- Email: [email protected]
Concentration sensor based on a tilted fiber Bragg grating for anions monitoring
Optical Fiber Technology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Optical Fiber Technology www.elsevier.com/locate/yofte
Concentration sensor based on a tilted fiber Bragg grating for anions monitoring L.B. Melo a, J.M.M. Rodrigues b, A.S.F. Farinha b, C.A. Marques a, L. Bilro a, N. Alberto a, J.P.C. Tomé b, R.N. Nogueira a,⇑ a b
Instituto de Telecomunicações, 3810-193 Aveiro, Portugal QOPNA, Department of Chemistry, University of Aveiro, 3810-193 Aveiro, Portugal
a r t i c l e
i n f o
Article history: Received 7 December 2013 Revised 28 March 2014 Available online xxxx Keywords: Sensors Chemical sensors Tilted fiber Bragg gratings Bio-chemical compounds Anions Refractive index
a b s t r a c t The ubiquity and importance of anions in many crucial roles accounts for the current high interest in the design and preparation of effective sensors for these species. Therefore, a tilted fiber Bragg grating sensor was fabricated to investigate individual detection of different anion concentrations in ethyl acetate, namely acetate, fluoride and chloride. The influence of the refractive index on the transmission spectrum of a tilted fiber Bragg grating was determined by developing a new demodulation method. This is based on the calculation of the standard deviation between the cladding modes of the transmission spectrum and its smoothing function. The standard deviation method was used to monitor concentrations of different anions. The sensor resolution obtained for the anion acetate, fluoride and chloride is 79 105 mol/ dm3, 119 105 mol/dm3 and 78 105 mol/dm3, respectively, within the concentration range of (39– 396) 105 mol/dm3. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Considering that anions play an important role in biological and environmental applications, the development of sensors for anion monitoring is of a great interest by the scientific community. Among different anions, acetate (CH3CO 2 ), fluoride (F ) and chloride (Cl) are of extreme importance in various situations. Acetate is present in numerous metabolic processes [1] as well as can be an indicator of organic decomposition in marine sediments [2]. The monitoring of fluoride in water is important to prevent dental caries [3] and its detection can help in clinical treatments such as osteoporosis [4]. With regard to chloride, a small change of the anion flow across the cell membranes can be an indicator of a disease named cystic fibrosis [5]. From this short list, it is notorious that the monitoring of anions is highly desirable. Anions sensing is traditionally performed by host molecules that are designed to recognize a target anion within a given concentration range [6,7]. When the binding between the host and the anion occurs, the host molecule emits a signal which can be ⇑ Corresponding author. Fax: +351 234377901. E-mail addresses: [email protected] (L.B. Melo), [email protected] (J.M.M. Rodrigues), [email protected] (A.S.F. Farinha), [email protected] (C.A. Marques), lucia. [email protected] (L. Bilro), [email protected] (N. Alberto), [email protected] (J.P.C. Tomé), [email protected] (R.N. Nogueira).
either colorimetric [8], fluorescent [9] and/or an electrochemical response [10]. In the last decade, the development of optical fiber sensors for the detection of chemical compounds has become an interesting topic of investigation due to the possibility of producing small sized sensors that are able to remotely and continuously operate in contact with the sample. A number of sensor configurations have been proposed for detection of chemical compounds by monitoring the change of the surrounding refractive index (SRI). These include tapers [11], fiber Bragg gratings (FBGs) [12,13], tilted fiber Bragg gratings (TFBGs) [14], long period gratings (LPGs) [15,16] and surface plasmon resonance (SPR) sensors [17]. Fiber tapering has demonstrated high sensitivity to the SRI; however, these sensors are significantly weaker than unmodified fiber due to the reduction of the fiber diameter [11]. FBGs can also be used to measure SRI by etching the cladding, which leads to a decrease of the fiber mechanical strength [12,13]. LPGs are sensitive to the SRI due to the coupling of radiation from the propagating core mode to the forward-propagating cladding modes [18]. The most prominent drawback of LPGs is the high sensitivity to other parameters such as temperature and strain, which can cause undesirable effects as an SRI sensor. SPR sensors exhibit very high sensitivity to the SRI; however, these sensors normally operate in the visible region which presents higher optical losses compared to the infrared region [17].
http://dx.doi.org/10.1016/j.yofte.2014.05.002 1068-5200/Ó 2014 Elsevier Inc. All rights reserved.
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002
L.B. Melo et al. / Optical Fiber Technology xxx (2014) xxx–xxx
A TFBG consist of a periodic modulation tilted with respect to the longitudinal axis of the fiber core which enables the coupling between the forward-propagating core mode and the forward/ backward-propagating cladding modes [19]. The core and cladding modes appear as a series of multiple dips in the transmission spectrum. The resonance wavelength and the intensity of the dips change as a function of the SRI. Different demodulation methods can be employed to determine the changes of the SRI in a TFBG [20]. Initially, Laffont and Ferdinand reported a technique based on the envelope area of the transmission spectrum [21]. Later, Caucheteur and Mégret improved this method through the correlation between the transmission spectrum and the calculation of two statistical parameters, namely Skewness and Kurtosis [22]. Another widely used method is based on the calculation of the transmission spectrum area [23,24]. In 2007, Chan et al. reported a method in which the wavelength separation between selected cladding modes was calculated [25]. More recently, an alternative demodulation method based on the measurement of the single wavelength time delay has been exploited [26]. In this paper, the use of a TFBG as a concentration sensor for three different anions species namely CH3CO and Cl is 2, F reported. The detection of anions is performed by measuring the changes of the SRI caused by different anion concentrations. For this purpose a new demodulation method based on the calculation of the standard deviation between the transmission spectrum of a TFBG and its smoothing function was developed.
2. Theory of TFBG TFBGs are classified as short period gratings whose refractive index modulation is tilted with respect to the longitudinal axis of the fiber. TFBGs written with a small tilt angle promote the coupling of the forward-propagating core mode to the backwardpropagating cladding modes and decrease the coupling to the backward-propagating core mode [19]. As a consequence, a number of resonances of the cladding modes as well as the resonance of the core mode can be observed in the transmission spectrum (Fig. 1). The resonances can be explained by the phase match condition through the following equations:
kB ¼
2nneff ;core KT cosðhÞ
kiclad ¼
ðneff ;core þ nieff ;clad ÞKT cosðhÞ
ð1Þ
ð2Þ
where neff ;core and nieff ;clad are the effective refractive indices of the core mode and the ith cladding mode, respectively. The nominal grating period is represented by KT and can be expressed as K cosðhÞ; where K is the pitch along the fiber axis and h is the tilt angle. The effective refractive index of the cladding modes depends on the SRI. Therefore, a change of the SRI affects the transmission spectrum of a TFBG. When the grating is immersed in a solution with a small SRI, numerous resonances are visible as a result of the coupling to the backward-propagating cladding modes. As the SRI increases, the cladding modes are gradually cutoff from lower to higher wavelengths and the transmission spectrum gradually approaches its smoothing function. When SRI equals nieff ;clad , the cladding modes are no further guided and a smoothed spectrum is observed.
5
Transmission (dB)
2
&ODGGLQJ0RGHV
0 -5 -10 *KRVW0RGH
-15
&RUH0RGH
-20 -25 1500
1520
1540
1560
Wavelength (nm) Fig. 1. Transmission spectrum of a tilted fiber Bragg grating.
