TEMPERATURE SENSORS

Experimental results of ratio-based erbium-doped-silica temperature sensor

[+] Author Affiliations
Gonzalo Paez, Marija Strojnik

Centro de Investigaciones en Optica, Apartado Postal 1-948, 37000?Leo´n, Gto., Me´xico E-mail: mstrojnik@aol.com

Opt. Eng. 42(6), 1805-1811 (Jun 01, 2003). doi:10.1117/1.1571830
History: Received Sep. 3, 2002; Revised Nov. 22, 2002; Accepted Nov. 27, 2002; Online May 21, 2003
Text Size: A A A

Open Access Open Access

We describe experimental results demonstrating the performance of an erbium-doped silica fiber as a remote temperature sensor in the temperature interval 21 to 96°C. We present the measured fluorescence spectrum corresponding to the energy levels 2H11/2 and 4S3/2. This sensor features simple signal detection in a band and its data analysis system, incorporating a power ratio to reduce noise effects. The channel responsivity, power ratio, and sensitivity for a number of possible filters are presented. By using filters transmitting the 527- to 537-nm and 545- to 555-nm spectral bands, the following figures of merit are achieved: the responsivity is greater than 0.2 μW/K, and the sensitivity is 0.0065K1. With a custom-made filter centered on 545 nm, even higher sensitivity is predicted. © 2003 Society of Photo-Optical Instrumentation Engineers.

Knowledge of temperature is important for operation of optical instruments, in controlling industrial processes, and in maintaining proper operational environments. Likewise, in space applications, including infrared astronomy, telescope operation, and data acquisition systems, the temperature of the telescope facility needs to be known at all times to control the thermal noise12 and to increase the signal-to-noise ratio in applications such as exoplanet detection.34 The most common method of temperature measurement is to put a transducer in contact with the piece whose temperature needs to be known, wait until their temperatures are equilibrated, and read the temperature on a corresponding scale. This method works well with thermocouples, within their operational range. It is necessary, however, that the thermal mass of the sensor be such that achieving equilibrium does not interfere with the measurement. Recently, there has been much emphasis on making the instruments intelligent567 and autonomous,8910 so that they can function independently of human supervision and intervention.

Most often, radiative methods of temperature measurements are considered highly advantageous, because they require neither physical contact nor temperature equilibrium between different objects with diverse thermal masses, effective temperatures, and emissivities.11

The problem of temperature variation in time has been found to be challenging in a number of studies,12131415 which require understanding of the thermal properties not only of the heated object, but also of the illumination source.16 Radiation measurements routinely achieve accuracy on the order of 0.1 K or better. However, these techniques employ calibration procedures requiring an accurate temperature source and human involvement.17

Other methods have been proposed for temperature measurements when the line of sight is obstructed. Using a fluorescence-based temperature sensor, two implementations are possible. The first one is to measure the emitted spectral power at two peaks and to find the ratio, as in the fluorescence intensity ratio technique, whose implementation with some modifications is reported in this paper.

The other one is the fluorescence decay-time method. At temperatures of interest in our application, from 20 to about 100°C, the former exhibits good variation with temperature, while the sensitivity of the latter approaches zero.1819 Using decay lifetime measurements, a standard-deviation error of 1.2 K has been reported in the temperature interval of interest.20 An improvement by a factor of 3 over the performance of sensors incorporating ruby is achieved by a newly grown material, chromium-doped spinel crystal.21 Using chromium in YAG, temperature measurement reproducibility of 1.2 K has recently been reported with lifetime measurements at room temperature.22 In the case of Yb-doped silica, the ratio technique is independent of strain, but not independent of temperature.23 Er–Yb-codoped samples exhibit an error in measurement of ±5 K within the temperature interval studied here.24

First we summarize the advantages of the fiber optic sensors for remote applications; then we comment on the significant advantages of the rare-earth-doped materials25 and their suitability for upconversion, and finally we describe our proposed sensor.

1.1 Fiber Optic Temperature Sensors

Fiber optic sensing devices are particularly appropriate for the operation in, and transmission across, thermally, chemically, and electromagnetically hazardous environments. These include applications involving highly corrosive gases, electrical and nuclear power stations, oil refineries, coal mines, and fire detection in such spaces. A review paper highlighting the importance and versatility of fiber optic sensors has been published recently.26

In the past, research on fiber-optic sensing and, in particular, on temperature measurement and its commercial development has been hindered by cost and reliability issues.2728 Temperature sensors based on interferometric techniques29 and the temperature dependence of the fluorescence decay time3031 have also been investigated. In general, these procedures require expensive and bulky equipment, they involve time-consuming measurements, and they are very sensitive to fluctuations in the excitation power.

1.2 Rare-Earth-Doped Silica Temperature Sensors

Current research is focused on developing reliable, cost-effective fiber temperature sensors.32 Rare-earth-doped silica fibers have been investigated previously for the possibility of developing new temperature sensors, utilizing the fluorescence power ratio technique,3334353637 which takes advantage of temperature dependence of the fluorescence emission spectrum of rare-earth-doped silica. The fluorescence power ratio method results in a superior sensor, because the ratio techniques are relatively immune to noise and fluctuations in the pump power.

