**4. Dew sensor based on PCFI**

Dew (condensed moisture) is a problem in the fields of precision electrical devices, automobiles, air conditioning systems, warehouses and domestic equipment, etc. High humidity and condensation can create an environment where the development of mould on the wooden parts can take place and it can also cause corrosion of iron parts. This is a major

Photonic Crystal Fibre Interferometer for Humidity Sensing 169

dew point temperature, Td, is defined as the temperature to which this quantity of air must be cooled down such that, at a constant pressure, condensation occurs (RH = 100%). In terms of relative humidity RH and temperature T, the dew point temperature is given as:

ln

ln

where, α=243.12 OC and β=17.62 are the so-called Magnus parameters for the temperature range -45 to 60 OC. There fore decreasing the temperature of the PCFI increases the relative humidity close to it. At a certain stage of decreasing the temperature the relative humidity becomes 100% or reaches the dew point temperature and hence the water vapour starts to condense. The condensed water vapour on the PCFI causes a large change in the effective refractive index of the interfering cladding mode (ncl) which in turn causes a large phase change between the interfering modes and therefore a large wavelength shift of the

The dew response of the PCF interferometer was studied by placing it on a thermoelectric cooler (TEC) as shown in Fig. 8. In order to study the influence of dew on the PCFI, it was decided to limit the PCFI length used to 42 mm or less, to suit the size of the available TEC used for temperature control. The temperature of the TEC element was controlled by a temperature controller. A thermistor was used to provide temperature feedback to the controller from the TEC element. An additional handheld thermometer was used to confirm the temperature on the TEC surface. The entire setup was placed inside a controlled environmental chamber. The inside relative humidity and the temperature of the chamber can be controlled with an accuracy of ±2 %RH and ±1 OC respectively. For the purpose of

this experiment the ambient temperature inside the chamber was fixed at 25 OC.

Fig. 8. Experimental arrangement for the calibration of PCFI based dew sensor.

*<sup>T</sup> T RH*

100%

*RH T*

*RH T*

*T*

100% T( , )

d

interference peaks is expected.

**4.2 Experimental characterization of the sensor** 

problem in the case of the works of art in the museums and churches [Camuffo & Valcher, 1986]. So there is a strong demand for a sensor able to accurately detect a high humidity or dew condensation state.

Fig. 7. Interference peak shift of the photonics crystal fibre interferometers with L= 40.5 mm and 17 mm with respect to relative humidity.

Approaches to dew detection using optical fibre have been previously reported in [Baldini et al., 2008; Kostritskii et al., 2009]. The working principle of these sensors is based on the change in the reflectivity which is observed on the surface of the fibre tip, when a water layer is formed on its distal end. The dependence on reflected power measurement scheme used in [Baldini et al., 2008; Kostritskii et al., 2009] increases the chance of measurement error due to source power fluctuations. Recently we have demonstrated a simple sensor head for dew detection based on a photonic crystal fibre interferometer (PCFI) operated in reflection mode [Mathew et al., 2011], with the advantage of good dew point measurement accuracy. The fabrication of such a sensor is very simple since it only involves cleaving and fusion splicing. Furthermore, the spectral measurement technique utilized in this work is free from errors due to source power variations. In the following section of the chapter a dew sensor based on PCFI is explained, including a study of temperature dependence of the device with different lengths of PCF. Since the sensor head is fabricated from a single material, silica, its temperature dependence is very low. From the results of the dew sensor performance with different lengths of PCF it was shown that a device with a compact length of PCF is suitable for dew sensing albeit with a reduction in the speed of response. The response of the sensor at different ambient humidity values is also included in this section.

#### **4.1 Operating principle of the sensor**

To study the response of the PCFI to dew formation it is required to set the temperature of the PCFI to dew point temperature, which is obtained from the values of ambient relative humidity and temperature. To do this let us consider a quantity of air with a constant water vapour concentration at a certain temperature, T, and relative humidity, RH < 100%. The

problem in the case of the works of art in the museums and churches [Camuffo & Valcher, 1986]. So there is a strong demand for a sensor able to accurately detect a high humidity or

Fig. 7. Interference peak shift of the photonics crystal fibre interferometers with L= 40.5 mm

