![TABLE II SENSITIVITY DATA OF THE D-SHAPED PCF-SPR SENSOR (ANALYTE RANGES FROM 1.33 TO 1.45 [Tbinding IS 500 NM])](/figures/table-ii-sensitivity-data-of-the-d-shaped-pcf-spr-sensor-21rozxg3.webp)
![TABLE I SENSITIVITY DATA OF THE D-SHAPED PCF-SPR SENSOR (ANALYTE RANGES FROM 1.33 TO 1.41[Ta IS 1500 NM])](/figures/table-i-sensitivity-data-of-the-d-shaped-pcf-spr-sensor-2lno6lz8.webp)
1
Influence of the Sub-peak of Secondary Surface
Plasmon Resonance onto the Sensing Performance
of a D-shaped Photonic Crystal Fibre Sensor
Suoda Chu, K. Nakkeeran, Senior Member, IEEE, Abdosllam M. Abobaker, Member, IEEE, Sumeet S. Aphale,
Senior Member, IEEE, S. Sivabalan, P. Ramesh Babu and K. Senthilnathan
Abstract—In this paper, we design a 6-fold D-shaped photonic
crystal fibre sensor based on the surface plasmon resonance
(SPR). The coupling between fundamental core mode and three
surface plasmonic modes which have different electric filed
distributions for analytes of various refractive indices. We observe
two different types of SPRs, namely, ‘dielectric like’ resonance
with low-loss peak and ‘plasmon like’ resonance with high-loss
peak, by analyzing the electric field distribution of the fibre
modes. Further, we discuss the influence of the secondary SPR
over the main SPR which is directly related to the detection
performance of the proposed sensor. In order to mitigate the
adverse effect of the sub-peak of the secondary SPR on the
sensor’s dynamic sensing range (DSR), we reduce the thickness of
analyte’s binding layer from 1500 nm to 500 nm. Thus, DSR can
be extended to 44.4% from 1.33-1.41 to 1.33-1.45 at the cost
of a reduced maximum sensitivity from 7900 nm/RIU to 5300
nm/RIU. Owning to the simple structure design of the proposed
sensor, we envisage that this highly sensitive D-shaped PCF-
SPR sensor could be developed as a versatile and competitive
instrument with a large and flexible refractive index detection
range.
Index Terms—Sensor, Photonic crystal Fibre, Refractive Index
Sensor, Sensitivity, Surface Plasmon Resonance
I. INTRODUCTION
T
HE sensing applications based on the optical excitation
and detection of the surface plasmon resonance (SPR)
phenomenon have been widely studied. It is proven to be a
commercially successful technology in the field of medical
diagnostics, chemical detection, bio-chemical reaction recog-
nition, food safety control, environment monitoring, etc. The
high sensitivity to the change of refractive index of the medium
in contact with the surface of thin metal film (typically gold
or silver with dielectric), has been widely utilized for those
sensing applications [1], [2].
The SPR phenomenon is generally defined as the strong
coupling between the electricmagnetic wave and surface plas-
mon wave at the interface of dielectric and metal [3]. The
Suoda Chu, K. Nakkeeran and Sumeet S. Aphale are with School of
Engineering, Fraser Noble Building, University of Aberdeen, Aberdeen
AB24 3UE, UK (email: r05sc15@abdn.ac.uk; K.Nakkeeran@abdn.ac.uk;
S.Aphale@abdn.ac.uk).
Abdosllam M. Abobaker is with Department of Communications En-
gineering, College of Electronic Technology, Bani Walid, Libya (email:
almahjub11@gmail.com).
S. Sivabalan is with School of Electrical Engineering, VIT University,
Vellore-632 014, Tamil Nadu, India (email: ssivabalan@vit.ac.in).
P. Ramesh Babu and K. Senthilnathan are with Department of Physics,
School of Advanced Sciences, VIT University, Vellore-632 014, Tamil Nadu,
India (email: prameshbabu@vit.ac.in; senthee@gmail.com)
development of SPR sensor started from the traditional struc-
ture of Kretschmann-Raether prism configuration in 1960.
