The emissive properties of a great number of luminescent materials can alter as a result of changes in temperature, pressure, or another chemical species present. For a long time, these properties have been used in the development of luminescence-based detectors. Besides chemical sensing, Luminescence thermometry, alongside chemical sensing, is one of the most frequently applied varieties of sensing.
In contrast to other methodologies, there is no need for a macroscopic probe in engage in physical contact with the area to be studied. This offers great benefits in instances where the sample is tricky to access, as well as in biological and microscopy samples.
To illustrate, luminescent nanoparticles can be functionalized to enter a biological target, and fluorescence microscopy can subsequently be employed to investigate the temperature at various points with great precision. This type of nanometric thermometry could be applied with great success in the medical field, for tasks such as imaging cancer cells with higher than average temperatures.1
Alterations in intensity, linewidth, photoluminescence lifetime, or spectral shifts can all offer a basis for luminescence thermometry. Lanthanide ions are frequently made use of in temperature detection applications, due to their steadiness and fine spectral features that allow for easy detection of changes.2
Additionally, lanthanide-doped materials exhibit upconversion luminescence, meaning they can be stimulated with near-infrared (NIR) light, and radiate within the visible portion of the spectrum. NIR excitation prompts reduced levels of self-absorption and scattering by biological tissue, and thus it is simpler to carry out remote excitation.
In light of this property, a growing number of temperature bioimaging studies by means of upconversion nanoparticles (UCNPs), inorganic nanoparticles doped with lanthanide ions are taking place.3
In Figure 1, the upconversion mechanism for a common lanthanide-doped phosphor, NaY0.77Yb0.20Er0.03F4 can be seen. A minimum of two 980 nm photons are necessary to prompt emission in the visible spectrum.
Figure 1: Schematic of upconversion mechanism in NaY0.77Yb0.20Er0.03F4, showing green and red luminescence transtitions. Grey arrows indicate nonradiative processes.
Aside from direct stimulation of the Er3+ ion, energy transfer from excited Yb3+ to excited states of Er3+ is detected. Depending on the upper Er3+ level in the transition, emission can be seen in either the blue, green and red areas of the visible spectrum.
Upconversion thermometry frequently concentrates on the two transitions emitting at 525 nm and 540 nm, i.e. 2H11/2 → 4I15/2 and 4S3/2 → 4I15/2. The 2H11/2 and 4S3/2 levels are closely spaced in energy, essentially placing them in thermal balance. As a result, their population ratio can be communicated through a Boltzmann distribution:
Where Ni is the population in level i, ΔE is the spacing between the levels, k is Boltzmann’s constant and C is a constant, contingent on degeneracies.
On this basis, the ratio of 525 nm and 540 nm luminescence intensities, RHS, can be used to surmise the ratio of 2H11/2 to 4S3/2, which in turn, offers the temperature of the sample. The Edinburgh Instruments FLS1000, or other photoluminescence spectrometers can be engaged to accurately depict this ratio as a function of temperature.
As well as broadly applied cryostats, the FLS1000 can be attached to temperature-managed stages through fibers. Stages such as this allow for the study of a single sample in both the FLS1000 and a microscope, without any transitional sample mounting steps.
This application note summarizes temperature-dependent upconversion luminescence results for NaY0.77Yb0.20Er0.03F4 gathered with the use of an FLS1000 spectrometer coupled to a temperature stage.
Materials and Methods
A commercial upconversion phosphor, NaY0.77Yb0.20Er0.03F4 (Sigma Aldrich) was positioned as delivered on a quartz tray within a Linkam HFS350EV-PB4 stage. The stage was optically linked to an FLS1000 Photoluminescence Spectrometer through the use of a custom mount (N-Linkam) and a fiber coupling accessory for the sample chamber.
Steady state photoluminescence spectra were gathered using a Xe2 lamp for excitation at 980 nm, the excitation wavelength of Yb3+. The excitation power sufficiently low to avoid inducing sample heating, a phenomenon frequently encountered with laser excitation.4 The FLS1000 was prepared with double monochromators as well as a standard PMT-900 detector.
With the use of a pulsed Xe lamp (µF2) as the excitation source, time-resolved photoluminescence decays were measured within the same device. Multi-channel scaling (MCS) electronics were engaged to obtain the decays, and the trial data was fitted using the Edinburgh Instruments FAST software bundle.
Results and Discussion
With the help of the temperature mapping option included in the Fluoracle software, red and green upconversion intensities were categorized from -100 °C to +80 °C in intervals of 20 °C. Figures 2 and 3 respectively demonstrate the results for green upcon version) and red upconversion.
Figure 2: Temperature emission map of NaY0.77Yb0.20Er0.03F4 measured with Linkam stage in FLS1000, green upconversion emission. Conditions: λex = 980 nm, Δλex = 10 nm, Δλem = 0.20 nm, step size = 0.10 nm, integration time = 1 s/step. The inset shows the 2H11/ 4I15/2 region in detail.
