Time-resolved universal temperature measurements using NaYF4:Er3+,Yb3+ upconverting nanoparticles in an electrospray jet

Abstract

Hexagonal upconverting nanoparticles (UCNPs) of NaYF₄:Er³⁺,Yb³⁺ (ca. 300 nm) have been widely used to measure the temperature at the nanoscale using luminescence ratio thermometry. However, several factors limit their applications. For example, changes in the peak shape, mainly is the S-band emission, hinders their ability to be used as a universal temperature sensor. Herein, we introduce a universal calibration protocol for NaYF₄:Er³⁺,Yb³⁺ upconverting nanoparticles that is robust to environmental changes and gives a precise temperature measurement. We used this new procedure to calculate the temperature profile inside a Taylor cone generated with an electrospray jet. Inside the Taylor cone the fluid velocity increases toward the tip of the cone. A constant acquisition length leads to a decrease in excitation and acquisition time. This decrease in excitation time causes a peak shape change that corrupts the temperature measurement if the entire peak shape is integrated in the calibration. Our universal calibration circumvents this problem and can be used for time-resolved applications. The temperature at the end of the Taylor cone increases due to the creation of a whispering gallery mode cavity with 980 nm excitation. We use time-resolved energy balance equations to support our optical temperature measurements inside the Taylor cone. We believe that the findings of this paper provide a foundation for time-resolved temperature measurements using NaYF₄:Er³⁺,Yb³⁺ upconverting nanoparticles and can be used to understand temperature-dependent reactions such as protein unfolding inside microjet/microdroplets and microfluidic systems.

Keywords: electrospray; microjet; nanothermometry; temperature measurement; time-resolved measurement; upconverting nanoparticles.

Figure 1

(A) SEM image of synthesized NaYF₄:Er³⁺,Yb³⁺ UCNPs using thermal decomposition method. (B) An image of a single nanoparticle. The nanoparticle has a hexagonal shape with a diameter of ca. 300 nm and a thickness of ca. 80 nm.

Figure 2

(A) Upconversion emission spectra of UCNPs at different temperatures. A temperature-dependent green emission is measured for a temperature range of 306–493 K. (B) A plot of ln(H/S) vs 1/T for calibration of UCNPs. The H- (514–534 nm) and S- (536–548 nm) band area selected for the calibration are shaded in green and brown, respectively, in (A). The slope of the line calculated is −1182 ± 8 K and the intercept is 3.002 ± 0.021.

Figure 3

Measurement of temperature of UCNPs at different laser intensities of 980 nm. (A) Photoluminescence spectra of a single cluster of UCNPs of ca. 1 μm size at different laser intensities. With the decrease in intensity, there is also decrease in an integrated area of S-band after 548 nm. The inset shows an enlarged view of a change in peak intensity with respect to laser intensity. (B) A plot of an average temperature plotted as a function of 980 nm laser intensity. The temperature calculated remains constant with a decrease in standard deviation upon an increase in laser intensity. An emission spectrum of a given color in (A) represents a temperature in the plot in (B).

Figure 4

A plot of the temperature of water calculated as a function of the 980 nm laser intensity using two different nanothermometers, namely erbium oxide (Er₂O₃) and UCNPs. The temperature of UCNPs is calculated using both the full S-band from 535 to 570 nm and the partial S-band from 535 to 548 nm. The temperature calculated using S_535–548 is in good agreement with our reference thermal sensor using Er₂O₃ nanoparticles [37].

Figure 5

(A) An image showing an electrospray jet of 0.8 pM upconverting solution (yellow lines) from a glass pipette (green lines). The red circle represents the tip of a Taylor cone. (B) Emission spectra of UCNPs at different regions as marked in panel (A) with respective colors. It shows a marked difference in the nature of spectra of UCNPs when they are in a glass pipette or flowing in a Taylor cone.

Figure 6

(A) A plot of intensity vs wavelength/wavenumber exhibiting whispering gallery modes at the tip of a Taylor cone indicated by a red circle in Figure 5A. The inset represents a peak profile, fitted with Gaussian to calculate the Q-factor. The values in red represent l + 1/2 values calculated using the respective peaks. (B) A plot of wavenumber vs l + 1/2 with slope 413 ± 3 and intercept of −137 ± 158. The cavity radius is 2.87 ± 0.02 µm.

Figure 7

(A) Plot of sampling time and total photoluminescence intensity of UCNPs versus distance from the tip of a Taylor cone (x_tip = 0). Both sampling time and intensity follow an r²-dependence along the jet where r is the radius of the cross section perpendicular to x. Distance x and radius r are related by r = x tanθ where θ is the Taylor cone angle. The sampling time reduces to 2 μs at the tip. (B) A plot of temperature calculated along the trajectory of a Taylor cone at room temperature (296 K) and at elevated temperature (306 K). The sampling time at a given distance is indicated in the figure.

Figure 8

A schematic illustration for a setup of temperature measurement in a Taylor cone, formed by electrospray of an upconverting solution. A 980 nm laser illuminates from the top with a 50× lens. A white light illuminates from the bottom using a 20× lens for imaging. The image on the top right shows a glass capillary pulled to form a ca. 15 μm pipette. The scale bar is 2 mm.