Marco
Bazi
a,
Matteo
Tomassoni
a,
Luca
Bellucci
a,
Gregorio
Bottaro
*b,
Marzio
Rancan
b,
Simona
Samaritani
a,
Lidia
Armelao
cd and
Luca
Labella
*a
aDipartimento di Chimica e Chimica Industriale and CIRCC, Università di Pisa, via Giuseppe Moruzzi 13, I-56124, Italy. E-mail: luca.labella@unipi.it
bCNR ICMATE and INSTM, presso Dipartimento di Scienze Chimiche, Università di Padova, via Marzolo 1, I-35131, Italy. E-mail: gregorio.bottaro@cnr.it
cDipartimento di Scienze Chimiche and INSTM, Università di Padova, via Marzolo 1, I-35131, Italy
dDipartimento di Scienze Chimiche e Tecnologie dei Materiali (DSCTM) Consiglio Nazionale delle, Ricerche, Piazzale A. Moro 7, 00185 Roma, Italy
First published on 30th May 2025
Lanthanide dinuclear complexes [Ln2(tta)6(μ-pyrzMO)2] (Ln3+ = Eu3+ and Gd3+; Htta = thenoyltrifluoroacetone; pyrzMO = pyrazine-N oxide) can behave as ligands for the synthesis of molecular heterometallic complexes. Fast reactions at room temperature occur with [Pt(μ-Cl)Cl(EPh3)]2 (E = P, As) affording as a single product [Ln2(tta)6(μ-pyrzMOPtCl2EPh3)2] (E = P, Ln = Eu, 1; Ln = Gd, 2); (E = As, Ln = Eu, 3; Ln = Gd, 4). Similarly, the new platinum compound [PtClppy(pyrzMO)], 5, obtained reacting [Pt(μ-Cl)ppy]2 (ppy = 2-phenylpyridine) with pyrzMO and structurally characterized through XRD, has been used to prepare [Ln2(tta)6(μ-pyrzMOPtClppy)2] (Ln = Eu, 6; Ln = Gd, 7) starting from the anhydrous formally unsaturated [Ln(tta)3] fragments. XRD studies on all heterometallic compounds have established a tetranuclear Ln2Pt2 molecular structure with the platinum fragment coordinated to the nitrogen atom of the pyrazine while the N-oxide donor functionality is bridging two lanthanide centres as in the dinuclear lanthanide precursors. PL and thermometric properties of the Eu3+ complexes are strongly correlated with the nature of the ligand coordinated to Pt that is PPh3, AsPh3 and ppy. Upon UV excitation, 1 and 3 exhibit intense red emission, typical of Eu3+ ions, with a luminance of ∼10 cd m−2 significantly higher than that of 6 (∼0.3 cd m−2). Temperature-induced variations in emission intensity have been exploited to develop a series of luminescent thermometers. The integrated intensity of the 5D0 → 7F2 europium transition can be used as the thermometric parameter (Δ). 1 and 3 reach Sr = 1 at around 200 K, while 6 exhibits a steeper decay in Δ, achieving Sr = 1 at approximately 100 K. For 1 and 3, the temperature sensitivity is related to non-radiative deactivation channels involving the platinum-containing fragment and tta ligands. For 6, instead, the interactions are limited only to the Pt-containing fragment.
A family of luminescent Eu3+ dinuclear complexes with general formula [Eu2(β-dike)6(N-oxide)y] (Hβ-dike = Htta; dibenzoylmethane, Hdbm; benzoyltrifluoroacetone, Hbta; N-oxide = pyrazine-N-oxide, pyrzMO; 4,4′-bipyridine-N-oxide, bipyMO); (y = 2); (Hβ-dike = Hhfac; N-oxide = pyrzMO; bipyMO); (y = 3) has been recently prepared to study the effect of the β-diketonato and N-oxide ligands on their thermometric properties.50,51 The products were obtained in mild conditions and in high yields reacting stoichiometric amounts of anhydrous lanthanide β-diketonates and the heterotopic ligands in toluene. X-ray single crystal studies revealed that, in all compounds, the heterotopic ligands exhibit hypodentate coordination, bridging the two lanthanide centres exclusively through the oxygen atom while featuring a pendant nitrogen atom.51 The luminescence and thermometric properties have been found to be dependent both from the diketonate and the N-oxide so that they can be modulated using different building blocks. An alternative approach to controlling emission properties could involve coordinating a second metal center—platinum, in our choice—to the nitrogen atom of the hypodentate ligand, utilizing the dinuclear lanthanide precursors as metallo-ligands. Notably, a recent paper reports that heterometallic Ln4Pt6 molecular aggregates (Ln = Eu, Yb) exhibit temperature-dependent emission.52 However, to the best of our knowledge, no studies in the literature have explored the modulation of thermometric properties through the incorporation of different platinum fragments into a molecular heterometallic d/f architecture. In this work, we report that the simple dinuclear complexes [Ln2(tta)6(N-oxide)2], (Ln = Gd, Eu) can be used as metallo-ligands for the synthesis of heterometallic tetranuclear lanthanide platinum complexes [Ln2(tta)6(μ-pyrzMOPtCl2EPh3)2] (Ln = Eu, E = P, 1, E = As, 3; Ln = Gd, E = P, 2, E = As, 4) involving the coordination of a platinum fragment to each pendant nitrogen donor, Scheme 1a. The same tetranuclear array [Ln2(tta)6(μ-pyrzMOPtClppy)2] (Ln = Eu, 6; Gd, 7, Hppy: 2-phenylpyridine) can be obtained from a pyrzMO-coordinated platinum precursor, Scheme 1b.
