Chain‐Mobility‐Enabled 1D Lanthanide Coordination Polymer Glassy Scintillators for Underwater X‐Ray Videography

ABSTRACT

Lanthanide(III) coordination scintillators feature efficient triplet harvesting and narrow‐band emission, making them promising for X‐ray imaging applications. However, their rigid coordination environments hinder melt‐processing into large‐area, transparent glassy scintillator screens. Herein, a “rigid‐node flexible‐linker” molecular design strategy that facilitates chain‐mobility‐enabled glass‐forming in lanthanide coordination scintillators by constructing one‐dimensional (1D) coordination chains is proposed. By integrating dibenzoylmethane antenna with flexible dual‐phosphine‐oxide linkers (OP‐Cn, n = 2, 4, 6, 8), a series of 1D Eu‐OP‐Cn coordination polymers is constructed, enabling the simultaneous realization of efficient ligand‐sensitized radioluminescence and the chain‐mobility required for vitrification, thereby allowing transformation from crystalline powders into glassy states. Benefiting from rigid local coordination environments that suppress nonradiative decay, crystalline Eu‐OP‐C2 exhibits a near‐unity photoluminescence quantum yield (97.5%) and an ultrahigh relative light yield of 70379 photons MeV⁻¹. Besides, elongating the alkyl‐chain length increases segmental flexibility, allowing Eu‐OP‐C6/C8 to form water‐stable, transparent glassy scintillators via melt‐quenching method. Notably, Eu‐OP‐C8 glass delivers radioluminescence intensity 12.1 times higher than Bi₄Ge₃O₁₂, enabling high‐resolution X‐ray imaging (> 30 lp mm⁻¹) and real‐time underwater X‐ray videography (2K, 60 fps). Moreover, this strategy is readily extendable to Tb³⁺, Sm³⁺ and Dy³⁺, establishing a general molecular‐design paradigm for melt‐processable lanthanide coordination glassy scintillators.

A novel “rigid‐node flexible‐linker” molecular design enables melt‐processable one‐dimensional lanthanide coordination polymer scintillators by balancing efficient ligand‐sensitized radioluminescence with chain mobility for glass formation. The obtained Eu(III)‐based coordination polymer glassy scintillator delivers high‐resolution X‐ray imaging and real‐time underwater videography, establishing a versatile platform for processable lanthanide scintillators in complex environments for advanced X‐ray imaging.

1. Introduction

Luminescent lanthanide(III)‐based coordination compounds (e.g. Eu³⁺, Tb³⁺), assembled through the interaction between Ln³⁺ centers and organic ligands, constitute an important class of luminescent materials owing to their narrow‐band emissions, long‐lived excited states, and highly tailorable coordination environments [1, 2]. These characteristics have enabled broad applications in optical sensing, molecular lighting, anticounterfeiting, and energy conversion [3, 4]. In recent years, the efficient ligand‐to‐metal “antenna effect”, combined with the intrinsically high atomic numbers of lanthanides, has further highlighted their potential in X‐ray responsive photophysics, positioning lanthanide coordination systems as promising candidates for high‐performance scintillators [5, 6].

Recently, significant progress has been achieved in lanthanide coordination scintillators based on zero‐dimensional (0D) molecular complexes and three‐dimensional (3D) coordination frameworks [7, 8]. Discrete 0D lanthanide complexes enable efficient ligand‐to‐metal energy transfer and enhanced triplet exciton harvesting [9, 10], while 3D lanthanide metal–organic frameworks (Ln‐MOFs) benefit from rigid coordination environments and high structural stability [11, 12], collectively demonstrating the feasibility of lanthanide‐based systems for static X‐ray imaging applications. However, both 0D and 3D lanthanide coordination scintillators are inherently crystalline in nature [13, 14]. In practical device fabrication, large‐area scintillation screens usually require dispersing crystalline powders into polymer matrices, which inevitably dilutes emissive components and introduces optical scattering due to particle aggregation and refractive‐index mismatch (Figure 1a) [15, 16]. These factors severely compromise transparency, scintillation efficiency, and imaging resolution, particularly under low‐dose conditions, and fundamentally hinder the realization of real‐time X‐ray videography.

