Open Access
Add to BibSonomy
Abstract
The rate of vapor condensation on a solid surface depends on the ambient relative humidity (RH). Also, surface plasmon resonance (SPR) on a metal layer is sensitive to the refractive index change of its adjacent dielectric. The SPR effect appears as soon as a small amount of moisture forms on the sensor, resulting in a decrease in the amount of light transmitted due to plasmonic loss. Using this concept, we developed a fiber optic humidity sensor based on SPR. It can measure the ambient RH over a dynamic range from 10% to 85% with an accuracy of 3%.
© 2017 Optical Society of America
1. Introduction
Various types of hygrometers, including mechanical hygrometers, wet and dry bulb psychrometers, infrared optical absorption hygrometers, and optical waveguides, can operate as humidity sensors [1, 2]. Although optical humidity sensors have advantages such as immunity to electromagnetic interference and fast responses, electrical humidity sensors remain more commonly used since optical hygrometers and humidity sensors do not satisfy most requirements in terms of cost, ease of operation, maintenance, and remote operability. In contrast, fiber optic humidity sensors offer features such as small sizes, multiplexing, and remote sensing capabilities. Hence, numerous investigations have been conducted on the use of fiber optic sensors for humidity and moisture measurements since the beginning of this millennium [3–5]. In general, these sensors utilize direct spectroscopic [6], evanescent wave [7], in-fiber grating [8,9], interferometric, or hybrid [10–12] methods to measure humidity. In this paper, we propose a surface plasmon resonance (SPR)-based fiber optic humidity sensor using water vapor condensation on solid surfaces.
The SPR effect results in the high sensitivity to the dielectric adjacent to a metal layer to refractive index changes and has numerous applications, from biochemical sensing to environmental monitoring [13–19]. The previously proposed plasmonic fiber optic humidity sensors work based on the SPR concept in which a plasmonic peak in the loss spectrum is deformed or displaced because of refractive index changes in the adjacent dielectric due to ambient relative humidity (RH) variations [20–23]. In these kinds of sensors, the difference between the plasmonic peaks is measured. In contrast, the characteristics of the plasmonic peaks are assessed in our technique, thereby, the maximum capacity of the difference is earned.
In our measurement method, the SPR effect appears as soon as a small amount of moisture forms on the fiber optic sensor, resulting in a decrease in the amount of light transmitted due to plasmonic loss. We discovered that the rate of this decrease depends on and enables easy measurement of the ambient humidity.
The paper is organized as follows. Section 2 begins with a brief discussion of the sensitivity of surface plasmon waves to moisture formation on top of a gold layer, which is followed by a theoretical description of the dependences of vapor condensation on the ambient humidity and temperature. In section 3, we discuss how ambient humidity is sensed in our innovative method and present the measurement results obtained using our proposed sensor. Finally, section 4 describes our conclusions.
2. Theory
2.1 Moisture detection by the SPR probe
Surface plasmon waves are surface electromagnetic waves that propagate parallel to a metal/dielectric interface. They are excited by an electromagnetic field perpendicular to the metal surface at the phase-matching point between an electron wave and an optical wave. In a typical SPR fiber optic sensor, this phase-matching occurs when the propagation constant of a guided mode in the fiber is equal to that of the surface plasmon wave on top of the metal layer coated on the surface of the fiber [20]. The tail of the field of each plasmon wave diminishes gradually in the dielectric, indicating how far the plasmon field penetrates into the dielectric and serving as an SPR depth probe. As we found previously [24], SPR can be used to probe depths of up to 1 μm when the operating wavelength is in the visible wavelength range. Thus, SPR enables the sensing of dielectric index changes very close to the metal/dielectric boundary.