![]() | ||
| Scheme 1 Access to d/f heteronuclear Ln2Pt2 arrays. In (b) quotation marks indicate an unsaturated fragment containing Ln. | ||
In these compounds, the photophysical and temperature-dependent emission properties are modulated by the introduction of different platinum fragments. In fact, upon illumination with ultraviolet light, europium complex powders emit characteristic red light due to Eu3+ ions. The luminescence of 1 and 3, visible to the naked eye, is more intense than that of 6, highlighting the crucial role of platinum-containing fragments in determining emission properties. A similar behaviour is observed when the luminescence of these complexes is studied as a function of temperature between 80 and 300 K.
SG. In each case, the structures feature an inversion centre located at the midpoint of the Ln2O2 core, stabilized by two bridging pyrzMO ligands. The crystal packing results in a network of voids that accommodate solvated chloroform molecules that are lost upon drying. In the molecular structures of the heterometallic compounds, the [Ln2(tta)6(pyrzMO)2] dinuclear core of the precursor is preserved, with pyrzMO bridging the two lanthanide centres solely through its oxygen atom (μ-O), while its nitrogen site is coordinated to platinum. Thus, the heterotopic divergent ligand effectively bridges the lanthanide and platinum centers by exploiting their differing affinities for oxygen and nitrogen donors. The Ln⋯Ln and Pt⋯Pt distances are 4.2101(7) Å and 14.9870(7) Å for 1, 4.2617(12) Å and 14.7089(8) Å for 2, 4.2999(8) Å and 14.8335(9) Å for 3.
In these compounds, the Ln centres show a coordination number of 8 due to the three chelating β-diketonates and the two bridging pyrzMO units. A stereochemical study of the eight-vertex polyhedra was performed by means of continuous shape measures (CShM). Calculations (SHAPE 2.1)63 of the degree of distortion of the LnO8 core with respect to an ideal eight-vertex polyhedron show that the coordination geometry is closer to a triangular dodecahedron for 1 and 3, and to a square antiprism for 2 (Table S3†). A square planar geometry of the trans isomer is observed for platinum as expected. Compound 4 is isostructural with its europium analogue (Tables S1† and S2).
Experimental evidence confirms the incorporation of a platinum fragment at room temperature appended to an almost unchanged dinuclear lanthanide core. Remarkably, the lanthanide precursor [Ln2(tta)6(μ-pyrzMO)2] behaves here as a f-based metallo-ligand.
The dinuclear core with O-bridging N-oxide fragments has frequently been observed in previous studies with tris-diketonato lanthanide fragments,64–68 demonstrating its high stability. In principle, a platinum precursor with a N-coordinated hypodentate pyrzMO ligand is expected to behave as free pyrzMO in the coordination to the formally unsaturated lanthanide fragment thereby stabilizing a dinuclear lanthanide unit.51 For this reason, a different synthetic approach was envisaged when the sparingly soluble dinuclear [Pt(μ-Cl)(ppy)]2,69–71 (ppy = orthometalated phenylpyridine) was considered. As a matter of fact [Pt(μ-Cl)(ppy)]2 was normally obtained as a greenish and not easy to purify solid and was not reasoned to be a suitable precursor for a reaction with a lanthanide metallo-ligand to be carried out with rigorous stoichiometric molar ratios of the reactants. A procedure starting from a formally unsaturated lanthanide fragment and a platinum metallo-ligand with a N-coordinated hypodentate pyrzMO can afford the same molecular architecture for heterometallic Ln2Pt2 complexes.
