Sensors and Actuators A 236 (2015) 38–43

Contents lists available at ScienceDirect

Sensors and Actuators A: Physical

journal homepage: www.elsevier.com/locate/sna

Experimental realization and performance evaluation of refractive

index SPR sensor based on unmasked short tapered multimode-fiber
operating in aqueous environments
Yusser Al-Qazwini a,∗ , A.S.M. Noor a,b,∗∗ , Mohd H. Yaacob a,b , S.W. Harun c , M.A. Mahdi a,b
a
Wireless and Photonics Networks Research Centre of Excellence (WiPNet), Faculty of Engineering, Universiti Putra Malaysia, 43400 UPM Serdang,
Selangor, Malaysia
b
Department of Computer and Communication Systems Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor,
Malaysia
c
Department of Electrical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: In this work, we fabricate and characterize an SPR sensor based on unmasked short tapered multimode
Received 10 August 2015 fiber. The entire taper region is coated with gold without masking offering a simpler approach towards
Received in revised form 12 October 2015 the sensor fabrication. The sensor was tested by immersion in sensing solutions of different refractive
Accepted 14 October 2015
indices and the transmittance is measured using a spectrometer system. The sensor performance in
Available online 20 October 2015
terms of sensitivity, detection accuracy, and robustness can be controlled by adjusting the taper physical
parameters such as waist diameter, waist length, and transition length. We realize a compact, sensitive,
Keywords:
cost-effective and robust SPR sensor suitable for aqueous media sensing by reducing the waist diameter
Refractive index sensor
Surface plasmon resonance
up to 25–45 ␮m with a total taper length of 3–5 mm coated with a thin gold film of around 55 nm.
Tapered multimode fiber The demonstrated sensor exhibits a sensitivity of 1600–2000 nm/RIU and a full width at half maximum
Sensitivity (FWHM) of 140–220 nm.
FWHM © 2015 Elsevier B.V. All rights reserved.

1. Introduction of SPR to occur in optical fiber, an evanescent field is generally

needed to generate surface plasmon waves (SPW) at the interface
In the past two decades, optical sensors based on surface plas- of a metal layer and the surrounding medium. For that purpose,
mon resonance (SPR) technique have been widely applied in a wide several structures of fiber optic SPR sensors have been proposed
range of sensing applications. Refractive index sensors based on SPR and developed. Some examples of these structures include partial
are capable of detecting small changes in the refractive index (RI) removal of fiber cladding [5–6], side-polishing [7], and tapering
of the surrounding media. The theory and principles behind this [4,8]. Photonic crystal fiber [9], D-shaped optical fiber [10], and fiber
phenomenon have been described elsewhere [1,2]. grating SPR sensors [11] have also been reported to improve the
The introduction of SPR-based optical fiber as an optical sensor reliability of fiber optic SPR sensors. More recently, SPW confined
in the early 90s has received a lot of attention because it provides to a nanostructure (also called localized surface plasmons) has been
a high level of miniaturization of SPR devices as well as sensing receiving increasing attention but their practical implementation
capability in inaccessible locations. Since then, optical fibers have still faces considerable challenges [12,13].
been utilized as SPR sensors in numerous applications and tremen- A typical SPR fiber optic sensor is typically made from silica
dous effort has been carried out to improve the performance of fiber, either single-mode (SMF) or multimode (MMF) fiber. Some
the fiber optic SPR sensors [3–5]. In order to increase the chances researchers reported the use of SMF instead of MMF to enhance
the detection accuracy of the SPR sensor [7]. However, due to the
small core of SMF, it can only accommodate very small fraction
of power, so the sensor design requires very high precision. Fur-
∗ Corresponding author.
∗∗ Corresponding author at: Research Centre of Excellence for Wireless and Pho-
thermore, the polarization of light has to be precisely controlled
tonics Network (WiPNet), Faculty of Engineering, Universiti Putra Malaysia, 43400
to minimize fluctuations in the sensor output which requires the
Serdang, Selangor, Malaysia use of extra components and thus increases the system complexity
E-mail addresses: [email protected] (Y. Al-Qazwini), [email protected] and cost [14]. Another drawback of SMF-based optical sensors is
(A.S.M. Noor).

