hygroscopic strain measurement by fibre bragg gratings sensors in organic matrix composites –...

Composites: Part B 58 (2014) 76–82

Contents lists available at ScienceDirect

Composites: Part B

journal homepage: www.elsevier .com/locate /composi tesb

Hygroscopic strain measurement by fibre Bragg gratings sensorsin organic matrix composites – Application to monitoring of acomposite structure

1359-8368/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.compositesb.2013.10.014

⇑ Corresponding author. Tel.: +33 472436179/612808764.E-mail addresses: hossein.ramezani-dana@insa-lyon.fr, hossein.ramezani_

dana@yahoo.com (H. Ramezani-Dana).

H. Ramezani-Dana ⇑, P. Casari, A. Perronnet, S. Fréour, F. Jacquemin, C. LupiLUNAM Université, Université de Nantes, Centrale Nantes, Institut de Recherche en Génie Civil et Mécanique (UMR CNRS 6183), 58, rue Michel Ange – BP 420,44606 Saint-Nazaire, France

a r t i c l e i n f o

Article history:Received 25 February 2013Received in revised form 30 August 2013Accepted 16 October 2013Available online 29 October 2013

Keywords:A. Polymer–matrix composites (PMCs)B. Optical properties/techniquesB. Environmental degradation

a b s t r a c t

This study is devoted to the identification of the moisture expansion coefficients of composite materialsby means of a novel measurement technique. This method is based on the insertion of Fibre Bragg Grating(FBG) sensors between composite layers. The sensor enables to measure the hygroscopic strain inducedby moisture diffusion in the plane of the laminated composite. Experimental results from immersed sam-ples, varying both the direction of measurement and the fibre volume fraction are given according to thewater uptake, and leading to the characterisation of moisture expansion coefficient.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Organic matrix composites are extensively used in several engi-neering applications due to their exceptional stiffness to weight ra-tio. However, when such materials are exposed to humidenvironments, they absorb moisture which can severely affecttheir hygro-mechanical properties [1–3] as well as the multi-scaleinternal mechanical states experienced by the constituents of poly-mer matrix composites [4,5]. Consequently, durability and reliabil-ity of structural parts made of such materials can be impaired inhumid environments. In the glass/polyester composites used inthis study, the matrix absorbs a significant amount of waterwhereas absorption by fibres is almost null. One key issue is to pre-dict the stress state of the material submitted to such environmentby means of moisture expansion coefficients. The purpose of thiswork is exactly to provide such materials parameters and a tech-nique has been developed for it. Numerous authors have used FibreBragg Grating (FBG) sensors [6–11] in order to achieve local strainmeasurements, temperatures and other quantities of engineeringinterest. An FBG clearly provides many advantages compared toother strain measurement techniques, since it enables to investi-gate internal mechanical states in the depth of the specimen with-out suffering from an ageing of the sensor itself.

Then, an experimental study of both the time-dependent mois-ture uptake and strain field in composite containing FBG sensorshas been achieved. Fibre Bragg Grating (FBG) sensors were usedto characterise the evolution of hygroscopic strain field in unidirec-tional fibre-reinforced composite specimens immersed in de-ion-ised water, during both the transient and the permanent stagesof the moisture diffusion process. Besides, the time-dependentaverage moisture uptake over the volume of the samples was fol-lowed through the classical gravimetric method until the perma-nent regime has been reached. Both longitudinal and transversalmoisture expansion coefficients (CME) of these composite sampleswere determined by using the data collected through the opticalsensors oriented either in a direction parallel or perpendicular tothe reinforcing fibres.

2. Materials and testing methods

2.1. Materials and specimens

The tested unidirectional composite samples were made of E-glass fibres embedded in an ortho-phthalic polyester resin (POLY-LITE 420-731 from REICHOLD) polymerising at room temperature.The specimens were manufactured through the vacuum assistedresin infusion (VARI) method.