However, this configuration has its own drawbacks such as
bulk size, complicated structure design, high cost and low
reliability [4]. Eventually, optical fibre evolved as the dielectric
medium to overcome the said drawbacks. Due to its design
flexibility and compactness [3], the metal coated optical fibres
that emerged as an alternative to the prism have been used
[5]. As a result, low-cost, highly integrable, and portable
optical fibre based SPR sensors have been developed. The most
common schemes of SPR fibre sensors include a modified fibre
end [6], tapered fibre [7], D-shaped fibre [8], and fiber grating
[9].
In recent times, photonic crystal fibre (PCF), also called
holey fibre (HF) or micro-structured fibre (MF) has been
widely used as a novel class of optical fibre for different
sensing applications. Apart from the inherent advantages as a
kind of optical fibres, PCFs exhibit their own pros, especially
the ability to control its optical characteristics by manipulating
the structural parameters of the optical fibre [10]. The PCF-
SPR based sensors are now widely studied because of their
simple and compact probe designed for high sensitivity, ro-
bustness, cost effectiveness, fast response, label-free detection
[11]. Moreover, it is possible to enhance the sensitivity and the
sensing range by optimizing the structural parameters, namely,
air holes diameters and the distance between two adjacent air
holes.
In the last decade, many metal coating schemes were
reported for the PCF-SPR sensor that include completely air
hole coating method by Shuai et al in 2012 [12], selectively
coated method with the air holes by Yu et al in 2009 [13],
the outside metal coating method by Hasan and his research
group in the year of 2018 [14] and D-shaped fibre side surface
coating by Tian and his research group in 2012 [15]. However,
the completely and selectively air holes coating methods are
not practically viable besides the challenges of coating the
thin metal film homogeneously and uniformly in the inner
micrometer-scaled air holes with the existing fabrication tech-
nologies. Further, the liquid analyte has to be injected into the
air holes by employing those two metal coating methods that
increase their complicity and production cost. Although the
outside metal coating scheme does not suffer from the above
drawbacks, owing its working principle, the sensors based
on such methods usually have a very weak surface plasmon
resonance which is not observable for sensing. Among these
2
metal coating schemes, the D-shaped fibre side surface coating
is proven scheme [16]. The hexagonal D-shaped PCF is one
of the most frequently used fibre media for common PCF-SPR
sensor applications. As the side-polished flat surface is coated
with thin metal film that is in direct contact with the liquid
analyte, the D-shaped PCF is proposed in this paper for the
feasibility of sensor fabrication.
We numerically investigate the sensing performance of
a standard fabricable 6-fold hexagonal D-shaped PCF-SPR
sensor on a large analyte refractive index range from 1.33
to 1.48. Further, we find that this sensor has a large DSR
from 1.33 to 1.41 with a high sensitivity and a linear sensing
performance with a liquid layer thickness of 1500 nm. It is
worth noting that for lower refractive index analyte (1.33 ≤
RI ≤ 1.41), only one ‘dielectric-like’ SPR with a low-loss
peak is observed at 0.6 µm wavelength. Hence, one single
loss peak is found in the loss spectrum for such detection
range. However, for higher refractive index analyte (1.42 ≤
RI ≤ 1.48), two different ‘plasmon-like’ SPRs with high-loss
peaks are observed with two peaks in the same wavelength of
0.6 µm. Due to the overlap of loss peak and the existence of
the sub-peak, as a single-peak detection device, the sensing
performance is literally affected and this drawback makes
this RI range undetectable. Moreover, the two resonance
coupling mechanisms of ‘dielectric like’ SPR and ‘plasmon
like’ SPR for low and high RI analyte ranges are different. The
wavelength interrogation method is applied for SPR optical
fibre sensing to evaluate the sensitivity of the proposed sensor
due to the availability of affordable small optical spectrum
analyzers [17]. The simulation results indicate that the highest
sensitivity for an RI range of 1.33-1.41 is up to 7900 nm/RIU.
