A new type of fluorescence fiber optic temperature sensor based on the detection of fluorescence intensity ratio (FIR) is proposed. Benefited from the temperature-dependent characteristic of upconversion luminescence (UCL), it can be applied in fiber optic temperature sensors. Rare earth doped upconversion nanoparticles (UCNPs) NaYF4:Er3+,Yb3+ are embedded in a multi-mode quartz fiber through the technology of fiber fusion as the sensing unit of temperature sensors. A 980 nm laser is used to stimulate UCL in a temperature range from 40 °C to 100 °C. Experimental validation and spectral analysis are carried out to confirm the rationality of sensors’ design. Results show that FIR changes with the temperature in Boltzmann distribution law. The sensitivity of the temperature sensors can reach the value between 0.0087 and 0.0144 K−1.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Temperature index is an indispensable parameter in many projects among the fields of natural science and engineering technology. Many temperature sensors are made to serve for scientific research and industrial production [1–3]. The fiber optic temperature sensor is one of the most important sensors. It is very popular in high direct-current (DC) electric field, high frequency field, microwave field and other environmental applications for the reason of insensitivity to electromagnetic field, small size, safety, wide working range and so on [4–6].
The fluorescence fiber optic temperature sensor, based on the temperature-dependent FIR of thermal coupled level (TCL), has become a research hotspot in the field of temperature measurement [7–10]. Fluorescence fiber optic temperature sensors, which are more superior than conventional temperature sensors, are accurate, stable, compactable and less dependent on environmental conditions. In recent years, temperature-dependent UCL characteristics in Er3+ doped or Er3+/Yb3+ codoped nanocrystals have been reported [11–13]. However, the technology is not mature enough to be widely used in temperature detection. The research work on fluorescence fiber optic temperature sensors is still in the experimental stage.
Research on fluorescence temperature sensors began in the late 1980s. In 1987, K. T. V. Grattan designed a temperature sensor with a ruby crystal as the sensing unit. Based on the fluorescence intensity technology, the temperature range was 293-433 K with an accuracy of ± 3 K [14,15]. In 2002, H. Aizawa reported fluorescence fiber optic temperature sensors based on the detection of fluorescence lifetime and fluorescence intensity. However, fluorescence lifetime-based temperature sensors required much more complicated instruments [16,17]. In 2014 and 2016, a new sort of fiber optic FIR-based temperature sensor was presented [18,19], which possessed the maximum sensitivity of 0.0099 K−1 at 583 K since the FIR of upconversion emissions can help to reduce the influence of measurement conditions. Some other scholars had conducted studies on this technology as well [20–24]. However, the research on fluorescence temperature sensors is still in the experimental stage and cannot be widely applied to industrial production. Nowadays, the fiber optic sensors widely used in industry are Bragg fiber grating sensors with high sensitivity , but the accuracy of such sensors can be easily affected by the fluctuations of exciting light. Besides, Bragg fiber grating sensors are not suitable for low temperature measurement because of the detection principle. Therefore, fluorescence fiber optic temperature sensors are regarded as one of the promising sensors because it can reduce the dependence of the measurement conditions, such as fluorescence loss, fluctuations of pump laser and electromagnetic compatibility problem. Fluorescence fiber optic temperature sensors are also very suitable for medium and low temperature detection.
In this paper, a new type of fluorescence fiber optic temperature sensor based on the detection of FIR has been proposed, fabricated, and analyzed. The sensor is equipped with NaYF4:Er3+,Yb3+ nanocrystals as the sensing unit. UCNPs NaYF4:Er3+,Yb3+ are embedded in the multi-mode quartz fiber through fiber fusion technology. Results show that this type of sensor has many advantages such as system reliability, small size, immunity to electromagnetic interference and ease of installation.
2. Detection principle
Photoluminescence is light emission from any form of an object after the absorption of photons, which can be classified as UCL, down conversion luminescence (DCL) and down shifting luminescence (DSL). UCL is a nonlinear photoluminescence process, which can convert long-wave radiation into short-wave radiation through multi-photon mechanism. Rare earth doped UCNPs usually consist of activator, sensitizer and matrix . In this paper, the NaY0.80F4:0.02Er3+,0.18Yb3+ is used as the UCNPs, in which NaYF4 is the matrix that provides a suitable crystal field for rare earth ions, Er3+ is acted as the activator to give off visible emission and Yb3+ is acted as the sensitizer to absorb excitation energy and then transfer to Er3+ through a multi-photon process.
