Beruflich Dokumente
Kultur Dokumente
This content has been downloaded from IOPscience. Please scroll down to see the full text.
(http://iopscience.iop.org/0957-0233/27/7/075107)
View the table of contents for this issue, or go to the journal homepage for more
Download details:
IP Address: 138.4.8.93
This content was downloaded on 31/03/2017 at 16:03
Intensity liquid level sensor based on multimode interference and fiber Bragg grating
Ricardo Oliveira, Stenio Aristilde, Jonas H Osrio et al.
Intensity modulated SMF cascaded tapers with hollow core PCF based micro cavity for curvature
sensing
Sumit Dass, Jitendra Narayan Dash and Rajan Jha
High force measurement sensitivity with FBGs fabricated in uniform-waist fiber tapers
Torsten Wieduwilt, Sven Brckner and Hartmut Bartelt
Fabrication and characterization of a metal-packaged regenerated fiber Bragg grating strain sensor
for structural integrity monitoring of high-temperature components
Yun Tu and Shan-Tung Tu
MachZehnder interferometer based on tapered PCF with an up-tapered joint for curvature, strain and
temperature interrogation
Jitendra Narayan Dash and Rajan Jha
Measurement Science and Technology
E-mail: oliveiraricas@av.it.pt
Abstract
We report the development of an optical fiber sensor capable of simultaneously measuring
strain, temperature and refractive index. The sensor is based on the combination of two fiber
Bragg gratings written in a standard single-mode fiber, one in an untapered region and another
in a tapered region, spliced to a no-core fiber. The possibility of simultaneously measuring
three parameters relies on the different sensitivity responses of each part of the sensor. The
results have shown the possibility of measuring three parameters simultaneously with a
resolution of 3.77 , 1.36 C and 5104, respectively for strain, temperature and refractive
index. On top of the multiparameter ability, the simple production and combination of all the
parts involved on this optical-fiber-based sensor is an attractive feature for several sensing
applications.
Keywords: MMI, multimode interference, fiber sensor, fiber Bragg grating
2
Meas. Sci. Technol. 27 (2016) 075107 R Oliveira etal
where, FBG1 and FBG2 represent the wavelength shifts water in it. The temperature was swept from 2285 C and no
of the FBG in tapered and untapered fiber, respectively; strain was imposed during the tests. For the refractive index
NCF denotes the wavelength shift of the NCF; and K, tests, the liquid container was filled with six different solu
KT and Kn are the sensitivity coefficients, corresponding tions of water/isopropyl alcohol whose refractive index was
to the changes of strain, temperature and refractive index, previously measured. For each solution, the temperature of
respectively. the bath was maintained at a constant 22 C and no strain was
Subscripts FBG1, FBG2 and NCF refer to the individual imposed on the sensor. The refractive index of the different
contribution of the FBGs in the untapered/tapered fiber and solutions was measured at 22 C using a Palm Abbe hand
SMS structure. The three parameters under study can, there held refractometer with a resolution of 1104 and a 590nm
fore, be calculated through the equivalent matrix as wavelength radiation.
FBG1
4. Results and discussion
T = M FBG2 ,
1
(3)
n NCF Regarding the results obtained for the FBGs strain test
(figures 2(a) and (c)), it can be observed that both FBGs pre
where M1 is the inverse coefficient matrix. sented a red shift with increasing strain, with values of 5.77
The wavelength measurement resolution (i.e. ( FBG1), and 0.92 pm 1, respectively for the FBG in tapered and
( FBG2) and (NCF)), given by the resolution of the acqui untapered fiber. The red shift was expected since the FBGs
sition system, will determine the uncertainty on the measured grating pitch increases with increasing strain.
strain, temperature and refractive index. The correspondent On other hand, the sensitivities obtained for the untapered
values can be calculated through and tapered FBGs were decreased and increased respectively,
( ) ( FBG1) when compared with the value of 1.2 pm 1 found for the
standard characterization of an FBG written in the same SMF-
(T ) = M ( FBG2 ) .
