT (2013) Au–Cu2–xSe heterodimer nanoparticles with broad localized surface plasmon
resonance as contrast agents for deep tissue imaging. Nano Lett 13:4333–4339
77. Fang C, Jia H, Chang S, Ruan Q, Wang P, Chen T, Wang J (2014) (Gold core)/(titania shell)
nanostructures for plasmon-enhanced photon harvesting and generation of reactive oxygen
species. Energy Environ Sci 7:3431–3438
78. Jiang R, Li B, Fang C, Wang J (2014) Metal/semiconductor hybrid nanostructures for
plasmon-enhanced applications. Adv Mater 26:5274–5309
79. Kuo C-H, Hua T-E, Huang MH (2009) Au nanocrystal-directed growth of Au Cu2O core
shell heterostructures with precise morphological control. J Am Chem Soc 131:17871–17878
80. Kuo C-H, Yang Y-C, Gwo S, Huang MH (2010) Facet-dependent and Au nanocrystal-
enhanced electrical and photocatalytic properties of Au Cu2O Core shell heterostructures.
J Am Chem Soc 133:1052–1057
81. Li M, Yu XF, Liang S, Peng XN, Yang ZJ, Wang YL, Wang QQ (2011) Synthesis of Au–CdS
Core–shell hetero-nanorods with efficient exciton–plasmon interactions. Adv Funct Mater
21:1788–1794
82. Ruan Q, Fang C, Jiang R, Jia H, Lai Y, Wang J, Lin H-Q (2016) Highly enhanced transverse
plasmon resonance and tunable double Fano resonances in gold@ titania nanorods. Nanoscale
8:6514–6526
83. Wang H, Sun Z, Lu Q, Zeng F, Su D (2012) One-pot synthesis of (Au Nanorod)–(metal
sulfide) core–shell nanostructures with enhanced gas-sensing property. Small 8:1167–1172
84. Meng X, Fujita K, Moriguchi Y, Zong Y, Tanaka K (2013) Metal–dielectric core–shell
nanoparticles: advanced plasmonic architectures towards multiple control of random lasers.
Adv Opt Mater 1:573–580
85. Zhang J, Tang Y, Lee K, Ouyang M (2010) Tailoring light-matter-spin interactions in colloidal
hetero-nanostructures. Nature 466:91–95
86. Baek S-W, Park G, Noh J, Cho C, Lee C-H, Seo M-K, Song H, Lee J-Y (2014) Au@Ag
core–shell nanocubes for efficient plasmonic light scattering effect in low bandgap organic
solar cells. ACS Nano 8:3302–3312
87. Majhi SM, Rai P, Raj S, Chon B-S, Park K-K, Yu Y-T (2014) Effect of Au nanorods on poten-
tial barrier modulation in morphologically controlled Au@ Cu2O core–shell nanoreactors for
gas sensor applications. ACS Appl Mater Interfaces 6:7491–7497
88. Li J, Zeng HC (2005) Size tuning, functionalization, and reactivation of Au in TiO2
nanoreactors. Angew Chem 117:4416–4419
89. Qi J, Chen J, Li G, Li S, Gao Y, Tang Z (2012) Facile synthesis of core–shell Au@CeO2
nanocomposites with remarkably enhanced catalytic activity for CO oxidation. Energy
Environ Sci 5:8937–8941
90. Zhang L, Blom DA, Wang H (2011) Au–Cu2O core–shell nanoparticles: a hybrid metal-
semiconductor heteronanostructure with geometrically tunable optical properties. Chem
Mater 23:4587–4598
91. Li B, Gu T, Ming T, Wang J, Wang P, Wang J, Yu JC (2014) (Gold core)@(ceria shell)

nanostructures for plasmon-enhanced catalytic reactions under visible light. ACS Nano
8:8152–8162
92. Kong X, Yu Q, Zhang X, Du X, Gong H, Jiang H (2012) Synthesis and application of surface
enhanced Raman scattering (SERS) tags of Ag@SiO2 core/shell nanoparticles in protein
detection. J Mater Chem 22:7767–7774
93. Li C, Xu L, Zhao Y, Sun J, He D, Jiao H (2015) Synthesis, morphology, photoluminescence
and photocatalytic properties of yolk–shell Au@Y2O3: Eu3C sub-microspheres. J Alloys
Compd 627:31–38
94. Li H, Kang J, Yang J, Wu B (2016) Fabrication of Au nanoparticle@mSiO2@ Y2O3: Eu
nanocomposites with enhanced fluorescence. J Alloys Compd 673:283–288
95. Min Y-L, Wan Y, Yu S-H (2009) Au@Y2O3:Eu3C rare earth oxide hollow sub-microspheres
with encapsulated gold nanoparticles and their optical properties. Solid State Sci 11:96–101
96. Sirinakis G, Siddique R, Monokroussos C, Carpenter MA, Kaloyeros AE (2005) Microstruc-
ture and optical properties of Au–Y2O3-stabilized ZrO2 nanocomposite films. J Mater Res
20:2516–2522
97. Yin D, Wang C, Ouyang J, Zhang X, Jiao Z, Feng Y, Song K, Liu B, Cao X, Zhang L
(2014) Synthesis of a novel core–shell nanocomposite Ag@SiO2@Lu2O3:Gd/Yb/Er for
large enhancing upconversion luminescence and bioimaging. ACS Appl Mater Interfaces
6:18480–18488
98. Zhang N, Fu X, Xu Y-J (2011) A facile and green approach to synthesize Pt@ CeO2
nanocomposite with tunable core-shell and yolk-shell structure and its application as a visible
light photocatalyst. J Mater Chem 21:8152–8158
99. Chen H, Shao L, Li Q, Wang J (2013) Gold nanorods and their plasmonic properties. Chem
Soc Rev 42:2679–2724
100. Pérez-Juste J, Pastoriza-Santos I, Liz-Marzán LM, Mulvaney P (2005) Gold nanorods:
synthesis, characterization and applications. Coord Chem Rev 249:1870–1901
101. Xu Z, Kang X, Li C, Hou Z, Zhang C, Yang D, Li G, Lin J (2010) Ln3C (Ln D Eu, Dy, Sm,
and Er) ion-doped YVO4 nano/microcrystals with multiform morphologies: hydrothermal
synthesis, growing mechanism, and luminescent properties. Inorg Chem 49:6706–6715
102. Brixner L, Abramson E (1965) On the luminescent properties of the rare earth vanadates. J
Electrochem Soc 112:70–74
103. Liu J, Li Y (2007) General synthesis of colloidal rare earth orthovanadate nanocrystals. J
Mater Chem 17:1797–1803
104. Bao L, Li Z, Tao Q, Xie J, Mei Y, Xiong Y (2013) Controlled synthesis of uniform LaF3
polyhedrons, nanorods and nanoplates using NaOH and ligands. Nanotechnology 24:145604
105. Wang J, Huang H, Zhang D, Chen M, Zhang Y, Yu X, Zhou L, Wang Q (2015) Synthesis of
gold/rare-earth-vanadate core/shell nanorods for integrating plasmon resonance and fluores-
cence. Nano Res 8:2548–2561
106. Liu G, Conn CE, Drummond CJ (2009) Lanthanide oleates: chelation, self-assembly, and
exemplification of ordered nanostructured colloidal contrast agents for medical imaging. J
Phys Chem B 113:15949–15959
107. Zhang Y, Xiao Q, He H, Zhang J, Dong G, Han J, Qiu J (2015) Simultaneous luminescence
modulation and magnetic field detection via magneto-optical response of Eu3C-doped
NaGdF4 nanocrystals. J Mater Chem C 3:10140–10145
108. Liu X-L, Liang S, Nan F, Pan Y-Y, Shi J-J, Zhou L, Jia S-F, Wang J-B, Yu X-F, Wang
Q-Q (2013) Stepwise synthesis of cubic Au-AgCdS core-shell nanostructures with tunable
plasmon resonances and fluorescence. Opt Express 21:24793–24798
109. Zhang J, Tang Y, Lee K, Ouyang M (2010) Nonepitaxial growth of hybrid core-shell
nanostructures with large lattice mismatches. Science 327:1634–1638
110. Song Y, Liu G, Dong X, Wang J, Yu W, Li J (2015) Au nanorods@ NaGdF4/Yb3C,Er3C
multifunctional hybrid nanocomposites with upconversion luminescence, magnetism, and
photothermal property. J Phys Chem C 119:18527–18536
111. Chen M, Wang J-H, Luo Z-J, Cheng Z-Q, Zhang Y-F, Yu X-F, Zhou L, Wang Q-Q (2016)
Facile synthesis of flower-shaped Au/GdVO4: Eu core/shell nanoparticles by using citrate as
stabilizer and complexing agent. RSC Adv 6:9612–9618
112. Sun Y, Chen Y, Tian L, Yu Y, Kong X, Zhao J, Zhang H (2007) Controlled synthesis
and morphology dependent upconversion luminescence of NaYF4: Yb, Er nanocrystals.
Nanotechnology 18:275609
113. Zhang H, Li Y, Ivanov IA, Qu Y, Huang Y, Duan X (2010) Plasmonic modulation of the
upconversion fluorescence in NaYF4:Yb/Tm hexaplate nanocrystals using gold nanoparticles
or nanoshells. Angew Chem 122:2927–2930
114. Wang Y-L, Estakhri NM, Johnson A, Li H-Y, Xu L-X, Zhang Z, Alù A, Wang Q-Q, Shih
C-KK (2015) Tailoring plasmonic enhanced upconversion in single NaYF4: Yb3C/Er3C
nanocrystals. Sci Rep 5:10196
115. Singh A, Jayabalan J, Chari R, Srivastava H, Oak S (2010) Tuning the localized surface
plasmon resonance of silver nanoplatelet colloids. J Phys D Appl Phys 43:335401
116. Washio I, Xiong Y, Yin Y, Xia Y (2006) Reduction by the end groups of poly (vinyl
pyrrolidone): a new and versatile route to the kinetically controlled synthesis of Ag triangular
nanoplates. Adv Mater 18:1745–1749
117. Zhang W, Ding F, Chou SY (2012) Large enhancement of upconversion luminescence of
NaYF4:Yb3C/Er3C nanocrystal by 3D plasmonic nano-antennas. Adv Mater 24:OP236–
OP241
118. Luu Q, Hor A, Fisher J, Anderson RB, Liu S, Luk T-S, Paudel HP, Farrokh Baroughi M,
May PS, Smith S (2014) Two-color surface plasmon polariton enhanced upconversion in
NaYF4:Yb:Tm nanoparticles on Au nanopillar arrays. J Phys Chem C 118:3251–3257
119. Wang Y, Nan F, Cheng Z, Han J, Hao Z, Xu H, Wang Q (2015) Strong tunability of cooperative
energy transfer in Mn2C-doped (Yb3C, Er3C)/NaYF4 nanocrystals by coupling with silver
nanorod array. Nano Res 8:2970–2977
Chapter 3
Sensing Through Surface Plasmon Resonance
Technique
R. K. Gupta
Abstract The optical phenomenon, surface plasmon resonance (SPR) has become
extremely popular owing to its high sensitivity, label-free and non-destructive mea-
surement towards any molecular specific interaction. This is one of the widely used
phenomena for biological, chemical and gas sensing devices. There are different
ways that the SPR phenomenon can be employed for such sensing applications.
In this chapter, we will review some of the important SPR techniques and their
applications in sensing.
Keywords Surface Plasmon Resonance • Angular interrogation • Wavelength

interrogation • SPR sensors
In the modern technologically advanced society, a large number of high performance

sensors are required. Sensors are required to monitor continuously health and
environment,and drive the advanced instruments. Sensors are integrated to many
devices e.g. smart watch, mobile phones, automobiles, potable water resources,
satellites etc and they are providing valuable information on the health and
environment. Sensors are ubiquitous and enormous amount of efforts are being
invested in this field. In general, most of the sensing technologies are based on
molecular specific interaction between the ligands and the analytes. In order to
build a sensor, an active area is created by immobilizing appropriate ligands. On
exposure to the sample possessing the specific analytes, the molecular specific
interaction establishes between the ligands and analytes. Such interaction may
change the optical, electrical, or other physical properties which can be perceived by
the transducer of the sensor. The resolution of a sensor depends on such perceptible
changes. At the same time, the sensitivity depends on the rate of change in the
perceptible-changes due to the change in the concentration. In addition to these,
the quality of sensor depends on the selectivity of the analytes. Therefore, any
sensor has to undergo a benchmarking on these qualities with reference to some
standards. With the development of nanoscience and nanotechnology and advanced
R.K. Gupta ()

Department of Physics, Birla Institute of Technology and Science, 333031, Pilani, India
e-mail: raj@pilani.bits-pilani.ac.in

DOI 10.1007/978-3-319-48081-7_3
40 R.K. Gupta
instrumentation, the field of sensor is growing exponentially. The sensor based

on the optical phenomenon, surface plasmon resonance (SPR) is gaining more
scientific attentions because it provides non-destructive, label free, very sensitive
and high resolution sensor. The instrumentation is simpler and can be miniaturized
for field deployment.
3.1 Principle of Surface Plasmon Resonance (SPR)
The SPR phenomenon was first observed by Otto in 1968 [1] however the
commercial realization took place in 1990 by Biacore of GE Healthcare. The surface
plasmon polaritons (SPPs) are transverse magnetic (TM) waves which arise due
to charge density oscillation at metal-dielectric interface.SPPs can be excited by a
polarized light wave incident onto the metallic layer (e.g. gold film) via a coupling
high refractive index medium (e.g. glass prism) [2, 3]. At SPR, the component of
wavevector of incident light planar to the interface (kx ) matches to that of surface
plasmon wave (kp ) resulting in the absorption of energy of the reflected beam [4].
Let i be the angle of incidence (greater than critical angle of a coupling prism), npri
is refractive index of the prism and is the wavelength of the incident light, then
2
kx D npri sin .i /

Here, kx can be changed by changing either angle of incidence ( i ) or the

wavelength of the light ().
The wavevector of the surface plasmon wave is given by
r
! "1 "2
kp D
c "1 C "2
Where ¨ is angular frequency, c is speed of light in vacuum, "1 & "2 are dielectric
constants of the metal film and the organic film deposited on the metal surface.
At surface plasmon resonance,
kx D kp
At the resonance, the reflected intensity diminishes to the minimum value due to
the maximum transfer of electromagnetic energy from the incident light to that of
surface plasmon wave.
For establishing the SPR at a given interface, the wavevector matching can take
place by two different ways – either by changing the angle of incidence (™) of a
monochromatic light beam and recording the resonance angle (™SPR ) or by allowing
a polychromatic light at a fixed ™ and recording the resonance wavelength (œSPR ).
3 Sensing Through Surface Plasmon Resonance Technique 41
Fig. 3.1 Dispersion curves of the surface plasmon polaritons (black solid line) and light wave
travelling in vacuum (n D 1, blue line) and in a coupling medium (n > 1, red line). Here ¨ is
angular frequencyand n is refractive index of the coupling medium
The first technique is popularly known as angular interrogation whereas the second
one is known as wavelength interrogation technique.
Figure 3.1 shows the dispersion curves for electromagnetic wave in free space
and in the coupling medium of refractive index, n. The curves clearly indicate
that non-zero solution can be achieved only when the electromagnetic wave travel
through a coupling medium (waveguide) with refractive index greater than that of
free space. Therefore, the wavevector matching can take place only if the light
is allowed to incident on the metal-surface (e. g. gold surface) via a waveguide
exhibiting a refractive index (n) greater than that of the free space [4].
Therefore, in SPR instrumentation the light should travel and incident on the
metallic surface through a coupling medium at an angle of incidence greater than
the critical angle of the air-medium interface. The evanescent wave interacts with
that of free charge density on the metal surface and can create the SPPs. SPR can
be established either by changing angle of incidence or changing wavelength of the
incident light in angular or wavelength interrogation modes, respectively.
At resonance, the intensity of the reflected light reduces to a minimum value.
The kp of the SPPs is extremely sensitive to the change in refractive index of
the dielectric over the metal surface. This is the underlying principle for any
biological, chemical or gas sensing application of the SPR. Due to molecular-
specific interaction, the refractive index of the film deposited on the metal surface
changes leading to a change in kp . Consequently, the resonance will occur at a
different angle of incidence in the angular interrogation mode. Recording the shift in
the resonance angle (RA), change in refractive index due to the molecular-specific
interaction can be estimated [5]. Such processes will be employed to understand the
molecular interactions and can be exploited for sensing applications. The schematic
42 R.K. Gupta
of the SPR spectra in angular interrogation method is shown in Fig. 3.2(a). The shift
in the resonance angle (™1 !™2 ) is observed due to adsorption of analytes onto the
gold surface causing a change in the dielectrics.
In wavelength interrogation method, polychromatic light is used to excite the
surface plasmon wave in the thin metallic layer. At SPR, the electromagnetic
wave at the resonance wavelength (œSPR ) exhibits minimum reflected intensity. The
schematic of the SPR spectra in the wavelength interrogation method is shown in
Fig. 3.2(b). The shift in the resonance wavelength (œSPR1 !œSPR2 ) is observed due
to adsorption of analytes on the gold surface.
The change in RA (™) due to analyte adsorption at a given œ is
™ D An C Bd
where A and B are constants, n and d are changes in refractive index and
thickness due to analyte adsorption, respectively. A quantitative understanding of
SPR data is reported in the reference [6].
The SPR phenomenon can be used for studying the molecular affinity and other
kinetic parameters during sensing. It can be employed for estimating the association
and dissociation constants and the concentration of the analytes. A schematic of the
kinetic curves for different concentration of the analyte is shown in Fig. 3.3. The
curves can be fit theoretically to estimate the association and dissociation constants
and hence the binding kinetic constant [7, 8].
3.2 Techniques Based on SPR for Sensing Applications
3.2.1 Angular Interrogation Based Sensor
In the angular interrogation mode, the excitation of surface plasmon with the
evanescent wave can be achieved employing two configurations: (1) Kretschmann-
Reather geometry and (2) Otto geometry. In Kretschmann-Reather geometry, a high
refractive index (RI) waveguide (nw ) is interfaced with a metal-dielectric waveguide
having thin metal film with permittivity "m and thickness (t), and a semi-infinite
dielectric with a refractive index nd (nd <nw ). The high RI waveguide is generally
glass prism. Here, the light propagating through the prism is made to incident on
the metal film. Above the critical angle of the waveguide, the light confines inside
the prism due to total internal reflection. In this condition, a part of the EM energy
propagates as evanescent wave in the metallic layer. When the metallic film is
sufficiently thin (t<100 nm), the evanescent wave couples with the surface plasmons
and excites the SPPs. The wavevector of surface plasmons along the metal surface is
influenced by the nature of dielectric onto it. The change in dielectrics on the metal
surface leads to the change in wavevector of the surface plasmon. This condition
alters the wave matching condition and thereby shifts the resonance condition. The
geometry is as shown in Fig. 3.4.
Fig. 3.2 The SPR spectra obtained in (a) angular and (b) wavelength interrogation methods. The
shift in resonance angle of incidence and resonance wavelength can occur due to adsorption of
molecules on the reference surface
44 R.K. Gupta
Fig. 3.3 The schematic representation of the kinetic curves obtained at different concentration of
the analytes
Fig. 3.4 Schematic diagram for SPR sensor in Kretschmann-Reather geometry. The geometry
involves a semicylindrical glass prism (waveguide) deposited with gold film of thickness 50 nm.
The surface of gold film can be functionalized appropriately for sensing application
In Otto geometry, the high RI waveguide is interfaced with dielectric-metal

waveguide consisting of a thin dielectric film with refractive index nd (nd <nw ) and
thickness (t) and semi infinite metal with permittivity, "m . Here, the thickness of
dielectric layer chosen to be appropriate for the coupling of evanescent wave to the
surface plasmon of the metal. The geometry is as shown in Fig. 3.5. Due to practical
limitations of integrating flowcell and ease of surface functionalization for sensing
applications, the Otto geometry is not very popular.
At the same time, the Krestchmann-Reather configuration is the most popular
mode of SPR sensor because it provides not only very high sensitivity but also
higher degree of practical flexibility to alter the sensing surface with immobilization
of appropriate ligands. In angular interrogation method, the angle of incidence of a
Fig. 3.5 Schematic diagram for SPR sensor in Otto configuration. The geometry requires a
semicylindrical glass prism (waveguide) and gold layer with a separation between the prism and
gold layer is 200 nm
tightly focused monochromatic light onto the sensing area via the coupling prism
is changed and reflected intensity is collected simultaneously. The dip in intensity
indicates the resonance. Any adsorption on the sensing element causes a shift in
RA. The angular interrogation can be achieved either by rotating prism-assembly or
by rotating source and detector simultaneously. Such a mechanical driven scanning
system makes the instrument bulky, costly and complex in operation. In general, the
basic SPR instrumentation involves the optical components to be mounted on the
arms of a goniometer. In this configuration, the angle of incidence and reflection
are changed equally, and the reflected intensity is measured simultaneously. In
another configuration, the prism-assembly and the detector are rotated and the light
source is kept fixed [9]. Goniometric SPR with a fixed photo-detector is achieved by
employing two prism configuration where the illuminated prism face is in contact
with another prism with a wedge like separation [10]. This configuration never
keeps the interrogation spot constant at the surface of illuminated prism. Constant
interrogation spot is obtained at the surface of the prism by fixing the reflecting
surface of the prism at the axis of rotation of the arms of goniometer. The inclusion
of optical components always causes the alignment problems and also leads to fall of
intensity after each refraction/reflection in the optical components. Two prism SPR
sensor with an extra prism results in lower intensity of the reflected beam onto the
detector as compared to the single prism configuration, and therefore degrades the
sensitivity of instrument. Recently, Devanarayanan et al have adopted a new opto-
mechanical scanning mechanism to obtain the SPR spectra in the high sensitive
angular interrogation mode without employing a goniometer. Here, the laser as well
as the detector is kept stationary. As the angle of incidence changes due to rotation
of the scanning mirror, the reference spot position on the gold surface shifts and
as a consequence the spot of reflected beam on the quadrant photodiode (QPD)
shifts. In order to retain the spot position fixed onto the gold surface, the prism-
assembly is translated vertically till original reference spot position is regained.
46 R.K. Gupta
Fig. 3.6 The schematic of an angular interrogation based SPR instrument. (A) Laser (B) mirror
(C) piezomotor, (D) prism assembly, (E) translation stage, (F) stepper motor (G) segmented
photodiode, (H) data acquisition hardware, (I) PC
Under this condition, the total reflected intensity and the angle of incidence are
recorded. Fig. 3.6 shows the schematic of the instrument [11, 12]. The proposed
scheme employs less number of optical components with low mass and therefore
the overall optical complexity, size and mass of the instrument reduces, and the
portability increases. The developed SPR instrument is employed to study the
dielectric layers of various materials e.g. water, liquid crystal, stearic acid, cadmium
stearate and aqueous solution of sucrose. The dielectric layers were formed onto the
gold surface by drop cast, spin coating and Langmuir-Blodgett (LB) techniques. The
angular resolution obtained in this configuration is 22 degrees which is comparable
to many commercial SPR instruments. Using the developed SPR instrument, the
lowest detection limit of the sucrose in aqueous solution was found to be 100 femto-
molar. The resolution of the instrument was thus found to be 1.92 RIU (micro
refractive index unit). The instrument is calibrated using sucrose solution. The
sensitivity of the instrument was found to be 52.6o /RIU. The developed instrument
is very low cost, portable and sensitive [13].
3.2.2 Wavelength Interrogation Based Sensor
In the wavelength interrogation mode, the light beam from a polychromatic source is
allowed to incident onto the sensing area at an angle greater than the critical angle of
the prism. The reflected beam is analyzed and resonance wavelength corresponding
to the minimum reflectivity (œSPR ) is recorded. The change in the refractive index
of the dielectric over the metal surface shifts the œSPR. The wavelength based SPR
sensors are portable and economical however the resolution and sensitivity are poor
Fig. 3.7 Schematic representation of fiber optics based SPR sensor. Gold layer followed by
appropriate ligands are deposited on the core
as compared to that of angular interrogation SPR instrument. In a brilliant attempt,

a very low cost and hand held wavelength interrogation based SPR sensor was
developed and it is popularly known as SPREETA. The instrument is equipped with
a light emitting diode light source, a waveguide, sensing element (gold layer) and
an array detector.
The wavelength interrogation based sensor can undergo further miniaturization
by using an optical fiber as a waveguide. The schematic of the optical fiber based
SPR sensor is shown in Fig. 3.7. Here, a small region of the core is deposited with
gold film followed by the appropriate ligands for a particular sensing application
after removal of the cladding of an optical fiber. The one end of the such assembly is
illuminated with the polychromatic light and the transmitted light is collected from
other side. The spectrum is generated and the œSPR is recorded from the spectrum.
The sensing region can be connected to a flow cell and the response curve can be
obtained. The kinetic parameters can be estimated from the response curves.
In an interesting work by Liu et al, a biosensor is fabricated using the optical fiber
on a mobile phone platform [14]. The LED flash light and camera of the mobile was
used as the source of light and detector, respectively. The flash light and the camera
were coupled through a SPR optical fiber sensor integrated with a flow cell. The
authors have demonstrated the sensing of bovine immunoglobulin G (IgG) using
the staphylococcal protein A (SPA).
3.2.3 SPR Imaging Based Sensor
Selectivity and multi-analyte detection is difficult with the conventional SPR

instruments. In order to address such issues, multichannel SPR instruments are
developed. However, such instruments are very expensive and massive. In 1989,
Hickel et al have demonstrated SPR microscopy wherein the different phases of
a lipid monolayer onto gold substrate yield different contrast [15]. This work was
followed up by studying the orientation of molecules in the ultrathin films obtained
by self-assembled monolayer [16, 17] and Langmuir-Blodgett technique [18, 19].
48 R.K. Gupta
Fig. 3.8 Schematic showing the basic principle of SPR imaging. The light is allowed to incident
at the RA of ligand A. The corresponding intensities IA and IB are recorded by the CCD camera
This work is adopted for the development of SPR imaging based sensor. The
basic principle of SPR imaging is shown in Fig. 3.8. As shown in the figure, the
gold surface is functionalized with two different ligands(A and B) in two different
regions of the gold surface. The monochromatic light is allowed to incident at SPR
angle of a reference (ligand A) and the reflected light is collected using a charge
coupled device (CCD) camera. The two regions on the gold surface with different
functionalization can adsorb different analytes. The intensity of such functionalized
spots may therefore vary according to the adsorbed species and their concentration.
In order to probe multispecies simultaneously, the gold surface can be function-
alized by creating an array of spots with different kinds of functionalization. The
data can be collected and analyzed in real time and decision can be taken rapidly
and more accurately. The SPR imaging is extremely popular in the bio-sensing and
biomedical application [20, 21].
3.3 Sensing Applications of SPR
The SPR technology is used successively for monitoring the environment, health,
food quality and safety. There are numerous reports in literature on sensing
applications of SPR. The biosensing application of SPR was first demonstrated
by Liedberg et at in 1983 [22] and since then the SPR phenomenon is used for
studying molecular specific interactions e.g. DNA-DNA, protein-protein, DNA-
protein, drug-protein and drug-membrane.
The sensing using the SPR instrument is listed in Table 3.1.
Table 3.1 Sensing applications using surface plasmon resonance phenomenon

References
Environmental monitoring
Detection of heavy metals e.g. Cd, Zn and Ni using a protein metallothinein. [23]
Lowest detectable concentration (LDC) was 100 ng/mL.
Detection of Cu2C ions using a polymer composite of squarylium dye deposited [24]
on the gold surface of a SPR instrument. The lowest detectable concentration
was found to be 1 pM.
Detection of organophosphate pesticide chlorpyrifos and carbaryl using [25, 26, 27]
pesticide sensitive SPR sensor. The detection limits were found to be around
1 ng/mL for carbaryl and 50 pg/mL for chlorpyrifos.
Detection of dichlorodiphenyltrichloroethane (DDT) in water using monoclonal [28]
antibodies specific to DDT and its derivatives immobilized over the gold surface
of SPR sensor. The LDC was found to be 15 pg/mL.
Detection of pesticide, 2,4-dichlorophenoxyacetic acid (2,4-D) using the [29]
indirect inhibition immunoassay by monitoring the interaction between
anti-2,4-D antibody and concanavalin A-2,4-D conjugate using SPR sensor. The
LDC was found to be 3 ppb.
The carcinogenic and environmental pollutant, benzo[a]pyrene was detected [30]
using the SPR sensor employing the indirect inhibition immunoassay format.
The LDC was 10 ppt.
The explosive material and environmental pollutant, trinitrotoluene (TNT) was [31]
detected by monitoring the immunoreaction between trinitrophenol-bovine
serum (TNP-BSA) conjugate and anti-TNP antibody using SPR sensor. The
LDC was reported to be 60 ppt.
Food quality and safety
Staphylococcal enterotoxin B (SEB) was detected in milk and mushroom [32, 33]
samples using the antibody of SEB immobilized onto gold surface of SPR
sensor. The LDC was found to be 1 ng/mL.
Domoic acid (DA) was detected in shellfish e.g. mussels, oysters and cockles in [34]
inhibition format of the SPR sensor. The detection limit of DA in the fishes were
found to be 1 g/g.
Deoxynivalenol was detected in wheat with LDC 2.5 ng/mL using the SPR [35]
sensor. Deoxynivalenol conjugated with casein was immobilized onto the gold
surface and the detection was done in inhibition format.
Steroid hormone, progesterone was detected in milk using inhibition format. [36]
The progesterone derivatives were immobilized onto the gold surface of the
SPR. The LDC was found to be 3.6 ng/mL.
Peanut allergens were detected using the immobilized specific antibody on the [37]
gold surface of a SPR sensor. The LDC was reported to be 700 ng/mL.
Detection of contaminants, 4-nonylphenol in shellfish using the monoclonal [38]
specific antibodies and inhibition detection format in the SPR sensor. The LDC
was found to be 2 ng/g.
The adulterants, non-milk proteins in dairy product is detected employing [39]
polyclonal specific antibodies using a SPR sensor. The LDC was found to be
200 ppb.
Detection of Escherichia coli O157-H7 based on immunoassay sensing protocol [40]
using SPR sensor. The detection limit was found to be 10 [4] cells/mL.
Foodborne pathogens, Salmonella paratyphi was detected using a SPR based [41]
immunosensor employing the protein G. The detection limit was 102 CFU/mL.
(continued)
50 R.K. Gupta
Table 3.1 (continued)

References
Bio-medical
Morphine-3-glucuronide, an indicator for morphine and heroine was detected [42]
using polyclonal antibodies in SPR sensor. The LDC was found to be 0.3 ppb.
The markers for cardiac muscle injury, myoglobin and cardiac troponin I were [43]
detected using the antigen specific antibodies employing the SPR sensor. The
LDC of myoglobin and troponin were 2.9 and 1.4 ng/mL.
The prostate cancer marker, prostate specific antigen (PSA) was detected using [44]
the monoclonal antibodies against PSA employing the SPR sensor. The LDC
was found to be 0.15 ng/mL.
The pancreatic cancer marker, carbohydrate antigen (CA 19-9) was detected by [45]
specific antibody against CA 19-9 using a SPR sensor. The LDC was reported to
be 410 U/mL.
Vascular endothelial growth factor protein (VEGF) can have role in lung, breast [46]
and colorectal cancers. VEGF was detected using SPR imaging by forming
microarray of the RNA. The sensor can detect 1 pM.
Anti-protein S antibodies against Varicella-Zolter virus infection were detected [47]
using SPR sensor. A high response was observed from the sample of infected
person whereas minimum response from that of healthy person.
Detection of antibodies against hepatitis G in serum using synthetic peptides as [48]
biorecognition element of the SPR sensor.
Detection of pregnancy marker, human chorionic gonadotropin hormone (hCG) [49]
was performed using a wavelength modulated SPR sensor with DNA-directed
antibody immobilization method. The LDC was 0.5 ng/mL.
Estradiol, a steroid hormone was detected by inhibition detection format using [50]
SPR sensor. The binding of antibody to estradiol conjugate was monitored using
the SPR sensor. The LDC was found to be 0.14 ng/mL.
The insulin was detected using SPR sensor in inhibition detection format. A [51]
LDC was reported to be 1 ng/mL.
3.4 Conclusion
SPR phenomenon has widely been used for the sensing of vital analytes. There
are few hundreds of articles in literature wherein the SPR technology has been
used for sensing application. It is one of the most successful technology for
sensing gas, chemical and biological samples. The technology associated with
SPR phenomenon is advancing rapidly and the scientific communities are working
together to introduce a low cost, rapid detectable, reliable and field deployable SPR
sensors. In this chapter, we have introduced a brief review of SPR technology and
few applications for monitoring environment, food safety, and bio-medical areas.
Acknowledgement We are thankful to Department of Science and Technology, India for financial
support for the development of a SPR instrument as in [13]. Thanks to Dr V. Manjuladevi for
discussion. Thanks are also due to University of Colorado for providing local infrastructure during
the manuscript preparation.
References
1. Otto A (1968) Excitation of nonradiative surface plasma waves in silver by the method of
frustrated total reflection. Z PhysA Hadrons Nucl 216:398
2. Raether H (1988) Surface plasmon on smooth and rough surfaces and grating. Springer, Berlin
3. Kretschmann E, Raether H (1968) Radiative decay of non-radiative surface plasmons excited
by light. ZNatureforsch 23A:2153
4. Novotny L, Hecht B (2008) Principles of nano-optics. Cambridge University Press, New York
5. Devanarayanan VP, Manjuladevi V, Poonia M, Gupta RK, Gupta SK, Akhtar J (2016)
Measurement of optical anisotropy in ultrathin films using surface plasmon resonance. J
MolStruct 1103:281
6. Jung LS, Campbell CT, Chinowsky TM, Mar MN, Yee SS (1998) Quantitative interpretation
of the response of surface plasmon resonance sensors to adsorbed films. Langmuir 14:5636
7. Daghestani HN, Day BW (2010) Theory and application of surface plasmon resonance,
resonant mirror, resonant waveguide grating, and dual polarization interferometry biosensors.
Sensors 10:9630
8. Hahnefeld C, Drewianka S, Herberg FW (2004) Determination of kinetic data using surface
plasmon resonance biosensors. Methods Mol Med 94:299
9. Tiwari K, Sharma SC (2014) Surface plasmon based sensor with order-of-magnitude higher
sensitivity to electric field induced changes in dielectric environment at metal/nematic liquid-
crystal interface. Sens Actuators A 216:128
10. Mohanty BC, Kasiviswanathan S (2005) Two-prism setup for surface plasmon resonance
studies. Rev Sci Instr 76:033103
11. Liang H, Miranto H, Granqvist N, Sadowski JW, Viitala T, Wang B, Ylipertulla M (2010)
Surface plasmon resonance instrument as a refractometer for liquids and ultrathin films. Sens
Actuators B 149:212
12. Liu Y, Xu S, Tang B, Wang Y, Zhou J, Zheng X, Zhao B, Xu W (2010) Simultaneous
measurement of surface plasmon resonance and surface-enhanced Raman scattering. Rev Sci
Instr 81:036105
13. Devanarayanan VP, Manjuladevi V, Gupta RK (2016) Surface plasmon resonance sensor based
on a new opto-mechanical scanning mechanism. Sens Actuators B 227:643
14. Liu Y, Liu Q, Chen S, Cheng F, Wang H, Peng W (2015) Surface plasmon resonance biosensor
based on smart phone platform. Sci Rep 5:12864
15. Hickel W, Kamp D, Knoll W (1989) Surface plasmon microscopy. Nature 339:186
16. Haussling L, Ringsdorf H, Schmitt FJ, Knoll W (1991) Biotin-functionalized self-assembled
monolayers on gold: surface plasmon optical studies of specific recognition reactions. Lang-
muir 7:1837
17. Evans SD, Flynn TM, Ulman A (1995) Self-assembled multilayer formation on predefined
templates. Langmuir 11:3811
18. HickelW KW (1990) Surface plasmon microscopy of lipid layers. Thin Solid Films 187:349
19. Duschl C, LileyM LH, Ghandi A, Zakeeruddin SM, Stahlberg H, Dabochet J, Nemetz A,
Knoll W, Vogel H (1996) Sulphur-bearing lipids for the covalent attachment of supported lipid
bilayers to gold surfaces: a detailed characterisation and analysis. Mater Sci Eng C 4:7–18
20. Brockman JM, Nelson BP, Corn RM (2000) Surface plasmon resonance imaging measure-
ments of ultrathin organic thin films. Annu Rev Phys Chem 51:41
21. Wong CL, Olivo M (2014) Surface plasmon resonance imaging:a review. Plasmonics 9:809
22. Liedberg B, Nylander C, Lundstrom I (1983) Surface plasmon resonance for gas detection and
biosensing. Sens Actuat 4:229
23. Wu LP, Li YF, Huang CZ, Zhang Q (2006) Visual detection of sudan dyes based on the plasmon
resonance light scattering signals of silver nanoparticles. Anal Chem 78:5570
24. Ock K, Jang G, Roh Y, KimS KJ, Koh K (2001) Optical detection of Cu2C ion using a SQ-dye
containing polymeric thin-film on Au surface. Microchem J 70:301
52 R.K. Gupta
25. Mauriz E, Calle A, Manclus JJ, Montoya A, Escuela AM, Sendra JR, Lechuga LM (2006)
Single and multi-analyte surface plasmon resonance assays for simultaneous detection of
cholinesterase inhibiting pesticides. Sens Actuators B 118:399
26. Mauriz E, Calle A, Abad A, Montoya A, Hildebrandt A, Barcelo D, Lechuga LM (2006)
Determination of carbaryl in natural water samples by a surface plasmon resonance flow-
through immunosensor. Biosens Bioelectron 21:2129
27. Mauriz E, Calle A, Lechuga LM, Quintana J, Montoya A, Manclus JJ (2006) Real-time
detection of chlorpyrifos at part per trillion levels in ground, surface and drinking water samples
by a portable surface plasmon resonance immunosensor. Anal Chim Acta 561:40
28. Mauriz E, Calle A, Manclus JJ, Montoya A, Hildebrandt A, Barcelo D, Lechuga LM (2007)
Optical immunosensor for fast and sensitive detection of DDT and related compounds in river
water samples. Biosens Bioelectron 22:1410
29. Svitel J, Dzgoev A, Ramanathan K, Danielsson B (2000) Surface plasmon resonance based
pesticide assay on a renewable biosensing surface using the reversible concanavalin a
monosaccharide interaction. Biosens Bioelectron 15:411
30. Miura N, Sasaki M, Gobi KV, Kataoka C, Shoyama Y (2003) Highly sensitive and selective
surface plasmon resonance sensor for detection of sub-ppb levels of benzo[a]pyrene by indirect
competitive immunoreaction method. Biosens Bioelectron 18:953
31. Shankaran DR, Gobi KV, Sakai T, Matsumoto K, Imato T, Toko K, Miura N (2005) A novel
surface plasmon resonance immunosensor for 2,4,6-trinitrotoluene (TNT) based on indirect
competitive immunoreaction: a promising approach for on-site landmine detection. IEEE Sens
5:616
32. Nedelkov D, Rasooly A, Nelson RW (2000) Multitoxin biosensor–mass spectrometry analysis:
a new approach for rapid, real-time, sensitive analysis of staphylococcal toxins in food. Int J
Food Microbiol 60:1
33. Nedelkov D, RWc N (2003) Detection of staphylococcal enterotoxin B via biomolecular
interaction analysis mass spectrometry. Appl Environ Microbiol 69:5212
34. Traynor IM, Plumpton L, Fodey TL, Higgins C, Elliott CT (2006) Immunobiosensor detection
of domoic acid as a screening test in bivalve molluscs: comparison with liquid chromatography-
based analysis. J AOAC Int 89:868
35. Tudos AJ, Lucas-van den Bos ER, Stigter EC (2003) Rapid surface plasmon resonance-based
inhibition assay of deoxynivalenol. J Agric Food Chem 51:5843
36. Gillis EH, Gosling JP, Sreenan JM, Kane M (2002) Development and validation of a biosensor-
based immunoassay for progesterone in bovine milk. J Immunol Methods 267:131
37. Mohammed I, Mullett WM, Lai EPC, Yeung JM (2001) Is biosensor a viable method for food
allergen detection. Anal Chim Acta 444:97
38. Samsonova JV, Uskova NA, Andresyuk AN, Franek M, Elliott CT (2004) Biacore biosensor
immunoassay for 4-nonylphenols: assay optimization and applicability for shellfish analysis.
Chemosphere 57:975
39. Haasnoot W, Olieman K, Cazemier G, Verheijen R (2001) Direct biosensor immunoassays for
the detection of nonmilk proteins in milk powder. J Agric Food Chem 49:5201
40. Oh BK, Kim YK, Bae YM, Lee WH, Choi JW (2002) Detection of Escherichia coli O157:H7
using immunosensor based on surface plasmon resonance. J Microbiol Biotechnol 12:780
41. Oh BK, Lee W, Kim YM, Lee WH, Choi JW (2004) Surface plasmon resonance immunosensor
using self-assembled protein G for the detection of salmonella paratyphi. J Biotechnol 111
42. Dillon PP, Daly SJ, Manning BM, Kennedy RO (2003) Immunoassay for the determination
of morphine-3-glucuronide using a surface plasmon resonance-based biosensor. Biosens
Bioelectron 18:217
43. Masson JF, Obando L, Beaudoin S, Booksh K (2004) Sensitive and real-time fiber-optic-based
surface plasmon resonance sensors for myoglobin and cardiac troponin. Talanta 62:865
44. Besselink GA, Kooyman RP, van Os PJ, Engbers GH, Schasfoort RB (2004) Signal amplifi-
cation on planar and gel-type sensor surfaces in surface plasmon resonance-based detection of
prostate-specific antigen. Anal Biochem 333:165
45. Chung JW, Bernhardt R, Pyun JC (2006) Additive assay of cancer marker CA 19-9 by SPR
biosensor. Sens Actuators B 118:28
46. Li Y, Lee HJ, Corn RM (2007) Detection of protein biomarkers using RNA aptamer
microarrays and enzymatically amplified surface plasmon resonance imaging. Anal Chem
79:1082
47. Regnault V, Boehlen F, Ozsahin H, Wahl D, de Groot PG, Lecompte T, de Moerloose P
(2005) Anti-protein S antibodies following a varicella infection: detection, characterization
and influence on thrombin generation. J Thromb Haemostasis 3:1243
48. Rojo N, Ercilla G, Haro I (2003) GB virus C (GBV-C)/Hepatitis G virus (HGV): Towards the
design of synthetic peptides-based biosensors for immunodiagnosis of GBV-C/HGV infection.
Curr Protein Pept Sci 4:291
49. Ladd J, Boozer C, Yu Q, Chen S, Homola J, Jiang S (2004) DNA-directed protein immobiliza-
tion on mixed self-assembled monolayers via a streptavidin bridge. Langmuir 20:8090
50. Miyashita M, Shimada T, Miyagawa H, Akamatsu M (2005) Surface plasmon resonance-based
immunoassay for 17beta-estradiol and its application to the measurement of estrogen receptor-
binding activity. Anal Bioanal Chem 381:667
51. Gobi KV, Iwasaka H, Miura N (2007) Self-assembled PEG monolayer based SPR immunosen-
sor for label-free detection of insulin. Biosens Bioelectron 22:1382
Chapter 4
Fractal Plasmonic Nanoantennae
Ravi Sadananda Hegde
Abstract The ability to precisely tailor lineshapes, operational bandwidth and

localized electromagnetic field enhancements (“hot spots”) in nanostructures is
currently of interest in advancing the performance of plasmonics based chemical
and biological sensing techniques as well as in plasmonics based energy harvesting
applications. Fractal geometries are an intriguing alternative in the design of
plasmonic nanostructures as they offer tunable multi-band response spanning the
visible and infrared spectral regions. This chapter reviews the recent developments
concerning the incorporation of fractal geometries into plasmonic nanostructures.
The scope is restricted to the review of fractal shaped antenna elements as opposed
to fractal based array placement methods. Beginning with a brief overview of
fractals and fractal based radio-frequency antenna engineering, the review focuses
on two canonical geometries: the Sierpinski carpet and the fractal tree. Fractal
geometries are promising for improving the performance of plasmonics based
optical applications like ultrasensing and energy harvesting.
Keywords Fractals • Nanoantenna • Surface-enhanced spectroscopy • Nanos-

tructures
4.1 Introduction
In the recent years, plasmonic nanostructures have excited interest in regards to

various applications like sensing, imaging and energy harvesting. The excitation of
localized surface plasmon resonances [1] leads to substantially enhanced and highly
localized electromagnetic fields [2]. Nanophotonic structures employing plasmonic
resonators are particularly important for chemical and biological sensing [3] and
various forms of surface enhanced spectroscopies such as Surface Enhanced Raman
Spectroscopy (SERS) [4, 5], Surface Enhanced Infrared Absorption Spectroscopy
(SEIRA) [6] and surface enhanced nonlinear harmonic generation [7]. In designing
a suitable plasmonic nanostructure for a particular application, the geometry and
R.S. Hegde ()

Indian Institute of Technology Gandhinagar, 382355, Palaj, Gandhinagar, India
e-mail: hegder@iitgn.ac.in; http://www.iitgn.ac.in/faculty/electrical/ravi.htm

DOI 10.1007/978-3-319-48081-7_4
56 R.S. Hegde
the dielectric environment are chosen to obtain the desired optical response in
the form of precisely tailored lineshapes [8–11] and controlled confinement of
electromagnetic energy (“hot-spots”) [12].
The wide variety of reported plasmonic geometries are fabricated [13] using
either lithography (periodic arrays of single nanoparticles or tightly coupled
nanoparticles [14], nanoapertures [15]) or chemical synthesis [12]. Using simple
nanoresonators does not lend enough flexibility to alter the frequency response
(either to broaden or narrow the response). There has been a lot of recent
interest towards compound plasmonic resonators [14, 16] that allow more design
flexibility. Additionally, nanoantenna arrays have mostly used periodic lattices.
Fractal geometries occupy the intermediate region between highly ordered periodic
arrays and completely random rough surfaces. The possibility of employing fractal
geometries [17] is of current interest as we design more and more complicated
plasmonic nanostructures.
The term fractal was coined by mathematician Benoit Mandelbrot [18] during
the 1970s when he was studying various kinds of geometrical patterns that are
found in nature. In contrast to smooth contours, typically found in Euclidian
objects like the sphere, disk and such, natural features like coastlines exhibit
jagged features. Mandelbrot coined the term fractal by basing it on the Latin
fractus meaning broken or fractured. Fractal and fractal-like geometries have a
long history in radio frequency antenna engineering [19, 20] where they have been
explored for their attractive properties like antenna size reduction and operational
bandwidth enhancement. Self-similarity is one of the key features of interest to
antenna designers; this implies that the fractal shape has multiple length scales
and the features at different length scales exhibit a degree of similarity. Unlike
mathematically idealized fractals that can exhibit self-similarity indefinitely; real
life objects are approximate fractals and exhibit self-similarity over finite scale
ranges. Figure 4.1 shows two natural objects that are readily seen to exhibit some
self-similiarity.
Fig. 4.1 Fractal like forms found in nature. (a) Romanesco broccoli edible flower clearly showing
self-similar repeating forms that is characteristic of fractals (Public domain image, courtesy of Jon
Sullivan) (b) Self-similarity observable in frost crystals formed on cold glass. This shape is the
inspiration for tree fractals like the Cayley tree (Image courtesy Wikimedia Commons, reproduced
under the Creative Commons License)
4 Fractal Plasmonic Nanoantennae 57
This chapter is motivated by the fact that a deeper understanding of the optical
properties of plasmonic nanostructures that are inspired from fractal geometries
may pave way for improved performance in engineered plasmonic nanostructures.
It reviews the recent developments in regards to fractal geometry inspired plasmonic
nanostructures. The chapter begins with a brief introduction to fractals (Sect. 4.1.1)
and fractal antennas (Sect. 4.1.1.1) to orient the reader. The Sierpinski triangle and
carpet geometries were the earliest explored in connection to plasmonics and these
geometries are explored in Sect. 4.2.1 and subsequently followed by a discussion on
fractal tree geometries in Sect. 4.2.2. The chapter concludes in Sect. 4.3.
4.1.1 Fractals
Rather than pursuing rigorous mathematical definitions, it is more helpful for

our purpose to understand the characteristics of the fractal geometry. Although,
real antenna geometries are only approximately self-similar, the concept of self-
similarity is nonetheless a useful construct in understanding their electromagnetic
behavior [21, 22]. Mathematical treatments generally concern with objects that
have infinite detail. According to Falconer [23], such “infinite” fractal geometries
generally exhibit the following characteristics:
• Self Similarity which implies that parts of the shape repeats at smaller and
smaller length scales.
• Detailed structure at arbitrarily small scales. Although, there is an increasing
amount of detail, the description of this proves to be simple owing to the self-
similarity.
• The geometry can be generated through an algorithmic iterative process. Succes-
sive recursive steps give closer approximations to the fractal.
• The geometry is not easily describable in classical terms using constructs of
Euclidean geometry.
• Its fractal dimension is different (typically larger) than its topological dimension.
Figure 4.2 shows the first four successive recursive stages of some canonical
fractals that are very relevant to antenna engineering. These stages successively
begin to approximate the final fractal. For instance, consider the Koch snowflake
(see Fig. 4.2b where the green colored lines serve to illustrate the construction pro-
cess) It can be constructed by starting with an equilateral triangle, then recursively
altering each line segment as below:
1. divide the line segment into three segments of equal length.
2. draw an equilateral triangle that has the middle segment from step 1 as its base
and points outward.
3. remove the line segment that is the base of the triangle from step 2.
The Koch snowflake is in fact the limiting case when the above steps or stages
are repeated infinitely. It is seen that the successive stages begin to reveal the criteria
58 R.S. Hegde
Fig. 4.2 Various iteration stages for some common fractals. Three common fractals that are most
relevant to antenna engineering are considered: (a) Sierpinski Triangle, (b) Koch snowflake and (c)
H-fractal. The typical procedure using iteration and motifs for the generation of various successive
stages for these fractals is also shown
described above like self-similiarity, fine detail and such. Consider the fractal
dimension of the Koch snowflake. It has a topological dimension DT of 1, being
a curve. However, the curve is not rectifiable as the length of the curve between
any two points is infinite. The Koch curve will be continuous everywhere but
differentiable nowhere. It has the fractal Hausdorff dimension given by log 4= log 3
or D 1:26186 (which is larger than its topological dimension D). Intuitively, the
Hausdorff dimension conveys the fact that although it is a curve, it is much more
detailed than a Euclidian curve, albeit, less detailed than a full planar object. Thus
its fractal dimension lies in between the topological dimensions of a line and that of
a plane.
4.1.1.1 Radio Frequency Fractal Antenna
A fractal antenna is an antenna that employs a fractal, self-similar design by

repeating a particular motif over several different length scales [17]. With increas-
ingly complicated requirement for cellular and wireless communications, fractal
antennas are of interest for antenna designers. Specifically, they are of interest
due to their ability [20] to provide: Compact size (up to factors of 2 to 4 without
performance degradation [17]); Low profile and conformal designs; and, multi-band
and broadband operational capability. The unique ability of the fractal antenna is
that it is capable of operating with good-to-excellent performance at many different
frequencies simultaneously. Such frequencies are not harmonically related and can
be designed at will by tuning the geometrical parameters of the fractal. The fractal
curves typically have longer electrical lengths for the same given footprint and are
thus able to miniaturize the antenna up to two to four times without noticeable
performance degradation. Additionally, the fractal nature of the antenna shrinks its
size, without the use of any additional components, such as inductors or capacitors.
Antenna systems can use either single radiating elements or an array of radiating
elements. There are two possible ways [20] of incorporating fractal geometries
in antenna systems: in the individual radiating elements themselves and having a
regular grid for the array; or, having simpler radiating elements but arranging them
in a irregular fractal inspired grid. In 1999 it was reported that self-similarity was
one of the underlying requirements to make antennas frequency and bandwidth
invariant [24, 25] with origin symmetry being another condition for frequency
independent operation. The so-called angle-defined antennas (e.g. log-periodic
antenna [26]) indeed are self-similar. This principle is now known as the HCR
principle after its proponents [24].
4.2 Fractal Nanostructures
As discussed earlier, there are two possible ways for incorporating fractal geome-
tries into plasmonic nanostructures and metasurfaces: either in the design of the
array elements themselves or in the formation of irregular arrays. Both of these types
have been reported in plasmonic nanostructures. Most of the fractals of interest to us
can be constructed through the so called subdivision rules or the iterative function
systems.
We will restrict our scope in this chapter to two canonical geometries: Sierpinski
triangle; and, fractal tree and their analogues and derivatives. Apart from these
canonical shapes, several workers have reported nanoantennae based on the log-
periodic antenna. The log-periodic antenna shares certain similarities with fractal
antennas, but can also be considered from the perspective of angle defined antennas
(See for example [24] for a detailed unifying framework). We will only briefly
discuss log-periodic geometries.
4.2.1 Sierpinski Triangle and Carpet
Most of the reported fractal plasmonic nanostructures till date have been based on
the Sierpinski carpet and Sierpinski triangle geometries. In 2010, Hsu and cowork-
ers [27] first reported the fabrication of a Sierpinski carpet plasmonic substrate and
used it to unravel the electromagnetic field dependent SERS enhancement effects.
60 R.S. Hegde
The multiscale nature of this geometry provided them with a unique platform
to understand the influence of geometrical effects on plasmon modes and thus
ultimately on SERS enhancement factors.
Volpe and coworkers [28] were the first to propose the use of engineered fractal
geometries as a general design tool for various kinds of plasmonic nanostructures.
They reported on a numerical study of the Sierpinski carpet geometry and explored
broadband spectral response and other properties like sub-diffraction focusing. In
contrast to the hot-spots obtained by plasmonic dimer geometries, the hot-spots
obtained via the carpet geometry was claimed to occur at an elevation away from the
metal surface with concomitant advantages in engineering light-matter interaction.
Rosa and co-authors [29] proposed using the plasmonic Sierpinski gasket based
replacement to the standard plasmonic bowtie and reported the lithography based
fabrication of the same. Using numerical simulations, they demonstrated that such
structures enable light field enhancement from visible to far-IR spectral range and
possibly scalable towards the Terahertz band. They claimed that these antennas
are attractive for IR and THz filtering, detection, and emission applications.
Independently, and near concurrently, Sederberg and coworkers [30] also proposed
to replace the solid bowtie elements with Sierpinski fractal features and proposed
its use for sensing molecular vibration modes in the near- to mid-infrared. Using
numerical simulations, it was shown that these antennas offer a compact device
footprint and an enhanced confinement factor compared to a bowtie antenna as
well as broadband operation from 700 nm to 3:4 µm. In [22], they investigated
the influence of a fractal nanostructure’s Hausdorff dimension on its broadband
spectral response. The proposed antennas with Pascal’s triangle geometry allowed
the resonance conditions to be varied while the antenna dimensions were kept
constant.
Chen and coworkers [31] reported the fabrication, characterization, and simula-
tion results for an optical antenna inspired by the Sierpinski carpet fractal geometry.
Measurements and simulations of the far-field scattering efficiency confirmed the
expected broadband optical response. Spatial maps of the near-field intensities
obtained through two-photon photoluminescence (TPPL) were reported as shown
in Fig. 4.3. The spatial maps clearly show that the fractal geometry is outperforming
the regular array and the fully random case. The authors reported up to 50 % larger
field enhancement in comparison to other two structures at the wavelength of 800 nm
and speculate that this could be arising due to the presence of long range spatial
order in the fractal case.
Cakmakyapan and coworkers reported on the experimental study of the Sierpin-
ski carpet based bowtie geometry [32] and are based on earlier simulation studies
by the same group [22, 30]. Figure 4.4 shows the fabricated structures where it is
seen that keeping the same footprint leads to a redshift for increasing fractal stages.
This is the familiar miniaturization property of fractal antenna discussed earlier
in connection with RF fractal antenna. SERS measurements show clearly that the
fractal shape leads to a higher field enhancement for successive stages. These results
are promising for developing SERS substrate with quantifiable enhancement.
Fig. 4.3 SEM and TPPL images of (a) Sierpinski carpet, (b) pseudo-random and (c) periodic gold
nanostructures at excitation wavelengths of 770, 780, 790, 800 nm. The double arrow indicates
the direction of polarization of the excitation beam. The color bar indicates the number of photon
counts and is indicative of the local field enhancement (Reproduced with permission [31] under
the Creative Commons Attribution 3.0 licence)
62 R.S. Hegde
Fig. 4.4 Schematics and SEM micrograph images of (a) Bowtie, (b) Fractal-1 that replaces the
bowtie with a stage 1 Sierpinski triangle, and (c) Fractal-2 that does so with a stage 2 triangle.
For (a), (b) and (c), g D 65 nm, r D 400 nm, y D 420 nm and the scale bar is 100 nm. (d)
Experimental transmission spectra of the antenna arrays for the various shapes (a), (b) and (c)
showing the red-shifting effect. (e) shows the SERS measurement results for Bowtie, Fractal-1 and
Fractal-2 geometries showing increased enhancement for successive fractal stages (Reproduced
with permission [32])
The Sierpinski carpet geometry in the bowtie form was reported to enhance
nanowire based photovoltaic power conversion efficiency [33] by Abdellatif and
coworkers. In their numerical simulations, the Sierpinski geometry based nanowire
performed the best providing about 24 % enhancement in power conversion effi-
ciency over the same nanowire without the antenna. However, the placement of the
antenna adds to the total footprint and thus the overall conversion efficiency per
unit area of the solar capturing surface is likely reduced as it leads to a reduced
packing density of the nanowires. Zhu and coworkers [34] reported quantum
efficiency enhancement in an ultra-thin silicon solar cell coated by a fractal-like
pattern of silver nano cuboids. However, the actual enhancement is seen only in
the overall optical absorption per unit area of the device. It is unclear how much
of this increased optical absorption actually translates into increased charge carrier
generation. The increase in optical absorption occurs as multiple antireflection

bands are achieved due to the self-similarity in the fractal-like structure. Plasmonic
Sierpinski nanocarpet as back structure [35] for a thin film Si solar cell has
been investigated and has been reported to lead to wideband optical absorption.
Additionally, adding a Sierpinski carpet like front electrode is also claimed to exhibit
a wide angle coverage in a polarization independent fashion in addition to being
wideband [36].
4.2.2 Fractal Tree Geometries
The fractal tree geometry is the wire equivalent to the Sierpinski triangle. H-fractal
based nano-apertured substrates have been explored by L. Zhou and coworkers [37].
It was shown that a metallic plate with periodic fractal-shaped slits can be
homogenized as a plasmonic metamaterial with plasmon frequency dictated by
the fractal geometry. This structure could focus light sources with all-dimensional
subwavelength resolution and enhanced field strengths as was demonstrated with
microwave experiments. Li and coworkers [38] have reported numerical studies on
a split complimentary version of this structure and show that it can be useful for
broadband strong field enhancement.
Branching self-similar structures like the Cayley tree [13] have been reported to
exhibit a broadband spectral response and “hot-spots” that are active over multiple
wavelengths making them attractive for engineering plasmonic substrates. The
effect of fractal order on the optical properties can be seen in Fig. 4.5, which shows
the SEM micrographs of three sequential stages and the corresponding experimental
and simulated transmission curves. The footprint of the nanoantenna is allowed
to increase in each successive stages and thus a red-shifting of the resonances is
expected. The noteworthy feature is the emergence of new resonance features in the
spectrum and the broad extent in which they span. Gottheim and coworkers [13]
performed an extensive analysis of the charge and near field enhancement of the
various fractal orders of the Cayley tree and they have catalogued the charge
distributions responsible for the various modes as seen in Fig. 4.6. Particularly
noteworthy is that the highest energy modes in any particular order involves charge
localization at the far corners of the structure while the lower energy modes assume
a more delocalized charge distribution. It is to be noted that for a Nth order tree,
there are N unique spectral signatures.
4.2.2.1 Ternary Tree Fractal
A systematic numerical study of the ternary tree [19] (three branched tree) fractal
plasmonic nanoantenna was reported by Hegde [39]. The ternary tree is a nanowire
equivalent of the Sierpinski carpet geometry [30]. Although the ternary tree is a
regular fractal, it nonetheless serves as a representative model for general stochastic
64 R.S. Hegde
Fig. 4.5 Scanning electron micrograph (SEM) images of 50 by 50 m sized 2D arrays of (a) first-,
(b) second-, and (c) third-order Cayley trees with the individual structures shown as insets. The
periodicity of the first-order Cayley tree (a) is 500 nm in both directions, while the (b) second-
and (c) third-order Cayley trees have 1 m periodicity. The array scale bars are 2 m; inset scale
bars are 300 nm. (d–f) Experimental Fourier transform infrared transmission spectra (solid lines)
of the first-, second-, and third-order Cayley tree structures under nonpolarized light excitation
with corresponding finite element method simulated transmission spectra (dashed lines) (Figure
copyright of American Chemical Society and reproduced with permission from [13])
fractals like the DLA fractal [40]. While possessing the same three-fold rotational
symmetry, the differences in branching angles in the ternary tree lead to a much
different behavior compared to the Cayley tree fractal [13]. The bounding area of a
ternary tree of a certain fractal order is smaller than that of a Cayley tree of the same
order and primary length and width. For the same fractal order and wire width, the
ternary tree fills the space more densely compared to a Cayley tree. This implies
that a higher hot-spot density can be achieved in the ternary tree geometry which is
of interest for designing SERS substrates.
The ternary tree follows an iterative process [40] in its construction beginning
with a bar (nano rod) in the zeroth iteration. The first design iteration places bars
of length L1 in such a way that an angle of 120ı is formed between adjacent arms
and creates a three branched tree-like structure such that the end points lie along
the vertices of an equilateral triangle (of side s D 1:5 L1 ); this triangle will be
referred to as the bounding triangle. The three bar structure forms the motif that is
scaled and translated to build the successive fractal orders. At the second iteration,
the motif with bars of length L2 is added to each of the 3 outer vertices. The process
continues by always adding the motif to the outermost vertices. By maintaining a
Fig. 4.6 Schematic of charge distribution leading to the formation of the multipeak plasmonic
spectra in the Cayley-tree geometry. In the more spatially extended plasmon modes, the charge
oscillates between the complex branches as shown with dashed ovals in the lower right corner
of the figure (Figure copyright of American Chemical Society and reproduced with permission
from [13])
constant ratio of lengths between successive iterations, a fully self-similar structure

results. The ratio LN =LN1 < 0:5 to ensure that the bars do not intersect at any stage.
The width of the arms at each stage can be kept constant or scaled proportionately
in successive iterations as well.
In their study, Hegde et al [39] deviate from the ideal fractal structure by making
certain changes to the ideal ternary tree structure so that it becomes easier to
fabricate. Firstly, since the size of the bounding equilateral triangle for any iteration
stage keeps growing successively, as seen in the previous discussion on the Cayley
tree, the resonances are red-shifted at each stage. In the interest of preserving activity
in the visible part of the spectrum, at each iteration the lengths were scaled in such
a way that the bounding equilateral triangle remained the same across iterations,
allowing better comparison between stages. Secondly, only two arm widths are used
throughout the structure such that W1 is the width of the innermost arms and W2 that
of all other arms. While the modified structure chosen was not strictly self-similar,
it retained the three fold symmetry properties of its ideal counterpart.
The first, second and third iterations of the modified ternary tree fractal plasmonic
nanoantenna and the corresponding absorption, scattering and extinction cross-
sections are shown in Fig. 4.7. The extinction efficiency ext D
ext =A is expressed
66 R.S. Hegde
Wavelength (um) Bounding