3. Experimental methods A TFBG was fabricated with a 248 nm KrF excimer laser in a standard single mode photosensitive fiber (PS1250/1500, Fibercore), using the phase mask technique. The fiber was previously hydrogen-loaded at 150 bars during two weeks to further enhance its photosensitivity. The phase mask, with a period of 1073.11 nm, was tilted in the perpendicular plane to the laser beam. The grating was written with a total length of 15 mm and an internal tilted angle of 8°. After the inscription, the grating was placed in a heat chamber at 80 °C for 24 h to lose the excess of hydrogen. The experimental setup used for anions monitoring is presented in Fig. 2. The grating was placed in a cylindrical glass container (reactor) and both sides of the fiber were fixed, one at a holder and the other at a manual linear stage. All experiments were performed under a temperature of 23.0 °C controlled by a climatic chamber (Challenge 340, Angelantoni Industrie) with a resolution of 0.1 °C. The manual linear stage was used to keep the fiber taut and the magnet was introduced in the reactor to ensure the homogeneity of the solutions. The transmission spectrum of the TFBG sensor was acquired by an Optical Network Analyzer (ONA) with a wavelength resolution of 4.2 103 nm. The TFBG was characterized to the SRI by using different concentrations of glycerin solutions, providing a refractive index range from 1.341 to 1.437 RIU (refractive index unit). The refractive index values were measured at ambient temperature by a commercial Abbe refractometer (AR4D, Krüss Optronic), at the wavelength of 589 nm and with a resolution of 1 103. The tests were performed by acquiring the transmission spectrum of the TFBG when immersed in each glycerin solutions, starting from the lowest to highest refractive index. After the first set of glycerin
Fig. 2. Experimental setup for anions monitoring.
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002
L.B. Melo et al. / Optical Fiber Technology xxx (2014) xxx–xxx Table 1 Final concentration of the solution in equilibrium after the injection of one anion dose. Anions
Stock solution concentration (mol/ dm3) 104
Injection dose (lL) (±0.5 lL)
Final concentration (mol/dm3) 106 (reactor)
CH3CO 2 F Cl
687 ± 4 300 ± 2 376 ± 3
15.0 34.0 27.0
206 ± 6 204 ± 7 203 ± 7
Table 2 Concentrations and volumes of the stock solutions used to characterize the TFBG sensor. Anions
Stock solution concentration (mol/dm3) 104
Injection dose (lL) (±0.5 lL)
Concentration range in the reactor (mol/dm3) 105
CH3CO 2 F Cl
395 ± 3 300 ± 2 376 ± 3
50.0 66.0 52.0
40–395 40–396 39–391
tests, the fiber was rinsed and two more trials were performed to ensure the reproducibility of the results. The first anion experiment was conducted to investigate the dissolution time of three different anions, namely CH3CO 2, F and Cl (Sigma–Aldrich) in ethyl acetate (analytical reagent grade, Fisher Scientific). Stock solutions of each anion species were prepared by adding the tetrabutylammonium salts of the anions to ethyl acetate (see ‘‘Stock solution concentration’’ on Table 1). The flasks of the stock solutions and the solvent (i.e. ethyl acetate) were placed inside of the climatic chamber under 23.0 °C prior to the beginning of the experiment, and kept in the chamber during the experiment to eliminate thermal fluid variations. The study was performed by initially pouring the reactor with 5 mL of ethyl acetate. To establish a baseline, the transmission spectrum of the TFBG in the presence of ethyl acetate was acquired at the start of the experiment and at the minute 10. Five minutes after the acquisition of the second spectrum, 15 lL of the stock solution of the anion CH3CO 2 (see ‘‘Injection dose’’ on Table 1) was injected into the reactor with a micropipette and the stirrer was turned on during 5 min. After that, the stirrer was turned off and a new TFBG spectrum was acquired. During the next 25 min the acquisition was repeated every 5 min. Afterwards, the reactor was cleaned with ethyl acetate and the same experimental procedure was repeated to measure the time response for the anions F and Cl. The injection dose of each anion was determined so that the concentration of the solution in the reactor after the injection of one anion dose was approximately 200 106 mol/dm3 (see ‘‘Final concentration’’ on Table 1). The second anion experiment was conducted to investigate the performance of the TFBG to different anions concentrations (see ‘‘Concentration range in the reactor’’ on Table 2). For this experiment, new stock solutions of the three anions were prepared (see ‘‘Stock solution concentration’’ on Table 2). Thermal fluid variations were eliminated by using the climatic chamber to keep the solutions at 23.0 °C. The study was performed by initially pouring the reactor with 5 mL of ethyl acetate. After this, a certain dose of the stock solution (see ‘‘Injection dose’’ on Table 2) was introduced into the reactor. The injection dose was calculated in order to increment the concentration of the anion in the reactor by approximately 40 105 mol/dm3. After the anion injection dose, the stirrer was turned on during 5 min followed by a stabilization period of 15 min. The transmission spectrum of the TFBG was acquired 15 min from the time that the stirrer was turned off. The anion injection procedure was repeated giving a total of 10 injections
3
doses. Three trials were performed for each anion species to ensure the reproducibility of the results. 4. Demodulation method In this section, a new demodulation method to analyze the transmission spectrum of a TFBG is presented. This method is based on the calculation of a standard deviation estimator (S), aiming to provide a practical and accurate algorithm to calculate SRI changes. Since the change of the SRI on the TFBG spectrum has only impact on the cladding modes, the wavelength range from 1500.000 nm to 1552.520 nm was selected. This accounts for the transmission spectrum of the grating used in the experiments (Fig. 1). The acquired transmission spectra of the TFBG immersed in different glycerin solutions were converted to a linear scale (Fig. 3). The four spectra of Fig. 3 show that the amplitude of the cladding modes decreases as the SRI increases. In the Fig. 3(d), the SRI is close to the refractive index of the fiber cladding and as a consequence, most of the cladding modes are no longer guided. In this situation the coupling is continuously done by the radiation modes. For each acquired spectrum, a smoothing function was calculated by the Savitzky–Golay method [27]. This consists of a polynomial regression to the data points in order to determine a smoothed value. This numerical tool enables the calculation of a set of points that replicate the trend of the experimental data. The smoothing of the transmission spectra is represented by the dashed red line in Fig. 3. After the calculation of the smoothing function for each spectrum corresponding to a different glycerin solution, a standard deviation analysis was applied. This common estimator is defined as:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u N u 1 X S¼t ðxi xi Þ2 N 1 i¼1
ð3Þ
where, N is the total number of points that constitutes the transmission spectra, xi corresponds to the ith values of the transmission spectrum expressed in the linear scale and xi corresponds to the ith values of the smoothing function. The performance of the standard deviation method was compared with two other intensity-based methods, namely the calculation of the area of the transmission spectrum, and the calculation of the envelope’s area of the transmission spectrum. Fig. 4 shows the SRI characterization of the TFBG analyzed by the three demodulation methods. The Y error bars represent the error determined for the three trials. Considering that the SRI range of interest in this work is around the refractive index of ethyl acetate (i.e. 1.370), the sensitivity of each method was calculated between the two experimental values in the vicinity of 1.370. For the SRI range between 1.364 and 1.381, the sensitivity of the standard deviation, area and envelope is 8.5/ SRI, 6.6/SRI and 7.0/SRI, respectively. Fig. 4 shows that for a wide range of SRI (i.e. between 1.341 and 1.437), the standard deviation offers higher sensitivity than the area. For the same range of SRI, the performance of the standard deviation is similar to the envelope. Nevertheless, the standard deviation does not require the calculation of the peaks of the transmission spectrum whereas the envelope requires this calculation in order to determine the upper and lower envelopes of the TFBG spectrum. Therefore, the standard deviation might become extremely useful when the grating spectrum presents a low signal-to-noise ratio. In this case, the detection of peaks of the transmission spectrum is difficult to implement, which makes
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002
4
L.B. Melo et al. / Optical Fiber Technology xxx (2014) xxx–xxx
1.0
Refractive index: 1.345 Smoothing
1.0 0.8
Transmission
Transmission
0.8 0.6 0.4 0.2 0.0
0.6 0.4 0.2 0.0
1500 1510 1520 1530 1540 1550 1560
1.0
1500 1510 1520 1530 1540 1550 1560
Wavelength (nm)
Wavelength (nm)
(a)
(b)
Refractive index: 1.410 Smoothing
1.0
Transmission
0.8
Transmission
Refractive index: 1.381 Smoothing
0.6 0.4 0.2
Refractive index: 1.437 Smoothing
0.8 0.6 0.4 0.2
0.0 1500 1510 1520 1530 1540 1550 1560
0.0
1500 1510 1520 1530 1540 1550 1560
Wavelength (nm)
Wavelength (nm)
(c)
(d)
Normalized S and Area
Fig. 3. Transmission spectrum and the calculated smoothing function for different surrounding refractive indices: (a) 1.345; (b) 1.381; (c) 1.410; and (d) 1.437.