http://dx.doi.org/10.1016/j.sna.2015.10.030
0924-4247/© 2015 Elsevier B.V. All rights reserved.
 Y. Al-Qazwini et al. / Sensors and Actuators A 236 (2015) 38–43 39

Fig. 1. (a) Microscopic image of the tapered optical fiber illustrating the waist and transition regions. (b) Photo of the actual fabricated sensor.

that they are very fragile which makes the sensor fabrication more The experimental setup used to measure the SPR transmission
difficult. As mentioned earlier, tapered optical fibers are among the spectra is depicted in Fig. 2. It consists of a white light source (Ocean
structures of fiber optic SPR sensors that have been proposed. The Optics HL-2000), fiber optic connectors, TOF sensor based on SPR
evanescent field generated after tapering can be used to excite the mounted in a u-groove holder, and a spectrometer. The spectrom-
SPW in metal films around the tapered optical fibers [15]. Generally, eter was connected to a computer installed with the SpectraSuite
tapered-based sensors are more sensitive, compact and simpler software to calculate and plot the transmitted SPR spectra in a
to make compared to those side-polished in which the cladding wavelength range from 350 to 1100 nm. The transmitted SPR spec-
is removed [16,17]. Even though many studies have been done trum is the sensor response in terms of its normalized transmission
on tapered optical fibers, they did not provide sufficient infor- with respect to wavelength. The normalized transmission T , as a
mation on the impact of tapering parameters. In addition, most percentage relative to a standard substance (e.g., air) at wavelength
studies applied some masking methods to coat certain lengths of  is estimated by SpectraSuite as
the tapered region which make the sensor fabrication more sophis-
ticated [18,19]. (S␭ − D␭ )
%T ␭ = × 100% (1)
In this paper, we report the fabrication and characterization of (R␭ − D␭ )
an RI sensor based on SPR in tapered MMF. Our fabricated sensor
exhibits high sensitivity and detection accuracy, robustness, and where S is the sample intensity at wavelength  when the sensing
low cost compared to other tapered fiber-based SPR sensors which solution is present, D is the intensity of the dark spectrum (when
use complicated masking techniques or SMF instead. The sensing no light passes through the fiber) and R is the reference intensity
area is prepared by coating the whole tapered area of the MMF that is stored before adding any sample.
by a thin layer of gold without any masking. The resonance wave- A series of aqueous solutions made from NaCl (sodium chloride)
lengths are determined by placing sensing solutions of different RIs in water dissolved in dionized water was papered to be used as
around the sensing area. The sensor is based on wavelength inter- sensing solutions. The refractive index of these solutions in a range
rogation method. We experimentally investigate the influence of from 1.33 to 1.37 was determined using ATAGO pocket refractome-
waist diameter and taper length (waist and transition lengths) on ter at room temperature (i.e., 25 ◦ C).
the performance of the fabricated SPR sensor.