FIGURE 1.

(a) Structural characteristics of 0D and 3D crystalline lanthanide coordination scintillators and schematic illustrations of their corresponding scintillation screen fabrication. (b) Schematic illustration of scintillator screen fabrication based on chain‐mobility‐enabled 1D Ln‐CPs scintillators designed via the rigid‐node flexible‐linker molecular strategy and processed by the melt‐quenching method. (c) Chemical structure of Eu‐CPs (n = 1, 2, 3, 4). (d) Crystal structure of the one‐dimensional coordination chain for Eu‐OP‐C2, ignoring H atoms.

To address these intrinsic limitations, one‐dimensional lanthanide coordination polymers (1D Ln‐CPs) provide an alternative and largely unexplored structural paradigm. Unlike discrete molecular complexes or rigid 3D frameworks, metal‐organic‐metal bridged 1D coordination chains exhibit anisotropic topologies that enable polymeric chain entanglement [17, 18]. Such entanglement facilitates the formation of stable amorphous networks that effectively suppress molecular rearrangement and recrystallization, thereby enabling melt‐processable, transparent glassy states with polymeric characteristics [19, 20]. This structure‐driven glass‐forming ability has been experimentally demonstrated in several 1D systems via the melt‐quenching process [21, 22], confirming the potency of this topology in accessing glassy states. This chain‐based glass physics can impart 1D Ln‐CPs glasses with long‐term structural stability and optical uniformity comparable to conventional polymer glasses, while mitigating spontaneous crystallization and optical degradation commonly encountered in small‐molecule glassy scintillators. However, most reported 1D lanthanide coordination polymers are constructed from rigid aromatic linkers, which lack the segmental mobility required for vitrification and therefore impede melt processing and glass‐state formation [23, 24].

Motivated by these considerations, we propose a “rigid‐node flexible‐linker” molecular design to realize chain‐mobility‐enabled 1D Ln‐CP glassy scintillators (Figure 1b). In this architecture, rigid lanthanide‐based emissive nodes preserve a well‐defined local coordination environment to ensure efficient ligand‐sensitized radioluminescence, while flexible bridged linkers impart sufficient segmental mobility to enable polymeric chain entanglement and vitrification [25, 26]. Using dibenzoylmethane (DBM) as an efficient triplet‐harvesting antenna and a homologous series of flexible linkers, namely 1, n‐bis(diphenylphosphoryl)[X]ane (OP‐Cn, n, [X] = 2, eth; 4, but; 6, hex; 8, oct), a family of 1D Eu(III) coordination polymers (Eu‐OP‐Cn, n = 2, 4, 6, 8) is constructed as a model system (Figure 1c,d). The short‐chain Eu‐OP‐C2 crystallizes into a rigid coordination framework, exhibiting a near‐unity photoluminescence quantum yield (PLQY, 97.5%) and an ultrahigh relative light yield of 70379 photons MeV⁻¹. In contrast, increasing linker flexibility enables Eu‐OP‐C6 and Eu‐OP‐C8 to undergo melt‐quenching into optically transparent glasses. Notably, the Eu‐OP‐C8 glass exhibits radioluminescence (RL) intensity an order of magnitude higher than that of Bi₄Ge₃O₁₂ (BGO), together with excellent environmental and water stability, enabling high‐resolution static X‐ray imaging (30 lp mm⁻¹). And it is capable of real‐time underwater X‐ray videography (2K 60 fps), allowing clear visualization of the internal structures of the chip in the recorded videos. Moreover, this strategy is readily extendable to Tb‐, Sm‐, and Dy‐based analogues, establishing a versatile molecular‐design paradigm for melt‐processable, lanthanide coordination glassy scintillators.