In our fiber optic sensor, condensation and evaporation occur near the gold layer coated on the side-polished surface, and plasmon waves can be excited by the light transmitted into the sensor. The condensation process in our sensor is illustrated schematically in Fig. 1(a), which depicts surface plasmon waves excited by the fundamental mode of the guided light and an approximately 1-μm-thick layer of water formed on top of a 50-nm-thick gold layer due to vapor condensation. Since SPR sensors are very sensitive to tiny changes in the water layer thickness, it is possible to measure small amounts of water vapor condensation in the sensing area by measuring the plasmonic loss (Iin-Iout).
Fig. 1 (a) Side view of the fiber optic humidity sensor, where the SPR probe senses moisture that condenses on top of the gold layer coated on the side-polished surface. (b) Top view microscopic image of the sensing area in our sensor, where the surface is wet due to the drop-wise condensation of water molecules in the adjacent air.
Figure 1(b) presents a top view of the sensing area in our sensor, where the surface is wet due to the drop-wise condensation of water molecules in the adjacent air. As shown, the water droplets range in size from a few hundred nanometers to a few microns, implying that the thickness of the water layer on the surface of the sensing area should be on the order of micron or less. Hence, our SPR-based fiber optic sensor enables the immediate detection of moisture formation and its thickness variations. In the following section, we describe how the vapor condensation rate depends on the ambient RH, enabling the sensor to measure the ambient RH.
2.2 Dependence of vapor condensation rate on ambient humidity
It is well known that moisture forms more slowly on a solid surface as the surrounding environment becomes warmer and dryer, as verified experimentally and by thermodynamics. Hence, the rate of water vapor condensation on a solid surface should depend on the ambient RH. During condensation, the free energy of the water molecules from the air is absorbed by the solid surface because of the phase shift from gas to liquid. Hence, the condensation rate on the surface ṁ can be calculated using the following equation, which was previously presented by Incropera et al. [25]:
where heff is the effective heat transfer coefficient, As is the surface area exposed to the ambient water vapor, Tdew is the ambient dew point temperature, Ts is the surface (or sensor) temperature, and hfg is the specific enthalpy of the saturated vapor. The Magnus formula, Eq. (2), can be used to calculate the dew point temperature Tdew based on the known temperature T and the RH of ambient air adjacent to the surface [26]:Substituting Eq. (2) into Eq. (1) to obtain Eq. (3) demonstrates that the condensation rate depends on the ambient RH and temperature:Figure 2(a), which was generated using Eq. (3), illustrates the dependence of the water vapor condensation rate on the ambient RH when the ambient temperature is 10°C, 20°C, 30°C, 40°C, and 50°C. As shown, the higher the ambient RH, the higher the condensation rate. We employed Eq. (3) again while varying the ambient temperature and leaving the ambient RH fixed to obtain the results presented in Fig. 2(b), which illustrates the dependence of the condensation rate on the ambient temperature when the ambient RH is 30%, 40% and 50%. As shown, the higher the temperature, the higher the condensation rate when the ambient RH is constant. This finding implies that air contains more evaporated water molecules, i.e., has higher absolute humidity, at higher temperatures. In the following sections, we demonstrate how this relationship between condensation rate and ambient RH, i.e., Eq. (3), can be employed to obtain the ambient RH.
Fig. 2 (a) Theoretical dependence of water vapor condensation rate on ambient RH, obtained using Eq. (3), with fixed ambient temperatures of 10°C, 20°C, 30°C, 40°C, and 50°C. (b) Dependence of water vapor condensation rate on ambient temperature with ambient RHs of 30%, 40%, and 50%.
3. Results and discussion
3.1 Measurement setup
Our SPR fiber optic sensor uses a standard multimode fiber (62/125 μm), where up to a 15 mm length the fiber is side-polished, with a remaining cladding thickness of 1–2 μm. We have higher guided modes in the multi-mode fiber causing modal loss or partial reduction of the SPR effect. If we use single mode fiber instead of multi-mode fiber, the SPR effect is even enhanced. Nevertheless, multi-mode fiber has been chosen because of its high numerical aperture helping to easily couple the light from a simple light-emitting diode (LED) into a fiber optic channel leading to a cost-effective device. Also, to reach an optimum SPR sensing performance, a 45-nm-thick layer of gold is coated on the polished area by sputtering method with depositing rate of 0.3 A/sec by plasma power of 5 Watts and chamber’s vacuum pressure of 1.2*10−6 Torr at room temperature.