3. Performance parameters
2. Experiments
The performance of SPR sensors is typically evaluated using two
The tapered optical fiber (TOF)-based SPR sensor was fabri- main parameters: sensitivity and detection accuracy which both
cated from a standard graded-index MMF with silica core and should be kept as high as possible to achieve higher performance.
cladding diameters of 62.5 and 125 ␮m, respectively, using a glass- The sensitivity depends on the amount of wavelength shift associ-
processing machine. About 50–70 mm of the outer jacket length ated with the change in the refractive index of the sensing medium
was removed from the multimode fiber by a specialized fiber strip- and it is given by Ref. [23]
per. The bare fiber was then cleaned with isopropyl alcohol and
tapered using a tapering machine. Fig. 1(a) shows the microscopic ␭res
S␭n = (2)
image of the tapered optical fiber which consists of three con- ns
tiguous regions: one taper waist region with a small and uniform
diameter and two narrowed transition regions with a gradually where ␭res is the shift in resonance wavelength and ns is the
decreased diameter (downtaper) and an increased diameter (upta- change in RI of the surrounding medium. According to Eq. (2),
per) adjacent to two untapered ends. After that, the whole tapered the larger the shift, the better the sensitivity. Studies concerning
region was coated with a thin gold film (around 55 nm) using a sput- the sensitivity of the SPR sensors aim to strengthen the coupling
ter coater. Gold thickness of 40–60 nm has been reported to realize between the evanescent field waves and SPW which depends on
high-sensitivity SPR sensors [20–22]. A photo of the actual fabri- light wavelength, fiber parameters, fiber geometry and metal layer
cated sensor is shown in Fig. 1(b). The tapered fibers were carefully properties. The sensitivity also depends on other aspects such as
mounted in a horizontal orientation inside the sputter chamber the noise induced by the light source, the spectrometer, fiber con-
and rotated to produce symmetric coating. The film thickness was nectors, temperature, humidity, and the accuracy of the sensing
measured using Alpha Step IQ Surface Profiler. liquids.
40 Y. Al-Qazwini et al. / Sensors and Actuators A 236 (2015) 38–43

Fig. 2. Schematic diagram of the experimental setup. The inset on the right shows a photo of the actual setup.

The detection accuracy or the signal-to noise ratio (SNR), on the 4. Results and discussion
other hand, depends on how accurately the SPR sensor can detect
the resonance wavelength and it is described by Ref. [24] 4.1. Waist diameter
 
␭res To investigate the effect of taper-waist diameter, Dw on the SPR
SNR (ns ) = (3)
␭1/2 resonance characteristics, we prepared and compared three sen-
sors made with different values of Dw (25, 35, and 85 ␮m) coated
where ␭1/2 is the full width at half maximum (FWHM) of the
with a 55-nm gold film without masking. The sensing length was
transmission dip.
set to 10 mm (uniform waist length of 6 mm and up/down taper-
As Eq. (3) shows, the detection accuracy depends on the spectral
ing regions of 2 mm each). The RI of the sensing medium was
width of the SPR curve; the narrower the SPR curve, the higher the
changed from 1.3353 to 1.3453. The measured SPR spectra are plot-
detection accuracy. Generally, the spectral width increases with the
ted in Fig. 4(a–c). The figures show that when we decreased the
number of reflections a ray undergoes which is given by Ref. [18]
taper-waist diameter from 85 ␮m to 25 ␮m, the resonance depth
LW dropped from 92% to 80.5%. This, in turn, leads to higher sensitiv-
Nref (, z) = (4)
2(Z)tan( + ) ity as illustrated in Fig. 5. The SPR sensitivity increased from 1395
to 1914 nm/RIU and the average FWHM also increased from 167
where (z) is the fiber taper radius at distance z from the input end to 217 nm when we reduced the taper-waist diameter from 85 to
of the optical fiber,  is the angle between the ray and the normal 25 ␮m. This is due to the increase in the angular range of guided rays
to the interface in the input fiber, LW is the taper-waist length, and launched into the fiber. The increasing number of guided modes
˝ is the taper angle which can be expressed by Ref. [19] strengthens the coupling between the evanescent waves and the
  SPs which is responsible for the enhancement in the SPR sensi-
(i − o )
˝ = tan−1 (5) tivity and broadening in the spectral width. Sensors with thinner
Lt
tapered regions, therefore, have higher sensitivity towards changes
where i is the radius of the fiber core and o is the radius of the in RI of the surrounding medium but they tend to be fragile and easy
tapered fiber (i.e., taper-waist radius), and Lt is the transition region to break. A tradeoff between the sensitivity and robustness is thus
length. A schematic diagram of a tapered optical fiber is shown in required when the sensor is designed.
Fig. 3.
Eq. (5) shows that the taper angle is inversely proportional to the
taper-waist radius and the transition region length. The increase 4.2. Waist length
in the taper angle thereby broadens the angular range of guided
rays launched into the fiber. It also makes the incident angle of the To investigate the effect of taper-waist length, Lw on the SPR
rays closer to the critical angle leading to deeper penetration of resonance characteristics, we used a set of two sensors made
the evanescent wave inside the sensing region. This strengthens with different values of Lw (3 and 9 mm) coated without masking
the coupling between the evanescent waves and the SPs, and thus,
enhances the SPR sensitivity and broadens of the SPR spectrum [19].
From Eq. (4), it can also be seen that the number of reflections the
ray undergoes in the fiber is directly proportional to the taper-waist
length and inversely proportional to the taper-waist diameter.
In our experiment, the SPR is active along the entire length of
the taper as the gold film is deposited on the whole tapered region
without masking. Masking out parts of the taper region decreases
the range of incident angles. This, in turn, decreases the minimum.
However, the deposition of an additional masking layer is another
parameter to control which makes the sensor design more complex. Fig. 3. Schematic diagram of a tapered optical fiber.
 Y. Al-Qazwini et al. / Sensors and Actuators A 236 (2015) 38–43 41