30.0 3.0 1.5 1.0 0.75 0.6
Triangle
15
Extinction (X pol)
Extinction (Y pol)
Scattering
Absorption s W1
L1
Normalized Cross sections
10
L1
s
L2
W2
5
L1
W3
s
L3
L2
0
Y
0 100 200 300 400 500
Frequency (THz) X
Fig. 4.7 Spectral variation of the extinction, absorption and scattering cross sections for the first
(top), second (middle) and third (bottom) iteration of the modified ternary tree plasmonic fractal
structure. The top-view (x y plane) of resulting tree structure at each iteration is shown at the
right of the corresponding spectra. The structure is excited with linearly polarized plane waves
propagating in the z direction with the electric field vector lying in the x y plane. The cross
sections have been normalized with respect to the area of the bounding equilateral triangle (see the
dotted lines that lie on the periphery of the tree structures with a side s). The extinction spectra
are shown when the exciting plane wave is polarized along x and y, whereas only the polarization
averaged values are shown for the absorption and the scattering spectra. Each set of spectra for
a particular iteration have been offset for clarity (first iteration spectra offset by 10 and second
iteration by 5). The bounding triangle length s, the lengths of the arms L1 ; L2 and L3 , and the widths
of the arms W1 ; W2 and W3 remain the same for all the iterations. For the set of spectra shown in
the figure, L1 D 300 nm, L2 D 0:45L1 and L3 D 0:15L1 and W1 D 40 nm, W2 D 0:5W1 and
W3 D W2 and the thickness of the structure is 50 nm. Gold [13] is used for the plasmonic material
and is assumed to be surrounded by a fictitious dielectric medium with refractive index value of
0:5 .1:0 C 1:45/, which is the average value of vacuum and a silica substrate (Reproduced with
permission from [39])
as the ratio of the extinction cross-section

ext to the surface area A of the resonator.
It is more convenient to consider instead 0ext D
ext =Ab , where Ab is the area of
the bounding triangle as it facilitates easier comparison between different fractal
orders. Although the fractal structure fills approximately a third of the area of the
bounding equilateral triangle (for the first iteration, the value of A=Ab is 30.8 %),
the various cross-sections are up to 3 to 5 times the area of the bounding triangle.
The enhanced cross sections are typical of plasmonic nanoparticles [3], the fractal
nanoantenna exhibit an increased ext in relation to the nanodisk and the nanorod
of the same bounding dimension [13]. The most striking property of the spectra is
the near complete polarization independence despite the structure possessing only
a three-fold symmetry. The small observable differences can be attributable to the

numerical errors arising due to the meshing in the finite element calculation.
The structure of the zeroth iteration (plasmonic nanorod) and the first iteration
(three branched tree) is somewhat different than the second and subsequent
iterations. Self-similarity (the characteristic signature of fractal geometry) in the
extinction spectra, begins to emerge only after the second iteration as seen in
Fig. 4.7. The extinction spectrum is characterized by a broad lowest frequency peak
followed by peaks that appear at higher frequencies. Unlike, the lowest frequency
peak, the peaks appearing at higher frequencies are not single peaks but are seen
to be two closely spaced peaks ( for example in stage 2, two closely spaced peaks
at 210 and 250 THz can be seen). In addition to the dominant peaks, some smaller
peaks are also seen (for example in stage 2,385 and 473 THz). At any iteration stage
N 1 the number of peaks P follows the rule:
P D 1 C 2 .N 1/ C s.N/; (4.1)
where 1 C 2 .N 1/ denote the dipolar modes associated with the largest

radiation coupling strength. Other smaller peaks s.N/ observable in the spectra can
be attributable to the higher order modes which exhibit a characteristic reduction in
the scattering cross-section in comparison to the dipolar modes.
The geometrical self-similarity leads to the spectral self-similarity. Plasmon
oscillations of different lengths create similar spectral features at different frequen-
cies. A qualitative understanding of the physical origin of the many peaks in the
spectra for any particular fractal order can be obtained by considering the possible
alternatives for dipole like charge distributions as has been previously discussed
for the Cayley tree. In Fig. 4.8, a schematic of the distribution of the charges is
seen for the first and the second fractal orders; these distributions assume that the
induced charges are primarily concentrated on the outer tips of any structure (and
consequently, the amount of induced charges in the internal parts of the structure
Fig. 4.8 Schematic describing the charge configuration in the plasmonic modes of a ternary fractal
tree nanoantenna for the first and the second iteration. In the first stage, (i) and (ii), two possibilities
with the charges localized to the corners exist. For the second stage, (iii) shows the lowest order
dipole mode where the charges are no longer localized over the corners but are instead spread out.
(iv) and (v) show the higher frequency modes involving primarily the outer arms (Reproduced with
permission from [39])
68 R.S. Hegde
is small) [13]. The zeroth iteration dipole mode is the simplest with only one
degeneracy. In the first stage, two alternatives are possible as seen in (i) and (ii);
in the absence of symmetry breaking the resulting modes are degenerate.
The assumption that charge concentration occurs primarily on the tips is not
generally true in higher order modes. A sharply localized charge distribution leads
to a narrower linewidth and a smeared-out charge distribution of induced charges
leads to a broader linewidth and redshift in the center wavelength. Distribution (iii)
leads to a mode that is similar to that of (ii); however, the induced charges are not
sharply localized. In the limit of small L2 =L1 , the lowest order mode will have a
distribution (iii) that is similar to (i) and the resulting linewidth of the mode is also
similar. For increasingly larger values of L2 =L1 a redshift in the center wavelength
and significant linewidth broadening is expected due to the wider spread of the
induced charge distribution. The higher frequency modes of the second stage are
primarily localized to the outer arms and have charge distributions approximately
similar to (i) and (ii). Although, localization is an approximation to the true picture
where the charge density becomes smeared out; the peaks of the smeared out charge
distributions are expected to follow the depiction in (iv) and (v). The more important
point to note about distributions (iv) and (v) is that the resulting modes are no longer
degenerate. The closely located twin-peak spectral feature seen in the second and
third fractal orders (Fig. 4.7) results from this degeneracy lifting.
Further insight into the nature of the plasmon resonances can be obtained with
the maps of the electric field enhancement in the near-field of the structures at
the peak frequencies. The ternary tree has a three fold symmetry leading to mode
degeneracies and thus the field maps are not simply the eigenmodes but rather a
spatially superimposed version [41]. Figure 4.9 shows the near field enhancement
(i.e. jE=Einc ) for various peak frequencies for the first, second and third iterations
(due to symmetry, only one half of the structure is shown) for two orthogonally
polarized electric field excitations. The lowest frequency mode of all the fractal
orders is considered first. Figure 4.9 (a1) and (a2) are seen to correlate with
the distributions (i) and (ii) seen in Fig. 4.8 respectively. Figure 4.9 (b1) and
(b2) correspond to distribution (iii) in Fig. 4.8 where it is noticed that the charge
distribution is not localized to the tips, but instead is spread out. The spreading leads
to a red-shift in the center frequency and significant broadening of the linewidth
even though the bounding triangle is the same for the first and the second orders.
Although, the spreading is more prominent in the lowest frequency mode of the third
order fractal, (Fig. 4.9 (d1) and (d2)), as evident in the extinction spectra (Fig. 4.7),
the overall intensity enhancement is significantly weaker. The higher frequency
mode of the second stage seen in Fig. 4.9 (c1) and (c2) seems to primarily be
localized to the outer arms and correspond to distributions (iv) and (v) in Fig. 4.8. It
is noted that in the third iteration, the hot-spots are primarily formed at the outermost
tips and are active over multiple frequency bands similar to the hot-spots in the
Cayley tree fractal [13].
(a1) 100 THz (b1) 45 THz (c1) 200 THz
(a2) 100 THz (b2) 45 THz (c2) 200 THz
(d1) 55 THz (e1) 185 THz (f1) 400 THz
(d2) 55 THz (e2) 185 THz (f2) 400 THz

Y
Electric field enhancement E/E inc
X
0 12.5 25
Fig. 4.9 Electric field enhancement in the near field of the first, second and third iterations of the
modified ternary tree plasmonic fractal structure at the various extinction peaks. The geometrical
parameters are the same as in Fig. 4.7. Subfigures (a) denotes the first iteration, (b) and (c) the
second iteration and (d), (e) and (f) the third iteration. The structure is excited with linearly
polarized plane waves propagating in the z direction with the electric field vector lying in the
x y plane. The electric field is polarized along the x axis in subfigures (a1), (b1) : : : (f1) and
along y axis for subfigures (a2), (b2) : : : (f2) (Reproduced with permission from [39])
The tunability of the plasmon resonances in the second and third iterations
by scaling the arm lengths while preserving the three-fold rotational symmetry
of the structure is shown in Fig. 4.10. For the second stage (Fig. 4.10a), the
spectral features show an increasing red-shift for L1 D 300 nm in comparison to
L1 D 250 nm while keeping L2 =L1 constant as expected and is consistent with
the red-shifts seen in plasmonic nanorods [10] and nanodisks [41]. For a given L1 ,
increasing L2 =L1 also leads to red-shifting of the peaks and is the result of spatial
70 R.S. Hegde
a Wavelength (um) b Wavelength (um)

30.0 3.0 1.5 1.0 0.75 0.6 30.0 3.0 1.5 1.0 0.75 0.6
32
24
L2/L1
L1(nm)
0.45
28
300
20
0.40
250

24
16
L2/L1
20
0.48 L3/L1
16
12
0.47
0.15
0.46
0.14
0.45
12
0.13
0.43
8
0.12
0.42
0.11
8
0.41
0.10
0.40
4
4
0
0 100 200 300 400 500 0 0 100 200 300 400 500
Frequency (THz) Frequency (THz)
Fig. 4.10 Plasmon tunability for the second and third iteration ternary tree fractal nanoantenna
for changes in the arm lengths and arm length ratios. The symmetry in the structure is preserved
by changing the arm lengths uniformly. The extinction cross sections are normalized with respect
to the area of the bounding equilateral triangle (as in Fig. 4.7) and are vertically offset for clarity.
The excitation conditions are similar to those described in Figs. 4.7 and 4.9 with the electric field
polarized along x. The tunability of the second iteration is shown in (a) where the cross sections are
plotted for different values of L2 =L1 while keeping the length L1 constant (the solid lines represent
L1 D 300 nm and the dashed lines L1 D 250 nm). The tunability of the third iteration is shown
in (b) where the cross sections are plotted for different values of L3 =L1 for a length L1 of 300 nm
and a constant L2 =L1 (the solid lines represent L2 =L1 D 0:45 and the dashed lines L2 =L1 D 0:40)
(Reproduced with permission from [39])
spreading in the induced charge distribution; the spreading increases continuously

as L2 =L1 approaches 0.5. For a given L1 as L2 increases, the higher frequency
peaks are also red-shifted and experience linewidth broadening. Additionally, the
influence of the asymmetry becomes more prominent; this leads to the observed
increasing separation between the peaks. For the third iteration (Fig. 4.10b), the
scaling behavior is not clearly observed for the lowest frequency mode. We note
that for the sizes used in this simulation, retardation effects are likely to occur and
that the shifts will not vary linearly in relation to the length scaling for the lowest
frequency mode [2]. The absence of a clear scaling behavior is also evident for the
peaks in the region between 180 and 250 THz. A clear red-shifting trend is observed
for the high frequency modes near the 500 THz region. However, near field coupling
effects [42, 43] are likely to be dominant for larger values of L3 =L1 in this region.
By a proper choice of the various arm lengths and arm widths of various stages, the
individual resonances can be independently tuned allowing for a precise tailoring of
the complete spectrum.
The three-fold rotational symmetry has been preserved in the results presented
thus far. The study of the influence of structural asymmetry on the spectra is
important because fabrication tolerances and the presence of substrate cannot
be avoided in practice. The influence of structural asymmetries on the resulting
lineshape also allows extrapolation to the case of stochastic fractals. Symmetry
breaking can be introduced in a number of ways: by altering the lengths, length
ratios and the angle between the arms; by the addition of a substrate [44–46]; and,
by altering the parameters in one branch with respect to another.
In Fig. 4.11a, the outer arms are all of the same dimension and only one inner
arm length is varied; in Fig. 4.11b, all the inner arms are of different lengths in
addition to having different lengths for the outer arms (the three fold symmetry is
not broken for the outer arms, however). As expected, the spectra begin to show
a dependence on incident wave polarization as the length L begins to change in
relation to L1 in Fig. 4.11a. The lowest frequency resonance, however, does not
redshift much when the length L is changed. A similarly unexpected behavior
is observed for the lowest frequency resonance in Fig. 4.11b. Strong polarization
dependence is observed for the dipolar mode in plasmonic nanorods [47, 48]. As
discussed previously the charge distribution in the lowest frequency mode is not
well localized but spread out; the mode thus acquires the isotropic character like
the mode in a nanodisk [41]. As the structure is somewhat large, retardation effects
are also another likely effect accounting for the smaller observed shifts in peak
frequency. The higher frequency modes on the other hand are more localized in
comparison and exhibit strong polarization dependence as seen in Fig. 4.11b. It is
expected that in the third and subsequent iteration stages, only the highest frequency
modes will exhibit polarization dependence in spite of length related asymmetries.
It can be speculated that even in the presence of large structural asymmetries, the
response of the fractal tree is nearly isotropic for lower frequency modes and that
stronger polarization dependence is seen as the mode frequency is increased. In
Fig. 4.12, introducing the substrate and nearest neighbour coupling effects result in
red-shifting of the lowest frequency mode when the distance between neighbouring
elements is decreased. The presence of the substrate can lead to lifting of the
degeneracy in modes [44] and is seen in the characteristic polarization dependencies
in the higher frequency regions. The lowest frequency modes are more robust to the
influence of the substrate in comparison to higher frequency modes.
4.3 Conclusion
Exploitation of fractal geometries for the design of plasmonic nanostructures offers

a large number of degrees of freedom and will permit the tailoring of complex
plasmonic lineshapes optimized for any particular application. Self-similar spectral
features are observed in the spectra in comparing structures at different iteration
72 R.S. Hegde
(a) L1 (b)
250
75
200
Y
L2 X 100
145
L
105 θ
Wavelength (um) Wavelength (um)

30.0 3.0 1.5 1.0 0.85 30.0 3.0 1.5 1.0 0.85
35
12 16 20 24 28 32 36 40 44 48 52
L (nm)
θ
30
400
−60
375
−50

350
25
−40
325
−30
300
−20
20
275
−10
250
0
15
+10
+20
+30
10
Pol +40
+50
X
8
+60
5
Y
4
0
0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350
(c) (d)
Fig. 4.11 Polarization dependence in the extinction spectra resulting from breaking the arm length
symmetry in the branches of the second iteration ternary tree fractal plasmonic nanoantenna.
Similar to the case shown in Fig. 4.7 a linearly polarized z traveling plane wave with its polarization
vector lying in the x y plane is considered. The dielectric environment is similar to Fig. 4.10 but
the symmetry is broken by introducing inter-branch arm-length differences. In (c), the spectral
variation of the normalized extinction cross-sections (similar to Fig. 4.7) for the structure (a) are
shown for x and y polarized light for different lengths L. In (a), the lengths L1 and L2 are kept
constant (L1 D 300 nm and L2 =L1 D 0:4). In (d), the spectral variation of the cross-sections for
the structure (b) is shown for different inclinations of the polarization vector with respect to the x
axis. The structure (b) has different inner arm lengths as well as the outer arm lengths; however,
the arm lengths of the outer arms for a particular branch are the same. In (d), red arrowed lines are
shown for some cases denoting the polarization vector inclination. In (c) and (d), the curves are
offset for clarity and are normalized with respect to the area of the bounding equilateral triangle of
a symmetrical tree with L1 D L1 D 300 nm (Reproduced with permission from [39])
(a) (b) gx
Incidence
gy
x
z T x
y
Wavelength (um) Wavelength (um)

30.0 3.0 1.5 1.0 0.85 30.0 3.0 1.5 1.0 0.85
(c) g_y = 150nm, g_x = 50nm

g_y = 150nm, g_x = 150nm
(d) g_x = 150nm, g_y = 50nm
g_x = 150nm, g_y = 100nm
2.0
2.0
Transmittance
Transmittance
1.5
1.5
1.0
1.0
0.5
0.5
0.0
0.0
0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350
Fig. 4.12 Anisotropy resulting from substrate and neighbour element coupling related symmetry
breaking for the second iteration ternary tree fractal plasmonic nanoantenna. A periodic array of
gold nanoantenna arranged in a rectangular lattice on a silica substrate is considered with a plane
wave excitation from the air as seen in (a). In (b), the bounding triangle of each nanoantenna is
denoted by red dotted lines and the spacing of the bounding triangles is quantified with gx and gy
as shown. The transmittance spectra is shown for a set of curves with different values of gx and gy
in (c) and (d). In (c), gy is kept constant at 150 nm, while in (d), gx is kept constant at 150 nm. For
all the curves L1 D 300 nm and L2 =L1 D 0:45. The direction of polarization of the electric field of
the exciting wave is shown in red and curves are offset for clarity. The transmission dips seen here
coincide with the extinction peaks seen in Fig. 4.7 and the sharp spectral features seen at shorter
wavelengths results from diffractive effects (Reproduced with permission from [39])
stages resulting in multiple resonant effects spanning a wide spectral gamut from
the visible to the mid infrared region. It is difficult [13] to obtain such features in
simpler configurations such as nanorods or coupled nanorods, as we have to rely
on the excitation of higher order modes. The hot-spot features in the third order
fractal tree are distributed nearly uniformly on the peripheral arms as opposed to
localized hot-spots that are seen in single resonators like nanodisks [41]. While hot
spots occur at many wavelengths, the mode-volumes [49] at different wavelengths
may be different. The spectra are also found to be robust in the face of structural
asymmetry.
Although, Sierpinski carpet and the fractal tree geometry are the most explored
ones; there have been several reports on log-periodic plasmonic nanoantennae [26,
50–54] The space filling fractal shapes like the Hilbert space filling curve are
of interest particularly as transparent conducting electrodes and as SERS sub-
strates [55, 56]. There is also interest in shapes like the Appolonian gasket [33].
74 R.S. Hegde
The structures reported in the recent past have mostly concerned with regular
deterministic fractals with simplified studies of the influence of random config-
uration changes on the spectral behavior. Electrochemically grown metal fractals
are a particularly interesting case study; they have already been explored in the
context of radio frequency antennae [57]. It was reported that with the multitude
of length scales present in such structures, denser multiband spectral response can
be expected. Electrochemically grown plasmonic fractal nanostructures have been
reported to exhibit extremely high metal-enhanced fluorescence [58–60] and Raman
scattering [60]. The study of regular tree fractal structure can be extended towards
stochastic fractals like the DLA fractal and in further understanding of the optical
behavior in such random structures [61].
References
1. Maier SA (2007) Plasmonics: fundamentals and applications. Springer, New York. doi:
10.1007/0-387-37825-1
2. Ciraci C, Hill RT, Mock JJ, Urzhumov Y, Fernandez-Dominguez AI, Maier SA, Pendry JB,
Chilkoti A, Smith DR (2012) Science 337(6098):1072. doi:10.1126/science.1224823
3. Anker JN, Hall WP, Lyandres O, Shah NC, Zhao J, Van Duyne RP (2008) Nat Mater 7(6):442.
doi: 10.1038/nmat2162
4. Stockman MI (2006) Electromagnetic theory of SERS. Top Appl Phys. 103:47. Springer. doi:
10.1007/11663898_3
5. Jackson JB, Halas NJ (2004) Proc Natl Acad Sci U S A 101(52):17930. doi:
10.1073/pnas.0408319102
6. Kundu J, Le F, Nordlander P, Halas NJ (2008) Chem Phys Lett 452(1–3):115. doi:
10.1016/j.cplett.2007.12.042
7. Scalora M, Vincenti Ma, de Ceglia D, Roppo V, Centini M, Akozbek N, Bloemer MJ (2010)
Phys Rev A 82:33. doi: 10.1103/PhysRevA.82.043828
8. Fischer H, Martin OJF (2008) Opt Express 16(12):9144. doi: 10.1364/OE.16.009144
9. Gallinet B, Martin OJF (2011) ACS Nano 5(11):8999. doi: 10.1021/nn203173r
10. Becker J, Trügler A, Jakab A, Hohenester U, Sönnichsen C (2010) Plasmonics 5(2):161. doi:
10.1007/s11468-010-9130-2
11. Hegde RS, Mesch M, Giessen H (2014) In: Proceedings of META’14, the 5th international
conference on metamaterials, photonic crystals and plasmonics, Singapore
12. Liu S, Sinclair MB, Mahony TS, Jun CY, Campione S, Ginn J, Bender DA, Wendt JR, Ihlefeld
JF, Clem PG, Wright JB, Brener I (2014) Optical magnetic mirrors without metals. Optica
1:250–256
13. Gottheim S, Zhang H, Govorov AO, Halas NJ (2015) ACS Nano 9(3):3284. doi:
10.1021/acsnano.5b00412, http://pubs.acs.org/doi/abs/10.1021/acsnano.5b00412
14. Luk’yanchuk B, Zheludev NI, Maier Sa, Halas NJ, Nordlander P, Giessen H, Chong CT (2010)
Nat Mater 9(9):707. doi: 10.1038/nmat2810, http://www.ncbi.nlm.nih.gov/pubmed/20733610
15. Brolo AG, Gordon R, Leathem B, Kavanagh KL, Langmuir (2004) 20(12):4813. doi:
10.1021/la0493621
16. Gallinet B (2012) Fano resonances in plasmonic nanostructures: fundamentals, numerical
modeling and applications. Ph. D. thesis, EPFL. doi: 10.5075/epfl-thesis-5299
17. Cohen N (2015) In: Frame M, Cohen N (eds) Benoit Mandelbrot: a life in many dimensions,
chap 8, World Scientific Publishing Company, Singapore, pp 207–229
18. Fowler AD (1991) The fractal geometry of nature (revised edition), W.H. Freeman and Com-
pany, San Francisco, vol. 17. doi: 10.1016/0098-3004(91)90106-N, http://www.sciencedirect.
com/science/article/pii/009830049190106N
19. Werner DH, Haupt RL, Werner PL (1999) IEEE Antennas Propag Mag 41(5):37. doi:
10.1109/74.801513
20. Werner DH, Ganguly S (2003) IEEE Antennas Propag Mag 45(1):38. doi:
10.1109/MAP.2003.1189650
21. Shalaev VM, Moskovits M, Golubentsev Aa, John S (1992) Phys A Stat Mech Appl 191(1–
4):352. doi: 10.1016/0378-4371(92)90551-Z
22. Sederberg S, Elezzabi AY (2011) Appl Phys Lett 98(26):1. doi: 10.1063/1.3605570
23. Falconer KJ (1990) Fractal geometry: mathematical foundations and applications. John
Wiley and Sons, New York doi: 10.2307/2532125, http://books.google.ca/books/about/Fractal_
geometry.html?id=_WrvAAAAMAAJ&pgis=1
24. Hohlfeld RG, Cohen N (1999) Fractals 7(Compendex), 79 doi: 10.1142/S0218348X99000098
http://dx.doi.org/10.1142/S0218348X99000098
25. Hohlfeld RG, Cohen N (2003) IEEE International Symposium on Phased Array Systems and
Technology 2003-January, 588 doi:10.1109/PAST.2003.1257047
26. Gonz??lez FJ, Boreman GD (2005) Infrared Phys Technol 46(5):418. doi:
10.1016/j.infrared.2004.09.002
27. Hsu KH, Back JH, Fung KH, Ferreira PM, Shim M, Fang NX (2010) J Raman Spectrosc
41(10):1124. doi: 10.1002/jrs.2581
28. Volpe G, Volpe G, Quidant R (2011) Opt Express 19(4):3612. doi: 10.1364/OE.19.003612
29. Rosa L, Sun K, Juodkazis S (2011) Phys Stat Solidi Rapid Res Lett 5(5–6):175. doi:
10.1002/pssr.201105136
30. Sederberg S, Elezzabi aY (2011) Opt Express 19(11):10456. doi: 10.1364/OE.19.010456
31. Chen TL, Dikken DJ, Prangsma JC, Segerink F, Herek JL (2014) N J Phys 16(9):093024.
doi: 10.1088/1367-2630/16/9/093024 http://stacks.iop.org/1367-2630/16/i=9/a=093024?key=
crossref.eb87b5b053d2a20a3b13c027c3153f1d
32. Cakmakyapan S, Cinel Na, Cakmak AO, Ozbay E (2014) Opt Express 22(16):19504. doi:
10.1364/OE.22.019504 http://www.osapublishing.org/viewmedia.cfm?uri=oe-22-16-19504&
seq=0&html=true
33. Abdellatif S, Kirah K (2013) Opt Lett 38(18):3680. doi: 10.1364/OL.38.003680 http://ol.osa.
org/abstract.cfm?URI=ol-38-18-3680
34. Zhu LH, Shao MR, Peng RW, Fan RH, Huang XR, Wang M (2013) Opt Express 21 Suppl
3(May), A313 doi: 10.1364/OE.21.00A313, http://www.ncbi.nlm.nih.gov/pubmed/24104419
35. Kazerooni H, Khavasi A (2014) Opt Quantum Electron 46(6):751. doi:
10.1007/s11082-013-9783-0
36. Eshaghian A, AbdollahRamezani S, Chizari A, Mehrany K (2016) pp 1–16 http://arxiv.org/
abs/1605.07208
37. Huang X, Xiao S, Ye D, Huangfu J, Wang Z, Ran L, Zhou L (2010) Opt Express
18(10):10377. doi: 10.1364/OE.18.010377, http://www.osapublishing.org/viewmedia.cfm?
uri=oe-18-10-10377&seq=0&html=true
38. Li G, Chen X, Ni B, Li O, Huang L, Jiang Y, Hu W, Lu W (2013) Nanotechnology
24(20):205702. doi: 10.1088/0957-4484/24/20/205702 http://www.ncbi.nlm.nih.gov/pubmed/
23598737
39. Hegde RS, Khoo EH (2015) Plasmonics doi: 10.1007/s11468-015-0059-3, http://dx.doi.org/
10.1007/s11468-015-0059-3
40. Nagatani T (1989) Phys Rev A 39(1):438
41. Kuttge M, García De Abajo FJ, Polman A (2010) Nano Lett 10(5):1537. doi:
10.1021/nl902546r
42. Willingham B, Brandl DW, Nordlander P (2008) Appl Phys B Lasers Opt 93(1):209. doi:
10.1007/s00340-008-3157-5
43. Funston AM, Novo C, Davis TJ, Mulvaney P (2009) Nano Lett 9(4):1651. doi:
10.1021/nl900034v
44. Knight MW, Wu Y, Lassiter JB, Nordlander P, Halas NJ (2009) Nano Lett 9(5):2188. doi:
10.1021/nl900945q
45. Powell DA, Kivshar YS (2010) Appl Phys Lett 97(9):1. doi: 10.1063/1.3486480
76 R.S. Hegde
46. Zhang J, Ou JY, Papasimakis N, Chen Y, MacDonald KF, Zheludev NI (2011) Opt Express
19(23):23279. doi: 10.1364/OE.19.023279
47. Cui Y, Phang IY, Hegde RS, Lee YH, Ling XY (2014) ACS Photonics 1(7):631. doi:
10.1021/ph5001154, http://pubs.acs.org/doi/abs/10.1021/ph5001154
48. Cui Y, Hegde RS, Phang IY, Lee HK, Ling XY (2014) Nanoscale 6(1):282. doi:
10.1039/c3nr04375d.Please
49. Maier Sa (2006) Optics Express 14(5):1957. doi: 10.1364/OE.14.001957
50. Pavlov RS, Curto AG, Van Hulst (2012) NF Opt Commun 285(16):3334. doi:
10.1016/j.optcom.2012.04.010
51. Solís DM, Taboada JM, Araújo MG, Obelleiro F, Rubiños López JO (2013) Opt Commun (July
2015) 61:301–302. doi: 10.1016/j.optcom.2013.03.047
52. Pavlov RS, Curto AG, Van Hulst (2012) NF Opt Commun 285(16):3334. doi:
10.1016/j.optcom.2012.04.010
53. Yang J, Kong F, Li K, Sheng S (2015) Opt Commun 342:230. doi:
10.1016/j.optcom.2014.12.075, http://dx.doi.org/10.1016/j.optcom.2014.12.075
54. Aouani H, Rahmani M, Šípová H, Torres V, Hegnerová K, Beruete M, Homola J, Hong M,
Navarro-Cía M, Maier Sa (2013) J Phys Chem C 117(36): 18620. doi: 10.1021/jp404535x
55. Grigorenko I (2013) Appl Phys Letters 103(4):2014. doi: 10.1063/1.4816726
56. Afshinmanesh F, Curto AG, Milaninia KM, Van Hulst NF, Brongersma ML (2014) Nano Lett
14(9):5068. doi:10.1021/nl501738b
57. Puente C, Claret J, Sagues F, Romeu J, Lopez-Salvans MQ, Pous R (1996) Electron Lett
32(25):2298. doi: 10.1049/el:19961579
58. Luchowski R, Shtoyko T, Matveeva E, Sarkar P, Borejdo J, Gryczynski Z, Gryczynski I (2010)
Appl Spectrosc 64(6):578. doi: 10.1366/000370210791414317
59. Dong J, Qu S, Zheng H, Zhang Z, Li J, Huo Y Li G (2014) Sens Actuators B: Chem 191:595.
doi: 10.1016/j.snb.2013.09.088
60. Stockman MI, Shalaev VM, Moskovits M, Botet R, George TF (1992) Phys Rev B 46(82):2821
61. Agrawal A, Matsui T, Zhu W, Nahata A, Vardeny ZV (2009) Phys Rev Lett 102:113901
Chapter 5
Compact Slow-Light Enhaced Plasmonic
Waveguide Refractive Index Sensors
Yin Huang, Changjun Min, and Georgios Veronis
Abstract We introduce slow-light enhanced subwavelength scale refractive index

sensors which consist of a plasmonic metal-dielectric-metal (MDM) waveguide
based slow-light system sandwiched between two conventional MDM waveguides.
We first consider a MDM waveguide with small width structrue for comparison, and
then consider two MDM waveguide based slow light systems: a MDM waveguide
side-coupled to arrays of stub resonators system and a MDM waveguide side-
coupled to arrays of double-stub resonators system. We find that, as the group
velocity decreases, the sensitivity of the effective index of the waveguide mode
to variations of the refractive index (RI) of the fluid filling the sensors as well as
the sensitivities of the reflection and transmission coefficients of the waveguide
mode increase. The sensing characteristics of the slow-light waveguide based sensor
structures are systematically analyzed. We show that the slow light enhanced sensors
lead to not only 3.9 and 3.5 times enhancements in the RI sensitivity, and therefore
in the minimum detectable refractive index change, but also to two and three times
reductions in the required sensing length, respectively, compared to a sensor using
a MDM waveguide with small width structure. We further investigate slow-light-
enhanced ultracompact plasmonic Mach–Zehnder interferometer (MZI) sensors.
The sensing arm consists of either a conventional MDM plasmonic waveguide
or a MDM waveguide side-coupled to arrays of double-stub resonators system.
We show that the MZI sensor using the double-stub resonators system leads to
approximately an order of magnitude enhancement in the RI sensitivity compared to
the sensor using a conventional MDM plasmonic waveguide. We also find that the
Y. Huang ()
Department of Optoelectrics Information Science and Engineering,
School of Physics and Electronics, Central South University, 410012 Changsha, Hunan, China
e-mail: yhuan15@csu.edu.cn
C. Min
Institute of Micro-Nano Optics, Key Laboratory of Optoelectronic Devices
and Systems of Ministry of Education and Guangdong Province, College of Optoelectronic
Engineering, Shenzhen University, 518060 Shenzhen, China
G. Veronis
School of Electrical Engineering and Computer Science, and the Center for Computation
and Technology, Louisiana State University, 70803 Baton Rouge, LA, USA

DOI 10.1007/978-3-319-48081-7_5
78 Y. Huang et al.
RI sensitivity for such a plasmonic MZI sensor is approximately two times larger
than the sensitivity of a waveguide cavity sensor in which the MDM waveguide
side-coupled to arrays of double-stub resonators system is sandwiched between two
conventional MDM waveguides.
Keywords Plasmon resonances • Guided waves • Photonic sensors • Slow light

waveguides • Mach–Zehnder interferometer
5.1 Introduction
The unique properties of surface plasmons, which are light waves that propagate
along metal surfaces, enable a wide range of practical applications, including light
guiding and manipulation at the nanoscale [1]. In recent years, surface plasmon
resonance (SPR) based sensors have been widely employed and investigated [2–7],
especially the refractive index (RI) sensing. Both propagating surface plasmon
resonances (PSPRs) and localized surface plasmon resonances (LSPRs) exhibit
great potentials for sensing applications due to their susceptibility to the changes in
the RI of the surrounding environment [8]. Among different plasmonic waveguiding
structures, metal-dielectric-metal (MDM) plasmonic waveguides are of particular
interest [9–14], because they support modes with deep subwavelength scale over
a very wide range of frequencies extending from DC to visible, and are relatively
easy to fabricate [15]. In a MDM waveguide the modal fields are highly confined
in the dielectric region. This characteristic also makes the MDM waveguides very
attractive for sensing applications. In addition, slow light offers the opportunity
for compressing the local optical energy density, which enhances light-matter
interactions, and thereby improves the performance of nanoscale plasmonic devices
[16–20]. Therefore, it is essential to investigate the sensing characteristics of
plasmonic MDM waveguide based slow-light RI sensors.
We investigate RI sensors consisting of a plasmonic slow-light waveguide
sandwiched between two conventional MDM waveguides. In these structures,
light is coupled from an input MDM plasmonic waveguide to a plasmonic slow-
light waveguide system, and then coupled back to an output MDM plasmonic
waveguide. We first consider a MDM waveguide with small width as the plasmonic
waveguide sensing system. We next consider two different plasmonic slow-light
waveguide sensing systems: the MDM waveguide side-coupled to arrays of MDM
stub resonators system [19] and the MDM waveguide side-coupled to arrays of
MDM double-stub resonators system [20]. We find that, decreased group velocity
vg in slow-light systems significantly enhances not only the sensitivity of the
effective index of optical mode to variations of the refractive index of the fluid
filling in the sensors, but also the sensitivities of the transmission and reflection
coefficients to variations of the RI of the fluid. The two optimized slow-light
enhanced subwavelength plasmonic RI sensors result in not only 3.9 and 3.5
times enhancements in the sensitivity, and therefore in the minimum detectable
RI change, but also two and three times reductions in the optimal sensing length,
5 Compact Slow-Light Enhaced Plasmonic Waveguide Refractive Index Sensors 79
respectively, compared to a sensor using a MDM waveguide with small width

system. Although the two optimized slow-light enhanced sensors have comparable
performance in sensitivity, the double-stub resonator system exhibits a small group
velocity dispersion over a broader wavelength range and has less power loss,
features which are highly desirable for practical sensing applications.
We found that the attenuation associated with material loss in the metal limits
the performance of the slow-light enhanced plasmonic sensors to a great extent
because there is a trade-off between propagation loss and the group velocity [20].
In Mach-Zehnder interferometer (MZI) based sensor structures, light from a laser
enters the input waveguide and is split equally to the sensing and reference branches
[21]. Here, we would like to investigate MZI based slow light enhanced plasmonic
waveguide sensors. The sensing branch will be implemented with a MDM based
slow-light waveguide to enhance the performance of the sensor, while the reference
branch will be implemented with a low loss conventional MDM waveguide. Thus,
compared to the slow-light based plasmonic sensors that we proposed previously, we
expect that the overall power loss of the MZI based sensing system will decrease,
and therefore the sensitivity of the sensor will further increase. In addition, in
principle, an MZI based sensor system is relatively robust since the reference
arm can be used to cancel out common mode noise terms, such as temperature
modulations [22].
We first consider a MZI sensor with a conventional nanoplasmonic MDM
waveguide as the sensing waveguide. We then consider MZI based sensors in
which the sensing arm consists of a slow-light waveguide based on a MDM
waveguide side-coupled to arrays of MDM double-stub resonators system [20].
We show that an MZI based sensor using such a MDM waveguide side-coupled
to arrays of MDM double-stub resonators system leads to approximately an order
of magnitude enhancement in the refractive index sensitivity, and therefore in the
minimum detectable refractive index change, compared to a MZI based sensor
using a conventional MDM waveguide. In addition, we find that the refractive index
sensitivity of the optimized MZI-based slow-light plasmonic waveguide system is
also 1.9 times larger than the sensitivity of the previously introduced waveguide-
cavity sensor in which the MDM waveguide side-coupled to arrays of MDM
double-stub resonators system is sandwiched between two conventional MDM
waveguides.
5.2 Slow-Light Enhanced Plasmonic Waveguide Refractive

Index Sensors
For application of dense chip-scale integration, we consider compact subwavelength

scale biosensor. In all cases, the total length of the sensing structure is limited to
less than 1.1 m, which approximately corresponds to one wavelength in water
(w D 0 /nw , where nw D 1.332), when operating at the optical communication
80 Y. Huang et al.
wavelength (œ0 D 1.55 m). To characterize the sensing capability of the proposed
sensors, we define the following figure of merit (FOM) in terms of the relative
change in the output power that occurs for a change in the refractive index
ˇ ˇ ˇ ˇ
1 ˇˇ dPout .n/ ˇˇ ˇˇ dT.n/ ˇˇ
FOM D D (5.1)
Pin ˇ dn ˇ ˇ dn ˇ
where n is the RI of fluid and T D PPout

in
is the power transmission. The input power
Pin is given as a constant. The output power Pout is the only measurable quantity
.n/
in such a sensor. The changes in Pout are related to n via Pout D dPoutdn
[21].
Denoting the smallest measurable change in output power as Pout,min , we obtain
following expression for the detection limit nmin of the sensor
ˇ ˇ
1 ˇ Pout;min ˇ
jnmin j D ˇ ˇ (5.2)
Pin ˇ FOM ˇ
It is noted that the detection limit jnmin j decreases as the FOM increases. We
use a two-dimensional finite-difference frequency-domain (FDFD) method [23, 24]
to numerically calculate the transmission in the MDM plasmonic waveguide. This
method allows us to directly use experimental data for the frequency-dependent
dielectric constant of metals such as silver [25], including both the real and
imaginary parts, with no approximation. Perfectly matched layer (PML) absorbing
boundary conditions are used at all boundaries of the simulation domain [26].
5.2.1 MDM Waveguide with Small Width System
We first consider a RI sensor consisting of a MDM waveguide with small width

sandwiched between two conventional MDM waveguides (Fig. 5.1a). The width of
the sensing MDM waveguide is w0 D 50 nm. Since MDM waveguides with width
w 140 nm were used to guide optical mode in several plasmonic nanocircuits
both theoretically and experimentally [27, 28], here the widths of input and output
waveguides are also set to be w D 140 nm. All of the MDM waveguides in this
structure (Fig. 5.1a) have subwavelength widths, so that only the fundamental TM
mode is propagating in them. Thus, we can use single-mode scattering matrix theory
to account for the behavior of the system [29, 30]. We use FDFD to extract the
complex magnetic field reflection coefficient r1 and transmission coefficient t1 of
the fundamental mode of a MDM waveguide at the input interface between the
two MDM waveguides with different width (Fig. 5.2a), as well as the reflection
coefficient r2 and transmission coefficient t2 at the output interface (Fig. 5.2b).
The FOM of biosensor structure of Fig. 5.1a can then be calculated using
scattering matrix theory as [28, 29]
Fig. 5.1 (a) Schematic of the direct-coupled-cavity biosensor structure consisting of a MDM
waveguide with small width sandwiched between two MDM waveguides. (b) FOM for the
structure of (a) as a function of the sensing length d calculated using FDFD (black solid line)
and scattering matrix theory (red circles). Results are shown for w D 140 nm and w0 D 50 nm at
0 D1.55 m. The metal is silver and the fluid is water.
Fig. 5.2 (a) Schematic defining the reflection coefficient r1 , transmission coefficients t1 and power
transmission T1 when the fundamental TM mode of the input MDM waveguide is incident at an
interface between input and sensing waveguide. (b) Schematic defining the reflection coefficient
r2 , transmission coefficients t2 and power transmission T2 when the fundamental TM mode of the
MDM sensing waveguide is incident at an interface between and sensing and output waveguide
ˇ ˇ
ˇ dA dB d jt1 t2 j 2 da db ˇˇ
FOM D ˇˇe2A C˛ C Cˇ C CT C Ca C Cb (5.3)
dn dn dn dn dn ˇ
where
2
2 jt1 t2 j 2 a C b2 e4A 1
C˛ D ; (5.4)
2
4e2A jt1 t2 j 2 Œb cos .2B/ a sin .2B/

Cˇ D ; (5.5)
2
1
CT D ; (5.6)

82 Y. Huang et al.

2e2A jt1 t2 j 2 sin .2B/ be2A
Cb D ; (5.7)
2

2e2A jt1 t2 j 2 cos .2B/ ae2A
Ca D ; (5.8)
2
and the coefficient is defined as follow

ˇ ˇ
D ˇ1 r2 2 e2d ˇ 2 (5.9)
D 1 2ae4A cos .2B/ 2be2A sin .2B/ C a2 C b2 e4A ;
where D ˛ C iˇ is the complex wave vector of the fundamental propagating TM

mode in a sensing waveguide, ˛ is the attenuation constant, ˇ is the phase constant,
AD˛d, BDˇd, and a and b are real and imaginary part of r2 2 , respectively. In Eqs.
(5.5), (5.6), (5.7), (5.8), and (5.9), jt1 t2 j can be further calculated as
s R
Re f 1 "2 g j'1 j 2 dx
jt1 t2 j D T1 T2 R ; (5.10)
Re f 2 "1 g j'2 j 2 dx
where T1 and T2 are the power transmissions at the input and output interfaces
(Fig. 5.2), respectively, "i (i D 1,2) are complex dielectric constant of input and
output MDM waveguide, respectively, i and ®i (i D 1,2,) are complex wave vector
and field profile of the fundamental TM mode in the input and output plasmonic
MDM waveguide, respectively. Due to the symmetry of all biosensor structures
considered in this paper, we have
p
jt1 t2 j D T1 T2 (5.11)
2
1 t2 j
Denoting S D e2A C˛ dA C Cˇ dB , ST D e2A djtdn , and SR D
2A
da db
dn dn
e Ca dn C Cb dn , the FOM becomes
ˇ ˇ
FOM D ˇS C ST C SR ˇ (5.12)
A and B are related to the effective index of the sensing waveguide system,
2
1 t2 j
hence S is named as the index sensitivity coefficient. dn ; dn ; and djtdn
da db
are factors
associated with sensitivity of the reflection and transmission coefficient of the mode
to refractive index variations at the interfaces between MDM waveguides. ST and
SR will heretofore be referred to as the transmission sensitivity coefficient and
reflection sensitivity coefficient, respectively. We note that is a function of the
reflection coefficient r2 of the sensing waveguide system, and also observe that
the resonance enhancement factor 1/ exhibits a maximum when the Fabry-Perot
resonance condition 2arg(r2 )2ˇd D 2m is satisfied, where m is an integer. Thus,
C˛ , Cˇ , Ca , Cb , and CT are factors associated with the Fabry-Perot resonances of the
sensing waveguide. In addition, since A is directly related to the attenuation constant

of the effective wave vector, so the factor e2A is associated with the attenuation of
the optical power in the sensing waveguide system.
Figure 5.1b shows the FOM for the structure of Fig. 5.1a as a function of the
sensing length d at operating wavelength 0 D 1.55 m. For the range of length
shown, the maximum FOM (2.66) is obtained for such a structure at d D 910 nm
(Fig. 5.1b). The FOM is computed directly by approximating the differential in
.n/
Eq. (5.1) with the finite-difference formula dTout dn
D Tout .nCn/T
2n
out .nn/
[21,
31]. This approximation improves in accuracy as n ! 0. In the computations,
we use n D 104 << nw [21]. Figure 5.1b also shows the FOM calculated using
scattering matrix theory. It is observed that there is an excellent agreement between
the scattering matrix theory results and the exact results obtained using FDFD. Thus,
based on this analytical model, we can investigate the relative contributions of the
three sensitivity coefficients to the overall RI sensitivity of the sensor [Eq. (5.12)].
For the optimized MDM waveguide with small width biosensor structure, the
index sensitivity coefficient S is 2.5686 (Table 5.1). We note the sensitivity dB dn
(5.5068, Table 5.1) dominates over the sensitivity dA dn
(0.0459, Table 5.1), which
indicates the change in the phase constant of the mode induced by a RI variation
2
1 t2 j
is important in such a structure. On the other hand, the sensitivities djtdn da
, dn ,
db
and dn are 0.0404, 0.0130, and 0.0565 (Table 5.1), respectively, which means
there are almost no relative change in the power transmission and reflection at the
interface between MDM waveguides for a change in the refractive index of fluid,
and therefore, the transmission sensitivity coefficient ST (0.0422, Table 5.1) and
reflection sensitivity coefficient SR (0.0491, Table 5.1) are negligible.
5.2.2 MDM Side-Coupled to Arrays of Stub Resonators System
To enhance the FOM, we next consider a plasmonic waveguide system consisting

of a MDM waveguide side-coupled to a periodic array of stub resonators (stub-
resonator system) [19] with stub width w1 D 50 nm (Fig. 5.3a). N periods of the
structure are included in the sensing region, and the periodicity P is 150 nm. As
before, the total length of the sensing structure is limited to less than 1.1 m, the
width w and w0 are 140 nm and 50 nm, respectively. The group velocity of the
optical mode in this system at a given wavelength can be tuned by adjusting the stub
length L [19]. Figure 5.3b shows the FOM for the structure of Fig. 5.3a as a function
of the stub length L and the number of periods N. For the range of parameters
shown, we observe the optimized FOM of such a biosensor structure obtained at
L D 150 nm and N D 3 is 10.34, which is 3.9 times larger than that of the optimized
MDM waveguide with small width system (2.66, Table 5.1). Figure 5.3c shows the
first band of dispersion relation of MDM waveguide side coupled to arrays of stubs
system. We find such a system do support a slow-light mode for L D 150 nm at
84 Y. Huang et al.
Table 5.1 Attenuation factor e2A , effective index sensitivities dA dB

, ,
dn dn
transmission sensitivity
djt1 t2 j2 da db
dn
,
reflection sensitivities , ,
dn dn
corresponding Fabry-Perot factors C˛ , Cˇ , CT , Ca , Cb , index
sensitivity coefficient S , transmission sensitivity coefficient ST , reflection sensitivity coefficient
SR , and figure of merit FOM of biosensors calculated using scattering matrix theory. Results are
shown for the optimized systems of Figs. 5.1a, 5.3a, and 5.5a, respectively
MDM Side coupled to stubs EIT
e2A 0.9017 0.4699 0.5194
d˛ dA
,
dn dn
5.0485*104 , 0.0459 1.6570*107 , 7.4567 1.0263*107 , 3.0789
C’ 1.7371 1.3627 1.4956
dˇ dB 6 7
,
dn dn
6.0515*10 , 5.5068 6.4604*10 , 29.0718 9.7606*107 , 29.2818
C“ 0.5028 0.1073 0.1736
S 2.5686 6.2406 5.0320
djt1 t2 j2
dn
0.0404 6.5733 6.9367
CT 1.1589 0.9381 0.9864
ST 0.0422 2.8976 3.5539
da
dn
0.0130 3.8833 4.0082
Ca 0.9221 0.4053 0.0798
db
dn
0.0565 1.8689 0.8233
Ca 1.1764 0.5244 0.7808
SR 0.0491 1.2001 0.5000
FOM 2.6599 10.3383 9.0859
œ0 D 1.55 m. To investigate how the enhanced FOM actually depends on the slow
light effect, the FOM can also be expressed as
ˇ ˇ ˇ ˇ
ˇ dT .!/ ˇ 1 ˇˇ dT .!/ d! ˇˇ
ˇ
FOM D ˇ ˇ D (5.13)
dn ˇ P ˇ dn dn ˇ
in
where d! is the spectral shift resulting from a small variation dn. Figure 5.3d shows
the power transmissions of the sensing stub resonator system with L D 150 nm
for N D 3 and 4 obtained by FDFD. It is clear that there is no transmission when
frequency is within the band gap [beyond 200 THz, Fig. 5.3c]. In the vicinity
of a resonance frequency ! 0 , the sensing stub resonator system is analogous to
a photonic waveguide-cavity-waveguide system shown in the inset of Fig. 5.3d
approximately [32]. Using coupled-mode theory (CMT) [9, 10, 32, 33], the energy
amplitudes A, S1 C and S2 for the cavity, input and output waveguides can be
described by
s
A A A 2 C
i!A D i!0 A C S ; (5.14)
d 1 2 1 1
s
2
S2 D A; (5.15)
2
Fig. 5.3 (a) Schematic of the plasmonic biosensor structure consisting of a MDM waveguide side-
coupled to arrays of MDM stub resonators sandwiched between two MDM waveguides. (b) FOM
for the structure of (a) as a function of the stub length L and the number of periods N of the
sensing waveguide system. Results are shown for 0 D1.55 m, P D 150 nm, w0 D w1 D 50 nm
and w D 140 nm. The metal is silver and the fluid is water. (c) Dispersion relation of the
system consisting of a MDM waveguide side-coupled to arrays of stub resonators for stub length
L D 150 nm, L D160 nm, and L D 170 nm, respectively. All other parameters are as in (b). (d)
FOM for the structure of (a) as a function of the stub length L and the width w of the input and
output MDM waveguide. All other parameters are as in (b)
Here 1/ d is the decay rate due to the intrinsic loss, 1/ 1 and 1/ 2 are the
decay rates into the input and output waveguides. Again, 1 D 2 by symmetry,
and we denote the total decay rate into the input and output waveguides by
1/ w D 1/ 1 C 1/ 2 (with quality factor Qw D ! 0 w /2) and the total decay rate
by 1/ D 1/ w C 1/ d (with quality factor Q D ! 0 /2 ). Thus, the transmission
spectrum can be obtained by
1
dT .!/ 4Q2w
D
2 2 ; (5.16)
d! !!0 1
!0
C 4Q2
86 Y. Huang et al.
dT.!/
and the derivative d!
in Eq. (5.13) is
! !0
dT .!/ 2!02 Q2w
D
2 2 ; (5.17)
d! !!0 1
!0
C 4Q2
Moreover,
thephase of the output propagating mode at the cavity/waveguide

S
interface ˆ C 2
is arctan 2Q !!
!0
0
. Thus, the group delay g experienced by
S1
the propagating mode is given by
2Q
dˆ .!/ !0
g D D h
i2 ; (5.18)
d!
1 C 2Q !!
!0
0
At the resonance frequency ! 0 , the group delay is g D2Q/ ¨0 . Substituting Eq.

