sensors Article
Ultra-Weak Fiber Bragg Grating Sensing Network Coated with Sensitive Material for Multi-Parameter Measurements Wei Bai 1 , Minghong Yang 1,2, *, Chenyuan Hu 1, *, Jixiang Dai 1 , Xuexiang Zhong 1 , Shuai Huang 1 and Gaopeng Wang 1 1
2
*
National Engineering Laboratory for Fiber Optic Sensing Technologies, Wuhan University of Technology, Wuhan 430070, China; [email protected] (W.B.); [email protected] (J.D.); [email protected] (X.Z.); [email protected] (S.H.); [email protected] (G.W.) Key Laboratory of Fiber Optic Sensing Technology and Information Processing, Ministry of Education, Wuhan, 430070, China Correspondence: [email protected] (M.Y.); [email protected] or [email protected] (C.H.); Tel.: +86-139-8624-5199 (M.Y.); +86-135-4500-0634 (C.H.)
Received: 11 May 2017; Accepted: 20 June 2017; Published: 26 June 2017
Abstract: A multi-parameter measurement system based on ultra-weak fiber Bragg grating (UFBG) array with sensitive material was proposed and experimentally demonstrated. The UFBG array interrogation principle is time division multiplex technology with two semiconductor optical amplifiers as timing units. Experimental results showed that the performance of the proposed UFBG system is almost equal to that of traditional FBG, while the UFBG array system has obvious superiority with potential multiplexing ability for multi-point and multi-parameter measurement. The system experimented on a 144 UFBG array with the reflectivity of UFBG ~0.04% for the four target parameters: hydrogen, humidity, temperature and salinity. Moreover, a uniform solution was customized to divide the cross-sensitivity between temperature and other target parameters. It is expected that this scheme will be capable of handling thousands of multi-parameter sensors in a single fiber. Keywords: fiber optics sensors; relativity; multi-parameter
1. Introduction The development trends towards the next generation of optical fiber sensing networks aim for huge capacity, long distance, and high precision. The huge capacity of optical fiber sensing networks includes two aspects of content, i.e. more points and more parameters of measurement. Jiang et al. proposed an on-line writing no-weld UFBG array through fiber drawing tower [1]. Then, Hu et al. demonstrated a single optical fiber network with over 1000 UFBG sensors by combining two semiconductor optical amplifiers (SOAs) and a time-division multiplexing (TDM) technique. Multi-point measurement based on UFBG array has made great improvement, while the multi-parameter measurement is still challenging [2,3]. In this paper, a serial TDM sensing network for multi-parameter measurement based on UFBGs coated with sensitive materials is proposed and experimentally demonstrated [4]. A UFBG array sensing system for hydrogen, relative humidity (RH), temperature and salinity measurements in a single fiber by coating different sensitive materials on different UFBGs is realized [5,6]. It can be concluded that thousands of UFBG sensors multiplexed in series with multi-parameter measurements could be possible, which shows promising applications for the future.
Sensors 2017, 17, 1509; doi:10.3390/s17071509
www.mdpi.com/journal/sensors
Sensors 2017, 17, 1509
2 of 10
2017, 17, 1509 2.Sensors Principle
2 of 10
2.1. 2.1.Sensing Sensingwith withCoated CoatedFBG FBG When Whenthe theFBG FBGisissubjected subjectedtotoexternal externalfield fieldeffect effect(such (suchasasstress, stress,temperature), temperature),the thespatial spatial period periodofofgrating gratingwill willchange, change,leading leadingtotothe thereflection reflectionorortransmission transmissiongrating gratingcentral centralwavelength wavelength drift. drift.Since Sincethey theyare areimmune immunetotochemical chemicalparameters, parameters,such suchas ashumidity, humidity,hydrogen, hydrogen,salinity, salinity,etc. etc.,the the FBG-based sensors areare fabricated by by coating corresponding sensitive materials on the FBG-basedmulti-parameter multi-parameter sensors fabricated coating corresponding sensitive materials on surface of theof fiber covering the Bragg instance, an RH-sensitive polymer the surface thecladding fiber cladding covering thegrating Bragg section. grating For section. For instance, an RH-sensitive ispolymer coated on the FBGonfiber measure as shown is coated the to FBG fiber toRH, measure RH,in asFigure shown1.in Figure 1.
FBG
Figure 1. Structure of a coated FBG sensor.
Figure 1. Structure of a coated FBG sensor.
When the RH and salinity variation changes, the volume expansion or volume contraction of When the RH and salinity the strain volume or volume contraction the sensitive polymer coating willvariation introducechanges, mechanical on expansion the FBG fiber, and therefore induceof the sensitive polymer coating will introduce mechanical strain on the FBG fiber, and therefore FBG central wavelength shift [7]. The measurement of the hydrogen concentration is based on the induce FBG central wavelength shift The measurement of the hydrogen concentration is based same principle, for which the sensor is [7]. covered with Pd. The accumulating central wavelength shift on the same principle, for which the sensor is covered with Pd. The accumulating central associated with multi-parameter (strain and temperature) measurement of FBG can be analytically wavelength shift associated calculated from Equation (1). with multi-parameter (strain and temperature) measurement of FBG can be analytically calculated from Equation (1). 2 2 2 a2 )·Y ∆λ B C · (α M − α) + (α + ξ ) · ∆T (1) (1 − Pe ) · β · a2 Y(b+(−2ba2 −)·aY22C)·Y · ∆M + (1 − Pe ) · a2 Y(b+(− 2 2 22 λ B = b − a )· Y F (b a ) Y F(b a ) YCC B CC 1 Pe 2 M 1 P T (1) e M a YF (b2 a 2 ) YC a 2Y (b2 a 2 ) Y Where λ BB is the central wavelength of the FBG; Pe is Fthe effectiveC photo-elastic constant; α M is the thermal β is an average of the target parameter; ∆M B is thecoefficient; central wavelength of theexpansion FBG; Pe coefficient is the effective photo-elastic constant; Whereexpansion isthe is normalized change of the target parameter; α and ξ are the thermal expansion coefficient and the thermal expansion coefficient; is an average expansion coefficient of the target M the thermo-optic coefficient of the single mode fiber, respectively; a and b are the cladding and total parameter; M is the normalized change of the target parameter; and are the thermal sensor diameters, respectively; YC and YF are the Young’s moduli of the coating materials and the expansion coefficient and the thermo-optic coefficient of the single mode fiber, respectively; a and silica fiber; and ∆T is the temperature change. For general silica fiber, Pe ≈ 0.22, α ≈ 5.5 × 10−7 ◦ C−1 , b are the cladding and total sensor diameters, respectively; YC and YF are the Young’s moduli of ξ ≈ 6.67 × 10−6 ◦ C−1 [8–10]. the coating materials and the silica fiber; and T is the temperature change. For general silica The desired parameter ∆M can be calculated by ∆λ B and ∆T. ∆λ B can be read by the central fiber, Pe ≈ 0.22, ≈ 5.5 × 10−7 °C−1, ≈ 6.67 × 10−6 °C−1 [8–10]. wavelength demodulation system. As FBG is inherently temperature sensitive, ∆T can be obtained M can be calculated by B and T . B can be read by the central The desired parameter and compensated by an FBG temperature sensor without any sensitive material coating. This is wavelength demodulation As FBG isofinherently temperature sensitive, be obtained T can also the typical solution forsystem. the separation strain and temperature cross-effect in mechanical and compensated by an FBG temperature sensor without any sensitive material coating. This is also parameter measurements. the typical solution for the separation of strain and temperature cross-effect in mechanical parameter 2.2. Interrogation System measurements. A schematic of the multi-parameter sensing network interrogation system for identical UFBGs is 2.2. Interrogation System illustrated in Figure 2. Field Programmable Gate Array (FPGA) takes the role of crucial scheduler of the central wavelength system, which includes three functions: (1) optical controller A schematic of thereadout multi-parameter sensing network interrogation system forpulse identical UFBGs for switching SOA1 and2.SOA2 (2) data acquiring controller for trigging digital is illustrated in Figure Fieldon/off, Programmable Gate Array (FPGA) takes the rolethe of analog crucial to scheduler of the central wavelength readout system, which includes three functions: (1) optical pulse controller for switching SOA1 and SOA2 on/off, (2) data acquiring controller for trigging the analog to digital (A/D) to sample the photoelectric conversing signals at a given time, and (3) communication with the personal computer (PC) for working environment configuration and data transition [11].