Normalized Transmission (a.u.)

100 1870 25 μm
Dw = 25 μm

Sensitivity (nm/RIU)
1700
90
35 μm

1530
80
Reference 85 μm
1.3353 1360
1.3453 1.3450 1.3500 1.3550 1.3600 1.3650
70 Refractive Index
500 600 700 800
Wavelength (nm) Fig. 5. Relationship between the sensitivity and the refractive index of measuring
solutions for three waist diameters (25, 35, and 85 ␮m) of the sensing region of
(a) the sensor. The waist length is 6 mm and the up and down tapers length are 2 mm
respectively.
Normalized Transmission (a.u.)

100
Dw = 35 μm

Normalized Transmission (a.u.)

100
Lw= 3 mm
90

90
80
Reference
1.3353 80
1.3453 Reference
70 1.3365
500 600 700 800 1.3435
Wavelength (nm) 70
500 600 700 800
(b) Wavelength (nm)
Normalized Transmission (a.u.)

100 (a)
Dw = 85 μm
Normalized Transmission (a.u.)

100
Lw=9 mm
90
90

80
Reference
80
1.3353 Reference
1.3453 1.3365
70 1.3435
500 600 700 800
70
Wavelength (nm) 500 600 700 800
Wavelength (nm)
(c)
(b)
Fig. 4. (a–c) SPR transmitted spectra at three different waist diameters, Dw : 25 ␮m
in (a), 35 ␮m in (b), and 85 ␮m in (c) with Au thickness of 55 nm; the RI of the Fig. 6. (a,b) SPR transmitted spectra at two different waist lengths, Lw : 3 mm in (a)
ambient is changed from 1.3353 to 1.3453. The waist length is 6 mm and the up and and 9 mm in (b) with Au thickness of 55 nm. The waist diameter is 35 ␮m and the
down tapers length are 2 mm respectively. up and down tapers length are set to 3 mm respectively.

with 55 nm of gold. The waist diameter and the transition length

(up/down tapering regions) were set to 35 ␮m and 3 mm, respec- 185 to 138 nm by reducing the taper-waist length from 9 to 3 mm.,
tively. We measured the transmitted SPR spectra and plotted them This is because when the taper waist length is shorter, the number
in Fig. 6(a,b). We varied the RI of the sensing medium from 1.3365 of reflections the ray undergoes in the fiber is minimized as illus-
to 1.3435, but in Fig. 6 we only show two values to illustrate the trated in Eq. (4). Furthermore, when the taper-waist diameter is
impact on the resonance shift. This figure shows that the reso- considerably large as in our experiment, the light is confined into
nance depth changed from 87.5% to 85.5% with the decrease of the core of the optical fiber. Thus, any further increase in the waist
waist length 9–3 mm. Fig. 7 shows that the sensitivity increased length will weaken the coupling between the evanescent field and
from 1255 to 1547 nm/RIU and the average FWHM dropped from the SPWs which degrades the sensitivity of the SPR sensor.
42 Y. Al-Qazwini et al. / Sensors and Actuators A 236 (2015) 38–43

Normalized Transmission (a.u.)

