Up-conversion emission and in vitro cytotoxicity characterization of blue emitting, biocompatible SrTiO₃ nanoparticles activated with Tm³⁺ and Yb³⁺ ions

Abstract

Functional SrTiO₃ nanoparticles activated with a broad concentration range of Tm³⁺ and Yb³⁺ ions were obtained utilizing the citric route. The effect of the sintering temperature and optically active co-dopant concentration on the structural and up-conversion emission properties was studied by using XRD (X-ray powder diffraction) and spectroscopic techniques (emission, power dependence, emission kinetics). The particle size was verified by the TEM technique being in the range of 20–90 nm depending on the annealing temperature. It was shown that cross-relaxations contribute to depopulation of both ¹G₄ and ³H₄ levels, whereas some of them populate the ³H₄ level. These processes are strongly dependent on the concentration of Tm³⁺ and are responsible for the specific interplay between blue and NIR emissions. Extraordinary short decays were recorded due to the relatively high concentrations of rare earths, structural features of the host matrix, extended surface area and contribution of the non-radiative processes. The cytotoxic activity of the nanoparticles to J774.E murine macrophages and U2OS human osteosarcoma cells was assessed showing that SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles are biocompatible and can be potentially used in further bio-related applications such as active implant layers or injectable medical cement ingredients.

---

1. Introduction

Nano-perovskites, described with the general formula ABO₃, are of intensified interest in the scientific community due to their extraordinary properties and wide application possibilities especially in the fields of catalysts,¹ membranes,² solar cells³ and etc. Most of these compounds crystallize in the Pm3̄m space group and can be divided, in terms of cation oxidation state, into two main subgroups – A³⁺B³⁺O₃ (e.g. LaAlO₃, GdFeO₃) and A²⁺B⁴⁺O₃ (e.g., BaTiO₃, MgSiO₃)⁴ as well. Strontium titanate (SrTiO₃) is one of the best known perovskites within the latter subgroup. Its physical and chemical stability,⁵ low phonon energy⁶ and high transparency in the VIS region makes this matrix very attractive for a phosphor application. Surprisingly, not so many articles can be found in the literature on SrTiO₃ rare earth doping. In fact, main focus was concentrated on the evaluation of the luminescence properties of the Eu^{3+ 7–9}, Pr^{3+ 8,10–12} and Er^{3+ 13–15} cations. Only a few papers were devoted to the up-conversion induced by mutual co-doping with Er³⁺/Yb³⁺.^13,16 Actually, no single scientific report on the up-conversion properties of the Tm³⁺/Yb³⁺ in the nanocrystalline SrTiO₃ can be found. Other, worth to mention, applications of the SrTiO₃ are in the broad field of electronics^17,18 and thermoelectric power generators.^19,20

Recently, it was shown that the SrTiO₃ is an attractive bioactive material being important ingredient of injectable acrylic bone cements for vertebroplasty²¹ as well as active-coating of Ti-based implants releasing Sr²⁺ ions.²² The latter ability stimulates the development of new nanomaterials for osteoporosis treatment²³ showing extended surface area assuring better contact of the particle surface with biological environment benefiting in enhanced cationic release. Moreover, addition of another functionality i.e. the up-conversion luminescence could be advantageous, especially in tracking of the ageing process in dental implants with non-invasive optical methods.²⁴ It was already shown that the up-converting nanoparticles are of great significance in biological applications due to that the near infrared (NIR) wavelength used for excitation is located within the range covering so-called biological optical window.^25,26 It means that the absorption of NIR by biological systems is very low and therefore deep tissue imaging is possible. Moreover, the anti-Stokes emission can be detected in the visible and the NIR spectral regions. In fact, co-doping with Tm³⁺/Yb³⁺ ions make possible a direct detection of the blue emission and NIR emission as well.^27,28 Usefulness of such functional materials for the cell and small animal imaging was already proven by many research groups.^29–31

The main goal of this research is focused on detailed characterization of structural properties of the SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles in connection with up-conversion luminescence as well as in vitro assessment of nanoparticles cytotoxicity to J774.E murine macrophages and U2OS human osteosarcoma cells being of great significance in estimation of the SrTiO₃:Tm³⁺/Yb³⁺ potential in bio-related applications.

2. Experimental

Analysis of the structural properties of SrTiO₃:Tm³⁺/Yb³⁺ nanopowders was based on measurement of XRD patterns by data collection in the 2θ range of 5–80° using X'Pert PRO X-ray diffractometer (Cu Kα₁: 1.54060 Å) (PANalytical). The average particle size was determined following the well known Scherrer's equation:where D is the mean grain size (nm); β₀ – apparatus broadening; β – full width at half maximum; θ – reflection angle; k constant (here set as 0.9), and λ is an X-ray source wavelength.³² The morphology and microstructure of final products were studied using transmission electron microscope (HR-TEM) Philips CM-20 Super Twin microscope operating at 200 kV. For the measurement powder was suspended in ethanol, de-agglomerated for 3 min in a ultrasound bath. Finally, droplet of nanoparticle dispersion was deposited on a copper grid with perforated carbon and gently dried. The average size of particles was determined using volume weighted formula:where d_av is the average particle size, n number of particles and d represents particle diameter. The diameter of particles was estimated using so-called Martin's diameter which can be defined as the distance between opposite sides of a particle measured crosswise of the particle and on a line bisecting the projected area. Elemental composition was checked using a scanning electron microscope FEI Nova NanoSEM 230 equipped with EDX spectrometer (EDAX PegasusXM4). Fifteen individual measurements were made from randomly chosen sample area in order to achieve reliable statistics. The up-conversion luminescence spectra were recorded using Jobin Yvon THR 1000 monochromator equipped with a Hamamatsu R928 photomultiplier and a 1200 grooves per mm holographic grating within the spectral range of the 450–900 nm. As an excitation source, continuous 975 nm line of 1.5 W diode laser (CNI laser, China) was used. The luminescence kinetics was measured using a LeCroy Wave Surfer oscilloscope with 975 nm line of 10 mJ Ti:sapphire pulse laser pumped by the second harmonic of the YAG:Nd³⁺ laser (532 nm line). The power dependence of the up-conversion emission intensity was measured using a miniature fiber spectrometer (Avantes, Netherlands, spectral resolution ∼ 3 nm) and 975 nm laser diode.