Several composites, as well as neat resin square plates 1.3 mmthick were fabricated. After cutting and polishing, the samples sur-faces were cleaned with ethanol in order to remove any residual oiland dirt resulting from machining. The initial weights and

H. Ramezani-Dana et al. / Composites: Part B 58 (2014) 76–82 77

dimensions of the specimens were recorded. The FBGs employed inthis study were printed on standard SMF-28e single mode opticalfibres with a 250 lm diameter. Bragg gratings (10 mm long) wereuniform, and centered on a 1555 ± 0.2 nm wavelength.

The use of FBG in composite samples is limited by the ability toinsert this sensor in the composite material in order to retrieve theoptical signal without (or the less as possible) modifying themechanical behaviour of the composite structure considered.

In order to address this issue, the optical fibre containing Bragggrating was inserted between two plates, bonded by polyesterresin. The FBG were either aligned with the reinforcing fibres orset perpendicularly to them.

In the area of the grating, the FBGs acrylate coating wasremoved in order to get the most direct strain provided by the sur-rounding material towards the sensor.

This type of specimen architecture enabled us to obtain instru-mented samples, the final size of which is 90 � 90 � 3 mm3. Inaddition, this kind of architecture allows us to limit the stressesexperienced by optical fibre during the fabrication process.

The main parameters controlling moisture diffusion kinetics,namely the diffusion coefficients and the maximum moistureabsorption capacity, do strongly depend on the volume fractionof the constituents (i.e., those of the reinforcing fibres and the poly-meric matrix). Therefore, the reinforcing fibres content of the com-posite affects the magnitude of the multi-scale mechanical statesto be monitored during the present work.

As a consequence, the accurate characterisation of the volumefraction occupied by glass fibres in each investigated sample wasmandatory in the context of the present work. Several techniqueshave historically been employed in order to determine the fibrevolume fraction (vf) of organic matrix composites [12,13]. In thepresent work, the identification of the glass–fibre volume fractionin the manufactured samples will be handled according to adimension and weight analysis method.

Optical microscopy and Scanning Electron Microscopy wereused for investigating the microstructure of the manufacturedsamples (see Fig. 1). According to both characterisation methods,specimens are void-free.

As a result, the total volume of the composite (Vt) is equal to thesum of the volume occupied by the glass fibres (Vf), on the onehand and by the resin (Vr), on the other hand:

Vt ¼ Vf þ Vr ð1Þ

A similar relation is also valid for the corresponding masses:

mt ¼ mf þmr ¼ qf Vf þ qrVr ¼ qtVt ð2Þ

Fig. 1. SEM micrograph showing an optical fibre embedded perpendicularly to thereinforcing fibres (white dashed line circle illustrates the optic fibre coating).

where q denotes the mass density. The total density respects thefollowing classical mixture law:

qt ¼ qtv f þ qrv r ¼ qf v f þ qrð1� v f Þ ð3Þ

The reinforcing fibres volume fraction corresponds to the fol-lowing ratio:

v f ¼Vf

Vt¼ 1

qf � qr

mt

Vt� qr

� �¼ qt � qr

qf � qrð4Þ

Dimensions and mass of the specimens were measured in orderto determine their total volume (Vt) and their total mass (mt), fromwhich the total mass density calculation is straightforward. Thesame method was applied to the neat resin samples, so that thenumericalvalue of qr could be deduced. Glass fibres mass densityis a well known parameter that is widely available in the literature[14]. The results obtained through this calculation method, for theglass fibres volume fraction, are displayed in Table 1.

2.2. Experimental set up and hygroscopic ageing process

Two series of specimens were manufactured. The first subset ofsamples was instrumented by FBGs sensors. FBGs were orientedperpendicular or parallel to the reinforcing fibres. This group ofsamples was intended to provide the time-dependent evolutionof the internal strains states throughout the moisture diffusionprocess. The second subset of specimens, which did not containany optical fibre, was devoted to moisture uptake characterisationby means of periodic mass measurements. The samples were alsoimmersed in de-ionised water.