2. Results and Discussion

2.1. Synthesis and Structural Characterization of 1D Eu‐CPs

As shown in Figure S1, a homologous series of dual‐phosphine‐oxide ligands OP‐Cn (n = 2, 4, 6, 8) was synthesized via direct oxidation of the corresponding phosphines using hydrogen peroxide. The ¹H NMR spectra are in good agreement with the proposed structures of the OP‐Cn ligands, exhibiting all the expected chemical shifts and multiplicities (Figure S2). Subsequently, 1D Eu(III)‐based coordination polymers, denoted as Eu‐OP‐Cn (n = 2, 4, 6, 8), were constructed by coordination of Eu³⁺ ions and DBM ligands with OP‐Cn linkers. Detailed synthetic procedures and crystallization conditions are provided in the Supporting Information. Single crystals suitable for X‐ray diffraction were obtained by slow solvent evaporation, enabling unambiguous structural determination. It reveals that Eu‐OP‐C2, Eu‐OP‐C4, and Eu‐OP‐C8 crystallize in the triclinic P‐1 space group, whereas Eu‐OP‐C6 adopts a monoclinic P2₁/c structure (Table S1). Despite differences in space group and alkyl‐chain length, all four compounds share a common 1D coordination polymer topology.

Taking Eu‐OP‐C2 as a representative example, each Eu³⁺ ion is coordinated by six carbonyl oxygen atoms from three DBM ligands and two P = O groups from two OP‐C2 ligands, forming a distorted square antiprismatic (SAPR‐8) geometry. The phosphine oxide ligands act as bridging units between adjacent Eu³⁺ centers, generating infinite metal–organic–metal chains (Figures S4–S7). As the alkyl‐chain length increases from OP‐C2 to OP‐C8, the coordination chains gradually evolve from relatively linear arrangements to more pronounced zigzag conformations, reflecting increasing segmental flexibility along the polymer backbone. Besides, powder X‐ray diffraction (PXRD) patterns of the crystalline powder samples are in well agreement with those simulated from single‐crystal data (Figure S8), confirming the phase purity and structural integrity of the coordination polymers. X‐ray photoelectron spectroscopy (XPS) further supports these findings. As shown in Figure S9, the photoemission peaks for Ln‐CPs at 1163.1 eV (Eu 3d_5/2), 1133.1 eV (Eu 3d_3/2), 530.1 eV (O 1s), 284.1 eV (C 1s), 132.1 eV (P 2p) are observed [27]. These analyses further confirm that europium remains in the trivalent oxidation state and that the organic ligands are not reduced during synthesis, indicating the excellent chemical stability of the 1D Ln‐CPs [28].

2.2. Scintillation Performance of 1D Eu‐CPs Crystalline Powders

The RL spectra of Eu‐OP‐Cn (n = 2, 4, 6, 8) crystalline powders measured under an X‐ray dose of 278 µGy s⁻¹ are shown in Figure 2a. They were corrected using the factory‐generated emission‐correction file of the Edinburgh Instruments FS5 system, which accounts for the wavelength‐dependent optical throughput and PMT quantum efficiency (Figure S3). All samples exhibit the characteristic Eu³⁺ emission peaks at 581, 594, 615, 654, and 701 nm, corresponding to the 5D₀ → 7F_J (J = 0 – 4) transitions, with the dominant emission centered at 616 nm (5D₀ → 7F₂) [5]. As the alkyl‐chain length increases, the RL intensity decreases monotonically, following the order Eu‐OP‐C2 (330844 a.u.) > Eu‐OP‐C4 (318670 a.u.) > Eu‐OP‐C6 (183583 a.u.) > Eu‐OP‐C8 (151620 a.u.). Under identical irradiation conditions, Eu‐OP‐C2 exhibits an RL intensity approximately 57 times that of the commercial scintillator BGO (6085 a.u.), while Eu‐OP‐C8 still delivers over 30‐fold higher intensity. The relative light yields were calculated by using commercial BGO as a reference scintillator under identical X‐ray irradiation and detection conditions [29], Eu‐OP‐C2 exhibits an exceptionally high relative light yield of 70379 photons MeV⁻¹ (Figure 2b and Figure S10) [24]. Notably, even with extended alkyl‐chains, crystalline Eu‐OP‐C8 maintains a high relative light yield of 42430 photons MeV⁻¹, outperforming most reported lanthanide coordination scintillators (Table S2) [30].

FIGURE 2.