All recorded spectra were corrected according to the characteristic apparatus response.

Synthesis of SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles

The SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles were prepared using previously described citric route¹⁶ as a function of co-dopant concentration (Tm³⁺ 0.5–2 mol%, Yb³⁺ 5–20 mol%) as well as sintering temperature (600–1000 °C). The general preparation procedure will be given on the example of SrTiO₃ doped with 0.5 mol% of the Tm³⁺ and 5 mol% of the Yb³⁺, respectively. The main substrates were as follows Sr(NO₃)₂ (99.999% Alfa Aesar), Ti(OC₄H₉)₄ (99% Alfa Aesar), Yb₂O₃ (99.99% Alfa Aesar) and Tm₂O₃ (99.99% Alfa Aesar). The citric acid C₆H₈O₇ (99.95% Alfa Aesar) plays a mutual role of metal cations complexing agent and organic filler limiting diffusion of substrates, influencing morphology and particle size. The water suspension (10 ml in total) containing stoichiometric amounts of Yb₂O₃ 0.0493 g (0.125 mmol) and Tm₂O₃ 0.00484 g (0.00125 mmol) were prepared and subsequently digested in excess of the HNO₃ (ultra-anal Avantor Performance Materials). The reaction was carried out at 80 °C in a wide glass beaker until transparent solution containing lanthanide nitrates was obtained. Afterwards Yb³⁺ and Tm³⁺ nitrates were dissolved in water and re-crystallized three times and mixed with water solution containing 1 g (4.725 mmol) of the Sr(NO₃)₂. In a separate glass beaker 1.7 ml (4.98 mmol) of the Ti(OC₄H₉)₄ was mixed with 2 ml of 2,4-pentanadione (99% Alfa Aesar) in order to prevent titanium precursor from fast and uncontrolled hydrolysis. After that both solutions were mixed together and 19.21 g (100 mmol) of anhydrous citric acid was added. However, the ratio between metal cations and citric acid can be tuned further for getting optimal balance between particle size and morphology. Additional portions of water can be added for complete dissolution of all reactants. Prior the annealing steps at the temperature range of 600–1000 °C lasting for 3 h glass beaker was heated up at 90 °C until black-brown resin was obtained. The final product was in the form of very fine, white powder. The synthesis protocol was further on repeated in the same manner for rest of the SrTiO₃:Tm³⁺/Yb³⁺ samples.

In vitro assessment of SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles cytotoxicity to J774.E murine macrophages and U2OS human osteosarcoma cells

Cytotoxicity assessment was carried out on murine macrophage (J774.E) and human osteosarcoma cell lines (U2OS). The choice of the in vitro model was based on the fact that under in vivo conditions, macrophages form the primary line of response to particulate matter. Thus, they are responsible for the distribution and clearance of nanoparticles and their agglomerates. On the other hand, U2OS is a cancer cell line derived from the bone tissue which is rich in hydroxyapatites that play a pivotal role in the extracellular matrix formation. This cell line belongs to most widely used models for studies on bone tissue in vitro.^33,34 The cells were cultured in RPMI-1640 medium (Institute of Immunology and Experimental Therapy, Wrocław, Poland) supplemented with 10% fetal bovine serum (FBS, Sigma, USA), L-glutamine (Sigma, UK) and antibiotics (penicillin and streptomycin, Sigma, Germany). For the cytotoxicity assessment, cells were seeded in 96-well-plates (NUNC, Denmark) at a density of 10 × 10³ (J774.E) or 3 × 10³ (U2OS) cells per well and pre-incubated at 37 °C for 4 h in a humidified atmosphere of 5% CO₂. After that, nanoparticle dispersions were added. Stock dispersions of SrTiO₃ nanoparticles were prepared based on a simplified version of the NANOGENOTOX dispersion protocol.³⁵ Nanoparticles were suspended in 0.05% BSA water solution and bath-sonicated at room temperature for 1 min. Next, the stock solutions were further diluted in 0.05% BSA and fresh dispersions in complete culture medium were prepared before each experiment. Cells were exposed to the nanoparticle dispersions for 72 h (5% CO₂, 37 °C). After that, the MTT assay was carried out. The test is based on the enzymatic reduction of the tetrazolium salt MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazoliumbromide] in living, metabolically active cells. Preliminary experiment showed no interference of nanoparticles with either MTT or formazan in a cell-free system. After 2 h of incubation, cells were lysed and left for complete dissolution of purple-colored metabolite. After 24 h, the optical density (OD) was measured by means of a microplate reader (ELx800, BioTek, USA) at the wavelength of 570 nm (reference 630 nm). The OD of control cells was taken as 100%. The results were obtained from at least 3 independent experiments.

3. Results and discussion

Structure analysis

Basing on the crystallographic data the SrTiO₃ has a cubic structure ascribed to the Pm3̄m space group (no. 221) at room temperature.