(5.18) into Eq. (5.17) gives
dT .!/ Q2
D 2 .! !0 / g2 .!/ 2 ; (5.19)
d! Qw
In terms of T(¨0 )DjQ/Qw j2 , dT.!/

d!
further becomes dT.!/
d!
D 2 .! !0 / 2g .!/ T 0 ,
where T0 is T(¨0 ). In the lossless case, QDQw , T(! 0 )D1, and dT.!/ d!
reduces
to 2 .! !0 / 2g .!/. Note that the group velocity vg of the propagating mode
is inversely proportional to the group delay g . Now, we have the following
relationship
dT .!/ T0
2 .! !0 / ; (5.20)
d! vg .!/
On the other hand, in the vicinity of resonance frequency ¨0, the rate of spectral
shift d!
dn
in Eq. 5.11 is the same as the rate of shift of the resonance frequency
approximately [34], that is, d!dn
d! dn
0
. For the resonance closest to the band gap
[slow light resonance, i.e., P1 , P2 in Fig. 5.3d], ˇ(! 0 ) /P as shown in Fig. 5.3c
[19]. Thus, we have
d! d!0 !0
' ' ; (5.21)
dn dn n
where n is the effective index of sensing waveguide. The derivative d!

dn
0
only depends
on material parameters [34]. Combining Eqs. (5.20) and (5.21), Eq. (5.13) becomes
ˇ ˇ ˇ ˇ
ˇ dT .!/ d! ˇ ˇˇ T0 !0 ˇ
ˇ
FOM D ˇˇ ˇˇ
ˇ .! !0 /ˇ (5.22)
dn dn ˇ nvg .!/
2 ˇ
It is shown that the FOM is inversely proportional to the square of the

group velocity vg . That is, for a given frequency !, the FOM increases as the
group velocity vg of the optical mode decreases due to the enhanced light-matter
interactions.
In Fig. 5.3d, we observe that the overlap between two adjacent resonance peaks
due to a strong cavity-waveguide coupling leads that the lineshape of a peak
(particularly, a peak with a resonance frequency far from the band gap) to some
extent departs from a Lorenzian. Note that the key assumption for the CMT is weak
coupling. In practice, it is typically found that the CMT is to be nearly exact for Q
30 (i.e., the quality factors of slow light resonance peaks P1 and P2 (Fig. 5.3d),
but also often qualitatively accurate for smaller Q [32]. Figure 5.3d shows that
the on resonance transmission coefficient is unity when the metallic loss is not
included, as predicted from the CMT. A slow light resonance (high £g) results in
light being ”trapped” in the cavity for a longer duration which results in a weaker
cavity-waveguide coupling and a higher Q (narrower spectral width of resonance,
Fig. 5.3d) and thus higher light-matter interactions. This is consistent with the
equation g (! 0 ) D 2Q/! 0 obtained by the CMT as well. In addition, although the
metal loss causes a power penalty relative to lossless case, the basic dependence in
T0
v2 g
(Eq. (5.22)) still holds. Thus, the enhancement in FOM of the stub-resonator
system here is an outcome of the slow-light effect. Based on this scattering matrix
theory modal, we next investigate how the slow-light effect affects the sensitivities
2
1 t2 j
( dA , dB , djtdn
dn dn
, dn ), attenuation factor e2A , and therefore the FOM of such
da db
, dn
biosensor system by adjusting stub length L.
Figure 5.4a shows the sensitivities d˛ dn
and dˇ
dn
of the stub resonator system in
structure of Fig. 5.3a as a function of the stub length L for optimized N D 3. We
find as L increases, both sensitivity d˛ dn
and sensitivity dˇdn
first increases, and then
decreases after a specific length L. More specifically, the sensitivity dˇ dn
achieves the
maximum for L D 155 nm, and then decreases to zero. To explain this, in Fig. 5.3c,
we also display the first band of dispersion relation of the stub-resonator system
for L D 160 nm and 170 nm, which exhibit band edge frequencies at 187 THz and
180 THz, respectively. As L increases, the band edge frequency of such system
decreases, and as the operating frequency approaches the band edge frequency,
the group velocity vg decreases [19]. Thus, at the beginning, by increasing the
stub length L from 0 nm to 155 nm, the sensitivity dˇ dn
increases due to the slow-
light enhanced light-matter interaction. However, as L further increases such as
L D 160 nm, the group velocity begins to increase at the operating frequency
(1.55 m) (Fig. 5.3c), so the sensitivity dˇ dn
decreases. When L D 170 nm, the
dˇ
sensitivity dn even becomes negative because the dispersion relation experience
back-bending with negative group velocity at 1.55 m (Fig. 5.3c).
da db
Figure 5.4b shows the sensitivities dn (black line) and dn (red line) of the MDM
waveguide side coupled to arrays of stubs system in structure of Fig. 5.3a as a
function of the stub length L for optimized N D 3. Recall that a and b are real and
imaginary part of r2 2 , that is, a D rp 2 –ra 2 , and b D 2rp ra . Here r2 D rp C ira is the
complex reflection coefficient of the fundamental mode of the stub-resonator system
88 Y. Huang et al.
Fig. 5.4 (a) FOM for the structure of Fig. 5.3a as a function of the stub length L calculated using
FDFD (black solid line) and scattering matrix theory (red circles). Results are show for optimized
dˇ
N D 3. All other parameters are as in Fig. 5.3b. (b) Sensitivities d˛ dn
(black line) and dn (red line) of
the MDM waveguide side coupled to arrays of stubs system in structure of Fig. 5.3a as a function
da db
of the stub length L. All parameters are as in (a). (c) Sensitivities dn (black line) and dn (red line)
of the slow light waveguide system in structure of Fig. 5.3a as a function of the stub length L. All
2
1 t2 j
parameters are as in (a). (d) Sensitivity djtdn (black line) and factor e2A (red line) of the slow
light waveguide system in structure of Fig. 5.3a as a function of the stub length L. All parameters
are as in (a)
at the output interface (Fig. 5.2b). The real and imaginal parts of this reflection
coefficient rp and ra are associated with the phase change and attenuation of the
reflected sensing mode at the output interface (Fig. 5.2b), respectively [37]. In
da db
Fig. 5.4b, we find that the sensitivities dn and dn experience a significant changes in
da db
the slow-light region. The reflection coefficient sensitivities dn and dn of the sensing
mode is directly related to the attenuation sensitivity dn and phase sensitivity dˇ
d˛
dn
of the sensing mode, respectively. When L <D 150 nm, as the group ˇvelocity ˇ
decreases, the slow light effect leads to an enhancement in both sensitivities ˇ dn
db ˇ
and
da dˇ d˛
dn
. Moreover, since the difference dn dn achieves the maximum at L D 150 nm,
da dr2 dr2
so the maximum sensitivity dn D dnp dna is also obtained at L D 150 nm. When
db
150 nm <L<D 155 nm, as L increases the dn keeps increasing as a result of the
slower group velocity (Fig. 5.4a). However, the difference dˇ
dn
d˛
dn
becomes smaller
da
as L increases at this length range, so that dn begins to decrease. Finally, when
ˇ db ˇ
L increases beyond 155 nm, the group velocity begins to increase, so ˇ dn ˇ rapidly
dˇ d˛ da
decreases. In addition, dn dn < 0 leads to negative sensitivity dn , and as L increases,
ˇ da ˇ
the difference d˛
dn
dˇ
dn
increases (Fig. 5.4a), an therefore ˇ dn ˇ increases as well.
2
1 t2 j
Figure 5.4c shows the sensitivity djtdn of the stub-resonator system in structure
of Fig. 5.3a as a function of the stub length L for optimized N D 3. We observe
2
1 t2 j
that sensitivity djtdn also experiences a significant enhancement as group velocity
tuned by the ˇ ˇstub length L. Note ˇ ˇthat t1 ' t2 due to reciprocity ˇ ˇ [37]. Thus, we obtain
jt1 t2 j 2 ' ˇt22 ˇ 2 . Consider that ˇt22 ˇ is approximated as 1 ˇr22 ˇ at the output interface
2
1 t2 j
2 d jt 2 j2 d.1jr22 j/
(Fig. 5.2b), we have djtdn dn2 dn
. With the help of r22 D a C ib, the
djt1 t2 j2
sensitivity dn
can be expressed as
ˇ ˇ
d jt1 t2 j 2 1 ˇr22 ˇ da db
' 2 ˇ 2 ˇ a Cb : (5.23)
dn ˇr ˇ dn dn
2
2
1 t2 j
It is noted that the sensitivity djtdn is opposite but proportional to the sum of
da db
sensitivities dn and dn as shown in Fig. 5.4c. Figure 5.4c also shows the attenuation
factor e2A of the stub-resonator system in structure of Fig. 5.3a as a function of
the stub length L for optimized N D 3. As the stub length L increases, the operating
frequency (0 D 1.55 m) approaches the band edge frequency, the field intensity in
the stub resonators is enhanced, the real part of wave vector increases, and therefore
the attenuation of the sensing mode increases [19]. For large L, the decrease in the
attenuation factor even dominates the increase in the slow-light enhanced sensitivity,
so the mode with an extremely low group velocity may not always be the one that
yields a high FOM. This is also consistent with Eq. (5.22) obtained by CMT. The
FOM is not only inversely proportional to the square of the group velocity vg,
but also proportional to the resonance transmission T0 . According to Eq. (5.3), the
overall FOM is thereby obtained maximum for L D 150 nm.
We find that for the optimized stub-resonator system the attenuation factor e2A
is 0.47 (Table 5.1) which is 1.9 times smaller compared to that of the MDM
waveguide with small width system 0.90 (Table 5.1). However, the absolute value
2
1 t2 j
of the sensitivities of the former system dA , dB , djtdn
dn dn
da
, dn db
, and dn are 7.46, 29.07,
6.5733, 3.8833, and 1.8689 (Table 5.1), which are 162.5, 5.3, 162.7, 298.7, and
33.1 times larger compared to those of the latter one 0.046, 5.51, 0.04, 0.013, and
0.057, respectively. Thus, the slow-light effect enhanced transmission sensitivity
coefficient ST (2.9, Table 5.1), reflection sensitivity coefficient SR (1.2, Table 5.1)
and contribution of the attenuation sensitivity dA dn
cannot not be negligible here. It
is noted that the absolute value of factors C˛ , Cˇ , CT , Ca , and Cb of the optimized
stub-resonator system are all smaller than those of the MDM waveguide with small
width system (Table 5.1), which indicates that the enhanced sensitivity coefficients
of former system are not due to a better Fabry-Perot resonance condition matching.
In addition, we observe that all factors C˛ , Cˇ , CT , Ca , and Cb are functions of
the attenuation factor e2A , therefore, the decrease in them is partly because of
90 Y. Huang et al.
the increase in attenuation of the slow-light mode. Overall, the optimized slow-
light enhanced MDM waveguide side-coupled to arrays of stub resonators system
(Fig. 5.3a) results in 3.9 times larger FOM compared to the optimized MDM
waveguide with small width system (Fig. 5.1a). It is worth noting that dT.n/ dn
[Eq.
(5.1)] can be positive or negative, which corresponds to increased or decreased
power transmission, respectively, for a change in the RI. A good design for a sensing
structure is to maximize such a variation in the power transmission. If all terms
2
1 t2 j
in Eq. (5.3), C˛ dA
dn
, Cˇ dB
dn
, CT djtdn da
, Ca dn db
, and Cb dn have the same sign, a high
FOM is achieved. Otherwise, they may cancel each other, and slow light enhanced
sensitivities may decrease instead of increasing the overall FOM. Since the MDM
waveguides with different widths are used in various optical nanocircuits [27, 28,
35, 36], we also show the FOM for the structure of Fig. 5.3a as a function of the stub
length L and width w of the input and output MDM waveguides for the optimized
N D 3 (Fig. 5.4d). We note that the optimized stub length L is around 150 nm for any
width, which suggests that the Fabry-Perot resonance effect here is less important
than the slow-light effect. Due to the same reason, in Fig. 5.3b, we observe that
for different number of periods N, the corresponding maximum FOM is always
obtained around L D 150 nm as well.
5.2.3 MDM Side-Coupled to Arrays of Double-Stub

Resonators System
We finally consider a slow-light plasmonic waveguide system consisting of a MDM

waveguide side-coupled to arrays of double-stub resonators (double-resonator
system) (Fig. 5.5a) based on our previous work [20], which was also implemented
experimentally by a recent work [28]. Unlike the proposed MDM waveguide side-
coupled to arrays of stub resonators system, such system can exhibits a small
group velocity dispersion and a large slowdown factor over a broad wavelength
range, features which are highly desirable for practical applications of slow-light
devices [19, 20]. As before, the total length of the structure is limited to less than
1.1 m. The transmission spectra of such a double-stub resonator structure feature
a transparency peak centered at a frequency which is tunable through the length of
the composite cavity formed by the two stub resonators L1 C L2 C w [20]. Here we
choose L1 C L2 C w D 420 nm and w0 D 50 nm, so that the transparency peak is
centered at the operating frequency of f D 194 THz (0 D 1.55 m) approximately.
Figure 5.5b shows the FOM for the structure of Fig. 5.5a as a function of the
stub length L1 and the number of periods N of the sensing waveguide system. For
the range of parameters shown, we observe one FOM peak for such a biosensor
structure obtained at L1 D 145 nm and N D 2 is 9.09, which is 3.5 times larger
than that of the optimized MDM waveguide with small width system (2.66, Table
5.1). We also observe that increasing the number of periods N in the sensing region
decreases the optimized stub length L1 of biosensor. For a plasmonic EIT system,
as L1 decreases, L2 increases, since L1 C L2 C w0 is fixed. Thus, the stub lengths

difference L D L2 L1 increases, hence the frequency spacing between the stub
resonances ! increases. As a result, the slowdown factor of the mode decreases,
the corresponding propagation length and attenuation factor e2A increases [20].
When the number of periods N increases, the optimized stub length L1 has to
decrease to match this increase in the length of sensing region. Figure 5.5c shows the
second band of dispersion relation of such sensing waveguide system, and we find
the sensing waveguide do support a slow light mode for L1 D 145 nm at operating
wavelength 0 D 1.55 m. Figure 5.5c shows the second band of dispersion relation
of such sensing waveguide system, and we find the sensing waveguide supports a
slow light mode for L1 D 145 nm at 0 D 1.55 m. As we discussed above, the
3.5 times enhancement is an outcome of the slow light effect. Figure 5.5d also
show the FOM for the structure of Fig. 5.5a as a function of the stub length L1
and width w of the input and output MDM waveguide for optimized N D 2. Like
the stub resonator system (Fig. 5.4a), we observe that for different width w, the
corresponding maximum FOM is always obtained around L1 D 145 nm as well.
Again, we use single-mode scattering matrix theory to account for the behavior of
the system.
Figure 5.6a shows the sensitivities d˛ dn
and dˇdn
of the double-stub resonator system
in structure of Fig. 5.5a as a function
ˇ of the stub length L1 for optimized N D 2. A
ˇ dˇ
similar trend is seen: both j d˛
dn ˇ
and dn
first increase, and then decrease as L1 further
increases. The sensitivity dˇdn
achieves the maximum when L1 D 165 nm, and then
goes to zero. To explain this, in Fig. 5.5c, we display the second band of dispersion
relation of the double-stub resonator system for L D 165 nm and 172.5 nm,
respectively. The stub lengths difference L D L2 L1 decreases, the frequency
spacing between the stub resonances ! decreases, hence the group velocity of
the mode decreases [20]. As the stub length L1 increases from 0 nm to 165 nm,
the sensitivity dˇ
dn
increases due to the slow-light effect. As L1 further increases
to L1 D 172.5 nm, the group velocity begins to increase at the œ0 D 1.55 m
(Fig. 5.5c), so the sensitivity dˇ
dn
decreases.
2
da db djt1 t2 j
Figure 5.6b and c also show the sensitivities dn , dn , dn , and the attenuation
factor e2A of the sensing waveguide system in structure of Fig. 5.5a as a function of
the stub length L1 for optimized N D 2. The effect of slow-light on the performance
of such a sensing system can be explained as in the case of the single-stub resonator
system (Sect. 5.2.2).
We find that for the optimized double-stub resonator system the attenuation factor
e2A is 0.52 (Table 5.1) which is 1.7 times smaller compared to that of the MDM
waveguide with small width system (0.90, Table 5.1). However, like previously
investigated stub-resonator system, the absolute of sensitivities of the double-stub
2
1 t2 j
resonator system j dA
dn
j, j dB
dn
j, j djtdn da
j, j dn db
j, and j dn j (3.0789, 29.2818, 6.9367, 4.0082
and 0.8233, Table 5.1) are enhanced by the slow-light effect. The corresponding
enhancements with respect to those of the MDM waveguide with small width system
are 66.9, 5.3, 171.7, 308.3, and 14.4, respectively. We note that the absolute values
92 Y. Huang et al.
Fig. 5.5 (a) Schematic of the plasmonic slow-light biosensor structure consisting of a MDM based
plasmonic analogue of EIT system sandwiched between two MDM waveguides. (b) FOM for the
structure of (a) as a function of the stub length L1 and the number of periods N of the sensing
waveguide. All parameters are as in Fig. 5.3b. The metal is silver and the fluid is water. (c)
Dispersion relation of the slow-light waveguide based on a plasmonic analogue of EIT system
for stub length L1 D 145 nm, L1 D165 nm, and L1 D 172.5 nm, respectively. All other parameters
are as in Fig. 5.3b. (d) FOM for the structure of (a) as a function of the stub length L1 and the
width w of the input and output MDM waveguide. All other parameters are as in Fig. 5.3b
of factors C˛ , Cˇ , CT , Ca , and Cb of such a double-stub resonator system are not

as good as those of MDM waveguide with small width system so that the enhanced
sensitivity coefficients ST , SR and S (3.5539, 0.5000 and 5.0320, Table 5.1) of
the double-stub resonator system are a result of the slow-light effect rather than a
better Fabry-Perot resonance enhancement. We also observe that the FOM of the
double-stub resonator system is slightly smaller than that of MDM side-coupled to
arrays of stubs system just due to a smaller factor Ca (0.0798, Table 5.1). Since
the attenuation factors e2A of these two systems are comparable, so the sensitivity
da
dn
of the double-stub resonator system has a worse Fabry-Perot resonance condition
matching.
Table 5.2 summarizes the optimized design for each structure at operating
wavelength 0 D 1.55 m. The detection limit nmin is computed using Eq. (5.2)
assuming input power of Pin D 1 mW and smallest measurable change in output
Fig. 5.6 (a) FOM for the structure of Fig. 5.5a as a function of the stub length L1 calculated using
FDFD (black solid line) and scattering matrix theory (red circles). Results are show for optimized
dˇ
N D 2. All other parameters are as in Fig. 5.5b. (b) Sensitivities d˛ dn
(black line) and dn (red line)
of the slow light waveguide system in structure of Fig. 5.5a as a function of the stub length L. All
da db
parameters are as in (a). (c) Sensitivities dn (black line) and dn (red line) of the plasmonic EIT
system in structure of Fig. 5.5a as a function of the stub length L1 . All parameters are as in (a).
2
1 t2 j
(d) Sensitivity djtdn (black line) and factor e2A red line) of the slow light waveguide system in
structure of Fig. 5.5a as a function of the stub length L1 . All parameters are as in (a)
Table 5.2 Summary of waveguide designs and operating parameters at 0 D 1.55 m. The
optimal sensing lengths de , power transmission coefficients and detection limits nmin of sensors
are shown for the optimized systems of Figs. 5.1a, 5.3a, and 5.5a, respectively
de Transmission nmin
MDM 910 nm 0.663 3.8 106
Stub resonator 450 nm 0.349 9.7 107
Double-stub resonator 300 nm 0.363 1.1 106
power of Pout,min D 10 nW [21]. Based on the same equations [Eqs. (5.1) and (5.2)]
and conditions (Pin D 1 mW and Pout,min D 10 nW), Berini investigated the sensing
performance of a generic Mach-Zehnder interferometer (MZI) implemented with
plasmonic waveguides, such as metal-dielectric single interface waveguide, thin
DMD waveguide (width of 20 nm) and thin MDM waveguide (width of 20 nm), at
the operating wavelength 0 D 1.31 m [21]. The metal was gold and the dielectric
also was water. The detection limits of these structures were 3.6 107 , 1.5 108
and 6.6 106 with optimal sensing lengths de D 82.9 m, 2039 m and 2.5 m,
94 Y. Huang et al.
respectively. For a comparison, although the detection limits of the first two MZI
based structures are 2.7 and 64.7 times smaller than that of the slow light enhanced
stub resonator structure (9.7 107 , Table 5.2), the required sensing lengths of these
two MZI based structures are 184.2 and 4531.1 times larger than that of the slow
light enhanced stub resonator structure (450 nm, Table 5.2), respectively. In other
words, these two structures are not suitable for ultradense chip-scale integration. In
Table 5.2, we also observe that the slow-light enhanced sensors lead to not only 3.9
and 3.5 times enhancements in the detection limit, but also to two and three times
reductions in the required sensing length, respectively, compared to the sensor using
a MDM waveguide with small width structure. It is due to the fact that the slow
light enhanced light-matter interactions can not only enhance the sensitivity, but can
also greatly reduce the required sensing length, thereby enabling the realization of
miniature sensors [38].
Finally, the power loss is of practical importance for application of plasmonic
sensors. In Table 5.1, we observe that the attenuation factor e2A of optimized
double-stub resonator system (0.5194) is higher than that of stub resonator system
(0.4699) at 0 D 1.55 m. In Table 5.2, it is also observed that the transmission
of the former optimized system (0.363) is higher than that of the latter one (0.349).
In Fig. 5.6d, we show the real part of the wave vector (attenuation constant) as
a function of frequency of the optical mode ’ for the optimized stub resonator
and double-stub resonator systems. In the stub resonator system, the attenuation
constant ’ greatly increases as the frequency approaches the band edge frequency.
On the other hand, in the double-stub resonator system the attenuation constant ’
is relatively small in the corresponding slow light frequency range (186–210 THz,
Fig. 5.5c). Figure 5.6d demonstrates that the attenuation constant ’ in the double-
stub resonator system is smaller than that in the single stub resonator system at the
operating wavelength of 0 D 1.55 m, in other words, the double-stub resonator
system is less lossy. This is due to the fact that the composite cavity formed by
two stubs in the double-stub resonator system has a weak resonance at a slow light
wavelength such as 0 D 1.55 m.
5.3 Mach-Zehnder Interferometer Based Slow-Light

Enhanced Plasmonic Waveguide Sensors
We consider a MZI, implemented with plasmonic waveguiding structures, which is

used as a refractive index sensor (Fig. 5.7a). The fluid to be sensed flows over the
sensing arm of the MZI. The fluidic channel and the plasmonic waveguide structure
in the sensing arm overlap in the sensing region. The length and width of the sensing
region are LS and WS , respectively.
The modes of the plasmonic waveguides in the sensing and reference arms
are excited by an optical source such as a laser (Fig. 5.7a). The output power,
Fig. 5.7 (a) Schematic of a Mach-Zehnder interferometer (MZI) used as a refractive index sensor.
Fluidic channel overlaps with the sensing arm of the MZI. The fluid to be sensed flows over the
sensing arm. The sensing fluidic channel and the plasmonic waveguide structure in the sensing
arm overlap in the sensing region. The length and width of the sensing region are LS and WS ,
respectively. (b) Schematic of the plasmonic waveguide system used in the sensing region of
the MZI sensor. It consists of a periodic array of two MDM stub resonators side-coupled to a
MDM waveguide. N periods of the structure are included in the sensing region of the sensor. (c) A
conventional MDM waveguide used in the sensing region of the MZI sensor
which is the only measurable quantity in such a MZI structure, is collected by a

photodetector, and is given by [39]
ˇ ˇ2
ˇ1 1 ˇ
Pout .n/ D Pin ˇˇ eŒA.n/CjB.n/ C e.A0 CjB0 / ˇˇ ; (5.24)
2 2
where n is the refractive index of the biological or chemical analyte in the sensing
arm, eA(n) and eA0 is the attenuation of the optical mode field in the sensing and
reference arms, respectively, while B(n) and B0 is the insertion phase of the sensing
and reference arms, respectively.
For maximum sensitivity the reference arm insertion phase B0 is therefore chosen
so that in the absence of analyte (n D n0 ), we have B(n0 ) D B0 ˙ m/ 2, where
m D 1, 3, 5,.. [21]. In addition, unlike the symmetric case in which the attenuation
in the two sensor arms in the absence of analyte is limited to be the same [39], here
the attenuation in the absence of analyte in the two sensor arms (eA(n0) and eA0 ) is
not limited to be the same. We find that
96 Y. Huang et al.
ˇ ˇ ˇ !
@Pout .n/ ˇˇ 1 @A ˇˇ @B ˇˇ
D eAA0 Pin e ACA0
: (5.25)
@n ˇnDn0 2 @n ˇnDn0 @n ˇnDn0
Thus, the sensor’s normalized power sensitivity FOM is defined as

ˇ ˇ ˇ ˇ !ˇˇ
ˇ 1 @P .n/ ˇˇ ˇ ˇ 1 @A
ˇ
ˇ @B ˇ
ˇ out ˇ ˇˇ D ˇˇ eAA0 ˇ
FOM D ˇ eACA0 ˇˇ ˇ ˇ D jT Sj :
ˇ Pin @n ˇnDn0 ˇ ˇ 2 @n nDn0 @n ˇnDn0 ˇ
(5.26)
where transmission coefficient T D eAA0 is the geometric mean of the power

transmission coefficients through
the two sensor arms in the absence of analyte,
and sensitivity coefficient S D 12 eACA0 @A j
@n nDn0
@B
@n nDn0
is associated with
the sensitivity of A(n) and B(n) to refractive index variations. More specifically,
when the difference between the sensing and reference arms insertion phases in the
absence of analyte is positive (negative), S is the difference between (sum of) the
relative attenuation sensitivity @A
@n
eACA0 and the insertion phase sensitivity @B
@n
.
We also use a two-dimensional FDFD method, to numerically calculate the
transmission through the sensing arm of the sensor. The transmission through the
sensing arm of the sensor is t(n) D exp [A(n) jB(n)]. We use our FDFD code
to calculate the transmission through the sensing arm. The sensitivities of A(n)
and B(n)to refractive index variations can then be calculated by approximating the
derivatives by finite differences: @A
@n
A.nCn/A.nn/
2n
and @B@n
B.nCn/B.nn/
2n
.
4 @A @B
In our calculations, we also use n D 10 << n0 . Once @n and @n are calculated,
the FOM is calculated using Eq. (5.26).
5.3.1 Conventional MDM Waveguide in the Sensing Arm
We first consider the case where the plasmonic waveguiding structure used in
the sensing arm of the MZI sensor is a conventional MDM plasmonic waveguide
(Fig. 5.7c). The metal and the dielectric are assumed to be silver and water,
respectively. In Fig. 5.2, we first consider a deep subwavelength sensing region
with length Ls D 200 nm, and show the FOM for such a structure as a function
of the width of the MDM waveguide w, calculated using FDFD and Eq. (5.26) (red
solid line). We observe that the FOM increases as the width of the waveguide w
decreases. This is due to a fact that, as w decreases, the group velocity of the optical
mode supported by the waveguide decreases [19]. The sensitivity of the effective
index of the mode to variations of the refractive index of the dielectric therefore
increases [39, 40], so that the sensor’s normalized power sensitivity increases.
Note that the practical lower limit for the MDM waveguide width is w 20 nm
[21]. Thus, the maximum FOM when a conventional MDM plasmonic waveguide
is used in the sensing region is 0.7 for sensing length Ls D 200 nm. Also note
that when a translationally invariant waveguide such as the conventional MDM

plasmonic waveguide is used in the sensing region of the MZI sensor we have
A D ˛Ls and B D ˇLs , where ˛ and ˇ D 2/0 neff are the real and imaginary
parts, respectively, of the propagation constant of the fundamental MDM waveguide
mode. In addition, for the range of parameters considered here, we can neglect the
sensitivity @˛
@n
so that Eq. (5.26) reduces to [21, 39]
ˇ ˇ ˇ
ˇ 1 2 @neff ˇˇ ˇ
ˇ ˛Ls A0 ˇ
FOM D ˇ e Ls ˇ: (5.27)
ˇ 2 0 @n ˇnDn0 ˇ
In Fig. 5.8, we show the FOM calculated with the approximation of Eq. (5.27)
(black circles), as well and we observe that it is in excellent agreement with the
FOM calculated using FDFD. Here neff is calculated using FDFD by exciting the
fundamental mode of the waveguide with a dipole source [41]. In addition, we also
consider the FOM as a function of the width of the MDM waveguide w for different
sensing lengths of such structure. We found that the FOM increases as Ls increases.
Even though the sensitivity of the effective index of the mode to variations of the
@n
refractive index of the dielectric @neff does not depend on the sensing length Ls ,
the FOM still increases as Ls increases as can be seen from Eq. (5.27). On the
other hand, the transmission coefficient T of the sensor decreases as the sensing
length Ls increases for all these lossy waveguides. However, the increase in the
sensitivity coefficient S dominates the decrease in the transmission coefficient T of
the sensor, so that overall the FOM also increases as the sensing length Ls of the
MDM plasmonic waveguide increases.
Fig. 5.8 Figure of merit

@P .n/
FOM P1in out @n
of a MZI
refractive index sensor
implemented with MDM
plasmonic waveguides as a
function of the width w of the
waveguides for different
sensing region lengths Ls.
Results are calculated at
f D 194 THz using the
analytical approximation of
Eq. (5.27) (circles) and FDFD
(solid lines). The metal and
dielectric are assumed to be
silver and water, respectively
98 Y. Huang et al.
5.3.2 MDM Side-Coupled to Arrays of Double-Stub

Resonators System in the Sensing Arm
We now consider the case where the plasmonic waveguide structure used in
the sensing arm of the MZI sensor is a slow-light plasmonic waveguide system
consisting of a MDM waveguide side-coupled to arrays of double-stub resonators
(double-resonator system) (Fig. 5.7b). N periods of the structure are included in
the sensing region. We first consider a MZI sensor in which a single period of the
double-resonator system is included in the sensing region (N D 1). The transmission
spectra of such a structure features a transparency peak centered at a frequency
which is tunable through the length L1 C L2 C w of the composite cavity formed
by the two stub resonators [20, 39]. Here we choose L1 C L2 C w D 420 nm
and w D 50 nm, so that the transparency peak is approximately centered at the
operating frequency of f D 194 THz (0 D 1.55 m) [39]. Our proposed structure
can be fabricated using techniques such as focused ion beam milling, electron
beam lithography, and nano-imprint lithography [42–44]. The width of the MDM
waveguide in the reference arm is also set to be 50 nm. In Fig. 5.9a, we show
the FOM and enhancement for the MZI sensor in which a single period of the
double-resonator system is included as a function of the stub resonator length L1
with negative sign (“” solid line) and positive sign (“C” dashed line) in Eq. (5.26).
The enhancement here is calculated with respect to a reference MZI sensor with
a conventional MDM waveguide in the sensing arm which has the same sensing
region length. In addition, the width of the conventional MDM waveguide in this
reference MZI sensor is chosen w D 20 nm so as to maximize its FOM. We also
note that the choice of the sign in Eq. (5.26) depends on whether the difference
between the sensing and reference arms insertion phases in the absence of analyte,
which as mentioned above is chosen to be an odd multiple of / 2, is chosen to be
positive or negative. In both the positive and negative sign cases, as L1 increases, the
FOM first increases, and then decreases, as L1 further increases (Fig. 5.9a).
We also investigate the effect of varying the number of periods N of the
plasmonic EIT system in the sensing region. In all cases, we found that the FOM as
a function of the stub length L1 has a similar trend as in the N D 1 case (Fig. 5.9b,
c, d, e and f), and the maximum FOM of 17.6 is obtained when N D 10. If
the number of periods N is further increased, the FOM for the optimum stub
length L1 decreases. This is because for large sensing lengths the decrease in the
transmission coefficient T dominates the increase in the sensitivity coefficient S.
Figure 5.9 also demonstrates that increasing the number of periods N in the sensing
region decreases the optimum stub length L1 of the sensor. This is due to the fact
that when lossy waveguides are used in the arms of a MZI sensor the optimum
length of the sensing region is equal to the propagation length of the optical mode
in the waveguides [22, 23]. For the plasmonic EIT system considered here, as L1
decreases, L2 increases, since L1 C L2 C w is fixed as mentioned above. Thus,
the stub lengths difference L D L2 L1 increases, hence the frequency spacing
L1 L1
L1 L1
L1 L1
@P .n/
Fig. 5.9 Figure of merit FOM P1in out @n
and enhancement of a MZI refractive index sensor
implemented with a double-resonator system in the sensing arm (Fig. 5.7b) as a function of the
stub resonator length L1 with w D 50 nm for negative sign (“” solid line) and positive sign
(“C” dashed line) in Eq. (5.26). Results are calculated for (a) a single period, (b) two periods, (c)
three periods, (d) four periods, (e) six periods and (f) ten periods of the double-resonator structure
included in the sensing region at f D 194 THz using FDFD. The period is d D 100 nm, and
the sensing region width is WS D L1 C L2 C w D 420 nm. The width of the MDM waveguide
in the reference arm is also 50 nm. The metal and dielectric are assumed to be silver and water,
respectively
between the stub resonances ! increases. As a result, the slow down factor of the
optical mode decreases and the corresponding propagation length increases.
Therefore, when the number of periods N increases, the optimized stub length
L1 has to decrease to match this increase in the length of the sensing region. In
100 Y. Huang et al.
addition, for comparison, the enhancement (shown on the right vertical axis in
Fig. 5.9a–f) with respect to a conventional MDM sensor with width w D 20 nm
is also calculated, and is proportional to the FOM. In Fig. 5.9, we observe that the
maximum enhancement of 16.5 is obtained for N D 1. Thus, even though the
maximum FOM is obtained for N D 10, the maximum enhancement with respect
to the reference MZI sensor with a conventional MDM waveguide in the sensing
arm is obtained for N D 1. In other words, the enhancement due to the slow light
effect is more pronounced for smaller sensing lengths. In short, we find that for
subwavelength sensing regions, using the plasmonic EIT system in the sensing arm
of the MZI sensor results in significantly larger FOM, compared to a sensor in which
a conventional MDM waveguide is used in the sensing arm [21]. As mentioned
above, a MZI sensor using the plasmonic EIT system results in a maximum FOM
of 6.0 and 17.6 for N D 1 and N D 10, respectively, while the maximum FOM
for a MZI sensor using a conventional MDM waveguide is 0.36 and 2.9 for
Ls D 100 nm and Ls D 1000 nm, respectively. Thus, the maximum enhancements
are 16.5 and 6 for Ls D 100 nm and Ls D 1000 nm, respectively. In other words,
a MZI sensor using the plasmonic EIT system leads to approximately an order
of magnitude enhancement in the FOM, and therefore in the minimum detectable
refractive index change, compared to a MZI sensor using a conventional MDM
waveguide.
Since the maximum FOM and maximum enhancement are obtained for N D 10
and N D 1 (Fig. 5.9), respectively, we next investigate the transmission coefficient
T and sensitivity coefficient S of these two cases in detail. In Fig. 5.10a and b, we
show the transmission coefficient T, the sensitivity coefficient S and the relative
attenuation sensitivity @A
@n
eACA0 as well as the insertion phase sensitivity @B @n
, as a
function of the stub resonator length L1 for N D 1. As L1 increases, L2 decreases,
since L1 C L2 C w is fixed, as mentioned above. Thus, the stub lengths difference
L D L2 L1 decreases, and therefore the frequency spacing between the stub
resonances ! decreases. As a result, the field intensity in the stub resonators is
enhanced, the real part of the wave vector increases, and the transmission coefficient
T, which corresponds to the transparency peak, decreases due to the metallic loss
(Fig. 5.10a) [19, 20, 39]. On the other hand, as L decreases, the slowdown
ˇ ˇ factor
increases [20]. Thus, the amplitude of the insertion phase sensitivity ˇ @B @n
ˇ, increases
for L1 < 152 nm. We note that, if L1 > 152 nm, we have L2 < 218 nm so that L
and therefore ! are small. In this regime we found that the group velocity
ˇ ˇ actually
increases [21], so that the amplitude of the insertion phase sensitivity ˇ @B
@n
ˇ decreases,
ˇas @A
L1 increases
ˇ (Fig. 5.10b). The amplitude of the relativeˇ attenuation
ˇ sensitivity
ˇ eACA0 ˇ as a function of L1 has a similar trend as ˇ @B ˇ (Fig. 5.10b). The
@n @n
amplitude of the overall sensitivity coefficient jSC j (jS j) [Eq. (5.26)] is maximized
for L1 D 152 nm (144 nm). As mentioned above, SC (S ) corresponds to the
difference between the sensing and reference arms insertion phases in the absence
of analyte being a negative (positive) odd multiple of / 2. In addition, the FOM is
the amplitude of the product of the transmission coefficient T, and the sensitivity
coefficient S ((4)). For L1 < 152 nm (144 nm) there is a trade-off between T which
decreases, and jSC j (jS j) which increases, as L1 increases (Fig. 5.10b). For small L1
L1 L1
L1 L1
Fig. 5.10 (a and c) Transmission coefficient T for a MZI refractive index sensor in which the
sensing region consists of a double-resonator waveguide system (Fig. 5.7b) with N D 1 and N D 10
as a function of the stub resonator length L1 . Results are calculated for w D 50 nm, and sensing
region width WS D L1 C L2 C w D 420 nm. All other parameters are as in Fig. 5.9. (b and d)
Sensitivity coefficients jS j (black solid line), jSC j (black dashed line), and relative attenuation
sensitivity @A
@n
eACA0 (red solid line), and insertion phase sensitivity @B@n
(red dashed line) as a
function of the stub resonator length L1 for N D 1 and N D 10. All other parameters are as in (a)
the increase in jSC j (jS j) dominates the decrease in T, so that the FOM increases,
as L1 increases. However, for larger L1 the decrease in T becomes significant, and
the FOM eventually decreases (Fig. 5.9a). The maximum FOM of 6.0 is obtained
for L1 148 nm.
In Fig. 5.10c and d, we also show the transmission coefficient T, the sensitivity
coefficient S, the relative attenuation sensitivity, @A
@n
eACA0 and the insertion phase
sensitivity @B , for N D 10 as a function of the stub resonator length L1 . The
@n ˇ ˇ
amplitude of the insertion phase sensitivity ˇ @B
@n
ˇ initially increases with L1 , and then
decreases after reaching the maximum as in ˇthe ˇN D 1 case. In the N D 10 case, the
amplitude of the insertion phase sensitivity ˇ @B ˇ is much larger than the amplitude
ˇ @A ACA
@n ˇ
of the relative attenuation sensitivity ˇ @n e 0 ˇ, so that the overall sensitivity
coefficient depends almost only on the insertion phase sensitivity (Fig. 5.10d). We
also observe that the dependence of the transmission coefficient T (Fig. 5.10c), the
sensitivity coefficient S (Fig. 5.10d), and therefore of the FOM (Fig. 5.9f) on L1
102 Y. Huang et al.
for N D 10 is very similar to the single period case (N D 1) (Figs. 5.10a, b, and
5.9a). For a given L1 , increasing the number of periods N of the double-resonator
system, increases both the attenuation and the insertion phase of the optical mode
propagating through the structure. Thus, the transmission coefficient T decreases,
while the sensitivity coefficient S increases (Fig. 5.10), as the number of periods
N increases. The maximum FOM of 17.6 for the N D 10 case is obtained for
L1 96 nm (Fig. 5.9f).
In Fig. 5.11a and c, we show the FOM of the MZI sensor based on the plasmonic
EIT system as a function of frequency for N D 1 and N D 10, respectively.
In Fig. 5.11b and d, we also show the transmission (black) and reflection (red)
spectra calculated using FDFD for the sensing arm of the sensors of Fig. 5.11a
and c, respectively. When N D 1, we observe that, as expected, the FOM of
the sensor exhibits a peak at f D 192 THz (Fig. 5.11a), which coincides with
the transparency peak in the transmission spectra of the double-resonator system
(Fig. 5.11b). Note that the transparency peak is actually centered at f D 192 THz
for L1 C L2 C w D 420 nm with L1 D 148 nm, and therefore the maximum FOM
is slightly higher than the one at f D 194 THz shown in Fig. 5.9a (6.0). When
N D 10, we observe that the FOM and transmission spectra are both much broader
than those of N D 1. In addition, the FOM exhibits a peak at f D 194 THz, whereas
the transmission peak is at f D 210 THz. These results can be explained as follows.
First, the stub length L1 used in the case of N D 10 (L1 D 96 nm) is smaller than
the one used for N D 1 (L1 D 148 nm). As L1 decreases, the stub lengths difference
L D L2 L1 increases. Thus, the frequency spacing between the stub resonances
! increases, and therefore the width of the transmission band increases. Second,
when we consider a multi-period structure in the sensing region, the interaction
between periods can give rise to a shift of the resonance peak and also broaden the
transmission band of the entire system.
Figure 5.12a,c, e show the magnetic field profiles associated with the optimized
structures of Fig. 5.9a for L1 D 148 nm (double-resonator system with N D 1
in the sensing region), Fig. 5.9f for L1 D 96 nm (double-resonator system with
N D 10 in the sensing region), and Fig. 5.8 for sensing length Ls D 1000 nm
(conventional MDM waveguide in the sensing region), at f D 194 THz. Figure
10b and d demonstrate that the contribution to the sensitivity coefficient S from
the insertion phase sensitivity @B
@n
is larger than that from the relative attenuation
sensitivity @A
@n
e ACA0
for a double-resonator slow light waveguide system in the
sensor arm, and for N D 10 the latter one is negligible compared to the former
one. In Fig. 12b and d, we clearly observe a change in the phase of the mode in
the output waveguide when there is an refractive index change of 0.1 introduced in
the sensing region for N D 1 and N D 10, respectively. Furthermore, the change
in phase for N D 10 is significantly larger than the one for N D 1, which is also
consistent with the results shown in Fig. 10b and d. On the other hand, Fig. 5.12f
shows that there is a much smaller change in the phase of the mode in the output
waveguide for the optimized conventional MDM waveguide sensing system with the
same index change and sensing length as in Fig. 5.12d. In short, the field profiles of
.n/
Fig. 5.11 (a and c) Figure of merit FOM P1in @Pout @n
of a MZI refractive index sensor, in which
the sensing region consists of a double-resonator waveguide system (Fig. 5.7b), as a function of
frequency. Results are shown for L1 D 148 nm (N D 1) and 96 nm (N D 10). All other parameters
are as in Fig. 5.10a. (b and d) Transmission (black) and reflection (red) spectra for the sensing arm
of the sensors of Fig. 5.5a and c, respectively.
Fig. 5.12 demonstrate that the induced change in the phase of the waveguide mode
propagating through the sensing arm of the MZI sensor, when the refractive index
in the sensing region is varied, is much larger if a double-resonator system is used
in the sensing arm, compared to if a conventional MDM waveguide is used.
Table 5.3 Summarizes the optimized design for the MZI based sensor using
the double-resonator system with N D 10 and the sensor consisting of a double-
resonator system sandwiched between two conventional MDM waveguides ˇ at theˇ
1 ˇ Pout;min ˇ
operating wavelength of œ0 D 1.55 m. The detection limit nmin D p ˇ FOM ˇ
in
is computed using Eq. (5.26) assuming input power of Pin D 1 mW and smallest
measurable change in output power of Pout,min D 10 nW [21]. The detection
limit of the optimized sensor structure consisting of a double-resonator system
sandwiched between two conventional MDM waveguides is 1.1 106 with
optimal sensing length Ls D 300 nm and transmission coefficient T D 0.52. For
comparison, although the required sensing length of the optimized MZI based
structure (1000 nm) is 3.3 times larger than that of the previously proposed structure
104 Y. Huang et al.
Fig. 5.12 (a) Magnetic field

profile for the structure of
Fig. 5.9a with L1 D 148 nm
at f D 194 THz when the
fundamental TM mode of the
MDM waveguide is incident
from the left. (b) Same as in
(a) except there is a refractive
index change of 0.1 in the
sensing region. (c) Magnetic
field profile for the structure
of Fig. 5.9f with L1 D 96 nm.
(d) Same as in (c) except
there is a refractive index
change of 0.1 in the sensing
region. (e) Magnetic field
profile for the structure of
Fig. 5.8 with Ls D 1000 nm
at f D 194 THz. (f) Same as
in (e) except there is also a
refractive index change of 0.1
in the sensing region.
(300 nm), the detection limit of the MZI based structure (5.8107 ) is 1.9
times smaller than that of the previously proposed structure (1.1106 ). In other
words, the MZI based slow light sensor can achieve smaller detection limit while
still keeping the size of the device at the subwavelength scale. In Table 5.3, we
also observe that although the required sensing length of the MZI based structure
(1000 nm) is 3.3 times larger than that of the previously proposed structure
(300 nm), the transmission coefficient T of the MZI based structure (0.43) is only
slightly smaller than that of the previously proposed structure (0.52). This indicates
that the MZI based slow-light system has a smaller insertion loss. This is due to
the fact that in the MZI based system the input energy is partially fed into the
low-loss reference arm, which leads to decreasing the overall power loss of the
sensor system. Thus, the effective sensing length and therefore the sensitivity of the
sensor system can be increased. In other words, for a large eA0 (the attenuation of
the optical mode field in the low loss reference arm), the transmission coefficient
Table 5.3 Summary of Parameter MZI Double-stub resonatora

optimized designs and
operating parameters at Ls 1000 nm 300 nm
œ0 D 1.55 m. The optimum T 0.43 0.52
sensing lengths Ls , nmin 5.8107 1.1106
transmission coefficients T, a
The optimum sensing lengths Ls , transmission
and detection limits nmin of coefficients T, and detection limits nmin of sensors
sensors are shown for the are shown for a slow-light plasmonic waveguide
optimized system of Fig. 5.7b system consisting of a MDM waveguide side-
coupled to arrays of double-stub resonators sand-
wiched between two conventional MDM waveg-
uides (double-stub case) in Sect. 5.2.
T of the MZI based sensor structure (eAA0 ) will be much larger than that of a
sensor structure consisting of a plasmonic EIT system sandwiched between two
conventional MDM waveguides (e2A , corresponding to the attenuation factor in
Sect. 5.2). In addition, the maximum FOM of the asymmetric case investigated here
(in which the attenuation in the two sensor arms in the absence of analyte is not
limited to be the same) is 2.6 times larger than that of the symmetric case (in which
the attenuation in the two sensor arms in the absence of analyte is limited to be the
same) [39]. Note that, even in the case of the same sensing length Ls D 200 nm, the
maximum FOM of the asymmetric case (Fig. 5.9b) is still 1.6 times larger than
that of the optimized symmetric case [39].
As final remarks, the sensing performance of our proposed plasmonic MZI sensor
can be measured using a mircofluidic platform [45–47]. The sensing channel can be
defined by fabricating a metal-air-metal slot waveguide using lithographic methods.
The sensing solution can flow into the sensing arm. A material such as CYTOP (a
fluoropolymer with a refractive index close to that of biologically compatible fluids,
such as water [47, 48]) can be used in the dielectric core of the reference arm. Thus,
the refractive index of the core of the reference arm remains constant, when different
sensing solutions flow into the sensor. The propagation constant of the fundamental
waveguide mode of the reference arm is entirely dependent on the refractive indices
of CYTOP and silver, and has no dependence on the refractive index of the fluid
to be sensed. The light can be injected and collected using grating or slits coupler
structures. A nanoslit or grating structure connected via a triangular air groove can
be etched in the input port of the plasmonic waveguide to effectively excite surface
plasmons. Similarly, a decoupling grating can be etched in the output port to scatter
the surface plasmons into free space for measurement purposes. The scattered light
can then be collected by a long working distance objective [28, 45, 49].
In addition, while here we design our structure so as to optimize the figure of
merit at 1550 nm, the structure can be designed to operate at different wavelengths
by properly choosing the MDM stub resonator lengths. Due to imperfections
introduced during the fabrication process, the operating wavelengths of fabri-
cated waveguide-cavity structures usually differ from the theoretically designed
106 Y. Huang et al.
wavelengths [50]. As an example, photonic crystal resonator sensors and ring

resonator sensors generally suffer from fabrication imperfections [51, 52]. One
way to address this issue is to use a post fabrication tuning mechanism. As an
example, it has been demonstrated that differential thermal tuning can be used to
align the resonance wavelengths of photonic crystal waveguide-cavity structures
[50]. This thermal tuning technique requires no extra materials or structures, and
thus avoids the potential quality degradation of the photonic crystal microcavities
and excess fabrication complexity [50]. This post-fabrication tuning method relaxes
the fabrication requirements.
We also note that it has been shown that, if the metal film is thick enough,
2-D simulations of MDM plasmonic waveguide devices (which correspond to
infinite metal film thickness) are in good agreement with three-dimensional (3-D)
simulations (which take into account the finite thickness of the metal film). In
fact, for a metal film thickness of 1 m it was shown that 2-D simulations of
MDM plasmonic waveguide-cavity devices are in excellent agreement with 3-D
simulations [53]. In addition, we recently showed that, with proper choice of their
design parameters, a 3-D plasmonic two-conductor waveguide-cavity device and
a 2-D MDM device can have nearly identical transmission spectra [54]. In other
words, given a 2-D MDM plasmonic waveguide device, it is possible to find in
a systematic way a 3-D plasmonic two-conductor waveguide-cavity device with
nearly identical transmission spectra. Using this approach, the 2-D plasmonic device
design that we propose here can be implemented using 3-D plasmonic waveguides.
References
424:824–830
2. Hao F, Nordlander P, Sonnefraud Y, van Dorpe P, Maier SA (2009) Tunability of subradiant
dipolar and fano-type plasmon resonances in metallic ring/disk cavities: implications for
nanoscale optical sensing. ACS Nano 3:643–652
3. Liu N, Weiss T, Msche M, Langguth L, Eigenthaler U, Hirscher M, Sonnichsen C, Giessen
H (2010) Planar metamaterial analogue of electromagnetically induced transparency for
plasmonic sensing. Nano Lett 10:1103–1107
4. Gao Y, Gan Q, Xin Z, Cheng X, Bartoli FJ (2011) Plasmonic Mach-Zehnder interferometer for
ultrasensitive on-chip biosensing. ACS Nano 5:9836–9844
5. Fu YH, Zhang JB, Yu YF, Lukyanchuk B (2012) Generating and manipulating higher order
fano resonances in dual-disk ring plasmonic nanostructures. ACS Nano 6(6):5130–5137
6. Fan F, Chen S, Wang XH, Chang SJ (2013) Tunable nonreciprocal terahertz transmission and
enhancement based on metal/magneto-optic plasmonic lens. Opt Express 21(7):8614–8621
7. Zhou JH, Xu XP, Han WB, Mu D, Song H, Meng Y, Leng X, Yang J, Di X, Chang Q (2013)
Fano resonance of nanoparticles embedded in Fabry-Perot cavities. Opt Express 21(10):12159–
12164
8. Brolo AG (2012) Plasmonics for future biosensor. Nat Photonics 6:709–713
9. Yu Z, Veronis G, Fan S, Brongersma ML (2008) Gain-induced switching in metal-dielectric-
metal plasmonic waveguides. Appl Phys Lett 92:041117
10. Min C, Veronis G (2009) Absorption switches in metal-dielectric-metal plasmonic waveguides.

