1. Introduction

The luminescent nanoparticles can be used in a wide range of photonic applications, for instance in near-field detection of plasmon polariton waves [1], nanolasers [2], nanolabels [3–6], and thermal nanoimaging [7]. The rare-earth doped oxyfluoride nano-glass-ceramics [8,9] and rare-earth doped fluoride nanoparticles [5,6] are of special interest due to their high, up to 15%, quantum yield of up-conversion luminescence [6,9]. This high quantum yield is due to low phonon energy and short inter-dopant distances in the fluoride nanoparticles hosting the rare-earth dopants [5–9].

Yb³⁺-Er³⁺ co-dopant couple is known for its efficient pumping at 975 nm by commercial laser diodes into the absorption band of Yb³⁺, which has the highest absorption cross-section amongst the rare-earth dopants. An excited Yb³⁺ transfers the energy to a highly luminescent Er³⁺ co-dopant [10]. The high cross-sections of absorption of Yb³⁺ and luminescence of Er³⁺ make the Yb³⁺-Er³⁺ co-dopant couple especially promising for doping of nanoparticles aiming at highly efficient luminescence.

In this work, we have found that a pump at 975 nm by laser diode, in addition to a strong up-conversion luminescence signal [6], also results in a substantial heating, up to several hundred degrees, of the Yb³⁺-Er³⁺ co-doped nanoparticles. The temperature rise in the nanoparticles is estimated based on the measurements of the intensity ratio of up-conversion luminescence bands ²H_11/2→⁴I_15/2 and ⁴S_3/2→⁴I_15/2 of Er³⁺ [7]. The up-conversion visible luminescence allows distant optical detection of heated nanoparticles and their surrounding nano-volumes.

2. Experimental and Results

Transparent 32(SiO₂)9(AlO_1.5)31.5(CdF₂)18.5(PbF₂)5.5(ZnF₂):3.5(REF₃), mol%., nano-glass-ceramics (NGC), have been prepared as described in [8], where RE stands for the Yb³⁺-Er³⁺ co-dopants with a molar ratio Yb³⁺/Er³⁺ equal to 6. The free standing Yb³⁺-Er³⁺ co-doped crystalline nanoparticles have been extracted from the NGC template by chemical etching, using the procedure described in [5]. The typical transmission electron microscope (TEM) images of the nanoparticles have been shown in [6], where chemical formula of the nanoparticles was estimated to be about RE₁₀Pb₂₅F₆₅ with some admixture of oxygen on their surface shell.

The luminescence spectra have been excited by a laser diode operating at 975 nm and up to 260 mW power. The laser beam was Gaussian-shaped as observed on the screen placed in the far field. The laser beam was focused on the sample in the spot of about 0.4 mm diameter using collimating optics of the laser, so that radiation with power of 250 mW resulted in pump power density in the focal point of about 200 W/cm². The power of the laser beam on the sample was tuned from zero to 240 mW using a Glan-Thomson prism by crossing its optical axis with polarisation vector of the laser. The Yb³⁺-Er³⁺ co-doped nanopowder was kept either in ampoules sealed under vacuum or was free-standing in the air while fixed between silica glass plates.

Figure 1 presents a visible green part of up-conversion luminescence spectra. The bands from 510 nm to 535 nm, and from 535 nm to 565 nm in Fig. 1 correspond to the ²H_11/2→⁴I_15/2 and ⁴S_3/2→⁴I_15/2 transitions of Er³⁺, as indicated in the energy level diagram in Fig. 2. We focus on these emission bands since their relative intensity has been shown in numerous papers to be dependent on the temperature of the emitting sample [7, 12–14 and refs therein]. In particular, the formula (1) is valid since the levels ²H_11/2 and ⁴S_3/2 are close to each other and therefore they are in thermal equilibrium described by the Boltzman factor:

where I_H and I_S stand for integral intensity, hν_H and hν_S for quantum energy of the ²H_11/2→⁴I_15/2 and ⁴S_3/2→⁴I_15/2 emission bands, respectively, g_H and g_S stand for degeneracy and A_H an A_S stand for spontaneous emission rates of the ²H_11/2 and ⁴S_3/2 levels, respectively [14]; Δ is a gap between the ²H_11/2 and ⁴S_3/2 levels (indicated in Fig. 2), which is invariant with a change of the dopant host; k is the Boltzmann constant and T is the temperature. The parameters g_H, g_S, hν_H and hν_S are also invariant regardless of the host material, while A_H and A_S show only a minor change with a host [11–13]. Therefore, Eq. (1) transforms to a simple Eq. (2),

where B and C are the constants, B≈1100 K, while C varies between 1.5 to 2.5 [12–14]. It is seen from Eq. (2) that the ratio IH/I_S is defined by the temperature T of emitting sample. However, the exact values of constants B and C should be experimentally found for a specific sample before Eq. (2) could be applied to evaluate the sample temperature precisely.

 

Fig. 1. Up-conversion luminescence spectra of Yb³⁺-Er³⁺ co-doped crystalline nanoparticles aggregated as nanopowder (a) and bulk nano-glass-ceramics (b), excited at 975 nm. Excitation power of laser diode is indicated while intensity of luminescence increases with indicated pump power, respectively. The nanopowder was placed in pyrex ampoule sealed under vacuum, the nano-glass-ceramics was placed in ambient air.

Download Full Size | PDF

Therefore, first we have measured a temperature dependence of the ratio I_H/I_S versus sample temperature T when these emission bands were pumped via the higher laying ⁴F_3/2 absorption level of Er³⁺ (see diagram in Fig. 2). The pump was done by a monochromatised Xe lamp radiation at such power density, that the pump beam does not heat the sample at all. The temperature of sample (nanoparticles aggregated as nanopowder and placed in air between thin silica plates) was controlled by means of optical cryostat. The stars in Fig. 3 show the experimental data in case of such pump, while a straight line is a linear fit to these data. A good linear fit proves that Eq. (2) is indeed valid for the nanopowder sample. Red circles and a black square indicate the values ln(I_H/I_S), when the luminescence was pumped at 975 nm resulting in a substantial heating of the sample. In the latter case, the values ln(I_H/I_S) are specially placed onto the linear fit to evaluate the temperature rise. An evaluated temperature rise is indicated for three selected pump powers for the nanopowder and bulk nano-glass-ceramics samples. It is seen from these selected data, that a 230 mW pump heats the nanopowder to 400°C, while the 230 mW pump heats the bulk nano-glass-ceramics sample only to 90°C, which is comparable to 60°C temperature rise in nanopowder when it is heated with pump of only 12 mW. This proves that laser heating of free-standing nanoparticles (nanopowders) is substantially stronger than for the bulk nano-glass-ceramics sample hosting the same nanoparticles, due to confinement of phonons in the free-standing nanoparticles and, contrary, due to efficient heat removal from nanoparticles embedded in bulk nano-glass-ceramics.

 

Fig. 2. Yb³⁺ and Er³⁺ energy level diagrams. Diode laser excitation transitions at 975 nm to the ²F_5/2 and ⁴I_11/2 levels of Yb³⁺ and Er³⁺ are showed by solid black arrows. Dash red arrows show energy transfer processes between Yb³⁺ and Er³⁺ ions resulting in up-conversion luminescence of Er³⁺ and heating the sample. Wavy lines shows phonon emission processes by Er ³⁺ ions, which mostly heat the sample due to energy mismatch with a levels ⁴F_9/2 and ⁴F_7/2 of Er³⁺. Δ is a gap between the ²H_11/2 and ⁴S_3/2 levels, while green arrows indicate the thermalised emission transitions from these levels.