The platinum complex [PtCl(ppy)pyrzMO] (5) is synthesized by reaction of the dinuclear [Pt(μ-Cl)(ppy)]2 with two equivalents of pyrzMO in THF, in a straightforward reaction. The soluble product is separated by traces of a dark solid and precipitated from a bright yellow solution with heptane. Analytical and spectroscopic data agree with the expected composition, showing through NMR the presence of a single species in solution (Fig. S4 and S5†), the SP4-4 isomer, identified through SC-XRD. The molecular structure (Fig. 2) shows the pyrzMO ligand bonded to platinum only through the nitrogen atom. The oxygen donor site is not involved in any significant bonding interaction. In the crystal structure, complex 5 assembles into one-dimensional supramolecular chains through staggered stacking interactions between the phenylpyridine rings, with an average inter-ring distance of 3.5 Å (Fig. 2b).
![]() | ||
| Fig. 2 (a) Molecular structure of 5 and (b) staggered stacking. Colour code: C, grey; O, red; N, blue; Cl, dark green; Pt, dark blue. | ||
In agreement with our data, several literature reports present a bridge-cleavage reaction of the dinuclear [Pt(μ-Cl)(ppy)]2 complex for a pyridine-based ligand yielding an initial mononuclear SP4-3 isomer, due to the strong trans effect exerted by the aryl donor, and to the successive isomerization to the thermodynamically more stable SP4-4 isomer form.72–74
5 promptly reacts with one equivalent of [Ln(tta)3] (Ln = Eu, Gd) in dichloromethane at room temperature with an initial bright yellow solution turning into a bright red solution. The reaction product, yellow-orange in colour, is precipitated by adding heptane and presents analytical data supporting the stoichiometry 1 to 1 of the two metal fragments. The IR spectra showed several similarities with the dinuclear complex [Ln2(tta)6(μ-pyrzMO)2] suggesting the formation of a dinuclear lanthanide core (Fig. S3†). The molecular structure of the tetranuclear Pt2Eu2 complex 6 was elucidated through SC-XRD showing the presence of the dinuclear lanthanide core in which the oxygen atoms of two pyrzMO bridge two europium centres similarly to the molecular precursor [Eu2(tta)6(μ-pyrzMO)2] (Fig. 3). Compound 6 crystallizes in the P21/n space group. The Eu⋯Eu and Pt⋯Pt distances are 4.270(4) Å and 14.648(3) Å, respectively. As found in the above discussed structure, an inversion centre is located at the midpoint of the Eu2O2 core. A stereochemical study of the eight-vertex polyhedron by means of CShM shows that the minimal degree of distortion of the EuO8 core with respect to an ideal eight-vertex polyhedron is reached for the square antiprism geometry (Table S3†). The nitrogen donor atom of the heterotopic divergent ligand is bonded to platinum in a SP4-4 isomer. The molecular structure of 6 closely resembles that of 1 and 2 despite the different synthetic approach. [Gd2(tta)6(μ-pyrzMOPtClppy)2], 7, has been found to be isostructural with 6 (Tables S1 and S2†). 5 uses the oxygen donor functionality to bind two lanthanide unsaturated fragments in fast reactions and mild conditions, behaving as the free pyrzMO moiety.
These results show that the dinuclear lanthanide unit [Ln2(tta)6(μ-pyrzMO)2] is sufficiently inert to be used as a f-based metallo-ligand and that this dinuclear core is sufficiently stable to be formed in mild conditions also starting from the platinum precursor 5 and [Ln(tta)3] (Ln = Eu, Gd).
Pt(II) complexes have been extensively studied due to their distinctive photophysical behaviour as triplet emitters. Their absorption and emission characteristics can be finely tuned by modifying the electronic and steric properties of the coordinating ligands. In particular, bulky ancillary ligands are often employed to suppress Pt⋯Pt interactions, which are favoured by the nearly square-planar geometry of Pt(II) centres. These interactions can promote aggregate formation, leading to red-shifted emissions and reduced colour purity. Additionally, ligands containing heavy atoms are frequently introduced to enhance spin–orbit coupling, thereby facilitating intersystem crossing and improving triplet emission efficiency.78–81
The range of available Pt(II) complexes with well-characterized electronic properties is vast. Therefore, selecting an appropriate Pt complex for the synthesis of heterobimetallic Eu2Pt2 complexes is nontrivial. In this study, we focused on two structural classes: (i) complexes featuring pnictogen-based neutral donors (e.g., PtCl2(PPh3) and PtCl2(AsPh3)), and (ii) cyclometalated Pt(II) complexes, specifically PtCl(ppy), where ppy = 2-phenylpyridine. The latter represent one of the most thoroughly investigated families of Pt(II) emitters, with ppy often considered the prototypical cyclometalating ligand. In contrast, ligands such as triphenylarsine (AsPh3) have been much less explored compared to their phosphorus analogue (PPh3). Nonetheless, we found them intriguing because in Pt(II) complexes of the general formula PtL(PnPh3) (Pn = P, As, Sb), a systematic red shift in the energy of electronic transitions is observed along the pnictogen series from P to Sb, and a hyperchromic shift (increase in the intensity of a spectral band) can be observed as well.80–83 The bathochromic shift, typically on the order of several hundred cm−1, can significantly influence the luminescence behaviour of Eu2Pt2 systems, particularly when the relevant excited-state energies fall within 17
000–18
000 cm−1, i.e., near the 5D0 emitting level of Eu3+.