Sensors 2017, 17, 1509
3 of 10
(A/D) to sample the photoelectric conversing signals at a given time, and (3) communication with the personal computer Sensors 2017, 17, 1509 (PC) for working environment configuration and data transition [11]. 3 of 10 ASE
Erbium-doped fiber amplifier
SOA (1)
Optical circulator
uFBGs Pulse generator PC
SOA (2) photoelectric conversion
FPGA Parameters configuration
Sampling controller
A/D
Figure2.2.Schematic Schematicofofthe theinterrogation interrogationsystem. system. Figure
3. Experiment 3. Experiment 3.1.Sensor SensorFabrication Fabrication 3.1. Thefabrication fabrication system for on-line FBGduring arraysfiber during fiberhasdrawing has been The system for on-line writingwriting FBG arrays drawing been successfully successfully developed in our all of the UFBGs aresequenced continuously sequenced along the developed in our laboratory; all laboratory; of the 144 UFBGs are 144 continuously along the fiber, the peak fiber, the peak wavelengths are ~1552 nm, and the peak reflectivity is about 0.04%. wavelengths are ~1552 nm, and the peak reflectivity is about 0.04%. Withoutloss loss of generality, thefour first four are UFBGs are chosen to measureRH, hydrogen, RH, Without of generality, the first UFBGs chosen to measure hydrogen, temperature, temperature, and salinity,Prior respectively. Prior tomaterials the sensitive materials coating, thesections surface to of be the and salinity, respectively. to the sensitive coating, the surface of the ◦ sections to be coated was cleaned by anhydrous alcohol and then the ultrasonic cleaner for 10 min coated was cleaned by anhydrous alcohol and then the ultrasonic cleaner for 10 min under 30 C. ◦ C for 20 min under 30°C. Subsequently, the sensitive coatings were heat-treated at 85 for 20 min in the thermostat. Subsequently, the sensitive coatings were heat-treated at 85 in °C thermostat. For UFBG3 For the UFBG3 that was to be used for temperature sensing, no more treatment was necessary. that was to be used for temperature sensing, no more treatment was necessary. Polyimide(ZKPI-305IIE, (ZKPI-305IIE, POME Sci-tech Ltd., Beijing, solid content: Polyimide POME Sci-tech Co., Co., Ltd., Beijing, China; China; solid content: 12~13%, 12%~13%, viscosity: viscosity: 5000–6000 cp) was chosen as the polymeric layer material because of its linear and 5000–6000 cp) was chosen as the polymeric layer material because of its linear and reversible response reversible response to humidity change. UFBG2 and UFBG4 were fabricated as RH and salinity to humidity change. UFBG2 and UFBG4 were fabricated as RH and salinity sensors, respectively, with sensors, respectively, with ZKPI-305IIE polyimide as awere sensitive coating. The were first ZKPI-305IIE polyimide as a sensitive coating. The UFBGs first dip-coated withUFBGs silane coupling dip-coated silane agent: coupling agent (silane coupling agent: alcohol: deionized waterto= enhance 20%:72%:8%) agent (silanewith coupling alcohol: deionized water = 20%:72%:8%) for 10 min the for 10 min to enhance the adhesion at the polymer interface. Second, the UFBGs were ◦ C for 1inh,a adhesion at the polymer interface. Second, the UFBGs were placed in a drying cabinet at 80 placed drying cabinet 80 °C for ready 1 h, following which they readyThe for fiber the polymer coating.into The following whichatthey were for the polymer layerwere coating. grating layer was dipped fiber grating was dipped into the polyimide solution for 5–10 min and dried in a drying cabinet for the polyimide solution for 5–10 min and dried in a drying cabinet for a short thermal treatment ata short treatment at 150 °C.several This process wasthe repeated times the sake obtaining ◦ C.thermal 150 This process was repeated times for sake ofseveral obtaining thefor desired filmofthickness. the desired film of thickness. With the purpose of fabricating a uniform surface With the purpose fabricating a uniform polyimide film on the surface polyimide of the fiber,film the on ratethe of UFBG of the fiber, the rate of UFBG fiber rising-up and dropping-down were set at 600 um/min. Finally, fiber rising-up and dropping-down were set at 600 um/min. Finally, the coating profile of the sensor the checked coating profile of the sensor was checked by an optical microscope, and as shown in Figure 3a, was by an optical microscope, and as shown in Figure 3a, the film thickness of UFBG2 and the film thickness UFBG2 andrespectively UFBG4 are 14.1 UFBG4 are 14.1 µm of and 13.4 µm, [12]. μm and 13.4 μm, respectively [12]. Pd/Ni composite film is an ideal candidate for hydrogen hydrogen sensors sensors due duetotoits itsdurability, durability,fast fast Pd/Ni composite film is an ideal candidate for response,and and relatively cost. Pd/Ni composite sputtered the etched by response, relatively lowlow cost. Pd/Ni composite filmfilm was was sputtered on theon etched UFBG1UFBG1 by using using a BESTECH sputtering system. First, a 10-nm Cr film was deposited on the side-face of the a BESTECH sputtering system. First, a 10-nm Cr film was deposited on the side-face of the FBG as a FBG layer as a by basal layer by the radio-frequency sputtering Second, 100-nm Pd/Ni basal the radio-frequency (RF) sputtering(RF) process. Second,process. a 100-nm Pd/Niacomposite film composite film was sputtered on the etched FBG by a co-sputtering process. Under 0.5 Pa sputtering was sputtered on the etched FBG by a co-sputtering process. Under 0.5 Pa sputtering pressure of Ar, pressure of Ar, the deposition for Pd targets 100 and 50which W, respectively, the deposition power for Pd andpower Ni targets areand 100Ni and 50 W,are respectively, correspondswhich to a corresponds to a deposition rate of 0.14 and 0.01 nm/s, respectively. With this sputtering process, the deposition rate of 0.14 and 0.01 nm/s, respectively. With this sputtering process, the atomic ratio of atomic of Pd Ni iscomposite about 91:9 Pd/Ni composite film. Thesensitive thickness the Pd and Niratio is about 91:9 and in Pd/Ni film.inThe thickness of the hydrogenfilmofwas hydrogensensitive film was monitored by the quartz crystal method, which could ensure the monitored by the quartz crystal method, which could ensure the thickness of the hydrogen-sensitive thickness of the hydrogen-sensitive film as shown in Figure 3b [13,14]. film as shown in Figure 3b [13,14].
Sensors 2017, 17, 1509 Sensors 2017, 17, 1509 Sensors 2017, 17, 1509 Sensors 2017, 17, 1509
4 of 10
(a)
4 of 10 4 of 10 4 of 10
(b)
(a) (b) Figure 3. (a) (b) Pd/Ni-covered UFBG. (a)Polyimide-covered UFBG;(b) Figure 3. (a) Polyimide-covered UFBG; (b) Pd/Ni-covered UFBG.
Figure 3. (a) Polyimide-covered (b) Pd/Ni-covered 3. (a) Polyimide-covered UFBG;UFBG; (b) Pd/Ni-covered UFBG.UFBG. 3.2. ExperimentalFigure Results
3.2. Experimental Results The experiment and interrogation system for multi-parameter measurement system is shown in 3.2. Experimental Results 3.2. Experimental Results Figure 2. During the initialization stagesystem of thefor interrogation system, the delay timer forisSOA1 and The experiment and interrogation multi-parameter measurement system shown in The experiment and interrogation system for multi-parameter measurement system is shown in The experiment and interrogation system for multi-parameter measurement system is shown in SOA2 was stepped 1 ns in every scanning period. Then, the local maximum of the reflected optical Figure 2. During the initialization stage of the interrogation system, the delay timer for SOA1 and Figure 2. During the initialization stagelinear of theimage system, the delay timer for SOA1 and and Figure 2. During the initialization stage ofinterrogation the interrogation system, the and delay forpositions SOA1 pulse from theinInGaAs detector calculated, thetimer space of SOA2 spectrum was stepped 1 ns every scanning period. Then, was the local maximum of the reflected optical SOA2SOA2 was stepped 1 ns in every scanning period. Then, the local maximum of the reflected optical was stepped 1 ns in every scanning period. Then, the local maximum of the reflected optical every were determined by Equation (1). detector was calculated, and the space positions of pulse UFBG spectrum from the InGaAs linear image pulsepulse spectrum from the InGaAs linear image detector was positions of ofevery spectrum from the InGaAs linear image detector wascalculated, calculated,and andthe the space space positions Setting the delay timer equal to 30 ns, every UFBG were determined by Equation (1). 50 ns, 70 ns, and 90 ns, respectively, the reflected everyUFBG UFBG were determined by Equation (1). were determined by Equation (1). spectrum of the UFBG1, UFBG3, andns,UFBG4 All 144 Setting delayUFBG2, timer equal to 30 50 ns, were 70 ns,obtained and 90 accordingly. ns, respectively, the reflected reflected Setting the delay timer equal to 30 ns, 50 ns, ns, and 90 ns, respectively, the reflected Setting the delay timer equal to 30 ns, 50 ns, 70 ns, and 90 ns, respectively, the reflected spectrum spectrums were normalized and are shown in Figure 4. The first nine signal spectrums were spectrum of UFBG1, UFBG2, UFBG3, and UFBG4 were obtained accordingly. All 144 reflected spectrum of UFBG1, UFBG2, UFBG3, and UFBG4 wereinobtained accordingly. All 144spectrums reflected were of UFBG1, UFBG2, UFBG3, and UFBG4 were obtained accordingly. All 144 reflected amplified partially for visible details and are shown Figure 5. spectrums were normalized and are shown in Figure 4. The first nine signal spectrums were spectrums were normalized andinare shown in first Figure The first nine signal spectrumspartially were for normalized and are shown Figure 4. The nine4.signal spectrums were amplified amplified partially for visible details and are shown in Figure 5. amplified partially details are5.shown in Figure 5. visible details for andvisible are shown in and Figure
Figure 4. Reflective spectrum of low reflectivity UFBG array. Figure 4. Reflective spectrum of low reflectivity UFBG array. Figure 4. Reflective spectrum of low reflectivity UFBG array. Figure 4. Reflective spectrum of low reflectivity UFBG array.
Figure 5. Reflective spectrum of UFBG1~UFBG9. Figure 5. Reflective spectrum of UFBG1~UFBG9.
Figure 5. Reflective spectrum of UFBG1~UFBG9. Figure 5. Reflective spectrum of UFBG1~UFBG9.
Sensors Sensors 2017, 2017, 17, 17, 1509 1509 Sensors 2017, 17, 1509
55of of10 10 5 of 10
3.2.1. Temperature Measurement 3.2.1. Temperature Measurement 3.2.1. Temperature Measurement The temperate and humidity chamber model 101-0AB was used for the temperature and RH The temperate and humidity chamber model 101-0AB for temperature and RH The temperate and humidity chamber 101-0AB waswas used thethe temperature andat RH test. The adjusted temperature range of themodel chamber was limited toused 20for °C–250 °C with ±1 °C room ◦ C–250 ◦ C with ±1 ◦ C at test. The adjusted temperature range of the chamber was limited to 20 test. The adjusted temperature range of the chamber was limited to 20 °C–250 °C with ±1 °C at room temperature. UFBG3 was placed in the while the other UFBGs were kept at 25 °C. The ◦ C. The room temperature. UFBG3 placed in the while thethe other UFBGs were atis°C. 25 temperature. UFBG3ofwas in the chamber the other UFBGs were keptkept at 25 The in spectrum shifting the placed firstwas nine sensors duechamber towhile the change in chamber temperature shown spectrum shifting of the first nine sensors due to the change in the chamber temperature is shown in spectrum shifting of the first nine sensors due to the change in the chamber temperature is shown in of Figure 6. The shape of the reflected spectrum of UFBG3 remained unchanged with the increase Figure 6. The shape of the reflected spectrum of UFBG3 remained unchanged with the increase of Figure 6. The shape the reflected spectrum UFBG3 with the of temperature. The of central wavelength can beofcould beremained calculatedunchanged from Equation (1). increase As expected, temperature. The central wavelength canbe beofcould could be from (1).(1). AsAs expected, apart temperature. The central wavelength can be calculated calculated fromEquation Equation expected, apart from UFBG3, the reflected spectrum the other UFBGs remained the same. from UFBG3, thethe reflected spectrum of the other UFBGs remained the the same. apart from UFBG3, reflected spectrum of the other UFBGs remained same.
Figure6. 6.Reflective Reflectivespectrum spectrumshift shiftof ofUFBG3 UFBG3during duringtemperature temperature measurement. measurement. Figure Figure 6. Reflective spectrum shift of UFBG3 during temperature measurement.