1600
3 mm 100
Sensitivity (nm/RIU) Lt = 1 mm
1500
90
1400

80
1300 Reference
9 mm
1.3350
1.3450
1200 70
1.3450 1.3500 1.3550 1.3600 1.3650 500 600 700 800
Refractive Index Wavelength (nm)
Fig. 7. Relationship between sensitivity and refractive index of measuring solutions (a)
for two waist lengths (3 and 9 mm) of the sensing region of the sensor. The waist
diameter is 35 ␮m and the up and down tapers are 3 mm respectively.

Normalized Transmission (a.u.)

100
Lt = 2 mm
4.3. Transition (downtaper/uptaper) lengths

To investigate the effect of transition length (up/down tapers), 90

Lt on the SPR response, we used a set of three sensors made with
different values of Lt (1, 2, and 4 mm) coated with 55 nm of gold
without masking. The waist diameter and length were set to 35 ␮m
80
and 3 mm, respectively. The transmitted SPR spectra are shown Reference
in Fig. 8(a–c). The figures show that the shift in resonance wave- 1.3350
length increases with the decrease of transition length which leads 1.3450
to higher sensitivity as demonstrated in Fig. 9. The SPR sensitivity 70
increased from 1485 to 1591 nm/RIU whereas the average FWHM 500 600 700 800
exhibited a marginal increment from 143 to 147 nm when we Wavelength (nm)
reduced the transition length from 4 to 1 mm. This is attributed
to the increase in the taper angle (see Eq. (5)) which leads to wider (b)
angular range of guided rays into the fiber and also stronger cou-
Normalized Transmission (a.u.)

pling between the evanescent waves and the SPs. Therefore, sensors 100
with shorter transition regions have higher magnitude of evanes-
Lt = 4 mm
cent field emanating from the optical fiber. This leads to higher
sensitivity to changes in the RI of the surrounding medium. In terms
of detection accuracy, the broadening of SPR response curves is not 90
that significant.

5. Conclusions 80
Reference
1.3353
50
In this paper, we fabricated an SPR sensor based on unmasked
1.3453
1.3450
short tapered multimode fiber. The sensing area was prepared
70
by depositing gold on the entire tapered region without mask- 500 600 700 800
ing which significantly simplifies the sensor fabrication. We Wavelength (nm)
investigated the impact of the taper parameters including taper
waist-diameter, waist length, and transition lengths on the perfor- (c)
mance of the fabricated sensor. We have found that sensors with
thinner tapered regions have higher sensitivity towards changes in Fig. 8. (a–c) SPR transmitted spectra at three different values of up/down taper
lengths, Lt : 1 mm in (a), 2 mm in (b), and 4 mm in (c) with Au thickness of 55 nm;
RI of the surrounding medium. However, they display lower detec-
the RI of the ambient is changed from 1.3350 to 1.3450. The waist diameter is 35 ␮m
tion accuracy as the spectral width is broadened with shorted waist and the waist length is 3 mm.
diameter. In addition, they tend to be fragile and easy to break. We
have also demonstrated that the waist length should be kept short
to enhance both the sensitivity and detection accuracy of the sen- to achieve a sensitivity of 1600–2000 nm/RIU and a spectral width
sor as well as the resonance depth. Similarly, sensors with shorter of 140–220 nm.
transition regions have a higher magnitude of evanescent field
emanating from the optical fiber. This thereby increases the sen-
sitivity of the sensor without significant broadening in the spectral Acknowledgments
width. Therefore, we can optimize the sensor performance in terms
of sensitivity, detection accuracy, design simplicity, and robust- This work was partly supported by Universiti Putra Malaysia’s
ness by tapering our standard multimode fiber up to 25–45 ␮m Research University Grant Schemes (Ref: 05-01-12-1626RU and
while maintaining a short taper with a total length of about 3–5 mm 05-02-12-2015RU) and Ministry of Higher Education, Malaysia’s
coated with a thin film of gold of around 55 nm without masking Fundamental Research Grant Scheme (Ref: 03-04-10-795FR). Good
 Y. Al-Qazwini et al. / Sensors and Actuators A 236 (2015) 38–43 43