The study of the absorption spectra of the Gd2Pt2 and Eu2Pt2 complexes provides information on the nature of the electron transitions in this family of complexes.
Simply looking at the sample powders we can divide them into two groups based on their colours. In fact, 1 and 3 appear as bright yellow powders while 6 is darker with orange tones.
To evaluate the effect of Pt-based fragments, it is useful to first identify the spectral features associated with the [Ln2(tta)6(μ-pyrzMO)2] core. The solid-state absorption spectrum of [Gd2(tta)6(μ-pyrzMO)2] (Fig. S6†) displays a broad band primarily located in the UV region, with an absorption onset around 420 nm. In contrast, the spectrum of [Eu2(tta)6(μ-pyrzMO)2] exhibits a markedly different profile, characterized by a sharp peak at 415 nm and a pronounced tail extending across most of the visible range. This feature is attributed to ligand-to-metal charge transfer (LMCT) transitions.84–91
The diffuse reflectance spectra of the Ln2Pt2 complexes (Fig. 4 and Fig. S7†) are significantly broadened and extend further into the visible region. The absorption onsets are observed around 500 nm for complexes 1 and 2, 530 nm for complexes 3 and 4, and approximately 600 nm for complexes 6 and 7.
![]() | ||
| Fig. 4 Diffuse reflectance absorption spectra of the powders of (a) Gd2Pt2 and (b) Eu2Pt2 complexes. | ||
In general, Pt(II) complexes exhibit intense absorptions in the UV region (ca. 260–330 nm), attributed to ligand-centred π → π* and n → π* transitions, along with broad LMCT bands that extend into the visible region.41,78–83 Accordingly, the absorption spectra of Gd2Pt2 (Fig. 4a) and Eu2Pt2 (Fig. 4b) complexes can be conceptually divided into two regions: below 400 nm, where ligand-centred transitions dominate, and above 400 nm, where the absorptions are primarily associated with MLCT transitions.
For the gadolinium series, the visible-region absorptions are exclusively attributed to MLCT transitions involving the Pt(II) moieties. The absorbed portion of visible light varies with the nature of the Pt(II) fragment, following the trend: PtCl2(PPh3) < PtCl2(AsPh3) < PtCl(ppy). In the Eu2Pt2 complexes, the overall band assignment parallels that of the Gd analogues. However, the high-energy portion of the MLCT envelope results from a combination of transitions involving both Pt and Eu centres (Fig. S6 and S7†).
Therefore, similarly to what observed in other heterobimetallic complexes,84 Eu3+ emission can be sensitized through two excitation paths involving the two kinds of ligands. In general, ligand-centred energy levels close to Eu3+ excited states favour non-radiative deactivation mechanisms. While these mechanisms may reduce the brightness of the emitters, they can positively influence their thermometric properties.92
The excitation spectra (PLE) of 1 and 3 (Fig. 5a) show broad signals in the ultraviolet region with a tail extending up to approximately 450 nm. Below 400 nm, the PLE spectrum of 6 is similar to those of 1 and 3. However, 6 exhibits a shoulder in the visible region that is absent in the spectra of the other two complexes, which is ascribed to transitions centred on the Pt-moiety. In fact, while the emission spectra of 1 and 3 show the typical well-resolved multiplets of europium, in the emission spectrum of 6 the signals due to the transitions from the 5D0 excited state to the fundamental multiplets 7FJ (J = 0, 1, 2, 3 and 4) of Eu3+ are broader and emerge from a background due to the emission, of comparable intensity, of the Pt-containing fragment (Fig. 5b and S8†). The emission of europium occurs by indirect excitation exploiting the absorption properties of the organic ligands and the energy transfer from ligand-centred triplet states to the emitting levels of the metal ion according to the well-known scheme that characterizes the “antenna process”. In addition to the shape of the spectra, the different behaviour of the complexes is also visible to the naked eye. In fact, when 1 and 3 powders are illuminated with ultraviolet light a bright red luminescence is observed. In contrast, under the same illumination conditions the light emission from 6 powders is faint.