1.0 0.8 0.6 0.4 0.2 0.0
Standard deviation method Area method Envelope method 1.340 1.360 1.380 1.400 1.420 1.440
Surrounding refractive index Fig. 4. Comparison between the area, the envelope and the standard deviation methods as regards to the change of the surrounding refractive index.
the use of the envelope more complex than the standard deviation. It is worthwhile noting that only intensity-based methods are compared in this article. Another demodulation option that offers high signal-to-noise ratio is the method that used the wavelength separation between a selected cladding mode and the Bragg mode [25].
5. Experimental results and discussion This section presents the results of the TFBG as a sensor for anions monitoring using the standard deviation method presented in Section 4.
The dissolution time of the anions CH3CO and Cl was 2, F investigated by monitoring the variation of the standard deviation caused by an injection dose (Fig. 5). The experimental values depicted in Fig. 5 were normalized by the value of the standard deviation of the grating when immersed in ethyl acetate. Fig. 5 shows that the value of the standard deviation remains constant in the first 10 min corresponding to the detection of ethyl acetate. The major change of the standard deviation occurs 5 min after the anion injection dose. The graph also shows that the equilibrium is obtained 15 min after the time that the stirrer was turn off. This behavior was observed for the three anions under study. Fig. 6 shows the variation of standard deviation as a function of anions concentration. Each point corresponds to an anion injection dose. The standard deviation value of each injection dose was normalized to the value calculated for the first injection, i.e., relative to the concentration of 40 105 mol/dm3. The error bars represent the deviation (r) associated with the three trials performed for each anion. Fig. 6 shows that the standard deviation decreases with the increase of the anion concentration. The variation of the standard deviation between the first and last injection dose of the anions CH3CO 2 , F and Cl is 0.0359, 0.0421 and 0.0502, respectively. By using the sensitivity value of the TFBG calculated in Section 4 (i.e. 8.5/SRI), the variations of the standard deviation correspond to an SRI variation of 0.0042 RIU (CH3CO 2 ), 0.0050 RIU (F ) and 0.0059 RIU (Cl ). The third-degree polynomial fitting was found to be the best model to describe the sensor behavior. The coefficients of each fit ting equation for CH3CO 2 , F and Cl are presented in the Table 3. It can be observed in Fig. 6 that each anion species has a different characterization curve based on the mean values of each trial.
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002
5
L.B. Melo et al. / Optical Fiber Technology xxx (2014) xxx–xxx
&+&2
1.00
)
0.99
&O
0.99
Normalized S
Normalized S
1.00
-
,QMHFWLRQGRVH 6WLUUHURQ 6WLUUHURII
0.98
(TXLOLEULXPWLPH
0.97 0.96
CH3 CO2 -
F Cl
-
0.98 0.97 0.96 0.95
0
10
20
30
40
50
0.0
Time (min)
-3
1.0x10
-3
2.0x10
-3
3.0x10
-3
4.0x10
Concentration (mol/dm3)
Fig. 5. The dissolution of the anions in ethyl acetate. Fig. 7. Sensor resolution for each anion under study.
6. Conclusions and outlook
Normalized S
1.00 0.99 0.98 0.97
&+&2
0.96
)
0.95
&O 7KLUGGHJUHHSRO\QRPLDOILWWLQJV
0.94 0.0
-3
1.0x10
-3
2.0x10
-3
-3
3.0x10
4.0x10
Concentration (mol/dm3) Fig. 6. Variation of the normalized standard deviation with the change of anion concentrations.
Table 3 Third-degree polynomial fitting coefficients in the form S = a + bx + cx2 + dx3 for the anions under study. R2 represents the determination coefficient associated with each anion concentration fitting.
CH3CO 2 F Cl
a (104)
b
c (102)
d (105)
R2
10,077 ± 4 10,102 ± 3 10,133 ± 3
21 ± 1 28 ± 1 37 ± 1
40 ± 8 69 ± 7 80 ± 8
4 ± 1 8 ± 1 7 ± 2
0.9997 0.9997 0.9998
However, the error bars associated to each experimental point suggest that the TFBG is only able to discriminate between the anions CH3CO 2 and Cl and between the anions F and Cl . The step of each concentration point depicted in Fig. 6 corresponds to an incremental injection dose of 40 105 mol/dm3. However, the value of the error bars of the three trials for each anion suggests that the increment of the anion injection dose is lower than the realistic sensor resolution that can be achieved by the TFBG proposed here. Therefore, the sensor resolution for each anion species was investigated by selecting the concentration increment that corresponds to a higher variation of the standard deviation between two consecutive values than the error bars of the selected experimental points. Fig. 7 shows the final selected concentration points for each anion species, which correspond to a resolution of 79 105 mol/dm3, 119 105 mol/dm3 and 78 105 mol/dm3 for the CH3CO 2 , F and Cl , respectively.
A preliminary study of an optical fiber sensor for detection of anions is reported. In this article, each type of anion was investigated individually. However, a real environment is generally characterized by a medium with complex solutions (i.e. different types of anions in solution). Therefore, the selectivity of TFBG sensors must be investigated to allow utilization of these sensors in practical applications. One possibility to achieve selectivity to a target anion is by attaching specific chemical compounds to the surface of an optical fiber by a process so-called functionalization [3,6]. The chemical compounds are fabricated to exhibit affinity for capturing target anions. Fortunately, there is precedent for optical fiber functionalization [13,14,28]. Examples of chemical compounds in the literature that show selectivity to the investigated anions include: sensors based on quinonehydrazone group for selectivity to the anion CH3CO 2 [29], sensors based on derivatives of meso-octamethylcalix[4]pyrrole for selectivity to the anion F [3], and sensors based on derivatives of calix[4]arene for selectivity to the anion Cl [30]. The authors are currently developing chemical compounds that can selectively capture specific anions [3,31]. These chemical compounds will be covalently attached to the surface of the optical fiber in the region of the TFBG. The selectivity of functionalized TFBG sensors will be analyzed in a future work. Additionally, a new demodulation method was developed in order to quantify the SRI changes in the transmission spectrum of the TFBG. This method is based on the calculation of a statistical parameter, namely standard deviation estimator, using the smoothed transmission spectrum as the reference. A sensor resolution of 79 105 mol/dm3, 119 105 mol/dm3 and 78 105 mol/dm3 was achieved for the anions CH3CO and Cl, 2, F respectively. Acknowledgments The authors gratefully acknowledge funding provided by European Regional Development Fund (ERDF) and Science and Technology Foundation, Ministry of Education and Science (FCT, Portugal). In particular, the financial support provided by the Projects PTDC/ CTM/101538/2008 and PTDC/EEA/122792/2010. The authors João Rodrigues, Andreia Farinha, Carlos Marques, Lúcia Bilro and Nélia Alberto acknowledge their doctoral (BD) and post-doctoral (BPD) support provided by FCT, namely by the fellowships SFRH/BD/ 81014/2011, SFRH/BPD/73060/2010, SFRH/BD/70661/2010, SFRH/ BPD/78205/2011 and SFRH/BPD/78141/2011.