By reducing the liquid layer thickness to 500 nm, the DSR can
be extended to 1.33-1.45. But on the downside, the maximum
sensitivity reduces to 5300 nm/RIU.
II. GEOMETRIC STRUCTURE AND NUMERICAL
MODELLING
The schematic of the proposed sensor is shown in Fig. 1.
It comprises of 4 layers of air holes arranged in a six-fold
hexagonal PCF structure of air hole diameter, d = 1.2 µm in
a solid core. The distance between two holes (pitch), Λ, is 2.5
µm and the radius of the whole sensor is set as 11 µm. An
open D-shaped analyte channel is designed at the top part of
the fibre cross-section so that the analyte can be infused in
the channel. The height of the D-shaped channel, d
a
, is 8 µm.
An uniform nano-scale gold metal film is deposited on the flat
side-polished surface with its layer thickness of t
Au
= 45 nm
for surface plasmon polaritons generation. The thickness of
liquid analyte layer in the D-shaped channel is kept at 1500
nm.
In this structure, fused silica is chosen as the cladding
material and the wavelength dependence of the refractive index
of the silica glass is calculated by the Sellmeier equation [18].
The dielectric constant of the thin gold film is calculated using
the Drude model that is characterized in [19].
The proposed structure can be fabricated using the state-
of-the-art technique of stack-and-draw [20] and side polishing
Fig. 1. Cross-section of the proposed six-fold D-shaped hexagonal PCF-SPR
sensor.
methods [21]. The fabrication process of proposed sensor is
also based on D-shaped side-channel PCF fabrication which
has been successfully reported by Chen et al. They used two
steps stack-and-draw process and the channel was created by
removing the top of the holy lattice cladding as illustrated in
Fig. 2 [16]. During the fibre drawing process, the pressure,
vacuum and the stability of drawing tower need to be well
controlled to maintain the structure of the PCF [22]. In order
to achieve a D-shaped channel, the side polishing method can
be used to remove the top part. The rugged surface of the
D-shaped upper side can be processed by a focused high-
power laser beam through the cavity to achieve flat surface
of the channel. Finally, a thin gold layer is coated on the
sensing surface with a chemical deposition method, described
by Jonathan Boehm [23].
Fig. 2. Stacked preform of the proposed PCF and side polishing method for
the top part removal.
To numerically investigate the sensing performance of
this sensor, FEM is used to find the effective refractive
indices through COMSOL Multiphysics software. A perfectly
matched layer (PML) is considered as the boundary condition
for the outside edges to absorb the radiated light energy for
the simulation to mimic the real situation [24].
III. ANALYSIS OF MODES
For any SPR based sensor, it is well known that the surface
plasmon mode (PM) is generated and coupled with the main
core-guided fundamental mode (FM) at a particular resonant
wavelength [25]. In essence, there occurs a transfer of energy
from FM to PM. Thus, at the resonance wavelength, the loss
spectrum exhibits a sharp resonance-peak characteristic [17].
Simulations results of the designed PCF-SPR sensor for the
formation of the PM are shown in Figs. 3. The refractive index
of the liquid analyte, n
a
, is 1.38 in the lower RI range (1.33
≤ RI ≤ 1.41). Figures 3 illustrate the distribution of the light
of the FM as well as the PM of the proposed sensor at various
resonance wavelengths of 600 nm, 690 nm and and 770 nm.
3
Fig. 3. Distribution of light in the cross-section of D-shaped PCF-SPR sensor
for different wavelengths for the analyte RI of n
a
= 1.38. (a) and (d) are the
FM and PM at 600 nm (shorter wavelength with respect to the resonance).
(b) and (e) are the FM and PM at resonant wavelength of 690 nm. (c) and
(f) are the FM and PM at 770 nm (longer wavelength with respect to the
resonance).The arrows indicate the direction of the electric field.
Here, the arrows indicate the direction of the electric field.