To study the humidity response of the sensor, we developed the setup depicted in Fig. 3, which included a light source (a red color LED), a sensor, an environmental chamber, a control and data acquisition (DAQ) module, and some electro-optic devices. The environmental chamber was equipped with the required environmental sensors, which were used to obtain reference measurements, and a conditioning module consisting of a heater, an ultrasonic air humidifier, an air dryer, and a thermoelectric cooler.
Fig. 3 Scheme of our experimental setup including a light source (red LED), a sensor, an environmental chamber, and a control and DAQ module. The conditioning module consisted of a heater, an ultrasonic air humidifier, an air dryer, and a thermoelectric cooler.
The first experiment was performed at room temperature and atmospheric pressure but with different ambient RHs. To determine Tdew, Ts was reduced from the ambient temperature until a thin layer of water formed on top of the gold layer, as illustrated in Fig. 1(b), indicating that Tdew had been reached. The physical effect of this moisture formation was that the refractive index of the adjacent dielectric medium changed from 1 to approximately 1.33, causing SPR excitation and, consequently, loss of the transmitted light.
3.2 Real-time measurements
By monitoring the transmitted light intensity, water layer thickness changes on the sub-micron scale can be detected. Figure 4 shows the real-time measurements obtained while increasing the ambient RH from 10% to 80%, as measured by a reference electrical humidity sensor (iTHXD-OMEGA Inc.) and shown by the dotted-dashed green curve in the diagram. In addition, Ts was decreased to Tdew and is represented by the dotted black curve, while T was fixed at 23°C. The dramatic decreases in the transmitted light intensity, represented by the solid blue curve, occur when Ts reached Tdew, at which point condensation began. By using the known values of Tdew and T and employing Eq. (2), the ambient RHs could be obtained.
Fig. 4 Real-time measurements obtained inside the environmental chamber: ambient RH measured by the reference humidity sensor (dotted-dashed green curve); Ts (dotted black curve), ambient temperature (double-dashed orange curve), transmitted light signal (solid blue curve), and Tdew (dashed red curve).
The RHs acquired using this method were then compared with those measured by the reference humidity sensor installed inside the environmental chamber. The results of this comparison are presented in Fig. 5, where the estimated ambient RHs are represented by the red circles and the measured reference ambient RHs are indicated by the black squares, with a standard error of 1.3% throughout the depicted RH range. To enable local comparisons to be made within the RH range, the normalized error is also shown (thin blue curve). Based on the statistical definition, normalized errors between −1 and + 1 indicate acceptable accuracy. Thus, high accuracy is evident, especially in the RH range from 10% to 50%. Although the exact signal-to-noise ratio of our sensor is not presented here, Figs. 4, 6, and 9 clearly indicate that it is high since the change in the transmitted light intensity due to condensation (signal) is significantly greater than the light intensity fluctuations (noise).
Fig. 5 Comparison between estimated (filled-in red circles) and reference (black squares) RHs inside the chamber.
To perform a more thorough investigation, we separately considered the transmitted light variation again in the time interval from 1300 s to 1400 s, which is depicted in Fig. 6, in the case in which the ambient RH and temperature were 32% and 23°C, respectively, and Ts was reduced to TD = 4.8°C. As Fig. 6 demonstrates, the transmitted light intensity decreases rapidly due to the SPR loss during condensation and then increases back to its initial value due to evaporation when Ts ˃ TD. This diagram contains valuable information about the ambient humidity; by extracting this information, the RH of the environment in which the sensor is placed can be obtained, as explained in Section 3.3.