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002
6
L.B. Melo et al. / Optical Fiber Technology xxx (2014) xxx–xxx
References [1] G. Liu, J. Shao, Fluorescence anion sensors based on combination of conformational restriction and photo-induced electron transfer, J. Lumin. 135 (2013) 312–317. [2] H. Maeda, Y. Ito, BF2 complex of fluorinated dipyrrolyldiketone: a new class of efficient receptor for acetate anions, Inorg. Chem. 45 (2006) 8205–8210. [3] N.I.P. Valente, P.V. Muteto, A.S.F. Farinha, A.C. Tomé, J.A.B.P. Oliveira, M.T.S.R. Gomes, An acoustic wave sensor for the hydrophilic fluoride, Sens. Actuators, B 157 (2011) 594–599. [4] D. Sharma, S.K. Sahoo, R.K. Bera, R. Kamal, Spectroscopic and computational study of a naphthalene derivative as colorimetric and fluorescent sensor for bioactive anions, J. Fluoresc. 23 (2013) 387–392. [5] N. Busschaert, P.A. Gale, Small-molecule lipid-bilayer anion transporters for biological applications, Angew. Chem. Int. Ed. 52 (2013) 1374–1382. [6] L.E. Santos-Figueroa, M.E. Moragues, E. Climent, A. Agostini, R. MartínezMáñez, F. Sancenón, Chromogenic and fluorogenic chemosensors and reagents for anions. A comprehensive review of the years 2010–2011, Chem. Soc. Rev. 42 (2013) 3489–3613. [7] G.T. Spence, P.D. Beer, Expanding the scope of the anion templated synthesis of interlocked structures, Acc. Chem. Res. 46 (2013) 571–586. [8] E.M. Boyle, S. Comby, J.K. Molloy, Thorfinnur Gunnlaugsson, Thiourea derived Tröger’s bases as molecular cleft receptors and colorimetric sensors for anions, J. Org. Chem. 78 (2013) 8312–8319. [9] R.M. Duke, E.B. Veale, F.M. Pfeffer, P.E. Kruger, T. Gunnlaugsson, Colorimetric and fluorescent anion sensors: an overview of recent developments in the use of 1,8-naphthalimide-based chemosensors, Chem. Soc. Rev. 39 (2010) 3936– 3953. [10] N.H. Evans, H. Rahman, J.J. Davis, P.D. Beer, Surface-attached sensors for cation and anion recognition, Anal. Bioanal. Chem. 402 (2012) 1739–1748. [11] Y. Tian, W. Wang, N. Wu, X. Zou, X. Wang, Tapered optical fiber sensor for label-free detection of biomolecules, Sensors 11 (2011) 3780–3790. [12] S.S. Saini, C. Stanford, S.M. Lee, J. Park, P. DeShong, W.E. Bentley, M. Dagenais, Monolayer detection of biochemical agents using etched-core fiber Bragg grating sensors, IEEE Photonics Technol. Lett. 19 (2007) 1341–1343. [13] G. Ryu, M. Dagenais, M.T. Hurley, P. DeShong, High specificity binding of lectins to carbohydrate-functionalized fiber Bragg gratings: a new model for biosensing applications, IEEE J. Sel. Top. Quantum 16 (2010) 647–653. [14] S. Maguis, G. Laffont, P. Ferdinand, B. Carbonnier, K. Kham, T. Mekhalif, M.C. Millot, Biofunctionalized tilted fiber Bragg gratings for label-free immunosensing, Opt. Express 16 (2008) 19049–19062. [15] P. Pilla, V. Malachovská, A. Borriello, A. Buosciolo, M. Giordano, L. Ambrosio, A. Cutolo, A. Cusano, Transition mode long period grating biosensor with functional multilayer coatings, Opt. Express 19 (2011) 512–526. [16] P.F. Manzillo, P. Pilla, A. Buosciolo, S. Campopiano, A. Cutolo, A. Borriello, M. Giordano, A. Cusano, Self assembling and coordination of water nano-layers on
[17]
[18]
[19] [20] [21]
[22] [23]
[24]
[25]
[26]
[27] [28]
[29]
[30]
[31]
polymer coated long period gratings: toward new perspectives for cation detection, Soft Mater. 9 (2011) 238–263. S.K. Srivastava, B.D. Gupta, Influence of ions on the surface plasmon resonance spectrum of a fiber optic refractive index sensor, Sens. Actuators, B 156 (2011) 559–562. V. Bhatia, D.K. Campbell, T.D. Alberto, G.A.T. Eyck, D. Sherr, K. Murphy, R.O. Claus, Standard optical fiber long-period gratings with reduced temperature sensitivity for strain and refractive-index sensing, Proc. IEEE OFC’97, Dallas, TX, USA, 1997, pp. 346–347. T. Erdogan, J.E. Sipe, Tilted fiber phase gratings, J. Opt. Soc. Am. A 13 (1996) 296–313. J. Albert, L.Y. Shao, C. Caucheteur, Tilted fiber Bragg grating sensors, Laser Photonics Rev. 7 (2013) 83–108. G. Laffont, P. Ferdinand, Tilted short-period fiber-Bragg-grating-induced coupling to cladding modes for accurate refractometry, Meas. Sci. Technol. 12 (2001) 765–770. C. Caucheteur, P. Megret, Demodulation technique for weakly tilted fiber Bragg grating refractometer, IEEE Photonics Technol. Lett. 17 (2005) 2703–2705. N.J. Alberto, C.A. Marques, J.L. Pinto, R.N. Nogueira, Three-parameter optical fiber sensor based on a tilted fiber Bragg grating, Appl. Opt. 49 (2010) 6085– 6091. Y.P. Miao, B. Liu, Q. Zhao, Refractive index sensor based on measuring the transmission power of tilted fiber Bragg grating, Opt. Fiber Technol. 15 (2009) 233–236. C.F. Chan, C. Chen, A. Jafari, A. Laronche, D.J. Thomson, J. Albert, Optical fiber refractometer using narrowband cladding-mode resonance shifts, Appl. Opt. 46 (2007) 1142–1149. M. Pisco, A. Ricciardi, S. Campopiano, C. Caucheteur, P. Mégret, Time delay measurements as promising technique for tilted fiber Bragg grating sensors interrogation, IEEE Photonics Technol. Lett. 21 (2009) 1752–1754. A. Savitzky, M.J.E. Golay, Smoothing and differentiation of data by simplified least squares, Anal. Chem. 36 (1964) 1627–1639. J. Barnes, M. Dreher, K. Plett, R.S. Brown, C.M. Crudden, H.P. Loock, Chemical sensor based on a long-period fibre grating modified by a functionalized polydimethylsiloxane coating, Analyst 133 (2008) 1541–1549. S. Hu, Y. Guo, J. Xu, S. Shao, A selective chromogenic molecular sensor for acetate anions in a mixed acetonitrile–water medium, Org. Biomol. Chem. 6 (2008) 2071–2075. J.N. Babu, V. Bhalla, M. Kumar, R.K. Mahajan, R.K. Puri, A chloride selective sensor based on a calix[4]arene possessing a urea moiety, Tetrahedron Lett. 49 (2008) 2772–2775. J.M.M. Rodrigues, A.S.F. Farinha, P.V. Muteto, S.M. Woranovicz-Barreira, F.A.A. Paz, M.G.P.M.S. Neves, J.A.S. Cavaleiro, A.C. Tome, M.T.S.R. Gomes, J.L. Sessler, J.P.C. Tome, New porphyrin derivatives for phosphate anion sensing in both organic and aqueous media, Chem. Commun. 50 (2014) 1359–1361.