Figure showsthe thecontinuous continuous temperature test results of UFBG3 by the setting the Figure 77 shows 120120 minmin temperature test results of UFBG3 by setting chamber Figure 7 shows the continuous 120 min temperature test results of UFBG3 by setting the ◦ C,°C, ◦ C,°C, ◦ C,°C, ◦ C, chamber temperature °C, 6575 °C,◦ Cand 75 °C respectively (75% RH). It canthat be temperature at 25 ◦ C, at 35 25 45 35 5545 6555 and respectively (75% RH). It can be seen chamber temperature at 25 °C, 35 °C, 45 °C, 55 °C, 65 °C, and 75 °C respectively (75% RH). It can be ◦ seen that the temperature response curve was homogeneous and stable. The jitter at 35 °C can be the temperature response curve was homogeneous and stable. The jitter at 35 C can be attributed to seen temperature response curve wasofhomogeneous and stable. The jitter at 35 °C can be attributed to the first regulation the chamber temperature. thethat firstthe regulation fluctuation offluctuation the chamber temperature. attributed to the first regulation fluctuation of the chamber temperature.
Figure 7. Temperature measurement of UFBG3. Figure 7. Temperature measurement of UFBG3. Figure 7. Temperature measurement of UFBG3.
Figure 88 shows shows the the linear linear fitting fitting results results of of temperature temperature with with the the central central wavelength wavelength shift shift of of Figure UFBG3. fitting function 0.01074x + 1552.677, so the temperature sensitivitywas wasabout about FigureThe 8 shows linearwas fitting ofx temperature the central wavelength shift of x = 0.01074 1552.677 UFBG3. The fittingthe function wasf (fx)results , so with the temperature sensitivity ◦ C. 11 pm/ UFBG3. The fitting function was f x 0.01074 x 1552.677 , so the temperature sensitivity was about 11 pm/°C.
11 pm/°C.
Sensors 2017, 17, 1509 Sensors 2017, 17, 1509 Sensors 2017, 17, 1509 Sensors 2017, 17, 1509
6ofof 1010 66of 10 6 of 10
Figure 8. 8. Central wavelength vs.vs. temperature. Figure Central wavelength temperature. Figure 8. Central wavelength vs. temperature. Figure 8. Central wavelength vs. temperature.
3.2.2. 3.2.2.RH RHMeasurement Measurement 3.2.2. RH Measurement 3.2.2. RH Measurement UFBG2 UFBG2was wasplaced placedin thetemperature temperatureand andhumidity humiditychamber chamberwhile whilethe thetemperature temperaturewas was UFBG2 was placed ininthe the temperature and humidity chamber while the temperature was ◦ C. UFBG2 was placed in the and humidity chamber while the temperature was maintained at 30 Keeping the temperature constant and changing the relative humidity from maintained at 30 °C. Keeping the temperature constant and changing the relative humidity from maintained at 30 °C. Keeping the temperature constant and changing the relative humidity from maintained at 30 °C. Keeping the temperature constant and changing the relative humidity from 30% RH to 40% RH, 50% RH, 60% RH, 70% RH, 80% RH, and 90% RH, respectively, the shift of the 30%RH RHtoto40% 40%RH, RH,50% 50%RH, RH,60% 60%RH, RH,70% 70%RH, RH,80% 80%RH, RH,and and90% 90%RH, RH,respectively, respectively,the theshift shiftofofthe the 30% 30% RH to 40% RH, 50% RH, 60% RH, 70% RH, 80% RH, and 90% RH, respectively, the shift of the central wavelength was recorded over 600 min continually with an RH adjustment at about every central wavelength was recorded over 600 min continually with an RH adjustment at about every central wavelength was recorded over 600 min continually with an RH adjustment at about every central wasFigure recorded overresults 600 min continually RH adjustment at conditions about every 100 min aswavelength shown in Figure 9. The verify the factthe that the chamber presents unstable 100 min asshown shown The verify the fact factwith thatan the chamber presents unstable 100 min as ininFigure 9.9.results The results verify that the chamber presents unstable 100 min as shown in Figure 9. The results verify the fact that the chamber presents unstable at low or high RH settings while the temperature is fixed. conditionsatatlow lowororhigh highRH RHsettings settingswhile whilethe thetemperature temperatureisisfixed. fixed. conditions conditions at low or high RH settings while the temperature is fixed.
Figure9.9.UFBG2 UFBG2central centralwavelength wavelengthevolution evolutionwith withthe thechange changeofofrelative relativehumidity humidity Figure .. Figure 9. 9. UFBG2 central wavelength evolution with thethe change of of relative humidity. Figure UFBG2 central wavelength evolution with change relative humidity.
Figure10 10shows showsthe thefitting fittingcurve curveofofthe thecentral centralwavelength wavelengthwith withRH. RH.The Thefitting fittingfunction functionwas was Figure
Figure shows the fitting curve central wavelength with RH. The fitting function was Figure 1010 shows the fitting curve of of thethe central wavelength with RH. The fitting function was 0.001258 1552.917 and theRH RH sensitivity wasabout about 1.26pm/%RH. pm/%RH. 0.001258 f f x x x x1552.917 , ,and the sensitivity was 1.26 0.001258 x 1552.917 , and and the RH sensitivity waswas about 1.261.26 pm/%RH. 0.001258x f ( xf ) x= + 1552.917, the RH sensitivity about pm/%RH.
Figure10. 10.Central Centralwavelength wavelengthvs. vs.humidity. humidity. Figure Figure Central wavelength humidity. Figure 10.10. Central wavelength vs.vs. humidity.
3.2.3.Hydrogen HydrogenConcentration ConcentrationMeasurement Measurement 3.2.3. 3.2.3. Hydrogen Concentration Measurement
Sensors 2017, 17, 1509
7 of 10
Sensors 2017, 17, 17, 1509 Sensors 2017, 1509
7 of7 10 of 10
3.2.3. Hydrogen Concentration Measurement
UFBG1 was place a chamber with varying concentrations of of hydrogen, while the other UFBG1 was place a chamber with varying concentrations hydrogen, while the other UFBG1 was place inina in chamber with varying concentrations of hydrogen, while the other ambient ambient parameters of the UFBGs were fixed. Figure 11 shows the central wavelength shift of of ambient parameters ofwere the fixed. UFBGsFigure were11 fixed. Figure 11 shows the central wavelength shift parameters of the UFBGs shows the central wavelength shift of UFBG1 with the UFBG1 with thethe of of hydrogen 1%, 1.5%, and 2%, respectively. UFBG1 with concentrations hydrogen at respectively. 1%, 1.5%, and 2%, respectively. concentrations ofconcentrations hydrogen at 1%, 1.5%, and at 2%,
Figure 11.11. Hydrogen concentration measurement. Figure Hydrogen concentration measurement. Figure 11. Hydrogen concentration measurement.
The mean value of of samples at at 7~97~9 min at at thethe same hydrogen concentration was calculated. The mean value samples min same hydrogen concentration was calculated. The mean value of samples at 7~9 min atthe same hydrogen concentration was calculated. f x 0.115376 x 1552.258 Figure 12 shows the fitting function to be . When hydrogen concentrations Figure 12 shows the fitting function to be f x 0.115376 x 1552.258 . When hydrogen concentrations Figure 12 shows the fitting function to be f ( x ) = 0.115376x + 1552.258. When hydrogen concentrations were from 1%1% to to 1.5% and from 1.5% to to 2%, thethe wavelength shifts of of thethe FBG areare 40 40 pmpm and 75 75 pm. were from 1.5% and from 1.5% 2%, wavelength shifts FBG and pm. were from 1% to 1.5% and from 1.5% to 2%, the wavelength shifts of the FBG are 40 pm and 75 pm.
Figure 12. Central wavelength vs. hydrogen concentration. Figure Central wavelength hydrogen concentration. Figure 12.12. Central wavelength vs.vs. hydrogen concentration.
3.2.4. Salinity Measurement 3.2.4. Salinity Measurement 3.2.4. Salinity Measurement For salinity measurement, the UFBG4 was placed into aacontainer with 30 ml deionized water, For salinity measurement, the UFBG4 was placed into a container with deionized water, For salinity measurement, the UFBG4 was placed into container with 3030 mlml deionized water, which means that the salinity was 0 mol/L. Then, 10.8 g NaCl was added to the water, bringing which means that the salinity was 0 mol/L. Then, 10.8 g NaCl was added to the water, bringing which means that the salinity was 0 mol/L. Then, 10.8 g NaCl was added to the water, bringing the thethe concentration of saturated salt solution to 6.154 mol/L, and the central wavelength was shifted down concentration saturated salt solution 6.154 mol/L, and the central wavelength was shifted down concentration of of saturated salt solution to to 6.154 mol/L, and the central wavelength was shifted down to 1552.038 nm. Following this, deionized water was injected into the container to obtain a salinity to 1552.038 nm. Following this, deionized water was injected into the container to obtain a salinity to 1552.038 nm. Following this, deionized water was injected into the container to obtain a salinity of of of 44 mol/L and 2 mol/L, and the drift of the central wavelength was recorded, respectively. The linear 4 mol/Land and2 2mol/L, mol/L,and andthe thedrift driftofofthe thecentral centralwavelength wavelengthwas wasrecorded, recorded, respectively. The linear mol/L respectively. The linear fitting curve of the central wavelength with salinity changes is shown in Figure 13. It can be be fitting curve of the central wavelength with salinity changes is shown in Figure 13. It can fitting curve of the central wavelength with salinity changes is shown in Figure 13. It can be expressed −1. −1 f xf x 0.017 x 1552.142 expressed as , with the salinity sensitivity about ~−17 pm/mol·L − 1 0.017 x 1552.142 expressed as , with the salinity sensitivity about ~−17 pm/mol·L . as f ( x ) = −0.017x + 1552.142, with the salinity sensitivity about ~−17 pm/mol·L .
Sensors 2017, 17, 17, 1509 Sensors 2017, 1509
8 of8 10 of 10
Figure Central wavelength salinity. Figure 13.13.Central wavelength vs. vs. salinity.