Fig. 6 Real-time measurements obtained by the sensor inside the environmental chamber for the interval time from 1300 s to 1400 s.
3.3 Dependence of transmission reduction rate on ambient RH
It is obvious that slope of the transmitted light intensity as it decreases is directly related to the water vapor condensation rate. Furthermore, the condensation rate is related to the ambient RH through Eq. (3). Thus, the transmission reduction rate should be related to the ambient RH. This point was verified by investigating the decreases in transmitted light intensity with different ambient RHs. Our experimental measurements, which are presented in Fig. 7(a), clearly demonstrate that the transmitted light intensity decreases faster when the ambient RH is higher, as suggested by the theoretically obtained diagram in Fig. 2. Figure 7(b) depicts the relationship between the transmission reduction rate and ambient RH during moisture formation at two slightly different ambient temperatures of 23°C and 25°C showing a similar trend. Hence, the light transmitted into the sensor offers a powerful tool for measuring ambient humidity. To have an estimation for the RH measurement speed of the sensor, we consider the time interval in which the transmitted light intensity drops by 1%. As Fig. 7(a) shows, the waiting time to see such a drop is between 2 to 30 seconds, depending on the ambient RH of 80% to 15%.
Fig. 7 (a) Transmitted light intensity reduction during moisture formation, which demonstrates that the slope, i.e., the condensation rate, increases significantly as the ambient RH increases. (b) Dependence of transmission reduction rate on the ambient RH at two different ambient temperatures.
In addition, Fig. 2(b) suggested that the condensation rate was related to the ambient temperature. We investigated this point in greater detail by measuring the vapor condensation rate at different ambient temperatures while leaving the ambient RH constant. Figure 8 shows the transmitted light intensity at temperatures of 10°C, 20°C, 30°C, and 40°C when the ambient RH was fixed around 30%. Clearly, the transmitted light intensity decreases more rapidly when the ambient temperature is higher. The inset in Fig. 8 also demonstrates that the relationship between the transmission reduction rate and the ambient temperature is approximately linear, as theoretically suggested by Fig. 2(b).
Fig. 8 Transmitted light intensity during moisture formation at temperatures of 10°C, 20°C, 30°C, and 40°C when the ambient RH was fixed near 30%. The inset shows an approximately linear relationship between the transmission reduction rate and the ambient temperature.
3.4 Sensor response to breath exposure
Finally, we analyzed the response of our proposed sensor to breath exposure. The results, which were obtained using an ambient RH and a temperature of 43% and 22°C, respectively, are presented in Fig. 9. To perform a more thorough evaluation, we also obtained the response of the commercial reference sensor, whose probe was adhered to that of the proposed sensor. We simultaneously brought the probes of our fiber optic sensor and the commercial one close to the mouth of a person for breath exposure. Since some humid and warm air is released from the mouth while exhaling, some moisture formed on the surface of each sensor due to condensation and then evaporated into the ambient air. This phenomenon is typically revealed by transmitted light intensity fluctuations. Figure 9 demonstrates that our proposed fiber optic sensor can respond as much as 8 s faster than the commercial one to the same breath exposure.
Fig. 9 Response of the proposed sensor to human breath exposure when placed at different distances from the mouth of a person. The dashed black curve represents the response of a commercial humidity sensor when used in the same conditions as the proposed sensor.
It can also be deduced that our proposed optical sensor has a better reversibility than the commercial sensor, because its peaks are narrower. Furthermore, Fig. 9 demonstrates that the response of our sensor is proportional to amount of moisture released from the mouth, which depends on the distance of the probe from the mouth, implying the accuracy of the proposed fiber optic humidity sensor.
Finally, we emphasized here that dry dirties such as dust and soil will not disturb the function of the proposed humidity sensor. The partial problem will appear if some soluble materials such as salt or sugar accumulate on the sensing area after water drying. Regarding chemical durability, there is no problem for the sensor to work for a long time, as long as 1-2 years. This is because the sensing layer metal is gold, which is immune to rust and corrosion, and the body of the transducer is silica which is chemically durable.