Figure 3: Temperature emission map of NaY0.77Yb0.20Er0.03F4 measured with Linkam stage in FLS1000, green upconversion emission. Conditions: λex = 980 nm, Δλex = 10 nm, Δλem = 0.20 nm, step size = 0.10 nm, integration time = 1 s/step.
Two Er3+ transitions add to the green upconversion emission detected in Figure 2: 4S3/2 → 4I15/2 and 2H11/2 → 4I15/2. The red emission band in Figure 2 matches up to the 4F9/2 → 4I15/2 transition.
The emission strength of 4S3/2 → 4I15/2 and 4F9/2 → 4I15/2 becomes lower with temperature, but the 2H11/2 → 4I15/2 band demonstrates differing behavior: it grows monotonically up to 273 K, at which point the growth decelerates.
As the temperature gets higher, nonradiative relaxation processes lessen the total upconversion luminescence. The forces at work within this temperature reduction progression can be examined using a temperature lifetime map, like that displayed in Figure 4, in which there is a noticeable reduction in photoluminescence lifetime, from 640 µs to 530 µs, as T increases.
Figure 4: Temperature lifetime map of NaY0.77Yb0.20Er0.03F4 measured with Linkam stage in FLS1000, 2H11/2 → 4I15/2 emission. Conditions: λex = 980 nm, Δλex = 15 nm, λem = 541 nm, Δλem = 10 nm, repetition rate = 100 Hz, acquisition time = 5 minutes/decay. The red and blue curves show the decays at -100 °C and 40 °C, respectively.
Examining Figures 2 and 3 once again, relaxation from 4S3/2, 2H11/2 to 4F9/2 results in a comparative increase in red emission. This is demonstrated by plotting the ratio of integrated intensities as a function of temperature Rrg, seen in Figure 5 (a). As outlined earlier in this piece, the ratio of 2H11/2 → 4I15/2 to 4S3/2 → 4I15/2, RHS, is a reliable gauge of temperature.
Figure 5 (b) plots RHS alongside temperature, while Figure 5 (c) displays this same set of data in logarithmic form. Curiously, RHS does not track a straight Boltzmann plot: further relaxation mechanisms occur and stimulate the “slow growth” behavior of 4S3/2 → 4I15/2 at increased temperatures.
Figure 5: Integrated intensity ratios of upconversion emission bands as a function of temperature: (a) red vs green ratio, (b) 2H11/2 → 4I15/2 vs 4S3/2 → 4I15/2 ratio, (c) logarithmic (Boltzmann) plot of the dataset in (b). A linear fit to the first part of the Boltzmann plot is shown in (c).
This consolidates the findings of earlier reports5,6 and demonstrates the intricate dynamics of the upconversion progression: nonradiative relaxation from 2H11/2 to 4S3/2 increases in importance at greater temperatures, so that the population ratio is not equivalent to RHS. A further key factor to note is that the RHS vs T is greatly dependent on the size of the particles.
To demonstrate the idea of upconversion thermometry, the low temperature area of the curve is fixed to a straight Boltzmann plot in Figure 5 (c). From this plot, the comparative sensitivity of the luminescence thermometry system, S, can be determined. This is a helpful parameter with which to evaluate a luminescence thermometer system, and is determined in the following manner:
Taking the slope in Figure 5 as -∆E/k, the sensitivity at 20 °C is 1.0% K-1. This result aligns with comparable upconversion thermometry systems.3
Using an FLS1000 Photoluminescence Spectrometer with a Linkam stage, the upconversion luminescence strength and lifespan of NaY0.77Yb0.20Er0.03F4 has been categorized as a function of temperature. The ratio of 2H11/2 → 4I15/2 to 4S3/2 → 4I15/2 can be used to investigate the temperature in luminescence thermometry experiments with a sensitivity of 1.0% K-1.
The fiber coupling accessory for the Linkam temperature stage allows users to easily switch between microscope characterization and luminescence measurements, with no requirement for transitional sample transfer steps.
 C. D. S. Brites, et al., Nanoscale 4, 4799-4829 (2012)
 M. D. Dramianin, Methods Appl. Fluoresc. 4, 042001 (2016)
 M. González-Béjar and J. Pérez-Prieto, Methods Appl. Fluoresc. 3, 042002 (2015)
 S. Zhou, et al., Optics Communications 291, 138-142 (2013)
 X. Bai, et al., J. Phys. Chem. C 111, 13611-13617 (2007)
 W. Yu, et al., Dalton Trans. 43, 6139-6147 (2014)
This information has been sourced, reviewed and adapted from materials provided by Edinburgh Instruments.
For more information on this source, please visit Edinburgh Instruments.