Cross-Sensitivity Investigations Thermal Compensation 3.3.3.3. Cross-Sensitivity Investigations andand Thermal Compensation expected,the thecross-sensitivity cross-sensitivitybetween between the the temperature including RH, AsAs expected, temperatureand andother otherparameters parameters including salinity, and clearthat thatwithout without temperature RH, salinity, andhydrogen hydrogenwas wasobserved. observed. From From Equation Equation (1), (1), itit isisclear temperature compensation, a correct multi-parameter measurement is not possible. Generally, apart from compensation, a correct multi-parameter measurement is not possible. Generally, apart from thethe temperature effect, the cross-sensitivity among the other parameters measures can be negligible. temperature effect, the cross-sensitivity among the other parameters measures can be negligible. Since FBGs are intrinsically temperature sensors, it is to natural to solve thistaking issuethe bymeasured taking the Since FBGs are intrinsically temperature sensors, it is natural solve this issue by measured temperature near the test points as the role of temperature compensation in this UFBG temperature near the test points as the role of temperature compensation in this UFBG sensing array. sensing Basedarray. on Equation (1), the central wavelength is influenced by the temperature change ∆T and T Based on Equation (1), parameter the centralchange wavelength the(fiber temperature the strain induced by the target ∆M of is theinfluenced transducerby layer coating). change The stress M of the transducer layer (fiber coating). and the strain induced by the target parameter change effects induced by temperature change are described as α M − α. Then the superposition of the central The stressshift effects induced by change are described wavelength λ recorded by temperature the interrogation system is given as:as M . Then the superposition t
of the central wavelength shift t recorded by the interrogation system is given as: (b2 − a2 )·Y
λt = λ B · (1 − Pe ) · β · a2 Y +(b2(−ba2 2C)·Ya 2 )· YM + Ct C Mt t B 1 Pe F 2 2 2 h i a Y ( b a ) Y F C (b2 − a2 )·Y (2) λ B · (1 − Pe ) · a2 Y +(b2 −a2C2)·Y 2· (α M − α) + (α + ξ ) · Tt + F (b Ca ) YC n h i o T B 1 Pe 2 M t 2 2 2 2 (2) a ·YF 2((bb 2−a2 a)·2Y2)C YC · M0 − λ B · (1 −Pe ) · 2 (b − a2 )·Y2C λ0 − λ B · (1 − Pe ) · β · (α M − α) + (α + ξ ) · T0 a YF +(b − a )·YC a YF +(b − a )·YC (b 2 a 2 ) YC (b 2 a 2 ) YC M 0 B 1 Pe 2 M T0 0 B 1 Pe 2 2 2 Where λ0 , M0 and T0 aare target parameter, and (b initial a ) YC value of the central YF the a YF (wavelength, b 2 a 2 ) YC
temperature, by theofinterrogation system; Tt can be calculated 0 , M 0 and λTt0 can Where respectively. arebe theobtained initial value the central wavelength, target parameter,by and UFBG3, the temperature sensor. temperature, respectively. t can be obtained by the interrogation system; Tt can be calculated by Without loss of generality, the measurement of RH and temperature is taken as example. Sweeping UFBG3, the temperature sensor. the ambient temperature from 25 ◦ C to 75 ◦ C with a step of 10 ◦ C and RH from 30% to 90% with a step Without loss of generality, the measurement of RH and temperature is taken as example. of 10%, the central wavelength of UFBG2 was recorded from the interrogation system, as shown in Sweeping the ambient temperature from 25 °C to 75 °C with a step of 10 °C and RH from 30% to 90% Figure 14 after least-squares curve-fitting. with a step of 10%, the central wavelength of UFBG2 was recorded from the interrogation system, as shown in Figure 14 after least-squares curve-fitting.
Sensors 2017, 1509 Sensors 2017, 17,17, 1509
9 9ofof1010
Figure14. 14.Central Centralwavelength wavelengthshift shiftwith withboth bothRH RHand andtemperature temperaturechange. change. Figure
4. Conclusions 4. Conclusions An ultra-weak FBG array has been proved to be one of the most promising solutions for An ultra-weak FBG array has been proved to be one of the most promising solutions for huge-capacity fiber sensing networks. In this work, we proposed and demonstrated a sensing array huge-capacity fiber sensing networks. In this work, we proposed and demonstrated a sensing array for multi-parameter measurements based on a UFBG with sensitive material coating. The central for multi-parameter measurements based on a UFBG with sensitive material coating. The central wavelength readout system employed uses two SOAs to separate the optical pulse from the sensors wavelength readout system employed uses two SOAs to separate the optical pulse from the sensors while rejecting unwanted signals caused by noise, crosstalk, and interference. This setup provides while rejecting unwanted signals caused by noise, crosstalk, and interference. This setup provides the advantage of testing various parameters in a single fiber without limiting temperature and strain the advantage of testing various parameters in a single fiber without limiting temperature and strain measurement, and allows for a larger scale because of the strong multiplexing capability of UFBG. measurement, and allows for a larger scale because of the strong multiplexing capability of UFBG. This method was performed on a 144 UFBG sensors array with the reflectivity of UFBG ~0.04% This method was performed on a 144 UFBG sensors array with the reflectivity of UFBG ~0.04% for the four target parameters: hydrogen, RH, temperature, and salinity. The performance of for the four target parameters: hydrogen, RH, temperature, and salinity. The performance of multi-parameter sensing array was almost equal to result of the single-point FBG with general multi-parameter sensing array was almost equal to result of the single-point FBG with general reflectivity. In order to solve the cross-sensitivity of multi-parameter sensors, with the temperature reflectivity. In order to solve the cross-sensitivity of multi-parameter sensors, with the temperature taking the primary effect to other parameters, a uniform solution was customized to divide the taking the primary effect to other parameters, a uniform solution was customized to divide the central central wavelength shift caused by target parameters from that caused by the temperature effect. wavelength shift caused by target parameters from that caused by the temperature effect. Based on the Based on the report of the multiplexing capacity of ultra-weak FBGs, it is expected that this scheme report of the multiplexing capacity of ultra-weak FBGs, it is expected that this scheme of UFBG-based of UFBG-based sensing network can potentially multiplex thousands of multi-parameter sensors in sensing network can potentially multiplex thousands of multi-parameter sensors in single fiber, which single fiber, which shows promising prospects for the future. shows promising prospects for the future. Acknowledgments:This Thiswork workisisfinically finicallysupported supportedbybythe theProject ProjectofofNational NationalNatural NaturalScience ScienceFoundation Foundationofof Acknowledgments: China,NSFC NSFC (Number: 51402228, 61290311, 61475121, 61575151), National Natural Science China, (Number: 51402228, 61290311, 61475121, 61575151), Project ofProject NationalofNatural Science Foundation ofFoundation Hubei Provincial Government (Number: 2014CFB260), the Creativethe theCreative Creativethe Group Project of Hubei of Hubei Provincial Government (Number: 2014CFB260), Creative Group Project Provincial Science Foundation (Project Number: of Hubei Natural Provincial Natural Science Foundation (Project2015CFA016). Number: 2015CFA016). Author Contributions: Wei Bai, Minghong Yang, and Chenyuan Hu conceived and designed the experiments; Author Contributions: Wei Bai, Minghong Yang, and Chenyuan Hu conceived and designed the experiments; Jixiang Dai, Xuexiang Zhong and Shuai Huang performed the experiments; Xuexiang Zhong and Gaopeng Wang Jixiang Dai, Xuexiang Zhong Shuai Huang performed the experiments; Xuexiang Zhong and Gaopeng analyzed the data; Wei Bai wroteand the paper. Wang analyzed the data; Wei Bai wrote the paper. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest.
References References
1. Yang, M.; Bai, W.; Guo, H.; Wen, H.; Jiang, D. Huge Capacity Fiber-Optic Sensing Network Based on 1. Ultra-Weak Yang, M.; Draw Bai, W.; Guo, H.; Wen, H.; Jiang, Huge Capacity Fiber-Optic Sensing Network Based on Tower Gratings. Photonic Sens.D. 2016, 6, 26–41. [CrossRef] Ultra-Weak Sens. 2016, 6, 26–41. 2. Hu, C.Y.; Wen,Draw H.Q.;Tower Bai, W.Gratings. A NovelPhotonic Interrogation System for Large Scale Sensing Network with Identical 2. Ultra-Weak Hu, C.Y.; Fiber Wen, Bragg H.Q.; Gratings. Bai, W. A Novel Interrogation for Large Scale Sensing Network with J. Lightwave Technol. 2014,System 32, 1406–1411. [CrossRef] Identical Ultra-Weak Fiber Bragg Gratings. J. Lightwave Technol. 2014, 32, 1406–1411. 3. Wang, Y.M.; Gong, J.M.; Wang, D.Y.; Dong, B.; Bi, W.; Wang, A.B. A quasi-distributed sensing network with 3. time-division-multiplexed Wang, Y.M.; Gong, J.M.; fiber Wang, D.Y.;gratings. Dong, B.; Bi, Photonic W.; Wang, A.B. A quasi-distributed sensing network Bragg IEEE Technol. 2011, 23, 70–72. [CrossRef] with time-division-multiplexed fiber Bragg gratings. IEEE Photonic Technol. 2011, 23, 70–72.
Sensors 2017, 17, 1509
4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
10 of 10
Wang, Y.M.; Gong, J.M.; Wang, D.Y.; Shillig, T.J.; Wang, A. A large Serial time-division multiplexed fiber Bragg grating sensor network. J. Lightwave Technol. 2012, 30, 2751–2756. [CrossRef] Wang, S.; Liao, Y.; Yang, H.; Wang, X.; Wang, J. Modeling seawater salinity and temperature sensing based on directional coupler assembled by polyimide-coated. Appl. Opt. 2015, 54, 10283–10289. [CrossRef] [PubMed] Luo, D.; Ma, J.X.; Ibrahim, Z.; Ismail, Z. Etched FBG coated with polyimide for simultaneous detection the salinity. Opt. Commun. 2017, 392, 218–222. [CrossRef] Liu, X.; Zhang, X.; Cong, J.; Xu, J.; Chen, K. Demonstration of etched cladding fiber Bragg grating-based sensors with hydrogel coating. Sens. Actuators 2003, 96, 468–472. [CrossRef] Dai, J.X.; Yang, M.H.; Chen, Y.; Cao, K.; Liao, H.; Zhang, P. Side-polished fiber Bragg grating hydrogen sensor with WO3 -Pd composite film as sensing materials. Opt. Express 2011, 19, 6141–6148. [CrossRef] [PubMed] Wanasekara, N.; Chalivendra, V.; Calvert, P. Sub-micron scale mechanical properties of polypropylene fibers exposed to ultraviolet and thermal degradation. Polym. Degrad. Stab. 2011, 96, 432–437. [CrossRef] Sutapun, B.; Tabib-Azar, M.; Kazemi, A. Pd-coated elastooptic fiber optic Bragg grating sensors for multiplexed hydrogen sensing. Sens. Actuators B Chem. 1999, 60, 27–34. [CrossRef] Zhang, Y.J.; Xie, X.P.; Xu, H.B. Distributed Temperature Sensor System Based on Weak Reflection Fiber Gratings Combined with WDM and OTDR. J. Opt. Electr. Eng. 2012, 39, 69–74. Alwis, L.; Sun, T.; Grattan, K.V. Analysis of Polyimide-Coated Optical Fiber Long-Period Grating-Based Relative Humidity Sensor. IEEE Sens. J. 2013, 13, 767–771. [CrossRef] Dai, J.; Yang, M.; Yu, X.; Cao, K.; Liao, J. Greatly etched fiber Bragg grating hydrogen sensor with Pd/Ni composite film as sensing material. Sens. Actuators 2012, 174, 253–257. [CrossRef] Yang, M.H.; Dai, J.X. Fiber Optic Hydrogen Sensors: A Review. Photonic Sens. 2014, 4, 300–324. [CrossRef] © 2017 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 (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Ultra-Weak Fiber Bragg Grating Sensing Network Coated with Sensitive Material for Multi-Parameter Measurements Wei Bai 1 , Minghong Yang 1,2, *, Chenyuan Hu 1, *, Jixiang Dai 1 , Xuexiang Zhong 1 , Shuai Huang 1 and Gaopeng Wang 1 1
2
*
National Engineering Laboratory for Fiber Optic Sensing Technologies, Wuhan University of Technology, Wuhan 430070, China; [email protected] (W.B.); [email protected] (J.D.); [email protected] (X.Z.); [email protected] (S.H.); [email protected] (G.W.) Key Laboratory of Fiber Optic Sensing Technology and Information Processing, Ministry of Education, Wuhan, 430070, China Correspondence: [email protected] (M.Y.); [email protected] or [email protected] (C.H.); Tel.: +86-139-8624-5199 (M.Y.); +86-135-4500-0634 (C.H.)