[22] H.R. Gwon, S.H. Lee, Spectral and angular responses of surface plasmon
1 mm resonance based on the Kretschmann prism configuration, Mater. Trans. 51
1600
(2010) 1150–1155.
Sensitivity (nm/RIU) 2 mm [23] J. Homola, On the sensitivity of surface plasmon resonance sensors with
spectral interrogation, Sens. Actuators B 41 (1997) 207–211.
[24] J. Sun, K. Shida, Multilayer sensing and aggregation approach to
1550 environmental perception with one multifunctional sensor, IEEE Sens. J. 2
(2002) 62–72.

1500 4 mm Biographies

Yusser Al-Qazwini received her Bachelor of Engineer-

1450 ing (Computer and Communication Systems Engineering)
1.3450 1.3500 1.3550 1.3600 1.3650
from Universiti Putra Malaysia (UPM), Malaysia (2006).
Refractive Index She is currently undertaking her Ph.D. at UPM, Malaysia on
the design and fabrication of surface plasmon resonance
Fig. 9. Relationship between the sensitivity and refractive index of measuring solu- (SPR) sensors based on optical fibers. Her research inter-
tions at three different values of up/down taper lengths (1, 2, and 4 mm) of the ests include SPR sensors, refractive-index sensors, and
sensing region of the sensor. The waist diameter is 35 ␮m and the waist length is nanotechnology.
3 mm.