Figures 3(d), 3(e) and 3(f) show the formation of the PM
for all the three wavelengths. In Fig. 3(b), a weak surface
plasmon resonance can be observed at the inference between
the thin gold film and the dielectric. Figs. 3(a) to 3(c) clearly
demonstrate the complete mechanism of SPR wherein a part
of light energy of FM leaks into the gold film sensing area as
the light signal of PM.
Figure 4 shows the confinement loss spectrum (solid curve),
dispersion relations of the FM (dashed curve) and PM (dot-
dashed curve) for the D-shaped hexagonal PCF-SPR sensor
for an analyte of RI of 1.38. Here, the confinement loss is
calculated in dB/cm using the relation [26],
α
loss
= 8.686 ×
2π
λ
=(n
eff
) × 10
4
, (1)
where =(n
eff
) is the imaginary part of the effective refractive
index of the FM. As it is illustrated in Fig. 4, a sharp and deep
single resonance peak is observed at the resonance wavelength
of 690 nm.
0 . 5 0 0 . 5 5 0 . 6 0 0 . 6 5 0 . 7 0 0 . 7 5 0 . 8 0 0 . 8 5
1 . 4 0
1 . 4 2
1 . 4 4
1 . 4 6
1 . 4 8
1 . 5 0
R e f r a c t i v e I n d e x ( r e a l p a r t )
W a v e le n g th [μ m ]
C o r e m o d e
P l a s m o n i c m o d e
L o s s
0
2
4
6
8
1 0
1 2
1 4
1 6
1 8
2 0
L o s s [ d B / c m ]
Fig. 4. The confinement loss and the dispersion relations of FM and PM for
the analyte RI of n
a
= 1.38.
However, as shown in Fig. 5, two resonance peaks are
observed in the confinement loss spectrum at resonance wave-
lengths of 1337.5 nm and 1515 nm for an higher refractive
index analyte of 1.46. The surface plasmon modes in these
two SPRs exhibit different patterns as depicted in Figs. 6
and Figs. 7 when compared to what has been illustrated in
Figs. 3. The electric field distributions of FM and PM are
shown in Figs. 6 and Figs. 7 for the first and second resonance
wavelengths, respectively, that include both shorter and longer
wavelengths.
1 . 0 1 . 1 1 . 2 1 . 3 1 . 4 1 . 5 1 . 6 1 . 7 1 . 8
1 . 4 2
1 . 4 4
R e f r a c t i v e I n d e x ( r e a l p a r t )
W a v e le n g th [μ m ]
C o r e M o d e
H i g h O r d e r P l a s m o n i c M o d e
P l a s m o n i c m o d e
L o s s
2 n d P e a k
1 s t P e a k
0
2 0 0
4 0 0
6 0 0
8 0 0
1 0 0 0
1 2 0 0
L o s s [ d B / c m ]
Fig. 5. The confinement loss and the dispersion relations of FM and PM for
the analyte RI of n
a
= 1.46.
Fig. 6. Distribution of light in the cross-section of D-shaped PCF-SPR sensor
for different wavelengths for the analyte RI of n
a
= 1.46. (a) and (d) are the
FM and PM at 1100 nm (shorter wavelength with respect to the resonance).
(b) and (e) are the FM and PM at resonant wavelength of 1337.5 nm. (c) and
(f) are the FM and PM at 1400 nm (longer wavelength with respect to the
resonance). The arrows indicate the direction of the electric field.
As is evident in Figs. 3(b), Figs. 6(b) and Figs. 7(b),
the electric field distribution of SPR is differen for different
resonance wavelengths. We note that the electric field of
‘dielectric-like’ resonance is mainly confined in the solid core
area in the case of analyte of lower RI. In this case, the
photons from the fundamental mode propagate through the
thin gold film and get coupled with the coherent delocalized
electrons of the gold film in the form of surface plasmons that
are generated at the upper boundary between thin gold film
and the analyte layer. Here, the strength of sensing signal is
low but detectable. At its resonance wavelength, the electric
field pattern of surface plasmon does not change apprecialbly
as illustrated in Figs. 3(e) when compared to Figs. 3(d) and
(f).