Received: 11 May 2017; Accepted: 20 June 2017; Published: 26 June 2017
Abstract: A multi-parameter measurement system based on ultra-weak fiber Bragg grating (UFBG) array with sensitive material was proposed and experimentally demonstrated. The UFBG array interrogation principle is time division multiplex technology with two semiconductor optical amplifiers as timing units. Experimental results showed that the performance of the proposed UFBG system is almost equal to that of traditional FBG, while the UFBG array system has obvious superiority with potential multiplexing ability for multi-point and multi-parameter measurement. The system experimented on a 144 UFBG array with the reflectivity of UFBG ~0.04% for the four target parameters: hydrogen, humidity, temperature and salinity. Moreover, a uniform solution was customized to divide the cross-sensitivity between temperature and other target parameters. It is expected that this scheme will be capable of handling thousands of multi-parameter sensors in a single fiber. Keywords: fiber optics sensors; relativity; multi-parameter
1. Introduction The development trends towards the next generation of optical fiber sensing networks aim for huge capacity, long distance, and high precision. The huge capacity of optical fiber sensing networks includes two aspects of content, i.e. more points and more parameters of measurement. Jiang et al. proposed an on-line writing no-weld UFBG array through fiber drawing tower [1]. Then, Hu et al. demonstrated a single optical fiber network with over 1000 UFBG sensors by combining two semiconductor optical amplifiers (SOAs) and a time-division multiplexing (TDM) technique. Multi-point measurement based on UFBG array has made great improvement, while the multi-parameter measurement is still challenging [2,3]. In this paper, a serial TDM sensing network for multi-parameter measurement based on UFBGs coated with sensitive materials is proposed and experimentally demonstrated [4]. A UFBG array sensing system for hydrogen, relative humidity (RH), temperature and salinity measurements in a single fiber by coating different sensitive materials on different UFBGs is realized [5,6]. It can be concluded that thousands of UFBG sensors multiplexed in series with multi-parameter measurements could be possible, which shows promising applications for the future.
Sensors 2017, 17, 1509; doi:10.3390/s17071509
www.mdpi.com/journal/sensors
Sensors 2017, 17, 1509
2 of 10
2017, 17, 1509 2.Sensors Principle
2 of 10
2.1. 2.1.Sensing Sensingwith withCoated CoatedFBG FBG When Whenthe theFBG FBGisissubjected subjectedtotoexternal externalfield fieldeffect effect(such (suchasasstress, stress,temperature), temperature),the thespatial spatial period periodofofgrating gratingwill willchange, change,leading leadingtotothe thereflection reflectionorortransmission transmissiongrating gratingcentral centralwavelength wavelength drift. drift.Since Sincethey theyare areimmune immunetotochemical chemicalparameters, parameters,such suchas ashumidity, humidity,hydrogen, hydrogen,salinity, salinity,etc. etc.,the the FBG-based sensors areare fabricated by by coating corresponding sensitive materials on the FBG-basedmulti-parameter multi-parameter sensors fabricated coating corresponding sensitive materials on surface of theof fiber covering the Bragg instance, an RH-sensitive polymer the surface thecladding fiber cladding covering thegrating Bragg section. grating For section. For instance, an RH-sensitive ispolymer coated on the FBGonfiber measure as shown is coated the to FBG fiber toRH, measure RH,in asFigure shown1.in Figure 1.
FBG
Figure 1. Structure of a coated FBG sensor.
Figure 1. Structure of a coated FBG sensor.
When the RH and salinity variation changes, the volume expansion or volume contraction of When the RH and salinity the strain volume or volume contraction the sensitive polymer coating willvariation introducechanges, mechanical on expansion the FBG fiber, and therefore induceof the sensitive polymer coating will introduce mechanical strain on the FBG fiber, and therefore FBG central wavelength shift [7]. The measurement of the hydrogen concentration is based on the induce FBG central wavelength shift The measurement of the hydrogen concentration is based same principle, for which the sensor is [7]. covered with Pd. The accumulating central wavelength shift on the same principle, for which the sensor is covered with Pd. The accumulating central associated with multi-parameter (strain and temperature) measurement of FBG can be analytically wavelength shift associated calculated from Equation (1). with multi-parameter (strain and temperature) measurement of FBG can be analytically calculated from Equation (1). 2 2 2 a2 )·Y ∆λ B C · (α M − α) + (α + ξ ) · ∆T (1) (1 − Pe ) · β · a2 Y(b+(−2ba2 −)·aY22C)·Y · ∆M + (1 − Pe ) · a2 Y(b+(− 2 2 22 λ B = b − a )· Y F (b a ) Y F(b a ) YCC B CC 1 Pe 2 M 1 P T (1) e M a YF (b2 a 2 ) YC a 2Y (b2 a 2 ) Y Where λ BB is the central wavelength of the FBG; Pe is Fthe effectiveC photo-elastic constant; α M is the thermal β is an average of the target parameter; ∆M B is thecoefficient; central wavelength of theexpansion FBG; Pe coefficient is the effective photo-elastic constant; Whereexpansion isthe is normalized change of the target parameter; α and ξ are the thermal expansion coefficient and the thermal expansion coefficient; is an average expansion coefficient of the target M the thermo-optic coefficient of the single mode fiber, respectively; a and b are the cladding and total parameter; M is the normalized change of the target parameter; and are the thermal sensor diameters, respectively; YC and YF are the Young’s moduli of the coating materials and the expansion coefficient and the thermo-optic coefficient of the single mode fiber, respectively; a and silica fiber; and ∆T is the temperature change. For general silica fiber, Pe ≈ 0.22, α ≈ 5.5 × 10−7 ◦ C−1 , b are the cladding and total sensor diameters, respectively; YC and YF are the Young’s moduli of ξ ≈ 6.67 × 10−6 ◦ C−1 [8–10]. the coating materials and the silica fiber; and T is the temperature change. For general silica The desired parameter ∆M can be calculated by ∆λ B and ∆T. ∆λ B can be read by the central fiber, Pe ≈ 0.22, ≈ 5.5 × 10−7 °C−1, ≈ 6.67 × 10−6 °C−1 [8–10]. wavelength demodulation system. As FBG is inherently temperature sensitive, ∆T can be obtained M can be calculated by B and T . B can be read by the central The desired parameter and compensated by an FBG temperature sensor without any sensitive material coating. This is wavelength demodulation As FBG isofinherently temperature sensitive, be obtained T can also the typical solution forsystem. the separation strain and temperature cross-effect in mechanical and compensated by an FBG temperature sensor without any sensitive material coating. This is also parameter measurements. the typical solution for the separation of strain and temperature cross-effect in mechanical parameter 2.2. Interrogation System measurements. A schematic of the multi-parameter sensing network interrogation system for identical UFBGs is 2.2. Interrogation System illustrated in Figure 2. Field Programmable Gate Array (FPGA) takes the role of crucial scheduler of the central wavelength system, which includes three functions: (1) optical controller A schematic of thereadout multi-parameter sensing network interrogation system forpulse identical UFBGs for switching SOA1 and2.SOA2 (2) data acquiring controller for trigging digital is illustrated in Figure Fieldon/off, Programmable Gate Array (FPGA) takes the rolethe of analog crucial to scheduler of the central wavelength readout system, which includes three functions: (1) optical pulse controller for switching SOA1 and SOA2 on/off, (2) data acquiring controller for trigging the analog to digital (A/D) to sample the photoelectric conversing signals at a given time, and (3) communication with the personal computer (PC) for working environment configuration and data transition [11].