![]() | ||
| Fig. 5 (a) Photoluminescence Excitation (PLE) spectra monitored on the most intense Eu3+ emission peak. (b) Emission spectra of Eu3+-containing complexes excited at 375 nm. | ||
To compare the emission performance of different emitters, quantification of the emitted light in certain conditions is required. To this aim, the most frequently used parameter is the photoluminescence quantum yield (PLQY) which for solid samples should be measured using an integrating sphere. Besides problems related to the measurements that can be tricky, the major drawback of PLQY is that it provides a partial indication. PLQY gives us information “only” on the fraction of absorbed photons that are re-emitted, but it does not provide information about the number of absorbed photons nor on the capability of the sample to absorb light. Or better, these data are used to calculate PLQY but seldom explicitly reported. In other words, to have an intense emission, a molecule or a material should absorb as many photons as possible and the PLQYs should be equal to 1. In solution the ability of a molecule to absorb light is given by the molar extinction coefficient. Therefore, for luminescent molecules in solution, a good parameter to quantify the amount of emitted light is the brightness (B = PLQY × ε), i.e. the product between PLQY and the molar absorption coefficient. In solid state the situation is more complicated, and the determination of an absorption coefficient is not trivial. In addition, in different fields of application, brightness can have different definitions. These aspects have been critically analyzed recently.93,94 Alternatively, to quantify the intensity of the emitted light and relate it to human perception of light and colours, one can use luminance (expressed in cd m−2), a photometric quantity that weights the radiation intensity based on the spectral response of the human eye.93,94Table 1 presents a collection of spectral parameters for the examined compounds, enabling the comparison of their spectroscopic properties.
| Sample |
T a (103 cm−1) |
τ exp (μs) |
τ
rad b (μs) |
Φ b % |
PLQY % |
η c (%) |
Luminanced (cd m−2) |
|---|---|---|---|---|---|---|---|
| a Determined from 80 K emission spectra (Fig. S9†) of Gd-complexes (2, 4, 7).95 Since the 0–0 vibronic band is not clearly observed the interval between the peak onset and first observed maximum is reported. b Value calculated following a well-established procedure.84,95 c η = PLQY/Φ × 100. d Measured illuminating the samples with a 375 nm LED with a power of 2.4 mW. | |||||||
| 1 | 18.2–20 | 206 | 1114 | 18 | 2.5 | 14 | 9.3 |
| 3 | 18.2–20 | 216 | 1139 | 19 | 2.5 | 13 | 10.5 |
| 6 | 18.2–15.8 | 52 | 810 | 6 | — | — | 0.3 |
At room temperature 1 and 3 have the same absolute quantum yield (Table 1) and the sensitization efficiencies, calculated as the ratio of quantum efficiency to PLQY, is less than 20%. The europium emission is less intense than that observed in other dinuclear complexes connected by N-oxides, where the nitrogen donor atom is not coordinated.51 The different behaviour is due to the presence of platinum. Consistent with the hypothesis proposed during the discussion of the diffuse reflectance absorption spectra (Fig. 4), MLCT of platinum-containing moieties introduce low-energy levels that favour the non-radiative deactivation of europium excited states. A comparable situation, but with more pronounced Eu3+ emission quenching, is observed in 6 bearing ppy ligand, which exhibits extremely weak emission, rendering its PLQY unmeasurable. It is worth noticing that the absorption spectrum of 6 is wider in the visible range than that of 1 and 3 (Fig. 4). The remarkable similarity in ligands employed for 1 and 3 accounts for their identical PLQY values and luminance, approximately 10 cd m−2 under 375 nm LED irradiation (2.4 mW) also indicating comparable light absorption capabilities. In contrast, 6 exhibits a significantly lower luminance of 0.3 cd m−2.
400 cm−1, making it an effective sensitizer for europium.51,91 Based on the observations at room temperature, the excitation channel involving Pt fragment is likely to determine the emission properties. The 80 K spectra of the Gd-based complexes (2, 4, 7, Fig. S9†) are entirely located at energies below 20
000 cm−1 with the first maximum at 18
200 cm−1 for 2 and 4 and 15
800 cm−1 for 7 (Table 1). The observed bands, which do not change regardless of the excitation wavelength, are attributable to transitions centred on the Pt-containing fragments, while signals at an energy compatible with the tta triplet are not clearly distinguishable, presumably shadowed by the transitions centred on Pt. Energy levels in close proximity to Eu3+ 5D0 will certainly play a primary role in determining the thermometric properties of this family of complexes.
As thermometric parameter (Δ) we used the integrated intensity of the 5D0 → 7F2 transition. The Δ vs. T curves and the relative thermal sensitivity (Sr, eqn (1)) trend are reported in Fig. 6.
