Luminescence thermometry driven by a support vector machine: a strategy toward precise thermal sensing.

Luminescence thermometry is a promising non-contact temperature measurement technique, but improving the precision and reliability of this method remains a challenge. Herein, we propose a thermal sensing strategy based on a machine learning. By using Gd3Ga5O12: Er3+-Yb3+ as the sensing medium, a support vector machine (SVM) is preliminarily adopted to establish the relationship between temperature and upconversion emission spectra, and the sensing properties are discussed through the comparison with luminescence intensity ratio (LIR) and multiple linear regression (MLR) methods. Within a wide operating temperature range (303-853 K), the maximum and the mean measurement errors actualized by the SVM are just about 0.38 and 0.12 K, respectively, much better than the other two methods (3.75 and 1.37 K for LIR and 1.82 and 0.43 K for MLR). Besides, the luminescence thermometry driven by the SVM presents a high robustness, although the spectral profiles are distorted by the interferences within the testing environment, where, however, LIR and MLR approaches become ineffective. Results demonstrate that the SVM would be a powerful tool to be applied on the luminescence thermometry for achieving a high sensing performance.

Valley-to-peak intensity ratio thermometry based on the red upconversion emission of Er3+.

Upon 976 nm diode laser excitation, the temperature dependence of the red upconversion emission of Er3+ in CaWO4:Yb3+/Er3+ phosphor was studied from 298 to 478 K. The spectrum was verified to consist of two Stark components originating from two Stark sublevels of 4F9/2 excited state to 4I15/2 ground state of Er3+. The valley-to-peak intensity ratio (VPR) of this double-peak spectrum was found to increase linearly with the rise of temperature. The maximum relative sensitivity of this VPR method was obtained to be about 0.20% K-1 at 298 K. Moreover, a study on the power dependence was also performed, suggesting that VPR method is immune to the pump power and is thus suitable for monitoring the temperature.

Optical thermometry based on the red upconversion fluorescence of Er3+ in CaWO4:Yb3+/Er3+ polycrystalline powder.

Based on the fluorescence intensity ratio method, the temperature-sensing behavior through thermally coupled levels (TCL) of the H11/22 and S3/24 states as well as the sub-levels of the F9/24 state of Er3+ has been studied. The thermometry is observed to be dependent on the pump power for the H11/22 and S3/24 states, leading to an error of more than 20 K at 478 K. By utilizing the sub-levels of the F9/24 state, such a problem could be solved. The maximum sensitivity is about 0.15% K-1 at 298 K. This will provide guidance on selecting appropriate and practical TCL for precisely sensing the temperature.

Multifunctional nanoparticles based on the Nd³âº/Yb³âº codoped NaYF4.

Broadband near-infrared luminescence (NIR) from 720 to 950 nm, which is located in the biological window, has been successfully achieved from Nd3+/Yb3+ codoped hexagonal NaYF4 nanoparticles when excited by 980 nm diode laser. Using the fluorescence intensity ratio technique, the temperature sensing behavior of Nd3+ NIR emissions exhibits various advantages over other rare earth ion based nanothermometers. The light-induced thermal loading for the 980 nm excited NaYF4:Nd3+/Yb3+ was also investigated. The results illustrate the multifunctionality of such fluoride nanoparticles, which could simultaneously act as the luminescent nanothermometers and nanoheaters and find potential application in photothermal therapy.

[Application of fluorescence spectra and parallel factor analysis in the classification of edible vegetable oils].

