Escolar Documentos
Profissional Documentos
Cultura Documentos
Laser & Plasma Technology Division, Bhabha Atomic Research Centre, Mumbai-400085, India
3
Electrical Communication Engineering Department, Indian Institute of Science, Bengaluru-560012, Karnataka, India
*Corresponding author: usraykar_kud@yahoo.co.in
Received 31 May 2011; revised 20 August 2011; accepted 8 September 2011;
posted 12 September 2011 (Doc. ID 148458); published 2 November 2011
Through this paper we experimentally demonstrate the fabrication of a fiber Bragg grating (FBG)
chemical sensor to detect and determine the manganese concentration in water and compare our results
with sophisticated spectroscopic methods, such as atomic absorption spectrometry and the inductively
coupled plasma method. Here we propose a simple method to develop a thin layer of gold nanoparticles
above the etched grating region to enhance the sensitivity of the reflected spectrum of the FBG. By doing
so, we achieve a sensitivity of 1:26 nm=parts per million in determining the trace level of Mn in water.
Proper reagents are used to detect manganese in water. 2011 Optical Society of America
OCIS codes: 230.0250, 230.1150, 230.2285, 060.2370, 060.3735.
1. Introduction
gases from the ground water, these salts can still occur in large dissolved concentrations. Manganese
content in water produces unpleasant dark stains
due to oxidation on every surface with which it comes
into contact, even at low concentrations. Because of
the precipitated manganese deposits collected in pipelines, tap water may contain black sediment and
turbidity [1]. Higher concentrations of Mn in water
may bring about various health-related problems
in both animals and plants. Adverse neurological
effects [decreased performance in school and in neurobehavioral examinations of the World Health
Organization (WHO) core test battery] were reported
in 11- to 13-year-old children, who were exposed to
excess manganese through ingestion of contaminated water and from wheat fertilized with sewage
water [25]. Long-term studies concluded that progressive increase in the manganese concentration
in drinking water is associated with a higher prevalence of neurological problems of chronic manganese
10 November 2011 / Vol. 50, No. 32 / APPLIED OPTICS
6033
The FBG is prepared by inscribing a periodic refractive index modulation in the core of an optical fiber,
with periodicity of a few micrometers. When a broadband light is injected into the FBG, light in a narrowband of wavelengths centered at a particular
wavelength known as the Bragg wavelength (B ) is
reflected because of contradirectional coupling of
the fundamental mode introduced by refractive index modulation in the core [7]. The FBG works on
the principle of Bragg reflection by reflecting wavelength B , given by
B 2neff ;
1
refractive index of the surroundings, since the reflected spectrum is the result of coupling the corebound mode with the backward propagating modes.
But in FBG-based chemical sensors, the change of
Bragg wavelength is induced by change in the chemical composition around the sensor. In order to measure small changes in composition, it is important for
the optical mode to penetrate evanescently into the
surrounding solution. To design a FBG-based chemical sensor, the fiber cladding in the grating region
should be removed by etching so that the core mode
interacts directly with the external medium. In this
configuration, the value of the effective refractive index of the waveguide mode is directly affected by the
refractive index of the medium in which the fiber is
immersed. When the optical fiber cladding is etched
to a depth until the fundamental waveguide mode is
affected, the propagation constant of the mode is
modified, as can be written as [15,19]
0 kp nsur ncld ;
2
Ap
2 dA
2 dA
3
A is the area of the cross section of the unetched fiber and is the wave function of the fundamental
mode in a weakly guiding circular fiber with a step
index profile. Because of the power loss that occurs in
the etched region of the fiber, the reflected power of
the FBG decreases when the fiber radius becomes
smaller. If p1 and p2 are two values of factor p at
two different diameters of etched fiber, from Eq. (2)
the associated variation of the modified propagation
constant of the fundamental mode would be
2 1 kp1 p2 nsur ncld :
4
5
6
8
Power (in W)
B 2neff :
0.00003
0.00002
0.00001
0.00000
1545
1546
1547
1548
1549
3. Experiment
A.
Bragg wavelength
Peak power
Reflectance
FWHM
1546:96 nm
30:16 W
75%
0:2 nm
6035
1547.0
0.000025
Experimental data
0.000020
1546.9
Air
DMWater
0.04ppm
0.06ppm
0.09ppm
Test Sample
1546.8
1546.7
0.000015
0.000010
0.000005
1546.6
0.000000
1546.5
0
10
20
30
40
50
Core
1546
1547
Wavelength (nm)
Cladding
1545
60
Cladding etched
grating region
Bragg WL
Linear Fit
1546.18
1546.16
1546.14
1546.12
1546.10
1546.08
1546.06
0.00
(a)
0.02
0.04
0.06
0.08
0.10
Concentration (ppm)
(b)
Power
Linear Fit
20
19
18
17
16
15
14
13
added. These test solutions were used for measurements with our FBG-based sensor. The FBG sensor
was coupled to a broadband light source and immersed into a test tube containing the manganese solution. The spectrum of the Bragg-reflected light was
recorded on an OSA through a 3 dB coupler. Figure 6
shows a schematic diagram of the experimental arrangement. The reflected spectra were recorded for
6036
0.00
0.02
0.04
0.06
0.08
0.10
different concentrations of manganese solution ranging from 0.04 to 0:09 ppm. As the concentration of
the manganese solution surrounding the sensor head
Table 2.