![]() | (1) |
![]() | ||
| Fig. 6 (a) Thermometric parameter (Δ) and b) relative thermal sensitivity (Sr) for Eu3+-containing complexes 1, 3 and 6 (λexc = 375 nm). | ||
The Δ(T) curves of 1 and 3 exhibit very similar trends, as one would expect given the great similarity of the two compounds. The value of Δ decreases gradually with increasing temperature with a pseudo-sigmoidal trend. Regarding relative thermal sensitivity, it is noted that for these two molecules the quality factor Sr ≥ 1 (quality criterion for using these compounds as highly sensitive luminescent thermometers) is reached at 200 K. In 6, we note a steeper decrease in the value of Δ, Sr reaches the value of 1 already at about 100 K. This characteristic means that when 6 is used as a luminescent molecular thermometer, it has a minimum operating temperature 100 degrees lower than that of 1 and 3 and a much wider operating range (temperature range with Sr ≥ 1). Previously, we studied other compounds of the same family51 in which the β-diketonato units are interconnected by pyrazine N-oxide and 4,4′-bipyridine N-oxide in which Sr became greater than one at significantly higher temperatures between 250 and 300 K. In that case, the thermometric properties were determined by the energy of the triplets of the β-diketonates and by the presence of LMCT transitions.
All ligand-centred and tta-to-Eu charge transfer (CT) transitions are located at high energies (above approximately 24
000 cm−1), with the latter exhibiting a long tail extending into the visible region (Fig. S6†). These energy levels are too high to play a direct role in the observed thermally induced variations in emission intensity.
At lower energies, absorption bands corresponding to metal-to-ligand charge transfer (MLCT) transitions centred on the Pt(II) moieties are observed (Fig. 4 and Fig. S6, S7†). The spectral and thermometric behaviour of the complexes is strongly influenced by the nature of the ligand coordinated to Pt(II)
:
PPh3 in complex 1, AsPh3 in complex 3, and ppy in complex 6. Triphenylphosphine and triphenylarsine ligands are electronically and sterically similar, and as a result, the spectroscopic features of complexes 1 and 3 are nearly identical (Fig. 4, 5 and Table 1). This similarity is also reflected in their thermometric performance, with Δ(T) and Sr(T) curves showing matching trends. These results support the hypothesis that the thermometric behaviour of the Eu2Pt2 complexes is primarily governed by the photophysical properties of the Pt-based moiety.
In contrast, substituting the EPh3 ligand with a cyclometalating ppy ligand, characterized by lower-energy MLCT states, as confirmed by both absorption and emission spectra (Fig. 4 and Fig. S7, S9†), results in markedly different thermal behaviour. Specifically, complex 6 exhibits a sharp decrease in integrated emission intensity at lower temperatures compared to complexes 1 and 3. This leads to enhanced thermometric sensitivity in the low-temperature range.
Overall, the coordination of N-oxide ligands to Pt(II) to form heterobimetallic assemblies has proven to be an effective strategy to tailor the operational temperature range of luminescent thermometers.
To obtain information on the luminescence quenching mechanisms that determine the thermometric response of the complexes, we fitted the experimental Δ vs. T curves using the Mott–Seitz model (eqn (2)) where Δ0 is the value of Δ at 0 K, ΔEi the activation energy, αi the ratio between the nonradiative and radiative deactivation probabilities for the considered path, kB the Boltzmann constant, and T the temperature in K.
![]() | (2) |
The experimental curves of 1 and 3 can be well reproduced using a two-term model (Fig. S10†), while only one is sufficient for 6. The activation energies are reported in Table 2.
| Sample | ΔE1 (cm−1) | ΔE2 (cm−1) |
|---|---|---|
| 1 | 475 | 2274 |
| 3 | 296 | 2187 |
| 6 | 357 | — |
Noteworthy, in all complexes there is a process with an activation energy between 300 and 500 cm−1 while only in 1 and 3 a second channel with a significantly higher energy (ΔE > 2000 cm−1) is active. To identify the possible non-radiative deactivation pathways of the excited emitting europium states, we have constructed a simplified energy level diagram (Fig. 7).
![]() | ||
Fig. 7 Simplified energy level diagram reporting the excited states of Eu3+, the triplet energy of the tta ligand, and the energy of Pt-centred MLCT states. All but tta triplet levels are indicated as boxes to evidence the uncertainties on the energy due to the lack of a vibronic progression in the spectra used for their determination (Fig. S9†). Eu3+-tta MLCT are not reported due to their high energy (24 000 cm−1). | ||
For 1 and 3, the two activation energies found are compatible with interactions between the excited states 5D0 → 7F0 and 5D0 → 7F1 of Eu3+ and MLCT states of –PtPPh3 and –PtAsPh3 moiety (Table 2, ΔE1) and of the tta triplet (ΔE2) indicating that both channels are active in determining the properties of these molecular thermometers. For 6, instead, the interactions only occur with the –PtClppy fragment, as supposed in the discussion of the emission properties at room temperature.
We have also investigated the emission properties and the possible application as a molecular thermometer of 5 (Fig. 8). The absorption spectrum shows a rather broad band that extends from 500 nm down to the ultraviolet. At room temperature, exciting at the maximum of absorption, a weak emission with a lifetime of 2 μs is observed (Fig. 8b). The emission band of 5 covers almost all the visible and shows a low-energy tail that extends into the near infrared.