The projection of the unit cell together with indication of coordination polyhedra is shown in Fig. 1. The Ti⁴⁺ (0.605 Å) ion is surrounded by six O²⁻ anions forming octahedra (TiO₆⁸⁻) whereas Sr²⁺ (1.44 Å) is in a twelve fold coordination (SrO₁₂²²⁻). The SrTiO₃ undergoes three phase transitions induced by lowering temperature leading to structural distortions from ideal cubic symmetry. Thus, at temperature range of 110–65 K SrTiO₃ has tetragonal symmetry (I4/mcm), between 55 and 35 orthorhombic and below 10 K dominates rhombohedral distortion.³⁶ Another factor contributing to the symmetry distortion is directly related with the presence of artificially introduced cations. It is well known that substitution of Sr²⁺ or Ti⁴⁺ with other cations with different oxidation states induces the charge compensation effects in a number of ways. The problem of the exact location of lanthanides cations in the perovskite structure is not yet clearly resolved. In our previous study,¹⁶ it was shown that ytterbium cations were located in the titanium sites predominantly, as evidenced by significant changes of the IR bands corresponding to the Ti–O stretching and bending vibrations. Unfortunately, the concentration of Er³⁺ ions was too low to deliver the clear evidence of their location. The FT-IR and Raman spectra did not show any indication on the presence of Er³⁺ or Yb³⁺ at Sr²⁺ ions. Since the ionic radii of Tm³⁺ at six fold coordination is comparable with Yb³⁺ and Er³⁺ (Er³⁺ – 0.89 Å, Yb³⁺ – 0.868 Å, Tm³⁺ – 0.88 Å)³⁷ it was assumed, that these rare earth cations will predominantly occupy Ti⁴⁺ site inducing formation of oxygen vacancies.³⁸

Fig. 1 Extended unit cell projection of SrTiO₃ perovskite with indication of coordination polyhedra of the Ti⁴⁺ and Sr²⁺.

The structural properties of the SrTiO₃:Tm³⁺/Yb³⁺ as a function of temperature (see Fig. 2) and co-dopants concentration effect (Fig. 3) were followed by means of the XRD technique. Since the formation of the SrTiO₃ was studied by us in detail previously¹⁶ 600 °C was chosen as the starting sintering temperature (below only multiphase products or amorphous phase can be detected). One can see that all the Bragg's reflections match very well with the reference standard of the cubic SrTiO₃ (ICSD 35-0734) up to 900 °C. This confirms phase purity of the final product with accuracy of the measuring technique. Increase of temperature above 900 °C results in appearance of extra peak at 31° 2θ. This reflection most probably corresponds to the SrYb₂O₄ compound confirming phase separation. The average crystallite size was calculated using Scherrer's equation by taking into consideration separated non-split peaks located at 32.6°, 40.1° and 46.6° 2θ corresponding with following crystallographic planes (110), (111) and (200), respectively. Mean crystallize size was estimated to be in the range from 19 to 39 nm pointing out on progressive grain growth with sintering temperature. Influence of the Tm³⁺ and Yb³⁺ concentration is presented in Fig. 3 showing that for the concentration range of Tm³⁺ from 0.5 to 1 mol% and fixed quantity of Yb³⁺ (5 mol%) in obtained perovskites were phase pure up to 900 °C. On the other hand, an increase of the Yb³⁺ concentration above 15 mol% caused phase separation and formation of the secondary phases. Investigation of the sintering temperature effect leads to the conclusions that the above 900 °C and 10 mol% of Yb³⁺ multiphase product will be obtained. These observations are consistent with our previous report on the SrTiO₃:Er³⁺/Yb³⁺ nanoparticles prepared using the same synthetic protocol.¹⁶ Therefore, phase separation was induced not only by high concentration of dopants (mainly Yb³⁺), due to the solubility limitation, but also by annealing temperature both affecting thermodynamic equilibrium of the given system.

Fig. 2 Evolution of crystal structure of SrTiO₃:0.5% Tm³⁺/5% Yb³⁺ sample induced by annealing temperature.

Fig. 3 Effect of the Tm³⁺ (left) and Yb³⁺ doping (right) on structure of the SrTiO₃ nanoparticles annealed at 600 °C.

Morphology, structure and primary particle size of the SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles were investigated by TEM and SAED techniques (Fig. 4). As it can be seen the mean particle size depends strongly on sintering temperature. In the case of the sample heat treated at 600 °C the grain size was found to be 20 nm whereas increase of temperature up to 1000 °C leads to the grain growth and particle size was estimated to be around 80–90 nm. What is interesting at 600 °C the fine particle with irregular shapes form 100–150 nm agglomerates. Eventually, they transform into elongated structures at 1000 °C forming non-agglomerated particles. This behavior points out on the typical Ostwald ripening grain growth mechanism where large particles are growing at the expense of small ones.

Fig. 4 TEM and SAED characterization of the SrTiO₃:0.5% Tm³⁺/5% Yb³⁺ nanoparticles sintered at 600 °C (a and b) and 1000 °C (c and d).

Presence of agglomerates of fine and small particles is a typical feature of nanomaterials with non-blocked surface. This is a natural tendency of all nanoparticles which allows them to achieve the minimum of surface energy. The discrepancy between the particle size calculated using Scherrer's method and TEM analysis, especially for high temperatures, relies on limitation of the latter. This approach offers only the calculation of the crystallite size and for polycrystalline materials the comparison between Scherrer's calculation and TEM analysis might lead to different values (see Table s1 in ESI†). Moreover, it does not take into account the particle strain, crystal lattice imperfections as well as the k constant depends on the particle shape (value of 0.9 approximates that all particles are spherical). Finally, omits completely crystallite distribution (peak broadening is significantly affected by smallest particles) as well as the upper limit of the grain size for Scherrer's method usability is 100 nm.³⁹ The comparison of the SAED images, presenting well developed spotty rings, with the reference standard of the SrTiO₃ confirms preparation of phase pure compounds. The elemental composition of the final products was confirmed by the SEM-EDS technique showing that content of lanthanides cations are close to the planned values with small deviations (see Fig. s1 and s2†).