4. Conclusion
In this article, we presented an SPR-based fiber optic sensor for performing ambient RH measurements using water vapor condensation. The experimental results agreed well with the theory and indicate the effectiveness of our proposed idea, because the ambient RH could be determined with a standard error of only 1.3%. The obtained results demonstrate the good performance of our fiber optic humidity sensor in terms of wide operating range, response speed, precision, repeatability, reversibility, working in the turbulent conditions. The performance limitation of our sensor is when its surface is made too dirty, so that a thin homogeneous dielectric layer is formed on the gold surface which may disturbing sensor’s function. As a scope of future works, there is developing a multi-channel fiber optic sensor for simultaneously monitoring of moisture in different locations, and improving the sensor performance via making its surface super hydrophilic.
Funding
Institut National de la Recherche Scientifique - Énergie, Matériaux et Télécommunications (INRS-EMT); Canada Foundation for Innovation – Major Science Initiatives; NSERC discovery grant; NSERC CREATE Integrated Sensor Systems (ISS) – grant number 371305-2010.
Acknowledgments
We would like to thank Guy Lebrun for technical assistance in this project.
References and links
1. C. Y. Lee and G. B. Lee, “Humidity sensors: A review,” Sens. Lett. 3(1), 1–14 (2005). [CrossRef]
2. Z. M. Rittersma, “Recent achievements in miniaturized humidity sensors—a review of transduction techniques,” Sens. Actuators A Phys. 96(3), 196–210 (2002). [CrossRef]
3. T. L. Yeo, T. Sun, and K. T. V. Grattan, “Fibre-optic sensor technologies for humidity and moisture measurement,” Sens. Actuators A Phys. 144(2), 280–295 (2008). [CrossRef]
4. L. Alwis, T. Sun, and K. T. V. Grattan, “Optical fibre-based sensor technology for humidity and moisture measurement: Review of recent progress,” Measurement 46(10), 4052–4074 (2013). [CrossRef]
5. S. A. Kolpakov, N. T. Gordon, C. Mou, and K. Zhou, “Toward a new generation of photonic humidity sensors,” Sensors (Basel) 14(3), 3986–4013 (2014). [CrossRef] [PubMed]
6. M. Bedoya, M. T. Diez, M. C. Moreno-Bondi, and G. Orellana, “Humidity sensing with a luminescent Ru(II) complex and phase-sensitive detection,” Sens. Actuators B Chem. 113(2), 573–581 (2006). [CrossRef]
7. Y. Liu, Y. Zhang, H. Lei, J. Song, H. Chen, and B. Li, “Growth of well-arrayed ZnO nanorods on thinned silica fiber and application for humidity sensing,” Opt. Express 20(17), 19404–19411 (2012). [CrossRef] [PubMed]
8. S. F. H. Correia, P. Antunes, E. Pecoraro, P. P. Lima, H. Varum, L. D. Carlos, R. A. S. Ferreira, and P. S. André, “Optical fiber relative humidity sensor based on a FBG with a di-ureasil coating,” Sensors (Basel) 12(7), 8847–8860 (2012). [CrossRef] [PubMed]
9. T. Venugopalan, T. Sun, and K. T. V. Grattan, “Long period grating-based humidity sensor for potential structural health monitoring,” Sens. Actuators A Phys. 148(1), 57–62 (2008). [CrossRef]
10. M. Consales, A. Buosciolo, A. Cutolo, G. Breglio, A. Irace, S. Buontempo, P. Petagna, M. Giordano, and A. Cusano, “Fiber optic humidity sensors for high-energy physics applications at CERN,” Sens. Actuators B Chem. 159(1), 66–74 (2011). [CrossRef]