Low-loss rare-earth-doped silica fibers are suitable for temperature-sensing applications because of the temperature dependence of their fluorescence emission spectrum. This behavior arises from the temperature dependence of the homogeneous broadening of the emission linewidth and the changing population of the energy levels, as illustrated in Fig. 1. When an erbium-doped fiber is pumped with photons of 2.484×1019-J energy, corresponding to a wavelength of approximately 800 nm (785 nm in our experiment), the 4I9/2 erbium level is excited and the 4I13/2 metastable level is quasiinstantaneously populated by nonradiative transitions. The 4I13/2 level absorbs pump photons, resulting in the excitation of levels 2H11/2 and 4S3/2, responsible for the emission at about 520 and 560 nm, respectively. These levels are said to be in quasi thermal equilibrium, because of the small energy gap between them (about 1.59×1020J) in comparison with the energy difference between them and the next lower level (about 5.9636×1020J).

Graphic Jump LocationF1 :

Generation of visible fluorescence radiation on pumping with near-IR radiation.

In silica, fast thermal coupling between these two levels has been studied theoretically3839 and observed experimentally.4041 The fluorescence of the rare-earth-doped silica, pumped with 800-nm radiation, has already been demonstrated with fiber amplifiers.424344 The ratio R of the powers P resulting from the transition between two levels is proportional to the ratio of their frequencies ν, to the ratio of their emission cross sections σe, and to the Boltzmann population distribution: Display Formula

1R=P(2H11/2)P(4S3/2)=ν(2H11/2)ν(4S3/2)σe(2H11/2)σe(4S3/2)exp(ΔEkBT).
In this expression, ΔE is the energy difference between two levels 2H11/2 and 4S3/2,kB is the Boltzmann constant, (kB=1.38054×1023J/K), and T is the absolute temperature in kelvins.

Figure 2(a) shows the experimental setup to characterize the erbium-doped fiber-optic sensor and to measure its fluorescence spectrum. Figure 2(b) shows a photograph of the experimental setup to evaluate the performance of the sensor. The temperature-sensing fiber is immersed in a liquid to increase the temperature stability of the experiment and to decrease any heat-transfer losses across boundaries. The laser beam at 785 nm passes through a dichroic mirror into a standard fiber. The erbium-doped (960 ppm) fiber of length 20 cm and core diameter 3.2 μm is located inside a water tank, whose temperature T is monitored with a thermocouple. The liquid in the water tank is slowly heated at the rate of approximately 0.1 K/min, while the spectrum is recorded every 2 K. The measured spectrum is normalized at each temperature due to the possibility of fluctuations in the pumping power, of fiber misalignment, and of coupling losses. The fluorescence radiative power is 270 μW at 27°C for 60-mW pump power, measured with a Newport research power meter (model 815).

Graphic Jump LocationF2 :

(a) Principle of operation of the erbium-doped fiber-optic sensor for remote temperature measurements, employing the fluorescence emission ratio in the wavelength interval 515 to 570 nm. (b) Photograph of the experimental setup to evaluate the performance of the erbium-doped fiber-optic temperature sensor. The temperature-sensing fiber is immersed in a liquid to increase the temperature stability of the measurements.

The dichroic mirror that transmits the pumping infrared laser radiation reflects the green fluorescence radiation. A filter wheel under computer control is used to select and transmit the narrow wavelength interval in the fluorescence spectrum. Thus, a number of spectral transmission curves are obtained, allowing for the selection of the one with the most desirable characteristics, as determined by an appropriate figure of merit. We propose to use a high signal-to-noise ratio for each channel, coupled with high sensitivity, rather than maximizing the power ratio.

A photodiode satisfies the principal requirements of the detection system for high sensitivity in the spectral interval 510 to 570 nm, low dark current, and compatibility with the optical fiber. In addition to the optical transmission losses at the beamsplitters, filters, and lenses, various noise sources also decrease the signal-to-noise ratio of the sensor.

Figure 3(a) shows the measured fluorescence emission spectrum of the erbium-doped silica optical fiber as a function of wavelength (500 to 600 nm) for different temperatures. The 2H11/2 emission is observed to increase nearly linearly with temperature in the interval 21 to 96°C. The 4S3/2 line decreases nearly linearly in the same temperature range.

Graphic Jump LocationF3 :

(a) The measured fluorescence emission spectrum of the erbium-doped silica optical fiber as a function of wavelength is shown over a temperature interval from 21 to 96°C. (b) The temperature dependence of the fluorescence spectrum, defined as the ratio of the spectral signal Pout(λ,T) to the spectral signal at the reference temperature of 21°C, Pout(λ,T=21°C), as a function of wavelength.

On the basis of the initial results, we analyzed in more detail the sensitivity of specific spectral emission subbands to the temperature changes. Figure 3(b) shows the normalized fluorescence spectrum as a function of wavelength for the 2H11/2 and 4S3/2 bands, displayed as the ratio of the spectral power at an elevated temperature, Pout(λ,T), to the spectral power at the reference temperature of 21°C, Pout(λ,T=21°C). The spectral bands sensitive to temperature changes are 510 to 535 nm, due to the erbium transition 2H11/2, and 535 to 575 nm, due to the transition 4S3/2. The emission of both transitions is nearly a linear function of temperature.

Finally, the peaks of both transitions shift with temperature. For this reason, it is difficult to choose the most appropriate spectral interval for the signal evaluation. Figure 4 shows the transmission curves of six commercially available filters, used in the search for the optimum spectral subbands for the power-ratio technique. The filters adequately cover both spectral lines. However, we do not have a filter that optimally transmits the peak of the 4S3/2 transition. The position of this peak appears to shift appreciably with temperature. In the temperature interval from room temperature to 96°C, the engineered sensor will have to include a custom-fabricated filter to bring out the overall optimized performance of the sensor.