The optical measurement system used is constituted as follows:The optical source is a broadband optical source in C-Band (1530–1560 nm) based on erbium doped Amplified Spontaneous Emis-sion. This light source is connected to a 2–1 optical coupler. Theoptical output coupler is connected to a FBGs realised in a standardSMF-28e single mode optical fibre. The Bragg grating length is10 mm. The reflected signal is received through the optical couplerby an Optical Spectrum Analyzer (Anritsu MS9710B) with 70 pmwavelength resolution. The spectral signature of the Bragg gratingis numerically analysed in order to determine the measured strainduring the hygroscopic ageing test imposed to the samples.

2.3. Technique of measurement

Mechanical strains measurements through FBG have been pre-sented in various papers [15,16]. A FBG consists in a series of grat-ing slices formed along the fibre axis. If the Bragg condition isfulfilled, a light signal propagating into the device can interfereconstructively to the waves reflected by each of the slices. As a re-sult, a back reflected signal with a center wavelength commonlyknown as the Bragg wavelength kb is formed. Bragg-wavelengthkb depends on the effective index of refraction (neff) and on theBragg period (K) of the grating according to the well known Braggequation (Eq. (5)).

By inscribing several FBGs with different grating periods in thesame optical fibre, an array of grating is manufactured. This allowsthe user to monitor different positions in the structure with onlyone sensor line [17].

Table 1Fibre volume content of the manufactured samples.

Specimen group no. 1 2 3

vf (%) 17 21 22

78 H. Ramezani-Dana et al. / Composites: Part B 58 (2014) 76–82

kb ¼ 2neffK ð5Þ

In the case when the FBG is submitted to a homogeneous axialstrain ez and uniform temperature change DT, the Bragg wave-length experiences a deviation Dkb from the reference value kb0

corresponding to the unloaded state (ez = 0; DT ¼ 0):

Dkb

kb¼ kb � kb0

kb¼ aez þ bDT ð6Þ

Coefficients a and b depend on the nature of the optical fibreand the FBG printing parameters. According to relation (6), underisothermal conditions, the Bragg wavelength shift Dkb is propor-tional to the axial strain (ez).

According to expression (6), one obvious major limitation ofFBG sensors in composites is the sensitivity to both the deforma-tion and the temperature. As a result to measure moisture absorp-tion through the deformation variation, the temperature must besimultaneously measured to determined thermo-optic effect andthermo-mechanical deformations to determine deformations onlydue do moisture absorption.

In the present study, the strong hygroscopic strain experiencedby organic matrix composites exposed to moisture diffusion is ex-pected to induce an axial elongation of the optic fibre containingthe Bragg grating [18], resulting in a significant Bragg wavelengthshift according to Eq. (6).

3. Results and discussion

3.1. Hygroscopic ageing tests

The curves of weight gain versus the square root of time ob-tained for the neat resin as well as those corresponding to the com-posite samples are shown in Fig. 2a and b, respectively.

The neat resin (Fig. 2a) exhibits a time-dependent moisture up-take typical from Fickian kinetics. According to Fig. 2b, the threecomposite samples exhibit an almost linear evolution of theirmoisture uptake for several months until a pseudo-plateau indicat-ing that the saturation of the diffusion process is reached. Thus, it isrealistic to also consider a fickian diffusion behaviour for the inves-tigated composite samples.

The identification method of the diffusion parameters used inthis study was explained in [19]. This method was consisting inidentifying the unknowns of the problem by minimising the

Fig. 2. The weight gain versus the square root of time (ffiffitp

) curves for resin

standard deviation between the calculated and measured quanti-ties using a Gauss–Newton algorithm.

The moisture diffusion parameters of both composite and neatresin specimen are then presented in Table 2.