1
(4) 28. This result is due to the unequal strain distributions along
(n ) (NCF ) the fiber sensor. The relation between the strain applied in
each part of the sensor and the correspondent cross-sectional
area can be written as
3. Experimental setup fiber Ataper
(5) = ,
taper Afiber
The experimental setup used to characterize the fiber sensor is
shown in figure1. Both fiber tips were glued to a fixed stage where fiber and taper refer to the strain on the fiber and taper,
and another to a motorized linear stage (UTS 150cc). The respectively; Afiber and Ataper refer to the cross-sectional area of
distance between the stages where the glue was inserted was the fiber and taper respectively. Thus, the applied strain will
27cm. On the central region of the setup, a liquid container be higher in the tapered region than in the untapered region,
was placed above a hot plate (IKAC-MAG HS7). The liquid since its area is smaller [20]. This leads to a decrease in the
container is used to surround the fiber sensor with a specific sensitivity on the untapered FBG, and to an increase in the
refractive index. The hot plate is, therefore, used to control the tapered FBG [20].
temperature of the solutions. In order to observe the FBG and Using the definition of strain and following the deductions
SMS spectral responses, an interrogator system (FS2200 given at [20], we can theoretically estimate the sensitivity for
Industrial BraggMETER, FiberSensing) was used to measure the FBG written on the untapered sectionof the fiber as
the reflection signal from the FBGs as well as the transmission
K1(Theo.) = K(FBG)
signal from the SMS structure. The fiber sensor configuration
used to measure the reflection signal from the FBGs is pre L FBG1 + L FBG2 + L fiber + L taper
,
sented in figure1, where the FBG precedes the SMS structure.
On other hand, if the opposite configuration is chosen (SMS
L taper ( ) +L ( ) +L
d FBG1 2
dtaper FBG2 d
d FBG1 2
FBG 2
fiber + L FBG1
followed by the FBGs), the FBG reflectivity will be modified (6)
by the envelope of the SMS spectra, that is dependent on the and for the FBG written on the tapered sectionas
external conditions [4, 16]. Therefore, the uncertainty present
K2(Theo.) = K(FBG)
on the peak wavelength detection with such a configuration
could be higher, leading us to exclude this scheme. L FBG1 + L FBG2 + L fiber + L taper
The sensor characterization was performed, getting the
sensitivity coefficients of each parameter individually: one
L FBG1 ( ) +L ( ) +L
d FBG2 2
d FBG1 fiber
d FBG2 2
dfiber taper + L FBG2
parameter changes while two others remain constant. For (7)
the strain test, the fiber sensor was kept in water, at a con where K(FBG) refers to the strain coefficient of a common
stant temperature of 22 C. The strain was imposed in steps FBG written in an SMF28 fiber (1.2 pm 1); L (FBG1),
of 92.6 in a range of 1389.0 . In order to characterize L (FBG2), L(fiber), and L(taper) are the lengths of each sensor sec
the sensor to temperature, the liquid container was left with tion; d FBG1, d FBG2, dfiber, and dtaper refer to the diameters of
3
Meas. Sci. Technol. 27 (2016) 075107 R Oliveira etal
20 = 0
= 0 SMF28 FBG
(a) 1 (b) 8 (c)
Normalized power
= 1389 NCF
= 1389
30 0.8 6
Power [dBm]
Figure 2. (a) Reflection spectra collected for both FBGs at different strain values; (b) normalized transmission spectra collected for the
NCF with different strain values; (c) wavelength shifts from the wavelength peak power for all the parts involved in the sensor at different
strain values.
n = 1.3598
tapered and untapered sectionof the SMF-28 will also occur 0.8 n = 1.3780
NCF
are comparable to the ones found for a standard single-mode
optical fiber, 5.5 m, as reported in [21].
Regarding the experimental characterization of the refrac 4
tive index, the resonant Bragg wavelength for the FBGs does
not change with the different solutions (see the data points 2
SNCF = 116.5 0.4 nm/RIU;
collected for the Bragg wavelength shifts on figure 4(b)).