Graphic Jump LocationF4 :

Measured transmission of the actual filters used in the experimental setup. [See filters F1 and F2 in Fig. 2(a).]

The spectral responsivity of the sensing element is considered next in order to identify the most appropriate spectral bands to be used in the sensor. First, we evaluate the spectral power incident on the detector, as limited by the filter transmission.

Figure 5 shows the fluorescence power Pout generated by the temperature-sensing element and transmitted by the filters, shown in Fig. 4, as a function of temperature. In some cases, the power decreases; in others, it increases. In two special cases, the power remains constant, indicating that the spectral power in the band transmitted by these filters is independent of temperature. The detected power in the interval increases from 2 to 53 μW at room temperature; it increases from 2 to 41 μW at 96°C. For the two spectral bands of highest emission, the power decreases from 53 to 41 μW at 25°C, and it increases from 26 to 39 μW at 96°C. In terms of the spectral transitions, these band emissions correspond to the available filters: 527 to 537 nm for the 2H11/2 transition, and 545 to 555 nm for the 4S3/2 transition. On examining the six curves in this diagram, we note that either of the two highest-power curves will serve adequately for the temperature sensor when a single transition is employed. (Clearly, the constant curve is not useful for temperature measurements, even though it is high.)

Graphic Jump LocationF5 :

The fluorescence power PΔλ generated by the sensor and transmitted by the filters shown in Fig. 4, as a function of temperature.

When two transitions are used in the power-ratio technique, employing a high-power curve for the numerator and a low-power curve for the denominator should result in the highest ratio. Using the highest- and the lowest-power emission curves results in an even higher value for the ratio. This approach, however, brings into the question the well-accepted idea of maximizing the ratio.

Rather, we propose maximizing the power in each band, even when using the ratio techniques. This is particularly important for low-light-level applications, where a high signal-to-noise ratio in the denominator may be of significantly more importance than a high power ratio, in designing a sensor with high measurement repeatability.45 Using the ratio of two high-power curves will result in the sensor with the highest measurement repeatability and lowest error.

For each filter transmission interval Δλ, the sensor responsivity ℜ may be presented as the ratio of the increase in the power output, ΔPΔλ, to the increase in the temperature, ΔT:Display Formula

2R=ΔPΔλΔT(W/K).

Figure 6 shows the channel responsivity ℜ, in watts per kelvin, of the erbium-doped silica as a function of temperature for the filters whose measured transmission is given in Fig. 4. For the curves with the highest responsivity, its absolute value decreases from approximately 0.28 to 0.13 μW/K for the spectral band from 527 to 537 nm, and from 0.23 to 0.16 μW/K for the band from 545 to 555 nm. In terms of the spectral transitions, these correspond to the available filters for the 2H11/2 and the 4S3/2 transition, respectively. The responsivity with other filters is lower by at least 25%. Again we note that even in a sensor employing the ratio technique the responsivity of each channel has to be maximized rather than looking for a sensor with a high ratio.

Graphic Jump LocationF6 :

The channel responsivity, defined as the ratio of the increase in power output (ΔPΔλ) to the increase in temperature (ΔT), as a function of temperature, for filters shown in Fig. 4.

Figure 7 shows the power ratio of two transitions, delineated by the filter transmission bands, as a function of temperature. When interpreting this set of graphs, it is important to keep in mind that we are presenting here the ratio of the power in a short wavelength interval to the power in a long wavelength interval. This ratio is smaller than 1 for all curves shown and for nearly all wavelengths. That may be seen as an undesirable characteristic. On the other hand, the results have increased repeatability due to small errors in the denominator. Also, its form offers a compact presentation.

Graphic Jump LocationF7 :

The measured power ratio varies nearly linearly with temperature over the temperature interval from 21 to 96°C, with slightly different slopes and nearly linear increase in y intercepts as a function of power ratio.

The power ratio varies about linearly with temperature in the interval from 21 to 96°C, with slightly different slopes for different bands, and nearly linear increase in the y intercepts as a function of power ratio. Table 1 shows the filters defining both bands in two emission peaks, and the parameters that describe the straight line approximating the ratio curve versus temperature: the y-intercept, R(T=0°C), and the slope dR/dT. The asterisk denotes the proposed filter combinations. Their sensitivity is evaluated in the next section and presented in Fig. 8.

Graphic Jump LocationF8 :

The sensitivity of the erbium-doped silica sensor, evaluated as the quotient of the increase in the power ratio in the band, ΔR(PΔλ1/PΔλ2), to the temperature increase ΔT, as a function of temperature.

Table Grahic Jump Location
Table 1 Filter combinations and the parameters that describe the straight line approximating the ratio curve versus temperature: the y intercept, R(T=0°C), and the slope dR/dT. The asterisk denotes the proposed filter combinations, whose sensitivity is reported in Fig. 8.

The highest ratio is achieved on employing the filter pair transmitting spectral bands from 527 to 537 nm and from 535 to 545 nm. It is followed by the pair transmitting from 527 to 537 nm and from 545 to 555 nm. The latter has the highest slope of ratio with respect to temperature. This is important for a sensor design, promising to have the highest sensitivity. Other bands may offer a better ratio, but either their signals will have lower value, or their temperature dependence is nonlinear.