Opt Express 17:10757
11. Cai W, White JS, Brongersma ML (2009) Compact, high-speed and power-efficient electro-
optic plasmonic modulators. Nano Lett 9:4403
12. Gramotnev DK, Bozhevolnyi SI (2010) Plasmonics beyond the diffraction limit. Nat Photonics
4:83
13. Zhan G, Liang R, Liang H, Luo J, Zhao R (2014) Asymmetric band-pass plasmonic nanodisk
filter with mode inhibition and spectrally splitting capabilities. Opt Express 22:9912
14. Wu T, Liu Y, Yu Z, Peng Y, Shu C, Ye H (2014) The sensing characteristics of plasmonic
waveguide with a ring resonator. Opt Express 22:7669
15. Economou EN (1969) Surface plasmons in thin films. Phys Rev 182:539–554
16. Karalis A, Lidorikis E, Ibanescu M, Joannopoulos J, Soljacic M (2005) Surface-plasmon-
assisted guiding of broadband slow and subwavelength light in air. Phys Rev Lett 95:063901
17. Sandtke M, Kuipers L (2007) Slow guided surface plasmons at telecom frequencies. Nat
Photonics 1:573
18. Gan Q, Ding YJ, Bartoli FJ (2009) “Rainbow” trapping and releasing at telecommunication
wavelengths. Phys Rev Lett 102:056801
19. Yang L, Min C, Veronis G (2010) Guided subwavelength slow-light mode supported by a
plasmonic waveguide system. Opt Lett 35:4184
20. Huang Y, Min C, Veronis G (2011) Subwavelength slow-light waveguides based on a
plasmonic analogue of electromagnetically induced transparency. Appl Phys Lett 99:143117
21. Berini P (2008) Bulk and surface sensitivities of surface plasmon waveguides. New J Phys
10:105010
22. Fan X, White IM, Shopova SI, Zhu H, Suter JD, Sun Y (2008) Sensitive optical biosensors for
unlabeled targets: a review. Anal Chim Acta 620:8–26
23. Wu SD, Glytsis EN (2002) Finite-number-of-periods holographic gratings with finite-width
incident beams: analysis using the finite-difference frequency-domain method. J Opt Soc Am
A 19:2018–2029
24. Veronis G, Dutton RW, Fan S (2004) Method for sensitivity analysis of photonic crystal
devices. Opt Lett 29:2288–2290
25. Palik ED (1985) Handbook of optical constants of solids. Academic, New York
26. Jin J (2002) The finite element method in electromagnetics. Wiley, New York
27. Melikyan A, Alloatti L, Muslija A, Hillerkuss D, Schindler PC, Li J, Palmer R, Korn D,
Muehlbrandt S, Van Thourhout D et al (2014) High-speed plasmonic phase modulators. Nat
Photonics 9:229
28. Zhu Y, Hu X, Yang H, Gong Q (2014) On-chip plasmon-induced transparency based on
plasmonic coupled nanocavities. Sci Rep 4:3752
29. Kocabas SE, Veronis G, Miller DAB, Fan S (2008) Transmission line and equivalent circuit
models for plasmonic waveguide components. IEEE J Sel Top Quantum Electron 14:1462–
1472
30. Huang Y, Min C, Veronis G (2012) Compact slit-based couplers for metal-dielectric-metal
plasmonic waveguides. Opt Express 20:22233
31. Taflove A (1995) Computational electrodynamics. Artech House, Boston
32. Joannopoulos J, Johnson S, Winn J, Meade R (2008) Molding the flow of light. Princeton
University Press, Princeton
33. Cao G, Li H, Zhan S, Xu H, Liu Z, He Z, Wang Y (2013) Formation and evolution mechanisms
of plasmoninduced transparency in MDM waveguide with two stub resonators. Opt Express
21:9198–9205
34. Digonnet K, Wen H, Terrel M, Huo Y, Fan S (2012) Slow light in fiber sensors. Proc SPIE
8273:82730W1
35. Huang KC, Seo M, Sarmiento T, Huo Y, Harris JS, Brongersma ML (2014) Electrically driven
subwavelength optical nanocircuits. Nat Photonics 8:244–249
36. Fu Y, Hu X, Lu C, Yue S, Yang H, Gong Q (2012) All-optical logic gates based on nanoscale
plasmonic slot waveguides. Nano Lett 12:5784–5790
108 Y. Huang et al.
37. Pozar DM (1998) Microwave engineering. Wiley, New York

38. Zhao Y, Zhang Y, Wang Q (2012) High sensitivity gas sensing method based on slow light in
photonic crystal waveguide. Sens Actuator B Chem 173:28–31
39. Huang Y, Dastmalchi P, Veronis G (2014) Slow-light enhanced nanoscale plasmonic waveguide
sensors and switches. Proc SPIE 8988:89880V1–10
40. Soljacic M, Johnson SG, Fan S, Ibanescu M, Ippen E, Joannopoulos JD (2002) Photonic crystal
slow light enhancement of nonlinear phase sensitivity. J Opt Soc Am B 19:2052–2059
41. Veronis G, Fan S (2005) Bends and splitters in subwavelength metal-dielectric-metal plasmonic
waveguides. Appl Phys Lett 87:131102
42. Salamin Y, Heni W, Haffner C, Fedoryshyn Y, Hoessbacher C, Bonjour R, Zahner M,
Hillerkuss D, Leuchtmann P, Elder DL et al (2015) Direct conversion of free space millimeter
waves to optical domain by plasmonic modulator antenna. Nano Lett 15:8342–8346
43. Huang X, Brongersma M (2013) Compact aperiodic metallic groove arrays for unidirectional
launching of surface plasmons. Nano Lett 15:5420–5424
44. Cattoni A, Ghenuche P, Haghiri-Gosnet A, Decanini D, Chen J, Pelouard J, Collin S (2011)
œ3 /1000 plasmonic nanocavities for biosensing fabricated by soft UV nanoimprint lithography.
Nano Lett 15:3557–3563
45. Gao Y, Gan Q, Xin Z, Cheng X, Bartoli FJ (2011) Plasmonic Mach-Zehnder interferometer for
ultrasensitive on-chip biosensing. ACS Nano 5:9836–9844
46. Krupin O, Asiri H, Wang C, Tait RN, Berini P (2013) Biosensing using straight long-range
surface plasmon waveguides. Opt Express 21:698–709
47. Wong W, Adikan FRM, Berini P (2015) Long-range surface plasmon Y-junctions for refer-
enced biosensing. Opt Express 23:31098–31108
48. Tencer M, Berini P (2009) Design of microfluidic channels separated by an ultra-thin free-
standing dielectric membrane. Microfluid Nanofluid 21:17–26
49. Feng J, Siu VS, Roelke A, Mehta V, Rhieu SY, Palmore GT, Pacifici D (2011) Nanoscale
plasmonic interferometers for multispectral, high-throughput biochemical sensing. Nano Lett
12:602–609
50. Pan J, Huo Y, Yamanaka K, Sandhu S, Scaccabarozzi L, Timp R, Povinelli ML, Fan S, Fejer
M, Harris JS (2008) Aligning microcavity resonances in silicon photonic-crystal slabs using
laser-pumped thermal tuning. Appl Phys Lett 92:103114
51. Tung B, Dao D, Ikeda T, Kanamori Y, Hane K, Sugiyama S (2011) Investigation of strain
sensing effect in modified single-defect photonic crystal nanocavity. Opt Express 19:8821–
8829
52. Donzella V, Sherwali A, Flueckiger J, Grist SM, Fard ST, Chrostowski L (2015) Design and
fabrication of SOI micro-ring resonators based on sub-wavelength grating waveguides. Opt
Express 23:4791–4803
53. Bahramipanah M, Abrishamian MS, Mirtaheri SA, Liu J (2014) Ultracompact plasmonic loop–
stub notch filter and sensor. Sensors Actuators B Chem 194:311–318
54. Mahigir A, Dastmalchi P, Shin W, Fan S, Veronis G (2015) Plasmonic coaxial waveguide-
cavity devices. Opt Express 23:20549–20562
Chapter 6
Fabrication, Properties and Applications
of Plasmene Nanosheet
Qianqian Shi, Dashen Dong, Kae Jye Si, and Wenlong Cheng
Abstract Plasmene nanosheets are free standing, one-particle thick superlattices

from the self-assembly of plasmonic nanoparticles. Such nanosheets are a new class
of two-dimensional (2D) nanomaterials with unique materials properties that differ
from the constituent nanoparticles or corresponding bulk materials. Their optical,
mechanical and ion-transporting properties could be tuned by adjusting the sizes
and shapes of constituent nanoparticles, and controlling the interparticle spacing
during self-assembly process. Here, we first discuss the viable plasmene nanosheet-
fabrication approaches including programmable assembly, micro-hole confined
assembly, liquid-liquid interfacial assembly and air-water interfacial assembly.
Then, we describe the unique plasmonic and mechanical properties of plasmene
nanosheets. Such soft nanosheets could further shaped into one-dimensional (1D)
nanoribbons and three-dimensional (3D) origami by focused ion beam lithography.
Finally, we show how these properties can be utilized for various technical
applications including mechanical resonator, surface-enhanced Raman scattering
(SERS) substrate, drug identification and anti-counterfeit security label. We believe
that plasmene nanosheets represent a “newborn” in the 2D materials family, which
may lead to the design of novel metamaterials with unexpected applications.
Keywords Plasmene nanosheets • Plasmonic nanoparticles • Nanoparticle super-

lattices • Two-dimensional nanomaterials • Self-assembly
Q. Shi • D. Dong • K.J. Si • W. Cheng ()

Department of Chemical Engineering, Faculty of Engineering, Monash University, 3800 Clayton,
VIC, Australia
The Melbourne Centre for Nanofabrication, 151 Wellington Road, 3168 Clayton, VIC, Australia
e-mail: wenlong.cheng@monash.edu

DOI 10.1007/978-3-319-48081-7_6
110 Q. Shi et al.
6.1 Introduction
Plasmon is a universal light-matter interacting property that is present in metallic

materials originating from the collective oscillation of conduction electrons [1].
Unlike bulk metals, plasmon in a nanosized metal is confined, leading to new
physics and novel applications. The past two decades have witnessed a series
of breakthroughs in the chemical synthesis of plasmonic nanoparticles, allowing
for formulation of a plasmonic periodic table in which geometrical order of
nanoparticles is evident [2]. However, it remains to be an unresolved challenge how
to build arbitrary plasmonic architectures with these “artificial atoms” – similar to
the way that a chemist builds versatile molecules, polymers and crystals. What is
encouraging is that theory advances significantly ahead. The plasmon hybridization
theory predicts that the coupling among plasmonic nanostructures is analogous to
electron coupling in a molecule described by the group theory [3, 4].
It is also encouraging to see that soft ligand approach has the great potential
in controlling self-assembly of plasmonic nanoparticles and tailoring the cou-
pling strength [2, 5–8]. As far as highly ordered nanoparticle arrays (termed
as nanoparticle superlattice) are concerned, DNA, polymer and molecules have
been successfully utilised to generate high-quality superlattices. It led to novel
functional materials with unique properties and optical responses as compared
to their disordered counterparts, thus offering high promise towards tomorrow’s
technological innovations, with a variety of applications in sensors, spintronics,
lightweight and adaptive optoelectronic and energy devices [5, 9–14].
Free-standing plasmonic nanoparticle superlattices are a particular class of 2D
materials due to their integration of plasmonic coupling and mechanical compliance
into one multifunctional metamaterials system, enabling their novel applications in
sensing, nanophotonics, and energy [15–18]. Recently, we have established a new
concept in superlattice plasmonics – ‘Plasmene’, which is defined in analogy to
graphene as free-standing, one-particle-thick, superlattice sheets of nanoparticles
(“meta-atoms”) from the “plasmonic periodic table” [6]. This term was coined with
a purpose to classify a new class of 2D plasmonic materials, which complement the
currently 2D molecular materials such as graphene, MoS2 , etc.
This review is dedicated to discussions of plasmene nanosheets with regards to
their fabrication methodologies, properties and various applications. We will start by
briefly reviewing the wet chemistry synthesis methods for a few plasmonic nanopar-
ticle building blocks. This is followed by a discussion on plasmene-fabrication
techniques and highlighting the ubiquitous folding capability of 2D nanosheets
into unique 1D nanoribbon and 3D origami nanostructures. Then, we will discuss
in detail plasmonic coupling (gap and propagating modes), mechanical, and ion-
transporting properties of 2D plasmene nanosheet. Finally, we will cover various
successful applications of plasmene nanosheets in photonic devices, sensing, and
encryption. The initial results of plasmene nanosheets are encouraging, which
predicts vast opportunities in future multifunctional 2D designer metamaterials.
The rich choices in artificial plasmonic atoms in the periodic table may enable
rational design of future 2D plasmene materials with high quality and accuracy.
6 Fabrication, Properties and Applications of Plasmene Nanosheet 111
Complementary features of plasmene and other 2D materials may be combined

rationally by layer-by-layer, leading to unusual metamaterials with exotic properties
and functionalities.
6.2 Synthesis of Constituent Plasmonic Elements
Over the past three decades, substantial progresses have been made to intensively
investigate and develop synthesis approaches to yield high quality nanoparticles
with well-defined shapes and sizes. Although top-down approaches which com-
monly utilize lithographic techniques to break down bulk materials into desired
nanomaterials are available [19], they are generally faced with issues such as
dimensional restriction and high instrumentation costs. Nowadays, wet chemical
bottom-up approaches appear to the most common nanoparticle synthesis routes
due to its simplicity, low cost effectiveness and reliability to generate particles
with tailorable plasmonic properties that can be controlled by optimization of the
synthesis conditions. Wet chemical methods generally involve the reduction of metal
ions into atoms, followed by nucleation and growth into desired nanoparticle. The
geometry of the synthesized metal nanoparticle is dependent on the reaction time,
strength of the reductant, and the choice of capping material.
Here in this section, we will focus on the synthesis methodology and growth
mechanism of spherical and non-spherical plasmonic nanoparticles, particularly
focused on gold and silver metal nanoparticles. To date, a lot of different published
protocols can be found for synthesis of nanoparticles with various sizes and shapes
[1, 19–25]. We would like to note that the following approaches discussed in Sects.
6.2.1 and 6.2.2 are by no means a complete listing of all available synthetic methods
towards plasmonic nanoparticles, but rather the approaches that we have utilized for
synthesis of building blocks for plasmonic nanosheets described in Sect. 6.3.
6.2.1 Synthesis of Spherical Plasmonic Nanoparticles
Considering all conventional wet chemical methods, the landmark for synthesis
of gold nanoparticles is the widespread single phase reduction of hot chloroauric
acid by citrate in water to yield the lowest-energy spherical nanoparticles. This
was initially introduced by Turkevitch in 1951, and further refined by Frens in
1973 to yield monodispersed spherical nanoparticles over a tunable size range of
10–20 nm, controlled by the ratio of gold to citrate [26, 27]. The kinetics of this
reaction can be explained by a nucleation growth mechanism [28], where initially
the thermodynamically stable clusters from partial reduction process form seed
particles with a radius of 1.5 nm. The remaining ions are then attached on to the
electronic double layer of the seeds, and subsequent reduction into gold monomers
that grow specifically on the seed surfaces. However, as the spherical sizes reaches
112 Q. Shi et al.
40 nm and above, this approach loses size specificity, leading to low yields and
polydispersed particles.
To yield larger monodispersed spherical particles, seeding approaches are gen-
erally required, which involving two or multiple-steps enlargement of preformed
nanoparticle nuclei (seeds) [19, 29–30]. The final sizes are solely determined by the
size of the initial seeds and amount of gold ions added for growth. Recently, we
have utilized a modified seed-mediated growth approach to yield monodispersed
nanoparticles up to sizes of 100 nm [31, 32]. First, spherical seeds of sizes 15
or 35 nm are produced, followed by simultaneous slow addition of chloroauric
acid and a mixture of sodium citrate and ascorbic acid to this preformed seed at
room temperature. This resulted in individual nucleation of gold atoms on each seed,
forming “blackberry-like” particles. Subsequent heating to boiling point prompted
the atomistic reorganization that leads to nanospheres (NSs) with smooth surfaces.
6.2.2 Synthesis of Anisotropic Plasmonic Nanoparticles
Similar to the case of large nanospheres, seed-mediated growth approaches are

the most efficient and popular protocol to manipulate the growth conditions for
affecting the nucleation and growth of anisotropic plasmonic nanoparticles [21, 24].
A typical seeded growth approach involves a minimal of two reaction stages – a
nucleation stage followed by a growth stage. The nucleation stage is associated with
the formation of small spherical nanoparticle seeds via strong reducing agent such
as sodium borohydride. In the subsequent growth stage, the seeds will continuously
grow in a reaction solution containing metal salt, reducing agent and surfactants
into the desired shape. A milder reducing agent such as ascorbic acid is generally
used during growth process to prohibit nucleation of new seeds and ensures gold
reduction only occurs on surface of the preformed seeds [33].
Since the nucleation and growth stage are separated in seed-mediated growth
process, one can possibly achieve a menagerie of nanoparticle shapes depending
on the nucleation and growth mechanisms (Fig. 6.1) [20, 34–35]. During the
nucleation stage, the initial crystalline nanostructure of the seeds can lead to growth
of different shapes. For instance, a single crystalline seed can evolve into nanorods
whereas a multiply twinned seed resulted in growth of bipyramids under similar
growth conditions [36]. To minimize total surface energy, nanoparticles generally
grow in accordance to thermodynamically favoured pathways during growth stage.
In general, for all the low-index facets, f110g facets possess the highest surface
energies and are the most favourable surface for gold deposition [37]. This results in
disappearance of f110g facets, and evolution of final particles with surfaces exposed
by the lowest energy f111g facets. In order to obtain different morphologies with less
stable facets, different capping surfactants and shape directing agents are generally
required to change the order of facet free energies and manipulate the growth rate
along specific crystalline facets. Therefore, it is apparent that to obtain plasmonic
nanoparticles with desired shape, one must carefully control both the crystalline
seed structures and the relative growth rates along different facet directions.
Fig. 6.1 Seed mediated approach towards synthesizing plasmonic nanoparticles with different
shapes and sizes. By adjusting different reaction parameters, different nanoparticles such as (a)
nanorods, (b) nanobipyramids, (c) rhombic dodecahedral, (d) nanostars, (e) nanocubes and (f)
nanobricks can be obtained. Scale bar represents 100 nm for all
6.2.2.1 Gold Nanorods
For a typical gold nanorod (GNR), the ends facets are usually f111g, with
longitudinal surfaces being either f100g or f110g facets, or both [38]. To grow such
structures, the seed-mediated approach introduced by Murphy’s group is one of the
114 Q. Shi et al.
vastly adapted techniques by using cetyltrimethylammonium bromide (CTAB) as

surfactant and silver nitrate as a shape directing agent [39, 40]. It is widely accepted
that CTAB acts as a cationic surfactant that preferentially attaches to the f100g facets
of the gold seeds, forming a micellar bilayer. This blocks and retard growth along
the side facets in the transverse axis, thus forcing rapid growth and elongation on
the longitudinal f111g ends [39, 41, 42]. It was also reported that silver ions act as
a specific shape directing agent by depositing silver selectively on the f110g facets
of the gold nanorods, allowing one dimensional growth in the f100g direction for
single crystalline nanorods, thus providing a better control over the final aspect ratio
[40, 42–44].
6.2.2.2 Gold Nanobipyramids
Similar to gold nanorods, bipyramid gold nanoparticles (BNPs) can be prepared

with a similar growth approach, but by utilizing multiply-twinned seeds instead
of single crystalline seeds [45, 46]. Consequently, the formation of BNP can be
explained in a similar way to nanorods, where the growth is induced by the
selective adsorption of CTAB on f100g facets. As the nanostructure grows along
the twinning axis, steps appear and are stabilized by Ag (I), providing sites with
larger underpotential deposition for silver deposition [36]. This leads to the unique
pentatwinned BNPs, with side f100g facets tilted towards f111g facets. However,
such approach towards BNP synthesis generally produces spherical nanoparticles as
side products. Hence, a post-synthetic procedure to undergo purification, reshaping
and transformation of high yield BNP is required [45]. By overcoating the whole
sample with silver and redispersed with a controlled amount of surfactant, the silver-
coated BNPs tend to aggregate and precipitate, easily being separated from the
spherical impurities. By etching away silver with hydrogen peroxide, the same sized
BNPs as grown from the seed solution can be obtained.
6.2.2.3 Gold Rhombic Dodecahedral
For synthesis of rhombic dodecahedral (RD) nanoparticles, which can be referred

to as a single crystalline nanoparticle bounded by 12 identical f110g facets [47],
a seeded growth approach was developed using cetyltrimethyl chloride (CPC) as
surfactant and single crystalline CPC-capped seeds [48]. Here, competition between
gold ions reduction and CPC capping plays the important role towards RD shape
formation. The reduction of gold ions at relatively high ascorbic acid concentration
tends to occur on the f111g gold facets, leading to disappearance and formation
of f111g and f100g facets, respectively. CPC capping alters the surface energy of
surface facets in the order of f100g > f110g > f111g, thus stabilizing f110g facets
and allowing RD to exist as the final shape.
6.2.2.4 Gold Nanostars
Gold nanostars (NStrs) with spiky structures are some unique and interesting
nanoparticles that can be synthesized with a surfactant-free seed mediated approach
with multiply-twinned citrate seeds, chloride and silver ions [49]. Here, the
underpotential deposition of silver ions on certain crystal facets of the gold seeds
helps to improve the shape. By hindering certain crystallographic facets, it promotes
anisotropic growth on the twinned seeds and leads to the formation of multiple sharp
branches. Without AgC or twinned seed structure, polydispersed rods and spheres
will be formed. The role of chloride ions can be said to be similar to the effect of
iodide ion influencing anisotropic growth in nanorods [50].
6.2.2.5 Core-Shell Au@Ag Nanocubes/Nanobricks
Core-shell Au@Ag nanoparticles can be prepared via a seed-mediated growth by

using cetyltrimethylammonium chloride (CTAC) as surfactant [51]. Here, CTAC
plays important roles in both the nucleation and growth stages. In nucleation stage,
the oxidative etching power of Cl- in CTAC helps in dissolving of twinned seeds,
thus leading to formation of high uniformity single-crystal gold nanoseeds enclosed
with a mix of f100g and f111g facets. In the growth stage, CTAC aids in directing
fast silver growth on the f111g facets of the spherical cores, producing final Au@Ag
core-shell nanocubes (NCs) with surfaces dominated by the f100g facets [51, 52]. In
addition, by substituting the nucleation cores with gold nanorods, one can easily coat
a layer of silver on gold nanorods with the same growth mechanism to form Au@Ag
core shell nanobricks (NBs) [53, 54]. Similar to NCs, the presence of CTAC during
the growth process is crucial for a uniform growth of silver layer [55].
6.3 Fabrication of Plasmonic Nanosheet
Assembling nanoparticles into superlattice structures generates remarkable col-

lective properties that are different from their isolated counterparts [2, 5, 7, 56,
57]. The quality of the resultant plasmonic nanosheets can be controlled by
choosing different fabrication methods, coating different kinds of soft ligands to
tune the complex nanoscale forces (such as van der Waals, electrostatic, solvation-
depletion, friction/lubrication and capillary forces) acting between nanoparticles,
and manipulating the size and shape of individual nanoparticles [2, 5–8].
At present, different kinds of assembly methods have been investigated to
fabricate superlattice nanosheets on a 2D interface, including Langmuir-Blodgett
technique, layer-by-layer assembly, drying mediated assembly, self-assembly at air–
water or liquid–liquid interface, and external field-driven assembly (e.g. electronic
or magnetic field) [14, 15, 56–71]. Here, successful examples of synthesizing
116 Q. Shi et al.
high-quality plasmene nanosheets including the programmable assembly, micro-

hole defined assembly, liquid-liquid interfacial assembly, and air-water interfacial
assembly will be discussed in detail below.
6.3.1 Programmable Self-Assembly
To manipulate the nanoparticle binding interactions for coordination of nanopar-

ticle assembly, organic ligands such as surfactants, polymers or biomolecules are
generally used. In particular, biomolecules such as DNA is of major interest due to
its: (1) controllable DNA length from nanoscale to microscale, (2) unique Watson-
Crick base pairing forces between complementary strands, and (3) structural rigidity
and physicochemical stability [2, 5, 7]. As such, DNA can be used as a powerful
nanomaterial that provides a pathway towards “programming” the assembly of
plasmonic nanoparticles into ordered superlattice nanosheets.
By functionalizing nanoparticles with biomolecular ligands, the nanoparti-
cles can be manipulated and located precisely into superstructures owing to the
high strength sequence-specific base pairing and intermolecular recognition. For
instance, by using thiolated DNA to functionalize Au nanoparticles, free-standing
nanoparticle superlattice sheets with conformational flexibility and extraordinary
environmental stability can be fabricated (Fig. 6.2a) [72]. The controllable length
and sequence of DNA ligands allow fine adjustment of the interparticle spacing,
thickness and lattice symmetry of the obtained superlattice sheet. Silica was then
used as a scaffolding material to encapsulate superlattice structure and provide
it with flexibility. Besides DNA, functional protein with recognition capabilities
can also be utilized to fabricate tightly packed free-standing gold nanoparticle
monolayer films [73]. Using a-Synuclein (aS) proteins as ligands, the stable aS-
aS interactions allows the interconnection of aS-functionalized nanoparticles into
a free-floating film that can expand up to a 4-inch diameter. The resulting protein
interactions resulted in “-sheet conformation between aS molecules, thus providing
mechanical stability for the film.
6.3.2 Microhole-Confined Self-Assembly
Besides the traditional use of DNA as a programmable ligand, DNA has also been
used in an unconventional manner as a ‘dry ligand’ [74]. This assembly process
starts with functionalizing gold nanoparticles with 50 -thiolated single-stranded DNA
(ssDNA). The Au nanoparticle colloidal solution was then dropped onto a holey
substrate to allow entropically driven nanoparticle self-assembly via non-specific
DNA-DNA interactions (Fig. 6.2b). In particular, the spatial confinement of the
drying mediated process within a microhole allows rational control over the local
nucleation and growth of superlattices, leading to high uniformity of the assembled
nanosheets.
Fig. 6.2 Schematic diagram of fabrication of nanoparticle superlattice sheets from (a) pro-
grammable assembly; (b) microhole-confined self-assembly; (c) liquid-liquid interfacial assembly;
(d) air-water interfacial self-assembly (Modified from Refs. [6, 72, 74, 79])
After the full evaporation of water, the interactions between the ssDNA, which
acts a soft corona similar to that of alkyl ligands lead to discrete 2D free-standing
superlattice nanosheets with pinning of their boundaries onto the microhole edges.
Despite some defects found around the boundary of the nanosheets, the assembled
nanosheet possess ultimate thickness limit, well-ordered hexagonal packing symme-
try, millimetre scale lateral dimension, paper-like mechanical properties and strong
plasmonic coupling (Fig. 6.3a). In addition, since DNA possesses a wide range of
length controllability, the as fabricated nanosheets are able to achieve finely tailored
edge to edge interparticle spacing from 1 up to 20 nm. Such tunability range are
much greater in comparison to those of alkyl-based ligands, which only possess a
small tunability regime of 1.2–2.3 nm [75, 76].
118 Q. Shi et al.
Fig. 6.3 (a) 3D STEM nanosheet from Au nanosphere; (b) Mirror like NB superlattice at the
hexane-water interface; (c) optical microscope images of NC plasmene nanosheet; (d) optical
images of bilayered NS nanosheets and corresponding nanoparticles (Modified from Refs. [6, 74,
79, 82])
6.3.3 Liquid-Liquid Interfacial Assembly
The liquid-liquid interface is an ideal 2D platform to obtain well-defined assembly

architectures from nanoparticles due to the nature of the interfaces. Such assembly
is driven by the minimization of free energy, including contributions from thermal
fluctuations and interfacial potential [77, 78]. Generally, the assembly process is
accomplished by introducing hydrophobic ligands into the surface of nanoparticles
or adding amphiphilic solvent, then hydrophilic nanoparticles can be trapped to
the liquid–liquid interface [56, 75]. Since the nanosheets are formed at a 2D fluid
interface, the superlattices can be directly transferred to any arbitrary substrates
without the requirement of any peel-off steps.
By using liquid–liquid interfacial assembly method, we fabricated plasmonic
superlattice sheets on hexane-water interface with long-range orders from both
anisotropic core-shell Au@Ag NCs and NBs as shown in Fig. 6.2c [79]. The
assembly process start with mixing the particle colloid solution of Au@Ag NC or
Au@Ag NB with the same amount of hexane with 1-dodecanethiol (DDT). Then
nanoparticles spontaneously self-assembled at the water/hexane interface and the
mirror-like sheets began to appear and become larger with the injection of ethanol.
During this process, ethanol facilitated the ligand exchanges between DDT and
CTAC surfactants (capped during the early-synthesis procedure) by reducing the
interfacial surface tension. Since the weak binding CTAC is replaced by strong
binding DDT, the particles are rendered to be hydrophobic, resulting the deposition
of nanoparticles at the interfaces. Finally, the fluid interface is all covered with
mirror-like superlattice membrane that is as large as 13 cm [2] as shown in Fig.
6.3b. The obtained nanosheets show larger scale compared with other nanosheets
as summarized in Table 6.1. After the self-assembly completed, the superlattice
nanosheets were transferred to PDMS via horizontal lifting.
Table 6.1 Summary of different plasmene nanosheets
Interparticle spacing Orientational Freestanding
Method Building blocks Dimension Ligand control control or not References
Microhole-confined Au NS 2 um DNA Yes No Yes [74]
assembly
Air –water interfacial Au NS 15 um Thiol-functionalized Yes No Yes [80]
assembly PS
Hexane/water interfacial Au@Ag NC and NB 13 cm 1-Dodecanethiol Not mentioned No No [79]
self-assembly
Air –water interfacial Au NR A few Thiol-functionalized Not mentioned Yes Yes [83]
assembly micrometers PS
Air –water interfacial Au@Ag NC 3 mm Thiol-functionalized Not mentioned No Yes [6]
assembly PS
Air –water interfacial Au@Ag NC 3 mm Thiol-functionalized Not mentioned No Yes [81]
assembly PS
Air –water interfacial Au NS, RD, NStr 3 mm Thiol-functionalized Yes No Yes [32]
assembly PS
6 Fabrication, Properties and Applications of Plasmene Nanosheet
Air –water interfacial Au@Ag NC and NB 3 mm Thiol-functionalized Yes No Yes [54]

assembly PS
Air –water interfacial Au BNP 3 mm Thiol-functionalized Yes Yes Yes [46]
assembly PS
Air –water interfacial Au NS 3 mm Thiol-functionalized Yes No Yes [82]
assembly PS
119
120 Q. Shi et al.
6.3.4 Air-Water Interfacial Self-Assembly
We have recently introduced a generic approach towards the fabrication of plasmene

nanosheets by using a polymer based entropy driven self-assembly approach at
air-water interface [6]. In general, pristine plasmonic nanoparticles are usually
stabilized by molecular ligands such as CTAC, CTAB or citrate. However, these
weak-binding ligands are unable to provide sufficient protection against random
nanoparticle aggregation, which leads to the need of grafting stronger binding
ligands via a ligand exchange process [7]. Thus, we choose a generic polymer – thi-
olated polystyrene (PS) as the capping ligands to construct superlattice nanosheets.
Compared with normally used alkyl ligands, the PS ligands have longer length,
enabling wider tailoring of interparticle spacing especially for large nanoparticles.
More importantly, freestanding nanoparticle nanosheets could be generated by
assembling nanoparticle over a holey solid substrate. These properties make it
possible to obtain soft plasmonic nanosheets with large scale and provide the
possibility to generate diverse nanosheets with novel plasmonic properties by
carefully controlling the assembly process, and enable transfer the nanosheets to
any substrates, which are essential in further plasmonic applications (see details in
Sect. 6.6)
6.3.4.1 General Approach for Plasmene Nanosheets
This air-water interfacial assembly is first used to assemble monolayered gold

nanospheres, in which the structural features can be adjusted by tunning the gold
nanoparticle sizes [80]. The assembly process begins with a ligand exchange
process, in which citrate-reduced gold nanoparticle was obtained by spun down
and redispersed in a mixture of THF and PS-SH. Then, the PS-capped gold
nanoparticles were concentrated and finally dispersed in chloroform. The free-
standing monolayered superlattice nanosheets are fabricated by dropping the gold
nanoparticle-chloroform solution onto the surface of Milli-Q water in a plastic Petri
dish. Since the particles are capped by the hydrophobic ligands, the solution is
quickly spread and solidify into a gold reflection on the surface of a water. The
plastic Petri dish was covered to slow fluid flows and minimize solvent fluctuations.
Finally, the monolayered particle film is generated at the air-water interface due to
the rapid chloroform evaporation. The obtained thin particle sheets floating at the
water surface can be as large as 15 m in lateral dimensions with a thickness of
13 nm.
After the successful assembly of plasmene nanosheets from gold nanosphere,
we further modify and extend this method into a general approach by confining
the assembly process on top of a water droplet as shown in Fig. 6.2d. There are
two key factors during the assembling process, which are the high concentration of
the nanoparticles and the proper length of ligands. When the highly concentrated
PS functionalized nanoparticles was spread onto the water surface, small patch of
islands are firstly formed due to the rapid chloroform evaporation. As the slow water
evaporation, these small islands are gradually forced to assemble into large scale
nanosheets due to the strong surface tension. The final lateral dimension of such kind
of plasmene sheets can reach up to 3 mm with single particle thickness (Fig. 6.3c).
Moreover, the packing order of the nanosheets can be adjusted by tuning the
length of the ligands, which is the result of the competition between van der Waals
attraction forces and the steric hindrance of ligands [6]. As the fabrication process
is processed on holey grids, the freestanding nanosheets are directly generated
without any need for further substrate removal, and are transferable to any arbitrary
substrates.
This assembly approach could be customized using various “meta-atoms” ele-
ments, such as Au@Ag NCs and NBs [6, 54, 81], bilayered Au NS [82] (Fig. 6.3d),
Au RD and NStr [32] to form high-quality plasmene nanosheets. Both SEM and
TEM characterizations revealed the assembled nanosheets are monolayered with
different packing types. For example, the gold spheres, RDs and NStrs exhibited a
hexagonally close-packed ordering, while NC tend to pack side by side, isotropic
NBs are aligned horizontally with random packing directions.
6.3.4.2 Orientational Control of Nanoparticle Self- Assembly
Anisotropic plasmonic nanoparticles have been successfully used as building blocks

for fabricating superlattice nanosheets (as summarized in Table 6.1). Further
orientational control over their spatial ordering provides more room for exploring
the diversity of plasmonic nanosheets [46, 83, 84]. As an extension of our previous
success in fabrication of plasmonic nanosheets from anisotropic nanoparticles,
GNRs and BNPs can align into monolayered nanosheets with different orientations
through the drying-mediated self-assembly process at the air-water interface. The
GNRs nanosheets can be either horizontally (H-sheets) or vertically (V-sheets)
aligned as shown in Fig. 6.4a, while the BNPs nanoparticle liquid crystalline
superstructure (NLCS) can generate four distinct orientational packing orders,
which are horizontal alignment (H-NLCS), circular arrangement (C-NLCS), slanted
alignment (S-NLCS), and vertical alignment (V-NLCS) (Fig. 6.4b).
Fig. 6.4 TEM and SEM images of (a) H- and V-sheets of PS-GNRs nanosheets and (b) H-, C-,
S-, and V-nanosheets made from BNPs. The scale bars are 100 nm (Modified from Refs. [46, 83])
122 Q. Shi et al.
For GNR assembly, we found that the assembly pattern is mainly affect directly
by operating temperature and ligand length [83]. For example, the majority of the
sheets are H-sheets at ambient temperatures while the yield of V-sheets improves
as the temperature or annealing time under the chloroform atmosphere is increased.
The annealing time of nanosheets were controlled by sealing and capping the beaker
with a lid immediately to build a slow solvent evaporation environment.
For BNPs assembly, a more precise orientation control are developed [46].
We have thoroughly investigated the effects of experimental parameters including
particle size, capping ligand length, evaporation rate, and substrates on the resulting
packing orders of the assembled superlattice. First, by increasing humidity during
self-assembly, dominant V-NLCSs are formed due to slow-evaporation process.
This is because the slow water evaporation can provide enough time to let each
BNP find the lowest energy state position. Second, we found that reducing the BNP
concentration can obtain an NLCS with nearly all circular arrangements. This is due
to the insufficient number of BNPs to form a continuous film during self-assembly
process. In addition, only H-NLCS and V-NLCS packing orders were observed
from the free-standing nanosheets formed on holy copper grid. This comes from
the strong surface tension and the absence of particlesubstrate interaction on the
holy substrate.
The mechanism of formation of nanosheets with different orientations could
be explained by both the application of Tirado’s model and entropy-driven self-
assembly [85–87]. According to the Tirado’s model, the ratio of temperature to
shear viscosity is a key parameter which can affect the rotational and translational
diffusion behaviours of cylindrical objects. To obtain vertical alignment of nanopar-
ticles, the temperature needs to be increased to decrease the shear viscosity, resulting
a smaller diffusion constant. Therefore, the system would gain more time to reach
the most thermodynamically stable state by correcting the disordered structures
[83]. We also attribute the formation of nanosheets from elongated nanoparticles to
an entropy-driven self-assembly confined to the 2D air-water interface, which can
be explained by Onsager’s model. According to Onsager’s model, there is a com-
petition between rotational and translational entropy during the assembly of hard
rods [86, 88]. Rotational entropy increased as the orientation possibility of the rods
increased, which decreases as alignment increases. Translational entropy increased
as the excluded volume decreased. In a dilute solution, the rods are possible to take
on arbitrary orientations and positions to maximize entropy. However, in a high-
concentration solution, both the translational and rotational freedom are restricted,
some of their orientational freedom was sacrificed to maximize the translational
freedom, so a parallel structure are formed. Although BNPs have a similar elongated
shape to nanorods, our BNP particles have two sharp tips and are capped with soft
ligands, which is different with the rigid rod or rod-like shapes in Onsager’s model.
Therefore, the excluded volumes need to be considered under confined environment,
in which the BNP particles are coated by an adaptive organic corona to form soft
nanocylinders that have an incompressible organic layer within thin chloroform
films. We found that the value of excluded volumes from V-NLCS is smaller than
that from S-NLCS and H-NLCS, indicating larger translational entropy. As a result,
more free volume is left to the system, facilitating a close packed V-NLCS.
6.4 Patterning of 2D Nanosheet into 1D Nanoribbon

and 3D Origami
Inspired by traditional Japanese art of paper folding which fashioned flat sheets of
papers into beautiful three dimensional origami designs, patterning of a planar 2D
nanosheet into different shapes or folding into 3D structures may, in principle, lead
to intriguing new plasmonic functionalities. However, patterning of such structures
with desired configuration still remains a major issue for modern nanotechnology
due to limitations such as complicated instrumentations and insufficient feature
controllability [89, 90].
Using a focused-ion beam (FIB) lithography-based approach, we have recently
demonstrated the first patterning and folding of 2D plasmene nanosheets into 1D
nanoribbons and 3D origami structures (Fig. 6.5a) [6]. Free-standing nanoribbons
with smooth surfaces and widths ranging from 100 to 500 nm were successfully
patterned. Analogous to graphene nanoribbons, our plasmene nanoribbons exhibited
width-dependent plasmonics, in which the resonance peak and peak intensity red-
shifted and grows linearly with increasing width, respectively. This can be attributed
to the increasing number of bonding plasmons interacting with each other as width
increases, leading lower resonant energy and spectral red-shift.
This FIB-milling can be further programmed in such a way that it only selectively
strips away surface binding polystyrene ligands to induce localized stresses. This
disrupts the internal balanced nanoscale forces, with van der Waals forces becoming
dominant and result in self-folding of the nanosheet. This localized stress-driven
self-folding can be designed in a variety of ways as hinges at predetermined
locations for folding into intricate 3D origami nanostructures such as heart, cube,
hexagon, diamond, and even mimicking a ‘flying bird’ as shown in Fig. 6.5b.
Fig. 6.5 (a) Folding of 2D plasmene nanosheets into 1D nanoribbons and 3D origami structures;
(b) A series of SEM images depicting the flapping of the wings of a plasmonic nanobird (Modified
from Ref. [6])
124 Q. Shi et al.
These folded 3D origami structures display intriguing folding-dependent optical

properties that are previously unattainable in 2D nanosheets. In particular, the
reduction in plasmonic interactions due to incoherent interparticle folding as a result
of folding leads to blue shifting of spectra with narrowing bandwidth. Undoubtedly,
such origami-inspired self-folding plasmene nanosheets would suggest an attractive
strategy towards facilitating the development of future smart functional devices with
unique plasmonic functionalities.
6.5 Properties of Plasmonic Nanosheets
Plasmonic nanosheet exhibits novel optical properties resulting from the interactions
between surface plasmons of individual particles [91]. More specifically, when
the incident light interacts with plasmonic nanoparticles, the free electrons would
oscillate resonantly in the finite volume around the nanoparticles leading to a strong
electromagnetic field, which is termed as localized surface plasmon resonance
(LSPR) [19]. Once the nanoparticles were assembled into superlattices, optical
properties would be altered due to the interaction and coupling of nanoparticles
plasmon with each other [92]. Therefore, plasmon peak shifts and various colours
could be observed by tunning the surrounding medium, size, shapes and composi-
tion of the metal particles.
6.5.1 Gap Plasmon Mode
Strong resonance peak could be found for Au@Ag core-shell nanocube plasmene
nanosheet which stems from the dipole-dipole coupling [6, 32, 46]. Comparing with
the solution spectra, the nanosheet shows a broader band and a red shift of resonance
peak on the spectra indicating the effect of interparticle coupling. To be noted, there
was a minor red shift for the Au@Ag core-shell nanocube when the thickness
of silver shell increased, which contributes from the scattering properties of
enlarger individual particles. From the simulation study on the plasmene nanosheet,
strong electromagnetic fields are highly localized in the interparticle gaps [6]. The
plasmonic coupling between the nanoparticles enhanced the localized field, and the
capacitive coupling between the adjacent nanoparticles support the plasmonic gap
resonance which is corresponding to the gap plasmon modes [93].
6.5.1.1 Particle Shape and Size Effect
The particle shape and size effect on plasmonic properties of plasmene nanosheet
was studied by comparing gold nanospheres, rhombic dodecahedrals and gold
nanostars [32]. Different to the smooth surface of gold spheres, rhombic dodecahe-
drals have 14 vertices and nanostars have numerous sharp tips. These sharp features
enhance the localized electromagnetic field under the knowledge of electromagnetic
theory and lighting rod effect [94, 95]. Thus, largest enhancement factor was found
for the nanostars sheet with gold spheres sheet showing weaker electromagnetic
field enhancement which is in consistent with both experimental observation and
simulation study.
For anisotropic elongated shape nanoparticles such as Au nanorods, it can
show different plasmon resonance peaks due to different alignments in the two
dimensional sheet [83]. It is found that Au nanorods superlattice nanosheet possess
two resonant peaks in the horizontal aligned sheet: (1). Transverse surface plasmon
resonance (2). Longitudinal surface plasmon resonance, while only the transverse
resonance peak could be observed in the vertical aligned sheet. This is due to the
absence of excited longitudinal plasmons within the vertical aligned nanoparticles.
These resonance peaks is red-shifted compared with that in solution phase which is
attributed to the interparticle coupling between nanorods. Similar optical behaviour
is found in our gold nanobipyramid sheet. Vertical aligned bipyramid nanosheet
only possesses the transverse peak as horizontal alignment, circular arrangement,
and slanted arrangement nanosheets show two peaks [46].
On the other hand, when the size of particles increase, van der Waals attraction
forces between the nanoparticles would become stronger, leading to the decrease in
the interparticle spacing [32]. Subsequently, the electric field would get enhanced
due to the increase in adjacent nanoparticles coupling, resulting a red shift on the
plasmon resonance peak.
6.5.1.2 External Strain Effect
Stretch-induced plasmonic property study further supports the investigation of gap

plasmon mode of our plasmene nanosheet system [79]. The superlattice sheet was
lifted up with a PDMS sheet and spin coated with another layer of PDMS on top
of sheet surface before conducting strain induced study. Trends of blue shift were
clearly observed on the plasmon resonance peak with increasing strain up to 35 %
due to the reduction of dipole-dipole coupling. The increase in the strain also led to
the decrease of plasmon resonance intensity and broadening of the resonance peak.
Such optical behaviour could be explained by the decay of the near-field coupling
strength between adjacent nanoparticles that is owing to the larger gap between
particles under the strain and less interacting particles over the beam spot area.
Remarkably, such responses can be fully reverted after releasing the applied strain.
Furthermore, the stretch induced plasmonic properties were found to be particles
shape dependent. For example, under 35 ˙ 1.5 % strain, 36 ˙ 5.0 nm blue-shift
was found for nanobrick superlattice sheet while only 12 ˙ 1.0 nm blue shift
could be observed for nanocube superlattice sheet.
126 Q. Shi et al.
6.5.2 Propagating Plasmon Mode
In addition to gap plasmon mode, plasmene also supported the propagating plasmon
mode as shown by both simulation and experimental work. From the simulation
results, a cylindrical wave like plasmon excitation was found to propagate across
the plamene nanosheet when the light falls on the small part of the centre of
the sheet [6]. The decay time of 11 fs for our plasmene nanosheet is found
through numerical calculation which is much longer than the decay time for isolated
metal nanoparticles. To further demonstrate the propagating nature of our plasmene
nanosheet, we transfer nanocube plasmene onto side polished D-shape optical
fibre to form a hybrid fibre-to-plasmene coupler. The nature of the propagating
surface plasmon was investigated by analyse the polarization of output light. It
was demonstrated that plasmene sheet selectively supports transverse magnetic
propagating waves and the electric field is perpendicular to the sheet surface. This
observation is because that metal-dielectric interface of the coupler forbids the
transverse electric surface waves. As a result, optical photons are only coupled into
vertically polarized surface plasmon and propagate across the nanosheet surface (as
shown in Fig. 6.6a).
Fig. 6.6 (a) Polar image measured at 532 nm from polished D-shape fibers with/without plasmene
sheet; (b) Diode like current curve obtained from plasmene bilayer structure; (c) Plasmene based
adhesive SERS substrate for chemical detection on a coin surface; (d) Left: Linear relationship
between SERS signal and drug concentration. Right: Drug barcode developed according to the
SERS spectra for Benzocaine; (e) Plasmene nanosheet as dual-coded anti-counterfeit security label
(Modified from Refs. [6, 32, 54, 81, 82])
6.5.3 Mechanical Properties
Though our plasmene nanosheet is only one particle thick, it shows strong tensile
strength due to the polystyrene ligand-ligand entanglement. AFM nanoindentation
experiments were carried to found out the mechanical properties of our plasmene
nanosheet. A series of indentation with increasing force at the centre of the
nanosheet was performed until the nanosheet ruptured. By fitting the force displace-
ment curve with both numerical model and theoretical model, a 1GPa Young’s
modulus for the nanocube based plasmene is derived which is similar to the alkyl-
nanoparticle superlattice sheet that evidently proves our plasmene nanosheet has
robust mechanical property [6]. In addition, the mechanical property of plasmonic
nanosheet could also be altered, which is shown by adjusting the DNA ligand length
[74]. A linear relationship between the force and displacement was found through
the AFM indentations. Through the investigation, the spring constant of nanosheet
is found to be inversely proportional to the ligand length which indicates that ligands
are governing the mechanical properties of nanosheets rather than the nanoparticle
cores. Remarkably, a Young’s modulus of 6.49 ˙ 1.57 GPa was obtained, where the
nanosheet was fabricated by using microhole-confined self-assembly process with
DNA as ‘dry ligand’.
6.5.4 Asymmetric Ionic Transport Behaviour
By fabricating one plasmene layer on top of another plasmene nanosheet, we

are able to fabricate free-standing superlattice nanosheet with bilayer structure.
In conjunction with an electrochemical system, novel asymmetric ionic transport
behaviour was found for this free-standing bilayer structure [82]. Diode-like current
curve was observed in the potential range from -1V to 1V for bilayer nanosheet,
which is different with the linear curve obtained by using monolayer nanosheet,
illustrating the ionic selectivity (Fig. 6.6b). The transport of the positive ions through
the cone-like nanochannel from the plasmene nanosheet with larger nanoparticles
to the plasmene with smaller nanoparticles was experimentally proved by the
larger current observation when applying positive scanning potential. In addition,
asymmetric ion transport behaviour could be further enhanced by adjusting both
nanoparticles size and ligand length, resulting a maximum current rectification
ratio of up to 13. This novel type of our plasmene bilayer structure demonstrates
chances towards nanoparticle based devices fabrication with excellent asymmetric
ionic selectivity.
128 Q. Shi et al.
6.6 Application of Plasmonic Nanosheets
The novel properties of plasmonic nanosheets enable its potential applications in the
field of sensors, optoelectronics and biomedicines [96–99]. Among those, surface
enhanced Raman scattering (SERS) chemical sensing is one of the most promising
applications [84, 100, 101]. With the detection limit down to single molecular
level, SERS serves as a powerful tool for application in chemistry, physics, forensic
and even life science. As shown in the previous section, the plasmene nanosheet
can localize and enhance the electric field due to the plasmon coupling between
the adjacent nanoparticles which make our plasmene nanosheet as ideal candidate
for SERS based detection. In conjunction with SERS, appealing applications of
plasmene nanosheet could be developed. In addition, our plasmene nanosheet has
the ability to customize the plasmonic property by tunning the design parameter
and it is free-standing with robust mechanical strength which further diverse its
applications [6, 32]. Here, the applications of our plasmene nanosheet in membrane
resonator, SERS substrates, quantitative drug identification, and anti-counterfeit
security label are mainly discussed.
6.6.1 Plasmene Based Membrane Resonator
As mentioned in Sect. 6.5, the plasmene nanosheet is mechanical strong which

enable us to use plasmene as a mechanical membrane resonator. The fundamental
resonance frequency of our PS capped nanoparticles superlattice sheet ranges from
40 to 220 kHz with varied sizes [6]. The highest measured quality factors of the
plasmene resonator can achieve 133 in air which is about 1 order of magnitude
larger than those of alky-nanoparticle based superlattice sheets [102]. This factor is
proportional to the ratio of energy stored in the resonator to the energy dissipated
per cycle and could be further enhanced in vacuum by avoiding air damping.
6.6.2 Plasmene Nanosheet as Next Generation SERS Substrate
The fabrication of plasmonic nanoparticle based SERS substrate has been drawn
much attention recently. Comparing with the nanoparticles randomly ordered
substrate, the self-assembled superlattice nanosheet could give better signal due to
the strong electromagnetic field enhancement(hot spots) provided by the nanometre
scale interparticle gap. Typically, rigid-opaque substrates, such as silicon and glass,
are used for the 2D nanosheet assembling which limits the ability of substrate to
contact with topological complex surfaces [103, 104]. Such substrate supported
SERS substrate could only work on the flat surface. However, the optical semi-
transparent, free-standing and mechanical flexible nature of our plasmene nanosheet
allows to transfer plasmene nanosheet to any arbitrary surfaces as an adhesive

SERS substrate. Moreover, SERS hot spots could be engineered to be excited under
specific wavelength by controlling the nanoparticles size and shape.
The Au@Ag nanocube/nanobrick nanosheets have been successfully fabricated
with robust protocol in our group. The preparation steps are as follows: first,
the nanosheets on copper grid are pressed with a PDMS elastomer sheet under
a moderate pressure for several minutes and then the sacrificial copper grid is
peeled off leaving a 7 7 m [2] square patterned nanosheet on PDMS. Then
the patterned nanosheets are able to be transferred to any arbitrary surface with
the stamping process due to the low surface tension of PDMS [54]. To utilize the
capability of plasmene based adhesive SERS substrate, various surfaces such as
Malaysian banknote (fibrous), Australian banknote (polymer) and Australian coin
(rough metal) were used. After the deposited 4-ATP (which is a Raman probe with
apparent SERS characteristic bands) solution totally dried, plasmene nanosheets are
stamped onto the treated region to have the intimate contact and exposed with a
514 nm excitation wavelength laser to acquire the SERS spectra. The results show
that the characteristic peak of 4-ATP could not be observed on the plain or 4-ATP
covered region, but obvious for the region attached with plasmene nanosheet due
to the signal enhancement (Fig. 6.6c). Furthermore, our plasmene nanosheet shows
multiple phase’s detection capability [54]. Both liquid phase and vapour phase of
analyst could be detected by using the plasmene based SERS substrate. Also, our
plasmene nanosheet has high structural homogeneity which can lead to uniform hot
spots distribution, therefore, resulting uniform SERS enhancement. As expected,
the SERS spectra from 20 different spots exhibited excellent signal uniformity and
reproducibility which proves the reliability of our plasmene nanosheet as adhesive
SERS substrate. Besides that, the plasmon modes and near field distribution can
be engineered by adjusting the sizes, shapes of building particles. All these distinct
functionalities enabled us to use the plasmene nanosheet as unique and appealing
SERS substrate towards chemical identification.
6.6.3 Quantitative Drug Identification Using Plasmene

Nanosheet
Conventional drug identification methods such as enzyme-linked immunosorbent

assay for screening, followed by verification and quantification in clinical labo-
ratories via techniques such as gas or liquid chromatography coupled with mass
spectrometry are time consuming and require multiple instrumentations, qualified
personnel and complex sample preparations [81, 105–108]. Although SERS is a
non-destructive technique offering detection limit down to single molecular level,
the conventional SERS substrate is normally opaque and rigid with poor signal
reproducibility resulting limited application. In contrast, the plasmene nanosheets
allows rapid and reliable drug detection due to the structural homogeneity and
uniformly generated Raman hot spots [54, 81].
130 Q. Shi et al.
The preparation and detection steps are simple and easy [81]: typically, a
short time oxygen plasma treatment is used onto plasmene nanosheet before test
to remove the surface polystyrene ligand to make the analyte being easier to
be absorbed; and then, after transferring the plasmene nanosheets onto silicon
substrates, a swipe based approach is used to load illicit drugs; lastly, the samples
are subjected into a Raman system to acquire the Raman spectra. By analysing
the unique vibration fingerprint of illicit drugs on the SERS spectra, the types
of drug can be easily identified. The plot of the dominant peak intensity versus
the drug concentration can be used to quantify the drug concertation. At the low
concentration regime, it is found out that the intensity, which is proportional to the
number of drug molecular absorbed on the nanosheet, shows a linear relationship
with the drug concentration. Several drugs including benzocaine, acetaminophen,
aspirin and ibuprofen were utilized to show the versatility of this quantification
approach [81]. The linear quantitative behaviour on SERS spectra within the low
concentration region was found for each drugs. Remarkably, the overdose limit
were all found to be within the liner region which enable its real world application
for drug overdose diagnose. To accomplish the real world drug detection, a drug
barcode system (Fig. 6.6d) is developed to identify drugs in a mixture. By comparing
the SERS spectra with the standard reference drug barcode, the composition of the
drug mixtures could be identified.
6.6.4 Plasmene Based Anti-counterfeit Security Label
The application of plasmene nanosheet is further extended to the anti-counterfeit

area. Utilizing the new class of plasmene as security labels can provide a new route
towards anti-counterfeit purpose. Here, plasmene nanosheet is dually coded with
plasmonic signatures and SERS fingerprints to serve as a unique anti-counterfeit
security label (Fig. 6.6e) [32].
The combination of plasmonic code and SERS barcode led to virtually unlimited
and unique authentication code. The plasmonic code is corresponding to signature
plasmon resonance peak of the plasmene nanosheets, while SERS coding is
developed by acquiring SERS spectra of the deposited Raman dye under specific
laser excitation wavelength and converting the molecular vibration fingerprint into
a SERS barcode. By altering the choices of different Raman dyes, such as 4-ATP,
malachite green, 4-mercaptobenzoic acid, 1-octadecanethiol and nile red, different
SERS barcodes can be developed for the same plasmene nanosheet.
The authentication codes are extremely challenged to be decrypted or forged
without knowing the nanoparticle shapes, sizes, choice of dyes and laser excitation
wavelength. After coating with a thin layer of PDMS, it was found to remain
a stable SERS performance after 3 months and survive after 100 folding cycle
which illustrates the stability and durability of our dual coded security label. This
dual coded security labels can offer unlimited optical signatures by engineering
the shapes and sizes of building blocks and numerous SERS codes by varying
the choices of Raman dyes. Besides that, the SERS barcode cannot be read
out without the enhancement of plasmene nanosheet with right laser wavelength.
Therefore, the long term stable and highly durable plasmene nanosheet with
excellent signal uniformity and reproducibility has great potential in future anti-
counterfeit technology.
6.7 Conclusion
This review highlighted recent advances in plasmene nanosheets which are a new
class of 2D soft metamaterials. Such novel materials enable simultaneous control
over plasmonic coupling and mechanical compliance in one multifunctional system.
The combination of both functionalities leads to novel applications in sensing,
nanophotonics, mechanical resonators and ion-transport regulations. For example,
the utilisation of plasmene nanosheets as soft SERS substrates enabled direct
identification of trace amount of chemicals sitting on real-world substrates – a
capability not possible to achieve with conventional rigid opaque gold standard
Klarite substrate. Soft organic ligands including DNA, polymers and molecules
play crucial roles in controlling plasmene growth and in regulating electromagnetic
coupling among plasmonic nanoparticles.
While past several years have witnessed exciting progress in fabrications and
applications of plasmene nanosheets, a few challenges remain to translate them
into real-world materials, such as scalability of production at macroscopic scale;
generation of plasmene nanosheets from multiple types of nanoparticles; accurate
plasmonic bandgap engineering by soft ligands. To overcome these materials
challenge requires multidisciplinary collaboration in developing viable simple yet
high-throughput, and large scale methodologies that can accommodate nanopar-
ticles with different size and shapes with finely and widely tunable materials
properties.
Acknowledgment These researches were financially supported under the Australian Research
Council’s Discovery projects funding schemes DP150103750 and DP140100052. This work was
performed in part at the Melbourne Centre for Nanofabrication (MCN) in the Victorian Node of
the Australian National Fabrication Facility (ANFF).
References
1. Xia Y, Halas NJ (2005) Shape-controlled synthesis and surface plasmonic properties of

metallic nanostructures. MRS Bull 30(5):338–348
2. Tan SJ, Campolongo MJ, Luo D, Cheng W (2011) Building plasmonic nanostructures with
DNA. Nat Nanotechnol 6(5):268–276
3. Prodan E, Radloff C, Halas NJ, Nordlander P (2003) A hybridization model for the plasmon
response of complex nanostructures. Science 302(5644):419–422
132 Q. Shi et al.
4. Nordlander P, Oubre C, Prodan E, Li K, Stockman MI (2004) Plasmon hybridization in