Sensors 2017, 17, 1509
3 of 10
(A/D) to sample the photoelectric conversing signals at a given time, and (3) communication with the personal computer Sensors 2017, 17, 1509 (PC) for working environment configuration and data transition [11]. 3 of 10 ASE
Erbium-doped fiber amplifier
SOA (1)
Optical circulator
uFBGs Pulse generator PC
SOA (2) photoelectric conversion
FPGA Parameters configuration
Sampling controller
A/D
Figure2.2.Schematic Schematicofofthe theinterrogation interrogationsystem. system. Figure
3. Experiment 3. Experiment 3.1.Sensor SensorFabrication Fabrication 3.1. Thefabrication fabrication system for on-line FBGduring arraysfiber during fiberhasdrawing has been The system for on-line writingwriting FBG arrays drawing been successfully successfully developed in our all of the UFBGs aresequenced continuously sequenced along the developed in our laboratory; all laboratory; of the 144 UFBGs are 144 continuously along the fiber, the peak fiber, the peak wavelengths are ~1552 nm, and the peak reflectivity is about 0.04%. wavelengths are ~1552 nm, and the peak reflectivity is about 0.04%. Withoutloss loss of generality, thefour first four are UFBGs are chosen to measureRH, hydrogen, RH, Without of generality, the first UFBGs chosen to measure hydrogen, temperature, temperature, and salinity,Prior respectively. Prior tomaterials the sensitive materials coating, thesections surface to of be the and salinity, respectively. to the sensitive coating, the surface of the ◦ sections to be coated was cleaned by anhydrous alcohol and then the ultrasonic cleaner for 10 min coated was cleaned by anhydrous alcohol and then the ultrasonic cleaner for 10 min under 30 C. ◦ C for 20 min under 30°C. Subsequently, the sensitive coatings were heat-treated at 85 for 20 min in the thermostat. Subsequently, the sensitive coatings were heat-treated at 85 in °C thermostat. For UFBG3 For the UFBG3 that was to be used for temperature sensing, no more treatment was necessary. that was to be used for temperature sensing, no more treatment was necessary. Polyimide(ZKPI-305IIE, (ZKPI-305IIE, POME Sci-tech Ltd., Beijing, solid content: Polyimide POME Sci-tech Co., Co., Ltd., Beijing, China; China; solid content: 12~13%, 12%~13%, viscosity: viscosity: 5000–6000 cp) was chosen as the polymeric layer material because of its linear and 5000–6000 cp) was chosen as the polymeric layer material because of its linear and reversible response reversible response to humidity change. UFBG2 and UFBG4 were fabricated as RH and salinity to humidity change. UFBG2 and UFBG4 were fabricated as RH and salinity sensors, respectively, with sensors, respectively, with ZKPI-305IIE polyimide as awere sensitive coating. The were first ZKPI-305IIE polyimide as a sensitive coating. The UFBGs first dip-coated withUFBGs silane coupling dip-coated silane agent: coupling agent (silane coupling agent: alcohol: deionized waterto= enhance 20%:72%:8%) agent (silanewith coupling alcohol: deionized water = 20%:72%:8%) for 10 min the for 10 min to enhance the adhesion at the polymer interface. Second, the UFBGs were ◦ C for 1inh,a adhesion at the polymer interface. Second, the UFBGs were placed in a drying cabinet at 80 placed drying cabinet 80 °C for ready 1 h, following which they readyThe for fiber the polymer coating.into The following whichatthey were for the polymer layerwere coating. grating layer was dipped fiber grating was dipped into the polyimide solution for 5–10 min and dried in a drying cabinet for the polyimide solution for 5–10 min and dried in a drying cabinet for a short thermal treatment ata short treatment at 150 °C.several This process wasthe repeated times the sake obtaining ◦ C.thermal 150 This process was repeated times for sake ofseveral obtaining thefor desired filmofthickness. the desired film of thickness. With the purpose of fabricating a uniform surface With the purpose fabricating a uniform polyimide film on the surface polyimide of the fiber,film the on ratethe of UFBG of the fiber, the rate of UFBG fiber rising-up and dropping-down were set at 600 um/min. Finally, fiber rising-up and dropping-down were set at 600 um/min. Finally, the coating profile of the sensor the checked coating profile of the sensor was checked by an optical microscope, and as shown in Figure 3a, was by an optical microscope, and as shown in Figure 3a, the film thickness of UFBG2 and the film thickness UFBG2 andrespectively UFBG4 are 14.1 UFBG4 are 14.1 µm of and 13.4 µm, [12]. μm and 13.4 μm, respectively [12]. Pd/Ni composite film is an ideal candidate for hydrogen hydrogen sensors sensors due duetotoits itsdurability, durability,fast fast Pd/Ni composite film is an ideal candidate for response,and and relatively cost. Pd/Ni composite sputtered the etched by response, relatively lowlow cost. Pd/Ni composite filmfilm was was sputtered on theon etched UFBG1UFBG1 by using using a BESTECH sputtering system. First, a 10-nm Cr film was deposited on the side-face of the a BESTECH sputtering system. First, a 10-nm Cr film was deposited on the side-face of the FBG as a FBG layer as a by basal layer by the radio-frequency sputtering Second, 100-nm Pd/Ni basal the radio-frequency (RF) sputtering(RF) process. Second,process. a 100-nm Pd/Niacomposite film composite film was sputtered on the etched FBG by a co-sputtering process. Under 0.5 Pa sputtering was sputtered on the etched FBG by a co-sputtering process. Under 0.5 Pa sputtering pressure of Ar, pressure of Ar, the deposition for Pd targets 100 and 50which W, respectively, the deposition power for Pd andpower Ni targets areand 100Ni and 50 W,are respectively, correspondswhich to a corresponds to a deposition rate of 0.14 and 0.01 nm/s, respectively. With this sputtering process, the deposition rate of 0.14 and 0.01 nm/s, respectively. With this sputtering process, the atomic ratio of atomic of Pd Ni iscomposite about 91:9 Pd/Ni composite film. Thesensitive thickness the Pd and Niratio is about 91:9 and in Pd/Ni film.inThe thickness of the hydrogenfilmofwas hydrogensensitive film was monitored by the quartz crystal method, which could ensure the monitored by the quartz crystal method, which could ensure the thickness of the hydrogen-sensitive thickness of the hydrogen-sensitive film as shown in Figure 3b [13,14]. film as shown in Figure 3b [13,14].
Sensors 2017, 17, 1509 Sensors 2017, 17, 1509 Sensors 2017, 17, 1509 Sensors 2017, 17, 1509
4 of 10
(a)
4 of 10 4 of 10 4 of 10
(b)
(a) (b) Figure 3. (a) (b) Pd/Ni-covered UFBG. (a)Polyimide-covered UFBG;(b) Figure 3. (a) Polyimide-covered UFBG; (b) Pd/Ni-covered UFBG.
Figure 3. (a) Polyimide-covered (b) Pd/Ni-covered 3. (a) Polyimide-covered UFBG;UFBG; (b) Pd/Ni-covered UFBG.UFBG. 3.2. ExperimentalFigure Results
3.2. Experimental Results The experiment and interrogation system for multi-parameter measurement system is shown in 3.2. Experimental Results 3.2. Experimental Results Figure 2. During the initialization stagesystem of thefor interrogation system, the delay timer forisSOA1 and The experiment and interrogation multi-parameter measurement system shown in The experiment and interrogation system for multi-parameter measurement system is shown in The experiment and interrogation system for multi-parameter measurement system is shown in SOA2 was stepped 1 ns in every scanning period. Then, the local maximum of the reflected optical Figure 2. During the initialization stage of the interrogation system, the delay timer for SOA1 and Figure 2. During the initialization stagelinear of theimage system, the delay timer for SOA1 and and Figure 2. During the initialization stage ofinterrogation the interrogation system, the and delay forpositions SOA1 pulse from theinInGaAs detector calculated, thetimer space of SOA2 spectrum was stepped 1 ns every scanning period. Then, was the local maximum of the reflected optical SOA2SOA2 was stepped 1 ns in every scanning period. Then, the local maximum of the reflected optical was stepped 1 ns in every scanning period. Then, the local maximum of the reflected optical every were determined by Equation (1). detector was calculated, and the space positions of pulse UFBG spectrum from the InGaAs linear image pulsepulse spectrum from the InGaAs linear image detector was positions of ofevery spectrum from the InGaAs linear image detector wascalculated, calculated,and andthe the space space positions Setting the delay timer equal to 30 ns, every UFBG were determined by Equation (1). 50 ns, 70 ns, and 90 ns, respectively, the reflected everyUFBG UFBG were determined by Equation (1). were determined by Equation (1). spectrum of the UFBG1, UFBG3, andns,UFBG4 All 144 Setting delayUFBG2, timer equal to 30 50 ns, were 70 ns,obtained and 90 accordingly. ns, respectively, the reflected reflected Setting the delay timer equal to 30 ns, 50 ns, ns, and 90 ns, respectively, the reflected Setting the delay timer equal to 30 ns, 50 ns, 70 ns, and 90 ns, respectively, the reflected spectrum spectrums were normalized and are shown in Figure 4. The first nine signal spectrums were spectrum of UFBG1, UFBG2, UFBG3, and UFBG4 were obtained accordingly. All 144 reflected spectrum of UFBG1, UFBG2, UFBG3, and UFBG4 wereinobtained accordingly. All 144spectrums reflected were of UFBG1, UFBG2, UFBG3, and UFBG4 were obtained accordingly. All 144 reflected amplified partially for visible details and are shown Figure 5. spectrums were normalized and are shown in Figure 4. The first nine signal spectrums were spectrums were normalized andinare shown in first Figure The first nine signal spectrumspartially were for normalized and are shown Figure 4. The nine4.signal spectrums were amplified amplified partially for visible details and are shown in Figure 5. amplified partially details are5.shown in Figure 5. visible details for andvisible are shown in and Figure
Figure 4. Reflective spectrum of low reflectivity UFBG array. Figure 4. Reflective spectrum of low reflectivity UFBG array. Figure 4. Reflective spectrum of low reflectivity UFBG array. Figure 4. Reflective spectrum of low reflectivity UFBG array.
Figure 5. Reflective spectrum of UFBG1~UFBG9. Figure 5. Reflective spectrum of UFBG1~UFBG9.
Figure 5. Reflective spectrum of UFBG1~UFBG9. Figure 5. Reflective spectrum of UFBG1~UFBG9.
Sensors Sensors 2017, 2017, 17, 17, 1509 1509 Sensors 2017, 17, 1509
55of of10 10 5 of 10
3.2.1. Temperature Measurement 3.2.1. Temperature Measurement 3.2.1. Temperature Measurement The temperate and humidity chamber model 101-0AB was used for the temperature and RH The temperate and humidity chamber model 101-0AB for temperature and RH The temperate and humidity chamber 101-0AB waswas used thethe temperature andat RH test. The adjusted temperature range of themodel chamber was limited toused 20for °C–250 °C with ±1 °C room ◦ C–250 ◦ C with ±1 ◦ C at test. The adjusted temperature range of the chamber was limited to 20 test. The adjusted temperature range of the chamber was limited to 20 °C–250 °C with ±1 °C at room temperature. UFBG3 was placed in the while the other UFBGs were kept at 25 °C. The ◦ C. The room temperature. UFBG3 placed in the while thethe other UFBGs were atis°C. 25 temperature. UFBG3ofwas in the chamber the other UFBGs were keptkept at 25 The in spectrum shifting the placed firstwas nine sensors duechamber towhile the change in chamber temperature shown spectrum shifting of the first nine sensors due to the change in the chamber temperature is shown in spectrum shifting of the first nine sensors due to the change in the chamber temperature is shown in of Figure 6. The shape of the reflected spectrum of UFBG3 remained unchanged with the increase Figure 6. The shape of the reflected spectrum of UFBG3 remained unchanged with the increase of Figure 6. The shape the reflected spectrum UFBG3 with the of temperature. The of central wavelength can beofcould beremained calculatedunchanged from Equation (1). increase As expected, temperature. The central wavelength canbe beofcould could be from (1).(1). AsAs expected, apart temperature. The central wavelength can be calculated calculated fromEquation Equation expected, apart from UFBG3, the reflected spectrum the other UFBGs remained the same. from UFBG3, thethe reflected spectrum of the other UFBGs remained the the same. apart from UFBG3, reflected spectrum of the other UFBGs remained same.
Figure6. 6.Reflective Reflectivespectrum spectrumshift shiftof ofUFBG3 UFBG3during duringtemperature temperature measurement. measurement. Figure Figure 6. Reflective spectrum shift of UFBG3 during temperature measurement.