Cytotoxic activity of SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles to J774.E murine macrophages and U2OS human osteosarcoma cells

The effects of the SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles of different sizes on cell viability for J774.E macrophages as well as U2OS osteosarcoma cells were presented in Fig. 5 and 6. One can note a tremendous difference in cytotoxic response of both cell lines. The J774.E cells seem not to be affected by the presence of nanoparticle suspension whereas U2OS cells show a distinct dose-dependent decrease in viability. It suggests that some cell types may be affected negatively to a higher extent than others. This result is very interesting because the J774.E macrophages are highly efficient phagocytes and tend to overload with particles when exposed to dispersions. This effect usually leads to the higher cytotoxicity as compared with a non-phagocytic cell line.⁴⁰ In the present study an opposite result was seen. In fact the SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles of different particle size did not produce much different responses. Thus, it is concluded that possible SrTiO₃ nanoparticle cytotoxicity is probably size-independent. It may be, however, related to ion release to the medium and, perhaps, associated with different vulnerability of both cell lines to the increase in free strontium ion concentration.

Fig. 5 Influence of SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles prepared at different temperatures on cell viability of J774.E murine macrophages as determined by the MTT assay after 72 h exposure.

Fig. 6 Influence of SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles prepared at different temperatures on cell viability of U2OS human osteosarcoma cells as determined by the MTT assay after 72 h exposure.

There are only few studies investigating SrTiO₃:Tm³⁺/Yb³⁺ cytotoxicity. It was clearly shown by Pazik¹⁶ that SrTiO₃:Er³⁺/Yb³⁺ nanoparticles do not cause red blood cell damage up to a concentration of 100 μg ml⁻¹. Tateyama et al.⁴¹ found that rat primary bone marrow cells show a reduced growth rate when seeded on SrTiO₃ plates (compared to biocompatible reference material). Furthermore, Wang et al.⁴² investigated the biocompatibility of SrTiO₃–TiO₂ nanoparticle–nanotube heterostructures and found that the potential of Sr²⁺ cation release is a major determinant of cellular response. One may conclude that SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles are biocompatible towards J774.E cells whereas in U2OS cells they cause selective cytotoxicity. This effect is particle size-independent and may be related to ion release and specific cell line sensitivity.

Optical properties

Up-conversion luminescence spectra of the representative SrTiO₃ nanocrystallites co-doped with 0.5% of Tm³⁺ and 5% of Yb³⁺ recorded at 300 K using as the excitation source 975 nm laser diode and measured as a function of annealing temperature are shown in Fig. 7. The spectra consist of characteristic of the Tm³⁺ anti-Stokes emission bands ascribed to the following electron transitions ¹G₄ → ³H₆ centered at 476 nm (21 008 cm⁻¹), so-called blue band, the ¹G₄ → ³F₄ transition (red band) at 650 nm (15 386 cm⁻¹) together with the ³F_2,3 → ³H₆ at 694 nm (14 410 cm⁻¹) as well as the near-infra red band (NIR) constituted of two electron transitions ³H₄ → ³H₆ and ¹G₄ → ³H₅ with maximum at 794 nm (12 594 cm⁻¹), respectively. All of the spectra were normalized to the ¹G₄ → ³F₄ transition of low intensity being practically independent on the concentration of co-dopants and sintering temperature (particle size).

Fig. 7 Anti-Stokes emission spectra of the SrTiO₃:0.5% Tm³⁺/5% Yb³⁺ as a function of sintering temperature.

All recorded bands are broad and do not split into separate Stark components. The reason of such behavior has its source in the distortion of the occupied site by rare earth cations (cationic size mismatch) as well charge compensation effect relying on creation of oxygen vacancies deepening the structural distortions. Another factor, which has to be taken into account, is the small particle size leading to the heterogeneous distribution of rare earths. This leads to the presence of Tm³⁺ and Yb³⁺ enriched surface regions. Therefore, the cations (Tm³⁺ and Yb³⁺) located in the near surface area might have lowered symmetry of their closest surrounding. These effects can contribute to the appearance of distorted or new sites of lower point symmetry causing spectral overlap of the Stark components of Tm³⁺ electron transitions. Such phenomena was confirmed by the analysis of the luminescence behavior of more sensitive to structural changes Eu³⁺ cations in Eu³⁺ doped nanocrystalline SrTiO₃.^43,44

As it can be seen, for the samples containing 0.5% of Tm³⁺ and 5% of Yb³⁺, the intensity of the blue emission is almost constant up to 800 °C and dominates over NIR band for the sample heat treated at 600 °C. Moreover, the intensity of the NIR band increases progressively with temperature and then drops down a little for high thermal treatment (above 900 °C). It is obvious, that in the case of the nanoparticles, sintering conditions are closely related to the particle size. Therefore, in order to explain the temperature effect on the behavior of the NIR band it is necessary to connect this phenomena with the particle size effect. It is well known that for the nanoparticles surface-to-volume ratio plays extremely important role in shaping the physicochemical properties of given material. It means, that the number of the surface atoms is higher for nanomaterials. Since the rare earth cations are exposed to the interaction with external environment like impurities (OH⁻ groups) the emission from surface cations is prone to be effectively quenched. Therefore, one may conclude that since nominal content of the Tm³⁺ in the SrTiO₃ is 0.5 mol%, thus the effective concentration of emitting centers will be lower in case of the smallest particles (at 600 °C). Further increase of thermal treatment temperature leads to the particle growth and decrease of Tm³⁺ fraction located at surface area resulting in increase of effective concentration of emitting ions. Summarizing, the number of quenched Tm³⁺ ions will be reduced as the sintering temperature rises. However, increase of the Tm³⁺ effective concentration can be seen as a double edge sword since Tm³⁺ ions are sensitive to cross-relaxation (described later in the text). This non-radiative process contributes to population of the ³H₄ level and increases the NIR band emission up to 800 °C. Therefore, it is assumed that above 900 °C structure imperfections are, to some extent, minimized in contrast to smaller nanoparticles. The same feature was discussed earlier upon investigation of Nd³⁺ doped nanopowders of the YVO₄.⁴⁵