R2 = 0.99
This result was expected because the fundamental mode on
0
the tapered/untapered regions of the fiber is strongly coupled
to the fiber core and, therefore, its evanescent field does not 1.34 1.35 1.36 1.37 1.38
interact with the surrounding medium. On other hand, the NCF Refractive Index
presents strong interaction with the surrounding environment Figure 4. (a) NCF-normalized transmission spectra collected for
(see the spectra change in figure4(a)). Since the fiber does not six different refractive index solutions; (b) wavelength shifts for all
present an outer cladding layer, the guided modes can strongly the parts involved in the sensor, for different index solutions.
4
Meas. Sci. Technol. 27 (2016) 075107 R Oliveira etal
20 T&n
(a) 1
T = 22 C
(b)
SMF28 FBG
Taper with FBG (c)
T T = 55 C
T = 22 C T = 85 C 2 NCFT,n
Normalized power
T
T = 85 C
1.5
0.6 SNCF = 34.82 0.45 pm/C;
40 T
Taper with FBG 1 2
R = 0.99
SMF28 FBG
0.4 SSMF28 FBG = 9.89 0.08 pm/C;
R2 = 0.99
50 0.5
0.2
S = 8.62 0.07 pm/C;
Taper FBG
60 0 0 R2 = 0.99
Figure 5. (a) Reflection spectra collected for both FBGs at different temperatures; (b) normalized transmission spectra collected for the
NCF with different temperatures, and considering the inherent refractive index change of water; (c) wavelength shifts collected from the
wavelength peak power for all the parts involved in the sensor at different temperatures.
interact with the external refractive index solutions. From the Table 1. System resolution comparison.
different NCF spectra collected, the peak wavelength change Proposed sensor Sensor 1 [19] Sensor 2 [10] Sensor 3 [11]
was calculated as the center wavelength at 30% of the max
imum peak. This was done in order to minimize the uncertainty () 3.77 7.71 1.96 140.77
on the peak wavelength position due to the power transfer T (C) 1.36 4.02 0.69 15.38
that appears between peaks (mainly on the last two spectra of N 5.0104 0.25 9.0104 5.9103
figure4(a)). The correspondent wavelength shifts were calcu
lated and can be seen in figure 4(b). Additionally, it can be almost four times higher than the ones found for the FBGs.
observed that the wavelength shift increases with increasing This higher value is probably due to the higher thermo-optic
refractive index and, as expected, the behavior is not linear coefficient of the NCF when compared to the SMF-28 [13].
under the range studied [1, 22]. In order to obtain a linear ten Using the above tests and linear fits, the correspondent ,
dency for the resonant wavelength shifts, a narrow refractive T and n, can be simultaneously calculated through
index range (~1.331.37) is considered. This range is still valid
0.92 9.89 FBG1
0
for several areas of engineering such as environmental sensing; 1
therefore, one can obtain a sensitivity of 116.5nm RIU1. (9)
T
= 5.77 8.62 0
FBG2 .
n 1.29 34.82 116 500 NCF
For the temperature characterization, the correspondent
spectra change for FBGs and NCF were obtained and can be
seen in figures 5(a) and (b)), respectively. From figure 5(c),
The sensor was experimentally tested, giving maximum
it can be noticed that the resonance wavelength shifts for the
relative errors of 0.4%, 11% and 8104 RIU, respectively,
FBGs are similar and positive with increasing temperature.
for strain, temperature and refractive index. The differences in
This red shift is mainly due to the thermo-optic coefficient
the maximum relative errors can be attributed to inaccuracy
(i.e. change of the refractive index of silica material with
and uncertainty during the calibration process and also on the
temperature). Regarding the results obtained for the NCF
peak wavelength detection, especially for the NCF spectra
(spectra in figure 5(b) and asterisk points on figure 5(c)), it
can be seen that there is a red shift of the peak wavelength that has a broadened shape. Considering a detection system
with increasing temperature. However, this test was done with with a wavelength resolution of 10 pm and using equation(4),
the fiber sensor immersed in water; therefore, the temperature the resolution of strain, temperature and refractive index can
changes imposed on the characterization test will indirectly be calculated respectively as 3.77 , 1.36 C and 5104.