On the other hand, for higher RI analyte, the electric fields
of two ‘plasmon-like’ resonances are mainly concentrated at
the gold film surface with extremely high loss, as shown in
Fig. 7. Distribution of light in the cross-section of D-shaped PCF-SPR sensor
for different wavelengths for the analyte RI of n
a
= 1.46. (a) and (d) are the
FM and PM at 1400 nm (shorter wavelength with respect to the resonance).
(b) and (e) are the FM and PM at resonant wavelength of 1515 nm. (c) and
(f) are the FM and PM at 1800 nm (longer wavelength with respect to the
resonance). The arrows indicate the direction of the electric field.
4
Figs. 6(b) and Figs. 7(b). On comparing the electric field
patterns of the surface plasmons displayed in Figs. 6(c),(e),(f)
and Figs. 7(c),(e),(f), we find that the surface plasmon can not
only absorb a small portion of leaked light at resonance, but
also radiates part of it to core to form a quasi-core distribution
at resonance wavelength. It is to be noted that this absorption
and radiation in both shorter and longer wavelengths are not
as strong as that of resonance wavelength. Moreover, the
strongest coupling/absorption happens at the lower boundary
of gold film with a weak surface plasmon generated on the
upper gold film surface which has the same phase [Figs. 6(d)]
or phase shift [Figs. 7(d)] with the lower boundary surface
plasmon.
Fig. 8. Dispersion relations of air (dashed line), fiber medium (blue) and upper
and lower surfaces of thin gold film (red). The circles indicate occurrence of
SPR in three different circumstances.
Three surface plasmon resonances with different electric
field distributions, as illustrated in Figs. 8, can be found by
the coulpings that happen between different surface plasmon
polaritons (SPPs) and the photons. Among them, the SPP
1 describes the surface plasmon mode of Figs. 3(d), SPP
2 corresponds to the surface plasmon mode of Figs. 6(d)
and SPP 3 is for the surface plasmon mode of Figs. 7(d).
This phenomenon can be explained by the high reflectance of
gold film at longer wavelength which is related to metallic
reflectance and the increased RI of the analyte. A standard
reflectance spectrum of gold shown in Fig. 9 is related to the
RI by the following equation [27]:
R =
(n − 1)
2
+ k
2
(n + 1)
2
+ k
2
, (2)
where R, n and k represent the reflectance, real part and
imaginary parts of refractive index, respectively. It is obvious
that, at higher wavelengths, the gold film will exhibit high
reflectance characteristic and vice versa.
For lower RI analytes, as their resonance wavelengths are
normally distributed in the shorter wavelengths region, the
gold film would have a lower reflectivity. Hence, most of the
leaked light energy gets transmitted through the gold film and
excites the SPR together with the delocalized electrons at the
upper surface of the gold film. However, the SPPs at the lower
surface of gold are too weak to observe. Moreover, as the RI of
analyte is lower than that of fibre silica, light gets confined well
in a fibre medium [28] and hence the confinement loss of this
sensor is very low. Therefore, a weak SPR takes place at the
upper surface of gold film. However, for higher RI analytes, a
larger portion of light leaks from FM to PM and it results in
0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0 1 . 2 1 . 4 1 . 6 1 . 8 2 . 0
0 . 0
0 . 2
0 . 4
0 . 6
0 . 8
1 . 0
l o n g e r w a v e l e n g t h ,
h i g h e r r e f l e c t a n c e
G o l d
W a v e le n g th ( μ m )
R e f l e c t a n c e ( % )
s h o r t e r w a v e l e n g t h ,
l o w e r r e f l e c t a n c e
Fig. 9. Standard reflectance spectrum for gold
high confinement loss as seen in Fig. 5. Besides, the leaked
light energy cannot easily transmit through the gold film as
the reflectivity of the metal is higher at longer wavelength.
Unlike the case of SPPs at the lower wavelength, most of the
leaked photons get coupled with the delocalized electrons at
the lower surface of gold film and only a small portion of
them can get transmitted through the gold film. As a result,
the gold film has stronger surface plasmon at lower surface
and weaker surface plasmon at the upper surface.