The observed emission band is likely due to 3MLCT [Pt(5d) → π*] as commonly observed in 2-phenylpyridine cased complexes.96 Lowering the temperature down to liquid nitrogen we observed an increase of the emission intensity as depicted in Fig. 8b suggesting a possible employment of this complex as molecular thermometer. Sr values are always slightly above 1% K−1 with small fluctuations indicating a good thermal sensitivity over a 200 K temperature range. The maximum Sr is 1.52% K−1 at 253 K.
72 and [Ln(tta)3]98 (Ln: Eu, Gd) were prepared according to literature methods. All other reagents were obtained from commercial sources (Merck, Stream Chemicals, Alfa Aesar) and used as received. NMR spectra for 5 were recorded at room temperature on Jeol400 spectrometer; IR spectra were recorded in ATR (Attenuated Total Reflectance) on PerkinElmer Spectrum One or Agilent Cary630FTIR spectrometer. Elemental analysis (C, H, N) was performed at the Dipartimento di Chimica e Chimica Industriale, Università di Pisa (Italy). Absorption spectra of powder samples were recorded using a Cary 5000 UV-vis spectrometer equipped with an integrating sphere. The spectra were normalized and plotted as F(R) vs. wavelength. F(R) is the Kubelka–Munk function. Room-temperature luminescence spectra of sample powders were recorded with a Horiba JobinYvon Fluorolog-3 spectrofluorimeter. Absolute photoluminescence quantum yields (PLQYs) were calculated from corrected emission spectra obtained by means of an integrating sphere. Estimated errors on PLQY and excited-state lifetimes are ±20% and 10%, respectively. Temperature-dependence experiments (223–373 K) were carried using a Horiba T64000 triple spectrometer and a Linkam THMS600 heating/freezing microscope stage having a temperature stability of <0.1 K over an 83–873 K temperature range. Luminance measurements were performed by means of a Konica-Minolta CS-160 Luminance and Colour Meter equipped with cs135 close-up lens. Ca. 3 mg of sample powders were mixed and finely grinded with 300 mg of BaSO4 and successively placed on watch glass. Luminance data, given as the average of 10 measurements, are collected normal to the sample at the minimum focusing distance of the CS-160. To excite samples luminescence a 375 nm (2.4 mW) LED powered with a benchtop stabilized power supply.
The Gd3+ derivative [Gd2(tta)6(μ-pyrzMOPtCl2PPh3)2] (2) has been obtained following a similar procedure starting from [Gd2(tta)6(μ-pyrzMO)2] (180 mg, 0.098 mmol), [PtCl(μ-Cl)PPh3]2 (103 mg, 0.097 mmol) in dichloromethane (10 mL). 2: 168 mg, 59.8%. Anal. calcd for [Gd2(tta)6(μ-pyrzMOPtCl2PPh3)2], C92H62N4Cl4F18Gd2O14Pt2P2S6, %: C, 38.2; H, 2.2; N, 1.9. Found: C, 37.9; H, 2.4; N, 1.8. IR-ATR (1700–1100 cm−1, Fig. S2†): 1615 (s), 1596 (s), 1579 (s), 1539 (s), 1507 (m), 1471 (m), 1436 (m), 1410 (s), 1355 (m), 1310 (s), 1294 (s), 1248 (s), 1231 (s), 1188 (s), 1138 (s), 1122 (s), 1100 (m). X-ray quality crystals were grown by slow diffusion of pentane vapour in a chloroform solution.