The effect of Tm³⁺ and Yb³⁺ ions concentration was shown in Fig. 8. As it was discussed earlier increase of Tm³⁺ ions concentration at fixed Yb³⁺ content (5 mol%) leads to decrease of the ¹G₄ → ³H₆ emission and increase of the NIR band due to the higher probability of the cross-relaxation. It seems that for prepared series of the SrTiO₃ samples optimal balance between blue and NIR emission was found for Tm³⁺ content of 0.5 mol%. In the case of the Yb³⁺ doping, concentration range of 5–20 mol% of sensitizer was tested showing constant improvement of the up-conversion luminescence. It can be concluded that increase of concentration of the Yb³⁺ content is beneficial, since more IR excitation can be harvest leading to enhancement of efficiency of the energy transfer.⁴⁶ However, it is important to note that exceeding concentration of 10 mol% of Yb³⁺ leads to phase separation of the SrTiO₃. If one considers only phase pure material this concentration is a critical one.

Fig. 8 Anti-Stokes emission spectra of the SrTiO₃:x% Tm³⁺/5% Yb³⁺ (left) and SrTiO₃:0.5% Tm³⁺/x% Yb³⁺ (right) prepared at 600 °C.

For better clarification of up-conversion luminescence properties of the SrTiO₃:Tm³⁺/Yb³⁺ dependence on particle size (sintering temperature), dopants concentration effect (Fig. 9), and to evaluate the efficacy of up-conversion process the near infrared to blue ratio was calculated using following equation:

where numerator defines integral intensity of the NIR band (794 nm) and denominator is an integral intensity of blue band (476 nm). It has to be underlined that the NIRB ratio is strongly dependent on the laser pump power influencing population of states, concentration of emitter and sensitizer, particle size which will modulate (together with co-dopant concentration) the probability of the radiative and non-radiative processes, synthetic parameters (for instance temperature), and properties of the host lattice (phonon energy, site symmetry). It is worth noting, that NIRB behavior vs. Tm³⁺ concentration shows clearly domination of the NIR band. Intensity of the NIR band is additionally enhanced (except two photon ETU) by the cross-relaxation processes populating the ³H₄ level. The samples heated at 600 °C showed the strongest blue emission for 0.5 mol% of Tm³⁺. In the case of Yb³⁺ concentration effect, the NIRB ratio directly points out on enhancement of energy transfer between sensitizer and emitter up to 800 °C and Yb³⁺ concentration of 20 mol%. Above 900 °C the NIR band intensity drops down due to the phase separation above 15% of the Yb³⁺. One can see that above 900 °C, for the largest particles and highest effective concentration of emitting Tm³⁺, decrease in NIR band intensity was observed due to higher contribution of the non-radiative processes (cross relaxations) as well as possible back-transfer. However, strong enhancement of the NIR band upon increase of Yb³⁺ content, almost in all cases, was found. The domination of the NIR emission is not a big surprise since the up-conversion process leading to the blue emission involves participation of three photons and for NIR luminescence only two are needed. Thus, the latter is more probable. Additionally, non radiative processes like cross relaxation and multiphonon relaxation can contribute to overall increase of the NIR emission as well. From a practical point of view the sample doped with 0.5 mol% of Tm³⁺ and varied concentration of Yb³⁺ are most promising for the bio-related application since the blue and NIR emissions are the most intense.

Fig. 9 Influence of doping levels on the NIR-to-blue ratio of the SrTiO₃:x% Tm³⁺/x% Yb³⁺. The lines are only to guide the eyes.

The power dependence of emission from the ¹G₄ → ³H₆ (blue band) and ³H₄ → ³H₆ (NIR band) was investigated as a function of the sintering temperature and Yb³⁺ content of SrTiO₃ sample doped with 0.5 mol% of Tm³⁺ ions (see Fig. 10 as well as Fig. s3† for respective spectra). The data was fitted according to the following formula:

I_UPC = I_in^N, (4)

allowing for estimation of the order of the up-conversion process N pointing out on the number of photons required for achieving the anti-Stokes emission.

Fig. 10 Power dependence of the blue and red emission of the SrTiO₃:0.5% Tm³⁺/5% Yb³⁺ as a function of annealing temperature (up) and Yb³⁺ concentration (bottom). Log–log plot.

Experimental data was fitted with linear function leading to the value of over or close to 2 in the case of blue band in the SrTiO₃:0.5% Tm³⁺/5% Yb³⁺ nanoparticles prepared as a function of annealing temperature to achieve blue emission at least two photons are required. In the case of NIR band values are close to 1.5 what suggests that more than one photon is needed. As it can be seen, increase of the Yb³⁺ concentration above 5 mol% reduces slope of power dependence of blue emission. The NIR Yb³⁺ concentration dependence does not change in the same way and slope is comparable with that of annealing temperature effect. The lower N values, than expected (3 for the blue and 2 for the red), could indicate on a higher probability of up-conversion that depletes intermediate states of Yb³⁺ and Tm³⁺ and thus saturates the power dependence more.⁴⁷ It is also interesting to note that all multiphonon emissions shown tendency to achieve plateau at high pumping powers (above 1 W).