At the resonance wavelength, due to maximum confinement
loss, the energy absorption of surface plasmon turns out to be a
maximum. Thus, the leaked photons can not fully couple with
electrons from metal layer. The rest uncoupled photons from
leaked light energy will be continually absorbed by the SPP.
However, the excess SPP energy is transformed into radiation
[29]. The radiation energy from SPP is reflected by the gold
film back into the fibre core. However, the radiated photons
that have the same dispersion relations with surface plasmon
mode will gather and form a quasi-core mode in the core area.
The surface plasmon modes shown in Figs. 6(e) and Figs. 7(e)
turn out to be hybrid modes of a surface-plasmon mode for
absorption at gold film and a quasi-core mode for radiation at
core area.
Moreover, in Fig. 7(d), the surface plasmons are formed by
the delocalized electrons at the lower boundary of gold film
and hence weak surface plasmons are formed with a phase
shift of π on the upper surface. Although a strong coupling
between fundamental mode and hybrid surface plasmon occurs
at the lower boundary of gold film, the weak surface plasmon
on upper side of gold film still gets involved in the resonance.
This results in ‘plasmon-like’ SPR which is also sensitive to
the RI changes of the analyte. However, due to the increasing
reflectance of gold in longer wavelength, the sensing ability
based on ‘plasmon-like’ SPR will decrease for higher RI
analyte.
Thus, for a large range of RI analytes, there are two different
types of SPRs, namely, the ‘dielectric-like’ and the ‘plasmon-
like’, respectively due to different resonance wavelengths.
Further, they are also known as ‘incomplete coupling’ and
‘complete coupling’ [12] [30] [31] [32]. We note that the phase
matching conditions for them are found to be different. For
lower RI of analyte, as its resonance wavelength is usually
located at shorter wavelength region, the ‘dielectric-like’ SPR
occurs at upper boundary of gold film surface due to the the
5
phase matching condition which demands that the real parts
of effective refractive indices (n
eff
) of FM and PM be equal.
However, for the analytes of higher RIs, as the resonance
wavelengths are usually located at longer wavelength region,
the ‘plasmon-like’ SPR occurs at lower surface of gold film.
As discussed above, the hybrid surface plasmon mode of
‘plasmon-like’ SPR actually combination of absorption and
radiation. In 2008, Zhang et al have noticed that there would
be two different coupling phenomenona between modes and
successfully proved them [33]. Based on their coupled-mode
theory, in any SPR, the parameters κ and δ
i
represent mode-
coupling strength, which is related to the real part of n
eff
, and
mode-absorption strength, which is related to the imaginary
part of n
eff
, respectively. For ‘dielectric-like’ SPR, as there is
only mode coupling between fundamental mode and upper
surface mode, the κ is always greater than δ
i
. Hence, the
incomplete coupling takes place under the condition with
equal real parts of two modes. For ‘plasmon-like’ SPR, the
mode-coupling strength κ is defined between fundamental
core mode and hybrid surface plasmon mode. However, the
mode-absorption strength δ
i
is related to the surface plasmon
mode absorption and the radiant energy. In this situation, δ
i
is
greater than κ. Therefore, a complete coupling occurs when
the imaginary parts of n
eff
of fundamental core modes are
equal to the n
eff
of hybrid surface plasmons. To the best of
our knowledge, this is the first time to reveal and explain these
phenomena by analysing the electric distribution of different
fibre modes. Also, this is the first time to observe the existence
of surface plasmon resonance at different surfaces of gold film.
IV. SENSING PERFORMANCE
In this section, we analyze the performance of the proposed
sensor. For a better performance of a refractive index based
SPR fibre sensor, the energy transferred to the PM needs
to be extremely sensitive to the RI changes of the aqueous
analyte [34]. When there are small RI changes in the analyte
due to chemical/biochemical interactions, the real part of the
n
eff
of the PM undergoes a shift its position with respect
to wavelength. As a consequence, the resonant wavelength
occurring at the phase matching condition will also have a
significant shift.