3.2. Internal strain measurement

The measured Bragg’s wavelength shift enables to determinestrain experienced in the center of the instrumented samples, fromrelation (6), assuming that the thermal contribution, namely, theproduct bDT is negligible by comparison to the quantity of interest:a � ez.

Fig. 3 shows the evolution as a function of the square root oftime of the hygroscopic strain, obtained for the three compositespecimens instrumented by FBG positioned perpendicularly tothe reinforcing fibres. Room temperature change of the ambientfluid monitored during the ageing test is also displayed on thesame figures. One can observe three steps on this figure. We cannote a very light induced strain during first day for the samplescontaining 17% and 22% of glass fibres in opposite to the samplewith 21% of glass fibres. Indeed this sample (vf = 21%) was releasedin order to characterise the Bragg grating properties before waterageing effects. This specimen sinks in again after some days. Thislast, led to the ageing period shifting for this specimen. In the sec-ond step, we can note a quasi linear behaviour of deformation dur-ing several months. Finally, the last step corresponds to thepermanent regime of the diffusion process. A significant scatteringof the hygroscopic strains can be noticed during last step. This scat-tering almost follows the temperature change of the ambient fluid,which ranges from 25 �C to 31 �C throughout the ageing test.

Such a temperature change strongly affects the Bragg wave-length and consequently the measured strain.

As an example, we found the hygroscopic strain close to4600 � 10�6 at 31 �C for the sample containing 17% of fibres. Thisvalue decreased to 3800 � 10�6 at 25 �C. This variation of the strainin the saturation state could be explained by a temperature change.Fig. 3b shows the evolution of the strain as a function of squareroot of time for the three composite specimens instrumented byFBG positioned in a direction parallel to the reinforcing fibres.The measured strain in this direction is smaller in comparison tothe first case (a). In fact, in the direction parallel to the reinforce-ments (b), the moisture expansion coefficient of unidirectionalcomposite is usually lower than in the direction perpendicular tothe reinforcements (a), which induces smaller hygroscopic strainduring the moisture diffusion process.

specimen (a) and composite specimens with different fibre content (b).

Table 2Moisture diffusion parameters of composite and neat resin.

Dapparent (mm2/s) D2 (mm2/s) MS (%)

Composite (vf = 17 %) 2.2 � 10-7 2.07 � 10-7 0.85Composite (vf = 21 %) 2.1 � 10-7 1.74 � 10-7 0.78Composite (vf = 22 %) 1.83 � 10-7 1.97 � 10-7 0.79Bulk resin 2.31 � 10-7 2.31 � 10-7 1.5

H. Ramezani-Dana et al. / Composites: Part B 58 (2014) 76–82 79

The effect of temperature variations on the measured strain isalso significant; therefore, the determination of purely hygroscopicstrain value is not easy, particularly in a direction parallel to thereinforcing fibres.

Fig. 4a shows the evolution, as a function of moisture uptake, ofthe hygroscopic strain obtained for the three composite specimensinstrumented by FBG oriented in a direction perpendicular to thereinforcing fibres. Such a curve has to be analysed in order to de-duce the coefficient of moisture expansion of the studiedspecimens.

According to Fig. 4, the measured strain remains close to zero atthe beginning of ageing process. A quasi-linear evolution of strainup to a sample saturation state was found. One can observe thatstrains keep on increasing while the specimen has reached satura-tion. This is explained because the measured moisture uptake is an

Fig. 3. Evolutions of the strain as a function of square root of time for the composite spe

Fig. 4. Strain as a function of moisture uptake for composite specimens instr

average value from the whole sample volume, whereas strains aremeasured in the center of composite specimen where FBGs werelocated. The strain value of specimen with 17% glass fibre contentis higher in this direction, which is due to a higher resin volumefraction.

When FBGs are oriented parallel to the reinforcing fibres (b),hygroscopic strains vary much less and exhibit a significant scatter.As temperature variations contribute to the Bragg peak shift, it isnecessary to correct this effect before analysing more accuratelyFig. 4a and b.