The principle of fluorescence fiber optic temperature sensors is based on the temperature-dependent UCL of NaYF4:Er3+,Yb3+ nanocrystals. The corresponding UCL spectrum and schematic energy levels involved in the upconversion process are shown in Fig. 1. On the pumping of 980 nm laser and the afterwards energy transfer from Yb3+ to Er3+, the radiative emission from different higher energy levels of Er3+ to its ground energy level may lead to several distinct emission bands. It can be clearly observed from Fig. 1 that the 2H11/24I15/2, 4S3/24I15/2 and 4F9/24I15/2 transitions of Er3+ give rise to visible emissions at 525, 545 and 666 nm, respectively .
As we know that the UCL intensity obtained from the spectral measurement can be affected by many factors, such as disturbance of pump laser source, noise interference and so on. The intensity-based method is not conducive to the accurate measurement of temperatures. Therefore, the measurement method based on temperature-dependent FIR of TCL is introduced. The intensity ratio of two emission bands peaked at 525 nm and 545 nm, namely RHS (I525/I545), is a single function of the temperature which conforms to the Boltzmann distribution. By calculating FIR, temperatures can be figured out from the formula. The advantage of this method is that it is not affected by the unstable intensity of pump laser source and insensitive to bending losses.
where RHS (0) is FIR when T = 0 K, is the energy difference between energy level 2H11/2 and 4S3/2, k is the Boltzmann constant, T is the absolute temperature.
The most important part of temperature sensors is the fabricated sensing unit. NaYF4:Er3+,Yb3+ nanocrystals embedded in optical fiber are shown in Fig. 2(a), which were synthesized in our joint lab . Image of fiber end face before fusion is shown in Fig. 2(b). A GT21-FC-FC-05 type multi-mode quartz fiber with 62.5 μm core diameter made by China Datang Storm Ltd. was cut in half before fusion, half flat half NaYF4:Er3+,Yb3+ nanocrystals mixed with anhydrous alcohol. Image of fiber end face after fusion is shown in Figs. 2(c) and 2(d), namely the prepared sensing unit. UCNPs were successfully embedded in 62.5 μm multi-mode quartz graded fiber through fiber fusion technology. Figures 2(c) and 2(d) are the schematic diagrams of X-axis and Y-axis direction of the fiber. Here, the thickness of the fused sensing unit is 12-15 μm.
The structure of the platform mainly includes (1) ADR-1805 980 nm single-mode laser (Radium Photoelectric Technology Co., Ltd., Changchun, China), (2) 62.5 μm multi-mode quartz graded fiber with NaYF4:Er3+,Yb3+ nanocrystals as the sensing unit, (3) bandpass optical filter with no lower than 90% light transmittance at 400-700 nm band, (4) BIM-6001 spectrometer (Brolight Technology Co., Ltd. Hangzhou, China), (5) DZF-250 temperature chamber (Yite Instrument Co., Ltd. Zhengzhou, China), the temperature resolution is 0.1 °C and the accuracy of the temperature control is ± 1%. (6) TYPE-81C direct core monitoring fusion splicer (Sumitomo Electric Industries, Kyoto, Japan). Fusion mode MM G651 Auto of automatic fiber identification function is chosen. The fusion splicer can determine the type of fiber, and automatically select the most suitable welding conditions. The welding parameters are mainly set as follows, alignment method: outer diameter alignment; discharge time: 3.0 s; fiber end face angle permissible value: 5.0°; allowable value of fiber fusion angle: 0.1°.
The proposed fiber optic temperature sensor in this paper is extrinsic that uses UCNPs as the sensing unit. In this condition, optical fiber serves as the path to transmit light.
The specific design of the experiment is shown in Fig. 3(a). A 980 nm single-mode laser is used to stimulate UCL. UCNPs are fused inside the 62.5 μm multi-mode graded quartz fiber as a sensing unit of the temperature sensor. At the beginning, the 980 nm laser emits laser light. Fiber that comes with laser is connected to the multi-mode fiber through the flange. Then, light enters the multi-mode fiber to excite UCL of NaYF4:Er3+,Yb3+ which converts the 980 nm infrared light into visible light. In order to enhance UCL phenomenon, the laser power is relatively large. Therefore, light passing through the sensing unit still has strong infrared component which affects the acquisition of final signals. The bandpass optical filter is placed in the bracket with collimator lenses aiming at filtering out the light at 980 nm. Finally, by setting the environmental temperature, spectral data under different temperatures are collected by the spectrometer.