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002
Contents lists available at ScienceDirect
Optical Fiber Technology www.elsevier.com/locate/yofte
Concentration sensor based on a tilted fiber Bragg grating for anions monitoring L.B. Melo a, J.M.M. Rodrigues b, A.S.F. Farinha b, C.A. Marques a, L. Bilro a, N. Alberto a, J.P.C. Tomé b, R.N. Nogueira a,⇑ a b
Instituto de Telecomunicações, 3810-193 Aveiro, Portugal QOPNA, Department of Chemistry, University of Aveiro, 3810-193 Aveiro, Portugal
a r t i c l e
i n f o
Article history: Received 7 December 2013 Revised 28 March 2014 Available online xxxx Keywords: Sensors Chemical sensors Tilted fiber Bragg gratings Bio-chemical compounds Anions Refractive index
a b s t r a c t The ubiquity and importance of anions in many crucial roles accounts for the current high interest in the design and preparation of effective sensors for these species. Therefore, a tilted fiber Bragg grating sensor was fabricated to investigate individual detection of different anion concentrations in ethyl acetate, namely acetate, fluoride and chloride. The influence of the refractive index on the transmission spectrum of a tilted fiber Bragg grating was determined by developing a new demodulation method. This is based on the calculation of the standard deviation between the cladding modes of the transmission spectrum and its smoothing function. The standard deviation method was used to monitor concentrations of different anions. The sensor resolution obtained for the anion acetate, fluoride and chloride is 79 105 mol/ dm3, 119 105 mol/dm3 and 78 105 mol/dm3, respectively, within the concentration range of (39– 396) 105 mol/dm3. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Considering that anions play an important role in biological and environmental applications, the development of sensors for anion monitoring is of a great interest by the scientific community. Among different anions, acetate (CH3CO 2 ), fluoride (F ) and chloride (Cl) are of extreme importance in various situations. Acetate is present in numerous metabolic processes [1] as well as can be an indicator of organic decomposition in marine sediments [2]. The monitoring of fluoride in water is important to prevent dental caries [3] and its detection can help in clinical treatments such as osteoporosis [4]. With regard to chloride, a small change of the anion flow across the cell membranes can be an indicator of a disease named cystic fibrosis [5]. From this short list, it is notorious that the monitoring of anions is highly desirable. Anions sensing is traditionally performed by host molecules that are designed to recognize a target anion within a given concentration range [6,7]. When the binding between the host and the anion occurs, the host molecule emits a signal which can be ⇑ Corresponding author. Fax: +351 234377901. E-mail addresses: [email protected] (L.B. Melo), [email protected] (J.M.M. Rodrigues), [email protected] (A.S.F. Farinha), [email protected] (C.A. Marques), lucia. [email protected] (L. Bilro), [email protected] (N. Alberto), [email protected] (J.P.C. Tomé), [email protected] (R.N. Nogueira).
either colorimetric [8], fluorescent [9] and/or an electrochemical response [10]. In the last decade, the development of optical fiber sensors for the detection of chemical compounds has become an interesting topic of investigation due to the possibility of producing small sized sensors that are able to remotely and continuously operate in contact with the sample. A number of sensor configurations have been proposed for detection of chemical compounds by monitoring the change of the surrounding refractive index (SRI). These include tapers [11], fiber Bragg gratings (FBGs) [12,13], tilted fiber Bragg gratings (TFBGs) [14], long period gratings (LPGs) [15,16] and surface plasmon resonance (SPR) sensors [17]. Fiber tapering has demonstrated high sensitivity to the SRI; however, these sensors are significantly weaker than unmodified fiber due to the reduction of the fiber diameter [11]. FBGs can also be used to measure SRI by etching the cladding, which leads to a decrease of the fiber mechanical strength [12,13]. LPGs are sensitive to the SRI due to the coupling of radiation from the propagating core mode to the forward-propagating cladding modes [18]. The most prominent drawback of LPGs is the high sensitivity to other parameters such as temperature and strain, which can cause undesirable effects as an SRI sensor. SPR sensors exhibit very high sensitivity to the SRI; however, these sensors normally operate in the visible region which presents higher optical losses compared to the infrared region [17].
http://dx.doi.org/10.1016/j.yofte.2014.05.002 1068-5200/Ó 2014 Elsevier Inc. All rights reserved.
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002
L.B. Melo et al. / Optical Fiber Technology xxx (2014) xxx–xxx
A TFBG consist of a periodic modulation tilted with respect to the longitudinal axis of the fiber core which enables the coupling between the forward-propagating core mode and the forward/ backward-propagating cladding modes [19]. The core and cladding modes appear as a series of multiple dips in the transmission spectrum. The resonance wavelength and the intensity of the dips change as a function of the SRI. Different demodulation methods can be employed to determine the changes of the SRI in a TFBG [20]. Initially, Laffont and Ferdinand reported a technique based on the envelope area of the transmission spectrum [21]. Later, Caucheteur and Mégret improved this method through the correlation between the transmission spectrum and the calculation of two statistical parameters, namely Skewness and Kurtosis [22]. Another widely used method is based on the calculation of the transmission spectrum area [23,24]. In 2007, Chan et al. reported a method in which the wavelength separation between selected cladding modes was calculated [25]. More recently, an alternative demodulation method based on the measurement of the single wavelength time delay has been exploited [26]. In this paper, the use of a TFBG as a concentration sensor for three different anions species namely CH3CO and Cl is 2, F reported. The detection of anions is performed by measuring the changes of the SRI caused by different anion concentrations. For this purpose a new demodulation method based on the calculation of the standard deviation between the transmission spectrum of a TFBG and its smoothing function was developed.
2. Theory of TFBG TFBGs are classified as short period gratings whose refractive index modulation is tilted with respect to the longitudinal axis of the fiber. TFBGs written with a small tilt angle promote the coupling of the forward-propagating core mode to the backwardpropagating cladding modes and decrease the coupling to the backward-propagating core mode [19]. As a consequence, a number of resonances of the cladding modes as well as the resonance of the core mode can be observed in the transmission spectrum (Fig. 1). The resonances can be explained by the phase match condition through the following equations:
kB ¼
2nneff ;core KT cosðhÞ
kiclad ¼
ðneff ;core þ nieff ;clad ÞKT cosðhÞ
ð1Þ
ð2Þ
where neff ;core and nieff ;clad are the effective refractive indices of the core mode and the ith cladding mode, respectively. The nominal grating period is represented by KT and can be expressed as K cosðhÞ; where K is the pitch along the fiber axis and h is the tilt angle. The effective refractive index of the cladding modes depends on the SRI. Therefore, a change of the SRI affects the transmission spectrum of a TFBG. When the grating is immersed in a solution with a small SRI, numerous resonances are visible as a result of the coupling to the backward-propagating cladding modes. As the SRI increases, the cladding modes are gradually cutoff from lower to higher wavelengths and the transmission spectrum gradually approaches its smoothing function. When SRI equals nieff ;clad , the cladding modes are no further guided and a smoothed spectrum is observed.
5
Transmission (dB)
2
&ODGGLQJ0RGHV
0 -5 -10 *KRVW0RGH
-15
&RUH0RGH
-20 -25 1500
1520
1540
1560
Wavelength (nm) Fig. 1. Transmission spectrum of a tilted fiber Bragg grating.