Approaches to dew detection using optical fibre have been previously reported in [Baldini et al., 2008; Kostritskii et al., 2009]. The working principle of these sensors is based on the change in the reflectivity which is observed on the surface of the fibre tip, when a water layer is formed on its distal end. The dependence on reflected power measurement scheme used in [Baldini et al., 2008; Kostritskii et al., 2009] increases the chance of measurement error due to source power fluctuations. Recently we have demonstrated a simple sensor head for dew detection based on a photonic crystal fibre interferometer (PCFI) operated in reflection mode [Mathew et al., 2011], with the advantage of good dew point measurement accuracy. The fabrication of such a sensor is very simple since it only involves cleaving and fusion splicing. Furthermore, the spectral measurement technique utilized in this work is free from errors due to source power variations. In the following section of the chapter a dew sensor based on PCFI is explained, including a study of temperature dependence of the device with different lengths of PCF. Since the sensor head is fabricated from a single material, silica, its temperature dependence is very low. From the results of the dew sensor performance with different lengths of PCF it was shown that a device with a compact length of PCF is suitable for dew sensing albeit with a reduction in the speed of response. The response of the sensor at different ambient humidity values is also included in this section.

To study the response of the PCFI to dew formation it is required to set the temperature of the PCFI to dew point temperature, which is obtained from the values of ambient relative humidity and temperature. To do this let us consider a quantity of air with a constant water vapour concentration at a certain temperature, T, and relative humidity, RH < 100%. The

dew condensation state.

and 17 mm with respect to relative humidity.

**4.1 Operating principle of the sensor** 

dew point temperature, Td, is defined as the temperature to which this quantity of air must be cooled down such that, at a constant pressure, condensation occurs (RH = 100%). In terms of relative humidity RH and temperature T, the dew point temperature is given as:

$$\mathrm{T\_d(T, RH)} = \alpha \frac{\ln\left(\frac{RH}{100\%}\right) + \frac{\mathrm{\beta T}}{\alpha + T}}{\beta - \ln\left(\frac{RH}{100\%}\right) - \frac{\mathrm{\beta T}}{\alpha + T}}$$

where, α=243.12 OC and β=17.62 are the so-called Magnus parameters for the temperature range -45 to 60 OC. There fore decreasing the temperature of the PCFI increases the relative humidity close to it. At a certain stage of decreasing the temperature the relative humidity becomes 100% or reaches the dew point temperature and hence the water vapour starts to condense. The condensed water vapour on the PCFI causes a large change in the effective refractive index of the interfering cladding mode (ncl) which in turn causes a large phase change between the interfering modes and therefore a large wavelength shift of the interference peaks is expected.

#### **4.2 Experimental characterization of the sensor**

The dew response of the PCF interferometer was studied by placing it on a thermoelectric cooler (TEC) as shown in Fig. 8. In order to study the influence of dew on the PCFI, it was decided to limit the PCFI length used to 42 mm or less, to suit the size of the available TEC used for temperature control. The temperature of the TEC element was controlled by a temperature controller. A thermistor was used to provide temperature feedback to the controller from the TEC element. An additional handheld thermometer was used to confirm the temperature on the TEC surface. The entire setup was placed inside a controlled environmental chamber. The inside relative humidity and the temperature of the chamber can be controlled with an accuracy of ±2 %RH and ±1 OC respectively. For the purpose of this experiment the ambient temperature inside the chamber was fixed at 25 OC.

Fig. 8. Experimental arrangement for the calibration of PCFI based dew sensor.

Photonic Crystal Fibre Interferometer for Humidity Sensing 171

interference between the cladding mode and the adsorbed water vapour. Hence the sensitivity to water vapour content and thus dew point temperature is high for a device

It is important to note that due to the large fringe spacing it is difficult to measure the peak shift accurately for a short PCFI, therefore the comparison of sensitivities for PCFIs with different lengths is not straightforward. It should also be noted that even a PCFI with a small length (3.5 mm, fringe spacing ~90 nm) when exposed to dew point temperature for a relatively long time i.e. several minutes will result in a measurable fringe shift as shown in Fig. 10(b). This is because an increasingly thicker adsorbed water layer is formed on the silica surfaces of the PCF as time progresses. Thus compared to 3.5 mm device the ~40.5 mm device is preferable for achieving a fast response time (in the order of seconds), but when a compact length is the main requirement a shorter PCFI also can be used as a dew sensor with a reduced measurement speed. The best range of lengths suitable for dew