The Gd3+ derivative [Gd2(tta)6(μ-PyrzMOPtCl2AsPh3)2] (4) has been obtained following a similar procedure starting from [Gd2(tta)6(μ-pyrzMO)2] (366 mg, 0.200 mmol), [PtCl(μ-Cl)AsPh3]2 (229 mg, 0.200 mmol). 4: 447 mg, 75.0%. Anal. calcd for [Gd2(tta)6(μ-pyrzMOPtCl2AsPh3)2], C92H62N4As2Cl4F18Gd2O14Pt2S6, %: C, 37.1; H, 2.1; N, 1.9. Found: C, 36.9; H, 2.2; N, 1.7. IR-ATR (1700–1100 cm−1, Fig. S2†): 1613 (s), 1594 (s), 1576 (s), 1538 (s), 1506 (m), 1467 (m), 1437 (m), 1408 (s), 1355 (m), 1307 (s), 1294 (s), 1247 (s), 1229 (s), 1186 (s), 1137 (s). Single crystals were grown by slow diffusion of pentane vapour in a dichloromethane solution. Although unsuitable for an X-ray study meeting publication standard, the crystals exhibited the same metric of 3. Unit cell parameters have been reported in Table S2.†
The Gd3+ derivative [Gd2(tta)6(μ-pyrzMOPtCl(ppy))2] (7) has been obtained following a similar procedure starting from [Gd(tta)3] (128 mg, 0.177 mmol) and [PtClppy(pyrzMO)] (74.7 mg, 0,155 mmol). 7: 142 mg, 70.4%. Anal. calcd for [Gd2(tta)6(μ-pyrzMOPtClppy)2], C78H48N6Cl2F18Gd2O14Pt2S6, %: C, 36.0; H, 1.9; N, 3.2. Found: C, 35.7; H, 1.9; N, 3.4. IR-ATR (1700–1100 cm−1, Fig. S3†): 1613 (w), 1600 (s), 1580 (m), 1539 (m), 1506 (w), 1487 (w), 1456 (m), 1442 (w), 1410 (m), 1353 (w), 1305 (s), 1248 (m), 1230 (m), 1188 (s), 1137 (s). Crystals were grown by slow diffusion of pentane vapour in a dichloromethane solution. Although unsuitable for an X-ray study meeting publication standard, the crystals exhibited the same metric of 6. Unit cell parameters have been reported in Table S2.†
The structures were solved with Olex299 by using ShelXT100 structure solution program by Intrinsic Phasing and refined with the ShelXL101 refinement package using least-squares minimization. In the last cycles of refinement, non-hydrogen atoms were refined anisotropically. Hydrogen atoms were included in calculated positions, and a riding model was used for their refinement. Refinement details for compounds 1, 2, 3, 5 and 6 are given in the ESI,† together crystallographic table (Table S1†) and asymmetric unit images for all the compounds studied.
Cambridge Crystallographic Data Centre (CCDC) numbers 2395042–2395045 and 2395089† contain the supplementary crystallographic data for this paper.
Concerning absorption and emission properties of Eu2Pt2 complexes, a straightforward visual inspection of the powdered samples reveals two distinct groups distinguishable by their colour. Compounds 1 and 3 appear as bright yellow powders, while compound 6 exhibits a darker orange hue. Notably, the absorption spectrum of 6 extends up to 600 nm, resulting in partial overlap with the most intense emission bands of Eu3+. Ligand-centred energy levels proximate to the Eu3+ excited states favour non-radiative deactivation pathways, which can be exploited to enhance thermometric properties near room temperature. At room temperature, the emission spectra of 1 and 3 display the characteristic, well-resolved multiplets of europium. Conversely, the emission spectrum of 6 exhibits broadened signals arising from transitions between the 5D0 excited state and the fundamental multiplets 7FJ (J = 0, 1, 2, 3, and 4) of Eu3+, superimposed on a background emission ascribed to Pt-containing fragment. The striking similarity in ligands between 1 and 3 accounts for their identical photoluminescence quantum yield values and luminance, approximately 10 cd m−2 under 375 nm LED irradiation (2.4 mW), indicating comparable light absorption capabilities. In contrast, 6 exhibits a significantly lower luminance of 0.3 cd m−2.
Regarding thermometric properties, the Δ(T) curves of 1 and 3 display highly similar trends, as expected given their structural similarity. The Δ value gradually decreases with increasing temperature, following a pseudo-sigmoidal pattern. The relative thermal sensitivity (Sr) reaches a value of 1 at 200 K and increases to 6 K−1% at 300 K. For compound 6, a steeper decrease in Δ is observed, and the temperature range in which Sr > 1 (T > 100 K) is approximately 100 K broader than that of compounds 1 and 3. Previous studies on analogous compounds featuring β-diketonato units interconnected by pyrazine N-oxide and 4,4′-bipyridine N-oxide, with non-coordinated nitrogen atoms, revealed significantly higher temperatures (250–300 K) for Sr > 1.
The photophysical properties of Eu2Pt2 complexes are strongly governed by the nature of the Pt-based moieties. While PPh3 and AsPh3 ligands lead to nearly identical absorption and emission features due to their similar electronic properties, cyclometalated –PtCl(ppy) complexes exhibit significantly lower energy MLCT transitions. These states influence the thermal response of the Eu3+-centred emission, with complex 6 (ppy) showing a steeper emission quenching at lower temperatures compared to complexes 1 and 3 (PPh3, AsPh3), thereby enhancing thermometric sensitivity. These findings highlight the key role of Pt-centred MLCT transitions in modulating the temperature sensitivity of Eu2Pt2-based luminescent thermometers and emphasize the importance of heterometallic architecture design in tailoring performance across different temperature ranges.
CCDC 2395042–2395045 and 2395089† contain the supplementary crystallographic data for this paper.
Footnote |
| † Electronic supplementary information (ESI) available: Additional spectroscopic and crystallographic data. CCDC 2395042–2395045 and 2395089. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5qi00893j |
| This journal is © the Partner Organisations 2025 |