The practical measurement system is shown in Fig. 3(b) and the temperature measurement range is between 40 °C and 100 °C. An image of the sensor’s sensing unit in the temperature chamber is shown in Fig. 3(c). The actual UCL spectra are obtained by collecting the spectra of the unfused sensors and the fused ones at the same temperature and then subtracting these two spectral data with each other. Because the original signals contain much direct-current component and high-frequency noise, so a band-pass filter technology is used in software to filter out parts of the noise. When the filter is finished, we can take the measurement method introduced in section 2. For the FIR method, the emissions peaked around 525 and 545 nm are selected to calculate intensity ratio for the measurement of temperatures, since the intensity at 566 nm is not easy to recognize from the noise signal. The original gathering UCL spectra in the temperature range from 40 °C to 100 °C are presented in Fig. 4.
As described in section 2, the UCL intensity obtained from the spectral measurement can be affected by many factors so that the relationship between intensity and temperature is not accurate enough to evaluate the testing temperature. Fortunately, the UCL intensity ratio of different emission peaks is stable with external disturbances. To prove the relationship between RHS (I525/I545) and the temperature of our fiber coincident with Eq. (1), the spectral data in Fig. 4 are processed and plotted as a two-dimensional curve of intensity ratio versus temperature, as shown in Figs. 5(a) and 5(b).
According to Eq. (2), we can know that the relationship between log plot of the intensity ratio RHS(I525/I545) and 1/T is substantially linear, which is consistent with the Boltzmann distribution law Eq. (1). Furthermore, we can obtain the absolute value of the fitting curve’s slope from the formula of the fitting curve, which is 1393. As for the temperature measurement, another important parameter that must be considered is the FIR for the rate of temperature changes, which decides the sensitivity of the sensor. The sensitivity can be described as follows:
According to Eq. (3), we can obtain the value of sensitivity between 0.0087 and 0.0144 K−1. The bigger the is, the more sensitive the sensor is. However, the energy should be limited to a certain range. Apart from this, sensitivity is also related to the temperature and the FIR.
According to the existing research results, if the temperature is lower, the temperature-dependent characteristic of UCNPs will be more obvious and the spectral intensity will be stronger . The value of linear correlation is 0.8348 in Fig. 5(a). Points of 90 °C and 100 °C slightly deviate from the fitting curve, while points of temperatures from 40 °C to 80 °C are relatively close to the fitting curve. The most possible reason responsible for this deviation may due to the more mixing of noise interruption with the actual intensity signals when the temperature is higher during the spectral measurement process. As we know that the UCL intensity decreases with increasing temperature due to the so called “thermal quenching”. In this case, the influence of the noise interruption becomes more obvious at high temperatures, which inevitably leads to a larger error estimation in the calculation of FIR. Therefore, the FIR method is more suitable for low temperature measurement. However, limited by experimental conditions, the temperature is set above room temperature in this experiment.
To show the reproducibility of the sensor for UCL spectra on time duration, the variance of its FIR is evaluated. As shown in Fig. 6(a), the reproducibility data of the FIR in 40 °C are collected every ten minutes and collected ten times in total. The variance of the FIR is 0.002018, showing a good stability. To show the reproducibility of the sensor for UCL spectra on heating cycle, the environment of temperature chamber is changed from 40 °C to 100 °C, then from 100 to 40 °C in reverse. Spectral data are collected every 10 °C. As shown in Fig. 6(b), the solid line represents the change of the FIR with the rise of temperatures. The dotted line represents the change of the FIR with the decrease of temperatures. It can be seen that the fabricated sensor has a good reliability on heating cycle.