Next, we also evaluate the sensitivity S of the power-ratio sensor. It has been defined as the increase in the power ratio integrated over the spectral intervals, ΔR(P1/P2), divided by the increase in the input temperature, ΔT:Display Formula

3S(Δλ1,Δλ2,T)=dP1(Δλ1,T)dTP2(Δλ2,T)P1(Δλ1,T)dP2(Δλ2,T)dT[P2(Δλ2,T)]2(K1).
Here, the subscripts 1 and 2 refer to the respective spectral channels defined by the filters. In deriving this expression, we canceled the wavelength interval in the numerator and the denominator, because they have the same spectral width in our measurements. Likewise, in a first approximation, we can equate and cancel the quantum efficiencies for wavelength bands separated by less than 30 nm.

Figure 8 shows the sensitivity of the erbium-doped fiber optic temperature sensor as a function of temperature, for five pairs of spectral intervals. The sensor sensitivity with the spectral bands from 527 to 537 nm and from 545 to 555 nm is approximately 0.0065 K−1, and it is nearly independent of temperature. This behavior is highly desirable, indicating a uniformity of the sensor’s performance over its proposed interval of operation. The second-best performance is offered by the spectral bands from 527 to 537 nm and from 535 to 545 nm. Here the sensitivity decreases from 0.008 K−1 at room temperature to about 0.004 K−1 at the high end of the temperature interval.

These results indicate that the optimal filter for the short-wavelength transition (2H11/2) is indeed the 527 to 537-nm one. However, neither of the two usable filters in the second transition is optimal. As noted earlier, neither of them transmits the peak that shifts by less than 10 nm in this temperature interval. This knowledge offers an opportunity to improve the sensor performance by incorporating a filter design tailored for the specific peak position. We may conclude that among the available filters, the filter transmitting the 545 to 555-nm spectral subband is the best one for the second transition (4S3/2).

The sensitivity is constant also when the filter in the short-wavelength transition (2H11/2) is changed to transmitting the 515 to 525-nm band, indicating that the power in the whole band increases about linearly with the temperature. However, the sensitivity is decreased when the transmission subband is shifted away from the fluorescence peak.

We have presented experimental results demonstrating the performance of the erbium-doped silica as a remote fiber-optic temperature sensor, using the fluorescence power-ratio technique. The spectral band from 527 to 537 nm, corresponding to the erbium transition 2H11/2, and the one from 545 to 555 nm corresponding to 4S3/2, are highly sensitive to temperature changes in the interval from 21 to 96°C. The sensitivity of the sensor is approximately 0.6%/K. With a custom-made filter centered on 545 nm, even higher sensitivity is predicted.