The fluorescence spectra of 22 samples of 8 kinds of edible vegetable oils (soybean oil, maize oil, olive oil, rice oil, peanut oil, walnut oil, sunflower oil and sesame oil) were measured with FS920 fluorescence spectrometer and the fluorescence matrixs (EEMs) were analyzed with parallel factor (PARAFAC) analysis model. To synthesize the capabilities of material characterization and component identification, fluorescence spectra combined with PARAFAC fulfill the classification of vegetable oils. The map feature (peak position, peak value and peak number) was obtained by analyzing three dimensional spectra and con tour maps in the range of emission wavelength from 260 to 750 nm, and excitation wavelengths from 250 to 550 nm. The fluorescent substances (unsaturated fatty acids, vitamin E and its derivatives, chlorophyll and carotenoid) corresponding to spectrum peaks were determined. The factor-number was selected and the components (vitamin E and its derivatives, linoleic acid and linolenic acid, fatty acid oxidation products, vegetable oil oxidation products) corresponding to each factor were ascertained. The four-factor excitation and emission profiles and projection score plots of PARAFAC model were plotted. Different vegetable oils can be characterized and distinguished with the map features of fluorescence spectra and sample projection plots of PARAFAC model. The results demonstrate the capability of the combination of fluorescence spectra technology and four-factor PARAFAC model for differentiating and characterizing vegetable oils.

Optical temperature sensing based on the near-infrared emissions from Nd³âº/Yb³âº codoped CaWO4.

Under a 980 nm diode laser excitation, the near-infrared (NIR) emissions from Nd3+:4F7/2, 4F5/2, and 4F3/2 states in Nd3+/Yb3+ codoped CaWO4 powder were studied at temperatures ranging from 303 to 873 K. As the temperature increased, the NIR luminescence intensity was significantly enhanced and nearly 190-fold enhancement was achieved at 873 K compared with that at 303 K. By using the fluorescence intensity ratio technique, the thermometry behaviors through the NIR emissions were investigated. The results illustrate that the sensitivity and the accuracy achieved here are much higher than temperature sensors based on other rare earth ion doped materials.

[The heating effect of the Er3+/Yb3+ doped Y2O3 nanometer powder by 980 nm laser diode pumping].

The Er3+ and Yb3+ doped Y2O3 Nano powder was prepared by sol-gel method. Based on 2H11/2 --> 4I15/2 and 4S3/2 --> 4I15/2 green conversion luminescence intensity rate of Er3+, the sample surface temperature changes caused by the increase in 980 nm diode laser pump power were studied. The results show that with pump power increasing, the sample surface temperature substantially rises. And the surface temperature reached to 820 K when the pump power was 1 000 mW. The phenomenon plays an important role in the analysis of upconversion process, especially with saturation power. And this feature has a potential application prospect in the biomedicine, soft tissue hole burning as well as the field of temperature sensing materials.

Multispectral plasmon-induced transparency in triangle and nanorod(s) hybrid nanostructures.

We theoretically investigate the optical properties of a hybrid nanostructure consisting of one triangle and one nanorod. A plasmon-induced transparency (PIT) resonance appears in the transmission spectrum, which is ascribed to the induced multipole plasmon mode of the nanorod. Multispectral PIT resonances are observed, when two or more nanorods are put in proximity to the triangle. It is proved that the combined effects of the induced multipole plasmon mode of nanorods and the cavity resonant mode contribute to these PIT peaks. The tunability of PIT on the geometrical parameters is also presented.

Short-wavelength upconversion emissions in Ho3+/Yb3+ codoped glass ceramic and the optical thermometry behavior.

Ho(3+)/Yb(3+) codoped glass ceramic was prepared by melt-quenching and subsequent thermal treatment. Under a 980 nm diode laser excitation, upconversion emissions from Ho(3+) ions centered at 540, 650, and 750 nm were greatly enhanced compared with those in the precursor glass. Especially, the short-wavelength upconversion emissions centered at 360, 385, 418, 445, and 485 nm were successfully obtained in the glass ceramic. An explanation for this phenomenon is given based on the fluorescence decay curve measurements. In addition, an optical temperature sensor based on the blue upconversion emissions from (5)F(2,3)/(3)K(8)→(5)I(8) and (5)F(1)/(5)G(6)→(5)I(8) transitions in Ho(3+)/Yb(3+) codoped glass ceramic has been developed. It was found that by using fluorescence intensity ratio technique, appreciable sensitivity for temperature measurement can be achieved by using the Ho(3+)/Yb(3+) codoped glass ceramic. This result makes the Ho(3+)/Yb(3+) codoped glass ceramic be a promising candidate for sensitive optical temperature sensor with high resolution and good accuracy.