3. Experimental methods A TFBG was fabricated with a 248 nm KrF excimer laser in a standard single mode photosensitive fiber (PS1250/1500, Fibercore), using the phase mask technique. The fiber was previously hydrogen-loaded at 150 bars during two weeks to further enhance its photosensitivity. The phase mask, with a period of 1073.11 nm, was tilted in the perpendicular plane to the laser beam. The grating was written with a total length of 15 mm and an internal tilted angle of 8°. After the inscription, the grating was placed in a heat chamber at 80 °C for 24 h to lose the excess of hydrogen. The experimental setup used for anions monitoring is presented in Fig. 2. The grating was placed in a cylindrical glass container (reactor) and both sides of the fiber were fixed, one at a holder and the other at a manual linear stage. All experiments were performed under a temperature of 23.0 °C controlled by a climatic chamber (Challenge 340, Angelantoni Industrie) with a resolution of 0.1 °C. The manual linear stage was used to keep the fiber taut and the magnet was introduced in the reactor to ensure the homogeneity of the solutions. The transmission spectrum of the TFBG sensor was acquired by an Optical Network Analyzer (ONA) with a wavelength resolution of 4.2 103 nm. The TFBG was characterized to the SRI by using different concentrations of glycerin solutions, providing a refractive index range from 1.341 to 1.437 RIU (refractive index unit). The refractive index values were measured at ambient temperature by a commercial Abbe refractometer (AR4D, Krüss Optronic), at the wavelength of 589 nm and with a resolution of 1 103. The tests were performed by acquiring the transmission spectrum of the TFBG when immersed in each glycerin solutions, starting from the lowest to highest refractive index. After the first set of glycerin
Fig. 2. Experimental setup for anions monitoring.
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002
L.B. Melo et al. / Optical Fiber Technology xxx (2014) xxx–xxx Table 1 Final concentration of the solution in equilibrium after the injection of one anion dose. Anions
Stock solution concentration (mol/ dm3) 104
Injection dose (lL) (±0.5 lL)
Final concentration (mol/dm3) 106 (reactor)
CH3CO 2 F Cl
687 ± 4 300 ± 2 376 ± 3
15.0 34.0 27.0
206 ± 6 204 ± 7 203 ± 7
Table 2 Concentrations and volumes of the stock solutions used to characterize the TFBG sensor. Anions
Stock solution concentration (mol/dm3) 104
Injection dose (lL) (±0.5 lL)
Concentration range in the reactor (mol/dm3) 105
CH3CO 2 F Cl
395 ± 3 300 ± 2 376 ± 3
50.0 66.0 52.0
40–395 40–396 39–391
tests, the fiber was rinsed and two more trials were performed to ensure the reproducibility of the results. The first anion experiment was conducted to investigate the dissolution time of three different anions, namely CH3CO 2, F and Cl (Sigma–Aldrich) in ethyl acetate (analytical reagent grade, Fisher Scientific). Stock solutions of each anion species were prepared by adding the tetrabutylammonium salts of the anions to ethyl acetate (see ‘‘Stock solution concentration’’ on Table 1). The flasks of the stock solutions and the solvent (i.e. ethyl acetate) were placed inside of the climatic chamber under 23.0 °C prior to the beginning of the experiment, and kept in the chamber during the experiment to eliminate thermal fluid variations. The study was performed by initially pouring the reactor with 5 mL of ethyl acetate. To establish a baseline, the transmission spectrum of the TFBG in the presence of ethyl acetate was acquired at the start of the experiment and at the minute 10. Five minutes after the acquisition of the second spectrum, 15 lL of the stock solution of the anion CH3CO 2 (see ‘‘Injection dose’’ on Table 1) was injected into the reactor with a micropipette and the stirrer was turned on during 5 min. After that, the stirrer was turned off and a new TFBG spectrum was acquired. During the next 25 min the acquisition was repeated every 5 min. Afterwards, the reactor was cleaned with ethyl acetate and the same experimental procedure was repeated to measure the time response for the anions F and Cl. The injection dose of each anion was determined so that the concentration of the solution in the reactor after the injection of one anion dose was approximately 200 106 mol/dm3 (see ‘‘Final concentration’’ on Table 1). The second anion experiment was conducted to investigate the performance of the TFBG to different anions concentrations (see ‘‘Concentration range in the reactor’’ on Table 2). For this experiment, new stock solutions of the three anions were prepared (see ‘‘Stock solution concentration’’ on Table 2). Thermal fluid variations were eliminated by using the climatic chamber to keep the solutions at 23.0 °C. The study was performed by initially pouring the reactor with 5 mL of ethyl acetate. After this, a certain dose of the stock solution (see ‘‘Injection dose’’ on Table 2) was introduced into the reactor. The injection dose was calculated in order to increment the concentration of the anion in the reactor by approximately 40 105 mol/dm3. After the anion injection dose, the stirrer was turned on during 5 min followed by a stabilization period of 15 min. The transmission spectrum of the TFBG was acquired 15 min from the time that the stirrer was turned off. The anion injection procedure was repeated giving a total of 10 injections
3
doses. Three trials were performed for each anion species to ensure the reproducibility of the results. 4. Demodulation method In this section, a new demodulation method to analyze the transmission spectrum of a TFBG is presented. This method is based on the calculation of a standard deviation estimator (S), aiming to provide a practical and accurate algorithm to calculate SRI changes. Since the change of the SRI on the TFBG spectrum has only impact on the cladding modes, the wavelength range from 1500.000 nm to 1552.520 nm was selected. This accounts for the transmission spectrum of the grating used in the experiments (Fig. 1). The acquired transmission spectra of the TFBG immersed in different glycerin solutions were converted to a linear scale (Fig. 3). The four spectra of Fig. 3 show that the amplitude of the cladding modes decreases as the SRI increases. In the Fig. 3(d), the SRI is close to the refractive index of the fiber cladding and as a consequence, most of the cladding modes are no longer guided. In this situation the coupling is continuously done by the radiation modes. For each acquired spectrum, a smoothing function was calculated by the Savitzky–Golay method [27]. This consists of a polynomial regression to the data points in order to determine a smoothed value. This numerical tool enables the calculation of a set of points that replicate the trend of the experimental data. The smoothing of the transmission spectra is represented by the dashed red line in Fig. 3. After the calculation of the smoothing function for each spectrum corresponding to a different glycerin solution, a standard deviation analysis was applied. This common estimator is defined as:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u N u 1 X S¼t ðxi xi Þ2 N 1 i¼1
ð3Þ
where, N is the total number of points that constitutes the transmission spectra, xi corresponds to the ith values of the transmission spectrum expressed in the linear scale and xi corresponds to the ith values of the smoothing function. The performance of the standard deviation method was compared with two other intensity-based methods, namely the calculation of the area of the transmission spectrum, and the calculation of the envelope’s area of the transmission spectrum. Fig. 4 shows the SRI characterization of the TFBG analyzed by the three demodulation methods. The Y error bars represent the error determined for the three trials. Considering that the SRI range of interest in this work is around the refractive index of ethyl acetate (i.e. 1.370), the sensitivity of each method was calculated between the two experimental values in the vicinity of 1.370. For the SRI range between 1.364 and 1.381, the sensitivity of the standard deviation, area and envelope is 8.5/ SRI, 6.6/SRI and 7.0/SRI, respectively. Fig. 4 shows that for a wide range of SRI (i.e. between 1.341 and 1.437), the standard deviation offers higher sensitivity than the area. For the same range of SRI, the performance of the standard deviation is similar to the envelope. Nevertheless, the standard deviation does not require the calculation of the peaks of the transmission spectrum whereas the envelope requires this calculation in order to determine the upper and lower envelopes of the TFBG spectrum. Therefore, the standard deviation might become extremely useful when the grating spectrum presents a low signal-to-noise ratio. In this case, the detection of peaks of the transmission spectrum is difficult to implement, which makes
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002
4
L.B. Melo et al. / Optical Fiber Technology xxx (2014) xxx–xxx
1.0
Refractive index: 1.345 Smoothing
1.0 0.8
Transmission
Transmission
0.8 0.6 0.4 0.2 0.0
0.6 0.4 0.2 0.0
1500 1510 1520 1530 1540 1550 1560
1.0
1500 1510 1520 1530 1540 1550 1560
Wavelength (nm)
Wavelength (nm)
(a)
(b)
Refractive index: 1.410 Smoothing
1.0
Transmission
0.8
Transmission
Refractive index: 1.381 Smoothing
0.6 0.4 0.2
Refractive index: 1.437 Smoothing
0.8 0.6 0.4 0.2
0.0 1500 1510 1520 1530 1540 1550 1560
0.0
1500 1510 1520 1530 1540 1550 1560
Wavelength (nm)
Wavelength (nm)
(c)
(d)
Normalized S and Area
Fig. 3. Transmission spectrum and the calculated smoothing function for different surrounding refractive indices: (a) 1.345; (b) 1.381; (c) 1.410; and (d) 1.437.