nanoparticle dimers. Nano Lett 4(5):899–903
5. Kinge S, Crego-Calama M, Reinhoudt DN (2008) Self-assembling nanoparticles at surfaces
and interfaces. Chemphyschem Eur J Chem Phys Phys Chem 9(1):20–42
6. Si KJ, Sikdar D, Chen Y, Eftekhari F, Xu Z, Tang Y, Xiong W, Guo P, Zhang S, Lu Y, Bao
Q, Zhu W, Premaratne M, Cheng W (2014) Giant plasmene nanosheets, nanoribbons, and
oirigami. ACS Nano 8(11):11086–11093
7. Klinkova A, Choueiri RM, Kumacheva E (2014) Self-assembled plasmonic nanostructures.
Chem Soc Rev 43(11):3976–3991
8. Liao J, Blok S, van der Molen SJ, Diefenbach S, Holleitner AW, Schönenberger C, Vladyka A,
Calame M (2015) Ordered nanoparticle arrays interconnected by molecular linkers: electronic
and optoelectronic properties. Chem Soc Rev 44(4):999–1014
9. Quan Z, Fang J (2010) Superlattices with non-spherical building blocks. Nano Today
5(5):390–411
10. Mattoussi H, Rubner MF, Zhou F, Kumar J, Tripathy SK, Chiang LY (2000) Photovoltaic
heterostructure devices made of sequentially adsorbed poly(phenylene vinylene) and func-
tionalized C60. Appl Phys Lett 77(10):1540–1542
11. Koktysh DS, Liang X, Yun BG, Pastoriza-Santos I, Matts RL, Giersig M, Serra-Rodríguez
C, Liz-Marzán LM, Kotov NA (2002) Biomaterials by design: layer-by-layer assembled ion-
selective and biocompatible films of TiO2 nanoshells for neurochemical monitoring. Adv
Funct Mater 12(4):255–265
12. Franke ME, Koplin TJ, Simon U (2006) Metal and metal oxide nanoparticles in chemiresis-
tors: does the nanoscale matter? Small 2(1):36–50
13. Zhou M, Lou XW, Xie Y (2013) Two-dimensional nanosheets for photoelectrochemical water
splitting: possibilities and opportunities. Nano Today 8(6):598–618
14. Sun S, Murray CB (1999) Synthesis of monodisperse cobalt nanocrystals and their assembly
into magnetic superlattices (invited). J Appl Phys 85(8):4325–4330
15. Dong A, Chen J, Vora PM, Kikkawa JM, Murray CB (2010) Binary nanocrystal superlattice
membranes self-assembled at the liquid-air interface. Nature 466(7305):474–477
16. Dong A, Ye X, Chen J, Murray CB (2011) Two-dimensional binary and ternary nanocrystal
superlattices: the case of monolayers and bilayers. Nano Lett 11(4):1804–1809
17. Cargnello M, Johnston-Peck AC, Diroll BT, Wong E, Datta B, Damodhar D, Doan-Nguyen
VVT, Herzing AA, Kagan CR, Murray CB (2015) Substitutional doping in nanocrystal
superlattices. Nature 524(7566):450–453
18. Talapin DV, Lee J-S, Kovalenko MV, Shevchenko EV (2010) Prospects of colloidal nanocrys-
tals for electronic and optoelectronic applications. Chem Rev 110(1):389–458
19. Eustis S, El-Sayed MA (2006) Why gold nanoparticles are more precious than pretty gold:
noble metal surface plasmon resonance and its enhancement of the radiative and nonradiative
properties of nanocrystals of different shapes. Chem Soc Rev 35(3):209–217
20. Lu X, Rycenga M, Skrabalak SE, Wiley B, Xia Y (2009) Chemical synthesis of novel
plasmonic nanoparticles. Annu Rev Phys Chem 60(1):167–192
21. Grzelczak M, Perez-Juste J, Mulvaney P, Liz-Marzan LM (2008) Shape control in gold
nanoparticle synthesis. Chem Soc Rev 37(9):1783–1791
22. You H, Yang S, Ding B, Yang H (2013) Synthesis of colloidal metal and metal alloy
nanoparticles for electrochemical energy applications. Chem Soc Rev 42(7):2880–2904
23. Csáki A, Thiele M, Jatschka J, Dathe A, Zopf D, Stranik O, Fritzsche W (2015) Plasmonic
nanoparticle synthesis and bioconjugation for bioanalytical sensing. Eng Life Sci 15(3):266–
275
24. Kretschmer F, Mühlig S, Hoeppener S, Winter A, Hager MD, Rockstuhl C, Pertsch T,
Schubert US (2014) Survey of plasmonic nanoparticles: from synthesis to application. Part
Part Syst Charact 31(7):721–744
25. Bigall NC, Eychmüller A (2010) Synthesis of noble metal nanoparticles and their non-ordered
superstructures. Philos Trans R Soc London Math Phys Eng Sci 368(1915):1385–1404
26. Turkevich J, Stevenson PC, Hillier J (1951) A study of the nucleation and growth processes
in the synthesis of colloidal gold. Discuss Faraday Soc 11:55–75
27. FRENS G (1973) Controlled nucleation for the regulation of the particle size in monodisperse
gold suspensions. Nat Phys Sci 241:20–22
28. Wuithschick M, Birnbaum A, Witte S, Sztucki M, Vainio U, Pinna N, Rademann K,
Emmerling F, Kraehnert R, Polte J (2015) Turkevich in new robes: key questions answered
for the most common gold nanoparticle synthesis. ACS Nano 9(7):7052–7071
29. Brown KR, Natan MJ (1998) Hydroxylamine seeding of colloidal Au nanoparticles in
solution and on surfaces. Langmuir 14(4):726–728
30. Jana NR, Gearheart L, Murphy CJ (2001) Seeding growth for size control of 540 nm
diameter gold nanoparticles. Langmuir 17(22):6782–6786
31. Ziegler C, Eychmüller A (2011) Seeded growth synthesis of uniform gold nanoparticles with
diameters of 15300 nm. J Phys Chem C 115(11):4502–4506
32. Si KJ, Sikdar D, Yap LW, Foo JKK, Guo P, Shi Q, Premaratne M, Cheng W (2015) Dual-
coded plasmene nanosheets as next-generation anticounterfeit security labels. Adv Opt Mater
3(12):1710–1717
33. Jana NR, Gearheart L, Murphy CJ (2001) Evidence for seed-mediated nucleation in the
chemical reduction of gold salts to gold nanoparticles. Chem Mater 13(7):2313–2322
34. Sun Y, Xia Y (2002) Shape-controlled synthesis of gold and silver nanoparticles. Science
298(5601):2176–2179
35. Grochola G, Snook IK, Russo SP (2008) Influence of substrate morphology on the growth of
gold nanoparticles. J Chem Phys 129(15):154708
36. Liu M, Guyot-Sionnest P (2005) Mechanism of silver(I)-assisted growth of gold nanorods
and bipyramids. J Phys Chem B 109(47):22192–22200
37. Wang ZL (2000) Transmission electron microscopy of shape-controlled nanocrystals and their
assemblies. J Phys Chem B 104(6):1153–1175
38. Murphy CJ, Sau TK, Gole AM, Orendorff CJ, Gao J, Gou L, Hunyadi SE, Li T (2005)
Anisotropic metal nanoparticles: synthesis, assembly, and optical applications. J Phys Chem
B 109(29):13857–13870
39. Jana NR, Gearheart L, Murphy CJ (2001) Wet chemical synthesis of high aspect ratio
cylindrical gold nanorods. J Phys Chem B 105(19):4065–4067
40. Busbee BD, Obare SO, Murphy CJ (2003) An improved synthesis of high-aspect-ratio gold
nanorods. Adv Mater 15(5):414–416
41. Nikoobakht B, El-Sayed MA (2001) Evidence for bilayer assembly of cationic surfactants on
the surface of gold nanorods. Langmuir 17(20):6368–6374
42. Gao J, Bender CM, Murphy CJ (2003) Dependence of the gold nanorod aspect ratio on the
nature of the directing surfactant in aqueous solution. Langmuir 19(21):9065–9070
43. Pérez-Juste J, Pastoriza-Santos I, Liz-Marzán LM, Mulvaney P (2005) Gold nanorods:
synthesis, characterization and applications. Coord Chem Rev 249(17–18):1870–1901
44. Jana NR, Gearheart L, Murphy CJ (2001) Seed-mediated growth approach for shape-
controlled synthesis of spheroidal and rod-like gold nanoparticles using a surfactant template.
Adv Mater 13(18):1389
45. Li Q, Zhuo X, Li S, Ruan Q, Xu Q-H, Wang J (2015) Production of monodisperse
gold nanobipyramids with number percentages approaching 100% and evaluation of their
plasmonic properties. Adv Opt Mater 3(6):801–812
46. Shi Q, Si KJ, Sikdar D, Yap LW, Premaratne M, Cheng W (2016) Two-dimensional bipyramid
plasmonic nanoparticle liquid crystalline superstructure with four distinct orientational
packing orders. ACS Nano 10(1):967–976
47. Personick ML, Langille MR, Zhang J, Harris N, Schatz GC, Mirkin CA (2011) Synthesis
and isolation of f110g-faceted gold bipyramids and rhombic dodecahedra. J Am Chem Soc
133(16):6170–6173
48. Niu W, Zheng S, Wang D, Liu X, Li H, Han S, Chen J, Tang Z, Xu G (2009) Selective
synthesis of single-crystalline rhombic dodecahedral, octahedral, and cubic gold nanocrystals.
J Am Chem Soc 131(2):697–703
134 Q. Shi et al.
49. Hsiangkuo Y, Christopher GK, Hanjun H, Christy MW, Gerald AG, Tuan V-D (2012)
Gold nanostars: surfactant-free synthesis, 3D modelling, and two-photon photoluminescence
imaging. Nanotechnology 23(7):075102
50. Grzelczak M, Sánchez-Iglesias A, Rodríguez-González B, Alvarez-Puebla R, Pérez-Juste J,
Liz-Marzán LM (2008) Influence of iodide ions on the growth of gold nanorods: tuning tip
curvature and surface plasmon resonance. Adv Funct Mater 18(23):3780–3786
51. Ma Y, Li W, Cho EC, Li Z, Yu T, Zeng J, Xie Z, Xia Y (2010) Au@Ag coreshell nanocubes
with finely tuned and well-controlled sizes, shell thicknesses, and optical properties. ACS
Nano 4(11):6725–6734
52. Tsuji M, Miyamae N, Lim S, Kimura K, Zhang X, Hikino S, Nishio M (2006) Crystal
structures and growth mechanisms of Au@Ag coreshell nanoparticles prepared by the
microwavepolyol method. Cryst Growth Des 6(8):1801–1807
53. Okuno Y, Nishioka K, Kiya A, Nakashima N, Ishibashi A, Niidome Y (2010) Uniform
and controllable preparation of Au–Ag core–shell nanorods using anisotropic silver shell
formation on gold nanorods. Nanoscale 2(8):1489–1493
54. Chen Y, Si KJ, Sikdar D, Tang Y, Premaratne M, Cheng W (2015) Ultrathin plasmene
nanosheets as soft and surface-attachable SERS substrates with high signal uniformity. Adv
Opt Mater 3(7):919–924
55. Park K, Drummy LF, Vaia RA (2011) Ag shell morphology on Au nanorod core: role of Ag
precursor complex. J Mater Chem 21(39):15608–15618
56. Guo S, Dong S (2011) Metal nanomaterial-based self-assembly: development, electrochemi-
cal sensing and SERS applications. J Mater Chem 21(42):16704–16716
57. Tao A, Sinsermsuksakul P, Yang P (2007) Tunable plasmonic lattices of silver nanocrystals.
Nat Nanotechnol 2(7):435–440
58. Zou J, Kim F (2012) Self-assembly of two-dimensional nanosheets induced by interfacial
polyionic complexation. ACS Nano 6(12):10606–10613
59. Muramatsu M, Akatsuka K, Ebina Y, Wang K, Sasaki T, Ishida T, Miyake K, Haga
M-a (2005) Fabrication of densely packed titania nanosheet films on solid surface by
use of langmuirblodgett deposition method without amphiphilic additives. Langmuir
21(14):6590–6595
60. Podsiadlo P, Michel M, Lee J, Verploegen E, Wong Shi Kam N, Ball V, Lee J, Qi Y, Hart
AJ, Hammond PT, Kotov NA (2008) Exponential growth of LBL films with incorporated
inorganic sheets. Nano Lett 8(6):1762–1770
61. Srivastava S, Kotov NA (2008) Composite layer-by-layer (LBL) assembly with inorganic
nanoparticles and nanowires. Acc Chem Res 41(12):1831–1841
62. Estephan ZG, Qian Z, Lee D, Crocker JC, Park S-J (2013) Responsive multidomain free-
standing films of gold nanoparticles assembled by DNA-directed layer-by-layer approach.
Nano Lett 13(9):4449–4455
63. Bigioni TP, Lin X-M, Nguyen TT, Corwin EI, Witten TA, Jaeger HM (2006) Kinetically
driven self assembly of highly ordered nanoparticle monolayers. Nat Mater 5(4):265–270
64. Cheng W, Park N, Walter MT, Hartman MR, Luo D (2008) Nanopatterning self-assembled
nanoparticle superlattices by moulding microdroplets. Nat Nanotechnol 3(11):682–690
65. Retsch M, Zhou Z, Rivera S, Kappl M, Zhao XS, Jonas U, Li Q (2009) Fabrication of
large-area, transferable colloidal monolayers utilizing self-assembly at the air/water interface.
Macromol Chem Phys 210(3-4):230–241
66. Mueggenburg KE, Lin X-M, Goldsmith RH, Jaeger HM (2007) Elastic membranes of close-
packed nanoparticle arrays. Nat Mater 6(9):656–660
67. Cheng L, Liu A, Peng S, Duan H (2010) Responsive plasmonic assemblies of amphiphilic
nanocrystals at oilwater interfaces. ACS Nano 4(10):6098–6104
68. Duan H, Wang D, Kurth DG, Möhwald H (2004) Directing self-assembly of nanoparticles at
water/oil interfaces. Angew Chem Int Ed 43(42):5639–5642
69. Reincke F, Hickey SG, Kegel WK, Vanmaekelbergh D (2004) Spontaneous assembly of
a monolayer of charged gold nanocrystals at the water/oil interface. Angew Chem Int Ed
43(4):458–462
70. Mittal M, Furst EM (2009) Electric field-directed convective assembly of ellipsoidal colloidal
particles to create optically and mechanically anisotropic thin films. Adv Funct Mater
19(20):3271–3278
71. Demirörs AF, Johnson PM, van Kats CM, van Blaaderen A, Imhof A (2010) Directed self-
assembly of colloidal dumbbells with an electric field. Langmuir 26(18):14466–14471
72. Ku JC, Ross MB, Schatz GC, Mirkin CA (2015) Conformal, macroscopic crystalline
nanoparticle sheets assembled with DNA. Adv Mater 27(20):3159–3163
73. Lee J, Bhak G, Lee J-H, Park W, Lee M, Lee D, Jeon NL, Jeong DH, Char K, Paik SR
(2015) Free-standing gold-nanoparticle monolayer film fabricated by protein self-assembly
of ’-synuclein. Angew Chem 127(15):4654–4659
74. Cheng W, Campolongo MJ, Cha JJ, Tan SJ, Umbach CC, Muller DA, Luo D (2009) Free-
standing nanoparticle superlattice sheets controlled by DNA. Nat Mater 8(6):519–525
75. Cheng W, Campolongo MJ, Tan SJ, Luo D (2009) Freestanding ultrathin nano-membranes
via self-assembly. Nano Today 4(6):482–493
76. Campolongo MJ, Tan SJ, Smilgies D-M, Zhao M, Chen Y, Xhangolli I, Cheng W, Luo D
(2011) Crystalline Gibbs monolayers of DNA-capped nanoparticles at the air–liquid interface.
ACS Nano 5(10):7978–7985
77. Booth SG, Dryfe RA (2015) Assembly of nanoscale objects at the liquid/liquid interface. J
Phys Chem C 119(41):23295–23309
78. Lin Y, Skaff H, Emrick T, Dinsmore A, Russell T (2003) Nanoparticle assembly and transport
at liquid-liquid interfaces. Science 299(5604):226–229
79. Guo P, Sikdar D, Huang X, Si KJ, Su B, Chen Y, Xiong W, Yap LW, Premaratne M, Cheng
W (2014) Large-scale self-assembly and stretch-induced plasmonic properties of core–shell
metal nanoparticle superlattice sheets. J Phys Chem C 118(46):26816–26824
80. Chen Y, Fu J, Ng KC, Tang Y, Cheng W (2011) Free-standing polymer–nanoparticle
superlattice sheets self-assembled at the air–liquid interface. Cryst Growth Des 11(11):4742–
4746
81. Si KJ, Guo P, Shi Q, Cheng W (2015) Self-assembled nanocube-based plasmene nanosheets
as soft surface-enhanced raman scattering substrates toward direct quantitative drug identifi-
cation on surfaces. Anal Chem 87(10):5263–5269
82. Rao S, Si KJ, Yap LW, Xiang Y, Cheng W (2015) Free-standing bilayered nanoparticle
superlattice nanosheets with asymmetric ionic transport behaviors. ACS Nano 9(11):11218–
11224
83. Ng KC, Udagedara IB, Rukhlenko ID, Chen Y, Tang Y, Premaratne M, Cheng W (2012)
Free-standing plasmonic-nanorod superlattice sheets. ACS Nano 6(1):925–934
84. Lee YH, Shi W, Lee HK, Jiang R, Phang IY, Cui Y, Isa L, Yang Y, Wang J, Li S, Ling XY
(2015) Nanoscale surface chemistry directs the tunable assembly of silver octahedra into three
two-dimensional plasmonic superlattices. Nat Commun 6:6990
85. Tirado MM, Martínez CL, de la Torre JG (1984) Comparison of theories for the translational
and rotational diffusion coefficients of rod-like macromolecules. Application to short DNA
fragments. J Chem Phys 81(4):2047–2052
86. Bishop KJM, Wilmer CE, Soh S, Grzybowski BA (2009) Nanoscale forces and their uses in
self-assembly. Small 5(14):1600–1630
87. Frenkel D (2015) Order through entropy. Nat Mater 14(1):9–12
88. Onsager L (1949) The effects of shape on the interaction of colloidal particles. Ann N Y Acad
Sci 51(4):627–659
89. Cho J-H, Gracias DH (2009) Self-assembly of lithographically patterned nanoparticles. Nano
Lett 9(12):4049–4052
90. Wu L, Dong Z, Kuang M, Li Y, Li F, Jiang L, Song Y (2015) Printing patterned fine 3D
structures by manipulating the three phase contact line. Adv Funct Mater 25(15):2237–2242
91. Nie Z, Petukhova A, Kumacheva E (2010) Properties and emerging applications of self-
assembled structures made from inorganic nanoparticles. Nat Nanotechnol 5(1):15–25
136 Q. Shi et al.
92. Rycenga M, Cobley CM, Zeng J, Li W, Moran CH, Zhang Q, Qin D, Xia Y (2011) Controlling
the synthesis and assembly of silver nanostructures for plasmonic applications. Chem Rev
111(6):3669–3712
93. Tao AR, Ceperley DP, Sinsermsuksakul P, Neureuther AR, Yang P (2008) Self-organized
silver nanoparticles for three-dimensional plasmonic crystals. Nano Lett 8(11):4033–4038
94. Boyack R, Le Ru EC (2009) Investigation of particle shape and size effects in SERS using
T-matrix calculations. Phys Chem Chem Phys 11(34):7398–7405
95. Tian F, Bonnier F, Casey A, Shanahan AE, Byrne HJ (2014) Surface enhanced Raman
scattering with gold nanoparticles: effect of particle shape. Anal Methods 6(22):9116–9123
96. Liu S, Tang Z (2010) Nanoparticle assemblies for biological and chemical sensing. J Mater
Chem 20(1):24–35
97. Shenhar R, Norsten TB, Rotello VM (2005) Polymer-mediated nanoparticle assembly:
structural control and applications. Adv Mater 17(6):657–669
98. Daniel M-C, Astruc D (2004) Gold nanoparticles: assembly, supramolecular chemistry,
quantum-size-related properties, and applications toward biology, catalysis, and nanotech-
nology. Chem Rev 104(1):293–346
99. Shipway AN, Katz E, Willner I (2000) Nanoparticle arrays on surfaces for electronic, optical,
and sensor applications. Chemphyschem Eur J Chem Phys Phys Chem 1(1):18–52
100. Jiang L, Liang X, You T, Yin P, Wang H, Guo L, Yang S (2015) A sensitive SERS substrate
based on Au/TiO2/Au nanosheets. Spectrochim Acta A Mol Biomol Spectrosc 142:50–54
101. Zhong L-B, Yin J, Zheng Y-M, Liu Q, Cheng X-X, Luo F-H (2014) Self-assembly of Au
nanoparticles on PMMA template as flexible, transparent, and highly active SERS substrates.
Anal Chem 86(13):6262–6267
102. Kanjanaboos P, Lin X-M, Sader JE, Rupich SM, Jaeger HM, Guest JR (2013) Self-assembled
nanoparticle drumhead resonators. Nano Lett 13(5):2158–2162
103. Polavarapu L, Liz-Marzan LM (2013) Towards low-cost flexible substrates for nanoplasmonic
sensing. Phys Chem Chem Phys 15(15):5288–5300
104. Palmieri V, Lucchetti D, Gatto I, Maiorana A, Marcantoni M, Maulucci G, Papi M, Pola R, De
Spirito M, Sgambato A (2014) Dynamic light scattering for the characterization and counting
of extracellular vesicles: a powerful noninvasive tool. J Nanopart Res 16(9):1–8
105. Farquharson S, Shende C, Sengupta A, Huang H, Inscore F (2011) Rapid detection and
identification of overdose drugs in Saliva by surface-enhanced Raman scattering using fused
gold colloids. Pharmaceutics 3(3):425–439
106. Andreou C, Hoonejani MR, Barmi MR, Moskovits M, Meinhart CD (2013) Rapid detection
of drugs of abuse in saliva using surface enhanced raman spectroscopy and microfluidics.
ACS Nano 7(8):7157–7164
107. Izquierdo-Lorenzo I, Sanchez-Cortes S, Garcia-Ramos JV (2011) Trace detection ofaminog-
lutethimide drug by surface-enhanced Raman spectroscopy: a vibrational and adsorption
study on gold nanoparticles. Anal Methods 3(7):1540–1545
108. Cîntǎ Pînzaru S, Pavel I, Leopold N, Kiefer W (2004) Identification and characterization of
pharmaceuticals using Raman and surface-enhanced Raman scattering. J Raman Spectrosc
35(5):338–346
Chapter 7
Experimental Observation of Melting
of the Effective Minkowski Spacetime
in Cobalt-Based Ferrofluids
Igor I. Smolyaninov and Vera N. Smolyaninova
Abstract Hyperbolic metamaterials were originally introduced to overcome the

diffraction limit of optical imaging. Soon thereafter it was realized that they
demonstrate a number of novel phenomena resulting from the broadband singular
behavior of their density of photonic states. These novel phenomena and applica-
tions include microscopy, stealth technologies, enhanced quantum-electrodynamic
effects, thermal hyperconductivity, superconductivity, and interesting gravitation
theory analogues. Here we describe the behaviour of cobalt nanoparticle-based
ferrofluid in the presence of an external magnetic field, and demonstrate that it forms
a self-assembled hyperbolic metamaterial, which may be described as an effective
3D Minkowski spacetime for extraordinary photons. Moreover, such photons
perceive thermal gradients in the ferrofluid as analogue of gravitational field, which
obeys the Newton law. If the magnetic field is not strong enough, the effective
Minkowski spacetime gradually melts under the influence of thermal fluctuations.
On the other hand, it may restore itself if the magnetic field is increased back to
its original value. Direct microscopic visualization of such a Minkowski spacetime
melting/crystallization is presented, which is somewhat similar to hypothesized
formation of the Minkowski spacetime in loop quantum cosmology and may mimic
various cosmological Big Bang scenarios.
Keywords Plasmonics • Hyperbolic metamaterials • Metallic nanoparticles •

Ferrofluid • Minkowski spacetime
I.I. Smolyaninov ()

Department of Electrical and Computer Engineering, University of Maryland, College Park,
MD, 20742, USA
e-mail: smoly@umd.edu
V.N. Smolyaninova
Department of Physics Astronomy and Geosciences, Towson University, 8000 York Rd,
Towson, MD, 21252, USA

DOI 10.1007/978-3-319-48081-7_7
138 I.I Smolyaninov and V.N Smolyaninova
7.1 Introduction
Hyperbolic metamaterials are extremely anisotropic uniaxial materials, which

behave like a metal in one direction and like a dielectric in the orthogonal direction.
Originally introduced to overcome the diffraction limit of optical imaging [1, 2],
hyperbolic metamaterials demonstrate a number of novel phenomena resulting
from the broadband singular behavior of their density of photonic states, which
range from super resolution imaging [2–4] to enhanced quantum-electrodynamic
effects [5, 6], new stealth technology [7], thermal hyperconductivity [8], high Tc
superconductivity [9], and interesting gravitation theory analogues [10]. Recent
developments in gravitation theory provide numerous clues, which strongly indicate
that the classic general relativity is an effective macroscopic theory, which will
be eventually replaced with a more fundamental theory based on yet unknown
microscopic degrees of freedom [11, 12]. Unfortunately, these true microscopic
degrees of freedom cannot be probed directly. Our ability to obtain experimental
insights into the future microscopic theory is severely limited by low energy
scales available to terrestrial physics and even to astronomical observations. In
order to circumvent this problem, it is instructive to look at various examples of
emergent gravity and analogue spacetimes [13], which appear in such solid state
systems as superfluid helium, electromagnetic metamaterials, and cold atomic Bose-
Einstein condensates. By looking at such systems we may better understand how
macroscopic effective gravity may arise from the well-known microscopic atomic
degrees of freedom.
Very recently magnetic ferrofluids subjected to external magnetic field have
emerged as an interesting example of an electromagnetic metamaterial, which
exhibits gravity-like nonlinear optical interactions, and which may be described by
an emergent effective Minkowski spacetime [14–16]. Metamaterials are artificially
structured materials, which are built from conventional microscopic materials in
order to engineer their electromagnetic, elastic, thermal, acoustic, etc. properties in
a desired way. Because of considerable 3D nanofabrication difficulties, electromag-
netic metamaterials are typically confined to small sizes in two spatial dimensions.
Recent demonstration that a ferrofluid subjected to strong enough magnetic field
self-assembles into a “wire array” hyperbolic metamaterial (see Fig. 7.1) [14, 17]
opens up many new directions in metamaterial research. Since such a metamaterial
is created by 3D self-assembly, its dimensions are not limited by nanofabri-
cation issues. Unlike other typical metamaterial systems, such a macroscopic
self-assembled 3D metamaterial may also exhibit reach physics associated with
topological defects [18] and phase transitions. Therefore, as was pointed out
recently by Mielczarek and Bojowald [19, 20], the properties of self-assembled
magnetic nanoparticle-based hyperbolic metamaterials exhibit strong similarities
with the properties of some microscopic quantum gravity models, such as loop
quantum cosmology.
Here we will describe direct microscopic visualization of phase transitions in
the cobalt nanoparticle based ferrofluids. In the presence of an external magnetic
7 Experimental Observation of Melting of the Effective Minkowski. . . 139
Fig. 7.1 (a) Schematic geometry of the metal nanowire-based hyperbolic metamaterial. (b)
Wavelength dependencies of the real parts of "z D "2 and "x,y D "1 for a cobalt nanoparticle-
based ferrofluid at ’ D 8.2 % volume fraction of nanoparticles. While "xy stays positive and
almost constant, "z changes sign to negative around œ D 1 m. (c) Experimentally measured
transmission of the cobalt based ferrofluid at œ D 1.55 m as a function of external magnetic
field and polarization angle. Zero degree polarization corresponds to E field of the electromagnetic
wave perpendicular to the direction of external magnetic field. The observed strong polarization
dependencies are consistent with the hyperbolic character of the ferrofluid. (d) Microscopic image
of cobalt nanoparticle-based ferrofluid in external magnetic field reveals nanoparticle alignment
along the field direction
field the ferrofluid forms a self-assembled hyperbolic metamaterial, which may be

described as an effective “3D Minkowski spacetime” for extraordinary photons.
Moreover, it appears that the extraordinary photons propagating inside the ferrofluid
perceive thermal gradients in the ferrofluid as analogue of gravitational field,
which obeys the Newton law. If the magnetic field is not strong enough, this
effective Minkowski spacetime gradually melts under the influence of thermal
fluctuations. On the other hand, it may restore itself if the magnetic field is increased
back to its original value. We will present direct microscopic visualization of

such a Minkowski spacetime melting/crystallization, which is somewhat similar to
hypothesized formation of the Minkowski spacetime in loop quantum cosmology.
It is also interesting to note that the physical vacuum appears to exhibit hyperbolic
metamaterial properties when subjected to a very strong magnetic field [21, 22].
Thus, the ferrofluid in an external magnetic field provides us with an easily
accessible experimental system, which lets us trace the emergence of gravity and
the effective Minkowski spacetime from the well-understood microscopic degrees
of freedom. In addition, the ferrofluid behaviour near the metric signature transition
may mimic various cosmological Big Bang scenarios. Unlike other systems exhibit-
ing analogue gravity behaviour, the unique feature of the ferrofluid consists in our
ability to directly visualize the Minkowski spacetime formation at the microscopic
level.
7.2 Electromagnetic Properties of Cobalt

Nanoparticle-Based Ferrofluid
Our experiments use the cobalt magnetic fluid 27-0001 from Strem Chemicals,
which is composed of 10 nm cobalt nanoparticles in kerosene coated with sodium
dioctylsulfosuccinate and a monolayer of LP4 fatty acid condensation polymer. The
average volume fraction of cobalt nanoparticles in this ferrofluid is p D 8.2 %.
Magnetic nanoparticles in a ferrofluid are known to form nanocolumns aligned
along the external magnetic field [23]. In addition, depending on the magnetic
field magnitude, solvent used, and the nanoparticle concentration, the ferrofluid
undergoes phase separation into nanoparticle rich and nanoparticle poor phases
[17]. The typical spatial scale of phase separation is 0.1–1 m. Therefore, the
microscopic structure of the ferrofluid may be visualized directly using an optical
microscope.
Macroscopic electromagnetic properties of such a ferrofluid may be understood
based on the Maxwell-Garnett approximation via the dielectric permittivities "m
and "d of cobalt and kerosene, respectively. Volume fraction of cobalt nanoparticles
aligned into nanocolumns by the external magnetic field, ˛(H,T), depends on
temperature and the field magnitude. At very large magnetic fields almost all
nanoparticles are aligned into nanocolumns, so that ˛(1, T) D p D 8.2 %, and
a 3D wire array hyperbolic metamaterial [17, 18] is formed as shown schematically
in Fig. 7.1a. We will assume that at smaller fields the difference ˛(1, T)˛(H,T)
describes cobalt nanoparticles, which are not aligned and distributed homoge-
neously inside the ferrofluid. Dielectric polarizability of these nanoparticles may
be included into "d , leading to slight increase in its value. Using this model, the
diagonal components of the ferrofluid permittivity may be calculated using the
Maxwell-Garnett approximation as follows [24]:
"z D "2 D ˛ .H; T/ "m C .1 ˛ .H; T// "d (7.1)
2˛ .H; T/ "m "d C .1 ˛ .H; T// "d ."d C "m /

"x D "y D "1 D (7.2)
.1 ˛ .H; T// ."d C "m / C 2˛ .H; T/ "d
Calculated wavelength dependencies of "2 and "1 at ˛(1, T) D p D 8.2% are plotted
in Fig. 7.1b. These calculations are based on the optical properties of cobalt in the
infrared range [25]. While "1 stays positive and almost constant, "2 changes sign to
negative around D 1 m. If the volume fraction of cobalt nanoparticles varies,
this change of sign occurs at some critical value ˛ H :
"d
˛ > ˛H D ; (7.3)
"d "m
so that the ferrofluid becomes a hyperbolic metamaterial [14, 17]. Alignment of

cobalt nanoparticle chains along the direction of external magnetic field is clearly
revealed by polarization dependencies of ferrofluid transmission and by microscopic
images of the ferrofluid, as shown in Fig. 7.1c, d. For example, Fig. 7.1c shows
experimentally measured transmission of the cobalt based ferrofluid at œD1.55 m
as a function of external magnetic field and polarization angle. Zero degree
polarization corresponds to E field of the electromagnetic wave perpendicular to the
direction of external magnetic field. The observed strong polarization dependencies
are consistent with the hyperbolic character of the ferrofluid. As far as magnetic
permeability is concerned, at the visible and infrared frequencies the ferrofluid may
be considered as a non-magnetic ( D 1) medium.
Propagation of monochromatic extraordinary photons inside the ferrofluid is
described by the wave equation, which is formally equivalent to a 3D Klein-Gordon
equation for a massive field ®! DEz in a 3D Minkowski spacetime:

@2 '! 1 @2 '! @2 '! !0 2 m2 c2
2
C C D '! D '! (7.4)
"1 @z ."2 / @x2 @y2 c2 ¯2
in which the spatial coordinate z behaves as a timelike variable, and the effective
mass is m*D-h! 0 /c2 [26, 27]. Equation (7.4) exhibits effective Lorentz invariance
under the coordinate transformation
1
z0 D q .z ˇx/
"1
1 ." 2 / ˇ
(7.5)
0 1 "1
x Dq xˇ z ;
"1
1 ." ˇ ."2 /
2/
where ˇ is the effective boost. Similar to our own Minkowski spacetime, the
effective Lorentz transformations in the xz and yz planes form the Poincare group
together with translations along x, y, and z axis, and rotations in the xy plane. Thus,
the wave equation (7.4) describes world lines of massive particles which propagate
in an effective 2C1 dimensional Minkowski spacetime. Components of the meta-
material dielectric tensor define the effective metric gik of this spacetime: g00 D"1
and g11 Dg22 D"2 . Nonlinear optical Kerr effect bends this spacetime resulting in
effective gravitational interaction between extraordinary photons [15, 16]. When
the ferrofluid develops phase separation into cobalt rich and cobalt poor phases, its
microscopic structure and the local field intensity in the hyperbolic frequency range
may be observed using an infrared microscope. As a result, microscopic properties
of the effective Minkowski spacetime, as well as its melting/crystallization may
be directly visualized in the experiment. In order to better understand these
experiments and their connection with gravity theory, let us consider in more detail
the emergence of analogue gravity in ferrofluids.
7.3 Analogue Gravity in Ferrofluids
When the nonlinear optical effects become important, they are described in terms of
various order nonlinear susceptibilities (n) of the metamaterial:
.1/ .2/ .3/

Di D ij Ej C ijl Ej El C ijlm Ej El Em C ::: (7.6)
Taking into account these nonlinear terms, the dielectric tensor of the metamate-
rial (which defines its effective metric) may be written as
.1/ .2/ .3/

"ij D ij C ijl El C ijlm El Em C ::: (7.7)
In a centrosymmetric material all the second order nonlinear susceptibilities

(2) ijl must be equal to zero. It is clear that Eq. (7.7) provides coupling between
the matter content (photons) and the effective metric of the metamaterial spacetime.
Nonlinear optical effects “bend” this effective Minkowski spacetime, resulting in
gravity-like interaction of extraordinary light rays. It appears that in the weak field
limit the nonlinear optics of hyperbolic metamaterials may indeed be formulated as
an effective gravity [15]. In such a limit the Einstein equation

8 1
Rki D 4 Tik ıik T (7.8)
c 2
is reduced to
1 1 8
R00 D 2
D g00 D 4 T00 (7.9)
c 2 c
where is the gravitational potential [28]. Therefore, the third order terms in Eq.
(7.7) may provide correct coupling between the effective metric and the energy-
momentum tensor. These terms are associated with the optical Kerr effect. The
detailed analysis in [15] indeed indicates that the Kerr effect in a hyperbolic
metamaterial leads to effective gravity.
Let us briefly reproduce this analysis, since it is important for the discussion of
connection between the effective gravity and the thermodynamics of a magnetized
ferrofluid. Since z coordinate plays the role of time, while g00 is identified with "1,
Eq. (7.9) may be re-written as
16 16
.2/ "1 D 4
Tzz D
zz (7.10)
c c4
where (2) is the 2D Laplacian operating in the xy plane, * is the effective

“gravitation constant”, and
zz is the zz component of the Maxwell stress tensor
of the electromagnetic field in the medium:

1 1
!
! !
!

zz D Dz Ez C Hz Bz DE CHB (7.11)
4 2
Taking into account Eq. (7.4), for a single plane wave Eq. (7.10) may be rewritten
as [15].

4 B2 k2
.0/
.2/ "1 D .2/ "1 C ı"1 D kx2 ı"1 D 2 2 z (7.12)
c ! "1
where without a loss of generality we have assumed that the B field of the wave
is oriented along y direction. We also assumed that the nonlinear corrections to "1
are small, so that we can separate "1 into the constant background value "1 (0) and
the weak nonlinear corrections. These nonlinear corrections do indeed look like
the Kerr effect assuming that the extraordinary photon wave vector components are
large compared to !/c:
4 B2 kz2 4 B2
ı"1 D D .3/ B2 (7.13)
c2 ! 2 "1 kx2 c2 ! 2 "2
This assumption has to be the case if extraordinary photons may be considered

as classic “particles”. Unlike the usual “elliptic” optical materials, this assumption
is justified by the hyperbolic dispersion relation of the extraordinary photons:
!2 kz2 kx2 C ky2

D (7.14)
c2 "1 j"2 j
which follows from Eq. (7.4). Equation (7.13) establishes connection between the
effective gravitation constant * and the third order nonlinear susceptibility (3)
of the hyperbolic metamaterial. Since "2 <0, the sign of (3) must be negative for
the effective gravity to be attractive. This condition is satisfied naturally in most
liquids, and in particular, in kerosene. Because of the large and negative thermo-
optic coefficient inherent to most liquids, heating produced by partial absorption of
the propagating beam translates into a significant decrease of the refractive index at
higher light intensity. For example, the reported thermo-optic coefficient of water
reaches n/T D 5.7 104 K1 [29]. Moreover, introduction of nanoparticles
or absorbent dye into the liquid allows for further increase of the thermal nonlinear
response [30]. Therefore, a ferrofluid-based self-assembled hyperbolic metamaterial
naturally exhibits effective gravity. The thermal origin of this effective gravity looks
interesting in light of the modern advances in gravitation theory [11, 12], which
strongly indicate that the classic general relativity description of gravity results from
thermodynamic effects.
As a next step, let us consider the effect of thermal gradients in the ferrofluid on
its effective metric. It appears that the extraordinary photons propagating inside the
ferrofluid perceive thermal gradients as an effective gravitational field. Indeed, Eqs.
(7.1) and (7.2) imply that "2 and "1 (which may be understood as the effective metric
coefficients g00 D"1 and g11 Dg22 D"2 of the metamaterial spacetime) depend
on the volume fraction ˛(H,T) of cobalt nanoparticles aligned into nanocolumns
by the external magnetic field, which in turn depends on the local temperature of
the ferrofluid. Since ferrofluids subjected to external magnetic field are known to
exhibit classical superparamagnetic behaviour [31], well established results from
the theory of magnetism may be used to calculate ˛(H,T). Superparamagnetism
is a form of magnetism that is exhibited by magnetic materials, which consist of
small ferromagnetic or ferrimagnetic nanoparticles. Superparamagnetism occurs
in nanoparticles which are single-domain, which is possible when their diameter
is below 50 nm, depending on the material. Since the typical size of magnetic
nanoparticles in ferrofluids is 10 nm, magnetic ferrofluids also belong to the
class of superparamagnetic materials. When an external magnetic field is applied
to an assembly of superparamagnetic nanoparticles, their magnetic moments tend to
align along the applied field, leading to a net magnetization. If all the particles may
be considered to be roughly identical (as in the case of a homogeneous ferrofluid),
and the temperature is low enough, then the magnetization of the assembly is [31]

H
M .H; T/ D n tanh ; (7.15)
kT
where n is the nanoparticle concentration, and is their magnetic moment.

Therefore, within the scope of our model of the dielectric response of the ferrofluid
(see Eqs. (7.1 and 7.2)), we may assume that

H
˛ .H; T/ D ˛1 tanh (7.16)
kT
where ˛ 1 D 0.082. Let us consider the typical case of "m >>"d and assume that
˛(H,T) is small (which is typically required for the Maxwell-Garnett approximation
to be valid). In such a case the effective metric coefficients are:

H
g11 D g22 D "2 ˛1 tanh "m "d (7.17)
kT

H
g00 D "1 "d 1 C 2˛1 tanh (7.18)
kT
The effective spacetime appears to be a Minkowski one if the temperature is low

enough, so that

H "d
˛1 tanh > (7.19)
kT ."m /
Thus, in the weak field limit the effective gravitational potential (see Eq. (7.9)) is

c2 g00 H
D 1 D ˛1 c2 tanh 1 (7.20)
2 "d kT
(where the value of the effective potential at TD0 is chosen as a reference) and the
!

effective gravitational field G is
!
˛1 c2 H
G D r D
rT (7.21)
kT 2 cosh2 H
kT
where we have assumed the external magnetic field H to be constant. Thus, the
extraordinary photons propagating inside the ferrofluid perceive thermal gradients
in the ferrofluid as an effective gravitational field. This observation correlates
nicely with the thermal origin of gravity-like nonlinear optical interaction of
the extraordinary light rays due to photo-thermal Kerr effect, which was noted
above.
Let us consider a linear source of heat q (see Fig. 7.2) placed inside the ferrofluid,
which is constant in time and aligned parallel to the external magnetic field H (and
therefore it is also parallel to the nanoparticle chains). Let us also assume that the
ferrofluid is kept at a constant temperature T0 . The temperature distribution around
the source is defined by the two-dimensional heat conductance equation

@2 @2 1

C T D
r .rT/ D q (7.22)
@x2 @y2 cp
Fig. 7.2 Schematic geometry

of the Newton law derivation
for effective gravity in a
ferrofluid
where
is the thermal diffusivity, cp is the heat capacity and is the density of
the ferrofluid. If the linear source of heat is weak enough, so that the temperature-
dependent terms in the denominator of Eq. (7.21) may be considered constant, Eq.
!

(7.22) may be re-written as an equation for the effective gravitational field G :
!
˛1 c2 H
rG D
q D q (7.23)
H

cp kT02 cosh2 kT 0
which has the form of the Newton law of gravity. As evident from Eq. (7.23),
the heat source q plays the role of a gravitational mass. Thus, a linear source of
heat parallel to the external magnetic field H behaves as a world line of massive
gravitating object in the metamaterial Minkowski spacetime. It is interesting that
unlike quantum mechanical derivation based on the holographic principle reported
in [12], the Newton law in a ferrofluid arises in a purely classical Boltzmann
system. In principle, a similar approach may be applied to other magnetic systems
at lower temperatures, where the thermodynamics description becomes explicitly
quantum mechanical [32]. However, direct visualization of gravity-like effects in
such systems appears to be much more difficult.
As far as the strong field limit is concerned (where the temperature may not
be considered almost constant) it was pointed out in ref. [15] that at large enough
power gravitational self-interaction of the extraordinary rays is strong enough, so
that spatial solitons may be formed. These spatial solitons also behave as world lines
of compact self-gravitating bodies in the effective 2C1 dimensional Minkowski
spacetime. This is not surprising since due to photo-thermal effect solitons may
also be considered as linear sources of heat. Moreover, upon increase of the optical
power, a spatial soliton may collapse into a black hole analogue [15]. Indeed, in
the presence of self-defocusing negative Kerr effect in the dielectric host "d , the
wave equation (Eq. (7.4)) must be modified. Assuming a spatial soliton-like solution
which conserves energy per unit length WP/c (where P is the laser power), the
soliton width and the magnetic field amplitude B of the extraordinary wave are
related as
B2 2 D P=c (7.24)
As a result, Eq. (7.4) must be re-written as

@2 '! 1 @2 '! @2 '! !0 2
C 2
C D '! (7.25)
.0/ ..3/ /P ." 2/ @x @y2 c2
"1 c2 @z2
where "1 (0) is the dielectric permittivity component at PD0 (note that the nonlinear
contribution to "2 ˛"m may be neglected). Effective metric described by Eq. (7.25)
has a black hole-like singularity at
.3/ !1=2
P
D .0/
(7.26)
c"1
The critical soliton radius at PD100 W was estimated to be 20 nm [15],

which does not look completely unrealistic from the fabrication standpoint. While
achieving such critical values with CW lasers seems implausible, using pulsed laser
definitely looks like a realistic option since thermal damage produced by a pulsed
laser typically depends on the pulse energy and not the pulse power.
7.4 Microscopic Investigation of Minkowski Spacetime

Melting
Another “strong field” situation corresponds to the ferrofluid behavior near the
metric signature transition. It appears that “melting” of the effective Minkowski
spacetime in such a case may mimic various cosmological Big Bang scenarios.
This is not surprising since symmetry breaking in magnetic systems is typically
described by the Mexican hat potential. Moreover, such a Big Bang-like behavior
may be observed directly using an optical microscope. Indeed, based on Eq. (7.4)
and Fig. 7.1b it is clear that the factor ("2 ) plays the role of a scale factor of the
effective Minkowski spacetime

ds2 D "1 dz2 C ."2 / dx2 C dy2 (7.27)
(since according to Eq. (7.18) "1 is positive and almost constant, as plotted in
Fig. 7.1b). The scale factor of the effective Minkowski spacetime calculated using
Fig. 7.3 (a) Scale factor of

the effective Minkowski
spacetime g11 D g22 D "2
(Calculated using Eq. (7.17))
plotted as a function of
kT/H at different values of
˛ 1 "m . It is assumed that
"d D 2. (b) Metric coefficient
g00 D "1 of the effective
Minkowski spacetime
calculated using Eq. (7.18)
Eq. (7.17) is plotted in Fig. 7.3 as a function of kT/H at different values of

˛ 1 "m . Let us assume that the temperature distribution inside the ferrofluid may
be described as.
T D Tc zrT D Tc z (7.28)
where Tc is the temperature of the metric signature transition (the "2 D0 point). Since
z coordinate plays the role of time in the effective spacetime described by Eq. (7.27),
such a temperature gradient will result in a Big Bang-like spacetime expansion
described by the scale factor –"2 (Tc - z) plotted in Fig. 7.3a. Indeed, as the ferrofluid
temperature falls away from the Tc boundary at zD0, the spacetime scale factor
increases sharply as a function of z. Note that at larger values of ˛ 1 "m expansion
of the effective spacetime accelerates at lower temperatures. As demonstrated below,
this “cosmological” spacetime expansion may be visualized directly using an optical

microscope.
We should also note that in the case of non-constant z-dependent "1 D"x D "y and
"2 D"z the electromagnetic field separation into the ordinary and the extraordinary
components remains well defined [33, 34]. Taking into account z derivatives of "1
and "2 , Eq. (7.4) becomes
2
@2 '! 1 @ '! @2 '! 1 @"1 2 @"2 @'!
C C C C
"1 @z2 ."2 / @x2 @y2 "21 @z "1 "2 @z @z
2
'! 1 @"1 @"2 @ "2 !0 2
C D '!
"1 "2 "1 @z @z @z2 c2
(7.29)
Since "1 remains almost constant in a very broad temperature range far from TD0
(as evident from Fig. 7.3b) its derivatives may be neglected. If we also neglect the
second derivative of "2 , the wave equation for the extraordinary field ®! DEz may
be re-written as
2
@2 '! 1 @ '! @2 '! 2 @"2 @'! !0 2
C C D '! (7.30)
"1 @z2 ."2 / @x2 @y2 "1 "2 @z @z c2
It is easy to verify that the latter equation coincides with the Klein-Gordon
equation [35] for a massive particle in a gravitational field described by the metric
coefficients g00 D"1 and g11 Dg22 D"2 :

1 @ ik p @ m2 c2
p g g k D 2 (7.31)
g @xi @x ¯
Indeed, for a wave field D ("2 )1/2 Eq. (7.30) is reproduced if we also neglect
the second derivative terms.
As demonstrated by Fig. 7.1d, the alignment of cobalt nanoparticles along the
timelike z direction is easy to visualize. Our experiments were performed using a
10 m pathlength optical cuvette placed under the infrared microscope. The cuvette
was filled with the ferrofluid and illuminated from below with a linear polarized
1.55 m laser light. The microscopic transmission images and movies were studied
as a function of magnitude and direction of the external magnetic field as it was
repeatedly turned on and off. Examples of the experimentally measured microscopic
transmission images of the cobalt nanoparticle-based ferrofluid obtained at 30
frames per second while the external magnetic field was gradually increased from
zero to about 300 G are shown in Fig. 7.4. The frame number is indicated in
each image. These images correspond to gradual “crystallization” of the effective
Minkowski spacetime starting from the initial state, which may be described as
an “Euclidean optical space” (since the initial state corresponds to a homogeneous
isotropic ferrofluid having "x D"y D"z >0). The pronounced periodic pattern of self-
assembled stripes visible in frame #295 and #315 appears due to phase separation.
Fig. 7.4 Experimentally measured microscopic transmission images of cobalt nanoparticle-based

ferrofluid obtained at 30 frames per second while the external magnetic field was gradually
increased. An infrared microscope was used in combination with a 1.55 m laser as an illumination
source. The frame number is indicated in each image. These images correspond to gradual
“crystallization” of the effective Minkowski spacetime
The periodic stripes are oriented along the direction of the magnetic field, leading
to the hyperbolic character ("2 <0 and "1 >0) of the ferrofluid. It is interesting to
note that a sudden spike in transmission is observed around frame #235, well in
advance of formation of the periodic long range order, which becomes visible after
frame #275. The temporal dependence of this short spike is presented in Fig. 7.5,
which shows temporal behaviour of the average light intensity integrated over the
frame, while the external magnetic field was gradually increased. This spike may
be attributed to short range ordering of the cobalt nanochains along the direction of
the external magnetic field. Since the individual nanochain size falls well below the
diffraction limit, they are not visible in the optical images. The similarly sudden
decrease in transmission indicates that the mutual alignment of this short range
ordered nanochains is short lived.
Fig. 7.5 Average light

intensity integrated over the
frame while the external
magnetic field was gradually
increased plotted as a
function of the frame number.
A sudden spike in
transmission is detected
around frame #235, well in
advance of formation of the
long range order visible after
frame #275. This spike is
attributed to formation of
short range order due to
external magnetic field
Figure 7.6 presents a similar set of microscopic frames recorded while the
external magnetic field was decreased from about 300 G to zero. If the external
magnetic field is not strong enough, the effective Minkowski spacetime gradually
melts under the influence of thermal fluctuations, as illustrated in Fig. 7.6. Compared
to ref. [34], such a melting represents a different “end of time” scenario, since
the spacetime metric signature changes, and the effective Minkowski spacetime
is completely replaced by the Euclidean optical space. Fourier analysis of the
frames from Fig. 7.6, which is shown in Fig. 7.7 confirms gradual disappearance
of the long range order. Gradual melting of the long range order is also clearly
visible in Fig. 7.8, which shows magnified images of the same corner area from
frames #564 and #580. These images reveal fine details of the effective Minkowski
spacetime melting. The top row shows these images in the usual greyscale format,
while the quasi-3D representation of the same images in the bottom row provides
better visualization of the actual nanoparticle filaments. The quasi-3D frame #580
in Fig. 7.8 reveals that instead of being aligned, the nanocolumns start to bend and
tend to form closed loops upon gradual melting of their long range order. This loop
formation is somewhat similar to the hypothesized asymptotic silence [19, 20] in
loop quantum gravity, which can be seen as a process in which continuous space
decouples into a set of not interacting points. A more detailed set of magnified
quasi-3D images of the ferrofluid obtained upon the long range order melting is
shown in Fig. 7.9. The frame number is indicated in each image. It is interesting
to note that a small region in frame #600 apparently remains in a microscopic
“Minkowski spacetime state” (while the rest of the original spacetime has already
melted). It is highlighted by a yellow circle in the frame. Existence of such
microscopic Minkowski spacetime regions inside the ferrofluid below the metric
signature change threshold was hypothesized in [14].
Fig. 7.6 Transmission images of cobalt nanoparticle-based ferrofluid obtained at 30 frames per
second while the external magnetic field was gradually decreased. The frame number is indicated in
each image. These images correspond to gradual “melting” of the effective Minkowski spacetime
It is also interesting to note that similar to [33], gradual melting of the effective
Minkowski spacetime within a single frame may be considered as a toy model
of inflation. Indeed, based on Eq. (7.27) and Fig. 7.1b it is clear that the factor
("2 ) plays the role of a scale factor of the effective Minkowski spacetime (since
"1 is positive and almost constant). Figure 7.10 presents an example of such an
inflation-like behaviour of the effective Minkwoski spacetime in frame #590. As
demonstrated above, visibility of the filaments in ferrofluid transmission images,
which may be assessed quantitatively by Fourier analysis, provides measurements
of the relative value of ˛(H,T) compared to ˛ 1 . A natural way to study ferrofluid
behaviour near the metric signature transition is to gradually reduce the external
magnetic field H until the periodic alignment of cobalt nanoparticles begins to
disappear. Since thermal and/or magnetic field gradients are unavoidable in such
an experiment, a ferrofluid region showing gradual disappearance of filament
Fig. 7.7 Fourier analysis of the frames from Fig. 7.4 confirms gradual disappearance of the long
range order
Fig. 7.8 Magnified images of the same area in frames #564 and #580 from Fig. 7.4 reveal fine
details of the effective Minkowski spacetime melting. Top row shows these images in the usual
greyscale format, while the quasi-3D representation of the same images in the bottom row provides
better visualization of the actual nanoparticle filaments
periodicity across the image is relatively easy to find. A microscopic transmission

image of such a region is presented in Fig. 7.10b. The filament periodicity (compare
Fig. 7.10a, b gradually disappears towards the top of the image, which is verified
by Fourier analysis in Fig. 7.10c (note that the contrast in these Fourier images
is internal). The measured dependence of the spacetime scale factor –"2 on the
effective time calculated at œD1.4 m is shown in Fig. 7.10d. This wavelength
is close enough to the œD1.5 m illumination wavelength used in the experiment,
while the experimental data for the electromagnetic properties of cobalt [25] may
also be used. Calculated values of –"2 are based on the Fourier analysis of smaller
regions in Fig. 7.10b (shown by the yellow boxes). The measured behaviour of
the scale factor in this experiment corresponds to small values of ˛ 1 "m , which
is expected at œD1.4 m based on the optical properties of cobalt [25]. As
evident from Fig. 7.10d, the measured data show good agreement with theoretical
calculations based on Eq. (7.17).
Fig. 7.9 Magnified quasi-3D images of the effective Minkowski spacetime melting. The frame
number is indicated in each image. A small region in frame 600, which remains in a microscopic
“Minkowski spacetime state” (while the rest of the original spacetime has already melted) is
highlighted by yellow circle. Existence of such miscroscopic Minkowski spacetime regions was
hypothesized in [14]
Fig. 7.10 Analysis of gradual melting of the effective Minkowski spacetime within a single frame:
(a) Microscopic transmission image of the cobalt nanoparticle-based ferrofluid taken in large
external magnetic field H using illumination with œ D 1.5 m light. (b) Analysis of gradual
melting of the effective Minkowski spacetime within a single microscopic image of the ferrofluid
(frame #590). The effective time direction is indicated by the red arrow. The effective spacetime
expansion is shown schematically by the yellow cone. (d) Measured dependence of the spacetime
scale factor –"2 on the effective time calculated at œ D 1.4 m. These calculations are based on
the Fourier analysis (c) of smaller regions of Fig. 7.10b shown by the yellow boxes. The measured
data are compared with theoretical calculations based on Eq. (7.17)
7.5 Final Remarks
Unfortunately, the described analogy between the extraordinary light propagation

inside the ferrofluid and the dynamics of massive particles in Minkowski spacetime
is in no way perfect. The main difficulty comes from the cross-coupling between
extraordinary and ordinary light inside the ferrofluid, which may be caused by
internal defects and strong gradients of the dielectric permittivity tensor compo-
nents. Since ordinary light “lives” in Euclidean space, such a cross-coupling breaks
the effective Lorentz symmetry (5) of the system. In addition, such a model is
necessarily limited to 2C1 spacetime dimensions. Nevertheless, these limitations
notwithstanding, it is interesting to trace the appearance of macroscopic “gravity”
(even though imperfect) to thermodynamics of well understood microscopic degrees
of freedom of the ferrofluid.
In conclusion, we have presented direct microscopic visualization of melt-
ing and crystallization of the effective Minkowski spacetime, which describes
extraordinary light propagation in cobalt ferrofluid-based self-assembled hyperbolic
metamaterials. As was pointed out recently by Mielczarek and Bojowald [19,
20], the properties of such self-assembled hyperbolic metamaterials exhibit strong
similarities with the properties of such microscopic quantum gravity models as loop
quantum cosmology. It appears that the effective Minkowski spacetime behaviour
near the metric signature transition may mimic various cosmological Big Bang
scenarios, which may be observed directly using an optical microscope. Even
though the considered model is applicable only to the extraordinary photons, it is
quite interesting since it permits both direct theoretical analysis at the level of well-
defined microscopic degrees of freedom, and it enables direct visualization of these
microscopic degrees of freedom using standard optical tools.
Acknowledgments This work is supported by the NSF grant DMR-1104676.
References
1. Podolskiy VA, Narimanov EE (2005) Strongly anisotropic waveguide as a nonmagnetic

left-handed system. Phys Rev B 71:201101
2. Jacob Z, Alekseyev LV, Narimanov E (2006) Optical hyperlens: far-field imaging beyond the
diffraction limit. Opt Express 14:8247–8256
3. Smolyaninov II, Hung YJ, Davis CC (2007) Magnifying superlens in the visible frequency
range. Science 315:1699–1701
4. Liu Z, Lee H, Xiong Y, Sun C, Zhang X (2007) Far-field optical hyperlens magnifying sub-
diffraction-limited objects. Science 315:1686
5. Jacob Z, Smolyaninov II, Narimanov EE (2012) Broadband Purcell effect: radiative decay
engineering with metamaterials. Appl Phys Lett 100:181105
6. Jacob Z, Kim J-Y, Naik GV, Boltasseva A, Narimanov EE, Shalaev VM (2010) Engineering
photonic density of states using metamaterials. Appl Phys B 100:215
7. Narimanov E, Noginov MA, Li H, Barnakov Y, Darker than black: radiation-absorbing
metamaterial. In: Quantum electronics and laser science conference, OSA Technical Digest
(CD) (Optical Society of America, 2010), paper QPDA6
8. Narimanov EE, Smolyaninov II, Beyond Stefan-Boltzmann law: thermal hyper-conductivity,

arXiv:1109.5444
9. Smolyaninov, I. I., Smolyaninova, V. N., Is there a metamaterial route to high temperature
superconductivity?, Adv Condens Matter Phys 2014, 2014, 479635
10. Smolyaninov II (2011) Critical opalescence in hyperbolic metamaterials. J Opt 13:125101
11. Jacobson T (1995) Thermodynamics of spacetime: the Einstein equation of state. Phys Rev
Lett 75:1260
12. Verlinde EP (2011) On the origin of gravity and the laws of Newton. J High Energy Phys
2011:29
13. Barcelo C, Liberati S, Visser M (2005) Analog gravity. Living Rev Relat 8:12
14. Smolyaninov II, Yost B, Bates E, Smolyaninova VN (2013) Experimental demonstration of
metamaterial “multiverse” in a ferrofluid. Opt Express 21:14918–14925
15. Smolyaninov II (2013) Analogue gravity in hyperbolic metamaterials. Phys Rev A 88:033843
16. Smolyaninov II (2014) Holographic duality in nonlinear hyperbolic metamaterials. J Opt
16:075101
17. Smolyaninova VN, Yost B, Lahneman D, Narimanov E, Smolyaninov II (2014) Self-assembled
tunable photonic hyper-crystals. Sci Rep 4:5706
18. Smolyaninov II, Smolyaninova VN, Smolyaninov AI (2015) Experimental model of topo-
logical defects in Minkowski spacetime based on disordered ferrofluid: magnetic monopoles,
cosmic strings and the spacetime cloak. Philos Trans R Soc A 373:20140360
19. Mielczarek J, Signature change in loop quantum cosmology, arXiv:1207.4657 [gr-qc]
20. Bojowald M, Mielczarek J, Some implications of signature-change in cosmological models of
loop quantum gravity, arXiv:1503.09154 [gr-qc]
21. Smolyaninov II (2011) Vacuum in strong magnetic field as a hyperbolic metamaterial. Phys
Rev Lett 107:253903
22. Chernodub MN, Van Doorsselaere J, Verschelde H (2012) Spontaneous electromagnetic
superconductivity and superfluidity of QCDxQED vacuum in strong magnetic field. Proc Sci
QNP2012:109
23. Gao Y, Huang JP, Liu YM, Gao L, Yu KW, Zhang X (2010) Optical negative refraction in
ferrofluids with magnetocontrollability. Phys Rev Lett 104:034501
24. Wangberg R, Elser J, Narimanov EE, Podolskiy VA (2006) Nonmagnetic nanocomposites for
optical and infrared negative-refractive-index media. J Opt Soc Am B 23:498–505
25. Johnson PB, Christy RW (1974) Optical constants of transition metals: Ti, V, Cr, Mn, Fe, Co,
Ni, and Pd. Phys Rev B 9:5056
26. Smolyaninov II, Narimanov EE (2010) Metric signature transitions in optical metamaterials.
Phys Rev Lett 105:067402
27. Smolyaninov II, Hung YJ (2011) Modeling of time with metamaterials. JOSA B 28:
1591–1595
28. Landau L, Lifshitz E (2004) Field theory. Elsevier, Amsterdam
29. Brochard P, Grolier-Mazza V, Cabanel R (1997) Thermal nonlinear refraction in dye solutions:
a study of the transient regime. J Opt Soc Am B 14:405–414
30. Alkeskjold TT, Lægsgaard J, Bjarklev A, Hermann DS, Anawati A, Broeng J, Li J, Wu ST
(2004) All-optical modulation in dye-doped nematic liquid crystal photonic bandgap fibers.
Opt Express 12:5857–5871
31. Kittel C (2004) Introduction to solid state physics. Wiley, New York
32. Smolyaninov II (2014) Quantum topological transition in hyperbolic metamaterials based on
high Tc superconductors. J Phys Condens Matter 26:305701
33. Smolyaninov II, Hung YJ, Hwang E (2012) Experimental modeling of cosmological inflation
with metamaterials. Phys Lett A 376:2575–2579
34. Smolyaninov II, Hwang E, Narimanov EE (2012) Hyperbolic metamaterial interfaces: hawking
radiation from Rindler horizons and spacetime signature transtions. Phys Rev B 85:235122
35. Davydov AS (1976) Quantum mechanics. Pergamon Press, New York
Chapter 8
Surface Plasmon Resonance Based Fiber Optic
Sensors Utilizing Zinc Oxide Thin Films
and Nanostructures
Banshi D. Gupta and Rana Tabassum
Abstract Metal oxides have been used in sensors based on surface plasmon reso-
nance to enhance the performance of the sensor and for the sensing of the analyte.
This is because a thin film of metal oxide enhances the electric field intensity at the
oxide-analyte interface due to its high refractive index and also changes its refractive
index due to its interaction with analyte to be sensed. Thus, the advantage of metal
oxide is twofold. Zinc oxide (ZnO) is one such metal oxide that has been widely
used in sensors to enhance the sensitivity and to detect the analyte. Apart from bulk
coating ZnO has been used as nanostructures such as nanoparticle, nanocomposite,
core-shell nanostructure, nanoflower, nanorod, nanobelt and nanowire. It has shown
its application in sensing of gases and various chemical and biochemical analytes.
In this chapter we have focused on the surface plasmon resonance based fiber optic
sensors utilizing zinc oxide as thin film and nanostructure. We have discussed its
application in sensing of hazardous and pollutant gases and pollutants in water.
In addition, its use in enhancing the performance parameter of the sensor by
optimizing its film thickness or composition ratio in the case of nanocomposite and
nanostructures has also been discussed. For the optimization of various parameters
and evaluation of the performance of the sensor simulations have been carried out
in various studies. These simulations use matrix method for the determination of
reflectivity and an algorithm for the determination of electric field intensity at the
interfaces of ZnO layer. Simulations, algorithm their implementation and the results
on some of the studies have been presented and discussed.
Keywords Surface plasmon resonance • Optical fiber • Sensor • Zinc oxide •