Fig. 10. (a). Interference spectra for a device with length 40.5 mm at room temperature and

(b). Interference spectra for a device with length 3.5 mm at room temperature and at dew

The dew sensing performance of a PCFI at different environmental conditions was determined by studying the dew response of a PCFI with L= 40.5 mm at three ambient humidity values of 40, 60 and 80 %RH. At each humidity value the temperature of the PCFI is reduced from 26 OC to the corresponding dew point temperature. The peak wavelength shift of the device is plotted against temperature in Fig. 11. The three curves represent the peak shift corresponding to the ambient relative humidity values of 40, 60 and 80 %RH. The onset of the dew formation is characterized by a large shift of the interference peak which is clear in Fig. 11. The dew point temperature calculated by using equation (1) based on the corresponding ambient conditions is marked on each curve in Fig. 11. For all these three ambient humidity values the continuous spectral shift starts exactly at the dew point temperature which confirms the high dew point

It is observed that at or below the dew point temperature the interference peak shifts continuously with time. This is because an increasingly thicker adsorbed water layer is formed on the silica surface of the PCF microholes as time progresses. By bringing the

sensing is the same as the one given above for humidity sensing.

measurement accuracy (estimated as ±0.1 OC) of the sensor.

with a longer length of PCF.

at dew point temperature.

point temperature.

Since the PCF is composed of only fused silica, it is expected to have minimal thermal sensitivity. The temperature dependence of the device was determined by observing the peak shift of the interference spectrum of the device for a temperature variation from 25 OC to 60 OC. The ambient humidity during the study was set to 40 % RH. When the temperature is increased from 25 OC to 60 OC the interference peak is shifted slightly to higher wavelengths. Fig. 9 shows this temperature dependence for two devices with L=17 mm and 40.5 mm. As expected the thermal sensitivity of the PCFI is very low and is further reduced for a device with the shorter length of PCF. The thermal sensitivity obtained in the experiment for a device with L= 40.5 mm is 9.5 pm/OC and that for L= 17 mm is 6.2 pm/OC.

The dew sensing experiments were carried out at an ambient temperature of 25 OC and at normal atmospheric pressure. To study the dew response of the device the temperature of the PCFI was decreased from ambient temperature (25 OC) to the dew point temperature at a fixed ambient relative humidity. It was found that the position of the interference peaks shifted to longer wavelengths with a decrease in temperature. This shift is similar to the humidity response of the PCFI as shown in Fig. 6 and 7. This occurs because the relative humidity inside the microholes and close to the PCFI increases with a decrease in temperature and causes a shift. At or below the dew point temperature (100% RH) water vapour condensation occurs, the condensed water vapour on the outer surface of the PCF also contributes to the change in the effective RI of the cladding mode, which results in a large spectral shift.

Fig. 9. Interference peak shift with respect to temperature for interferometers with PCF lengths L= 40.5 mm and 17 mm.

The spectra of two interferometers at room temperature and at the dew point temperature for devices fabricated with lengths 40.5 mm and 3.5 mm are shown in Fig. 10(a) & (b). The lengths selected are practically the largest and the smallest PCF lengths that can be studied using our experimental setup. The ambient humidity during this study was set at 60 % RH. From the Fig. 10 it is clear that relative to the period of the interferometer the shift will be larger for a longer PCFI due to a longer interaction length available for the

Since the PCF is composed of only fused silica, it is expected to have minimal thermal sensitivity. The temperature dependence of the device was determined by observing the peak shift of the interference spectrum of the device for a temperature variation from 25 OC to 60 OC. The ambient humidity during the study was set to 40 % RH. When the temperature is increased from 25 OC to 60 OC the interference peak is shifted slightly to higher wavelengths. Fig. 9 shows this temperature dependence for two devices with L=17 mm and 40.5 mm. As expected the thermal sensitivity of the PCFI is very low and is further reduced for a device with the shorter length of PCF. The thermal sensitivity obtained in the experiment for a device with L= 40.5 mm is 9.5 pm/OC and that for L= 17 mm is 6.2 pm/OC. The dew sensing experiments were carried out at an ambient temperature of 25 OC and at normal atmospheric pressure. To study the dew response of the device the temperature of the PCFI was decreased from ambient temperature (25 OC) to the dew point temperature at a fixed ambient relative humidity. It was found that the position of the interference peaks shifted to longer wavelengths with a decrease in temperature. This shift is similar to the humidity response of the PCFI as shown in Fig. 6 and 7. This occurs because the relative humidity inside the microholes and close to the PCFI increases with a decrease in temperature and causes a shift. At or below the dew point temperature (100% RH) water vapour condensation occurs, the condensed water vapour on the outer surface of the PCF also contributes to the change in the effective RI of the cladding mode, which results in a