Figure showsthe thecontinuous continuous temperature test results of UFBG3 by the setting the Figure 77 shows 120120 minmin temperature test results of UFBG3 by setting chamber Figure 7 shows the continuous 120 min temperature test results of UFBG3 by setting the ◦ C,°C, ◦ C,°C, ◦ C,°C, ◦ C, chamber temperature °C, 6575 °C,◦ Cand 75 °C respectively (75% RH). It canthat be temperature at 25 ◦ C, at 35 25 45 35 5545 6555 and respectively (75% RH). It can be seen chamber temperature at 25 °C, 35 °C, 45 °C, 55 °C, 65 °C, and 75 °C respectively (75% RH). It can be ◦ seen that the temperature response curve was homogeneous and stable. The jitter at 35 °C can be the temperature response curve was homogeneous and stable. The jitter at 35 C can be attributed to seen temperature response curve wasofhomogeneous and stable. The jitter at 35 °C can be attributed to the first regulation the chamber temperature. thethat firstthe regulation fluctuation offluctuation the chamber temperature. attributed to the first regulation fluctuation of the chamber temperature.
Figure 7. Temperature measurement of UFBG3. Figure 7. Temperature measurement of UFBG3. Figure 7. Temperature measurement of UFBG3.
Figure 88 shows shows the the linear linear fitting fitting results results of of temperature temperature with with the the central central wavelength wavelength shift shift of of Figure UFBG3. fitting function 0.01074x + 1552.677, so the temperature sensitivitywas wasabout about FigureThe 8 shows linearwas fitting ofx temperature the central wavelength shift of x = 0.01074 1552.677 UFBG3. The fittingthe function wasf (fx)results , so with the temperature sensitivity ◦ C. 11 pm/ UFBG3. The fitting function was f x 0.01074 x 1552.677 , so the temperature sensitivity was about 11 pm/°C.
11 pm/°C.
Sensors 2017, 17, 1509 Sensors 2017, 17, 1509 Sensors 2017, 17, 1509 Sensors 2017, 17, 1509
6ofof 1010 66of 10 6 of 10
Figure 8. 8. Central wavelength vs.vs. temperature. Figure Central wavelength temperature. Figure 8. Central wavelength vs. temperature. Figure 8. Central wavelength vs. temperature.
3.2.2. 3.2.2.RH RHMeasurement Measurement 3.2.2. RH Measurement 3.2.2. RH Measurement UFBG2 UFBG2was wasplaced placedin thetemperature temperatureand andhumidity humiditychamber chamberwhile whilethe thetemperature temperaturewas was UFBG2 was placed ininthe the temperature and humidity chamber while the temperature was ◦ C. UFBG2 was placed in the and humidity chamber while the temperature was maintained at 30 Keeping the temperature constant and changing the relative humidity from maintained at 30 °C. Keeping the temperature constant and changing the relative humidity from maintained at 30 °C. Keeping the temperature constant and changing the relative humidity from maintained at 30 °C. Keeping the temperature constant and changing the relative humidity from 30% RH to 40% RH, 50% RH, 60% RH, 70% RH, 80% RH, and 90% RH, respectively, the shift of the 30%RH RHtoto40% 40%RH, RH,50% 50%RH, RH,60% 60%RH, RH,70% 70%RH, RH,80% 80%RH, RH,and and90% 90%RH, RH,respectively, respectively,the theshift shiftofofthe the 30% 30% RH to 40% RH, 50% RH, 60% RH, 70% RH, 80% RH, and 90% RH, respectively, the shift of the central wavelength was recorded over 600 min continually with an RH adjustment at about every central wavelength was recorded over 600 min continually with an RH adjustment at about every central wavelength was recorded over 600 min continually with an RH adjustment at about every central wasFigure recorded overresults 600 min continually RH adjustment at conditions about every 100 min aswavelength shown in Figure 9. The verify the factthe that the chamber presents unstable 100 min asshown shown The verify the fact factwith thatan the chamber presents unstable 100 min as ininFigure 9.9.results The results verify that the chamber presents unstable 100 min as shown in Figure 9. The results verify the fact that the chamber presents unstable at low or high RH settings while the temperature is fixed. conditionsatatlow lowororhigh highRH RHsettings settingswhile whilethe thetemperature temperatureisisfixed. fixed. conditions conditions at low or high RH settings while the temperature is fixed.
Figure9.9.UFBG2 UFBG2central centralwavelength wavelengthevolution evolutionwith withthe thechange changeofofrelative relativehumidity humidity Figure .. Figure 9. 9. UFBG2 central wavelength evolution with thethe change of of relative humidity. Figure UFBG2 central wavelength evolution with change relative humidity.
Figure10 10shows showsthe thefitting fittingcurve curveofofthe thecentral centralwavelength wavelengthwith withRH. RH.The Thefitting fittingfunction functionwas was Figure
Figure shows the fitting curve central wavelength with RH. The fitting function was Figure 1010 shows the fitting curve of of thethe central wavelength with RH. The fitting function was 0.001258 1552.917 and theRH RH sensitivity wasabout about 1.26pm/%RH. pm/%RH. 0.001258 f f x x x x1552.917 , ,and the sensitivity was 1.26 0.001258 x 1552.917 , and and the RH sensitivity waswas about 1.261.26 pm/%RH. 0.001258x f ( xf ) x= + 1552.917, the RH sensitivity about pm/%RH.
Figure10. 10.Central Centralwavelength wavelengthvs. vs.humidity. humidity. Figure Figure Central wavelength humidity. Figure 10.10. Central wavelength vs.vs. humidity.
3.2.3.Hydrogen HydrogenConcentration ConcentrationMeasurement Measurement 3.2.3. 3.2.3. Hydrogen Concentration Measurement
Sensors 2017, 17, 1509
7 of 10
Sensors 2017, 17, 17, 1509 Sensors 2017, 1509
7 of7 10 of 10
3.2.3. Hydrogen Concentration Measurement
UFBG1 was place a chamber with varying concentrations of of hydrogen, while the other UFBG1 was place a chamber with varying concentrations hydrogen, while the other UFBG1 was place inina in chamber with varying concentrations of hydrogen, while the other ambient ambient parameters of the UFBGs were fixed. Figure 11 shows the central wavelength shift of of ambient parameters ofwere the fixed. UFBGsFigure were11 fixed. Figure 11 shows the central wavelength shift parameters of the UFBGs shows the central wavelength shift of UFBG1 with the UFBG1 with thethe of of hydrogen 1%, 1.5%, and 2%, respectively. UFBG1 with concentrations hydrogen at respectively. 1%, 1.5%, and 2%, respectively. concentrations ofconcentrations hydrogen at 1%, 1.5%, and at 2%,
Figure 11.11. Hydrogen concentration measurement. Figure Hydrogen concentration measurement. Figure 11. Hydrogen concentration measurement.
The mean value of of samples at at 7~97~9 min at at thethe same hydrogen concentration was calculated. The mean value samples min same hydrogen concentration was calculated. The mean value of samples at 7~9 min atthe same hydrogen concentration was calculated. f x 0.115376 x 1552.258 Figure 12 shows the fitting function to be . When hydrogen concentrations Figure 12 shows the fitting function to be f x 0.115376 x 1552.258 . When hydrogen concentrations Figure 12 shows the fitting function to be f ( x ) = 0.115376x + 1552.258. When hydrogen concentrations were from 1%1% to to 1.5% and from 1.5% to to 2%, thethe wavelength shifts of of thethe FBG areare 40 40 pmpm and 75 75 pm. were from 1.5% and from 1.5% 2%, wavelength shifts FBG and pm. were from 1% to 1.5% and from 1.5% to 2%, the wavelength shifts of the FBG are 40 pm and 75 pm.
Figure 12. Central wavelength vs. hydrogen concentration. Figure Central wavelength hydrogen concentration. Figure 12.12. Central wavelength vs.vs. hydrogen concentration.
3.2.4. Salinity Measurement 3.2.4. Salinity Measurement 3.2.4. Salinity Measurement For salinity measurement, the UFBG4 was placed into aacontainer with 30 ml deionized water, For salinity measurement, the UFBG4 was placed into a container with deionized water, For salinity measurement, the UFBG4 was placed into container with 3030 mlml deionized water, which means that the salinity was 0 mol/L. Then, 10.8 g NaCl was added to the water, bringing which means that the salinity was 0 mol/L. Then, 10.8 g NaCl was added to the water, bringing which means that the salinity was 0 mol/L. Then, 10.8 g NaCl was added to the water, bringing the thethe concentration of saturated salt solution to 6.154 mol/L, and the central wavelength was shifted down concentration saturated salt solution 6.154 mol/L, and the central wavelength was shifted down concentration of of saturated salt solution to to 6.154 mol/L, and the central wavelength was shifted down to 1552.038 nm. Following this, deionized water was injected into the container to obtain a salinity to 1552.038 nm. Following this, deionized water was injected into the container to obtain a salinity to 1552.038 nm. Following this, deionized water was injected into the container to obtain a salinity of of of 44 mol/L and 2 mol/L, and the drift of the central wavelength was recorded, respectively. The linear 4 mol/Land and2 2mol/L, mol/L,and andthe thedrift driftofofthe thecentral centralwavelength wavelengthwas wasrecorded, recorded, respectively. The linear mol/L respectively. The linear fitting curve of the central wavelength with salinity changes is shown in Figure 13. It can be be fitting curve of the central wavelength with salinity changes is shown in Figure 13. It can fitting curve of the central wavelength with salinity changes is shown in Figure 13. It can be expressed −1. −1 f xf x 0.017 x 1552.142 expressed as , with the salinity sensitivity about ~−17 pm/mol·L − 1 0.017 x 1552.142 expressed as , with the salinity sensitivity about ~−17 pm/mol·L . as f ( x ) = −0.017x + 1552.142, with the salinity sensitivity about ~−17 pm/mol·L .
Sensors 2017, 17, 17, 1509 Sensors 2017, 1509
8 of8 10 of 10
Figure Central wavelength salinity. Figure 13.13.Central wavelength vs. vs. salinity.