1.0 0.8 0.6 0.4 0.2 0.0
Standard deviation method Area method Envelope method 1.340 1.360 1.380 1.400 1.420 1.440
Surrounding refractive index Fig. 4. Comparison between the area, the envelope and the standard deviation methods as regards to the change of the surrounding refractive index.
the use of the envelope more complex than the standard deviation. It is worthwhile noting that only intensity-based methods are compared in this article. Another demodulation option that offers high signal-to-noise ratio is the method that used the wavelength separation between a selected cladding mode and the Bragg mode [25].
5. Experimental results and discussion This section presents the results of the TFBG as a sensor for anions monitoring using the standard deviation method presented in Section 4.
The dissolution time of the anions CH3CO and Cl was 2, F investigated by monitoring the variation of the standard deviation caused by an injection dose (Fig. 5). The experimental values depicted in Fig. 5 were normalized by the value of the standard deviation of the grating when immersed in ethyl acetate. Fig. 5 shows that the value of the standard deviation remains constant in the first 10 min corresponding to the detection of ethyl acetate. The major change of the standard deviation occurs 5 min after the anion injection dose. The graph also shows that the equilibrium is obtained 15 min after the time that the stirrer was turn off. This behavior was observed for the three anions under study. Fig. 6 shows the variation of standard deviation as a function of anions concentration. Each point corresponds to an anion injection dose. The standard deviation value of each injection dose was normalized to the value calculated for the first injection, i.e., relative to the concentration of 40 105 mol/dm3. The error bars represent the deviation (r) associated with the three trials performed for each anion. Fig. 6 shows that the standard deviation decreases with the increase of the anion concentration. The variation of the standard deviation between the first and last injection dose of the anions CH3CO 2 , F and Cl is 0.0359, 0.0421 and 0.0502, respectively. By using the sensitivity value of the TFBG calculated in Section 4 (i.e. 8.5/SRI), the variations of the standard deviation correspond to an SRI variation of 0.0042 RIU (CH3CO 2 ), 0.0050 RIU (F ) and 0.0059 RIU (Cl ). The third-degree polynomial fitting was found to be the best model to describe the sensor behavior. The coefficients of each fit ting equation for CH3CO 2 , F and Cl are presented in the Table 3. It can be observed in Fig. 6 that each anion species has a different characterization curve based on the mean values of each trial.
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002
5
L.B. Melo et al. / Optical Fiber Technology xxx (2014) xxx–xxx
&+&2
1.00
)
0.99
&O
0.99
Normalized S
Normalized S
1.00
-
,QMHFWLRQGRVH 6WLUUHURQ 6WLUUHURII
0.98
(TXLOLEULXPWLPH
0.97 0.96
CH3 CO2 -
F Cl
-
0.98 0.97 0.96 0.95
0
10
20
30
40
50
0.0
Time (min)
-3
1.0x10
-3
2.0x10
-3
3.0x10
-3
4.0x10
Concentration (mol/dm3)
Fig. 5. The dissolution of the anions in ethyl acetate. Fig. 7. Sensor resolution for each anion under study.
6. Conclusions and outlook
Normalized S
1.00 0.99 0.98 0.97
&+&2
0.96
)
0.95
&O 7KLUGGHJUHHSRO\QRPLDOILWWLQJV
0.94 0.0
-3
1.0x10
-3
2.0x10
-3
-3
3.0x10
4.0x10
Concentration (mol/dm3) Fig. 6. Variation of the normalized standard deviation with the change of anion concentrations.
Table 3 Third-degree polynomial fitting coefficients in the form S = a + bx + cx2 + dx3 for the anions under study. R2 represents the determination coefficient associated with each anion concentration fitting.
CH3CO 2 F Cl
a (104)
b
c (102)
d (105)
R2
10,077 ± 4 10,102 ± 3 10,133 ± 3
21 ± 1 28 ± 1 37 ± 1
40 ± 8 69 ± 7 80 ± 8
4 ± 1 8 ± 1 7 ± 2
0.9997 0.9997 0.9998
However, the error bars associated to each experimental point suggest that the TFBG is only able to discriminate between the anions CH3CO 2 and Cl and between the anions F and Cl . The step of each concentration point depicted in Fig. 6 corresponds to an incremental injection dose of 40 105 mol/dm3. However, the value of the error bars of the three trials for each anion suggests that the increment of the anion injection dose is lower than the realistic sensor resolution that can be achieved by the TFBG proposed here. Therefore, the sensor resolution for each anion species was investigated by selecting the concentration increment that corresponds to a higher variation of the standard deviation between two consecutive values than the error bars of the selected experimental points. Fig. 7 shows the final selected concentration points for each anion species, which correspond to a resolution of 79 105 mol/dm3, 119 105 mol/dm3 and 78 105 mol/dm3 for the CH3CO 2 , F and Cl , respectively.
A preliminary study of an optical fiber sensor for detection of anions is reported. In this article, each type of anion was investigated individually. However, a real environment is generally characterized by a medium with complex solutions (i.e. different types of anions in solution). Therefore, the selectivity of TFBG sensors must be investigated to allow utilization of these sensors in practical applications. One possibility to achieve selectivity to a target anion is by attaching specific chemical compounds to the surface of an optical fiber by a process so-called functionalization [3,6]. The chemical compounds are fabricated to exhibit affinity for capturing target anions. Fortunately, there is precedent for optical fiber functionalization [13,14,28]. Examples of chemical compounds in the literature that show selectivity to the investigated anions include: sensors based on quinonehydrazone group for selectivity to the anion CH3CO 2 [29], sensors based on derivatives of meso-octamethylcalix[4]pyrrole for selectivity to the anion F [3], and sensors based on derivatives of calix[4]arene for selectivity to the anion Cl [30]. The authors are currently developing chemical compounds that can selectively capture specific anions [3,31]. These chemical compounds will be covalently attached to the surface of the optical fiber in the region of the TFBG. The selectivity of functionalized TFBG sensors will be analyzed in a future work. Additionally, a new demodulation method was developed in order to quantify the SRI changes in the transmission spectrum of the TFBG. This method is based on the calculation of a statistical parameter, namely standard deviation estimator, using the smoothed transmission spectrum as the reference. A sensor resolution of 79 105 mol/dm3, 119 105 mol/dm3 and 78 105 mol/dm3 was achieved for the anions CH3CO and Cl, 2, F respectively. Acknowledgments The authors gratefully acknowledge funding provided by European Regional Development Fund (ERDF) and Science and Technology Foundation, Ministry of Education and Science (FCT, Portugal). In particular, the financial support provided by the Projects PTDC/ CTM/101538/2008 and PTDC/EEA/122792/2010. The authors João Rodrigues, Andreia Farinha, Carlos Marques, Lúcia Bilro and Nélia Alberto acknowledge their doctoral (BD) and post-doctoral (BPD) support provided by FCT, namely by the fellowships SFRH/BD/ 81014/2011, SFRH/BPD/73060/2010, SFRH/BD/70661/2010, SFRH/ BPD/78205/2011 and SFRH/BPD/78141/2011.