Fig. 9. Interference peak shift with respect to temperature for interferometers with PCF

The spectra of two interferometers at room temperature and at the dew point temperature for devices fabricated with lengths 40.5 mm and 3.5 mm are shown in Fig. 10(a) & (b). The lengths selected are practically the largest and the smallest PCF lengths that can be studied using our experimental setup. The ambient humidity during this study was set at 60 % RH. From the Fig. 10 it is clear that relative to the period of the interferometer the shift will be larger for a longer PCFI due to a longer interaction length available for the

large spectral shift.

lengths L= 40.5 mm and 17 mm.

interference between the cladding mode and the adsorbed water vapour. Hence the sensitivity to water vapour content and thus dew point temperature is high for a device with a longer length of PCF.

It is important to note that due to the large fringe spacing it is difficult to measure the peak shift accurately for a short PCFI, therefore the comparison of sensitivities for PCFIs with different lengths is not straightforward. It should also be noted that even a PCFI with a small length (3.5 mm, fringe spacing ~90 nm) when exposed to dew point temperature for a relatively long time i.e. several minutes will result in a measurable fringe shift as shown in Fig. 10(b). This is because an increasingly thicker adsorbed water layer is formed on the silica surfaces of the PCF as time progresses. Thus compared to 3.5 mm device the ~40.5 mm device is preferable for achieving a fast response time (in the order of seconds), but when a compact length is the main requirement a shorter PCFI also can be used as a dew sensor with a reduced measurement speed. The best range of lengths suitable for dew sensing is the same as the one given above for humidity sensing.

Fig. 10. (a). Interference spectra for a device with length 40.5 mm at room temperature and at dew point temperature.

(b). Interference spectra for a device with length 3.5 mm at room temperature and at dew point temperature.

The dew sensing performance of a PCFI at different environmental conditions was determined by studying the dew response of a PCFI with L= 40.5 mm at three ambient humidity values of 40, 60 and 80 %RH. At each humidity value the temperature of the PCFI is reduced from 26 OC to the corresponding dew point temperature. The peak wavelength shift of the device is plotted against temperature in Fig. 11. The three curves represent the peak shift corresponding to the ambient relative humidity values of 40, 60 and 80 %RH. The onset of the dew formation is characterized by a large shift of the interference peak which is clear in Fig. 11. The dew point temperature calculated by using equation (1) based on the corresponding ambient conditions is marked on each curve in Fig. 11. For all these three ambient humidity values the continuous spectral shift starts exactly at the dew point temperature which confirms the high dew point measurement accuracy (estimated as ±0.1 OC) of the sensor.

It is observed that at or below the dew point temperature the interference peak shifts continuously with time. This is because an increasingly thicker adsorbed water layer is formed on the silica surface of the PCF microholes as time progresses. By bringing the

Photonic Crystal Fibre Interferometer for Humidity Sensing 173

Awakuni, Y. & Calderwood, J.H. (1972). Water vapour adsorption and surface conductivity in solids. *Journal of Physics D: Applied Physics,* Vol.5, No.5, (May 1972), pp. 1038. Baldini, F. ; Falciai, R. ; Mencaglia, A. A. ; Senesi, F. ; Camuffo, D. ; Valle, A. D. & Bergsten,

Barrera, D. ; Villatoro, J. ; Finazzi, V. P. ; Cardenas-Sevilla, G. A. ; Minkovich, V. P. ; Sales, S. &

Birks, T. A.; Knight, J. C. & Russell, P. St. J. (1997). Endlessly single-mode photonic crystal

Bock, W. J.; Eftimov, T. A.; Mikulic, P. & Chen, J. (2009). An Inline Core-Cladding