The Bragg wavelengths from before and after ageing and thecorresponding strains were gathered in Table 3.

3.3. Correction from thermal expansion

A part of the strain scatter in the Bragg peak shifts is the effectof temperature leading to both a variation of the optical indexassociated to a global thermal expansion of both the compositelaminate and the optical fibre. In order to determine a better esti-mate of hygroscopic strains, temperature effects are corrected.

Several methods have been proposed in the literature to achievethis task, as an example by combining multiple sensors or by usinga specific sensor [20–27]. In this study, we propose a quick but

cimens instrumented perpendicularly (a) and parallely (b) to the reinforcing fibres.

umented perpendicularly (a) and parallel (b) to the unidirectional fibres.

Table 3Hygroscopic strain for all instrumented specimens.

vf (%) Angle (�) kb (nm) Before immersion kb (nm) After immersion Dkb (nm) e (10�6)

17 0 1554.68 1554.95 0.27 25517 90 1550.98 1556.01 5.03 415721 0 1554.68 1555.14 0.46 37921 90 1551.48 1555.93 4.45 367722 0 1554.68 1554.97 0.29 23922 90 1550.81 1555.18 4.37 3611

80 H. Ramezani-Dana et al. / Composites: Part B 58 (2014) 76–82

reliable method. Actually, additional information is necessary. Forthis purpose, a thermal loading has been conducted once the lam-inates reached a saturated state. The result on strain is plotted inFig. 5 for the three composite specimens instrumented by FBGpositioned perpendicularly (Fig. 5a) or parallel (Fig. 5b) to the rein-forcing fibres, respectively.

From the slope of the curves presented in Fig. 5a and b, one canseparate the effects induced by the temperature change from thatof the hygroscopic strain on the measured Bragg wavelength shift-ing kb. A linear change of the Bragg wavelength occurs as a functionof temperature from 23 �C to 35 �C in the studied directions. Thecomposites specimens, in which FBG sensors were positioned ina direction perpendicular to the reinforcing fibres (Fig. 5a) showgreater slope of Bragg shifting in comparison to the specimens, inwhich FBG sensors were oriented parallel to the reinforcing fibres(Fig. 5b). These results are used to correct the measured strain froma known temperature shift. Indeed by using the graph trend slopeof these curves (k ¼ f ðTÞ), sensitivity of Bragg wavelength to thetemperature was deduced. Accordingly, the contribution due tothe temperature on the Bragg peak shifts could be determinedand separated from the total strain, so that the neat hygroscopicstrain was eventually estimated through (7).

ehyg ¼ etot � ether ð7Þ

According to the measurements, the sensitivity is higher in thedirection perpendicular to the reinforcing fibres.

Fig. 6a represents strains before and after correction as a func-tion of the square root of time for the composites specimen with17% fibre contents instrumented by FBG oriented in a directionperpendicular to the reinforcing fibres. The saturation pattern ofstrains curve for this sample is better defined in comparison tothe uncorrected measurements. In contrary, the deformation ofthe composite specimen with 17% fibre content instrumented by

Fig. 5. Bragg wavelength as a function of temperature for the composite specime

FBG positioned parallel to the reinforcing fibres (Fig. 6b) remainsscattered after correction of the effect of temperature. This lastcould be explained by local heterogeneities of the lamination.Globally, measured strains decrease by removing the effect of tem-perature from the strain response to moisture.

Fig. 7 shows the corrected strains as a function of moisture con-tent for the three composite specimens instrumented by FBG posi-tioned either in a direction perpendicular (Fig. 7a) or parallel(Fig. 7b) to the reinforcing fibres, respectively.