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002
6
L.B. Melo et al. / Optical Fiber Technology xxx (2014) xxx–xxx
References [1] G. Liu, J. Shao, Fluorescence anion sensors based on combination of conformational restriction and photo-induced electron transfer, J. Lumin. 135 (2013) 312–317. [2] H. Maeda, Y. Ito, BF2 complex of fluorinated dipyrrolyldiketone: a new class of efficient receptor for acetate anions, Inorg. Chem. 45 (2006) 8205–8210. [3] N.I.P. Valente, P.V. Muteto, A.S.F. Farinha, A.C. Tomé, J.A.B.P. Oliveira, M.T.S.R. Gomes, An acoustic wave sensor for the hydrophilic fluoride, Sens. Actuators, B 157 (2011) 594–599. [4] D. Sharma, S.K. Sahoo, R.K. Bera, R. Kamal, Spectroscopic and computational study of a naphthalene derivative as colorimetric and fluorescent sensor for bioactive anions, J. Fluoresc. 23 (2013) 387–392. [5] N. Busschaert, P.A. Gale, Small-molecule lipid-bilayer anion transporters for biological applications, Angew. Chem. Int. Ed. 52 (2013) 1374–1382. [6] L.E. Santos-Figueroa, M.E. Moragues, E. Climent, A. Agostini, R. MartínezMáñez, F. Sancenón, Chromogenic and fluorogenic chemosensors and reagents for anions. A comprehensive review of the years 2010–2011, Chem. Soc. Rev. 42 (2013) 3489–3613. [7] G.T. Spence, P.D. Beer, Expanding the scope of the anion templated synthesis of interlocked structures, Acc. Chem. Res. 46 (2013) 571–586. [8] E.M. Boyle, S. Comby, J.K. Molloy, Thorfinnur Gunnlaugsson, Thiourea derived Tröger’s bases as molecular cleft receptors and colorimetric sensors for anions, J. Org. Chem. 78 (2013) 8312–8319. [9] R.M. Duke, E.B. Veale, F.M. Pfeffer, P.E. Kruger, T. Gunnlaugsson, Colorimetric and fluorescent anion sensors: an overview of recent developments in the use of 1,8-naphthalimide-based chemosensors, Chem. Soc. Rev. 39 (2010) 3936– 3953. [10] N.H. Evans, H. Rahman, J.J. Davis, P.D. Beer, Surface-attached sensors for cation and anion recognition, Anal. Bioanal. Chem. 402 (2012) 1739–1748. [11] Y. Tian, W. Wang, N. Wu, X. Zou, X. Wang, Tapered optical fiber sensor for label-free detection of biomolecules, Sensors 11 (2011) 3780–3790. [12] S.S. Saini, C. Stanford, S.M. Lee, J. Park, P. DeShong, W.E. Bentley, M. Dagenais, Monolayer detection of biochemical agents using etched-core fiber Bragg grating sensors, IEEE Photonics Technol. Lett. 19 (2007) 1341–1343. [13] G. Ryu, M. Dagenais, M.T. Hurley, P. DeShong, High specificity binding of lectins to carbohydrate-functionalized fiber Bragg gratings: a new model for biosensing applications, IEEE J. Sel. Top. Quantum 16 (2010) 647–653. [14] S. Maguis, G. Laffont, P. Ferdinand, B. Carbonnier, K. Kham, T. Mekhalif, M.C. Millot, Biofunctionalized tilted fiber Bragg gratings for label-free immunosensing, Opt. Express 16 (2008) 19049–19062. [15] P. Pilla, V. Malachovská, A. Borriello, A. Buosciolo, M. Giordano, L. Ambrosio, A. Cutolo, A. Cusano, Transition mode long period grating biosensor with functional multilayer coatings, Opt. Express 19 (2011) 512–526. [16] P.F. Manzillo, P. Pilla, A. Buosciolo, S. Campopiano, A. Cutolo, A. Borriello, M. Giordano, A. Cusano, Self assembling and coordination of water nano-layers on
[17]
[18]
[19] [20] [21]
[22] [23]
[24]
[25]
[26]
[27] [28]
[29]
[30]
[31]
polymer coated long period gratings: toward new perspectives for cation detection, Soft Mater. 9 (2011) 238–263. S.K. Srivastava, B.D. Gupta, Influence of ions on the surface plasmon resonance spectrum of a fiber optic refractive index sensor, Sens. Actuators, B 156 (2011) 559–562. V. Bhatia, D.K. Campbell, T.D. Alberto, G.A.T. Eyck, D. Sherr, K. Murphy, R.O. Claus, Standard optical fiber long-period gratings with reduced temperature sensitivity for strain and refractive-index sensing, Proc. IEEE OFC’97, Dallas, TX, USA, 1997, pp. 346–347. T. Erdogan, J.E. Sipe, Tilted fiber phase gratings, J. Opt. Soc. Am. A 13 (1996) 296–313. J. Albert, L.Y. Shao, C. Caucheteur, Tilted fiber Bragg grating sensors, Laser Photonics Rev. 7 (2013) 83–108. G. Laffont, P. Ferdinand, Tilted short-period fiber-Bragg-grating-induced coupling to cladding modes for accurate refractometry, Meas. Sci. Technol. 12 (2001) 765–770. C. Caucheteur, P. Megret, Demodulation technique for weakly tilted fiber Bragg grating refractometer, IEEE Photonics Technol. Lett. 17 (2005) 2703–2705. N.J. Alberto, C.A. Marques, J.L. Pinto, R.N. Nogueira, Three-parameter optical fiber sensor based on a tilted fiber Bragg grating, Appl. Opt. 49 (2010) 6085– 6091. Y.P. Miao, B. Liu, Q. Zhao, Refractive index sensor based on measuring the transmission power of tilted fiber Bragg grating, Opt. Fiber Technol. 15 (2009) 233–236. C.F. Chan, C. Chen, A. Jafari, A. Laronche, D.J. Thomson, J. Albert, Optical fiber refractometer using narrowband cladding-mode resonance shifts, Appl. Opt. 46 (2007) 1142–1149. M. Pisco, A. Ricciardi, S. Campopiano, C. Caucheteur, P. Mégret, Time delay measurements as promising technique for tilted fiber Bragg grating sensors interrogation, IEEE Photonics Technol. Lett. 21 (2009) 1752–1754. A. Savitzky, M.J.E. Golay, Smoothing and differentiation of data by simplified least squares, Anal. Chem. 36 (1964) 1627–1639. J. Barnes, M. Dreher, K. Plett, R.S. Brown, C.M. Crudden, H.P. Loock, Chemical sensor based on a long-period fibre grating modified by a functionalized polydimethylsiloxane coating, Analyst 133 (2008) 1541–1549. S. Hu, Y. Guo, J. Xu, S. Shao, A selective chromogenic molecular sensor for acetate anions in a mixed acetonitrile–water medium, Org. Biomol. Chem. 6 (2008) 2071–2075. J.N. Babu, V. Bhalla, M. Kumar, R.K. Mahajan, R.K. Puri, A chloride selective sensor based on a calix[4]arene possessing a urea moiety, Tetrahedron Lett. 49 (2008) 2772–2775. J.M.M. Rodrigues, A.S.F. Farinha, P.V. Muteto, S.M. Woranovicz-Barreira, F.A.A. Paz, M.G.P.M.S. Neves, J.A.S. Cavaleiro, A.C. Tome, M.T.S.R. Gomes, J.L. Sessler, J.P.C. Tome, New porphyrin derivatives for phosphate anion sensing in both organic and aqueous media, Chem. Commun. 50 (2014) 1359–1361.
Please cite this article in press as: L.B. Melo et al., Concentration sensor based on a tilted fiber Bragg grating for anions monitoring, Opt. Fiber Technol. (2014), http://dx.doi.org/10.1016/j.yofte.2014.05.002