induce a change in the refractive index of water [23]. This is In order to compare the proposed sensor with the ones
not the case with the FBGs since they are wavelength-inde found in literature, capable of simultaneously detecting
pendent of the external refractive index (figure 4(b)). strain, temperature and refractive index (i.e. [10, 11, 19], we
Therefore, the NCF wavelength shifts due to the temper have constructed a table where we can compare the system
ature changes alone (T), can be calculated as resolutions. The values were calculated using equation (4),
considering a detection system with a resolution of 10 pm
T = T, n Kn,NCF RWater T ,
(8) in wavelength and 0.02 dB in amplitude (for sensors based
where T, n is the overall shift due to the direct and indirect on power (i.e. [11, 19]). The values obtained for sensor 1,
contribution of temperature and water refractive index change, presented at [19], were calculated considering a silica fiber
respectively; Kn,NCF is the calculated refractive index sensi Youngs modulus of 70GPa. Additionally, the refractive index
tivity of the NCF (116.5nm RIU1); and RWater is the water value calculated for sensor 3 [11] was calculated individually
thermo-optic coefficient, calculated from the data points given since this parameter is taken independently from the temper
in [24] as2104 C1. The correspondent T found after ature and strain parameters. The correspondent system resolu
the correction can be seen on figure5(c) as the square points. tions can be seen in table1.
The correspondent NCF temperature sensitivity was then cal The values presented in table 1 show that the proposed
culated from the corrected wavelength shifts, giving a value sensor is well positioned among the different sensors found
5
Meas. Sci. Technol. 27 (2016) 075107 R Oliveira etal
in literature, achieving values very close to the ones presented [6] Antonio-LopezJE, LiKamWaP, Sanchez-MondragonJ and
for sensor 2 (i.e. [10]). May-ArriojaDA 2013 All-fiber multimode interference
The proposed fiber-optic sensor is simply produced due to micro-displacement sensor Meas. Sci. Technol. 24055104
[7] RanY, XiaL, HanY, LiW, RohollahnejadJ, WenY and
the easy concatenation of all the parts involved. However, the LiuD 2015 Vibration fiber sensors based on SM-NC-SM
NCF cleaving process needed to achieve a peak power wave fiber structure IEEE Photonics J. 76800607
length centered at the wavelength range of the measurement [8] LuoY, XiaL, YuC, LiW, SunQ, WangY and LiuD 2015
system needs to be done with special care. This can be easily Multi-parameter optical fiber sensor based on enhanced
solved using a learning process. multimode interference Opt. Commun. 3441204
[9] CusanoA, CutoloA and AlbertJ 2011 Fiber Bragg Grating
The capability to simultaneously discriminate strain, Sensors: Recent Advancements, Industrial Applications and
temperature and refractive index can be a useful tool; for Market Exploitation (Sharjah: Bentham Science Publishers)
instance, in the control of the cure processes of materials, [10] LeeS-M, SainiS and JeongM 2010 Simultaneous
where the knowledge of these three parameters is important. measurement of refractive index, temperature, and strain
using etched-core fiber Bragg grating sensors IEEE
Photonics Technol. Lett. 2214313
5.Conclusions [11] AlbertoNJ, MarquesCA, PintoJL and NogueiraRN 2010
Three-parameter optical fiber sensor based on a tilted fiber
Bragg grating Appl. Opt. 49608591
We have proposed and demonstrated a simple and effective [12] LiC, NingT, WenX, LiJ, ZhengJ, YouH, ChenH, ZhangC
all-fiber sensor to simultaneously measure strain, temper and JianW 2015 Strain and temperature discrimination
ature and refractive index. The ability to measure different using a fiber Bragg grating and multimode interference
variables without stabilization of the ambient conditions is effects Opt. Commun. 34369
[13] ZhouD-P, WeiL, LiuW-K, LiuY and LitJWY 2008
useful in different areas. The concatenation of the different Simultaneous measurement for strain and temperature
fiber technologies allows the possibility of simultaneously using fiber Bragg gratings and multimode fibers Appl. Opt.