In this paper, a new type of fluorescence temperature fiber optic sensor that uses NaY0.80F4:0.02Er3+,0.18Yb3+ as the sensing unit is proposed and fabricated. The temperature dependent FIR analysis and experimental test indicates that there exists a linear relationship between log of the intensity ratio and reciprocal of temperatures in the range of 40 °C to 100 °C, and the sensitivity can reach the value between 0.0087 and 0.0144 K−1. The fabricated sensor has good linearity in the range of 40 °C to 80 °C. Reproducibility of the sensor's FIR for UCL spectra on time duration and heating cycle indicates a good stability and reliability, which proves the rationality of fluorescence fiber optic sensor’s design and the feasibility of the scheme. It will promote the development of temperature sensors in industrial detection and other areas.
Fundamental Research Funds for the Central Universities (Grant no. 22120180299); Natural Science Foundation of Shanghai (Grant no.18ZR1441900); National Natural Science Foundation of China (Grant no. 51872200).
1. K. T. V. Grattan and T. Sun, “Fiber optic sensor technology: an overview,” Sens. Actuators A Phys. 82(1), 40–61 (2000). [CrossRef]
2. J. Castrellon, G. Paez, and M. Srtojnik, “Remote temperature sensor employing erbium-doped silica fiber,” Infrared Phys. Technol. 43(1), 219–222 (2002). [CrossRef]
3. A. H. Khalid and K. Kontis, “Thermographic phosphors for high temperature measurements: principles, current state of the art and recent applications,” Sensors (Basel) 8(9), 5673–5744 (2008). [CrossRef] [PubMed]
4. V. I. Busurin, A. S. Semenov, and N. P. Udalov, “Optical and fiber-optic sensors,” Sov. J. Quantum Electron. 15(5), 595–621 (1985). [CrossRef]
5. A. D. Kersey, “A review of recent developments in fiber optic sensor technology,” Opt. Fiber Technol. 2(3), 291–317 (1996). [CrossRef]
6. B. Lee, “Review of the present status of optical fiber sensors,” Opt. Fiber Technol. 9(2), 57–79 (2003). [CrossRef]
7. P. V. Dos Santos, M. T. De Araujo, A. S. Gouveia Neto, J. A. Medeiros Neto, and A. S. B. Sombra, “Optical temperature sensing using upconversion fluorescence emission in Er3+/Yb3+ co-doped chalcogenide glass,” Appl. Phys. Lett. 73(5), 578–580 (1998). [CrossRef]
8. V. K. Rai, D. K. Rai, and S. B. Rai, “Pr3+ doped lithium tellurite glass as a temperature sensor,” Sens. Actuators A Phys. 128(1), 14–17 (2006). [CrossRef]
9. W. Xu, X. Gao, L. Zheng, Z. Zhang, and W. Cao, “Short-wavelength upconversion emissions in Ho3+/Yb3+ codoped glass ceramic and the optical thermometry behavior,” Opt. Express 20(16), 18127–18137 (2012). [CrossRef] [PubMed]
10. K. Zheng, Z. Liu, C. Lv, and W. Qin, “Temperature sensor based on the UV upconversion luminescence of Gd3+ in Yb3+-Tm3+-Gd3+ codoped NaLuF4 microcrystals,” J. Mater. Chem. C Mater. Opt. Electron. Devices 1(35), 5502–5507 (2013). [CrossRef]
11. X. Yu, F. Song, C. Zou, L. J. Luo, C. G. Ming, W. T. Wang, Z. Z. Cheng, L. Han, T. Q. Sun, and J. G. Tian, “Temperature dependence of luminescence behavior in Er3+/Yb3+ co-doped transparent phosphate glass ceramics,” Opt. Mater. 31(11), 1645–1649 (2009). [CrossRef]
12. S. K. Singh, K. Kumar, and S. B. Rai, “Er3+/Yb3+ co-doped Gd2O3 nano-phosphor for optical thermometry,” Sens. Actuators A Phys. 149(1), 16–20 (2009). [CrossRef]
13. F. Huang, Y. Gao, J. Zhou, J. Xu, and Y. S. Wang, “Yb3+/Er3+ co-doped CaMoO4: a promising green upconversion phosphor for optical temperature sensing,” J. Alloys Compd. 639(1), 325–329 (2015). [CrossRef]