Thin film • Nanostructures • Hazardous gases • Water pollutants
B.D. Gupta () • R. Tabassum

Physics Department, Indian Institute of Technology Delhi, 110016, New Delhi, India
e-mail: bdgupta@physics.iitd.ernet.in

DOI 10.1007/978-3-319-48081-7_8
160 B.D. Gupta and R. Tabassum
8.1 Introduction
In last two decades surface plasmon resonance technique has been widely used
for varieties of sensing applications due to fast response, high sensitivity and
label free detection [1–12]. Surface plasmons are basically the collective oscil-
lations of free electrons at the metal-dielectric interface giving rise to transverse
magnetically polarized wave propagating along the metal-dielectric interface. The
surface plasmon wave can be excited by a p-polarized light by transferring some
of its energy/momentum to the surface plasmon wave. The resonance between
the two occurs when the propagation constant of excitation light becomes equal
to the propagation constant of the surface plasmon wave. Under this condition,
excitation light transfers most of its energy to surface plasmons. To excite surface
plasmons Kretschmann configuration is used [13, 14]. In the standard Kretschmann
configuration, a metal layer is coated over the base of a high index prism. The other
side of the film is kept in contact of a dielectric medium to be sensed/studied.
When a TM polarized light is incident through one of the faces of the prism at
the prism-metal interface at an angle of incidence greater than the critical angle,
the evanescent wave gets generated which propagates along the prism-metal layer
interface and excites the surface plasmons at the metal-dielectric interface. At
resonance, the intensity of the reflected light from the prism-metal interface is
minimum which occurs at a particular angle of incidence. In other words, a plot
between the intensity of the reflected light and the angle of incidence possesses
a sharp dip in the reflected light intensity at an angle called the resonance angle.
The change in the refractive index of the dielectric medium in contact of the metal
layer changes the resonance angle. Thus knowing the resonance angle, the refractive
index of the dielectric medium can be determined. The prism based surface plasmon
resonance (SPR) sensor has some disadvantages such as bulky size, expensive,
cannot be used for remote sensing applications. Many of the shortcomings are
overcome by replacing the prism with the core of an optical fiber and still using
Kretschmann configuration [1, 3–7, 15–17]. The miniaturization of the probe and
sensing of multianalyte are additional advantages of optical fiber based SPR sensor
[18]. In the case of optical fiber a range of guided rays are launched and hence to
achieve resonance condition wavelength interrogation instead of angular is used.
In a basic fiber optic SPR probe, cladding is removed from some part of the fiber
and the unclad core is coated with a thin film of metal. The dielectric medium to
be sensed is kept in contact of the other surface of the metal layer. Light from a
polychromatic source is launched in the fiber and the spectrum of the transmitted
power is determined. The wavelength corresponding to minimum transmission is
called the resonance wavelength and its value depends on the refractive index of the
dielectric medium around the metal layer. Similar to angular interrogation in prism,
an increase in refractive index of the dielectric medium increases the resonance
wavelength. Knowing the resonance wavelength, one can determine the refractive
index of the dielectric medium. There are various characteristic parameters on which
8 Surface Plasmon Resonance Based Fiber Optic Sensors Utilizing Zinc. . . 161
the performance of the sensor depends. These are shift in resonance wavelength
for a unit change in refractive index called the sensitivity and the spectral width
of the SPR curve on which the detection accuracy of the resonance wavelength
depends. These two performance parameters are combined to one called figure of
merit (FOM) and for a sensor its value should be large.
To enhance the performance of the SPR based fiber optic sensors various
modifications are being carried out in the design of the probe such as tapering
and bending of the sensing region [19–21]. The other modification is in terms of
coating of nanometer thick film of high index dielectric such as silicon over metal
layer which increases the electric field intensity at the interface of over-layer and
analyte (sensing medium) [22–24]. In some of the cases metal oxides (MOs) have
also been found very suitable for SPR based sensors [25, 26]. The advantage of
metal oxides is that, in addition to enhancing electric field intensity at the interface
due to high refractive index, they, in some of the cases, also change their refractive
index due to interaction with analyte to be sensed. Thus, their advantage is two-
fold. Zinc oxide (ZnO) is one such metal oxide. It is a transparent material in the
visible and IR region of the electromagnetic spectrum and does not agglomerate
in the form of islands and also does not involve band-to-band transitions. Apart
from bulk coating it has also been used as nanostructures such as nanoparticles,
nanocomposites, core-shell nanostructures, nanoflowers, nanorods, nanobelts and
nanowires to enhance the sensitivity of the SPR based sensors. ZnO has shown its
application in gas sensing and in the sensing of various chemical and biochemical
analytes.
In this chapter we shall focus on the surface plasmon resonance based fiber
optic sensors utilizing zinc oxide thin films and nanostructures. We shall discuss its
application in sensing of gases and pollutants in water. Further, we shall also discuss
its use in enhancing the performance parameter of the sensor by optimizing its film
thickness or composition ratio in the case of nanocomposite and nanostructures.
For the optimization of various parameters and for the performance evaluation
of the sensor simulations have been carried out using the matrix method for the
determination of reflectivity and an algorithm for the determination of electric field
intensity at the interface of the ZnO layer. These methods and their implementation
shall also be discussed.
8.2 SPR Based Sensors and Performance Parameters
Figure 8.1a shows the design of a simplest prism based SPR probe utilizing
Kretschmann configuration. The base of the prism is coated with a thin layer of
metal while the dielectric medium to be sensed is kept on the other side of the metal
layer. A p-polarized light is incident on the interface of the prism and the metal
layer through one of the faces of the prism at an angle, , greater than or equal
to the critical angle to produce an evanescent wave propagating along the prism-
metal interface. The electric field of evanescent wave decays exponentially in the
Fig. 8.1 (a) Prism based Kretschmann configuration for SPR based sensor, and (b) a typical plot
of reflectivity versus angle of incidence at prism-metal interface
metal and the dielectric medium. If "p is the dielectric constant of the material of the
prism then the propagation constant of the evanescent wave is given by
!p
Kev D "p sin (8.1)
c
where ! is the frequency and c is the velocity of light in vacuum. As mentioned in
previous section, the coherent oscillation of free electrons at the metal-dielectric
interface results in the propagation of surface plasmon wave along the metal-
dielectric interface. Its propagation constant is given by

! "m "s 1=2
KSP D (8.2)
c "m C "s
where "m and "s are the dielectric constants of the metal and the dielectric medium
respectively. For the excitation of surface plasmons, the propagation constants of
excitation light which is evanescent wave and the surface plasmon wave should be
equal. Thus, the resonance condition is given as

!p ! "m "s 1=2
"p sin res D (8.3)
c c "m C "s
To satisfy this condition, the first requirement is that "p should be greater than "s .
For a monochromatic source, the condition is satisfied at an angle called resonance
angle ( res ). At this angle the incident light transfers maximum light to surface
plasmon wave. In the case of prism based SPR sensor, the intensity of the reflected
light from the prism-metal interface exited from the face of the prism is measured
as a function of angle of incidence. At resonance angle, a sharp dip in reflected light
intensity is observed as shown in Fig. 8.1b. The resonance angle depends on the
refractive index of the dielectric medium (or sensing medium). Increase in refractive
(a) (b)
Transmitted Power
1.0
Cladding
θ
Light out ns
Light in ns+d ns
Metal Sensing
Core 0.0
layer region
l
dl res
1 2 3 4 N-2 N-1 N Transmitted

Reflected light
light
Incident light
(C)
Fig. 8.2 (a) Schematic of a fiber optic SPR probe, (b) the plot of normalized transmitted power
versus wavelength of excitation light for two different refractive indices of the dielectric (sensing)
medium around the metal layer, and (c) the two dimensional multilayer system for simulation
index increases the resonance angle. A plot of resonance angle versus refractive
index of the dielectric medium (calibration curve) is used to determine the refractive
index of the unknown dielectric medium around the metal layer if the value of the
resonance angle is known experimentally.
In the case of fiber optic SPR sensor, same configuration is applied because in
an optical fiber light guidance occurs due to the total internal reflection at the core-
cladding interface and the resulting evanescent wave propagates along this interface.
The SPR based fiber optic sensor is fabricated by removing cladding from a small
portion of the fiber and coating the unclad core with a metal film. The sensing
medium surrounds the metal layer as shown in Fig. 8.2a. In an optical fiber the
rays with angles greater than the critical angle are guided and, therefore, instead
of resonance angle the resonance wavelength is determined from the transmitted
spectrum. For this, instead of a monochromatic source, a polychromatic light source
is used to launch light into the fiber. The evanescent field of the guided rays excites
the surface plasmons at the interface of metal-sensing medium which depends on
the wavelength, fiber parameters, probe geometry, and the metal used. The SPR
spectrum is similar in shape to that shown in Fig. 8.1b. The sensor is calibrated by
plotting the resonance wavelength as a function of refractive index of the sensing
medium.
The performance of the SPR sensor is evaluated in terms of sensitivity, detection
accuracy (or the spectral width), figure of merit (FOM) and limit of detection (LOD)
in the case of chemical/biosensors. Sensitivity is defined as the shift in resonance
wavelength per unit change in the refractive index of the sensing medium. In the case
of chemical sensor, refractive index is replaced by the concentration of the analyte in
the definition. Figure 8.2b shows a SPR spectrum for two different refractive indices
of the sensing medium ns and ns C ıns . Increase in refractive index by ıns shifts the
resonance wavelength by ıres . The sensitivity is thus defined as
ıres
SD (8.4)
ıns
It is determined by differentiating the calibration curve of the sensor. The other

important parameter is the spectral width of the SPR curve. How accurately and
precisely the resonance wavelength can be determined is decided by the spectral
width of the SPR curve. The detection accuracy of the sensor is, thus, inversely
proportional to the width of the SPR spectrum. For the accurate determination of
the resonance wavelength the SPR spectrum should be narrow. Unfortunately when
sensitivity increases spectral width also increases. Therefore, another performance
parameter called as figure of merit is defined which takes into account both the
sensitivity and the spectral width and is mathematically written as
S
FOM D (8.5)

where œ is the spectral width of the SPR curve. The figure of merit should be
maximized to achieve the best performance of the sensor. In the case of chemical
sensor, limit of detection (LOD) is also an important parameter and is defined as
resolution
LOD D (8.6)
SCD0
where resolution is the resolution of spectrometer and SCD0 is the sensitivity of

the sensor near zero concentration of the analyte. For the sensor, the value of LOD
should be as low as possible.
8.3 Zinc Oxide: Background
Metal oxides have been widely used for the sensing of various chemical analytes in
sensors utilizing different methods and phenomena. In SPR based sensors these have
also been used along with the plasmonic metals. A number of metal oxides have
been used for this purpose and zinc oxide (ZnO) is one such metal oxide. It has many
advantages such as transparent in the visible and IR region of the electromagnetic
spectrum, absence of islands on a very thin layer of ZnO deposited on the dielectric
surface, and also its no involvement in band-to-band transitions. Basically, ZnO is
a wide-band gap semiconductor and is referred as n-type semiconductor because
of doping creating electron charge carriers. This has several favourable properties
towards sensing. Its nanostructures such as nanoparticles, nanocomposites, core-
shell nanostructures, nanoflowers, nanorods, nanobelts, nanowires etc can be used
to enhance the sensitivity of the SPR based sensor. Due to these advantages, ZnO
has been used in sensing of gases, chemical and biochemical analytes. It has been
used in different forms like thin film, nanoparticle, nanocomposite and moreover
the core-shell nanostructure. It has been used both as high index dielectric layer
over metal layer and as the sensing material.
8.3.1 ZnO as High Index Layer
It has been observed that the addition of high index over-layer over plasmonic metal
layer in the SPR based sensor enhances the evanescent field at the interface of high
index dielectric layer and analyte which in turn enhances the sensitivity of the sensor
[22, 23]. Zinc oxide is a high index material and hence if used over metal layer can
enhance the sensitivity of the sensor. Additionally, it is a semiconductor oxide with
high melting as well as boiling point and therefore does not get easily oxidized
which further improves the durability of the sensor probe.
8.3.2 ZnO as Sensing Material
There are various exciting properties of oxides which make them suitable in
sensing applications. The sensing characteristics of the sensor improve if the sensing
surface absorbs analyte properly and in more quantity. The properties of oxides that
improve the sensing are porosity in their structure, inherent oxygen vacancy in the
intermediate site and defects level in the structure. The defects level may be tailored
by adding some dopant in the oxide. The defects level is the favourable property
for sensing because the analyte can easily reside/adsorb over it for long time and
can get sensed. This property proves to be more prominent when the defects level
gets tailored by a selective material for a particular analyte. It not only enhances the
sensitivity but improves the selectivity of the sensor manifold. Zinc oxide possesses
all these characteristics and in more amount which make it very useful for the
sensing of gases and some of the analytes.
Before fabricating a sensor with ZnO layer or its nanocomposite, some of
the parameters such as the thickness of the ZnO layer and composition ratio are
optimized. For this, simulations are carried out to determine the performance of the
sensor. Below we discuss the methods/algorithm being used for this purpose.
8.4 Electric Field Intensity at Interfaces
For modelling and designing of the SPR based sensors, the first step is to obtain
the electric field intensity at the interfaces due to the penetration of the evanescent
field into the analyte region. More the field is at the interface of the last layer and
the analyte more is the sensitivity of the sensor. An algorithm has been reported
for the evaluation of electric field intensity in the multilayer structure shown in Fig.
8.2 [23]. The amplitudes of the electric and magnetic fields at the first interface are
connected with the last interface by an overall characteristic matrix according to the
following equation
" # " #
Hy1 HyN
DM (8.7)
Ex1 ExN
The amplitude reflection coefficient, rp , is expressed in terms of the incident and

reflected amplitude of the magnetic field as
Hyref
rp D (8.8)
Hyinc
where
Hyref D .M11 C M12 qN / q1 .M21 C M22 qN / (8.9)
and
Hyinc D .M11 C M12 qN / q1 C .M21 C M22 qN / (8.10)
The field within the 1st layer can be calculated from the following equation
" #
Hy1 .z/ 1 C rp
D P1 .z/ Hyinc ; z1
z
z2 (8.11)
Ex1 .z/ q1 1 rp
where P1 (z) is given as

" #
cos .k0 n1 z cos 1 / qi2 sin .k0 n1 z cos 1 /
P1 .z/ D (8.12)
iq1 sin .k0 n1 z cos 1 / cos .k0 n1 z cos 1 /
By using the values of constants, the amplitude of the electric field component
can be calculated by taking Hyinc D1 for the normalized magnetic field. The field
components at the succeeding layers are calculated using the following matrix
1
Y
Hyj .z/ 1 C rp Hyinc ; zj
z
zjC1
D Pj .z/ P1 .z D zl C dl //
Exj .z/ q0 1 rp
lDj1
(8.13)
where Pj (z) is given as

" #
cos k0 nj z zj cos j qij sin k0 nj z zj cos j
Pj .z/ D (8.14)
iqj sin k0 nj z zj cos j cos k0 nj z zj cos j
8.5 SPR Spectrum
In the case of fiber optic SPR sensor the spectrum of the transmitted power is
determined from the amplitude reflection coefficient at the interface. The N-layer
model shown in Fig. 8.2c and the matrix method are used for this [27]. In this model,
the layers are considered to be stacked along the Z-axis normal to the core-metal
interface. The kth layer is represented by the thickness dk , dielectric constant "k ,
permeability k , and refractive index nk . For p-polarized light, the tangential fields
at the first interface Z D Z1 D 0 are related to those at the final interface Z D ZN1
by

U1 UN1
DM (8.15)
V1 VN1
U1 and V1 are the tangential components of electric and magnetic fields at the first
interface, respectively; UN1 and VN1 are the corresponding fields at the (N1)
interface and M is a characteristic matrix of the combined structure given by
N1
Y
M11 M12
MD Mk D (8.16)
M21 M22
kD2
with

cos ˇk .i sin ˇk / =qk
Mk D (8.17)
iqk sin ˇk cos ˇk
where
1=2 1=2
k "k n21 sin2 1
qk D cos k D (8.18)
"k "k
2 2dk 1=2
ˇk D nk cos k .zk zk1 / D "k n21 sin2 1 (8.19)

k is the angle of the ray with the normal to the kth interface while 1 is the
angle of the ray with the normal to the core-metal interface. The amplitude reflection
coefficient (rp ) is given by
.M11 C M12 qN / q1 .M21 C M22 qN /

rp D (8.20)
.M11 C M12 qN / q1 C .M21 C M22 qN /
Finally, the intensity reflection coefficient (Rp ) is

ˇ ˇ2
Rp D ˇrp ˇ (8.21)
The angular power distribution of rays launched into the fiber using a collimated
source and a focusing lens is given by
n2 sin cos
dP / 1 2 d (8.22)
1 n21 cos2
The normalized transmitted power for all the guided rays launched in the fiber is
given by
Z
=2
N ./ n21 sin cos
Rp ref 2 d
1 n21 cos2
cr
Ptrans D (8.23)
Z
=2
n21 sin cos
2 d
1 n21 cos2
cr
where
L
Nref ./ D (8.24)
D tan

ncl
cr D sin1 (8.25)
n1
Nref is the total number of reflections made by a ray making an angle with the
normal to the core-metal interface in the sensing region while L and D are the length
and diameter of the fiber core sensing region respectively; cr is the critical angle of
the fiber whereas n1 and ncl are the refractive indices of the core and the cladding
respectively.
8.6 Dispersion Relations
To find the SPR spectrum by simulation one needs the dispersion relations of
materials of each layer used in the probe. Homogeneous as well as nanocomposite
layers are used in the SPR probes and the dispersion relations are different for
homogeneous and nanocomposite layers. The dispersion relations of the materials
that have been used in the probe fabrication of the sensors discussed in this chapter
are given below.
8.6.1 Homogeneous Film
Generally, plastic clad silica fiber (PCS) has been used for the fabrication of SPR
probe. The dispersion relation of the silica core is given by the following Sellmeier
relation
s
A1 2 A2 2 A3 2
n1 ./ D 1 C 2 2
C 2 2
C 2 (8.26)
B1 B2 B23
where is the wavelength in m and A1 , A2 , A3 , B1 , B2 and B3 are the Sellmeier

coefficients. The values of these coefficients are A1 D 0.6961663, A2 D 0.4079426,
A3 D 0.8974794, B1 D 0.0684043 m, B2 D 0.1162414 m and B3 D 9.896161 m
[28]. Over unclad core, a film of metal is coated in the SPR based fiber optic sensor.
The wavelength dependence of the dielectric function of the metal layer is obtained
using Drude formula given as
2 c
"1 ./ D 1 (8.27)
2p .c C i/
where p and c are the plasma and collision wavelengths of the plasmonic metal
respectively.
The wavelength dependence of the dielectric function of the succeeding layer,
ZnO, is obtained by fitting the data given in ref. [29] and is given by
0:879682
"ZnO ./ D 2:81418 2 (8.28)
0:30422 0:007112
In some of the studies a layer of nanocomposite of ZnO and metal is used. In that
case the dispersion relation of the effective dielectric constant is used.
8.6.2 Nanocomposite Film
In a nanocomposite, particles of one component material are dispersed in a

continuous host of the other component material. When the size of the particles
are much smaller than the wavelength of the light then the Maxwell–Garnett model
can be used to calculate the dielectric function of the nanocomposite [30]. For two
component materials with dielectric constants, "1 (component 1) and "2 (component
2, dielectric host material), the effective dielectric function of the composite material
is given by
"1 C 2"2 C 2f ."1 "2 /

"eff D "2 (8.29)
"1 C 2"2 f ."1 "2 /
where f is the volume-filling factor of component 1 into component 2.

For the experimental analysis, the value of f is obtained in terms of the mass of
the individual material. Let A and B are the component 1 and component 2 of the
nanocomposite. The mass, density and volume of component 1 are represented by
m1, d1 and V1 while that of component 2 by m2, d2 and V2 . If x is the relative mass
of component 1 then (1x) will be the relative mass of the other component and the
empirical
formula of the nanocomposite will be Ax B(1x) . Using the definition of
volume V D md , f can be related to x through the following equation
.x=d1 /
f D
(8.30)
x .1x/
d1
C d2
The value of x in Ax B(1x) nanocomposite can be chosen in the range (0

x
1).
8.6.3 Experimental Methods of Coating
The sensors discussed in this chapter involve the coatings of thin films of plasmonic
metal, ZnO, nanocomposite, and the core-shell nanostructure. Two kinds of tech-
niques have been used for coating depending on the material. The first technique
is thermal evaporation in high vacuum and the other is the dip coating. Thermal
evaporation technique is used to coat thin films of plasmonic metals and metal
oxides (ZnO). The technique offers accurate, stable, homogeneous and cost effective
coating in extremely high vacuum (5 106 mbars). The thickness of the deposited
film is monitored using quartz crystal monitor. The other technique, dip coating,
is used to coat nanostructure onto 3D objects such as core of the optical fiber. For
coating of a 3D object, it is simply inserted and removed from the bath of solution to
be coated. Although further steps are needed to improve the film adhesion capability
but the basic method is the insertion and removal only. The insertion and removal of
the substrate or the fiber is made at a constant speed and is allowed to remain inside
the solution for a while before pulling at a constant speed. Sometimes the process
is repeated number of times depending upon the requirement of the thickness of the
material. The thin layer gets deposited itself on the substrate while it is pulling up.
The pulling speed is a useful parameter in dip coating process because it determines
the thickness of the coating. Faster is the speed of withdrawal thicker is the coating
over the substrate. For the removal of the extra or the unwanted material, deionized
water is used. The substrate is washed number of times to drain the unwanted
particles from the substrate surface. Depending on the type of the material, the
substrate is allowed to dry in air as well as in oven. Thin films or gel type materials
can be dried in open air and get adhere over the substrate surface. Generally the
deposited nanostructure needs drying in oven typically at 60–90 ı C temperature for
more than 4–5 h. Although the temperature required in the drying depends upon the
type of material.
8.7 Zinc Oxide as Over Layer
As reported, introducing a high index layer over commonly used plasmonic metal
such as silver (Ag) and gold (Au) enhances the electric field intensity at the interface
of the high index layer and analyte. For high index layer ZnO has also been
chosen because of other properties which help in sensing of chemical analytes [31].
Instead of single metal layer, bimetallic layers have also been used for plasmons. In
this section we discuss a SPR based fiber optic refractive index sensor utilizing
single/bimetallic metal layers along with ZnO layer. Four types of sensors with
layers of Ag/ZnO, Au/ZnO, Au/Ag/ZnO and Ag/Au/ZnO over unclad core of
the fiber have been studied [32]. The electric field intensity as a function of the
distance from the core-metal interface for all the four sensors have been determined
for 600 nm wavelength and 1.33 refractive index of the sensing medium. The
thicknesses of the metallic layers are kept fixed while the thickness of the ZnO
layer is varied. In the case of bimetallic layer the thicknesses of individual layer
are equally chosen. It has been observed that the electric field intensity at the
interface of ZnO and sensing medium depends on the thickness of the ZnO layer
and is maximum for a particular thickness of the ZnO layer. Figure 8.3 shows the
maximum field intensity of all the aforesaid four sensors. It may be seen that the
sensor with bimetallic layer (Ag/Au) with 20 nm thickness of each metal along with
10 nm thickness of ZnO layer possesses the maximum field intensity at the interface
of ZnO and analyte (sensing medium). Since the field is correlated to the sensitivity
of the sensor [22] the combination of Ag/Au/ZnO layers with respective optimized
thickness of 20/20/10 nm should give the maximum sensitivity of the sensor.
A fiber optic SPR probe has been fabricated for refractive index sensing using
the optimized thicknesses of layer obtained above. The plastic clad silica fiber of
core diameter 600 m and NA 0.37 ˙ 0.02 is used. The experimental setup used
for the characterization of the fabricated probe is shown in the Fig. 8.4 [33]. The
experimental setup is consist of a tungsten halogen lamp, microscope objective,
30
ZnO-Analyte Interface Wavelength =0.600 mm
R.I. of Analyte =1.33
25
ZnO-Analyte Interface
Ag-ZnO
Interface
Ag(0.020 mm)/Au(0.020mm)/ZnO(0.010mm)
20
2
Ag(0.040mm)/ZnO(0.010mm)
|E|
15 Au(0.040mm)/ZnO(0.005mm)
Au(0.020mm)/Ag(0.020mm)
10 ZnO(0.005mm)
0
0 0.05 0.1 0.15 0.2 0.25
Distance from the core metal interface (mm)
Fig. 8.3 Electric field intensity along the normal to the interfaces with optimized thicknesses. The
refractive index of the analyte is 1.33 while the wavelength of the light source is 600 nm
Fig. 8.4 Schematic of the experimental setup for the characterization of the fiber optic SPR sensor
[33] (Reprinted with permission from Springer)
glass flow cell fixed with the fabricated sensing probe and a spectrometer interfaced
with the personal computer to record the spectrum for a given sample in the flow
cell.
Figure 8.5a shows the SPR spectra of fiber optic probe having the optimized
multilayer of silver, gold and zinc oxide with respective thicknesses of 0.020 m,
0.020 m and 0.010 m for solutions of refractive index lying in the range from
1.33 to 1.37 [32]. A red shift in the resonance wavelength is observed as the
refractive index of the solution is increased. Similar kinds of spectra have been
obtained using the theoretical analysis discussed above. The theoretical SPR spectra
are shown in Fig. 8.5b [32]. A quantitative difference between experimental and
theoretical spectra occurs due to the approximations taken in the theoretical analysis.
1.0
(a)
Normalized Transmitted Power 0.9
0.8
R I=1.37
R I=1.33
0.7 R I=1.36
R I=1.34 R I=1.35
0.5 0.6 0.7 0.8 0.9
Wavelength(mm)
1
(b)
Normalized Transmitted Power
0.9
0.8
0.7
RI=1.33
0.6
0.5
RI=1.34
0.4
RI=1.37
0.3
RI=1.36
RI=1.35
0.2
0.5 0.6 0.7 0.8 0.9 1 1.1 1.2
Wavelength (mm)
Fig. 8.5 (a) Experimental and (b) theoretical SPR spectra for different refractive indices of the
analyte solution in the case of probe coated with Ag (0.020 m)/Au (0.020 m)/ZnO (0.010 m)
layers [32] (Modified from Ref. [32])
These are (i) use of 2-dimensional geometry although the fiber is 3-dimensional and
(b) consideration of only meridional rays although the majority of rays guided in
the fiber are skew rays. The roughness of the metal surface can also be the reason of
mismatch which has not been considered in the theoretical modelling.
35
FOM 30
25
20
Theoretical points
Experimental points
15
1.33 1.34 1.35 1.36 1.37

Refractive index
Fig. 8.6 Variation of figure of merit with the refractive index of the analyte solution for the fiber
optic probe with Ag (0.020 m)/Au (0.020 m)/ZnO (0.010 m) (Modified from Ref. [32])
Table 8.1 Simulated and experimental values of probe parameters with different coatings [32]
FOM at 1.37 refractive index
Simulated
Electric field
Coatings on intensity at 1.33 Optimized
fiber optic SPR refractive index thickness of
probe (A.U.) Theo. Exp. ZnO film (nm)
Ag/ZnO 20.23 30 25 10
Au/ZnO 17.70 24 21 5
Au/Ag/ZnO 4.10 20 16 5
Ag/Au/ZnO 29.33 35 30 10
The performance parameters such as sensitivity, width of the SPR curve and
the figure of merit have been calculated from the theoretical and experimental SPR
spectra. Figure 8.6 shows the variation of FOM with the refractive index of the
solution obtained from the theoretical and experimental results [32]. It may be noted
that the FOM increases with the increase in the refractive index of the solution.
Although there is a slight mismatch in the theoretical and experimental curves, the
mismatch in values is again due to the approximations taken in the theoretical model
as mentioned above. For the comparison of all the aforesaid sensors, the values of
electric field intensity, FOM and the optimized thickness of ZnO layer are tabulated
in Table 8.1 [32]. It may be noted that the electric field intensity at the interface
of ZnO layer and sensing medium matches with the FOM values obtained both
theoretically and experimentally.
8.8 Hazardous and Pollutant Gases
In the previous section, it has been shown that the electric field intensity at the
interface, sensitivity and the figure of merit of the SPR based fiber optic refractive
index sensor improve by using bimetallic coating and the zinc oxide over-layer. For
the gas sensing applications, the electric field intensity is not the only parameter for
the sensitivity enhancement but the sensing surface also promotes gas molecules to
get adsorbed over it and get sensed. This idea has been explored for the sensing of
hydrogen sulphide (H2 S) and hydrogen (H2 ) gases. As mentioned in the beginning,
ZnO possesses large defects level which can adsorb/reside the gas molecules over
its surface leading to the change in the dielectric function of the ZnO and hence can
be used in the sensing of the gas. The concept of adsorbing gas molecules over the
defects level is only valid for the gases of higher molecular weight like H2 S. For the
lighter gases such as H2 which can easily migrate, sensing is not possible using ZnO.
Therefore, for the sensing of lighter gases some absorbent in coordination with ZnO
is required. Palladium and platinum are well known absorbents for hydrogen gas.
A schematic diagram of the experimental set up of the fiber optic SPR gas sensor
is shown in Fig. 8.7 [34]. For the study of the response of the sensor probe, the
fabricated probe is fixed inside the gas chamber which has the facility for gas inlet
and outlet. To evacuate the chamber, a rotary pump is connected to the chamber.
After every measurement, the gas is evacuated from the chamber using the rotary
pump. For the measurement of the vacuum pressure inside the chamber, a vacuum
gauge is also connected in the path of the rotary pump. In the experimental setup,
N2 gas assembly is also attached with the test gas to remove the undesirable gases
from the chamber. The concentration of the test gas inside the chamber is varied and
Fig. 8.7 Schematic of the experimental setup for the detection of gases [34] (Reprinted with
permission from IOP)
the SPR spectra are recorded. The resonance wavelength corresponding to different
concentrations of the test gas is determined from these SPR spectra to calibrate the
sensor. Here we shall describe the ZnO based fiber optic SPR sensors for both the
heavier and lighter gases.
8.8.1 Hydrogen Sulphide (H2 S) Gas Sensor
In this section, we discuss a SPR based fiber optic H2 S gas sensor. The probe
is fabricated by depositing plasmonic metal Cu (40 nm) with additional sensing
layer of ZnO (10 nm) over unclad core of the fiber. The principle of sensing of the
H2 S gas by the fabricated probe is based on the presence of defects level over the
sensing surface for residing the gas molecules. The defects level are maximum in
the case of ZnO which is evident from the photoluminescence spectra of thin films
of various oxides viz. TiO2 , SiO2 , SnO2 and ZnO shown in Fig. 8.8 [35]. The peak
photoluminescence intensity (PPI) implies the defects level. Higher the PPI more
is the defects level. Hence, amongst the four oxide materials, ZnO has maximum
number of defects over its surface and provides largest number of active sites for
H2 S sensing.
To study the response characteristics of the sensor SPR spectra are recorded for
different concentrations of H2 S gas around the sensing probe. Figure 8.9 shows the
SPR spectra of the fabricated fiber optic SPR probe for different concentrations of
the H2 S gas [35]. The SPR curve shifts towards red side of the electromagnetic
Fig. 8.8 Photoluminescence spectra of the films of different metal oxides [35] (Reproduced by
permission of the PCCP Owner Societies)
Fig. 8.9 SPR spectra of fiber optic SPR probe for H2 S gas [35] (Reproduced by permission of the
PCCP Owner Societies)
spectrum as the concentration of the gas increases. The reason of shift is that the
exposure of ZnO layer with H2 S changes its dielectric function according to the
following reaction
ZnO.s/ C H2 S.g/ • ZnS.s/ C H2 O.g/ (8.31)
In addition to the shift, the depth of the SPR spectrum also increases with
the increase in the concentration of H2 S gas. This implies that both the real and
imaginary parts of the refractive index of the zinc oxide change with the change in
the concentration of the H2 S gas. The change in the real part changes the resonance
wavelength whereas the change in the imaginary part changes the depth of the SPR
curve. The variation of resonance wavelength with the concentration of H2 S gas is
shown in Fig. 8.10a [35]. The resonance wavelength increases as the concentration
of H2 S gas increases. The rate of increase decreases as the concentration of H2 S
increases and saturates beyond 100 ppm. Thus, the H2 S gas sensor has the fixed
range of operation. The dependence of sensitivity of the sensor on the concentration
of H2 S gas is shown in Fig. 8.10b [35]. Increase in the concentration of H2 S gas
decreases the sensitivity and it becomes negligible at higher concentrations. The
reason of this is that as the concentration of the H2 S gas increases the number of
ZnO molecules available to react per H2 S molecule decreases and hence no further
change in optical properties of ZnO occurs after a particular concentration. Use
of resolution of the spectrometer (0.333 nm) and the sensitivity of the sensor near
zero concentration (0.667 nm/ppm) gives the limit of detection of the sensor as
0.499 ppm.
Fig. 8.10 Variation of (a) resonance wavelength, and (b) sensitivity of the sensor with the
concentration of the H2 S gas in the chamber [35] (Reproduced by permission of the PCCP Owner
Societies)
8.8.2 Hydrogen (H2 ) Gas Sensor
In this section we describe two types of fiber optic SPR sensors for the sensing
of small concentrations of hydrogen gas. The probe of one sensor has multilayer
of ZnO and palladium (Pd) while the other has layer of their composite over
a silver coated unclad core of the fiber. Palladium is well known for absorbing
hydrogen which influences its dielectric function. ZnO, on the other hand, is used in
Fig. 8.11 Mechanism of H2 gas sensing by the fiber optic SPR probe having coatings of Ag and
ZnO with Pd in the form of multilayer or the nanocomposite [34] (Reprinted with permission from
IOP)
coordination with Pd to combine the properties of ZnO and Pd so as to improve

the hydrogen sensing behaviour of the sensor. The ZnO being a high dielectric
constant material increases the evanescent field at the interface and has large number
of defects or oxygen vacancy on its surface. First, we shall discuss the theoretical
analysis of these probes reported in the literature. The coatings of these probes are
represented as Ag/ZnO/Pd and Ag/ZnO:Pd nanocomposite respectively.
In Fig. 8.11 we show the mechanism of sensing of hydrogen gas by these two
probes [34]. The absorption and desorption of H2 by Pd lead to the formation
of reversible palladium hydride (PdHx ) according to Eq. (8.32); ‘x’ represents
the atomic ratio (H/Pd). The absorption of H2 by Pd can also be related to a
crystallographic phase transitions represented by ‘’’ and ‘“’. In the absence of H2 ,
Pd is in the ’-phase. As H2 comes in contact with Pd layer the formation of PdHx
occurs which is a solid solution of H2 in the palladium lattice. The hydridation of
Pd leads to the “-phase. Hence, the principle behind the sensing of H2 gas by Pd
involves ’ to “ transition and vice versa.
x
Pd C H2 $ PdHx (8.32)
2
In terms of physical change, the complex dielectric function and the crystallo-
graphic phase of palladium change in the presence of hydrogen gas. The complex
dielectric function of palladium changes according to the following equation
"pd .c/ D h .c/ "pd .0/ (8.33)
where "pd (c) and "pd (0) represent the dielectric function of Pd in the presence and
absence of of H2 gas; h(c) is a variable which depends on the concentration (c) of
H2 gas. Its value for 0 % and 4 % H2 are 1.0 and 0.8, respectively. The dielectric
function of palladium before hydridation, is obtained by fitting the data given in ref.
[36] and is given by
h i
"Pd ./ D 24:96364 e.=1:09383/ C 28:65615
h i (8.34)
C i 15:31843 e.=0:76839/ 16:97702
The SPR spectra are generated using dielectric functions of materials of all
the layers for varying thickness of ZnO and fixed thickness of Pd in Ag/ZnO/Pd
multilayer fiber optic hydrogen gas sensor. The spectral width of the SPR spectrum
and the shift in resonance wavelength for the change in hydrogen concentration from
0 % to 4 % H2 are determined to calculate FOM.
The variation of FOM with the thickness of ZnO layer for different thicknesses
of Pd layer is shown in Fig. 8.12a [37]. It may be noted that FOM depends on
the thickness of the layers of both the materials. The maximum value of FOM is
obtained for the combination of 0.09 m thickness of ZnO and 0.05 m thickness of
palladium. The SPR spectra for probe with optimized thicknesses of all the layers for
0 % and 4 % H2 concentration are shown in Fig. 8.12b [37]. Increasing concentration
of H2 from 0 to 4 % shifts the SPR spectrum significantly. The shift in resonance
wavelength obtained is about 0.020 m.
Similar kind of study has been carried out by replacing multilayer of ZnO/Pd
by layer of nanocomposite of these materials (ZnO:Pd). The variation of FOM as
a function of volume fraction (defined in Eq. (8.30) for different thicknesses of
nanocomposite layer is shown in Fig. 8.13a [37]. It is observed that FOM depends
on the volume fraction of nanoparticles of Pd in ZnO as well as on the thickness of
the nanocomposite layer. The maximum value of FOM is obtained around 0.080 m
thickness of the nanocomposite film and 0.55 volume fraction of nanoparticles in
host material. Using these optimized values SPR spectra of the fiber optic hydrogen
gas sensor have been determined for 0 % and 4 % concentration of hydrogen gas
around the probe. A significant shift of 0.139 m in spectrum is observed for the
change in concentration from 0 % to 4 % H2 as shown in Fig. 8.13b [37]. This
is higher than that obtained in the case of multilayer probe. Thus replacement of
ZnO/Pd multilayer by ZnO:Pd nanocomposite layer is thus advantageous.
1.5
1.4 (a) Ag/ZnO/Pd multilayer model
1.3
Figure of merit 1.2 Pd = 0.050 m m
1.1
1.0
0.9
0.8
0.7
0.6 Pd = 0.060 m m
Pd = 0.040 m m
0.5
0.4
0.06 0.07 0.08 0.09 0.10 0.11 0.12
Thickness of ZnO (m m)
0.98
(b) Multilayer Model: Ag (0.040 m m)

ZnO (0.090 m m)
0.96 Pd (0.050 m m)
0.94
0 % H2
0.92
0.9
4 % H2
0.88
0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7

Wavelength (mm)
Fig. 8.12 (a) Variation of FOM with the thickness of ZnO layer for three different thicknesses
of Pd layer, and (b) SPR spectra for 0 % and 4 % concentration of hydrogen gas in the case of
Ag/ZnO/Pd multilayer coated probe [37] (Reprinted with permission from OSA)
An experimental study has been reported on hydrogen gas sensing using

nanocomposite layer of ZnO and Pd over silver coated unclad core of the fiber.
Palladium doped zinc oxide nanocomposites [ZnO(1-x) Pdx ] are prepared by chemical
route for different composition ratio. The fabrication of the sensor by varying the
fraction of Pd in ZnO is carried out using Eq. (8.30). The experimental setup for the
hydrogen gas sensing is similar to that shown in Fig. 8.7. Here the test gas is hydro-
gen. The lower explosion limit (LEL) of hydrogen is 4 % in nitrogen atmosphere.
3.5 ZnO:Pd=0.080 m m
(a) ZnO:Pd=0.070 m m ZnO:Pd=0.090 m m
3.0
2.5 ZnO:Pd=0.050 m m
Figure of merit
ZnO:Pd=0.060 m m
2.0
1.5 ZnO:Pd=0.1
ZnO:Pd=0.040 m m
1.0
0.5 ZnO:Pd=0.120 m m ZnO:Pd=0.030 m m

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Volume fraction of nanoparticles (f)
1
(b) Nanocomposite Model: Ag (0.040 m m)
ZnO:Pd (0.080m m)
0.95
f = 0.55
0.9
0.85
0.8
0 % H2
0.75
4 % H2
0.7
0.75 1.25 1.5
Wavelength (m m)
Fig. 8.13 (a) Variation of FOM with volume fraction of nanoparticles for different thickness of
nanocomposite film and (b) SPR spectra for 0 % and 4 % concentration of H2 gas in the case of
Ag/ZnO:Pd coated fiber optic hydrogen sensor [37] (Reprinted with permission from OSA)
Therefore, both the hydrogen and nitrogen gas cylinders are connected to the sensing
chamber. All the measurements are carried out at atmospheric pressure for pure
nitrogen i.e. 0 % H2 and 4 % H2 C96 % N2 . The position of the dip for 0 % and 4 %
concentrations of hydrogen are obtained from the recorded spectrum.
8000000 Ag(40nm)/ZnO(1-x)Pdx
7000000 x=0.75 x=0.55

PL Intensity (a.u.)
6000000
5000000
x=0.4
4000000 x=0.3
3000000 x=0.15 x=0 (pure ZnO)

2000000
1000000
0
360 380 400 420 440 460 480 500
Wavelength (nm)
Fig. 8.14 Photoluminescence spectra of pure ZnO and ZnO(1-x) Pdx for different values of x [34]
(Reprinted with permission from IOP)
As mentioned earlier, ZnO is known to have large number of defects on its

surface known as ‘interstitial defects’ or ‘oxygen vacancy’. The amount of defects
on the surface of any oxide is responsible for the gas sensing behaviour because gas
molecules reside on defect level and changes the dielectric function of the material.
The defects level can further be increased by adding impurities. To confirm this,
photoluminescence (PL) spectra have been recorded for pure ZnO and Pd doped
ZnO nanocomposite for different volume fractions of Pd in ZnO and are shown
in Fig. 8.14 [34]. In the present case, the impurity atom is Pd which also absorbs
H2 . It has been observed that the peak photoluminescence intensity is highest for
x D 0.4 in ZnO(1-x) Pdx and lowest for pure ZnO. The increase in value of x beyond
0.4 does not increase the peak intensity of PL spectrum and hence the defects level
gets saturated at x D 0.4.
The SPR spectra, shown in Fig. 8.15, are recorded for 0 % and 4 % concentrations
of the H2 gas in N2 atmosphere by the probe having multilayer of Ag (40 nm)
and ZnO(1x) Pdx nanocomposite (94 nm) with x D 0.4 [34]. When hydrogen gas
concentration changes from 0 % to 4 % in N2 atmosphere around the fabricated
probe the resonance wavelength shifts towards red side by 80 nm. This gives
the maximum sensitivity of the sensor as 20 nm/% H2 . The experimentally
determined shift in resonance wavelength is slightly lower than that obtained using
theoretical model due to the reasons mentioned earlier on the approximations taken
in theoretical modelling. The use of spectrometer resolution (0.333 nm) and the
sensitivity of the sensor near zero concentration give the value of the LOD of the
fabricated hydrogen gas sensor as 0.016 % H2 in nitrogen atmosphere.
0.95 Ag(40nm)/ZnO(1-x)Pdx (94 nm) @ x=0.4

Gas chamber presssure = 1 atm
0.90
0 % H2
0.85
0.80
0.75
4 % H2
0.70
0.65
0.60
600 700 800 900 1000 1100 1200
Wavelength (nm)
Fig. 8.15 SPR spectra of fiber optic SPR probe for 0 % H2 and 4 % H2 in N2 at 1 atmospheric
pressure [34] (Reprinted with permission from IOP)
8.9 Water Impurities Sensors
Results of previous sections show that the electric filed intensity, sensitivity and
figure of merit of the fiber optic SPR sensor improve manifold by adding high index
ZnO layer. These parameters can further be improved by the doping of metallic
absorbent materials (for example Pd for H2 ) into the high index ZnO. In this
section, we discuss the doping of conducting polymer into ZnO for the detection
of impurities in water. Three conducting polymers viz. polyvinylpyrollidone (PVP),
polypyrrole (PPy) and polyaniline (PANI) have been used for the detection of
chlorine content in water, heavy metal ions such as manganese (Mn2C ) in water and
the hardness of the water, respectively. The sensing occurs due to the change in the
dielectric function of the sensing surface due to the presence of analyte molecules
around it while the selectivity to a particular analyte is due to the specific reactions
of a particular analyte with a particular conducting polymer. For the sensing of
Mn2C the nanocomposite of polypyrrole with ZnO (ZnO(1x) PPyx ) has been used
[38]. The mechanism of Mn2C sensing is based on the interaction of outer most
electrons of Mn2C with the electrons due to the -orbital conjugation of PPy in ZnO
matrix. Manganese has three electrons in outer shell and this increases the amount of
charge carriers and hence the quanta of oscillations (plasmons). The paramagnetic
behaviour of Mn2C , -orbital conjugation in sensing material and the overall change
in the effective dielectric constant of the ZnO(1-x) PPyx nanocomposite result in the
shift in the SPR spectrum.
Ag(40nm)+ZnO(1-x)PPyx at x=0.55
1.0

0.64
0.63
0.9
0.62
0.61
0.8 0 mg/l
150 mg/l 720 750
10 mg/l
200 mg/l
0.7
20 mg/l
50 mg/l
0.6 100 mg/l
480 600 720 840 960

Wavelength (nm)
Fig. 8.16 SPR spectra of the fiber optic SPR probe for different concentrations of the Mn2C in
water. [38] (Reprinted with permission from Elsevier)
The experimental setup used to record the SPR spectrum of the sensor is similar
to that shown in Fig. 8.4. Figure 8.16 shows the SPR spectra of the fiber optic probe
having layers of Ag and ZnO(1x) PPyx with x D 0.55 for Mn2C concentration in
water varying from 0 to 200 g/l [38]. As the concentration of Mn2C increases, a red
shift in the SPR spectrum is observed. The variations of resonance wavelength and
the sensitivity of the sensor with the concentration of Mn2C are shown in Fig. 8.17a,
b, respectively [38]. The resonance wavelength saturates for higher concentrations
due to the decrease in the number of PPy molecules available per Mn2C . In terms
of optical properties, the effective refractive index of ZnO(1-x) PPyx increases with
the increase in the concentration of Mn2C and becomes constant after 200 g/l
concentration of Mn2C . Due to this reason the sensitivity decreases nonlinearly with
the increase in the concentration of Mn2C .
To sense the hardness of drinking water, the probe similar to manganese
ions has been fabricated. The difference is in conducting polymer. For hardness,
nanocomposite of polyaniline (PANI) with ZnO (ZnO(1-x) PANIx ) has been used.
To achieve the best performance of the sensor, first the composition ratio of the
nanocomposite has been optimized. Rest of the procedure such as fabrication and
measurement has been the same. The working principle involves the change in the
effective dielectric constant of the nanocomposite (ZnO(1-x) PANIx ) by CO3 2 or
HCO3 ions in aqueous atmosphere. Figure 8.18 shows the SPR spectra of the
fiber optic probe having layers of silver and ZnO(1x) PANIx nanocomposite with
x D 0.45 for CO3 2 concentration in water varying from 0 to 200 g/l [33]. The
SPR spectrum shifts toward red with the increase in the concentration of CO3 2
in water. This occurs due to the change in the effective dielectric constant of the
750 (a)
Resonance wavelength (nm)

740
730
Ag(40nm)/ZnO(1-x)PPyx at x=0.55
720 Exp points

Polynominal fit of exp points
710
700
0 50 100 150 200
Concentration of manganese ions in water (m g/l)
1.2 (b)
1.0 Ag(40nm)/ZnO(1-x)PPyx at x = 0.55
Sensitivity (nm/ m g/l)
0.8 Exp points

Polynomial fit of exp points
0.6
0.4
0.2
0.0
0 50 100 150 200

Concentration of manganese ions in water (m g/l)
Fig. 8.17 Variation of (a) resonance wavelength, and (b) sensitivity of the fiber optic SPR probe
for different concentrations of the Mn2C in water [38] (Reprinted with permission from Elsevier)
ZnO(1-x) PANIx nanocomposite layer by CO3 2 . The experiments on the same probe
are also carried out for HCO3 containing water samples and similar kind of trend
is obtained. The variation of shift in resonance wavelength with the concentration
for both kinds of ions is shown in Fig. 8.19 [33]. The shift in resonance wavelength
increases linearly with the increase in their concentration in the range from 0 to
200 g/l. The sensitivity of the sensor for CO3 2 and HCO3 are 0.094 nm/(g/l)
and 0.065 nm/(g/l) respectively. Since both, CO3 2 and HCO3 , are responsible
1.00 Probe=Ag(40nm)/ZnO(1-x)PANIx @ x=0.45

2-
0.98 Analyte=CO3 ions
0.96 0 m g/l
0.94 10 m g/l
50 m g/l
0.92
100 m g/l
0.90 150 m g/l
0.88 200 m g/l
650 700 750 800 850

Wavelength (nm)
Fig. 8.18 SPR spectra of the fiber optic SPR probe for different concentrations of the carbonate
ions in water [33] (Reprinted with permission from Springer)
30 Exp. pts of CO32- ions

Shift in resonance wavelength (nm)
25 Exp. pts of HCO3- ions

Linear Fit
20 Linear Fit
15
10
-5
0 50 100 150 200
Concentration of ions in water (mg/l)
Fig. 8.19 Shift in resonance wavelength as a function of concentrations of the carbonate and
bicarbonate ions in water [33] (Reprinted with permission from Springer)
for the hardness of the water it implies that as the hardness in water increases the
resonance wavelength increases. Knowing the shift in resonance wavelength, one
can find the hardness in water.
The sensing of chlorine content in water, particularly in swimming pool, is
important because its harmfulness if added in large quantity. SPR based fiber optic
chlorine sensor utilizing multilayer of Ag, ZnO and polyvinylpyrollidone (PVP) has
been reported. The absorption of chlorine by the PVP and the ZnO layers enhances
their effective dielectric constant and gives rise to shift in resonance wavelength of
the SPR spectrum. The experimental setup for the sensing of chlorine in water is
similar to that shown in Fig. 8.4. The fiber optic probe fabricated with multilayer of
Ag(40 nm)/ZnO(10 nm)/PVP(5 nm) over the unclad core of the optical fiber with
optimized thicknesses given in brackets is fixed in the flow cell. The SPR spectra
are recorded for different concentrations of chlorine in water and the resonance
wavelengths are noted from each spectrum. Figure 8.20a shows the calibration curve
of the sensor for 0 to 5 ppm concentration [39]. It shows a nonlinear increase in the
resonance wavelength with the increase in the concentration of the chlorine gas in
water. The nonlinearity is because of decrease in the available PVP molecules per
chloride ion with the increase in the concentration of the chlorine molecules. Hence,
at higher concentration of the chlorine the resonance wavelength saturates. In terms
of optical properties, the effective refractive index of PVP increases nonlinearly
with the increase in the concentration of the chlorine becomes constant after nearly
5 ppm concentration. The variation of the sensitivity with the concentration of the
chlorine gas is shown in Fig. 8.20b [39]. The sensitivity linearly decreases with the
concentration and becomes zero at higher concentrations of the chlorine.
8.10 Phenyl Hydrazine Sensor
In the sensors described above the sensing layer is either in the form of a thin
film of material or a film of nanocomposite. In this section we discuss fiber
optic SPR sensor for the detection of phenyl hydrazine (Ph-Hyd) using core-shell
nanostructure. For this study, aluminium (Al) is used in the form of central core into
ZnO shell (Al@ZnO). Phenyl hydrazine, a nitrogen containing aromatic compound,
plays an important role in the production of several drugs, pesticides, dyes, and
rocket propellant. It is found in atmosphere and in different food stuffs such as
tobacco and mushrooms. It may also be present in waste materials and water bodies.
When it comes in contact with the human body it irritates the skin, eyes and throat
and may cause dermatological diseases. It also creates a rapid haemolysis if exposed
to kidney or may cause liver destruction.
The core-shell nanostructure of Al in ZnO with varying volume fraction is
prepared using laser ablation technique. The schematic diagram of the experimental
setup for the preparation of Al@ZnO by laser ablation technique is shown in
Fig. 8.21 [40]. Nd-YAG (neodymium-doped yttrium aluminium garnet) laser of
wavelength 1064 nm, pulse width 10 ns and repetition rate 100 KHz is used for
ablation. It is focused using a lens of 40 mm focal length on the target, Al, immersed
in deionized water. The spot size of the laser on the target is kept around 5.55 m.
The level of the deionized water above the Al target is maintained as 4–5 mm so as
to focus the laser beam exactly onto the Al target. The fabrication of Al@ZnO is
carried out in two steps. In the first step, the laser beam is focused on Al target in
712
Resonance Wavelength (nm)

710 (a)
708
706
704
702
Experimental points
700 Polynomial fit
Ag(40nm)/ZnO(10nm)/PVP(5nm)
698
0 1 2 3 4 5
Concentration of chlorine (ppm)
4 (b)
Sensitivity (nm/ppm)
1
Experimental points
0 Linear fit
Ag(40nm)/ZnO(10nm)/PVP(5nm)
-1
0 1 2 3 4 5
Concentration of chlorine (ppm)
Fig. 8.20 Variation of (a) resonance wavelength and (b) sensitivity of the sensor with the
concentration of the chlorine gas [39] (Reproduced by permission of The Royal Society of
Chemistry)
deionized water to produce the Al core nanoparticles. In the second step, ZnO target
is immersed in the already prepared Al nanoparticles colloidal solution and the laser
is ablated onto the ZnO target to get ZnO shell over the Al core. The final solution
is then filtered so as to ensure the size of the particles in nanorange.
The fiber optic probe for the detection of Ph-Hyd is fabricated by depositing a
layer of Al@ZnO over the silver coated unclad core of the fiber using dip-coating
3-D stage Nd-Yag Laser

Laser beam
Mirror
Lens
Al@ZnO
Laptop to control
structure
laser
Fig. 8.21 Experimental setup for the preparation of Al@ZnO core-shell nanostructure using laser
ablation technique [40] (Reprinted with permission from IOP)
method. The sensing is based on the interaction of Ph-Hyd molecules with Al@ZnO
by a specific reaction/interaction. As mentioned above, ZnO alone possesses oxygen
vacancies over its surface. When Al is introduced within ZnO, it enhances the
adsorption capability by making significant changes in inherent oxygen vacancies
over Al@ZnO surface. The adsorbed oxygen is called as the chemisorbed oxygen.
When Ph-Hyd molecule comes in contact with Al@ZnO, the chemisorbed oxygen
reacts/interacts with Ph-Hyd molecule and oxidizes it resulting in the formation
of diazenyl benzene. The diazenyl benzene again reacts with chemisorbed oxygen
and forms benzenediazonium. The overall reaction between Ph-Hyd and Al@ZnO
changes the effective dielectric constant of Al@ZnO layer. In other words, oxidation
and reduction processes, given below, are the basis of the sensing of Ph-Hyd by
Al@ZnO.
.Al@ZnO/ C O2 ! 2O .Adsorbed oxygen by Al@ZnO/ (8.35)
1
Ph-Hyd C O2 .Adsorbed oxygen of Al@ZnO/ ! Diazenyl Benzene
2
C HC C OH C 2e (8.36)
1
Diazenyl benzene C O2 .Adsorbed oxygen of Al@ZnO/ C HC
2
! Benzenediazonium C H2 O (8.37)
Before fabricating the probe the thickness of the core-shell nanostructure film
and volume fraction are optimized by determining the electric field intensity (jEj2 )
at the interface of core-shell nanostructure layer and analyte sample. For this the
50
Wavelength = 0.633 m m
Probe=Ag(0.040m m)/Al@ZnO
Ag-Al@ZnO (0.040m m)
40 Interface
Al@ZnO-Analyte Interface
f=0.25
30 f=0.15
|E^|2
f=0.55
f=0.0
f=0.65
20 f=0.35 f=0.75
f=0.45
10
0
0 0.05 0.1 0.15 0.2
Distance from the Core-Metal interface (mm)
Fig. 8.22 Variation of electric field intensity with distance from the core-metal interface for
different values of f and 0.040 m thickness of core-shell nanostructre layer [40] (Reprinted with
permission from IOP)
thickness of the silver layer is taken as 40 nm and the wavelength of the light is
considered as 0.633 m. The electric field intensity is calculated for different values
of the volume fraction (f ) and the thickness of Al@ZnO layer. The amount of field
intensity at the interface is highest for 40 nm thickness of Al@ZnO layer and 0.25
volume fraction (or, x D 0.142). The plot of electric field intensity along the normal
to the interface is shown in Fig. 8.22 for different values of volume fraction and
fixed value of the thickness of the core-shell nanostructure layer [40]. As mentioned
earlier, the electric field intensity (jEj2 ) at the interface of sensing layer and analyte
is, theoretically, the measure of the sensitivity of the sensor. However, in a practical
sensor, in addition to (jEj2 ), the surface to volume ratio of the sensing layer also
affects the sensitivity at great extent. This is because larger the surface to volume
ratio of the sensing layer larger would be the exposure of the analyte molecules to
the sensing surface and more will be the sensitivity. In the nanostructured materials
such as core-shell nanostructure, the surface to volume ratio increases in comparison
to the bulk materials.
To fabricate the sensor, the Al@ZnO core shell structure is prepared using
laser ablation technique mentioned above. The TEM images of the core-shell
nanostructure are shown in Fig. 8.23a, b with different magnification to show the
clear image of core-shell nanostructure [40]. Here, Al occupies the position of the
core while ZnO as the shell. The prepared Al@ZnO structures are filtered by the
nano-sized pores filter paper to reduce the size of nanoparticles in the final solution.
The size distribution of Al@ZnO is shown in Fig. 8.23c with the maximum number
Fig. 8.23 TEM images of