measuring three parameters with a resolution of 3.77 , 47166872
1.36 C and 5104, respectively, for the strain, temper [14] YuS, PeiL, LiuC, WangY and WengS 2014 Simultaneous
ature and refractive index. The simple design and measure strain and temperature measurement using a no-core fiber-
based modal interferometer with embedded fiber Bragg
ment scheme is a highly promising feature for multiparameter grating Opt. Eng. 53087105
sensing applications. [15] WuQ, SemenovaY, YanB, MaY, WangP, YuC and FarrellG
2011 Fiber refractometer based on a fiber Bragg grating and
single-modemultimodesingle-mode fiber structure Opt.
Acknowledgments Lett. 3621979
[16] LiL, XiaL, WuangY, RanY, YangC and LiuD 2012 Novel
This work was funded by FINEP (#01.12.0393.00) and by FCT- NCF-FBG interferometer for simultaneous measurement of
Fundao para a Cincia e Tecnologia through Portuguese national refractive index and temperature IEEE Photonics Technol.
funds PEstOE/EEI/LA0008/2013 and UID/EEA/50008/2013 Lett. 24226871
[17] RongQ, QiaoX, GuoT, WangR, ZhangJ, HuM, FengZ,
(Swat), PhD Scholarship SFRH/BD/88472/2012, investigator WengY and MaY 2012 Temperature-calibrated fiber-optic
grant IF/01664/2014 and project INITIATE. J H Osrio acknowl refractometer based on a compact FBG-SMS structure
edges CNPq for financial support. Chin. Opt. Lett. 10030604
[18] LiC, NingT, WenX, LiJ, ZhangC and ZhangC 2015
Magnetic field and temperature sensor based on a no-core
References fiber combined with a fiber Bragg grating Opt. Laser
Technol. 721047
[19] MauJ, WuP, FuM and LiuW 2012 Simultaneous
[1] SilvaS, PachonEGP, FrancoMAR, HayashiJG, measurement of stress, temperature and refractive index
MalcataFX, FrazoO, JorgeP and CordeiroCMB 2012 using an PMFBG cascaded with an LPG Prog. Electromagn.
Ultrahigh-sensitivity temperature fiber sensor based on Res. Symp. Proc. (Kuala Lumpur, Malaysia) pp 15848
multimode interference Appl. Opt. 51323642 [20] FrazoO, SilvaSFO, GuerreiroA, SantosJL, FerreiraLA
[2] AndreRM, BiazoliCR, SilvaSO, MarquesMB, and ArajoFM 2007 Strain sensitivity control of fiber
CordeiroCMB and FrazaoO 2013 Strain-temperature Bragg grating structures with fused tapers Appl. Opt.
discrimination using multimode interference in tapered fiber 46857882
IEEE Photonics Technol. Lett. 251558 [21] AntunesP, LimaH, MonteiroJ and AndrePS 2008 Elastic
[3] ChenY, HanQ, LiuT and LuX 2015 Self-temperature- constant measurement for standard and photosensitive single
compensative refractometer based on singlemode- mode optical fibres Microw. Opt. Technol. Lett. 5024679
multimode-singlemode fiber structure Sensors Actuators B [22] BiazoliCR, SilvaS, FrancoMAR, FrazoO and
21210711 CordeiroCMB 2012 Multimode interference tapered fiber
[4] SunA and WuZ 2015 Multimode interference in single refractive index sensors Appl. Opt. 5159415
modemultimode FBG for simultaneous measurement of [23] YanJ, ZhangAP, ShaoL, DingJ and HeS 2007
strain and bending IEEE Sensors J. 1533904 Simultaneous measurement of refractive index and
[5] Antonio-LopezJE, Torres-CisnerosM, Arredondo-LucioJA, temperature by using dual long-period gratings with an
Sanchez-MondragonJ, LikamwaP and May-ArriojaDA etching process IEEE Sensors J. 713601
2010 Novel multimode interference liquid level sensors [24] LideDR 2004 CRC Handbook of Chemistry and Physics
Proc. SPIE 7839 78391V (Boca Raton, FL: CRC Press)