According to Fig. 7a, the evolution of hygroscopic strain and thesaturation pattern of the three samples are better defined in com-parison to non-corrected state. Unfortunately, at this step, we cansee also a bit of disturbance on the saturation pattern for the spec-imen with 17% fibre content. Knowing that this disturbance ap-pears in a saturated state, consecutive evolution of thedeformation is probably associated to a defect of the microstruc-ture, such as a cracking. Hygroscopic strain in the direction parallelto the reinforcements (Fig. 7b) is hard to explain, due to the verysmall strains experienced by the three composite specimens in thatdirection.

From the curves of the hygroscopic strain as a function of thewater uptake, one can also determine the coefficients of moistureexpansion of the studied samples. The moisture expansion coeffi-cients are determined by considering the strains and moisture con-tents in the steady state:

b11 ¼e11

MSð8Þ

b22 ¼e22

MSð9Þ

The values of these coefficients are given in Table 4:

ns instrumented perpendicularly (a) and parallel (b) to the reinforcing fibres.

Fig. 6. Strains evolution as a function of square root of time for the composite specimens instrumented perpendicularly (a) and parallel (b) to the reinforcing fibres.

Fig. 7. Evolutions of the strain as a function of moisture content for the composite specimens instrumented perpendicularly (a) and parallel (b) to the reinforcing fibres.

Table 4Mechanical and hygroscopical properties of elaborated samples.

vf (%) MS (%) e11 (10�6) e22 (10�6) b11 b22

17 0.85 180 4000 0.021 0.47121 0.79 330 3140 0.042 0.39722 0.78 200 3100 0.026 0.397

H. Ramezani-Dana et al. / Composites: Part B 58 (2014) 76–82 81

4. Conclusions

In this work, the hygroscopic strains experienced by compositesamples immersed in de-ionised water were followed from thetransient to the permanent stage of the diffusion process. The mea-surements were carried out owing to Fibre Bragg Grating (FBG)sensors. The results obtained by this method confirm the perti-nence of FBG sensors to measure the local strain.

In this study, the thermal strains due to the temperature changeduring the experimental investigation have been determined in or-der to identify a more realistic estimation of the hygroscopicstrains.

The hygroscopic strain has been identified in a direction parall-ely and perpendicularly to the reinforcing fibres. The combination

of the obtained results to the knowledge of the moisture uptakeenables the identification of the macroscopic coefficients of mois-ture expansion of the studied samples, also.

References

[1] Shen CH, Springer GS. Moisture absorption and desorption of compositematerials. J Compos Mater 1976;10:2–20.

[2] Weitsman YJ. Fatigue of composite materials. New York: Elsevier; 1991.[3] Selzer R, Friedrich K. Mechanical properties and failure behavior of carbon

fibers-reinforced polymer composites under the influence of moisture.Composite Part A 1976;28:595–604.

[4] Youssef G, Fréour S, Jacquemin F. Stress-dependent moisture diffusion incomposite materials. J Compos Mater 2011;43:1621–37.

[5] Sar BE, Fréour S, Davies P, Jacquemin F. Coupling moisture diffusion andinternal mechanical states in polymers – a thermodynamical approach. Eur JMech 2012;36:38–43.

[6] Karalekas D, Cugnoni J, Botsis J. Monitoring of hygrothermal ageing effects inan epoxy resin using FBG sensor: a methodological study. Compos Sci Technol2009;69:507–14.

[7] Pereira G, Frias C, Faria H, Frazão O, Marques AT. On the improvement of strainmeasurements with FBG sensors embedded in unidirectional composites.Polym Testing 2013;32:99–105.

[8] Mihailov SJ. Fiber Bragg grating sensors for harsh environments. Sensors2012;12(2):1898–918.

82 H. Ramezani-Dana et al. / Composites: Part B 58 (2014) 76–82

[9] Frieden J, Cugnoni J, Botsis J, Gmür T, Coric D. High-speed internal strainmeasurements in composite structures under dynamic load using embeddedFBG sensors. Compos Struct 2010;92:1905–12.