Scholl  M. S. , “ Autonomous star field identification using intelligent CCD-based cameras. ,” Opt. Eng. . 33, (1 ), 134–139  ((1994)).
Scholl  M. S. , “ Star field identification algorithm. ,” Opt. Lett. . 18, (3 ), 399–401  ((1993)).
Scholl  M. S. , “ Experimental demonstration of a star field identification algorithm. ,” Opt. Lett. . 18, (3 ), 402–404  ((1993)).
Scholl  M. S. , “ Optical processing for semi-autonomous terminal navigation and docking. ,” Appl. Opt. . 32, (26 ), 5049–5055  ((1993)).
Scholl  M. S. , and Eberlein  S. , “ Site characterization with robotic sample acquisition systems. ,” Opt. Eng. . 32, (4 ), 840–846  ((1993)).
Scholl  M. S. , “ Star field identification for autonomous attitude determination. ,” J. Guid. Control Dyn. . 18, (1 ), 61–65  ((1995)).
Scholl  M. S. , “ Design parameters for a two-mirror telescope for stray-light sensitive infrared applications. ,” Infrared Phys. Technol. . 37, , 251–257  ((1996)).
Scholl  M. S. , “ Stray light issues for background-limited far-infrared telescope operation. ,” Opt. Eng. . 33, (3 ), 681–684  ((1994)).
Scholl  M. S. , “ Infrared signal generated by a planet outside the solar system discriminated by rotating rotationally-shearing interferometer. ,” Infrared Phys. Technol. . 37, , 307–312  ((1996)).
Scholl  M. S. , “ Signal generated by an extra-solar-system planet detected by a rotating rotationally-shearing interferometer. ,” J. Opt. Soc. Am. A . 13, (7 ), 1584–1592  ((1996)).
Scholl  M. S. , “ Errors in radiance simulation and scene discrimination. ,” Appl. Opt. . 21, (10 ), 1839–1843  ((1982)).
Scholl  M. S. , and Wolfe  W. L. , “ An infrared target design—Fabrication considerations. ,” Appl. Opt. . 20, (12 ), 2143–2152  ((1981)).
Scholl  M. S. , “ Thermal considerations in the design of a dynamic IR target. ,” Appl. Opt. . 21, (4 ), 660–667  ((1982)).
Scholl  M. S. , “ Spatial and temporal effects due to target irradiation: a study. ,” Appl. Opt. . 21, (9 ), 1615–1620  ((1982)).
Scholl  M. S. , “ Target temperature distribution generated and maintained by a scanning laser beam. ,” Appl. Opt. . 21, (12 ), 2146–2152  ((1982)).
Scholl  M. S. , “ Measured spatial properties of the cw Nd-YAG laser beam. ,” Appl. Opt. . 19, (21 ), 3655–3659  ((1980)).
Scholl  M. S. , “ Temperature calibration of an infrared radiation source. ,” Appl. Opt. . 19, (21 ), 3622–3625  ((1980)).
Collins  S. F. , , Baxter  G. B. , , Wade  S. A. , , Sun  T. , , Grattan  K. T. V. , , Zhang  Z. Y. , , and Palmer  A. W. , “ Comparison of fluorescence-based temperature sensor schemes: theoretical analysis and experimental validation. ,” J. Appl. Phys. . 84, (9 ), 4649–4654  ((1998)).
Arnaud  A. , , Forsyth  D. I. , , Sun  T. , , Zhang  Z. Y. , , and Grattan  K. T. V. , “ Strain and temperature effects on erbium-doped fiber for decay-time based sensing. ,” Rev. Sci. Instrum. . 71, (1 ), 104–108  ((2000)).
Wade  S. A. , , Forsyth  D. I. , , Grattan  K. T. V. , , and Guofu  Q. , “ Fiber optic sensor for dual measurements of temperature and strain using a combined fluorescence lifetime decay and fiber Bragg grating technique. ,” Rev. Sci. Instrum. . 72, (8 ), 3186–3190  ((2001)).
Aizawa  H. , , Ohishi  N. , , Ogawa  S. , , Watanabe  E. , , Katsumata  T. , , Komuro  S. , , Morikawa  T. , , and Toba  E. , “ Characteristics of chromiumdoped spinel crystals for fiber-optic thermometer applications. ,” Rev. Sci. Instrum. . 73, (8 ), 3089–3092  ((2002)).
Fernicola  V. C. , , Rosso  L. , , Galleano  R. , , Sun  T. , , Zhang  Z. Y. , , and Grattan  K. T. V. , “ Investigations on exponential lifetime measurements for fluorescence thermometry. ,” Rev. Sci. Instrum. . 71, (7 ), 2938–2943  ((2000)).
Wade  S. A. , , Baxter  D. W. , , Collins  S. F. , , Grattan  K. T. V. , , and Sun  T. , “ Simultaneous strain-temperature measurement using fluorescence from Yb-doped silica fiber. ,” Rev. Sci. Instrum. . 71, (6 ), 2267–2269  ((2000)).
Sun  T. , , Zhang  Z. Y. , , and Grattan  K. T. V. , “ Erbium/ytterbium fluorescence based fiber optic temperature sensor system. ,” Rev. Sci. Instrum. . 71, (11 ), 4017–4021  ((2000)).
Scholl  M. S. , and Trimmier  J. R. , “ Luminescence of YAG:TM:Tb. ,” J. Electrochem. Soc. . 133, (3 ), 643–648  ((1986)).
Alasaarela  I. , and Karioja  P. , “ Comparison of distributed fiber optic sensing methods for location and quantity information measurements. ,” Opt. Eng. . 41, (1 ), 181–189  ((2002)).
Wickersheim  K. A. , and Hyatt  W. D. , “ Commercial applications of fiber optic temperature measurements. ,” in Fiber Optic Sensors IV , R. T. Kersten, Ed., Proc. SPIE . 1267, , 84–96  ((1990)).
F. C. Allard, Fiber Optic Handbook for Engineers and Scientists , McGraw-Hill, New York (1990).
Choi  H. S. , , Taylor  H. F. , , and Lee  C. E. , “ High-performance fiber optic temperature sensor using low-coherence interferometry. ,” Opt. Lett. . 22, (23 ), 1814–1816  ((1997)).
Zhang  Z.-Y. , , Grattan  K. T. V. , , Palmer  A. W. , , and Meggitt  B. T. , “ Thulium-doped intrinsic fiber optic sensor for high temperature measurements (>1100 °C). ,” Rev. Sci. Instrum. . 69, , 3210–3214  ((1998)).
Sun  T. , , Zhang  Z. Y. , , Grattan  K. T. V. , , and Palmer  A. W. , “ Ytterbium-based fluorescence decay time fiber optic temperature sensor systems. ,” Rev. Sci. Instrum. . 69, , 4179–4185  ((1998)).
Sun  T. , , Zhang  Z. Y. , , Grattan  K. T. V. , , Palmer  A. W. , , and Collins  S. F. , “ Temperature dependence of the fluorescence lifetime in Pr3+:ZBLAN glass for fiber optic thermometry. ,” Rev. Sci. Instrum. . 68, , 3447–3451  ((1997)).
E. Snitzer, W. W. Morey, and W. H. Glenn, “Fiber optic rare earth temperature sensors,” in Proc. 1st Int. Conf. on Optical Fiber Sensors , pp. 79–82, London (1983).
Berthou  H. , and Jorgensen  C. K. , “ Optical-fiber temperature sensor based on upconversion-excited state fluorescence. ,” Opt. Lett. . 15, , 1100–1102  ((1990)).
Farries  M. C. , , Fernmann  M. E. , , Laming  R. I. , , Poole  S. B. , , Payne  D. N. , , and Leach  A. P. , “ Distributed temperature sensor using Nd3+-doped optical fibre. ,” Electron. Lett. . 22, , 418–419  ((1986)).
Dos Santos  P. V. , , Araujo  M. T. , , Gouveia-Neto  A. S. , , Madeiros Neto  J. A. , , and Sombra  A. S. B. , “ Optical temperature sensing using upconversion fluorescence emission in Er3+/Yb3+-codoped chalcogenide glass. ,” Appl. Phys. Lett. . 73, , 578–580  ((1998)).
Maurice  E. , , Wade  S. A. , , Collins  S. F. , , Monnom  G. , , and Baxter  G. W. , “ Self-referenced point temperature sensor based on a fluorescence intensity ratio in Yb3+-doped silica fiber. ,” Appl. Opt. . 36, (31 ), 8264–8269  ((1997)).
Maurice  E. , , Monnom  G. , , Ostrowsky  D. B. , , and Baxter  G. W. , “ 1.2-μm transitions in erbium-doped fibers: the possibility of quasi-distributed temperature sensors. ,” Appl. Opt. . 34, (21 ), 4196–4199  ((1995)).
Maurice  E. , , Monnom  G. , , Dussardier  B. , , Saissy  A. , , Ostrowsky  D. B. , , and Baxter  G. W. , “ Erbium doped silica fibers for intrinsic fiber optic temperature sensors. ,” Appl. Opt. . 34, , 8019–8025  ((1995)).
Maurice  E. , , Monnom  G. , , Ostrowsky  D. B. , , and Baxter  G. W. , “ High dynamic interval temperature point sensor using green fluorescence intensity ratio in erbium-doped silica fiber. ,” J. Lightwave Technol. . 13, (7 ), 1349–1353  ((1995)).
Krug  P. A. , , Sceats  M. G. , , Atkins  G. R. , , Guy  S. C. , , and Poole  S. B. , “ Intermediate excited-state absorption in erbium-doped fiber strongly pumped at 980 nm. ,” Opt. Lett. . 16, , 1976–1978  ((1991)).
Maurice  E. , , Monnom  G. , , Saissy  A. , , Ostrowsky  D. B. , , and Baxter  G. W. , “ Thermalization effects between upper levels of green fluorescence in Er-doped silica fibers. ,” Opt. Lett. . 19, , 990–992  ((1994)).
Atkins  C. G. , , Armitage  J. R. , , Wyatt  R. , , Ainslie  B. J. , , and Craig Ryan  S. P. , “ Pump excited state absorption in Er3+ doped optical fibers. ,” Opt. Commun. . 73, , 217–222  ((1989)).
Betts  R. A. , , Kuhl  F. F. , , Kwok  T. M. , , and Zheng  G. F. , “ Optical amplifiers based on phosphorus co-doped rare-earth-doped optical fibers. ,” Int. J. Optoelectron. . 6, , 47–64  ((1991)).
G. Keiser, Optical Fiber Communications , 2nd ed., McGraw-Hill (1991). Biographies and photographs of the authors not available.
© 2003 Society of Photo-Optical Instrumentation Engineers