Chemical
Species
FBG Sensor
(ppm)
Atomic Absorption
Spectrometera (ppm)
Inductively Coupled
Plasmab (ppm)
WHO Permissible
Limit (ppm)
Manganese
0.0303
0.048
<0:05
0.05
a
b
From the spectral response of the FBG sensor at different Mn concentrations, as shown in Fig. 7, it is observed that as the concentration of the surrounding
medium increases, the reflected spectrum shifts toward a longer wavelength. With the increase in concentration of the solution surrounding the etched
FBG, neff of the core mode increases, resulting in a
shift of B toward a longer wavelength. Figure 8 represents the graph of the Bragg wavelength (B ) as a
function of the concentration of Mn solutions. The
shift in B is linear and the linear least-square data
fit with the experimental data gives a correlation factor of 0.95859 and a RMS error of 11:937 103 nm.
The shift of Bragg wavelength was 0:104 nm, when
the Mn concentration of the solution was increased
to 0:09 ppm. The concentration of Mn in the test samples can be determined using the least-square-fit
graph of B versus concentration. From the graph,
the concentration of the test samples from the local
ground water sources is found to be 0:0303 ppm. The
sensitivity of this sensor is found to be 1:26 nm
per ppm.
There is also a linear shift in the reflected light
intensity and power at B with the surrounding
concentration, as shown in Fig. 9. In this case, the correlation factor is 0:9803 and the RMS error is
51:33 102 W. It may be understood here that the
power mentioned in Fig. 9, in Figs. 7 and 2, and elsewhere in the text, is the power measured by the OSA,
which is proportional to the actual reflected power
from the FBG. The peak intensity at the Bragg wavelength decreases with increase in the concentration of
the manganese solution. This is due to the etchinginduced multimodal propagation conditions along
the etched region [22]. Using this graph, the concentration of manganese in our test sample is found to be
In summary, we have demonstrated a highly sensitive etched FBG sensor using BGe codoped fiber
to detect traces of Mn in drinking water. We have also
observed the enhancement in sensitivity of a chemical sensor by depositing gold nanoparticle layers on
the unclad portion of a FBG. FBG sensors can be
used for measuring the manganese compounds in
water with good sensitivity in the lower concentration range. The advantages of this type of sensor are
its simplicity in construction and ease of use. The
sensor can find many applications in pharmaceuticals and industrial units, including in food industries. By optimization of some key parameters,
such as the length of unclad fiber, refractive index
profile, and optical design, better measurement sensitivity may be achieved.
We gratefully acknowledge support by the Department of Atomic EnergyBoard of Research in Nuclear Sciences (DAE-BRNS) (BARC, Mumbai, India),
Grant No. 2006/34/22-BRNS/2801. The authors also
acknowledge Prof. Kattesh Katti and Prof. Raghuraman Kannan, University of MissouriColumbia,
Mo., USA for the courtesy of the TEM image of the
gold nanoparticles and Prof. V. J. Fulari, Shivaji
University, Kolhapur, India for providing the SEM
image of the gold nanoparticles.
References
1. APEC Water Systems, FreeDrinkingWater.com, http://www
.freedrinkingwater.com/water_quality/chemical/waterproblems
manganese.htm.
2. M. M. Canavan, S. Cobb, and C. Srinker, Chronic manganese
poisoning, Arch. Neurol. Psychiatry 32, 501512 (1934).
3. D. G. Cook, S. Fahn, and K. A. Brait, Chronic manganese
intoxication, Arch. Neurol. 30, 5964 (1974).
4. P. He, D. H. Liu, and G. Q. Zhang, Effects of high-level
manganese sewage irrigation on childrens neurobehavior,
Chin. J. Prev. Med. 28, 216218 (1994).
5. G. Zhang, D. Liu, and P. He, Effects of manganese on learning
abilities in school children, Chin. J. Prev. Med. 29, 156158
(1995).
6. X. Kondakis, N. Makris, M. Leotsinidis, M. Prinou, and
T. Papapetroplous, Possible health effects of high manganese
concentration in drinking water, Arch. Environ. Health 44,
175178 (1989).
10 November 2011 / Vol. 50, No. 32 / APPLIED OPTICS
6037
6038
16.
17.
18.
19.
20.
21.
22.