Al@ZnO core shell
nanostructure prepared for
f D 0.25 with magnification
(a) 70000X and (b) 200000X.
(c) Size distribution of
Al@ZnO core- shell
nanostructure [40] (Reprinted
with permission from IOP)
50 (c)
40
No. of particles
30
20
10
0
10 20 30 40 50 60 70
Particle size (nm)
0.955
Ag/Al@ZnO core shell

0.950
f = 0.25
0.945
0.940 100 mg/l

150 mg/l
0.935
0.930 0 mg/l
200 mg/l
0.925
10 mg/l
0.920 50 mg/l
0.915
0.910
650 700 750 800
Wavelength (nm)
Fig. 8.24 SPR spectra for 0 to 200 g/l concentration of Ph-Hyd for the probe having layers of
Ag and Al@ZnO with f D 0.25 [40] (Reprinted with permission from IOP)
of particles are of 40 nm diameter [40]. This is close to the optimized value of

thickness of Al@ZnO required for the maximum field at the interface of Al@ZnO-
analyte.
The Al@ZnO coated probe fabricated with optimized parameters, thickness D
40 nm and f D 0.25, is characterized for different concentrations of Ph-Hyd in
buffer solution. The experimental setup to record the SPR spectrum is the same as
in Fig. 8.4. The SPR spectra recorded for different concentrations of Ph-Hyd at pH
7.0 are shown in Fig. 8.24 [40]. Increase in the concentration of Ph-Hyd shifts the
SPR spectrum towards red side. This is due to the formation of diazenyl benzene
and benzenediazonium on the exposure of the probe to Ph-Hyd which increases
the dielectric constant of Al@ZnO layer and shifts the resonance wavelength.
A plot of resonance wavelength determined from the SPR spectrum with the
concentration of Ph-Hyd is shown in Fig. 8.25a [40]. In this case also the increase
in the concentration of Ph-Hyd increases the resonance wavelength nonlinearly.
The variation of sensitivity as a function of concentration and determined from the
slope of the curve shown in Fig. 8.25a is depicted in Fig. 8.25b [40]. The variation
implies a decrease in the sensitivity as the concentration of Ph-Hyd increases.
This is due to the limited number of core-shells in the sensing layer. The value
of limit of detection (LOD) of the sensor is determined using the resolution of
the spectrometer (0.333 nm) and the sensitivity of the Ph-Hyd sensor near zero
concentration (0.124 nm/g/l). These values give 2.685 g/l as the limit of detection
of the Ph-Hyd sensor.
747
(a)
744
Resonance Wavelength (nm)
741
738
735
732
729
Experimental points
726 Polynomial fit of exp point
0 50 100 150 200
Concentration of Ph-Hyd in buffer solution (mg/l)
0.14
(b)
0.12
Sensitivity (nm/mg/l)
0.10
0.08
0.06
Experimental points
0.04 Linear fit of exp points
0 50 100 150 200

Concentration of Ph-Hyd in buffer solution (mg/l)
Fig. 8.25 Variation of (a) resonance wavelength and (b) sensitivity with the concentration of Ph-
Hyd for the probe coated with Ag and Al@ZnO with f D 0.25 [40] (Reprinted with permission
from IOP)
8.11 Summary
In summary, we have described surface plasmon resonance based fiber optic sensors
for the detection of gases and pollutants in water using zinc oxide thin film,
nanocomposite and core-shell nanostructure. Introduction of zinc oxide film over
metal layer in the probe not only enhances the sensitivity of the sensor because
of being the high refractive index material but also increases the sensitivity due
to the presence of defects level. To achieve the best performance of the sensor
the thickness of the zinc oxide film in the probe is optimized using the algorithm
for the electric field intensity determination at the interface of the ZnO layer with
the sensing medium. In the case of nanocomposite film its composition ratio along
with the thickness is optimized. Experimental results on the sensitivity and figure of
merit of various zinc oxide based sensors utilizing optimized parameters have been
presented and compared with the simulated results.
References
1. Jorgenson RC, Yee SS (1993) A fiber-optic chemical sensor based on surface plasmon
resonance. Sensors Actuators B Chem 12:213–220
2. Homola J, Yee SS, Gauglitz G (1999) Surface plasmon resonance sensors: review. Sensors
Actuators B Chem 15:3–15
3. Bevenot X, Trouillet A, Veillas C, Gagnaire H, Clement M (2002) Surface plasmon resonance
hydrogen sensor using an optical fibre. Meas Sci Technol 13:118–124
4. Cennamo N, Massarotti D, Conte L, Zeni L (2011) Low cost sensors based on spr in a plastic
optical fiber for biosensor implementation. Sensors 11:11752–11760
5. Singh S, Gupta BD (2012) Fabrication and characterization of a highly sensitive surface
plasmon resonance based fiber optic ph sensor utilizing high index silicon layer and hydrogel.
Sensors Actuators B Chem 73:268–273
6. Bhatia P, Gupta BD (2012) Fabrication and characterization of a surface plasmon resonance
based fiber optic urea sensor for biomedical applications. Sensors Actuators B Chem 161:434–
438
7. Lin YC (2013) Characteristics of optical fiber refractive index sensor based on surface plasmon
resonance. Microw Opt Technol Lett 55:574–576
8. Hosoki A, Nishiyama M, Igawa H, Seki A, Choi Y, Watanabe KA (2013) Surface plasmon
resonance hydrogen sensor using Au/Ta2 O5 /Pd multi-layers on hetero-core optical fiber
structures. Sensors Actuators B Chem 185:53–58
9. Zeng S, Baillargeat D, Ho HP, Yong KT (2014) Nanomaterials enhanced surface plasmon
resonance for biological and chemical sensing applications. Chem Soc Rev 43:3426–3452
10. Shukla S, Sharma NK, Sajal V (2015) Sensitivity enhancement of a surface plasmon resonance
based fiber optic sensor using ZnO thin film: a theoretical study. Sensors Actuators B Chem
206:463–470
11. Liu Y, Liu Q, Chen S, Cheng F, Wang H, Peng W (2015) Surface plasmon resonance biosensor
based on smart phone platforms. Sci Rep 5:12864
12. Zhang NMY, Hu DJJ, Shum PP, Wu Z, Li K, Huang T, Wei L (2016) Design and analysis
of surface plasmon resonance sensor based on high-birefringent microstructured optical fiber.
J Opt 18:065005
13. Otto A (1968) Excitation of nonradiative surface plasma waves in silver by the method of
frustrated total reflection. Zeitschrift fur Physik 216:398–410
14. Kretschamann E, Raether H (1968) Radioactive decay of non-radiative surface plasmon
excitation by light. Natureforsch 23:2135
15. Aleixandre M, Corredera P, Hernanz ML, Gutierrez-Monreal J (2005) Development of fiber
optic hydrogen sensors for testing nuclear waste repositories. Sensors Actuators B Chem
107:113–120
16. Verma RK, Gupta BD (2008) On the performance of different bimetallic combinations in
surface plasmon resonance based fiber optic sensors. J Phys D: Appl Phys 41:095106
17. Verma R, Gupta BD (2015) Detection of heavy metal ions in contaminated water by surface
plasmon resonance based optical fibre sensor using conducting polymer and chitosan. Food
Chem 166:568–575
18. Verma R, Gupta BD (2013) A novel approach for simultaneous sensing of urea and glucose by
spr based optical fiber multianalyte sensor. Analyst 139:1449–1455
19. Bhatia P, Gupta BD (2011) Sensitivity enhancement of a surface plasmon resonance based
tapered fiber optic sensor using a high index dielectric layer. Proc of SPIE 81731T-1
20. Srivastava SK, Gupta BD (2011) A multi-tapered fiber optic spr sensor with enhanced
sensitivity. IEEE Photon Technol Lett 23:923–925
21. Verma RK, Gupta BD (2008) Theoretical modeling of a bi-dimensional u-shaped surface
plasmon resonance based fiber optic sensor for sensitivity enhancement. J Phys D Appl Phys
41:095106
22. Lahav A, Auslender M, Abdulhalim I (2009) Surface plasmon resonance sensor with enhanced
sensitivity using nano-top dielectric layer. J Nanophoton 3:031501
23. Shalabney A, Abdulhalim I (2010) Electromagnetic fields distribution in multilayer thin film
structures and the origin of sensitivity enhancement in surface plasmon resonance sensors. Sens
Actuators A 159:24–32
24. Bhatia P, Gupta BD (2013) Surface plasmon resonance based fiber optic refractive index sensor
utilizing silicon layer: effect of doping. Opt Commun 286:171–175
25. Singh S, Mishra SK, Gupta BD (2013) Sensitivity enhancement of a surface plasmon resonance
based fiber optic refractive index sensor utilizing an additional layer of oxides. Sens Actuators
A 193:136–140
26. Usha SP, Mishra SK, Gupta BD (2015) Fiber optic hydrogen sulphide gas sensor utilizing
ZnO thin films/ZnO nanoparticles: a comparison of surface plasmon resonance and lossy mode
resonance. Sensors Actuators B Chem 218:196–205
27. Sharma AK, Gupta BD (2005) On the sensitivity and signal to noise ratio of a step index fiber
optic surface plasmon resonance sensor with bimetallic layers. Opt Commun 245:159–169
28. Ghatak A, Thyagarajan K (1998) An introduction to fiber optics. Cambridge University Press,
Cambridge
29. Omar K, Johan MD, Hassin MM (2009) Investigation on dielectric constant of zinc oxide. Mod
Appl Sci 3:110–116
30. Ruppin R (2000) Evaluation of extended Maxwell–Garnett theories. Opt Commun 182:273–
282
31. Bao G, Li D, Jiang W, Cheng XM (2012) Surface plasmon optical sensor with enhanced
sensitivity using top ZnO thin film. Appl Phys A Mater Sci Process 107:279–283
32. Tabassum R, Gupta BD (2015) Performance analysis of bimetallic layer with zinc oxide for
SPR-based fiber optic sensor. IEEE J Lightwave Technol 33:4565–4571
33. Tabassum R, Gupta BD (2016) Tailoring the field distribution of ZnO by polyaniline for SPR-
based fiber optic detection of hardness of the drinking water. Plasmonics 2016(11):483–492
34. Tabassum R, Gupta BD (2016) Fiber optic hydrogen gas sensor utilizing surface plasmon
resonance and native defects of zinc oxide by palladium. J Opt A Pure Appl Opt 18:015004
35. Tabassum R, Mishra SK, Gupta BD (2013) Surface plasmon resonance based fiber optic
hydrogen sulphide gas sensor utilizing Cu–ZnO thin films. Phys Chem Chem Phys 15:11868–
11874
36. Sullivan BT (1990) Optical properties of palladium in the visible and near UV spectral regions.
Appl Opt 29:13–20
37. Tabassum R, Gupta BD (2015) Surface plasmon resonance based fiber optic hydrogen gas
sensor utilizing palladium supported zinc oxide multilayer and their nanocomposite. Appl Opt
54:1032–1040
38. Tabassum R, Gupta BD (2015) Fiber optic manganese ions sensor using SPR and nanocom-
posite of ZnO–polypyrrole. Sensors Actuators B Chem 220:903–909
39. Tabassum R, Gupta BD (2015) Surface plasmon resonance based fiber optic detection of
chlorine utilizing polyvinylpyrrolidone supported zinc oxide thin films. Analyst 140:1863–
1870
40. Tabassum R, Kaur P, Gupta BD (2016) Tuning the field distribution and fabrication of Al@ZnO
core shell nanostructure for SPR based fiber-optic phenyl hydrazine sensor. Nanotechnol
27:215501
Chapter 9
Frontiers of Light Dynamics in Photonic
Crystals
Renju Rajan, P. Ramesh Babu, and K. Senthilnathan
Abstract In this paper, we briefly review the growth and applications of two
dimensional photonic crystals. We begin with planar photonic crystals which are
useful in integrated photonic circuits. After analyzing various parameters which
determine the performance of these structures, we describe the evolution of photonic
crystal fibers. Photonic crystal fibers, namely, solid core and hollow core fibers,
and their working principles are discussed. Besides, we analyze the linear and
nonlinear properties exhibited by these fibers along with their applications. Design
considerations for these fibers are also briefly addressed. Further, application of
photonic crystal fiber as a sensor is illustrated using surface plasmon resonance
phenomenon. Finally, we delineate the negative index materials that are basically
periodic structures made of metallic wires and split-rings. One of the applications
of these structures in perfect imaging is briefly described.
Keywords Photonic crystals • Photonic crystal fibers • Surface plasmon reso-

nance • Negative index materials • Perfect lens
9.1 Introduction
Crystals are solid structures endowed with regularity in arrangement of their

constituents. They are much sought for their peculiar properties, such as luster in
jewelry industry, electronic properties in semiconductor industry, optical properties
in optoelectronics industry, mechanical properties in gas-turbine industry, to cite
a few. These varied properties arise out of the periodicity in the lattice of the
crystal. The integrity of the crystal depends on the binding force which holds the
constituents together to form the lattice of the crystal [16]. The electrons in a solid
are influenced by the presence of these lattice atoms which perturb the energy
levels of system leading to the formation of energy bands. Due to the interaction
of electron waves with ion cores of the crystal, standing waves are formed at
R. Rajan • P. Ramesh Babu • K. Senthilnathan ()

Department of Physics, School of Advanced Sciences, VIT University, Vellore,
Tamil Nadu 632014, India
e-mail: senthee@gmail.com

DOI 10.1007/978-3-319-48081-7_9
200 R. Rajan et al.
Energy
Third
allowed band
Forbidden
band
Second allowed
band
Forbidden
band
First allowed
band
-2π/a -π/a 0 π/a 2π/a

Wave vector
Fig. 9.1 An extended zone scheme illustrating the allowed and forbidden bands formed in a one
dimensional periodic lattice due to Bragg reflection
particular values of wave vector determined by the Bragg reflection condition [26].
As a result, as shown in Fig. 9.1, a discontinuity called band gap is formed in
the band structure at which electrons cannot propagate through the crystal. This
concept could also be extended to the periodically varying refractive index structures
made of dielectric materials known as photonic crystals (PCs) wherein the photonic
band gap appears due to the reflection of electromagnetic wave according to the
Bragg’s condtion [59]. The initial ideas in this direction were put forward by Eli
Yablonovitch and Sajeev John in 1987 [23, 58]. These materials can be used for
confining light in a region, thereby enabling new methods for controlling light,
analogous to electrons in a semiconductor. Since silicon is a semiconductor and
optical material, an integrated approach to develop hybrid systems containing both
photonic crystals and semiconductor devices has emerged as an area of research
known as silicon photonics [15]. Moreover, applications of photonic crystals in the
field of sensors and medical diagnostics are flourishing at a rapid rate [38]. In this
paper, we give an outline of the major photonic devices that have been materialized
using this concept.
9.2 Planar Photonic Crystals
The name photonic crystal is attributed to a class of artificial periodic dielectric

structures in two or three dimensions which can generate a photonic band gap
(PBG) for electromagnetic waves [60]. Since these structures do not allow light
to propagate in certain directions and frequency ranges, it can be used for spacial
localization of light. These type of structures require high refractive index contrast
along the spacial periods to establish a photonic band gap [35]. Further, periodicity
9 Frontiers of Light Dynamics in Photonic Crystals 201
Silicon
Air hole
Silica
Silicon
substrate
Fig. 9.2 Schematic diagram of a planar photonic crystal with a triangular lattice
of these structures should be of the order of wavelength of light used [37]. Generally,
in the case of 2D photonic crystal structures such as planar photonic crystal (PPC),
perforations made in a silicon slab result in periodic arrangement of air holes and
silicon as shown in Fig. 9.2. A PPC is also known as photonic crystal slab. In contrast
to 3D photonic crystal structures, 2D photonic crystal structures can be easily
fabricated using the planar microfabrication techniques used in semiconductor
industry [4]. This along with other fabrication techniques has resulted in a branch
of optics known as nanophotonics [29].
Due to the periodicity in a plane, a PPC creates a 2D photonic crystal. The third
dimension is nonperiodic and has a finite dimension based on the thickness of the
slab [34]. The leakage of light in this third dimension is prevented by placing the
slab in a lower refractive index material to help achieve the total internal reflection
for confining light within the slab. The band gap properties of PPC depend on many
factors such as the lattice type (e.g. triangular, square, honeycomb etc.), thickness of
the slab, refractive index of the slab and the surrounding medium, periodicity of the
lattice, and size of the air holes. In general, triangular lattice is preferred to square
lattice since the former offers wider band gap and ensures stronger light localization
due to the close packed arrangement.
The periodic structure in PPC creates PBGs which prevent light with certain
frequencies from propagating through the structure. If some defects are created in
the periodic structure, it can be used to localize light in that region. These defects can
be of various forms, such as a point defect which acts like microcavity, a line defect
which results in a waveguide, or a planar defect which could exhibit the properties
of a mirror [22]. All these defects can be put into use for specific applications in
optics. In this way, light is sought for replacing electrons in integrated circuits for
data transfer since it can travel faster than electrons in a dielectric, can carry more
202 R. Rajan et al.
bandwidth of information, and consume less energy due to non-interactive nature of

photons as against electrons. In the near future, photonic crystals in the form of PPC
may turn out to be the workhorse in the photonic integrated circuits.
9.3 Photonic Crystal Fibers
Photonic crystal fiber (PCF) is an extended analogue of planar photonic crystal [27,
59]. The third dimension in PCF can be extended infinitely as against a finite
dimension that is possible for a planar photonic crystal. In the cross section, PCF
consists of silica material in which air holes appear periodically in certain lattice
pattern (e.g. triangular, square, honeycomb etc.) [47]. A triangular lattice is also
known as a hexagonal lattice [65]. Light can be localized in a PCF by making defects
in the periodicity of the structure. A PCF is also known as microstructured fiber
(MSF) or microstructured optical fiber (MOF). According to the type of defect in
the central core region, two types of PCFs are possible, namely, solid-core PCF (SC-
PCF) and hollow-core PCF (HC-PCF). While in the case of SC-PCF, the air hole in
the core region is replaced by a defective solid, the air hole is replaced by a defective
air hole of larger diameter in the case of HC-PCF. The guidance mechanisms are
different for each of these PCFs. In the case of a SC-PCF, the guidance mechanism
is modeled after an effective index approach wherein an effective refractive index
is attributed to the photonic crystal cladding [5, 28]. Unlike index guiding SC-PCF,
guidance in a HC-PCF is due to the photonic band gap effect [3]. Both the PCFs
have their own merits and hence have various applications in photonics. Among the
various methods used for fabricating PCFs, stack-and-draw technique and extrusion
fabrication process are the most commonly used techniques [47]. Figure 9.3 shows
the schematic representation of a PCF with a triangular lattice.
It is well known that the guiding mechanism in a conventional optical fiber is
total internal reflection (TIR) which occurs when light inside the core is incident at
an angle greater than the critical angle at the core-cladding interface. If the angle
Silica
Air hole
Fig. 9.3 Schematic representation of a photonic crystal fiber with a triangular lattice
of incidence is less than the critical angle, a part of light is lost in the cladding due
to refraction. Depending upon the diameter of the fiber core, different modes (field
distributions) of propagation are possible in an optical fiber. Accordingly, the fibers
are classified into multimode fiber (MMF) and singlemode fiber (SMF) [50]. While
a MMF sustains many modes of propagation, a SMF supports only one mode called
fundamental mode. Since the material of the fiber (silica) is essentially dispersive
in nature, different modes travel with different velocities resulting in the spreading
of the pulse called dispersion. This is analogous to dispersion caused by a prism.
Dispersion is disadvantageous in a fiber-optic communication system due to the
overlapping of the pulses, which in turn adversely affects the information carrying
capacity of a fiber. In this way, SMF turns out to be appropriate for carrying data
over long distances due to its low dispersion characteristics.
The waveguiding properties of a PCF can be varied by a proper combination
of diameter of the air-hole .d/ and hole-to-hole spacing .ƒ/. Unlike conventional
SMF having restricted diameter of the core, PCF can have single mode condition as
long as the air-filling fraction (d=ƒ) is below 0.4 [47]. This permits for considering
larger diameter core without being concerned about single mode guidance [33]. This
comes in handy for optical communication where the need for repeaters at regular
intervals can be reduced due to the availability of large amount of power that can be
propagated through the core without being hindered by nonlinear effects [65]. On
the other hand, due to high refractive index contrast between core and cladding,
light is well confined in the core region of a small-core PCF. This opens door
for nonlinear effects that come into picture as a result of high intensity which is
facilitated by small core area and good light confinement in certain PCFs. As a
result, these PCFs are able to produce a wealth of nonlinear effects such as soliton
formation and supercontinuum (SC) generation with ease.
Solitons are localized wave packets that are produced due to a dynamical
balance between dispersion and nonlinearity which exists in an optical fiber. These
robust pulses which travel without spreading over long distances are routinely
used in long-haul fiber-optic communication systems [13]. Solitons can also be
used for generating ultrashort pulses in femtosecond and picosecond ranges [54].
These pulses are peculiar due to a large amount of average power they carry,
and find applications in micromachining, medical treatment and diagnostics [8].
Moreover, ultrashort pulses can be utilized for the generation of supercontinuum
which is a broadband coherent light source [6, 21]. The foremost application
of supercontinuum is in optical coherence tomography (OCT) [2, 19]. This is a
noninvasive technique for imaging biological tissues. Another notable application is
in optical metrology wherein it can be used as a frequency comb [61]. All the above
mentioned applications are also possible in photonic band gap-guided HC-PCF.
Further, due to the very low attenuation levels possible in a large core air-filled HC-
PCF, it can be used for effective beam delivery in high-power laser applications [20].
When the core of a HC-PCF is filled with appropriate nonlinear medium such as gas
or liquid, it exhibits higher order nonlinear effects at a much lower power threshold
when compared to a SC-PCF [7].
204 R. Rajan et al.
9.4 Surface Plasmon Resonance
It is well known that metals are usually not used in device fabrication at optical
frequencies due to their high absorption in this frequency range [30]. But, metals
are betting on a comeback in this frequency regime with their newfound role in
plasmonics [40]. Plasmonics deals with the study of plasmons, which are quanta of
collective electron oscillations (plasma oscillations) that arise in free electron rich
metals and doped semiconductors [10]. Plasmons can be described classically using
the Drude-Lorentz model [10, 36]. Of the variety of plasmons available, surface
plasmons (SPs) which propagate along the metal-dielectric interface are particularly
interesting [1]. Evanescent photons from a prism coupler can be used to excite these
surface plasmons. Due to the coupling between plasmon and evanescent photons
on the surface of a metal, these quasiparticles are referred to as surface plasmon
polariton (SPP) [24]. The wavelength of surface plasmon waves is very small when
compared to light waves. Thus, these waves can be used for miniaturization in the
device fabrication [1]. Currently, sensors form a main area of application for surface
plasmons [39, 55].
Since the wave vector of plasmons is greater than that of free space photons,
photons cannot be directly used for exciting the plasmons. Alternately, the mismatch
in wave vectors is compensated by passing light through a dielectric, resulting in
phase matching between surface plasmon and evanescent wave over the surface.
This results in surface plasmon resonance (SPR) at which plasmons oscillate
with maximum amplitude [24]. Among the various prism coupler based setups,
Kretschmann configuration which is the most widely used is shown in Fig. 9.4b.
This setup consists of a dielectric substrate (prism coupler) on which a thin metal
film is coated. The light undergoes total internal reflection at metal-dielectric
interface. A part of the incident light which evanescently decays (evanescent wave)
through the metal-dielectric interface is instrumental for the formation of surface
plasmon waves. The incident angle at which light falls on metal-dielectric interface
is important for inducing surface plasmons. When surface plasmons are induced
over the metal-dielectric interface, there is a dip in the reflected light intensity due
to the transfer of energy to SPR. The drop in intensity is a direct measure of the
sensitivity in SPR based sensors. The incident angle at which there is a maximum
drop in reflected light is called resonance angle or SPR angle [56]. The surface
plasmon wave spreads out into either sides of the metal layer. The presence of any
material (analyte) over the metallic surface influences the SPR angle. The shift in
the SPR angle with the presence of analyte on one side of the metallic surface can
give inference about its properties. This forms the working principle of a SPR based
sensor. Since this technology uses the changes in refractive index in the vicinity of
metal surface for sensing purpose, this sensor comes under the category of refractive
index sensor [32].
In addition to the prism coupler-based Kretschmann configuration, dielectric
optical waveguide-based sensors are also possible as shown in Fig. 9.4c. PCF
which is a dielectric waveguide can also be used in these sensors since it offers
(a) (b)
Reflected
light
Incident
light Analyte
Gold layer
Gold layer Surface plasmon
Analyte Prism coupler
Air hole
(c)
Surface plasmon mode
Analyte
Metal layer
Waveguiding layer
Substrate
Waveguide mode
Fig. 9.4 (a) PCF-based SPR sensor (b) Prism coupler-based SPR sensor (c) Optical waveguide-
based SPR sensor
miniaturization, ruggedness, and remote sensing possibilities [12, 48, 49]. For
making use of a waveguide for SPR, phase matching should be achieved between
the guided mode and the surface plasmon mode [17, 18]. In effect, this is achieved
by matching the effective refractive indices of both the modes. Figure 9.4a shows
the schematic representation of a PCF integrated with a SPR sensor. In this, the
metal used for generating surface plasmon is gold that is coated inside the outer
ring air holes. The analyte required to be examined is allowed to flow through these
large holes. When the light passes through the central core region, a fraction of the
incident light is utilized to excite the surface plasmon mode on the metallic gold
layer coating. After calibrating the PCF to show variation in the intensity of light
against the properties of the analyte, this can be used as a sensor for analyzing
chemical or biological samples.
9.5 Negative Index Materials
According to Maxwell’s electromagnetic theory, velocity of electromagnetic waves

in a medium can be calculated by its relative permittivity (r ) and permeability
(r ) which are related to the refractive index of the medium by the relation,
p
n D r r [9]. Materials can be classified into different classes based on their
r and r values as shown on Fig. 9.5a. Materials having positive refractive index
appear in the first quadrant. On the other hand, materials of negative refractive
index, also known as negative index materials (NIMs) occupy the third quadrant.
206 R. Rajan et al.
(a) µr (b)
II I
Plasmas, Most known
Metals at materials,
optical Dielectrics
frequencies
No naturally Ferromagnets in
occurring Ghz range,
materials Antiferromagnets
in Thz range
III IV
(c) NIM
Split rings
Object Metallic wire
Profile of evanescent wave
Fig. 9.5 (a) Classification of materials based on r and r values (b) Negative index material
composing of split rings and metallic wires (c) Schematic diagram of negative refraction in a
negative index material
These materials are also known as metamaterials since they exhibit some unique
properties due to their negative refractive index [51]. Metals exhibiting negative r
in the optical frequency range find place in the second quadrant. But, they have
positive value for r in the same frequency range [46]. Ferromagnets in GHz range
and antiferromagnets in THz range occupy fourth quadrant of the classification
scheme [11]. For the first time, it was Veselago who pondered over the concept
of negative index in 1968 [57]. These artificial materials are sought after for their
applications in perfect imaging, cloaking devices, and metamaterial antennas, to cite
a few [42, 45, 64].
Since there are no naturally occurring NIMs, they need to be created artificially.
This can be done by setting negative values for r and r . For this, the desired
electric and magnetic properties of the medium are identified and worked out at
microscopic level to obtain macroscopic property of the medium such as refractive
index. This is because electromagnetic waves do not distinguish between sub-
wavelength structures at the macroscopic level. Attempts in this direction were
initiated by Pendry et al. [43, 44]. The structures they designed were able to achieve
negative permittivity and permeability in microwave region of the electromagnetic
spectrum. While an array of metallic wire structure was used to achieve negative
permittivity, an array of metallic split-rings formed the structure for attaining
negative permeability. Later on, by combining both these structures, negative
refraction was achieved in the microwave region by Smith et al. [52, 53].
Modeling of these structures is based on the Drude-Lorentz model [10, 36].

According to this model, negative permittivity is feasible in metals at frequencies
below the plasma frequency and above the resonance frequency. Since plasma
frequency is a function of number density and effective mass of electrons, an array
of metallic wires with proper thickness and spacial separation distance can be used
for achieving negative permittivity [44]. Similarly, negative permeability can be
introduced by making use of the split-ring resonators (SRRs) composing of LC
oscillators which generate a resonance frequency for the LC circuit [43]. If the
frequency of the incoming magnetic field is greater than the resonance frequency,
the medium would exhibit negative permeability. In this way, an array of metallic
wires and split-rings forms a negative index material for the incoming microwaves
of the electromagnetic spectrum. Further design considerations have enabled to
extend these structures into the optical regime [31, 41, 51, 63]. Since metals have
negative permittivity in optical frequencies, split-ring like structures are suffice to
achieve negative refraction in those frequency range. Figure 9.5b illustrates the
arrangement of split-rings and metallic wires for achieving negative refraction.
One of the fascinating applications of NIMs is in perfect imaging. They can be
used to construct a perfect lens which can increase the sharpness of an image. As it
is known, electromagnetic waves in a medium consist of propagating waves and
evanescent waves. Evanescent waves decay exponentially and do not contribute
for the image formation with conventional refractive medium such as lens. Thus,
a significant amount of information is lost in reaching the image plane. But, in
a NIM, these waves can be amplified by the plasmonic oscillations. Since the
frequency of evanescent wave matches with that of the plasmonic oscillations, the
appearance of resonance leads to amplification of the evanescent wave [25, 42].
In this wave, the amplified evanescent wave also contributes in image formation
with a NIM which results in sharper images. Image formation using a NIM is
shown in Fig. 9.5c. As it is already known, the resolution of an image depends on
the wavelength of radiation [14]. As a result, radiation with smaller wavelengths
is capable of producing higher resolution images, as it is evident from electron
microscopic images. But, it is beneficial to get perfect images in optical frequencies
with as minimum loss of information as possible. Interestingly, NIMs can achieve
this and hence find particularly useful in biological imaging [62].
9.6 Conclusion
Photonic crystals are microfabricated structures whose lattice parameters are of

the order of wavelength of light. This dimensionality of the constituent structure
prevents light from entering these structures unless some defects are created
within the structure. With the introduction of these defective photonic structures,
localization of light has been realized in these structures. Based on the spacial
dimension of photonic crystals, they are classified as 1D, 2D and 3D photonic
crystals. Of these, 2D and 3D structures are considered as the promising products
208 R. Rajan et al.
with potential applications. Compared to 3D structures, fabrication of 2D structures

is easier and economical. This has led to commercialization of some of the 2D
photonic crystals such as planar photonic crystal, photonic crystal fiber, etc. Planar
photonic crystals are useful in semiconductor industry as waveguides for transfer
of data between various components. On the other hand, photonic crystal fibers find
applications in optical networks, nonlinear optics and medical diagnostics. Air holes
of photonic crystal fiber coated with noble metals are particularly useful as sensors
due to the surface plasmon resonance modes supported by them. Further, periodic
structures made of metallic wires and split-rings can be used for achieving negative
refraction. These structures can be used as a perfect lens for lossless imaging by
amplifying the evanescent waves which are usually lost in a conventional lens.
References
424:824
2. Begum F, Namihira Y (2012) Photonic crystal fiber for medical applications. In: Yasin M,
Harun SW, Arof H (eds) Recent progress in optical fiber research. InTech, Rijeka, pp 229–246
3. Bjarklev A, Broeng J, Bjarklev AS (2003) Photonic crystal fibres. Springer Science & Business
Media, Dordrecht
4. Bogaerts W, Wiaux V, Taillaert D, Beckx S, Luyssaert B, Bienstman P, Baets R (2002)
Fabrication of photonic crystals in silicon-on-insulator using 248-nm deep uv lithography.
IEEE J Sel Top Quantum Electron. 8:928
5. Broeng J, Mogilevstev D, Barkou SE, Bjarklev A (1999) Photonic crystal fibers: a new class
of optical waveguides. Opt Fiber Technol 5:305
6. Dudley JM, Genty G, Coen S (2006) Supercontinuum generation in photonic crystal fiber. Rev
Mod Phys 78:1135
7. Dudley JM, Taylor JR (2009) Ten years of nonlinear optics in photonic crystal fibre. Nat Photon
3:85
8. Fermann ME, Galvanauskas A, Sucha G (2003) Ultrafast lasers: technology and applications.
Marcel Dekker, New York
9. Fox M (2006) Quantum optics: an introduction. Oxford University Press, New York
10. Fox M (2010) Optical properties of solids, 2nd edn. Oxford University Press, Oxford
11. Gevorgian S (2009) Ferroelectrics in microwave devices, circuits and systems: physics,
modeling, fabrication and measurements. Springer Science & Business Media, London
12. Hassani A (2006) Skorobogatiy M, Design of the microstructured optical fiber-based surface
plasmon resonance sensors with enhanced microfluidics. Opt Express 14(11):616
13. Haus HA (1993) Optical fiber solitons, their properties and uses. Proc IEEE 81:970
14. Hecht E (2002) Optics, 4th edn. Addison-Wesley, San Francisco
15. Hochberg M, Baehr-Jones T (2010) Towards fabless silicon photonics. Nat Photon 4:492
16. Hofmann P (2008) Solid state physics: an introduction. Wiley-VCH, Weinheim
17. Homola J (2006) Electromagnetic theory of surface plasmons. In: Homola J (ed) Surface
plasmon resonance based sensors. Springer Science & Business Media, Berlin, pp 3–44
18. Homola J, Yee SS, Gauglitz G (1999) Surface plasmon resonance sensors: review. Sens
Actuator B-Chem 54:3
19. Huang D, Swanson EA, Lin CP, Schuman JS, Stinson WG, Chang W, Hee MR, Flotte T,
Gregory K, Puliafito CA et al (1991) Optical coherence tomography. Science 254:1178
20. Humbert G, Knight J, Bouwmans G, Russell PSJ, Williams D, Roberts P, Mangan B (2004)
Hollow core photonic crystal fibers for beam delivery. Opt Express 12:1477
21. Jiang X, Joly NY, Finger MA, Babic F, Wong GK, Travers JC, Russell PSJ (2015) Deep-
ultraviolet to mid-infrared supercontinuum generated in solid-core ZBLAN photonic crystal
fibre. Nat Photon 9:133
22. Joannopoulos JD, Villeneuve PR, Fan S (1997) Photonic crystals: putting a new twist on light.
Nature 386:143
23. John S (1987) Strong localization of photons in certain disordered dielectric superlattices. Phys
Rev Lett 58:2486
24. Kawata S (2012)New technique lights up the creation of holograms. Riken Res 7(3):14–17
25. Kim M, Rho J (2015) Metamaterials and imaging. Nano Converg 2:1
26. Kittel C (1996) Introduction to solid state physics, 7th edn. John Wiley & Sons, New York
27. Knight JC, Birks TA, Atkin DM, Russell PSJ (1996) Pure silica single-mode fibre with
hexagonal photonic crystal cladding. In: OFC Conference, vol. 2
28. Knight JC, Birks TA, Russell PSJ, De Sandro JP (1998) Properties of photonic crystal fiber and
the effective index model. J Opt Soc Am A 15:748
29. Koenderink AF, Alù A, Polman A (2015) Nanophotonics: shrinking light-based technology.
Science 348:516
30. Krauss TF, Richard M (1999) Photonic crystals in the optical regimepast, present and future.
Prog Quant Electron 23:51
31. Le-Van Q, Le Roux X, Aassime A, Degiron A (2016) Electrically driven optical metamaterials.
Nat Commun 7:1
32. Li M, Cushing SK, Wu N (2015) Plasmon-enhanced optical sensors: a review. Analyst 140:386
33. Li W, Chen D, Qinling Z, Hu L (2015) Large-mode-area single-mode-output neodymium-
doped silicate glass all-solid photonic crystal fiber. Sci Rep 5:1
34. Lončar M, Scherer A (2004) Microfabricated optical cavities and photonic crystals. In:Vahala
K (ed) Opt microcavities. World Scientific, Singapore, pp 39–93
35. López C (2003) Materials aspects of photonic crystals. Adv Mater 15:1679
36. Maier SA (2007) Plasmonics: fundamentals and applications. Springer Science & Business
Media, New York
37. Nair RV, Jagatap BN (2011) An engineered forbidden zone for light. BARC Newsletter, p 61
38. Nair RV, Vijaya R (2010) Photonic crystal sensors: an overview. Prog Quant Electron 34:89
39. Nguyen HH, Park J, Kang S, Kim M (2015) Surface plasmon resonance: a versatile technique
for biosensor applications. Sensors 15(10):481
40. Ozbay E (2006) Plasmonics: merging photonics and electronics at nanoscale dimensions.
Science 311:189
41. Padilla WJ, Basov DN, Smith DR (2006) Negative refractive index metamaterials. Mater Today
9:28
42. Pendry JB (2000) Negative refraction makes a perfect lens. Phys Rev Lett 85:3966
43. Pendry JB, Holden AJ, Robbins DJ, Stewart WJ (1999) Magnetism from conductors and
enhanced nonlinear phenomena. IEEE Trans Microw Theory Tech 47:2075
44. Pendry JB, Holden AJ, Stewart WJ, Youngs I (1996) Extremely low frequency plasmons in
metallic mesostructures. Phys Rev Lett 76:4773
45. Pendry JB, Schurig D, Smith DR (2006) Controlling electromagnetic fields. Science 312:1780
46. Pendry JB, Smith DR (2004) Reversing light with negative refraction. Phys Today 57:37–43
47. Poli F, Cucinotta A, Selleri S (2007) Photonic crystal fibers: properties and applications.
Springer, Dordrecht
48. Rifat AA, Mahdiraji GA, Ahmed R, Chow DM, Sua Y, Shee Y, Adikan FM (2016) Copper-
graphene-based photonic crystal fiber plasmonic biosensor. IEEE Photon J 8
49. Rifat AA, Mahdiraji GA, Chow DM, Shee YG, Ahmed R, Adikan FRM (2015) Photonic
crystal fiber-based surface plasmon resonance sensor with selective analyte channels and
graphene-silver deposited core. Sensors 15(11):499
50. Saleh BEA, Teich MC (2007) Fundamentals of photonics, 2nd edn. John Wiley & Sons, New
Jersey
51. Shalaev VM (2007) Optical negative-index metamaterials. Nat Photon 1:41
210 R. Rajan et al.
52. Shelby RA, Smith DR, Schultz S (2001) Experimental verification of a negative index of
refraction. Science 292:77
53. Smith DR, Padilla WJ, Vier DC, Nemat-Nasser SC, Schultz S (2000) Composite medium with
simultaneously negative permeability and permittivity. Phys Rev Lett 84:4184
54. Steinmeyer G, Sutter D, Gallmann L, Matuschek N, Keller U (1999) Frontiers in ultrashort
pulse generation: pushing the limits in linear and nonlinear optics. Science 286:1507
55. Troia B, Paolicelli A, De Leonardis F, Passaro VMN (2013) Photonic crystals for optical
sensing: a review. In: Passaro VMN (ed) Adv Photonic Crys. InTech, pp 241–295
56. Tudos AJ, Schasfoort RBM (2008) Introduction to surface plasmon resonance In: Schasfoort
RBM, Tudos AJ (eds.) Handbook of surface plasmon resonance, pp 1–14. The Royal Society
of Chemistry, Cambridge
57. Veselago VG (1968) The electrodynamics of substances with simulaneously negative values of
and . Sov Phys Usp 10:509
58. Yablonovitch E (1987) Inhibited spontaneous emission in solid-state physics and electronics.
Phys Rev Lett 58:2059
59. Yablonovitch E (2001) Photonic crystals: semiconductors of light. Sci Am 285:46
60. Yablonovitch E (2007) Photonic crystals: What’s in a name? Optics Photon News 18:12
61. Ye J, Schnatz H, Hollberg LW (2003) Optical frequency combs: from frequency metrology to
optical phase control. IEEE J Sel Top Quantum Electron 9:1041
62. Zhang X, Liu Z (2008) Superlenses to overcome the diffraction limit. Nat Mater 7:435
63. Zheludev NI, Plum E (2016) Reconfigurable nanomechanical photonic metamaterials. Nat
Nanotechnol 11:16
64. Ziolkowski RW, Erentok A (2006) Metamaterial-based efficient electrically small antennas.
IEEE Trans Antennas Propag 54:2113
65. Zolla F, Renversez G, Nicolet A, Kuhlmey B, Guenneau S, Felbacq D, Argyros A, Leon-Saval
S (2005) Foundations of photonic crystal fibres. Imperial College Press, London
Chapter 10
LSPR Biosensing: Recent Advances
and Approaches
Pranveer Singh
Abstract Localized surface plasmon resonance (LSPR) based biosensing provide a

sensitive, label-free, cost effective and rapid method to detect biomolecular interac-
tion with nanoscale spatial resolution. This technique has promising applications for
robust and sensitive detection of biomolecular interactions. The portability and nano
sized sensor enables the sensor miniaturization to a scale unapproachable by other
planar technique like surface plasmon resonance (SPR). LSPR-based biosensing
devices can be easily fabricated with inexpensive sensing platforms. The utility of
LSPR based sensing can be enhanced by integrating it with multiplexed microfluidic
devices. Multiplexing ensures accurate diagnostic measurements. Recent progress
in nano-optics has paved the route towards the development of highly sensitive and
label-free optical transducers using the LSPR of metal nanostructures. Nevertheless,
challenges remain in terms of sensitivity, detection limits and utility in point-of-care
diagnostic devices in helping achieving LSPR its full potential.
Keywords Nanoplasmonics • Electromagnetic radiation • Biosensors •

Microfluidics • Biomarker
10.1 Introduction
The metal nanocrystals are endowed with fascinating optical and electrical prop-
erties. The optical properties of metal nanocrystals owe to the interaction of
their free conduction electrons with the incident light. When the incident elec-
tromagnetic wave interacts with metal nanoparticles, localized surface plasmon
resonance (LSPR), that is a collective oscillation of conduction electrons is excited
[1, 2]. In the far field, LSPR leads to absorption and scattering of the incident
electromagnetic waves, which is shown by the emergence of an intense band in
the extinction spectrum. In the near field, intense local electromagnetic fields
within a few nanometers of a particle surface are generated. The near-field effect is
gaining attention in areas that include surface-enhanced Raman scattering (SERS),
P. Singh ()
Department of Zoology, Mahatma Gandhi Central University (MGCU), Motihari, Bihar, India
e-mail: mailpranveer@gmail.com

DOI 10.1007/978-3-319-48081-7_10
212 P. Singh
plasmon-enhanced catalysis and photocatalysis, plasmon-enhanced fluorescence,

the development of plasmonic solar cells, plasmonic solar water splitting process,
and biological imaging and sensing [3–7].
10.2 Advantage Over Other Planar Techniques
LSPR biosensing provides a label-free mode to detect biomolecule with nanoscale

spatial resolution. This technique has promising applications in the study of DNA-
protein interactions [8], toxins [9], proteins [10] and vesicles [11]. The portability
and small sensor size (40–150 nm in diameter) enables the sensor miniaturization
to a scale unapproachable by other planar technique like surface plasmon resonance
(SPR). Also, unique to LSPR is that the nanoplasmonic resonance condition is satis-
fied in a simple reflected or transmitted light geometry common to both microscopy
and spectroscopy applications, whereas SPR excitation requires incident light that
is totally internally reflected. This has given impetus to the LSPR spectroscopy
and imaging (LSPRi) integrated into live-cell microscopy studies for the detection
of secreted proteins while simultaneously imaging the cells with more traditional
techniques such as fluorescence and bright field [12, 13]. Reflective LSPR based
on silver nanoparticles show high stability, reproducibility and enhanced sensitivity.
It can be modified with antigen to act as biosensor [reviewed elsewhere 14].
Multiplexed LSPR is highly selective, show uniform sensitivity, and has provision
for customized optical properties thus making it suitable for high throughput
label-free kinetic analysis of biomolecular binding interactions. Gold-nano rod
based longitudinal LSPR could detect environmentally toxic metal ions [reviewed
elsewhere 14].
Among the label free biosensing techniques, LSPR has emerged as a leader
owing to sensitive, robust, and facile detection of biomolecular interactions. Con-
ventional LSPR-based biosensing technique utilizes the sensitivity of the plasmon
frequency to changes in local refractive index at the nanoparticle surface. LSPR
has now become a gold standard for label free biosensing, nevertheless challenges
remains in terms of sensitivity and limit of detection, selectivity in complex
biological solutions like blood and urine, compatibility with membrane proteins
particularly GPCRs and point of care (POC) diagnostic devices.
This article features the recent approaches and challenges in plasmonic biosens-
ing. This review describes the basis behind the LSPR sensing and summarizes the
latest progress regarding nanostructure fabrication techniques, biosensing appli-
cations, miniaturization and multiplexed sensors that constitute a cutting-edge
research in the LSPR biosensing field. Finally, the challenges that LSPR biosensor
is facing for its commercialization in future is discussed.
10 LSPR Biosensing: Recent Advances and Approaches 213
10.3 Theoretical Concepts
When incident light strikes the noble metal nanoparticles (NPs) like Ag, Au,
Cu and Al, a collective oscillation of conduction band electrons are produced
known as the localized surface plasmon resonance (LSPR), Figs. 10.1 and 10.2.
Resonance condition is met when the incident electromagnetic field matches that of
the oscillating electrons on the surface of the nanoparticle. This resonant oscillation
produces large wavelength-selective increases in absorption, scattering, and electro-
magnetic field at the NP surface. The increases in absorption and scattering have
Fig. 10.1 Localized surface plasmon on a nanoparticle surface (Reproduced from Ref. [98].
© 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license (http://
creativecommons.org/licenses/by/3.0/))
Fig. 10.2 Schematic description of the LSPR biosensing system (Reproduced from Ref. [99].
Copyright © 2008 by MDPI Reproduction is permitted for noncommercial purposes)
214 P. Singh
been utilized towards LSPR biosensing whereas the increases in electromagnetic

fields have been utilized towards surface enhanced Raman spectroscopic (SERS)
detection of biological analytes. Only materials with a negative real and small
positive imaginary dielectric constant are capable of supporting surface plasmons
[15].
10.3.1 Mie Theory
Mie theory explains the factors that affect the large increase in absorption and
scattering due to resonance [16]. Mie theory is an analytical solution to Maxwell’s
equations with spherical boundary conditions, which is used to describe the
extinction spectra of a given nanoparticle. However, for the precise calculation of
the dielectric constants at the different wavelength values and more complex shapes,
the Modified Long Wavelength Approximation (MLWA) of Mie theory is utilized
[17]:
3
24 2 R3 "m2 N "i
Cext D
ln.10/ ."r C "m /2 C "2i
where R is the radius of the particle, œ is the wavelength of the incident light, "m
is the dielectric constant of the surrounding medium, " D "r C i"i is the complex
dielectric constant of the bulk metal, N is the electron density, and ¦ accounts for
the shape of the particle. This “shape factor”, ¦, models the particles as an ellipse
and is proportional to a/b where a and b are the minor and major axis of the ellipse.
As shown in Equation (1), many factors such as shape of the NP, wavelength of
incident light, type of material, and the surrounding media influence the absorption
and scattering processes.
10.3.2 Factors Affecting LSPR Peak Frequency
Binding of a biological analyte to the surface of the NP leads to variation in

refractive index at the nanoparticle surface that in turn shifts the LSPR peak
frequency. The shift of LPSR frequency is also affected by the shape and makeup of
nanoparticle. Among all the plasmonic materials, Silver has the largest negative real
dielectric constant and consequently most sensitive to the changes in local refractive
index. Asymmetrical shape of NPs also affects the changes in biomolecular surface
binding than spherical colloids. The shift in LSPR frequency upon adsorbate binding
has been described by the following relation [10, 18]:

2d
D m .n/ 1 exp
ld
where m is the refractive index sensitivity, n is the change in refractive index

due to adsorbate binding, d is the effective adsorbate layer thickness and ld is
the electromagnetic field decay length (approximated as an exponential decay).
Quantum of LSPR shift is determined by the difference in refractive index of the
absorbate relative to solution (n) and the size of the analyte (d) that is binding
to the NP surface. A plot of LSPR frequency versus refractive index gives the
refractive index sensitivity (m), which is presumed to be linear (within a relatively
small range of index of refraction) according to the Drude model of the electronic
structure of metals [19]. The electromagnetic field decay length, ld , which can also
affect the LSPR shift observed is also sensitive to the shape of the nanoparticle.
The perceived similarity in size of electromagnetic field decay length for many
NPs (50–100 nm in diameter) to that of a protein molecule (5–10 nm) enables
for the unique sensitivity of noble metal NPs to sense the biomolecular binding to
the surface [20, 21]. Refractive index sensitivity is crucial for detecting an analyte
biomolecule; entails the need for NPs substrates that exhibit high sensitivity to
changes in refractive index that increases linearly with LSPR frequency towards
red region of the spectrum [22]. LSPR frequency shift to longer wavelengths can
be achieved by synthesizing larger and/or asymmetric structures and by plasmonic
coupling that couples their electromagnetic fields [23]. The effect of plasmonic
coupling on the LSPR frequency is more pronounced than variations in the refractive
index.
10.4 Biomolecular Biosensing and Nanoplasmonics
Biosensing based on nanoplasmonics is highly dependent on the surface chem-

istry design. Most important being how to maximize the accommodation of the
probability of analyte capture on the reduced surface area. This is crucial as the
most common ligands in biosensing are antibodies, surface conjugated through their
lysine residues or N-terminal amino groups. A typical antibody may have 60–80 of
such residues resulting in a range of surface orientations, many of which are not
optimal for analyte detection. LSPR optimization requires at a minimum sized, low
molecular weight highly oriented full antibodies as ligands which can be modified
to enhance preferential orientation [24].
There are a number of classes of biomolecules which have been engineered with
the goal of meeting these criteria including single chain antibodies (ScFvs), Fab
fragments, and aptamers. Orientated ScFvs, which are roughly one fifth the size of
full antibodies were recently shown to enhance LSPR sensitivity relative to whole
antibody counterparts. However, ScFvs and Fab fragments with molecular weights
of approximately 27 and 50 kDa respectively are still relatively large ligands for
many LSPR biosensing applications requiring further size reduction [25, 26].
When conjugated to plasmonic nanostructures, orientated single domain antibod-
ies (sdAb) exhibited enhanced LSPRi sensitivity which matched that of biotinylated
surfaces to neutravidin, a positive control model system for small and orientated
216 P. Singh
ligand studies. Further, sdAb expressed well in bacteria and yeast, are readily
modified and robust to denaturation make them well suited for a broad range of
SPR and LSPR biosensing applications. The application of LSPRi to ricin detection
is an important step forward in the design of small, readily deployable biosensors
against bio-threat agents [27].
10.4.1 Nanoplasmonics and LSPR

10.4.1.1 Nanoparticles
The plasmonic behavior of metallic NPs are explicitly dependent on their shape,
size and the surrounding dielectric space. The optical properties of the noble metal
nanoparticles (NPs) are dominated by LSPR (Fig. 10.1). The LSPR frequency of
gold and silver NPs are characteristically in the near UV, visible or infrared region
of the spectrum. However, the exact value of this frequency depends on the dielectric
constant of the surrounding media, material, size, shape, and aspect ratio of the NPs
distance between NPs and shape of the NPs clusters [28–30].
Among all metal nanoparticles, Ag nanoparticles (NPs) are important as silver
has the strongest plasmonic interaction with light, greater scattering cross-section,
sharper LSPR bands required for plasmonic application in sensors. The LSPR peak
wavelength of the AgNPs can be tuned throughout the visible and near-infrared
region by their shape, size, and dielectric environment. Compared with spherical or
quasi-spherical AgNPs, the anisotropic morphology such as nanotriangles will show
more LSPR bands with their decreased symmetry and the in-plane dipole resonance
can be easily tailed by their edge length [31]. Similarly, nanotip provides large local
field enhancement and high spatial resolution since most signals are generated from
the tip area. Hermoso et al. [32] have improvised the Ag NTs synthesis protocol by
adjusting the amount of the seeds, or the mole ratio between PVP and AgNO3, thus
tuning the edge length of the silver triangles.
The SPR used to vary with the edge length of the Ag NTs. By optimizing the
dropping speed, and concentration of the trisodium citrate, the Ag NTs with sharper
corner and narrower size distribution can be synthesized thus enhancing the sharper
LSPR bands.
10.4.1.2 Multiplexed Single Nanoparticle LSPR Sensing
LSPR biosensing measurements can be confined to a single chip, i.e. through single
NP-LSPR. This has advantage of less sample consumption, less use of NP material
and higher sensitivity to measure smaller amount of bound analyte. Single NPs
can be easily distinguished on a chip using dark-field microscopy [33]. Single
nanoparticle LSPR biosensing is successfully applied for detection of cancer cell
lines [34].
Diaz-Ezea and co-workers [35] have investigated the individual (single NP)
and collective (>200NPs) LSPR modes of Ag nanoparticle that are aggregated
by covalent binding through bifunctional molecular linkers. Both individual and
collective modes are studied at nanometer scale by means of energy-filtering
transmission electron microscopy (EFTEM) and compared to ultraviolet–visible
spectroscopy. Studies reveal that the aspect ratio and the shape of the clusters
are important contributors to the low-energy collective modes. Study found good
agreement with UV–vis absorption optical measurements.
Surface plasmon polaritons (SPPs) are the electromagnetic waves propagating
at the flat interface between a conductor and a dielectric. These are coupled to an
electric charge density fluctuation on the metallic surface. SPPs on the flat metal
surface, localized surface plasmons (LSPs) arise around the metallic nanoparticles
(NPs) or nanoshells (NSs) [36–38]. LSPs originate from the scattering effect of
a sub-wavelength conductive nanoparticle in an oscillating electromagnetic field.
Thus, different from SPPs, LSPs are non-propagating excitations of the conduction
electrons in metallic nanostructures coupled to the electromagnetic field.
10.4.1.3 Nickel Nanoparticles (NiNPs)
Rahman and co-workers [39] have shown the LSPR effect of Ni nanoparticles
(NiNPs) on the performance of dye-sensitized solar cells (DSSCs). Their study
revealed that NiNPs in a conventional I /I3 electrolyte (NiNPs@I /I3 ) increased
the net optical absorption of a N719 dye over a broad wavelength range by LSPR,
and concurrently improved the power conversion efficiency (PCE) in DSSCs. At
an optimized concentration of the NiNPs@I /I3 electrolyte (1 mg mL1 ), N719-
sensitized DSSCs with a photoanode thickness of ca. 2, 5, and 10 m, exhibited
net PCEs of 2.32, 6.02, and 9.83 %, respectively. These efficiencies were consistent
with a net improvement of 43.2, 20.4, and 12.7 %, respectively and were mainly
attributed to a significant enhancement of the short circuit current density (Jsc ) by
the LSPR from the NiNPs. Similar effects were observed for cells sensitized by the
N3, Ru505, and Z907 dyes. Furthermore, the NiNPs exhibited excellent resistance
to corrosion from a conventional I /I3 electrolyte over a period of 60 days [39].
10.4.1.4 Platinum Nanoparticles (PtNPs)
An improvement in light extraction efficiency due to the LSP of Pt NPs and the
surface roughing of Si nanopillars can have potential application in the fields of
silicon-based light-emitting applications, nanoscience and nanotechnology. LSP
is one crucial way to enhance light emission due to coupling between the LSP
of metallic NPs and the band-gap emission of materials. Surface roughening by
forming the anti-reflection surface and reducing the surface reflection is another
effective way to improve the light extraction ratio. These two methods were utilized
to enhance the light emission using a Si QDs /SiO2 multilayer film on a Pt
218 P. Singh
100 nm
nm
800
600
400
200
Fig. 10.3 Atomic force Microscopy of Pt nanoparticles coated Si nanopillar (Reproduced from
Ref. [100]. This work is licensed under a Creative Commons Attribution 4.0 International License.
The images or other third party material in this article are included in the article’s Creative
Commons License)
NPs-coated Si-nanopillar array substrate (Fig. 10.3). The Pt NPs were used for their
LSPs and the nanopillar array was used for surface roughening. The result shows
that electroluminescence intensity was significantly enhanced [40].
10.4.1.5 Silver and Gold Nanoparticles (AgNPs/AuNPs)
In a study by Yokota et al. [41], monolayer structures of Au and Ag nanoparticles