Citation

Gonzalo Paez and Marija Strojnik
"Experimental results of ratio-based erbium-doped-silica temperature sensor", Opt. Eng. 42(6), 1805-1811 (Jun 01, 2003). ; http://dx.doi.org/10.1117/1.1571830


Figures

Graphic Jump LocationF1 :

Generation of visible fluorescence radiation on pumping with near-IR radiation.

Graphic Jump LocationF2 :

(a) Principle of operation of the erbium-doped fiber-optic sensor for remote temperature measurements, employing the fluorescence emission ratio in the wavelength interval 515 to 570 nm. (b) Photograph of the experimental setup to evaluate the performance of the erbium-doped fiber-optic temperature sensor. The temperature-sensing fiber is immersed in a liquid to increase the temperature stability of the measurements.

Graphic Jump LocationF3 :

(a) The measured fluorescence emission spectrum of the erbium-doped silica optical fiber as a function of wavelength is shown over a temperature interval from 21 to 96°C. (b) The temperature dependence of the fluorescence spectrum, defined as the ratio of the spectral signal Pout(λ,T) to the spectral signal at the reference temperature of 21°C, Pout(λ,T=21°C), as a function of wavelength.

Graphic Jump LocationF4 :

Measured transmission of the actual filters used in the experimental setup. [See filters F1 and F2 in Fig. 2(a).]

Graphic Jump LocationF5 :

The fluorescence power PΔλ generated by the sensor and transmitted by the filters shown in Fig. 4, as a function of temperature.

Graphic Jump LocationF6 :

The channel responsivity, defined as the ratio of the increase in power output (ΔPΔλ) to the increase in temperature (ΔT), as a function of temperature, for filters shown in Fig. 4.

Graphic Jump LocationF7 :

The measured power ratio varies nearly linearly with temperature over the temperature interval from 21 to 96°C, with slightly different slopes and nearly linear increase in y intercepts as a function of power ratio.

Graphic Jump LocationF8 :

The sensitivity of the erbium-doped silica sensor, evaluated as the quotient of the increase in the power ratio in the band, ΔR(PΔλ1/PΔλ2), to the temperature increase ΔT, as a function of temperature.

Tables

Table Grahic Jump Location
Table 1 Filter combinations and the parameters that describe the straight line approximating the ratio curve versus temperature: the y intercept, R(T=0°C), and the slope dR/dT. The asterisk denotes the proposed filter combinations, whose sensitivity is reported in Fig. 8.

