Wide-range luminescent thermometers? – Quite a challenge!

Luminescent thermometers become increasingly important for remote, non-invasive temperature sensing at especially small lateral dimensions, at which other, contact-based detection principles only work in a limited way or even not at all. Typical application sectors are temperature recording on catalytic surfaces or in biomedical processes.

        Especially the lanthanoid ions have emerged as leading luminescent ions in this area due to their rich electronic structure with selected radiatively emitting excited levels having energy separations in the order of typical thermal energies. In this case, the intensity ratio of the related emission peaks from two thermally coupled excited levels from the 4fn configuration follows Boltzmann’s law. This is appealing for applications as the calibration model is simple, can be linearized, and the fitting parameters identified with independently measurable quantities.

        Unfortunately, Boltzmann thermometers only offer high measurement precision in a limited temperature range that depends on the employed energy gap. A simple thermodynamic solution to overcome this obstacle and to widen of the dynamic range of the luminescent thermometer is the usage of more than two excited states.

        Aiming at high-temperature luminescence thermometry introduces several additional practical issues, however. A prominent issue is the appearance of a sloping background due to black-body radiation. As the intensity of thermal radiation scales with the fourth power of temperature and the maximum of the related emission spectrum shifts towards the visible range at higher temperatures, an ideal wide-range luminescent thermometer should emit in the ultraviolet (UV) range.

        Gd³⁺ (4f⁷) fulfills all these requirements for effective wide-range luminescence thermometry but is unfortunately a weak UV absorber. Thus, we came up with an alternative concept. Since the blue InGaN LED emitting at around 450 nm has undergone a tremendous optimization that allows to reach high incident power densities with blue LEDs very easily nowadays, we used Pr³⁺ (4f²) as a sensitizer for Gd³⁺. Pr³⁺ absorbs at 450 nm and under sufficiently high irradiance, this ion is excited into 4f¹5d¹-based states with energies in the UV range, like the excited levels of Gd³⁺. This allows to excite Gd³⁺ by means of a cheap 450 nm light source.

        This conceptually different approach also limits the choice for a potential host compound. It should be stable over a wide range of temperatures, preferentially without any phase transition. In addition, the 5d orbitals are much more sensitive to the locally surrounding ligand field, which requires a host that shifts the energy of the excited 4f¹5d¹ configuration of Pr³⁺ to a similar range of the excited levels of Gd³⁺. Finally, the host structure should intrinsically keep the doped Gd³⁺ ions far apart. This allows to incorporate a higher fraction of Gd³⁺ ions to increase the absorption strength while avoiding energy migration processes that could lead to concentration quenching of the UV luminescence of Gd³⁺. The huntite-type crystallizing borate YAl₃(BO₃)₄ matches all these requirements and is an ideally suited host compound to test these requirements.

        Upon inclusion of all these design principles, we were able to demonstrate that YAl₃(BO₃)₄:Pr³⁺,Gd³⁺ can be used as an effective luminescent thermometer with constantly low relative measurement uncertainty (< 0.1%) in the range between 30 K and 800 K. The current work therefore represents the first example of a wide-range luminescent Boltzmann thermometer that also incorporates practical concerns in its design.

---

---