The simplified energy level diagram presenting the most important accompanying and contributing processes to the overall up-conversion emission in the SrTiO₃ nanoparticles doped with Tm³⁺ and Yb³⁺ ions was shown in Fig. 11. Auzel, in his review,⁴⁸ gave a detailed description of the APTE effect (addition de photon par transfer d'energie) known more commonly in literature as the energy transfer up-conversion process. The excitation paths in the system containing Tm³⁺ and Yb³⁺ cations are more complex than in the case of Er³⁺ and Yb³⁺ co-doping. It is very clear that the Tm³⁺ ion does not absorb at 975 nm which assures a selective excitation of the Yb³⁺ under this particular wavelength. Moreover large absorption cross section of Yb³⁺ ions allows very effective absorption of the excitation energy. All starts from the absorption of the NIR photons which moves the electrons from the ground state ²F_7/2 to the excited ²F_5/2 levels of Yb³⁺ by the ground state absorption process (GSA). They can immediately drop back to the ground state, but this step is followed by several non-resonant energy transfers to the neighboring Tm³⁺ ion. The first one excites the Tm³⁺ ground level ³H₆ to the excited state ³H₅ which is depopulated to the ³F₄ level by multiphonon relaxation process (MPR I). Next energy transfer excites the ³F₄ level to the ³F_2,3 levels from which second MPR process occurs (MPR II) feeding the ³H₄ excited level. The third step results in feeding of the ¹G₄ level of Tm³⁺. As it already can be seen the population of the ¹G₄ level is three photon process. There is also possibility of participation of fourth energy transfer which populates the ¹D₂ level from the ¹G₄. As a result of three photon process two emission bands are observed. First one corresponding with depopulation of the ¹G₄ namely ¹G₄ → ³H₆ (476 nm), and second the ¹G₄ → ³F₄ (650 nm). In fact, last one (overlapping) can be also ascribed to the emission from the ³F_2,3 → ³H₆ (light red dotted line – two photon process 650 nm). The emission in the NIR region is mainly ascribed to the ³H₄ → ³H₆ (749 nm) electron transition being a result of two photon process and MPR II. In addition, the ¹G₄ → ³H₆ transition (dark red dotted line – three photon process) can overlap with previous one as well. Interestingly, at relatively high Tm³⁺ concentration cross relaxation processes (CR) might occur leading to depopulation and population of certain energy levels in the following manner:

(¹G₄ + ³H₄) → (³F₄ + ¹D₂), (CR I)

(³H₄ + ¹G₄) → (³F₄ + ¹D₂), (CR II)

(¹G₄ + ³H₆) → (³H₅ + ³H₄), (CR III)

(³H₆ + ¹G₄) → (³H₅ + ³H₄), (CR IV).⁴⁹

Fig. 11 Simplified energy level scheme presenting up-conversion and possible accompanying processes in the SrTiO₃:Tm³⁺/Yb³⁺ system (description in the text).

Cross-relaxations (I) and (II) contribute to depopulation of both ¹G₄ and ³H₄ levels, whereas (III) and (IV) will populate the ³H₄ level. All of them are strongly dependent upon effective concentration of Tm³⁺ and are responsible for specific interplay between blue and NIR emissions.

Another possibility of getting the emission of Tm³⁺ with participation of Yb³⁺ is co-operative sensitization process (CS – right part of Fig. 11) which can take place in the host matrixes with high Yb³⁺ concentration.^50,48 This process requires successive energy transfer from Yb³⁺ to Tm³⁺ by which the first NIR photon converts system into an intermediate metastable state. The system is further on excited from the metastable state to a higher energy level due to absorption of the second photon. Thus after GSA absorption of the two Yb³⁺ ions to the excited ²F_5/2 levels energy is transferred to the ¹G₄ level in the process described as (²F_5/2 + ²F_5/2 + ³H₆ → ²F_7/2 + ²F_7/2 + ¹G₄).⁵¹ However, the CS process is less efficient (η – 10⁻⁶) in contrast to the up-conversion (η – 10⁻³) but its probability increases with concentration of sensitizer.⁴⁸

The luminescence kinetics of the SrTiO₃:0.5% Tm³⁺/5% Yb³⁺ as a function of the sintering temperature recorded for the blue (¹G₄ → ³H₆) as well as NIR (³H₄ → ³H₆) bands are presented in Fig. 12, whereas calculated values of decay times are shown in Fig. 13 (decay curves of Tm³⁺ and Yb³⁺ concentration dependence can be found in ESI Fig. s4†). In fact, the decay curves were clearly non-exponential for all samples, thus values of lifetimes were calculated as the effective emission decay time using given formula:

where I(t) represents the luminescence intensity at time t corrected for the background and the integrals are evaluated on a range 0 < t < t^max where t^max ≫ τ_m.⁵²

Fig. 12 Decay curves of blue (left) and red (right) emission of the SrTiO₃ doped with 0.5 mol% Tm³⁺ and 5 mol% Yb³⁺ ions as a function of annealing temperature.

Fig. 13 Luminescence lifetimes of the SrTiO₃ particles doped with Tm³⁺ (left) and Yb³⁺ (middle) as well as temperature effect on decays (right). Connecting lines are used only as a guide for the eye.

Obviously, the values of decay times of either the blue emission and NIR emission are dependent on sintering temperature as well as concentration of the Tm³⁺ and Yb³⁺ ions. In bulk type materials the Tm³⁺ decay is of the order of hundreds of μs.⁴⁹ However, in this study decay time values are extraordinary short (tens of μs). This can be directly related to the structural properties and nanosize character of compounds. The size effect is clearly seen since rather high fraction of optically active ions are placed near particle surface leading to the emission quenching due to the surface defects or impurities. The non-exponential character of decay curves reflects the non-homogenous distribution of dopants as well as presence of the Tm³⁺ ions located at multi-sites (crystal sites together with surface sites with lowered symmetry). Therefore, existence of the rare earth enriched surface regions significantly increases probability of the cross-relaxation processes leading to shortening of the decay times. Indeed, increase of the Tm³⁺ concentration results in higher probability of cross relaxation process and further shortening of decay times (see Fig. s4†). In the case of the Yb³⁺ concentration dependence, progressive decrease of lifetimes was observed which could be attributed to the back-energy transfer process between Yb³⁺ and Tm³⁺ (see ESI Fig. s5† for spectra). The back transfer effectively drains the population of the Tm³⁺ emitting states. Increase of temperature leads to decrease of surface atoms and improves distribution of emitters within particle. In the case of the 1000 °C sample the decay curve shows clearly two exponential components which could be assigned to presence of the second phase (SrYb₂O₄), as indicated by XRD analysis.