References

Scholl  M. S. , “ Autonomous star field identification using intelligent CCD-based cameras. ,” Opt. Eng. . 33, (1 ), 134–139  ((1994)).
Scholl  M. S. , “ Star field identification algorithm. ,” Opt. Lett. . 18, (3 ), 399–401  ((1993)).
Scholl  M. S. , “ Experimental demonstration of a star field identification algorithm. ,” Opt. Lett. . 18, (3 ), 402–404  ((1993)).
Scholl  M. S. , “ Optical processing for semi-autonomous terminal navigation and docking. ,” Appl. Opt. . 32, (26 ), 5049–5055  ((1993)).
Scholl  M. S. , and Eberlein  S. , “ Site characterization with robotic sample acquisition systems. ,” Opt. Eng. . 32, (4 ), 840–846  ((1993)).
Scholl  M. S. , “ Star field identification for autonomous attitude determination. ,” J. Guid. Control Dyn. . 18, (1 ), 61–65  ((1995)).
Scholl  M. S. , “ Design parameters for a two-mirror telescope for stray-light sensitive infrared applications. ,” Infrared Phys. Technol. . 37, , 251–257  ((1996)).
Scholl  M. S. , “ Stray light issues for background-limited far-infrared telescope operation. ,” Opt. Eng. . 33, (3 ), 681–684  ((1994)).
Scholl  M. S. , “ Infrared signal generated by a planet outside the solar system discriminated by rotating rotationally-shearing interferometer. ,” Infrared Phys. Technol. . 37, , 307–312  ((1996)).
Scholl  M. S. , “ Signal generated by an extra-solar-system planet detected by a rotating rotationally-shearing interferometer. ,” J. Opt. Soc. Am. A . 13, (7 ), 1584–1592  ((1996)).
Scholl  M. S. , “ Errors in radiance simulation and scene discrimination. ,” Appl. Opt. . 21, (10 ), 1839–1843  ((1982)).
Scholl  M. S. , and Wolfe  W. L. , “ An infrared target design—Fabrication considerations. ,” Appl. Opt. . 20, (12 ), 2143–2152  ((1981)).
Scholl  M. S. , “ Thermal considerations in the design of a dynamic IR target. ,” Appl. Opt. . 21, (4 ), 660–667  ((1982)).
Scholl  M. S. , “ Spatial and temporal effects due to target irradiation: a study. ,” Appl. Opt. . 21, (9 ), 1615–1620  ((1982)).
Scholl  M. S. , “ Target temperature distribution generated and maintained by a scanning laser beam. ,” Appl. Opt. . 21, (12 ), 2146–2152  ((1982)).
Scholl  M. S. , “ Measured spatial properties of the cw Nd-YAG laser beam. ,” Appl. Opt. . 19, (21 ), 3655–3659  ((1980)).
Scholl  M. S. , “ Temperature calibration of an infrared radiation source. ,” Appl. Opt. . 19, (21 ), 3622–3625  ((1980)).
Collins  S. F. , , Baxter  G. B. , , Wade  S. A. , , Sun  T. , , Grattan  K. T. V. , , Zhang  Z. Y. , , and Palmer  A. W. , “ Comparison of fluorescence-based temperature sensor schemes: theoretical analysis and experimental validation. ,” J. Appl. Phys. . 84, (9 ), 4649–4654  ((1998)).
Arnaud  A. , , Forsyth  D. I. , , Sun  T. , , Zhang  Z. Y. , , and Grattan  K. T. V. , “ Strain and temperature effects on erbium-doped fiber for decay-time based sensing. ,” Rev. Sci. Instrum. . 71, (1 ), 104–108  ((2000)).
Wade  S. A. , , Forsyth  D. I. , , Grattan  K. T. V. , , and Guofu  Q. , “ Fiber optic sensor for dual measurements of temperature and strain using a combined fluorescence lifetime decay and fiber Bragg grating technique. ,” Rev. Sci. Instrum. . 72, (8 ), 3186–3190  ((2001)).
Aizawa  H. , , Ohishi  N. , , Ogawa  S. , , Watanabe  E. , , Katsumata  T. , , Komuro  S. , , Morikawa  T. , , and Toba  E. , “ Characteristics of chromiumdoped spinel crystals for fiber-optic thermometer applications. ,” Rev. Sci. Instrum. . 73, (8 ), 3089–3092  ((2002)).
Fernicola  V. C. , , Rosso  L. , , Galleano  R. , , Sun  T. , , Zhang  Z. Y. , , and Grattan  K. T. V. , “ Investigations on exponential lifetime measurements for fluorescence thermometry. ,” Rev. Sci. Instrum. . 71, (7 ), 2938–2943  ((2000)).
Wade  S. A. , , Baxter  D. W. , , Collins  S. F. , , Grattan  K. T. V. , , and Sun  T. , “ Simultaneous strain-temperature measurement using fluorescence from Yb-doped silica fiber. ,” Rev. Sci. Instrum. . 71, (6 ), 2267–2269  ((2000)).
Sun  T. , , Zhang  Z. Y. , , and Grattan  K. T. V. , “ Erbium/ytterbium fluorescence based fiber optic temperature sensor system. ,” Rev. Sci. Instrum. . 71, (11 ), 4017–4021  ((2000)).
Scholl  M. S. , and Trimmier  J. R. , “ Luminescence of YAG:TM:Tb. ,” J. Electrochem. Soc. . 133, (3 ), 643–648  ((1986)).
Alasaarela  I. , and Karioja  P. , “ Comparison of distributed fiber optic sensing methods for location and quantity information measurements. ,” Opt. Eng. . 41, (1 ), 181–189  ((2002)).
Wickersheim  K. A. , and Hyatt  W. D. , “ Commercial applications of fiber optic temperature measurements. ,” in Fiber Optic Sensors IV , R. T. Kersten, Ed., Proc. SPIE . 1267, , 84–96  ((1990)).