[10] Mulle M, Collombet F, Olivier P, Zitoune R, Huchette C, Laurin F, et al.Assessment of cure-residual strains through the thickness of carbon–epoxylaminates using FBGs Part II: technological specimen. Compos A Appl SciManuf 2009;40(10):1534–44.

[11] Yeo TL, Sun T, Grattan KTV. Fibre-optic sensor technologies for humidity andmoisture measurement. Sensors Actuat A 2008;144:280–95.

[12] Ye BS, Svenson AL, Bank LC. Mass and volume fraction properties of pultrudedglass fibre-reinforced composites. Composites 1995;26(10):725–31.

[13] Green P. Fibre volume fraction determination of carbon–epoxy compositesusing an acid digestion bomb. J Mater Sci Lett 1991;10(19):1162–4.

[14] Herakovitch CT. Mechanics of fibrous composites. New York: John Wiley andSons Inc.; 1998.

[15] Helan R, Urban F. Principle of fiber Bragg gratings measurement. In:Proceeding of 30th international spring seminar on electronics technology(ISSE/30). Cluj Napoca, Romania; 2007.

[16] Giaccari P, Dunkel GR, Humbert L, Botsis J, Limberger HG, Salathé RP. On adirect determination of non-uniform internal strain fields using fibre Bragggrating. Smart Mater Struct 2005;14(1):127–36.

[17] Luyckx G, Voet E, Lammens N, Degrieck Y. Strain measurements of compositelaminates with embedded fibre bragg gratings: criticism and opportunities forresearch. Sensors 2011;11(1):384–408.

[18] Frieden J, Cugnoni J, Botsis J, Gmür T. High-speed internal strainmeasurements in composite structures under dynamic load using embeddedFBG sensors. Compos Struct 2010;92:1905–12.

[19] Ramezani dana H, Perronnet A, Freour S, Casari P, Jacquemin F. Identification ofmoisture diffusion parameters in organic matrix composites. J Compos Mater2013;47(9):1081–92.

[20] Li T, Dong X, Chan CC, Hu L, Qianà W. Simultaneous strain and temperaturemeasurement based on a photonic crystal fiber modal-interference interactingwith a long period fiber grating. Opt Commun 2012;285:4874–7.

[21] Xia L, Shuai B, Li W, Liu D. Simultaneous measurement of temperature andinfinitesimal displacement using the high cladding mode coupling loss in fiberBragg grating. Sensors Actuat A 2012;176:53–6.

[22] Valdivielso F, Matıas IR, Arregui FJ. Simultaneous measurement of strain andtemperature using a fiber Bragg grating and a thermochromic material.Sensors Actuat A 2002;101:107–16.

[23] Cavaleiro PM, Araújo FM, Ferreira LA, Santos JL, Farahi F. Simultaneousmeasurement of strain and temperature using Bragg gratings written ingermanosilicate and boron-codoped germanosilicate fibers. IEEE PhotonicsTechnol Lett 1999;11(12):1635–7.

[24] Enríquez DA, Cruz ARD, Thereza M, Giraldi MR. Hybrid FBG–LPG sensor forsurrounding refractive index and temperature simultaneous discrimination.Opt Laser Technol 2012;44(4):981–6.

[25] Kang HK, Ryu CY, Hong CS, Kim CG. Simultaneous measurement of strain andtemperature of structures using fiber optic sensor. J Intell Mater Syst Struct2001;12(4):277–81.

[26] Rao YJ, Yuan SF, Zeng XK, Lian DK, Zhu Y, Wang YP, et al. Simultaneous strainand temperature measurement of advanced 3-D braided composite materialsusing an improved EFPI/FBG system. Opt Lasers Eng 2002;38(6):557–66.

[27] Lopez-Higuera JM, Echevarria J, Quintela A, Jauregui C, Cobo A. Strain andtemperature transducer on one fiber Bragg grating. In: Proceedings of 8thannual international symposium on smart structures and materials (SPIE/8),Newport Beach, California, USA; March 2001. p. 192–98.