14. K. T. V. Grattan, R. K. Selli, and A. W. Palmer, “Ruby fluorescence wavelength division fiber‐optic temperature sensor,” Rev. Sci. Instrum. 58(7), 1231–1234 (1987). [CrossRef]
15. K. T. V. Grattan, R. K. Selli, and A. W. Palmer, “Ruby decay‐time fluorescence thermometer in a fiber‐optic configuration,” Rev. Sci. Instrum. 59(8), 1328–1335 (1988). [CrossRef]
16. H. Aizawa, N. Ohishi, S. Ogawa, T. Katsumata, S. Komuro, T. Morikawa, and E. Toba, “Fabrication of ruby sensor probe for the fiber-optic thermometer using fluorescence decay,” Rev. Sci. Instrum. 73(10), 3656–3658 (2002). [CrossRef]
17. H. Aizawa, T. Katsumata, J. Takahashi, K. Matsunaga, S. Komuro, T. Morikawa, and E. Toba, “Long afterglow phosphorescent sensor materials for fiber-optic thermometer,” Rev. Sci. Instrum. 74(3), 1344–1349 (2003). [CrossRef]
18. K. Morita, T. Katsumata, S. Komuro, and H. Aizawa, “Fiber-optic thermometry using thermal radiation from Tm end doped SiO2 fiber sensor,” Rev. Sci. Instrum. 85(4), 044902 (2014). [CrossRef] [PubMed]
19. X. N. Chai, J. Li, X. S. Wang, Y. X. Li, and X. Yao, “Color-tunable upconversion photoluminescence and highly performed optical temperature sensing in Er3+/Yb3+ codoped ZnWO4,” Opt. Express 24(20), 22438–22447 (2016). [CrossRef] [PubMed]
20. H. Q. Guo and S. Q. Tao, “An active core fiber-optic temperature sensor using an Eu(III)-doped sol-gel silica fiber as a temperature indicator,” IEEE Sens. J. 7(6), 953–954 (2007). [CrossRef]
21. P. A. S. Jorge, P. Caldas, J. C. G. E. D. Silva, C. C. Rosa, A. Oliva, J. L. Santos, and F. Farahi, “Luminescence-based optical fiber chemical sensors,” Fiber Integr. Opt. 24(1), 201–225 (2005). [CrossRef]
23. I. Hernández-Romano, M. A. Cruz-Garcia, C. Moreno-Hernández, D. Monzón-Hernández, E. O. López-Figueroa, O. E. Paredes-Gallardo, M. Torres-Cisneros, and J. Villatoro, “Optical fiber temperature sensor based on a microcavity with polymer overlay,” Opt. Express 24(5), 5654–5661 (2016). [CrossRef] [PubMed]
24. Y. Jiang, Z. Fang, Y. Du, E. Lewis, G. Farrell, and P. Wang, “Highly sensitive temperature sensor using packaged optical microfiber coupler filled with liquids,” Opt. Express 26(1), 356–366 (2018). [CrossRef] [PubMed]
25. J. Mandal, S. Pal, T. Sun, K. T. V. Grattan, A. T. Augousti, and S. A. Wade, “Bragg grating-based fiber-optic laser probe for temperature sensing,” IEEE Photonics Technol. Lett. 16(1), 218–220 (2004). [CrossRef]
27. D. D. Li, Q. Y. Shao, Y. Dong, and J. Q. Jiang, “Temperature sensitivity and stability of NaYF4: Yb3+,Er3+ core-only and core-shell upconversion nanoparticles,” J. Alloys Compd. 617(1), 1–6 (2014). [CrossRef]
28. Z. H. Feng, L. Lin, Z. Z. Wang, and Z. Q. Zheng, “Low temperature sensing behavior of upconversion luminescence in Er3+/Yb3+ codoped PLZT transparent ceramic,” Opt. Commun. 399(1), 40–44 (2017). [CrossRef]
29. W. Yu, W. Xu, H. Song, and S. Zhang, “Temperature-dependent upconversion luminescence and dynamics of NaYF4:Yb3+/Er3+ nanocrystals: influence of particle size and crystalline phase,” Dalton Trans. 43(16), 6139–6147 (2014). [CrossRef] [PubMed]
30. S. Ye, P. Xiao, H. Z. Liao, S. Li, and D. P. Wang, “Fast synthesis of sub-10 nm β-NaYF4: Yb3+, Er3+@NaYF4 core-shell upconversion nanocrystals mediated by oleate ligands,” Mater. Res. Bull. 103(1), 279–284 (2018). [CrossRef]