F. C. Allard, Fiber Optic Handbook for Engineers and Scientists , McGraw-Hill, New York (1990).
Choi  H. S. , , Taylor  H. F. , , and Lee  C. E. , “ High-performance fiber optic temperature sensor using low-coherence interferometry. ,” Opt. Lett. . 22, (23 ), 1814–1816  ((1997)).
Zhang  Z.-Y. , , Grattan  K. T. V. , , Palmer  A. W. , , and Meggitt  B. T. , “ Thulium-doped intrinsic fiber optic sensor for high temperature measurements (>1100 °C). ,” Rev. Sci. Instrum. . 69, , 3210–3214  ((1998)).
Sun  T. , , Zhang  Z. Y. , , Grattan  K. T. V. , , and Palmer  A. W. , “ Ytterbium-based fluorescence decay time fiber optic temperature sensor systems. ,” Rev. Sci. Instrum. . 69, , 4179–4185  ((1998)).
Sun  T. , , Zhang  Z. Y. , , Grattan  K. T. V. , , Palmer  A. W. , , and Collins  S. F. , “ Temperature dependence of the fluorescence lifetime in Pr3+:ZBLAN glass for fiber optic thermometry. ,” Rev. Sci. Instrum. . 68, , 3447–3451  ((1997)).
E. Snitzer, W. W. Morey, and W. H. Glenn, “Fiber optic rare earth temperature sensors,” in Proc. 1st Int. Conf. on Optical Fiber Sensors , pp. 79–82, London (1983).
Berthou  H. , and Jorgensen  C. K. , “ Optical-fiber temperature sensor based on upconversion-excited state fluorescence. ,” Opt. Lett. . 15, , 1100–1102  ((1990)).
Farries  M. C. , , Fernmann  M. E. , , Laming  R. I. , , Poole  S. B. , , Payne  D. N. , , and Leach  A. P. , “ Distributed temperature sensor using Nd3+-doped optical fibre. ,” Electron. Lett. . 22, , 418–419  ((1986)).
Dos Santos  P. V. , , Araujo  M. T. , , Gouveia-Neto  A. S. , , Madeiros Neto  J. A. , , and Sombra  A. S. B. , “ Optical temperature sensing using upconversion fluorescence emission in Er3+/Yb3+-codoped chalcogenide glass. ,” Appl. Phys. Lett. . 73, , 578–580  ((1998)).
Maurice  E. , , Wade  S. A. , , Collins  S. F. , , Monnom  G. , , and Baxter  G. W. , “ Self-referenced point temperature sensor based on a fluorescence intensity ratio in Yb3+-doped silica fiber. ,” Appl. Opt. . 36, (31 ), 8264–8269  ((1997)).
Maurice  E. , , Monnom  G. , , Ostrowsky  D. B. , , and Baxter  G. W. , “ 1.2-μm transitions in erbium-doped fibers: the possibility of quasi-distributed temperature sensors. ,” Appl. Opt. . 34, (21 ), 4196–4199  ((1995)).
Maurice  E. , , Monnom  G. , , Dussardier  B. , , Saissy  A. , , Ostrowsky  D. B. , , and Baxter  G. W. , “ Erbium doped silica fibers for intrinsic fiber optic temperature sensors. ,” Appl. Opt. . 34, , 8019–8025  ((1995)).
Maurice  E. , , Monnom  G. , , Ostrowsky  D. B. , , and Baxter  G. W. , “ High dynamic interval temperature point sensor using green fluorescence intensity ratio in erbium-doped silica fiber. ,” J. Lightwave Technol. . 13, (7 ), 1349–1353  ((1995)).
Krug  P. A. , , Sceats  M. G. , , Atkins  G. R. , , Guy  S. C. , , and Poole  S. B. , “ Intermediate excited-state absorption in erbium-doped fiber strongly pumped at 980 nm. ,” Opt. Lett. . 16, , 1976–1978  ((1991)).
Maurice  E. , , Monnom  G. , , Saissy  A. , , Ostrowsky  D. B. , , and Baxter  G. W. , “ Thermalization effects between upper levels of green fluorescence in Er-doped silica fibers. ,” Opt. Lett. . 19, , 990–992  ((1994)).
Atkins  C. G. , , Armitage  J. R. , , Wyatt  R. , , Ainslie  B. J. , , and Craig Ryan  S. P. , “ Pump excited state absorption in Er3+ doped optical fibers. ,” Opt. Commun. . 73, , 217–222  ((1989)).
Betts  R. A. , , Kuhl  F. F. , , Kwok  T. M. , , and Zheng  G. F. , “ Optical amplifiers based on phosphorus co-doped rare-earth-doped optical fibers. ,” Int. J. Optoelectron. . 6, , 47–64  ((1991)).
G. Keiser, Optical Fiber Communications , 2nd ed., McGraw-Hill (1991). Biographies and photographs of the authors not available.

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Related Book Chapters

Topic Collections

Advertisement
  • Don't have an account?
  • Subscribe to the SPIE Digital Library
  • Create a FREE account to sign up for Digital Library content alerts and gain access to institutional subscriptions remotely.
Access This Article
Sign in or Create a personal account to Buy this article ($20 for members, $25 for non-members).
Access This Proceeding
Sign in or Create a personal account to Buy this article ($15 for members, $18 for non-members).
Access This Chapter

Access to SPIE eBooks is limited to subscribing institutions and is not available as part of a personal subscription. Print or electronic versions of individual SPIE books may be purchased via SPIE.org.