(NPs) were fabricated by a dipping method to realize the control of LSPR energy.
The mean inter-particle distance in the monolayer was controlled by changing the
concentration of NPs in the colloidal solution used for the monolayer assembly. The
extinction-peak energy of the monolayer structure was red-shifted with decreasing
inter-particle distance, reflecting plasmon coupling between NPs [41].
Tanaka et al. [42] have investigated the characteristics of LSPR excited on
mixed monolayers composed of self-assembled Ag and Au nanoparticles (AgNPs
and AuNPs, respectively). Mixed monolayered films were fabricated at the air-
water interface at different mixing ratios. The films retained their phase-segregated
morphologies in which AuNPs formed several 10–100 nm island domains in a
homogeneous AgNP matrix phase. The LSPR bands originating from the self-
assembled domains shifted to longer wavelengths as the domain size increased,
as predicted by a finite-difference time-domain (FDTD) simulation. The FDTD
simulation also revealed that even an alternating-lattice-structured two-dimensional
(2D) AgNP/AuNP film retained two isolated LSPR bands, revealing that the
plasmon resonances excited on each particle did not couple even in a continuous
2D sheet, unlike in the homologous NP system. The fluorescence quenching test
of Cy3 and Cy5 dyes confirmed that the independent functions of AuNPs and
AgNPs remained in the mixed films, whereas the AuNPs exhibited significantly
higher quenching efficiency for the Cy3 dye compared with AgNPs due to the
overlap of the excitation/emission bands of the dyes with the AuNP LSPR band.
Various applications can be considered using this nano hetero-structured plas-
monic assembly to excite spatially designed, high-density LSPR on macroscopic
surfaces [42].
Hong et al. [43] reported the characteristics of LSP-enhanced near-UV light-
emitting diodes (NUV-LEDs) fabricated by colloidal silver nanoparticles (Ag NPs).
Colloidal Ag NPs were deposited on the p-GaN spacer layer using a spray process.
This has resulted in the enhanced optical output power of NUV-LEDs with colloidal
Ag NPs compared with NUV-LEDs without colloidal Ag NPs. The enhancement
was attributed to increased internal quantum efficiency caused by the resonance
coupling between excitons in the multiple quantum wells and the LSPs in the Ag
NPs [43].
10.4.2 Nanotubes (NTs) – Nanoparticle (NPs) Hybrids
Multicomponent assemblies like nanotube (NTs) /nanoparticle (NPs) hybrids are

crucial for the next generation optical platforms for enhanced biochemical sensing.
These nanostructures that link one-dimensional carbon nanotubes (CNTs) with
zero-dimensional gold nanoparticles (AuNPs) have potential application in nan-
odevices for biochemical sensing, cell imaging, drug delivery tools, and supports
for catalysts etc. Gold nanostructures alone and gold – polymer nanocomposites
are successfully utilized for the detection of biomolecules [44–48]. Similarly,
biosensors based on Carbon Nanotubes (CNTs) were also utilized with success for
the electrochemical detection of protein binding [49, 50], Fig. 10.4. The sensitivity
of detection could be enhanced further by augmenting the set up with AuNP-Ab
conjugates in a CNT field effect transistor [51].
Fig. 10.4 Carbon nanotubes

(CNT) coated with gold
nanoparticle for plasmonic
detection. The SEM image of
the Au-CNT is shown in inset
(Reproduced from Ref. [53].
This work is licensed under a
Creative Commons
Attribution 4.0 International
License. The images or other
third party material in this
article are included in the
article’s Creative Commons
License)
220 P. Singh
10.4.2.1 Single-Wall Carbon Nanotubes (SWCNTs)
LSPR affects the optical absorption and scattering of materials of nano sized dimen-
sion. The intensities and peak energies of the surface plasmons strongly depend on
the carrier density, which can be manipulated by the density of injected carriers. This
was demonstrated by Iragashi et al. [52], where high-carrier injection by electric
double layers can induce a transverse surface-plasmon peak with tunability in peak
position and intensities in aggregated, electricity-selected SWCNTs. This has the
advantage of controlling plasmonic optical processes by electric double-layer carrier
injections using ionic liquid. Single-wall carbon nanotubes (SWCNTs) have great
potential in plasmonic and optoelectronic applications [52].
10.4.2.2 Multi Wall Carbon Nanotubes (MWCNT)
Ozikandathil et al. [53] have synthesized Multi Wall Carbon Nanotubes (MWCNT)
exhibiting LSPR properties for biosensing study of traces of recombinant bovine
growth hormones (rbST). The sensitivity was found to be better than with AuNPs.
High surface to volume ratio and mobility of electrons in CNTs make them
very good sensing material which could be further enhanced by addition of Au
nanoclusters [54]. AuNPs provide novel attachment sites for implementing LSPR
phenomena important for range of sensing applications of the Au-MWCNT hybrids.
Integration of Au enhances the ultrasensitive detection of biomolecules utilizing the
plasmonic properties of gold. Plasmonic coupling of CNTs and AuNPs were utilized
for SERS sensing [55], Fig. 10.5.
10.4.3 Gold Nanorod (GNR)
Park et al. [56] have devised a sensor for detection of mercury in nano to milimolar
concentration range. This simple, receptor-free and regeneratable mercury sensor
is based on LSPR shifts of a gold nanorod (GNR). It works by the coordination
of mercury to citrate, which alters the local refractive index (RI) around the GNR
surface on glass slide, promoting a red-shift in its LSPR absorption peak. This
sensor can be reused sufficient times by repeated steps of washing and precipitation.
This sensor also displayed excellent recoveries in spiking test with real water
samples, such as tap water, lake and river [56].
Lee et al. [57] have devised a LSPR nanorod gas sensor for the detection of
volatile organic compound (VOC) gases. The Au-Ppy nanorods were synthesized
by an anodic aluminum oxide template by the electrochemical deposition. Field
emission scanning electron microscopy was employed to confirm the uniform size
fabrication of Au-Ppy nanorod arrays. Gold deposition ensured that the Au-Ppy
nanorods exhibited both optical and LSPR interference. This sensor was employed
1 Glass Substrate 2 Glass Substrate
3 Glass Substrate 4 Glass Substrate
Multiwall Carbon Nanotube

on glass substrate
Glass Substrate
Gold nanoparticle
Thiol linker
Activating linker(EDS-NHS)
5 Glass Substrate Anti-rbST
rbST
Fig. 10.5 Biosensing protocol using Au-MWCNT nanocomposite as a sensing platform (1. Au-
MWCNT deposited on the glass substrate, 2. Au nanoparticles with a thiol linker, 3. Activating
linker (EDS-NHS), 4. Anti-rbST is attached to gold through the linker, and 5. RbST (antigen)
is attached to the antibody (Reproduced from Ref. [53]. This work is licensed under a Creative
Commons Attribution 4.0 International License. The images or other third party material in this
article are included in the article’s Creative Commons License)
to test acetic acid, benzene, and toluene with a short response time (1 min) over
a wide concentration range. Acetic acid gave the highest sensitivity having linear
relation with the gas concentration, indicating that the system can be used as a gas
sensor [57].
10.4.3.1 Potential-Scanning LSPR Sensors
Nishi et al. [58] have proposed novel potential-scanning LSPR sensors where
potential scanning controls the electron density of the plasmonic nanoparticles. The
sensor exhibits a resonance peak during the potential scan that negatively shifts with
increasing refractive index. Therefore, it scans potential rather than the wavelength.
222 P. Singh
Portability and economy of cost are the advantages of potential-scanning LSPR

sensors over the conventional LSPR sensors. Use of nanorods and nanospheres
further improves the sensitivity of the sensor [58].
10.4.4 Nanowell
Nanowell structure has potential application in plasmonics like SERS nanocrystal

synthesis and biomolecular immobilization as a plasmonic sensors and biomolecular
docking system platforms. The nanowell dimensions can be controlled to produce
an optically tunable LSPR for its use as a plasmonic sensor. It was confirmed
by the AFM and LSPR spectroscopy that the triangular Au nanowell structures
are partially embedded in the surface of a transparent glass substrate to protect
the shape and sharp features of the nanowell. Experimental spectral results and
numerical calculations show that the extinction maximum of the LSPR is located
in the NIR range, and correlates linearly with the base thickness of nanowell.
Numerical calculations to analyze the extinction spectrum of gold nanowell show
that scattering of the nanostructure is dominant to compare to its absorption [59].
10.4.5 Platinum Nanorings
Shin et al. [40] have reported the enhancement of deep ultraviolet emissions
from AlGaN-based quantum wells (QWs) using energy-matched LSPs in platinum
(Pt) nanoring arrays. The peak resonances of the extinction spectra were shifted
to the red spectral region with increasing the nanoring diameters. The emission
enhancement is attributed to the strong coupling of QWs with LSPs from the Pt
nanorings [40].
10.4.6 Gold Nanoantennas
Sardana et al. [60] have prepared arrays of different sized disk shaped gold
nanoantennas on glass showing LSPR and Rayleigh anomalies in the near infrared
and telecom range between 1000 and 1500 nm wavelength. The spectral position of
these grating resonances depends critically on the period of the array and the size
of the nanoantennas. When PbS quantum dots embedded in PMMA surround the
nanoantennas, an up to fourfold enhancement of the photoluminescence is observed
at the grating resonances due to the constructive diffractive feedback among neigh-
boring antennas. In accordance with the grating resonances a shift of the emission
towards smaller wavelengths with decreasing disk diameter is demonstrated [60].
10.4.7 LSPR in Gold (Au) Clusters
Addition of an aqueous solution of NaBH4 to a dispersion of small (1.2 nm)

gold clusters stabilized by poly(N-vinyl-2-pyrrolidone) (Au:PVP) induced a LSPR
absorption for a certain period of time while maintaining the cluster size. The dura-
tion of the LSPR band could be lengthened by increasing the NaBH4 concentration
and shortened by increasing the concentration of dissolved O2 , and the LSPR band
could be made to appear and reappear repeatedly. The appearance of the LSPR band
is explained by the electron donation to the Au core from the adsorbed H atoms that
originate from NaBH4, whereas its disappearance is ascribed to the removal of H
atoms by their reaction with O2 . These results suggest that the transition between the
metallic and non-metallic electronic structures of the Au clusters can be reversibly
induced by the adsorption and desorption of H atoms, which are electronically
equivalent to Au [61].
10.5 Localized and Propagating SPR
Nezami and Gordon [62] have investigated and optimized criterion for the influence
of localized and propagating SPs on third harmonic generation from rectangular
apertures in metal films. To achieve this, optimal aperture array structures were
designed by using FDTD simulations with nonlinear scattering theory. This set up
was used to measure the third harmonic in the region of maximal performance. The
highest third harmonic conversion efficiency (double fold when both localized and
propagating are tuned to the fundamental wavelength) obtained when the localized
resonance is tuned to the fundamental wavelength and the propagating (Bragg)
resonance is tuned to the third harmonic. However, due to the inherent absorption of
gold at the third harmonic, tuning the LSPR to the third harmonic is less beneficial.
All cases showed quantitative agreement with the original theoretical analysis [62].
10.6 LSPR Imaging
LSPR spectroscopy is widely used for label free, high sensitive analysis of
interactions at the surface. LSPR spectroscopy integrates nanoparticle plasmon
resonance and adsorbate molecular resonance (Figs. 10.2 and 10.6). LSPR imaging
(LSPRi) enables high throughput multiplexed measurements by simultaneously
probing multiple sensors on single sample surface. LSPR in conjunction with
scanning tunneling microscopy (STM), electrochemical detection based on direct
electron transfer in virus, and SERS using plasmonic nanoparticle can be utilized
for detection of HIV-1 [63]. Similarly, integrating electrophoresis device with
224 P. Singh
Fig. 10.6 Localized surface plasmon resonance (LSPR) enabled label-free optical detection of
analyte biomolecules (Reproduced from Ref. [101]. Copyright © 2007 by MDPI (http://www.
mdpi.org). Reproduction is permitted for noncommercial purposes)
Light source
Inlet Outlet
Spectrometer
Glass on the top Glass in the middle Glass at the bottom
5 mm 5 mm
1 mm
Inlet Outlet 18 mm
Φ=1 mm Flow
channel
18 mm
Fig. 10.7 LSPR sensing system integrated with the microfluidic device synthesized on the glass
substrate. The top image shows an assembled LSPR sensor chip and the bottom three glass slices
with through holes and a fluidic channel (Reproduced from Ref. [102]. © 2014 by the authors;
licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the
terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/
licenses/by/3.0/))
microfluidics on LSPR can detect reactive oxygen species (ROS) with less sample
consumption and time, [reviewed elsewhere 14] (Figs. 10.7 and 10.8).
Vasconcelos and group [64] have devised a strategy to produce efficient scat-
tering type near-field optical microscopy probes utilizing focused-ion-beam milled
single groove near the apex of electrochemically etched gold tips. Electron energy-
loss spectroscopy and scanning transmission electron microscopy are employed to
obtain highly spatially and spectroscopically resolved maps of the milled probes,
Fig. 10.8 Schematic of the detection system with high sensitivity enhanced by LSPR. The DCHF
are loaded into a sample inlet and the ROS is trapped from puffs of cigarette smoke mainstream.
The generated DCF is quantified using the microfluidic electrophoresis system after the trapping
process. A laser-induced fluorescence (LIF) detection system is used to detect fluorescent signal
(Reproduced from Ref. [102]. © 2014 by the authors; licensee MDPI, Basel, Switzerland. This
article is an open access article distributed under the terms and conditions of the Creative Commons
Attribution license (http://creativecommons.org/licenses/by/3.0/))
revealing LSPR at visible and near-infrared wavelengths. By varying the distance

between the groove and the probe apex, the LSPR energy can be fine-tuned at a
desired absorption channel. The said method was validated by Tip-enhanced Raman
spectroscopy [64].
10.7 LSPR Biosensing in Point-of-Care Diagnostic Devices
LSPR-based biosensing devices can be easily fabricated with inexpensive sensing

platforms. This platform enables sensitive, label-free detection with cost effective,
rapid and portability in the measurement [65]. The utility of LSPR based sensing can
be enhanced by integrating it with multiplexed, microfluidic devices. Multiplexing
will ensure accurate diagnostic measurement as assay utilizing multiple biomarker
lead to false diagnosis than single biomarker tests.
Haes et al. [66] has pioneered the application of LSPR biosensing when
this technique was utilized for the detection of amyloid-derived diffusible ligand
(ADDL) marker for Alzheimer’s disease. This was followed by the surge in the
LSPR sensing for POC. Demirci and group have accurately identified subtypes
of HIV using LSPR [63, 67–70] Fig. 10.9. Other application of LSPR includes
detection of glucose in urine [71], blood plasma [72], serum [73] and cerebrospinal
fluid [74], detection of virus and cancer [75–78].
226 P. Singh
HIV - infected Patient Samples

6
Nanoplasmonic Virus Count)
log10 (HIV copies/mL by
1 Y = 2.50X - 4.93
R2 = 0.69
0
2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4
log10 (HIV copies/mL by RT-qPCR Virus Count)
Fig. 10.9 Schematic of HIV detection assay (b) HIV subtype detection (Reproduced from Ref.
[68]. © 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/3.0/))
10.7.1 LSPR Gold Nanobiosensor for Immunoassay
For cancer therapy, photothermal ablation/optical hyperthermia are a minimal inva-

sive approach. Photothermal transducer when combined with NIR (700–1100 nm
light is preferred for such an application as it can penetrate soft tissues deeply
owing to the relatively low absorption/scattering by hemoglobin and water in this
so-called transparent window), all of them were able to generate sufficient heat to
raise the local temperature and thus killing cancer cells. Recently, Au nanostructures
have received great interest due to the fact that their LSPR peaks can be easily
tuned to the NIR region by altering their size, shape, structure, or a combination
of these parameters. A wide variety of Au nanostructures, including aggregates
of colloidal particles, nanoshells, nanocages, nanorods, and nanocrosses have been
demonstrated for photothermal cancer therapy with NIR light. LSPR of plasmonic
nano-particles opens up new avenues for biosensing and nanomedicine, [reviewed
elsewhere 14].
Faridli et al. [79] have developed a LSPR based immunoassay to determine
prolactin hormone in human serum samples. This involved electrostatic adsorption
of antibody on the surfaces of gold nanoparticles acting as a probe optimizing the
pH and the antibody-to-gold nanoparticle ratio, as the factors affecting the probe
functions. This nanobiosensor was characterized by dynamic light scattering and
applied for the determination of prolactin antigen concentration and prolactin in
human control serum sample based on the amount of LSPR peak shift [79].
10.8 LSPR for Biomolecular Assays
10.8.1 Poly(methyl methacrylate) Polymer Based LSPR

Biosensor Chips
Poly(methyl methacrylate) polymer based LSPR biosensor chips were fabricated

using glycopolymer brushes (for enhancement of detection sensitivity) carrying
glucose moieties for the detection of concanavalin A. Poly(pentafluorostyrene),
with pre-determined polymer chain lengths, were synthesized via a reversible
addition-fragmentation chain transfer polymerization technique. The synthesized
poly(pentafluorostyrene), was subsequently converted into glycopolymers via a
para-fluoro-thiol click reaction and grafted onto the surface of sensor chips. The
glycocluster effect induced by pendent carbohydrate moieties enabled a stronger
affinity for concanavalin A binding, which resulted in a dramatic expansion of the
sensors’ response range. In addition to the superior performance, the capabilities
of the reported sensor chips can be easily manipulated to detect a diverse range of
analytes by clicking various sensing elements onto the polymer brushes [80].
10.8.2 LSPR Enhanced Visible-Light Photoelectrochemical

Biosensors
Yan et al. [81] have reported a novel visible-light photoelectrochemical (PEC)

biosensor based on LSPR enhancement and dye sensitization for highly sensitive
analysis of protein kinase activity with ultralow background. In this strategy,
DNA conjugated gold nanoparticles (DNA@AuNPs) were assembled on the phos-
phorylated kemptide modified TiO2/ITO electrode through the chelation between
Zr(4C) ions and phosphate groups followed by the intercalation of [Ru(bpy)3](2C)
into DNA grooves. The adsorbed [Ru(bpy)3](2C) can harvest visible light to
228 P. Singh
produce excited electrons that inject into the TiO2 conduction band to form
photocurrent under visible light irradiation. In addition, the photocurrent efficiency
was further improved by the LSPR of AuNPs under the irradiation of visible light.
Moreover, because of the excellent conductivity and large surface area of AuNPs
that facilitate electron-transfer and accommodate large number of [Ru(bpy)3](2C),
the photocurrent was significantly amplified, affording an extremely sensitive PEC
analysis of kinase activity with ultralow background signals. The biosensor also
showed excellent performances for quantitative kinase inhibitor screening and
PKA activities detection in MCF-7 cell lysates under for skolin and ellagic acid
stimulation. The developed dye-sensitization and LSPR enhancement visible-light
PEC biosensor shows great potential in protein kinase related clinical diagnosis and
drug discovery [81].
10.8.3 Electrofocusing-enhanced Localized Surface Plasmon

Resonance Biosensors
Zhang et al. [82] have devised a peptide-modified plasmonic gold nanohole (AuNH)
array for real-time detection of human troponin I. Applying a negative electric
bias on the AuNH sensor chip enables them to attract and concentrate human
troponin I at the sensor surface, while repelling other proteins thus decreasing
interferences due to nonspecific adsorption. This circumvents the disadvantage of
LSPR biosensors from diffusion limited mass transport and nonspecific adsorption
upon detection of biomolecules in real biofluids [82].
10.8.4 High-sensitivity Detection of ATP Using a Localized

Surface Plasmon Resonance (LSPR) Sensor and Split
Aptamers
LSPR aptasensor for adenosine triphosphate (ATP) detection in E.coli was devel-
oped by Park and group [83]. The sensor platform includes two split ATP aptamers,
one being the receptor fragment covalently attached to the surface of a gold nanorod
(GNR) and the other labeled with a random DNA sequence and probe fragment
(TAMRA dye). This results in shift in the wavelength of the LSPR band due to
decrease in the distance between GNR surface and TAMRA dye and an associated
LSPR wavelength. The sensitivity of the proposed sensor lies within the picomolar
to micromolar range and is reusable by salt addition and washing steps [83].
10.9 LSPR Biosensing: Challenges and Solutions
Recent progress in nano-optics has paved the route towards the development
of highly sensitive and label-free optical transducers using the LSPR of metal
nanostructures. The first major limitation is inadequate sensitivity and detection
limits, especially for refractive-index based biosensing. Sensitivity can be improved
by changing the nanoparticle shape and material but it will reduce the limit
of detection by as much as tenfold. Second critical issue is selectivity in com-
plex biological solutions like blood and urine. Selectivity can be improved by
self-assembled monolayers (SAM) on the NP surface, biological scaffolds, and
size/shape complementarity. Next, the challenges of carrying out biosensing with
membrane proteins and the need to mimic natural protein environment on the
sensor surface. The ultimate holy-grail is to integrate the microfluidic devices with
plasmonic sensing structures for simple POC detection.
10.9.1 Improving Limit of Detection
LSPR-based biosensing measurements detect changes in refractive index at the

nanoparticle surface. The detection of small molecules, which require large numbers
to coat the surface remains problematic. The first major limitation is inadequate
sensitivity and detection limits, especially for refractive-index based biosensing.
Sensitivity can be improved by changing the nanoparticle shape and material but
it will reduce the limit of detection by as much as tenfold. To increase the limit of
detection of analytes, many strategies have been implemented [27].
10.9.1.1 Improving Limit of Detection: Enzymatic Amplification
The idea is to trigger a reaction-cascade when analyte binds to plasmonic substrate.

The by-products from these reactions can yield an amplified response [27].
10.9.1.2 Improving Limit of Detection: Plasmonic Coupling of NPs
Plasmonic coupling occurs when NPs come into close proximity, which leads to
enhanced shifts in LSPR frequencies towards red region of the spectrum. This
happens due to the coupling of electromagnetic field on account of plasmonic
coupling. This can be achieved by putting the analyte on the surface of NP,
which then binds a fixed nanoparticle array, or through large-scale nanoparticle
aggregation in solution through analyte binding [27].
230 P. Singh
b
0.8
0.4
Δλ/λ(x10 )
0.0
-2
non-specific
-0.4
-0.8 target
-1.2
-1.6
0 5 10 15 20 25
Time [h]
Fig. 10.10 DNA detection in complex serum (Reproduced from Ref. [25]. © 2014 by the authors;
licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms
and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/
by/3.0/))
10.9.1.3 Improving Limit of Detection: Biomolecular

Conformationally-Gated Amplification
Signal amplification can also be achieved through the integration of NPs into
various biological species. For example, integration of DNA with NPs for DNA
detection through sequence specific hybridization or for detection of DNA binding
proteins through confirmation change mediated variation in plasmonic coupling
of bound NPs [84–89]. One such study by Xu et al. [88] involved the screening
of endonuclease activity through colorimentric measurements in the presence of
inhibiters using DNAse I in the presence of strong and weak inhibitors (Fig. 10.10).
Li et al. have utilized NP-DNA conjugate to detect restriction endonucleases at the
femtomolar level [89].
10.9.2 Selectivity Improvisation in Complex Biological

Solution
LSPR sensors when introduced to complex biological solutions may bind with
unwanted spurious compounds or analytes of non interest leading to false, sizable
red-shifts in the LSPR frequency. These compounds may also block the binding of
the analyte of interest. This could be circumvented by self-assembled monolayers

(SAM) on the NP surface, biological scaffolds, plasmonic coupling and size/shape
complementarity. Selectivity is largely determined by the functionalization layer on
the NP surface [27].
10.9.2.1 Selectivity Improvisation: Functionalization Layers
These functionalization layers are made up of small molecules that attract specific
protein of interest to bind. For example, boronic acid ligands are used for the
detection of glucose [90].
10.9.2.2 Selectivity Improvisation: Biological Scaffolds
LSPR-based biosensing does not provide any information about the molecular
identity of the species in binding reaction. In addition, the detection of unwanted
molecules is a teething issue. Non-specific binding of biomolecules or biofouling
of plasmonic materials results in a red shift of the LSPR wavelength as these
species increase the refractive index at the NP surface. As detection arises out of NP
coupling (or the space between NPs), the variation in LSPR frequency will be sub-
stantially larger than that observed for refractive index sensing. However, biofouling
related red-shifts in LSPR frequency only minimally affect the measurement. Also,
there can be a situation where a change in biological conformation may cause NPs
moving apart upon binding a specific substrate, which would cause a blue-shift in
the LSPR spectrum opposite of what is observed for biofouling. This can easily
distinguish binding of unwanted species from the analyte of interest [27].
10.9.2.3 Plasmon Ruler
The development of plasmon ruler has greatly facilitated the use of biological
scaffolds for selectivity. The plasmon ruler usually couples NPs together by dsDNA
[91, 92]. Blue shift in LSPR frequency occurs when an analyte specific for a DNA
sequence binds to the DNA that has coupled the two NPs together leading to
spacing between the NPs. This approach was successfully applied for the purified
DNA, DNA hybridization assays [92] and also in complex biological solutions
[93]. ssDNA is more flexible compared to the dsDNA. Upon DNA hybridization
a rigid DNA dimer is produced increasing the space between the two NPs,
decreasing the plasmonic coupling and causing a spectral blue shift that is specific
for a complementary strand of DNA. Ginger and co-workers have detected DNA
hybridization in serum before nonspecific binding overcame the spectral blue shifts
produced by reduced plasmonic coupling, and instead began to produce spectral red
shifts [93].
232 P. Singh
10.9.2.4 Selectivity Improvisation: Size-Selective Films or Shape

Complementarity
This can be achieved by incorporating size selectivity into the self-assembled

monolayer (SAM) on the NP surface. One approach is to create pores on the SAM
by which only certain molecules can pass. These pores can be created by using NPs
of different sizes [94]. This way SAM can effectively repel unwanted species [27].
10.9.3 LSPR Biosensing for Membrane-Associated Species
LSPR biosensing plays important role in biorecognition and protein-protein inter-

action [95]. Therefore, the application of LSPR biosensing needs to be extended to
the study of membrane protein interactions and protein receptors [27].
10.9.3.1 LSPR Biosensors Utilizing Supported Lipid Bilayers
Effective sensing volume measured by LSPR is smaller than that for SPR. There-
fore, except for liposomes based studies [11, 96], LSPR sensing has utilized the
supported lipid bilayer systems with lipids and associated species reside in close
proximity to the support. LSPR based membrane biosensors utilize supported
lipid bilayers and holes in thin gold film. Supported lipid bilayers mimic the
natural membrane environment [97]. Thin metal films perforated with nanoholes
exhibit LSPR confined to the nanohole. These structures were prepared over quartz
surfaces. Sensing platform was designed by encasing the nanohole structures in
thin film of silica that prevents quenching and supports fluid layer formation
confirmed by Fluorescence recovery after photobleaching measurements (FRAP).
This platform allows real time monitoring of supported lipid bilayers enabling
accurate determination of kinetic parameters. However, silica encasing leads to drop
in refractive index sensitivity [27].
10.10 Conclusion
There has been a spurt in label-free biosensing based on metallic nanoparticles

supporting localized surface plasmon resonances (LSPR). This renewed interest
has surpassed the utility of LSPR over surface plasmon resonance (SPR) approach
based on extended metal films. LSPR biosensing features a high-end miniatur-
ization potential and a significant reduction of the interrogation device bulkiness,
positioning itself as a promising candidate for point-of-care diagnostic and field
applications. However, much needs to be done to meet the challenges of detection
limit, sensitivity, point-of-care diagnostics, diffusion limited mass transport and

nonspecific adsorption upon detection of biomolecules in real Biofluids before
LSPR can attain the status of irreplaceable tool and technique in biomolecular
sensing.
Acknowledgement Valuable inputs from various sources are gratefully acknowledged.

Funding This research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors.
References
1. Lu X, Rycenga M, Skrabalak SE, Wiley B, Xia YN (2009) Chemical synthesis of novel

plasmonic nanoparticles. Annu Rev Phys Chem 60:167
2. Kelly KL, Coronado E, Zhao LL, Schatz GC (2003) The optical properties of metal
nanoparticles: the influence of size, shape, and dielectric environment. J Phys Chem B
107:668
3. Kaur R, Pal B (2013) Co-catalysis effect of different morphological facets of as prepared Ag
nanostructures for the photocatalytic oxidation reaction by Ag-TiO2 aqueous slurry. Mater
Chem Phys 143:393
4. Noh HS, Cho EH, Kim HM, Han YD, Joo J (2013) Organic solar cells using plasmonics of
Ag nanoprisms. Org Electron 14:278
5. Potara M, Boca S, Licarete E, Damert A, Alupei MC, Chiriac MT et al (2013) Chitosan-coated
triangular silver nanoparticles as a novel class of biocompatible, highly sensitive plasmonic
platforms for intracellular SERS sensing and imaging. Nanoscale 5:6013
6. Sun QC, Mundoor H, Ribot JC, Singh V, Smalyukh II, Nagpal P (2014) Plasmon
enhanced energy transfer for improved upconversion of infrared radiation in doped-lanthanide
nanocrystals. Nano Lett 14:101
7. Wu J, Peng Y, Susha AS, Sablon KA, Chen H, Zhou Z et al (2015) Broadband efficiency
enhancement in quantum dot solar cells coupled with multispiked plasmonic nanostars. Nano
Energy 13:827
8. Endo T, Kerman K, Nagatani N, Takamura Y, Tamiya E (2005) Label-free detection of peptide
nucleic acid-DNA hybridization using localized surface plasmon resonance based optical
biosensor. Anal Chem 77:6976
9. Park JH, Byun JY, Mun H, Shim WB, Shin YB, Li T, Kim MG (2014) A regeneratable, label-
free, localized surface plasmon resonance (LSPR) aptasensor for the detection of ochratoxin
A. Biosens Bioelectron 59:321
10. Haes AJ, Van Duyne RP (2002) A nanoscale optical blosensor: sensitivity and selectivity
of an approach based on the localized surface plasmon resonance spectroscopy of triangular
silver nanoparticles. J Am Chem Soc 124:10596
11. Dahlin AB, Jonsson MP, Hook F (2008) Specific self-assembly of single lipid vesicles in
nanoplasmonic apertures in gold. Adv Mater 20:1436
12. Raphael MP, Christodoulides JA, Delehanty JB, Long JP, Byers JM (2013) Quantitative
imaging of protein secretions from single cells in real time. Biophys J 105:602
13. Oh BR, Huang NT, Chen WQ, Seo JH, Chen PY, Cornell TT, Shanley TP, Fu JP, Kurabayashi
K (2014) Integrated nanoplasmonic sensing for cellular functional immunoanalysis using
human blood. ACS Nano 8:2667
14. Singh P (2016) Surface plasmon resonance: historical perspectives and current challenges.
Sensors Actuators B Chem 229:110
234 P. Singh
15. Myroshnychenko V, Rodriguez-Fernandez J, Pastoriza-Santos I, Funston AM, Novo C,

Mulvaney P, Liz-Marzan LM, de Abajo FJG (2008) Modelling the optical response of gold
nanoparticles. Chem Soc Rev 37:1792
16. Mie G (1908) Beiträge zur Optik trüber Medien, speziell kolloider Metallösungen. Ann Phys
25:377
17. Yang WH, Schatz GC, Vanduyne RP (1995) Discrete dipole approximation for calculating
extinction and Raman intensities for small particles with arbitrary shapes. J Chem Phys
103:869
18. Jung LS, Campbell CT, Chinowsky TM, Mar MN, Yee SS (1998) Quantitative interpretation
of the response of surface plasmon resonance sensors to adsorbed films. Langmuir 14:5636
19. Mayer KM, Hafner JH (2011) Localized surface plasmon resonance sensors. Chem Rev
111:3828
20. Bingham JM, Paige HW, van Duyne RP (2012) Exploring the unique characteristics of LSPR
biosensing. In: Nanoplasmonic Sensors. Springer, New York
21. Haes AJ, Zou SL, Schatz GC, van Duyne RP (2004) A nanoscale optical biosensor: The
long range distance dependence of the localized surface plasmon resonance of noble metal
nanoparticles. J Phys Chem B 108:109
22. Miller MM, Lazarides AA (2006) Sensitivity of metal nanoparticle plasmon resonance band
position to the dielectric environment as observed in scattering. J Opt A Pure Appl Opt
8:S239
23. Jain PK, MA EI-S (2010) Plasmonic coupling in noble metal nanostructures. Chem Phys Lett
487:153
24. Soler M, Estevez MC, Alvarez M, Otte MA, Sepulveda B, Lechuga LM (2014) Direct
detection of protein biomarkers in human fluids using site-specific antibody immobilization
strategies. Sensors 14:2239
25. Chen JIL, Chen Y, Ginger DS (2010a) Plasmonic nanoparticle dimers for optical sensing of
DNA in complex media. J Am Chem Soc 132:9600
26. Trilling AK, Beekwilder J, Zuilhof H (2013) Antibody orientation on biosensor surfaces: a
minireview. Analyst 138:1619
27. Raphael MP, Christodoulides JA, Byers JM, Anderson GP, Liu JL, Turner KB, Goldman ER,
Delehanty JB (2015) Optimizing nanoplasmonic biosensor sensitivity with orientated single
domain antibodies. Plasmonics 10:1649
28. Garcia MA (2011) Surface plasmons in metallic nanoparticles: fundamentals and applica-
tions. J Phy D-Appl Phy 44:283001
29. Halas NJ, Lal S, Chang W-S, Link S, Nordlander P (2011) Plasmons in strongly coupled
metallic nanostructures. Chem Rev 111:3913
30. Urban AS, Shen XS, Wang YM, Large N, Wang H, Knight MW et al (2013) Three
dimensional plasmonic nanoclusters. Nano Lett 13:4399
31. Liu X, Zhang SY, Tan PL, Zhou J, Huang Y, Nie Z, Yao SZ (2013) A plasmonic blood glucose
monitor based on enzymatic etching of gold nanorods. Chem Commun 49:1856
32. Hermoso W, Alves TV, Oliveira CCC, Moriya EG, Ornellas FR, Camargo PHC (2013)
Triangular metal nanoprisms of Ag, Au, and Cu: modeling the influence of size, composition,
and excitation wavelength on the optical properties. Chem Phys 423:142
33. Rodriguez-Lorenzo L, Alvarez-Puebla RA, Pastoriza-Santos I, Mazzucco S, Stephan O,
Kociak M, Liz-Marzan LM, de Abajo FJG (2009) Zeptomol detection through controlled
ultrasensitive surface-enhanced Raman scattering. J Am Chem Soc 131:4616
34. Hu R, Yong K-T, Roy I, Ding H, He S, Prasad PN (2009) Metallic nanostructures as localized
plasmon resonance enhanced scattering probes for multiplex dark-field targeted imaging of
cancer cells. J Phys Chem C 113:2676
35. Diaz-Egea C, Abargues R, Martínez-Pastor JP, Sigle W, van Aken PA, Molina SI (2015) High
spatial resolution mapping of individual and collective localized surface plasmon resonance
modes of silver nanoparticle aggregates: correlation to optical measurements. Nanoscale Res
Lett 10:2
36. Oldenburg SJ, Jackson JB, Westcott SL, Halas NJ (1999) Infrared extinction properties of
gold nanoshells. Appl Phys Lett 75:2897
37. Evlyukhin AB, Reinhard C, Chichkov BN (2011) Multipole light scattering by nonspherical
nanoparticles in the discrete dipole approximation. Phys Rev B 84:235429
38. Fu YH, Zhang JB, Yu YF, Luk’yanchuk B (2012) Generating and manipulating higher order
fano resonances in dual-disk ring plasmonic nanostructures. ACS Nano 6:5130
39. Rahman MM, Im SH, Lee JJ (2016) Enhanced photoresponse in dye-sensitized solar cells via
localized surface plasmon resonance through highly stable nickel nanoparticles. Nanoscale
8(11):5884–5891
40. Shin HW, Son KR, Kim TG (2016) Localized surface plasmon-enhanced light emission using
platinum nanorings in deep ultraviolet-emitting AlGaN quantum wells. Opt Lett 41:88–91
41. Yokota H, Taniguchi T, Watanabe T, Kim D (2015) Control of localized surface plasmon
resonance energy in monolayer structures of gold and silver nanoparticles. Phys Chem Chem
Phys 17:27077
42. Tanaka D, Imazu K, Sung J, Park C, Okamoto K, Tamada K (2015) Characteristics of
localized surface plasmons excited on mixed monolayers composed of self-assembled Ag
and Au nanoparticles. Nanoscale 7:15310
43. Hong SH, Kim JJ, Kang JW, Jung YS, Kim DY, Yim SY, Park SJ (2015) Enhanced optical
output of InGaN/GaN near-ultraviolet light-emitting diodes by localized surface plasmon of
colloidal silver nanoparticles. Nanotechnology 26:385204
44. Lee S, Perez-Luna VH (2005) Dextran-gold nanoparticle hybrid material for biomolecule
immobilization and detection. Anal Chem 77:7204
45. Ozhikandathil J, Badilescu S, Packirisamy M (2011) Synthesis and characterization of silver-
PDMS nanocomposite for the biosensing applications (Proceedings of SPIE Ser. 8007).
46. Ozhikandathil J, Badilescu S, Packirisamy M (2012) Gold nanoisland structures integrated in
a lab-on-a-chip for plasmonic detection of bovine growth hormone. J Biomed Opt 17:077001
47. Choi J et al (2013) Localized surface plasmon-enhanced nanosensor platform using dual-
responsive polymer nanocomposites. Nanoscale 5:7403
48. SadAbadi H, Badilescu S, Packirisamy M, Wuthrich R (2013) Integration of gold nanopar-
ticles in PDMS microfluidics for labon-a-chip plasmonic biosensing of growth hormones.
Biosens Bioelectron 44:77
49. Avouris P (2002) Molecular electronics with carbon nanotubes. Acc Chem Res 35:1026
50. Kim JP, Lee BY, Hong S, Sim SJ (2008) Ultrasensitive carbon nanotube-based biosensors
using antibody-binding fragments. Anal Biochem 381:193
51. Mao S, Lu G, Yu K, Chen J (2010) Specific biosensing using carbon nanotubes functionalized
with gold nanoparticle–antibody conjugates. Carbon 48:479
52. Igarashi T, Kawai H, Yanagi K, Cuong NT, Okada S, Pichler T (2015) Tuning localized
transverse surface plasmon resonance in electricity-selected single-wall carbon nanotubes by
electrochemical doping. Phys Rev Lett 114:176807
53. Ozhikandathil J, Badilescu S, Packirisamy M (2015) Plasmonic Gold Decorated MWCNT
Nanocomposite for Localized Plasmon Resonance Sensing. Sci Report 5:13181
54. Penza M, Rossi R, Alvisi M, Cassano G, Serra E (2009) Functional characterization of
carbon nanotube networked films functionalized with tuned loading of Au nanoclusters for
gas sensing applications. Sensors Actuators B Chem 140:176
55. Makaryan T et al (2014) Hybrids of carbon nanotube forests and gold nanoparticles for
improved surface plasmon manipulation. ACS Appl Mater Interfaces 6:5344
56. Park JH, Byun JY, Yim SY, Kim MG (2015a) A Localized Surface Plasmon Resonance
(LSPR)-based, simple, receptor-free and regeneratable Hg2C detection system. J Hazard
Mater 307:137
57. Lee JS, Yoon NR, Kang BH, Lee SW, Gopalan SA, Kim SW, Lee SH, Kwon DH, Kang SW
(2015a) Au-polypyrrole framework nanostructures for improved localized surface plasmon
resonance volatile organic compounds gas sensing. J Nanosci Nanotechnol 15:7738
236 P. Singh
58. Nishi H, Hiroya S, Tatsuma T (2015) Potential-scanning localized surface plasmon resonance
sensor. ACS Nano 9:6214
59. Jung B, Frey W (2015) Fabrication and localized surface plasmon properties of triangular
gold nanowell arrays in a glass substrate. J Nanosci Nanotechnol 15:688
60. Sardana N, Talalaev V, Heyroth F, Schmidt G, Bohley C, Sprafke A, Schilling J (2016)
Localized surface plasmon resonance in the IR regime. Opt Express 24:254
61. Ishida R, Yamazoe S, Koyasu K, Tsukuda T (2016) Repeated appearance and disappearance
of localized surface plasmon resonance in 1.2 nm gold clusters induced by adsorption and
desorption of hydrogen atoms. Nanoscale 8:2544
62. Nezami MS, Gordon R (2015) Localized and propagating surface plasmon resonances in
aperture-based third harmonic generation. Opt Express 23:32006
63. Wabuyele MB, vo-Dinh T (2005) Detection of human immunodeficiency virus type 1 DNA
sequence using plasmonics nanoprobes. Anal Chem 77:7810
64. Vasconcelos TL, Archanjo BS, Fragneaud B, Oliveira BS, Riikonen J, Li C, Ribeiro DS,
Rabelo C, Rodrigues WN, Jorio A, Achete CA, Cançado LG (2015) Tuning localized surface
plasmon resonance in scanning near-field optical microscopy probes. ACS Nano 9:6297
65. Tokel O, Inci F, Demirci U (2014) Advances in plasmonic technologies for point of care
applications. Chem Rev 114:5728
66. Haes AJ, Chang L, Klein WL, van Duyne RP (2005) Detection of a biomarker for Alzheimer’s
disease from synthetic and clinical samples using a nanoscale optical biosensor. J Am Chem
Soc 127:2264
67. Liu Y, Huang CZ (2012) One-step conjugation chemistry of DNA with highly scattered silver
nanoparticles for sandwich detection of DNA. Analyst 137:3434
68. Inci F, Tokel O, Wang SQ, Gurkan UA, Tasoglu S, Kuritzkes DR, Demirci U (2013)
Nanoplasmonic quantitative detection of intact viruses from unprocessed whole blood. ACS
Nano 7:4733
69. Lee JH, Kim BC, Oh BK, Choi JW (2013) Highly sensitive localized surface plasmon
resonance immunosensor for label-free detection of HIV-1. Nanomed 9:1018
70. Cecchin D, de la Rica R, Bain RES, Finnis MW, Stevens MM, Battaglia G (2014) Plasmonic
ELISA for the detection of gp120 at ultralow concentrations with the naked eye. Nanoscale
6:9559
71. Yetisen AK, Montelongo Y, Vasconcellos FD, Martinez-Hurtado JL, Neupane S, Butt H,
Qasim MM, Blyth J, Burling K, Carmody JB et al (2014) Reusable, robust, and accurate
laser-generated photonic nanosensor. Nano Lett 14:3587
72. Joshi GK, Johnson MA, Sardar R (2014) Novel pH-responsive nanoplasmonic sensor:
Controlling polymer structural change to modulate localized surface plasmon resonance
response. RSC Adv 4:15807
73. Liu L, Burnyeat CA, Lepsenyi RS, Nwabuko IO, Kelly TL (2013) Mechanism of shape
evolution in Ag nanoprisms stabilized by thiol-terminated poly(ethylene glycol): an in situ
kinetic study. Chem Mater 25:4206
74. Jiang Y, Zhao H, Lin YQ, Zhu NN, Ma YR, Mao LQ (2010) Colorimetric detection of glucose
in rat brain using gold nanoparticles. Angew Chem Int Ed 49:4800
75. Chang YF, Wang SF, Huang JC, Su LC, Yao L, Li YC, Wu SC, Chen YMA, Hsieh JP, Chou
C et al (2010) Detection of swine-origin influenza A (H1N1) viruses using a localized surface
plasmon coupled fluorescence fiber-optic biosensor. Biosens Bioelectron 26:1068
76. Zheng S, Kim DK, Park TJ, Lee SJ, Lee SY (2010) Label-free optical diagnosis of hepatitis
B virus with genetically engineered fusion proteins. Talanta 82:803
77. Draz MS, Fang BA, Li LJ, Chen Z, Wang YJ, Xu YH, Yang J, Killeen K, Chen FF (2012)
Hybrid nanocluster plasmonic resonator for immunological detection of hepatitis B virus.
ACS Nano 6:7634
78. Lee J, Ahmed SR, Oh S, Kim J, Suzuki T, Parmar K, Park SS, Lee J, Park EY (2015) A
plasmon-assisted fluoro-immunoassay using gold nanoparticle-decorated carbon nanotubes
for monitoring the influenza virus. Biosens Bioelectron 64:311
79. Faridli Z, Mahani M, Torkzadeh-Mahani M, Fasihi J (2016) Development of a localized

surface plasmon resonance-based gold nanobiosensor for the determination of prolactin
hormone in human serum. Anal Biochem 495:32
80. Jin Y, Wong KH, Granville AM (2016) Enhancement of localized surface plasmon resonance
polymer based biosensor chips using well-defined glycopolymers for lectin detection. J
Colloid Interface Sci 462:19
81. Yan Z, Wang Z, Miao Z, Liu Y (2016) Dye-sensitized and localized surface plasmon reso-
nance enhanced visible-light photoelectrochemical biosensors for highly sensitive analysis of
protein kinase activity. Anal Chem 88:922
82. Zhang J, Wang Y, Wong TI, Liu X, Zhou X, Liedberg B (2015) Electrofocusing-enhanced
localized surface plasmon resonance biosensors. Nanoscale 7:17244
83. Park JH, Byun JY, Shim WB, Kim SU, Kim MG (2015) High-sensitivity detection of ATP
using a localized surface plasmon resonance (LSPR) sensor and split aptamers. Biosens
Bioelectron 73:26
84. Elghanian R, Storhoff JJ, Mucic RC, Letsinger RL, Mirkin CA (1997) Selective colorimetric
detection of polynucleotides based on the distance-dependent optical properties of gold
nanoparticles. Science 277:1078
85. Storhoff JJ, Elghanian R, Mucic RC, Mirkin CA, Letsinger RL (1998) One-pot colorimetric
differentiation of polynucleotides with single base imperfections using gold nanoparticle
probes. J Am Chem Soc 120:1959
86. Bailey RC, Nam JM, Mirkin CA, Hupp JT (2003) Real-time multicolor DNA detection with
chemoresponsive diffraction gratings and nanoparticle probes. J Am Chem Soc 125:13541
87. Jin RC, Wu GS, Li Z, Mirkin CA, Schatz GC (2003) What controls the melting properties of
DNA-linked gold nanoparticle assemblies? J Am Chem Soc 125:1643
88. Xu XY, Han MS, Mirkin CA (2007) A gold-nanoparticle-based real-time colorimetric
screening method for endonuclease activity and inhibition. Angew Chem Int Ed 46:3468
89. Zhao J, Liu T, Fan Q, Li GX (2011) A new strategy for a DNA assay based on a target-
triggered isothermal exponential degradation reaction. Chem Commun 47:5262
90. Pandya A, Sutariya PG, Menon SK (2013) A non enzymatic glucose biosensor based on an
ultrasensitive calixarene functionalized boronic acid gold nanoprobe for sensing in human
blood serum. Analyst 138:2483
91. Reinhard BM, Siu M, Agarwal H, Alivisato AP, Liphardt J (2005) Calibration of dynamic
molecular rule based on plasmon coupling between gold nanoparticles. Nano Lett 5:2246
92. Sonnichsen C, Reinhard BM, Liphardt J, Alivisatos AP (2005) A molecular ruler based on
plasmon coupling of single gold and silver nanoparticles. Nat Biotechnol 23:741
93. Chen HX, Huang JY, Lee J, Hwang S, Koh K (2010b) Surface plasmon resonance spec-
troscopic characterization of antibody orientation and activity on the calixarene monolayer.
Sensors Actuators B Chem 147:548
94. Hong C, Lee J, Zheng H, Hong SS, Lee C (2011) Porous silicon nanoparticles for cancer
photothermotherapy. Nanoscale Res Lett 6:321
95. Anker JN, Hall WP, Lyandres O, Shah NC, Zhao J, van Duyne RP (2008) Biosensing with
plasmonic nanosensors. Nat Mater 7:442
96. Kumar K, Dahlin AB, Sannomiya T, Kaufmann S, Isa L, Reimhult E (2013) Embedded
plasmonic nanomenhirs as location-specific biosensors. Nano Lett 13:6122
97. Dahlin A, Zach M, Rindzevicius T, Kall M, Sutherland DS, Hook F (2005) Localized surface
plasmon resonance sensing of lipid-membrane-mediated biorecognition events. J Am Chem
Soc 127:5043
98. Unser S, Bruzas I, He J, Sagle L (2015) Localized surface plasmon resonance biosensing:
current challenges and approaches. Sensors 15:15684
99. Lin TJ, Chung MF (2008) Using monoclonal antibody to determine lead ions with a localized
surface plasmon resonance fiber-optic biosensor. Sensors (Basel) 8:582
238 P. Singh
100. Li W, Wang S, Hu M, He S, Ge P, Wang J, Guo YY, Zhaowei L (2015) Enhancement of elec-

troluminescence from embedded Si quantum dots/ SiO2multilayers film by localizedsurface-
plasmon and surface roughening. Sci Rep 5:11881
101. Vestergaard M, Kerman K, Tamiya E (2007) An overview of label-free electrochemical
protein sensors. Sensors (Basel) 7:3442
102. Lee H, Xu L, Koh D, Nyayapathi N, Oh KW (2014) Various on-chip sensors with
microfluidics for biological applications. Sensors 14:17008
Index
A M
Angular interrogation, 41–47, 160 Mach–Zehnder interferometer (MZI), 79,
93–106
Metallic nanocrystals, 16, 21, 27
B Metallic nanoparticles, 16, 21, 217, 232
Biosensors, 47, 79–84, 87, 90, 92, 163, 212, Microfluidics, 224
216, 219, 227–228, 232 Minkowski spacetime, 137–157
C N
Core-shell nanostructures, 21, 161, 165, 170, Nanoantenna, 28, 55–74, 222
188, 190–192, 194 Nanoparticle superlattices, 110, 116, 117, 127
Nanoplasmonics, 215–216
Nanostructures, 16, 17, 21–24, 26, 27, 29,
E 30, 55–57, 59–61, 71, 110, 112, 114,
Electromagnetic radiation, 211 123, 159–195, 212, 215, 217, 219, 222,
Extraordinary transmission, 7 226, 229
Nanowire waveguide, 1–12
Near-field scanning microscopy, 8
F Negative index materials, 25–207
Ferrofluid, 137–157
Fractals, 56–59, 64, 71, 74
O
Optical fiber, 47, 160, 163, 170, 188,
G 202, 203
Guided waves, 77–106
P
H Perfect lens, 207, 208
Hazardous gases, 175–176 Photonic crystals, 106, 199–208
Hybrid waveguide, 9, 11, 12 fibers, 202–203, 208
Hyperbolic metamaterials, 138–144, 157 Photonic sensors, 84

DOI 10.1007/978-3-319-48081-7
240 Index
Plasmene nanosheets, 109–131 204–205, 212, 213, 216, 223, 224,

Plasmonic nanoparticles, 16, 66, 110–113, 228, 232
116, 120, 121, 124, 128, 131, 221, 223 sensors, 44–47, 49, 50, 160, 162, 163, 167,
Plasmonics, 1–12, 16, 17, 21–25, 28, 30, 57, 172, 176, 178, 184, 188, 205, 221–222,
110, 123, 204, 215–223, 226–231 224, 228, 230
Plasmon resonances, 16, 29, 31, 39–51, Surface plasmons, 2, 12, 16, 28, 31, 39–50, 55,
55, 68, 69, 78, 124, 125, 130, 78, 105, 124–126, 159–195, 204–205,
159–195, 204–205, 212, 213, 218, 224, 208, 211–213, 217, 220, 224, 228, 232
228, 232
T
R Thin film, 47, 63, 159–194, 232
Rare earth nanocrystals, 15–31 Two-dimensional nanomaterials, 110, 124, 128
S U
Self-assembly, 110, 115–122, 127, 138 Upconversion fluorescence, 16–18, 26–27, 31
Sensor, 39, 40, 42, 44–50, 79, 80, 83, 94–105,
160–167, 169, 171, 172, 175–180,
182–186, 188–194, 204, 205, 212, W
220–222, 224, 227–229 Water pollutants, 161
Slow light waveguides, 78, 79, 88, 92, 93, 102 Wavelength interrogation, 41–43, 46–47, 160
Subwavelength confinement, 1–12
Surface-enhanced spectroscopy, 55
Surface plasmon resonance (SPR), 16, 21, Z
23–25, 31, 39–50, 55, 78, 124, 159–195, Zinc oxide (ZnO), 159–195

Reviews in Plasmonics

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Reviews in Plasmonics

Uploaded by

Copyright:

Available Formats

Reviews in Plasmonics

Reviews in Plasmonics 2016

ISSN 1555-7731 ISSN 1555-7677 (electronic)

Library of Congress Control Number: 2017930597

© Springer International Publishing AG 2017

Printed on acid-free paper

This Springer imprint is published by Springer Nature

This is the third volume in the Plasmonics series,

1 Plasmonic Nanowire Waveguide for Deep Subwavelength

9 Frontiers of Light Dynamics in Photonic Crystals . . . . . . . . . . . . . . . . . . . . . . 199

Keywords Extraordinary transmission • Hybrid waveguide • Nanowire

© Springer International Publishing AG 2017 1

high-efficiency solar cells [13, 14]. Many approaches to achieving propagation of

1.2 Cylindrical Waveguide

A dielectric cylinder of radius a surrounded by a cladding metal of finite conduc-

Fig. 1.1 Schematic of a

and Ez of the TM mode is given by:

where A, B, C and D are unknown coefficients to be determined, m is the azimuth

indicating a cutoff radius-to-wavelength ratio of 0.13. Therefore it may be possible

1.3 Plasmonic Nanowire Waveguide

As shown previously [36], when the following condition is satisfied:

"1 C "2  0; (1.5)

1.4 Hybrid Plasmonic Waveguide

The plasmonic nanowire waveguide can achieve deep subwavelength confinement,

The confinement of these modes can be characterized by examining the full

© Springer International Publishing AG 2017 15

Keywords Rare earth nanocrystals • Metallic nanocrystals • Core-shell

The collective oscillations of conduction electrons in metallic nanostructures,

2.2 Dopant-Controlled Synthesis of Upconversion Rare

Upconversion emission materials of RE NCs are proposed as a promising probe

2.3 Synthesis of Rare Earth Nanocrystal Shells on Plasmonic

Fluorescence enhancement and manipulation of RE NCs by plasmonic interaction

2.3.1 Synthesis of Metal/Oxide Core/Shell Nanocrystals

2.3.2 Synthesis of Au NR/Vanadate Core/Shell Nanocrystals

AuNR is a typical anisotropic metal NC with tunable longitudinal SPR in a wide

emission spectra, rare-earth vanadate has been widely-used as fluorescent [101–

2.3.3 Synthesis of Au Nanoparticle/Vanadate Core/Shell

Because of large lattice mismatch and interfacial energy, it is difficult to form

2.4 Plasmon-Enhanced Upconversion Fluorescence

2.5 Fluorescence Tunability of Rare Earth Upconversion

Plasmonic array is also an effective method to achieve upconversion enhancement.

RE nanocrystals possess characteristic fluorescence properties and are expected to

1. Li M, Cushing SK, Wu N (2015) Plasmon-enhanced optical sensors: a review. Analyst

10. Bauch M, Toma K, Toma M, Zhang Q, Dostalek J (2014) Plasmon-enhanced fluorescence

32. Li Z, Chen S, Li J, Liu Q, Sun Z, Wang Z, Huang S (2012) Plasmon-enhanced upconversion

91. Li B, Gu T, Ming T, Wang J, Wang P, Wang J, Yu JC (2014) (Gold core)@(ceria shell)

Keywords Surface Plasmon Resonance • Angular interrogation • Wavelength

In the modern technologically advanced society, a large number of high performance

R.K. Gupta ()

© Springer International Publishing AG 2017 39

instrumentation, the field of sensor is growing exponentially. The sensor based

3.1 Principle of Surface Plasmon Resonance (SPR)

Here, kx can be changed by changing either angle of incidence ( i ) or the

3.2 Techniques Based on SPR for Sensing Applications

3.2.1 Angular Interrogation Based Sensor

In Otto geometry, the high RI waveguide is interfaced with dielectric-metal

3.2.2 Wavelength Interrogation Based Sensor

as compared to that of angular interrogation SPR instrument. In a brilliant attempt,

3.2.3 SPR Imaging Based Sensor

Selectivity and multi-analyte detection is difficult with the conventional SPR

3.3 Sensing Applications of SPR

Table 3.1 Sensing applications using surface plasmon resonance phenomenon

Table 3.1 (continued)

Ravi Sadananda Hegde

"1 C "2 0; (1.5)

where 1 C 2 .N 1/ denote the dipolar modes associated with the largest

At the resonance frequency ! 0 , the group delay is g D2Q/ ¨0 . Substituting Eq.