Brunauer, S.; Emmett, P. H. & Teller, E. (1938). Adsorption of gases in multimolecular layers, *Journal of the American Chemical Society*, Vol.60, (February 1938), pp. 309-319. Camuffo, D. & Valcher, S. (1986). A dew point signaller for conservation of works of art, *Environmental Monitoring and Assessment*, Vol.6, No. 2, (1986), pp. 165-170. Cárdenas-Sevilla, G. A.; Finazzi, V. ; Villatoro, J. & Pruneri, V. (2011). Photonic crystal fiber

Choi, H.Y. ; Kim, M. J. & Lee, B. H. (2007). All-fiber Mach-Zehnder type interferometers formed in photonic crystal fiber, *Optics Express*, Vol. 15, No. 9, (April 2007), pp. 5711–5720. David, B. A. & Seong, H. K. (2005). Evolution of the adsorbed water layer structure on

Jha, R.; Villatoro, J. & Badenes, G. (2008). Ultrastable in reflection photonic crystal fiber

Káčik, D.; Turek, I.; Martinček, I.; Canning, J.; Issa, N. & Lyytikäinen, K. (2004). Intermodal

Knight, J. C. (2003). Photonic crystal fibres. *Nature*, Vol. 424, No. 6950, (Aug. 2003), pp. 847–851. Knight, J. C.; Birks, T. A.; Cregan, R. F.; Russell, P. St. J. & De Sandro, J. P. (1998). Large mode

Knight, J. C.; Birks, T. A.; Russell, P. S. J. & Atkin, D. M. (1996). All-silica single-mode optical fiber

Kostritskii, S. M.; Dikevich, A. A.; Korkishko Y. N. & Fedorov, V. A. (2009). Dew point

Lim, J. H.; Jang, H. S.; Lee, K. S.; Kim, J. C. & Lee, B. H. (2004). Mach–Zehnder interferometer

Mathew, J.; Semenova, Y.; Rajan, G. & Farrell, G. (2010). Humidity sensor based on photonic

Pipes, *Journal of Sensors*, Vol. 2008, Article ID 321065.

fiber, *Optics Letters*, Vol. 22, No. 13, (July 1997), pp. 961-963.

*Technology*, Vol. 27, No. 17, (Sept. 2009), pp. 3933–3939.

2011), pp. 7596-7602.

pp. 3465-3470.

(2009), pp. 73561K.

1341-1343.

(August 2005), pp. 16760–16763.

Vol. 93, No. 19, (November 2008), pp. 191106.

*Optics Letters*, Vol. 29, No. 4, pp. 346–348.

C. J. (2008). Miniaturised Optical Fibre Sensor for Dew Detection Inside Organ

Pruneri, V. (2010). Low-Loss Photonic Crystal Fiber Interferometers for Sensor Networks. *Journal of Lightwave Technology*, Vol. 28, No. 24, (Dec. 2010), pp. 3542–3547.

Intermodal Interferometer Using a Photonic Crystal Fiber, *Journal of Lightwave* 

sensor array based on modes overlapping, *Optics Express*, Vol. 19, No. 8, (April

silicon oxide at room temperature, *Journal of Physical Chemistry B*, Vol. 109, No. 35,

modal interferometer for accurate refractive index sensing, *Applied Physics Letters*,

interference in a photonic crystal fibre, *Optics Express*, Vol. 12, No. 15, (July 2004),

area photonic crystal fibre, *Electronics Letters*, Vol. 34, No. 13, (June 1998), pp. 1347-1348.

with photonic crystal cladding. *Optics Letters*, Vol. 21, No. 19, (Oct. 1996), pp. 1547–1549.

measurement technique utilizing fiber cut reflection. *Proceedings of SPIE*, Vol. 7356,

formed in a photonic crystal fiber based on a pair of long-period fiber gratings,

crystal fibre interferometer, *Electronics Letters*, Vol. 46, No. 19, (September 2010), pp.

**6. References** 

temperature of the PCFI back to room temperature the interference peaks also shift back to their initial position. This shows the reversibility of the sensor. Because of the small size of the sensor head and the high sensitivity to adsorbed water vapour the demonstrated sensor response time is in seconds which is relatively fast compared to existing dew point hygrometers that take several minutes for a single measurement. The simple fabrication method, small size and the all-silica nature of the demonstrated sensor head suggest that with some simple additions such as attaching a TEC element with temperature feedback on to the PCFI, the combination can be used as a dew point hygrometer.

Fig. 11. Interference peak shift of PCFI with respect to temperature at three ambient humidity values of 40, 60 and 80 %RH.