suggestions of Dr. Zaineb Al-Qazwini in National University of

Singapore are highly appreciated.
A.S.M. Noor received his B. Eng. degree in Electrical and
Electronics Engineering from Niigata University, Japan in
2002, M.Sc. from Universiti Putra Malaysia and Ph.D. in
References Engineering (Optical Science) from Shizuoka University,
in 2006 and 2009 respectively. He is the current Head of
[1] L.S. Jung, C.T. Campbell, T.M. Chinowsky, M.N. Mar, S.S. Yee, Quantitative Department for the Department of Computer and Com-
interpretation of the response of surface plasmon resonance sensors to munication Systems Engineering, Faculty of Engineering,
adsorbed films, Langmuir 14 (1998) 5636–5648. Universiti Putra, Malaysia. His current research interests
[2] K.S. Johnston, S.R. Karlsen, C.C. Jung, S. Yee, New analytical technique for include surface plasmon related phenomena and nonlin-
characterization of thin films using surface plasmon resonance, Mater. Chem. ear optics.
Phys. 42 (1995) 242–246.
[3] Y. Yuan, L. Ding, Z. Guo, Numerical investigation for SPR-based optical fiber
sensor, Sens. Actuators B 157 (2011) 240–245.
[4] S. Kumar, G. Sharma, V. Singh, Sensitivity of tapered optical fiber surface
plasmon resonance sensors, Opt. Fiber Technol. 20 (2014) 333–335. Mohd Hanif Yaacob received his Bachelor of Engineer-
[5] S. Shukla, N.K. Sharma, V. Sajal, Sensitivity enhancement of a surface plasmon ing (Electronic Computer Systems) and Master of Science
resonance based fiber optic sensor using ZnO thin film: a theoretical study, (Communication and Network Engineering) from Sal-
Sens. Actuators B 206 (2015) 463–470. ford University, UK (1999) and Universiti Putra Malaysia,
[6] Y. Al-Qazwini, A.S.M. Noor, M.H. Yaacob, S.W. Harun, M.A. Mahdi, Fabrication Malaysia (2002), respectively. He did his Ph.D. research
and characterization of a refractive index sensor based on SPR in an etched at RMIT University, Melbourne, Australia (2012) in the
plastic optical fiber, Procedia Eng. 120 (2015) 969–974. area of optical sensor based on nanomaterials for chemi-
[7] Y. Zhao, Z.-q. Deng, Q. Wang, Fiber optic SPR sensor for liquid concentration cal sensing applications. His research interests are optical
measurement, Sens. Actuators B 92 (2014) 229–233. sensors, optical communication systems and nanotech-
[8] A.B. Socorro, I. Del Villar, J.M. Corres, F.J. Arregui, I.R. Matias, Spectral width nology. Currently, he is a lecturer at the Department of
reduction in lossy mode resonance-based sensors by means of tapered optical Computer and Communication Systems Engineering and a
fibre structures, Sens. Actuators B 200 (2014) 53–60. principal researcher at the Wireless and Photonic Network
[9] Y. Zhao, Z.-q. Deng, J. Li, Photonic crystal fiber based surface plasmon Research Center, Universiti Putra, Malaysia.
resonance chemical sensors, Sens. Actuators B 202 (2014) 557–567.
[10] N. Cennamo, G. D’Agostino, M. Pesavento, L. Zeni, High selectivity and
sensitivity sensor based on MIP and SPR in tapered plastic optical fibers for Sulaiman Wadi Harun received the B.E. degree in electri-
the detection of l-nicotine, Sens. Actuators B 191 (2014) 529–536. cal and electronics system engineering from the Nagaoka
[11] C. Ribaut, et al., Small biomolecule immunosensing with plasmonic optical University of Technology, Nagaoka, Japan, in 1996, and
fiber grating sensor, Biosens. Bioelectron. 77 (2016) 315–322. the M.Sc. and Ph.D. degrees in photonics from the Uni-
[12] F. Gu, H. Zeng, L. Tong, S. Zhuang, Metal single-nanowire plasmonic sensors, versity of Malaya, Kuala Lumpur, Malaysia, in 2001 and
Opt. Lett. 38 (2013) 1826–1828. 2004, respectively. He is currently a full professor at Pho-
[13] J. Cao, T. Sun, K.T.V. Grattan, Gold nanorod-based localized surface plasmon tonics Research Centre, University of Malaya. His current
resonance biosensors: a review, Sens. Actuators B 195 (2014) 332–351. research interests include fiber-optic active and passive
[14] T. Anttalainen, V. Jaaskelainen, Introduction to Communication Networks, devices.
Artech House, 2014.
[15] A. Dhawan, Development of Robust Fiber Optic Sensors Suitable for
Incorporation into Textiles and a Mechanical Analysis of Electronic Textile
Circuits, ProQuest, 2007.
[16] W.B. Lin, N. Jaffrezic-Renault, A. Gagnaire, H. Gagnaire, The effects of
polarization of the incident light-modeling and analysis of a SPR multimode
optical fiber sensor, Sens. Actuators A 84 (2000) 198–204. Mohd Adzir Mahdi received his B.Eng. degree from the
[17] W. Henry, Evanescent field devices: a comparison between tapered optical Universiti Kebangsaan Malaysia, and M.Sc. and Ph.D.
fibres and polished or D-fibres, Opt. Quant. Electron. 26 (1994) S261–S272. degrees from the Universiti Malaya in 1996, 1999 and
[18] J. Villatoro, D. Monzón-Hernández, E. Mejía, Fabrication and modeling of 2002 respectively. He joined the Faculty of Engineering,
uniform-waist single-mode tapered optical fiber sensors, Appl. Opt. 42 (2003) Universiti Putra Malaysia in 2003. Since 1996, he has
2278–2283. been involved in photonics research specializing in opti-
[19] S. Ju, et al., Experimental demonstration of surface plasmon resonance cal amplifiers, lasers and sensors. He has authored and
enhancement of the tapered optical fiber coated with Au/Ti thin film, J. coauthored over 270 journal papers and 200 conference
Non-Cryst. Solids 383 (2014) 146–152. papers. His research interest includes optical communica-
[20] Y. Al-Qazwini, A. Noor, P. Arasu, A. Sadrolhosseini, Investigation of the tions, optical sensors and nonlinear optics.
performance of an SPR-based optical fiber sensor using finite-difference time
domain, Curr. Appl. Phys. 13 (2013) 1354–1358.
[21] R. Slavıı́k, J. Homola, J. Čtyroký, E. Brynda, Novel spectral fiber optic sensor
based on surface plasmon resonance, Sens. Actuators B 74 (2001) 106–111.