4. Conclusions

It was shown that the citric route can be successfully used for preparation of blue and NIR emitting SrTiO₃ nanoparticles co-doped with Tm³⁺ and Yb³⁺. The particle size depends strongly on sintering temperature being around 20 nm for the SrTiO₃ heat treated at 600 °C and 90 nm at 1000 °C. Phase separation was noticed for the Yb³⁺ concentration exceeding 10 mol% and annealing temperature above 900 °C. Structural modification of the SrTiO₃ host with Tm³⁺ and Yb³⁺ leads to the structure distortions by formation of oxygen vacancies due to the charge compensation effect.

It was shown that the SrTiO₃:Tm³⁺/Yb³⁺ nanoparticles are biocompatible towards J774.E cells whereas in U2OS cells they cause selective cytotoxicity. This particular result is very interesting due to that the J774.E macrophages are highly efficient phagocytes and tend to overload with particles when exposed to dispersions. In the case of U2OS, cytotoxicity was not particle size-dependent and could be related to ion release and specific cell line sensitivity.

Significant band broadening of up-conversion emission was found due to the structural distortions induced by optically active cations, heterogeneous distribution of the dopants, as well as existence of enriched surface area contributing to spectral overlap of the Stark components and quenching. The temperature change of the blue and NIR band intensity was explained basing on concept of effective concentration of emitters and enhanced role of surface-to-volume ratio in nanoparticles. Increase of sintering temperature leads to decrease of Tm³⁺ concentration at nanoparticles surface. This directly results in increase of effective concentration of emitting ions. Due to the specific energy level structure of Tm³⁺ increase of its concentration has a detrimental effect on blue luminescence due to higher probability of cross relaxation processes. It was shown that cross-relaxations contribute to depopulation of both ¹G₄ and ³H₄ levels. Extraordinary short decays (tens of μs) were recorded due to relatively high concentration of rare earths, structural features of host matrix as well as extended surface area and contribution of the non-radiative processes.

Acknowledgements

The authors would like to thank M.Sc. Ewa Bukowska for performing XRD measurements as well as Ph.D. J. Kowalczyk for technical assistance during particle synthesis. Financial support of the National Science Centre in course of realization of the Project ‘Smart nanoparticles for bio-imaging and drug delivery’ no. UMO-2011/01/D/ST5/05827 is gratefully acknowledged.