One can observe single resonance peaks from the loss
spectra of Fig. 10(a) for different analytes of RI ranging from
1.33 to 1.41 within the wavelength range from 0.5 to 1.0
µm. Here, the confinement loss increases with the increase
in RI of the analyte. It should be noted that the loss peaks
for RI=1.42 and 1.43 become more and more blunt and
broad due to increase in confinement loss and the existence
of a secondary SPR caused by FM and hybrid PM in the
longer wavelength. As a result, the resonance peak can hardly
be obtained and observed by a spectrometer. However, dual
resonance peaks are observed in Fig. 10(b) for a wavelength
range from 0.9 to 1.8 µm for the analytes of RI varying from
1.42 to 1.48. Due to the existence of SPR between fundamental
core mode and hybrid surface plasmon, a sub-peak is observed
in the confinement loss spectra. As the secondary SPR always
occurs at a longer wavelength, it is a ‘plasmon-like’ SPR with
very high loss. Although the sub-peak is still sensitive to the
RI changes of the analyte, the overlap of resonance peaks
causes an adverse effect on sensing performance. Thus, in the
detection process, dual-peak characteristics will cause errors
in the measurement within the certain sensing bandwidth.
It is clear that the uniqueness and accuracy of single peak
measurement cannot be guaranteed by the proposed sensor
for analytes whose RI is greater than 1.41 due to the overlap
of two resonance peaks. This is a limitation on detection range
and sensing performance of the proposed sensor.
0 . 5 0 . 6 0 . 7 0 . 8 0 . 9 1 . 0 1 . 1
0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
9 0
n
a
= 1 . 3 3
n
a
= 1 . 3 4
n
a
= 1 . 3 5
n
a
= 1 . 3 6
n
a
= 1 . 3 7
n
a
= 1 . 3 8
n
a
= 1 . 3 9
n
a
= 1 . 4 0
n
a
= 1 . 4 1
n
a
= 1 . 4 2
n
a
= 1 . 4 3
W a v e le n g th ( μ m )
L o s s [ d B / c m ]
N o t s h a r p
0 . 9 1 . 0 1 . 1 1 . 2 1 . 3 1 . 4 1 . 5 1 . 6 1 . 7 1 . 8
0
2 0 0
4 0 0
6 0 0
8 0 0
1 0 0 0
1 2 0 0
2 n d
P e a k s
n
a
= 1 . 4 2 n
a
= 1 . 4 3 n
a
= 1 . 4 4 n
a
= 1 . 4 5
n
a
= 1 . 4 6 n
a
= 1 . 4 7 n
a
= 1 . 4 8
W a v e le n g t h (μ m )
L o s s [ d B / c m ]
1 s t
P e a k s
Fig. 10. The loss spectra for the analyte refractive index (n
a
) varying from
(a) 1.33 to 1.43 in steps of 0.01 and (b) 1.42 to 1.48 in a step of 0.01.
By introducing the wavelength interrogation method, the
sensitivity, S
λ
(λ), of the sensor can be calculated using the
expression [2],
S
λ
(λ) =
∆λ
peak
∆n
a
, (3)
where ∆λ
peak
is the resonant wavelength shift and ∆n
a
is the analyte’s refractive index difference. For example, in
Fig. 10(a), the wavelength shift between the confinement loss
peaks for RI 1.40 and 1.41 is 79 nm. Hence, the sensitivity
of the PCF-SPR sensor for the analyte RI changing from 1.40
to 1.41 is calculated to be 7900 nm/RIU.
Table I reports the resonant wavelengths and the corre-
sponding sensitivities of the PCF-SPR sensor for various de-
tectable analytes’. The sensor exhibits a maximum sensitivity
of 7900 nm/RIU under ‘dielectric-like’ SPR with a low loss
for analyte’s RI ranging from 1.33 to 1.41 and the maximum
resolution is as high as 1.26 × 10
−5
RIU for a 0.1 nm
peak-wavelength resolution of the instrument for precise and
accurate detection.