Download Full Size | PDF

 

Fig. 3. Black stars show dependence ln(I_H/I_S) versus temperature 1/T in nanoparticles aggregated as nanopowder placed between silica plates in air, when luminescence was pumped at 450 nm into ⁴F_3/2 level of Er³⁺ (avoiding heating of sample due to the pump) while a straight line is a linear fit to these data. Red circles and a black square indicate the values ln(I_H/I_S), while placed on this linear fit line, when the luminescence was pumped at 975 nm (resulting in heating of sample) in nanopowders placed in evacuated pyrex ampoule (circles) and in bulk nano-glass-ceramics placed in air (square). An evaluated temperature rise is indicated for three selected pump powers of the laser diode.

Download Full Size | PDF

Here it is important to mention that due to a Gaussian character of intensity distribution in the cross-section of a diode laser beam the pump power in the central part of the beam was substantially higher than on its peripheral parts. Therefore the luminescence spectra shown in Fig. 1 are an average over varying pump power across the laser beam cross-section and the temperature rise in the central part of the excitation spot on the sample may be substantially higher than on the peripheral parts. When the pump power of the laser diode exceeded 250 mW (or 200 W/cm²) in data of Fig. 1, we have visually observed melting (formation of condensed drops) and even sublimation of nanoparticles in the central part of excitation spot indicative that the temperature rise in the central part reached value above 800°C, which corresponds to the melting temperature of the nanoparticles (which are PbF₂-based [6]). Further experiments when exciting a single nanoparticle or few nanoparticles are to provide more accurate evaluation of temperature rise in single nanoparticles. Noteworthy, the optical heating and up-conversion luminescence effects in the free-standing nanoparticles did not disappear after their ageing for half a year in air presumably due to chemical protection of the nanoparticles by their oxygen-containing shells [5,6].

3. Discussion

Energy level diagram in Fig. 2 indicates the routes for excitation of up-conversion luminescence with subsequent heating of Yb³⁺-Er³⁺ co-doped nanoparticles, when pumped by diode laser at 975 nm. The dashed lines show involved up-conversion luminescence energy transfer processes, which dominate in the present case of heavy rare-earth doping [6,9,15 and refs therein]. A high, up to 15%, quantum yield for light frequency up-conversion in these Yb³⁺-Er³⁺ co-doped nanoparticles [6,9] is a rather important factor since it ensures a strong up-conversion luminescence (which is a radiative part of this total quantum yield) and a strong heating of the sample due to phonon emission process indicated by wavy lines in Fig. 2 (which is a non-radiative part of this total quantum yield). This high quantum yield has been shown to be due to low phonon energy and short inter-dopant distances in these PbF₂-based nanoparticles [5,6,9]. While the Yb³⁺ co-dopant serves as efficient absorbent of laser diode energy, the Er³⁺ co-dopant is responsible for efficient heating of samples due to emission of phonons, as seen in diagram of Fig. 2 (processes indicated by wavy arrows).

A small size of nanoparticles hosting these dopants plays an important role in thermal release of absorbed energy into surrounding medium due to their large surface-to-volume aspect ratio. Indeed, should a good thermal contact be assured between the nanoparticle and surrounding volume this aspect ratio will facilitate higher heating of surroundings of nanoparticles compared to larger size particles. In addition, a quantum confinement of phonons and therefore enhanced electron-phonon interaction take place in these nanoparticles [16], resulting in extra heating rate of nanoparticles, which is a matter of future investigation.

4. Conclusion

In this work, an infrared laser diode induced heating of Yb³⁺-Er³⁺ co-doped nanoparticles up to 800°C has been detected by means of visible up-conversion luminescence and visual observation. The heating and up-conversion luminescence are essentially co-existing and simultaneous effects in these nanoparticles pointing out that both effects can be used simultaneously for heating of their surrounding nano-volumes and for detection of temperature rise and location of these nanovolumes. The reported nanoheater may be used in medicine for local hypothermal treatment of cells, for perforation of nanoholes in organics and metals.