References

1. N. K. Labhsetwar, A. Watanabe, R. B. Biniwale, R. Kumar and T. Mitsuhashi, Appl. Catal., B, 2001, 33, 165–173.
2. A. L. Ahmad, N. F. Idrus and M. R. Othman, J. Memb. Sci., 2005, 262, 129–137.
3. B.-W. Park, B. Philippe, X. Zhang, H. Rensmo, G. Boschloo and E. M. J. Johansson, Adv. Mater., 2015, 9, 6806–6813.
4. G. Nag Bhargavi and A. Khare, Opt. Spectrosc., 2015, 118, 902–917.
5. E. Heifets, S. Piskunov, E. A. Kotomin, Y. F. Zhukovskii and D. E. Ellis, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 75, 115417.
6. A. I. Lebedev, Phys. Solid State, 2009, 51, 362–372.
7. C. R. García, J. Oliva, M. T. Romero and L. A. García Trujillo, Adv. Mater. Sci. Eng., 2015, 2015, 1–7.
8. B. Marí, K. C. Singh, P. Cembrero-Coca, I. Singh, D. Singh and S. Chand, Displays, 2013, 34, 346–351.
9. J. A. Dawson, X. Li, C. L. Freeman, J. H. Harding and D. C. Sinclair, J. Mater. Chem. C, 2013, 1, 1574–1582.
10. E. A. Bondarenko, A. A. Skomorokhov, N. I. Kargin, A. S. Gusev and S. M. Ryndya, Opt. Spectrosc., 2015, 118, 729–734.
11. O. M. Marchylo, Y. Nakanishi, H. Kominami, K. Hara, L. V. Zavyalova, S. V. Svechnikov and B. A. Snopok, Theor. Exp. Chem., 2013, 49, 158–164.
12. W. Tang, Y. Sun, M. Yu, X. Liu, Y. Yin, B. Yang, L. Zheng, F. Qin, Z. Zhang and W. Cao, RSC Adv., 2015, 5, 27491–27495.
13. H. Guo, N. Dong, M. Yin, W. Zhang, L. Lou and S. Xia, J. Alloys Compd., 2006, 415, 280–283.
14. Z. Potuček, A. P. Skvortsov, N. K. Poletaev, Z. Bryknar, L. Jastrabik, A. Dejneka and V. A. Trepakov, J. Lumin., 2014, 154, 437–444.
15. J. Xu, J. Wang, Y. L. Wei, Z. X. Sun, X. Q. Zheng, X. L. Tao, L. Q. Fan, M. L. Huang and Y. F. Huang, J. Synth. Cryst., 2014, 43, 1954–1959.
16. R. Pazik, M. Maczka, M. Malecka, L. Marciniak, A. Ekner-Grzyb, L. Mrowczynska and R. J. Wiglusz, Dalton Trans., 2015, 44, 10267–10280.
17. C. Li, Z. Liu, W. Lü, X. R. Wang, A. Annadi, Z. Huang, S. Zeng, Ariando and T. Venkatesan, Sci. Rep., 2015, 5, 13314.
18. A. Annadi, S. Lu, G. Cheng, M. Tomczyk, M. Huang, H. Lee, S. Ryu, C.-B. Eom, P. Irvin and J. Levy, APS March Meeting. 2015, 2015.
19. Y. Lin, C. Norman, D. Srivastava, F. Azough, L. Wang, M. Robbins, K. Simpson, R. Freer and I. A. Kinloch, ACS Appl. Mater. Interfaces, 2015, 7, 15898–15908.
20. A. I. Abutaha, S. R. S. Kumar, K. Li, A. M. Dehkordi, T. M. Tritt and H. N. Alshareef, Chem. Mater., 2015, 27, 2165–2171.
21. R. G. Carrodeguas, B. V. Lasa and J. S. R. Del Barrio, J. Biomed. Mater. Res., Part B, 2004, 68, 94–104.
22. Y. Xin, J. Jiang, K. Huo, T. Hu and P. K. Chu, ACS Nano, 2009, 3, 3228–3234.
23. A. S. Hurtel-Lemaire, R. Mentaverri, A. Caudrillier, F. Cournarie, A. Wattel, S. Kamel, E. F. Terwilliger, E. M. Brown and M. Brazier, J. Biol. Chem., 2009, 284, 575–584.
24. R. J. Wiglusz, B. Pozniak, K. Zawisza and R. Pazik, RSC Adv., 2015, 5, 27610–27622.
25. Y. Pu, L. Shi, S. Pratavieira and R. R. Alfano, J. Appl. Phys., 2013, 114, 153102.
26. G. Tian, Z. Gu, L. Zhou, W. Yin, X. Liu, L. Yan, S. Jin, W. Ren, G. Xing, S. Li and Y. Zhao, Adv. Mater., 2012, 24, 1226–1231.
27. J. C. Boyer, L. A. Cuccia and J. A. Capobianco, Nano Lett., 2007, 7, 847–852.
28. V. Mahalingam, F. Vetrone, R. Naccache, A. Speghini and J. a. Capobianco, Adv. Mater., 2009, 21, 4025–4028.
29. D. Yang, P. Ma, Z. Hou, Z. Cheng, C. Li and J. Lin, Chem. Soc. Rev., 2015, 44, 1416–1448.
30. D. Chatterjee, A. Rufaihah and Y. Zhang, Biomaterials, 2008, 29, 937–943.
31. M. Nyk, R. Kumar, T. Y. Ohulchanskyy, E. J. Bergey and P. N. Prasad, Nano Lett., 2008, 8, 3834–3838.
32. H. P. Klug and L. E. Alexander, X-Ray Diffraction Procedures: For Polycrystalline and Amorphous Materials, Wiley, New York, 2nd edn, 1974.
33. V. Kartsogiannis and K. W. Ng, Mol. Cell. Endocrinol., 2004, 228, 79–102.
34. E. M. Czekanska, M. J. Stoddart, R. G. Richards and J. S. Hayes, Eur. Cells Mater., 2012, 24, 1–17.
35. http://www.nanogenotox.eu/, 2011.
36. F. W. Lytle, J. Appl. Phys., 1964, 35, 2212–2214.
37. R. D. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1976, 32, 751–767.
38. M. J. Akhtar, Z. U. N. Akhtar, R. A. Jackson and C. R. A. Catlow, J. Am. Ceram. Soc., 1995, 78, 421–428.
39. A. K. Singh, Advanced X-ray Techniques in Research and Industry, IOS Press, Amsterdam, 2005.
40. R. Pązik, A. Zięcina, E. Zachanowicz, M. Małecka, B. Poźniak, J. Miller, Z. Śniadecki, N. Pierunek, B. Idzikowski, L. Mrówczyńska, A. Ekner-Grzyb and R. J. Wiglusz, Eur. J. Inorg. Chem., 2015, 4750–4760.
41. Y. Tateyama, Y. Morita, K. Tanaka, T. Katayama and E. Nakamachi, Proceedings of the World Congress 2009 for Medical Physics and Biomedical Engineering, in IFMBE Proceedings, ed. O. Dossel and W. Schlegel, Springer, Berlin Heidelberg, 2010, pp. 279–281.
42. Y. Wang, D. Zhang, C. Wen and Y. Li, ACS Appl. Mater. Interfaces, 2015, 7, 16018–16026.
43. F. Fujishiro, T. Arakawa and T. Hashimoto, Mater. Lett., 2011, 65, 1819–1821.
44. I. K. Battisha, A. Speghini, S. Polizzi, F. Agnoli and M. Bettinelli, Mater. Lett., 2002, 57, 183–187.
45. R. J. Wiglusz, L. Marciniak, R. Pazik and W. Strek, Cryst. Growth Des., 2014, 14, 5512–5520.
46. J. Méndez-Ramos, P. Acosta-Mora, J. C. Ruiz-Morales and N. M. Khaidukov, J. Alloys Compd., 2013, 575, 263–267.
47. M. Pollnau, D. Gamelin, S. Lüthi, H. Güdel and M. Hehlen, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 3337–3346.
48. F. Auzel, Chem. Rev., 2004, 104, 139–173.
49. I. Etchart, I. Hernandez, A. Huignard, M. Berard, M. Laroche, W. P. Gillin, R. J. Curry and A. K. Cheetham, J. Appl. Phys., 2011, 109, 063104.
50. V. V. Ovsyankin and P. P. Feofilov, JETP Lett., 1966, 4, 317–318.
51. F. Guell, R. Sole, J. Gavalda, M. Aguilo, M. Galan, F. Diaz and J. Massons, Opt. Mater., 2007, 30, 222–226.
52. R. Pazik, D. Hreniak, W. Strek, a. Speghini and M. Bettinelli, Opt. Mater., 2006, 28, 1284–1288.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra03075k

---