Cross-Sensitivity Investigations Thermal Compensation 3.3.3.3. Cross-Sensitivity Investigations andand Thermal Compensation expected,the thecross-sensitivity cross-sensitivitybetween between the the temperature including RH, AsAs expected, temperatureand andother otherparameters parameters including salinity, and clearthat thatwithout without temperature RH, salinity, andhydrogen hydrogenwas wasobserved. observed. From From Equation Equation (1), (1), itit isisclear temperature compensation, a correct multi-parameter measurement is not possible. Generally, apart from compensation, a correct multi-parameter measurement is not possible. Generally, apart from thethe temperature effect, the cross-sensitivity among the other parameters measures can be negligible. temperature effect, the cross-sensitivity among the other parameters measures can be negligible. Since FBGs are intrinsically temperature sensors, it is to natural to solve thistaking issuethe bymeasured taking the Since FBGs are intrinsically temperature sensors, it is natural solve this issue by measured temperature near the test points as the role of temperature compensation in this UFBG temperature near the test points as the role of temperature compensation in this UFBG sensing array. sensing Basedarray. on Equation (1), the central wavelength is influenced by the temperature change ∆T and T Based on Equation (1), parameter the centralchange wavelength the(fiber temperature the strain induced by the target ∆M of is theinfluenced transducerby layer coating). change The stress M of the transducer layer (fiber coating). and the strain induced by the target parameter change effects induced by temperature change are described as α M − α. Then the superposition of the central The stressshift effects induced by change are described wavelength λ recorded by temperature the interrogation system is given as:as M . Then the superposition t
of the central wavelength shift t recorded by the interrogation system is given as: (b2 − a2 )·Y
λt = λ B · (1 − Pe ) · β · a2 Y +(b2(−ba2 2C)·Ya 2 )· YM + Ct C Mt t B 1 Pe F 2 2 2 h i a Y ( b a ) Y F C (b2 − a2 )·Y (2) λ B · (1 − Pe ) · a2 Y +(b2 −a2C2)·Y 2· (α M − α) + (α + ξ ) · Tt + F (b Ca ) YC n h i o T B 1 Pe 2 M t 2 2 2 2 (2) a ·YF 2((bb 2−a2 a)·2Y2)C YC · M0 − λ B · (1 −Pe ) · 2 (b − a2 )·Y2C λ0 − λ B · (1 − Pe ) · β · (α M − α) + (α + ξ ) · T0 a YF +(b − a )·YC a YF +(b − a )·YC (b 2 a 2 ) YC (b 2 a 2 ) YC M 0 B 1 Pe 2 M T0 0 B 1 Pe 2 2 2 Where λ0 , M0 and T0 aare target parameter, and (b initial a ) YC value of the central YF the a YF (wavelength, b 2 a 2 ) YC
temperature, by theofinterrogation system; Tt can be calculated 0 , M 0 and λTt0 can Where respectively. arebe theobtained initial value the central wavelength, target parameter,by and UFBG3, the temperature sensor. temperature, respectively. t can be obtained by the interrogation system; Tt can be calculated by Without loss of generality, the measurement of RH and temperature is taken as example. Sweeping UFBG3, the temperature sensor. the ambient temperature from 25 ◦ C to 75 ◦ C with a step of 10 ◦ C and RH from 30% to 90% with a step Without loss of generality, the measurement of RH and temperature is taken as example. of 10%, the central wavelength of UFBG2 was recorded from the interrogation system, as shown in Sweeping the ambient temperature from 25 °C to 75 °C with a step of 10 °C and RH from 30% to 90% Figure 14 after least-squares curve-fitting. with a step of 10%, the central wavelength of UFBG2 was recorded from the interrogation system, as shown in Figure 14 after least-squares curve-fitting.
Sensors 2017, 1509 Sensors 2017, 17,17, 1509
9 9ofof1010
Figure14. 14.Central Centralwavelength wavelengthshift shiftwith withboth bothRH RHand andtemperature temperaturechange. change. Figure
4. Conclusions 4. Conclusions An ultra-weak FBG array has been proved to be one of the most promising solutions for An ultra-weak FBG array has been proved to be one of the most promising solutions for huge-capacity fiber sensing networks. In this work, we proposed and demonstrated a sensing array huge-capacity fiber sensing networks. In this work, we proposed and demonstrated a sensing array for multi-parameter measurements based on a UFBG with sensitive material coating. The central for multi-parameter measurements based on a UFBG with sensitive material coating. The central wavelength readout system employed uses two SOAs to separate the optical pulse from the sensors wavelength readout system employed uses two SOAs to separate the optical pulse from the sensors while rejecting unwanted signals caused by noise, crosstalk, and interference. This setup provides while rejecting unwanted signals caused by noise, crosstalk, and interference. This setup provides the advantage of testing various parameters in a single fiber without limiting temperature and strain the advantage of testing various parameters in a single fiber without limiting temperature and strain measurement, and allows for a larger scale because of the strong multiplexing capability of UFBG. measurement, and allows for a larger scale because of the strong multiplexing capability of UFBG. This method was performed on a 144 UFBG sensors array with the reflectivity of UFBG ~0.04% This method was performed on a 144 UFBG sensors array with the reflectivity of UFBG ~0.04% for the four target parameters: hydrogen, RH, temperature, and salinity. The performance of for the four target parameters: hydrogen, RH, temperature, and salinity. The performance of multi-parameter sensing array was almost equal to result of the single-point FBG with general multi-parameter sensing array was almost equal to result of the single-point FBG with general reflectivity. In order to solve the cross-sensitivity of multi-parameter sensors, with the temperature reflectivity. In order to solve the cross-sensitivity of multi-parameter sensors, with the temperature taking the primary effect to other parameters, a uniform solution was customized to divide the taking the primary effect to other parameters, a uniform solution was customized to divide the central central wavelength shift caused by target parameters from that caused by the temperature effect. wavelength shift caused by target parameters from that caused by the temperature effect. Based on the Based on the report of the multiplexing capacity of ultra-weak FBGs, it is expected that this scheme report of the multiplexing capacity of ultra-weak FBGs, it is expected that this scheme of UFBG-based of UFBG-based sensing network can potentially multiplex thousands of multi-parameter sensors in sensing network can potentially multiplex thousands of multi-parameter sensors in single fiber, which single fiber, which shows promising prospects for the future. shows promising prospects for the future. Acknowledgments:This Thiswork workisisfinically finicallysupported supportedbybythe theProject ProjectofofNational NationalNatural NaturalScience ScienceFoundation Foundationofof Acknowledgments: China,NSFC NSFC (Number: 51402228, 61290311, 61475121, 61575151), National Natural Science China, (Number: 51402228, 61290311, 61475121, 61575151), Project ofProject NationalofNatural Science Foundation ofFoundation Hubei Provincial Government (Number: 2014CFB260), the Creativethe theCreative Creativethe Group Project of Hubei of Hubei Provincial Government (Number: 2014CFB260), Creative Group Project Provincial Science Foundation (Project Number: of Hubei Natural Provincial Natural Science Foundation (Project2015CFA016). Number: 2015CFA016). Author Contributions: Wei Bai, Minghong Yang, and Chenyuan Hu conceived and designed the experiments; Author Contributions: Wei Bai, Minghong Yang, and Chenyuan Hu conceived and designed the experiments; Jixiang Dai, Xuexiang Zhong and Shuai Huang performed the experiments; Xuexiang Zhong and Gaopeng Wang Jixiang Dai, Xuexiang Zhong Shuai Huang performed the experiments; Xuexiang Zhong and Gaopeng analyzed the data; Wei Bai wroteand the paper. Wang analyzed the data; Wei Bai wrote the paper. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest.
References References
1. Yang, M.; Bai, W.; Guo, H.; Wen, H.; Jiang, D. Huge Capacity Fiber-Optic Sensing Network Based on 1. Ultra-Weak Yang, M.; Draw Bai, W.; Guo, H.; Wen, H.; Jiang, Huge Capacity Fiber-Optic Sensing Network Based on Tower Gratings. Photonic Sens.D. 2016, 6, 26–41. [CrossRef] Ultra-Weak Sens. 2016, 6, 26–41. 2. Hu, C.Y.; Wen,Draw H.Q.;Tower Bai, W.Gratings. A NovelPhotonic Interrogation System for Large Scale Sensing Network with Identical 2. Ultra-Weak Hu, C.Y.; Fiber Wen, Bragg H.Q.; Gratings. Bai, W. A Novel Interrogation for Large Scale Sensing Network with J. Lightwave Technol. 2014,System 32, 1406–1411. [CrossRef] Identical Ultra-Weak Fiber Bragg Gratings. J. Lightwave Technol. 2014, 32, 1406–1411. 3. Wang, Y.M.; Gong, J.M.; Wang, D.Y.; Dong, B.; Bi, W.; Wang, A.B. A quasi-distributed sensing network with 3. time-division-multiplexed Wang, Y.M.; Gong, J.M.; fiber Wang, D.Y.;gratings. Dong, B.; Bi, Photonic W.; Wang, A.B. A quasi-distributed sensing network Bragg IEEE Technol. 2011, 23, 70–72. [CrossRef] with time-division-multiplexed fiber Bragg gratings. IEEE Photonic Technol. 2011, 23, 70–72.
Sensors 2017, 17, 1509
4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
10 of 10
Wang, Y.M.; Gong, J.M.; Wang, D.Y.; Shillig, T.J.; Wang, A. A large Serial time-division multiplexed fiber Bragg grating sensor network. J. Lightwave Technol. 2012, 30, 2751–2756. [CrossRef] Wang, S.; Liao, Y.; Yang, H.; Wang, X.; Wang, J. Modeling seawater salinity and temperature sensing based on directional coupler assembled by polyimide-coated. Appl. Opt. 2015, 54, 10283–10289. [CrossRef] [PubMed] Luo, D.; Ma, J.X.; Ibrahim, Z.; Ismail, Z. Etched FBG coated with polyimide for simultaneous detection the salinity. Opt. Commun. 2017, 392, 218–222. [CrossRef] Liu, X.; Zhang, X.; Cong, J.; Xu, J.; Chen, K. Demonstration of etched cladding fiber Bragg grating-based sensors with hydrogel coating. Sens. Actuators 2003, 96, 468–472. [CrossRef] Dai, J.X.; Yang, M.H.; Chen, Y.; Cao, K.; Liao, H.; Zhang, P. Side-polished fiber Bragg grating hydrogen sensor with WO3 -Pd composite film as sensing materials. Opt. Express 2011, 19, 6141–6148. [CrossRef] [PubMed] Wanasekara, N.; Chalivendra, V.; Calvert, P. Sub-micron scale mechanical properties of polypropylene fibers exposed to ultraviolet and thermal degradation. Polym. Degrad. Stab. 2011, 96, 432–437. [CrossRef] Sutapun, B.; Tabib-Azar, M.; Kazemi, A. Pd-coated elastooptic fiber optic Bragg grating sensors for multiplexed hydrogen sensing. Sens. Actuators B Chem. 1999, 60, 27–34. [CrossRef] Zhang, Y.J.; Xie, X.P.; Xu, H.B. Distributed Temperature Sensor System Based on Weak Reflection Fiber Gratings Combined with WDM and OTDR. J. Opt. Electr. Eng. 2012, 39, 69–74. Alwis, L.; Sun, T.; Grattan, K.V. Analysis of Polyimide-Coated Optical Fiber Long-Period Grating-Based Relative Humidity Sensor. IEEE Sens. J. 2013, 13, 767–771. [CrossRef] Dai, J.; Yang, M.; Yu, X.; Cao, K.; Liao, J. Greatly etched fiber Bragg grating hydrogen sensor with Pd/Ni composite film as sensing material. Sens. Actuators 2012, 174, 253–257. [CrossRef] Yang, M.H.; Dai, J.X. Fiber Optic Hydrogen Sensors: A Review. Photonic Sens. 2014, 4, 300–324. [CrossRef] © 2017 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 (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