11. J. Mathew, Y. Semenova, and G. Farrell, “Fiber optic hybrid device for simultaneous measurement of humidity and temperature,” IEEE Sensors 13(5), 1632–1636 (2013). [CrossRef]
12. J. Mathew, Y. Semenova, and G. Farrell, “Photonic crystal fiber interferometer for dew detection,” J. Lightwave Technol. 30(8), 1150–1155 (2012). [CrossRef]
13. H. Esmaeilzadeh, M. Rivard, E. Arzi, F. Légaré, and A. Hassani, “A super continuum characterized high-precision SPR fiber optic sensor for refractometry,” Sens. Actuators A Phys. 229, 8–14 (2015). [CrossRef]
14. M. H. Yang, H. L. Liu, D. H. Zhang, and X. L. Tong, “Hydrogen sensing performance comparison of Pd layer and Pd/WO 3composite thin film coated on side-polished single- and multimode fibers,” Sens. Actuators B Chem. 149(1), 161–164 (2010). [CrossRef]
15. J. Senosiain, I. Díaz, A. Gastón, and J. Sevilla, “High sensitivity temperature sensor based on side-polished optical fiber,” IEEE Trans. Instrum. Meas. 50(6), 1656–1660 (2001). [CrossRef]
16. F. Villuendas and J. Pelayo, “Optical fibre device for chemical sensing based on surface plasmon excitridon,” Sens. Actuators A Phys. 23(1-3), 1142–1145 (1990). [CrossRef]
17. R. Slavík, J. Homola, and E. Brynda, “A miniature fiber optic surface plasmon resonance sensor for fast detection of Staphylococcal enterotoxin B,” Biosens. Bioelectron. 17(6-7), 591–595 (2002). [CrossRef] [PubMed]
18. X. Fan, I. M. White, S. I. Shopova, H. Zhu, J. D. Suter, and Y. Sun, “Sensitive optical biosensors for unlabeled targets: a review,” Anal. Chim. Acta 620(1-2), 8–26 (2008). [CrossRef] [PubMed]
19. A. Gaston, I. Lozano, F. Perez, F. Auza, and J. Sevill, “Evanescent wave optical-fiber sensing (temperature, relative humidity, and pH sensors),” IEEE Sens. J. 3, 67–72 (2000).
20. J. Homola, “Optical fiber sensor based on surface plasmon excitation,” Sens. Actuators B Chem. 29(3), 401–405 (1995). [CrossRef]
21. P. J. Rivero, A. Urrutia, J. Goicoechea, and F. J. Arregui, “Optical fiber humidity sensors based on localized surface plasmon resonance (LSPR) and Lossy-mode resonance (LMR) in overlays loaded with silver nanoparticles,” Sens. Actuators B Chem. 173, 244–249 (2012). [CrossRef]
22. E. Klantsataya, P. Jia, H. Ebendorff-Heidepriem, T. M. Monro, and A. François, “Plasmonic fiber optic refractometric sensors: From conventional architectures to recent design trends,” Sensors (Basel) 17(1), 1–23 (2016). [CrossRef] [PubMed]
23. A. K. Sharma, R. Jha, and B. D. Gupta, “Fiber-optic sensors based on surface plasmon resonance: A comprehensive review,” IEEE Sens. J. 7(8), 1118–1129 (2007). [CrossRef]
24. H. Esmaeilzadeh, M. Rivard, E. Arzi, F. Légaré, and A. Hassani, “Smart textile plasmonic fiber dew sensors,” Opt. Express 23(11), 14981–14992 (2015). [CrossRef] [PubMed]
25. F. P. Incropera, D. P. DeWitt, T. L. Bergman, and A. S. Lavine, Fundamentals of Heat and Mass Transfer, 6th ed. (John Wiley and Sons, 2007).
26. W. Alduchov and R. E. Eskridge, “Improved Magnus form approximation of saturation vapor pressure,” J. Appl. Meteorol. 35(4), 601–609 (1996). [CrossRef]