(a) RL spectra and (b) relative light yield of Eu‐OP‐Cn measured under an X‐ray dose rate of 278 µGy s⁻¹, with BGO as reference. (c) Linear fitting of the maximum RL intensity from 4.58 to 278 µGy s⁻¹, with detection limits calculated using the 3σ/slope method. (d) Normalized RL intensity vs time (on/off cycles) of Eu‐CPs under X‐ray irradiation at a dose rate of 278 µGy s⁻¹. (e) PXRD patterns of Eu‐OP‐C2, comparing the simulated pattern derived from single‐crystal data with those of the crystalline powder before and after 30 min of X‐ray irradiation.

As shown in Figure 2c and Figure S11, all Eu‐OP‐Cn samples show excellent linearity over the dose‐rate range of 4.58 – 278 µGy s⁻¹. Based on the 3σ/slope criterion [31], the corresponding detection limits are calculated to be 40.5, 47.3, 54.6 and 62.8 nGy s⁻¹ for Eu‐OP‐C2, Eu‐OP‐C4, Eu‐OP‐C6, and Eu‐OP‐C8, respectively, which are more than two orders of magnitude lower than typical medical X‐ray dose rates (≈5.5 µGy s⁻¹) [32]. Besides, irradiation at 278 µGy s⁻¹ for 36 on‐off cycles results in negligible RL degradation, with all samples retaining over 96% of their initial intensity (Figure 2d). PXRD patterns collected before and after irradiation show no discernible changes (Figure 2e and Figure S12), confirming the excellent radiation robustness of the 1D Eu‐CPs.

To elucidate the origin of the alkyl‐chain length dependent scintillation behavior of the Eu‐OP‐Cn series, their photophysical properties were systematically investigated. As shown in Figure 3a, all Eu‐OP‐Cn compounds exhibit nearly identical excitation and emission profiles with large Stokes shifts exceeding 200 nm, which effectively suppresses self‐absorption and reabsorption losses [5, 33]. In view of the highly efficient dual‐antenna energy‐transfer pathway operative in the Eu‐OP‐Cn system, further analysis of the triplet energy levels reveals that the T₁ states of the OP‐Cn ligands are nearly identical (≈25 000 cm⁻¹), while the energy gap between the triplet state of the DBM ligand and the excited state of Eu³⁺ is approximately 4000 cm⁻¹ (Figure 3b and Figure S13) [10]. This gap lies within the optimal range for efficient ligand‐to‐metal energy transfer, indicating that the radioluminescence of the Eu‐OP‐Cn system is predominantly governed by an efficient ligand‐sensitized process [34, 35]. Specifically, excitation proceeds from the OP‐Cn to the DBM ligand via ligand‐ligand charge transfer (LLCT), followed by ligand‐to‐metal charge transfer (LMCT) from DBM to the Eu³⁺ center, enabling effective exciton utilization (Figure 3c) [5]. Density functional theory (DFT) calculations further corroborate this mechanism [36]. As shown in Figure 3d,e and Figures S14–S16, the valence bands are mainly composed of ligand‐centered p orbitals, whereas the conduction bands display pronounced hybridization between ligand states and Eu 5D orbitals. This electronic structure confirms efficient ligand‐to‐metal energy transfer, fully consistent with a dual‐antenna sensitization model [13].

FIGURE 3.

(a) The PL and Photoluminescence Excitation (PLE) spectra of Eu‐OP‐Cn. (b) Relative alignment of the triplet energy levels of ligands and the emissive energy level (⁵D₀) of the Eu³⁺ ion. (c) Schematic diagram of the X‐ray‐excited luminescence principle of 1D Eu‐CPs. (d) The band structure for the direct bandgap transition of Eu‐OP‐C2 with an E_g value of 2.42 eV and (e) the corresponding electronic density of states for the direct bandgap transition of Eu‐OP‐C2. (f) The PLQY of the 1D Eu‐CPs crystalline powders and (g) their time‐resolved luminescence decay profiles. (h) Metal center distance measurement diagram of Eu‐OP‐C2 and (i) distance between adjacent Eu³⁺ atoms in Eu‐OP‐Cn (n = 2, 4, 6, 8).

All samples exhibit PLQYs exceeding 50%, with Eu‐OP‐C2 achieving a near‐unity PLQY of 97.5%. With increasing alkyl‐chain length, the PLQY gradually decreases to 81.3% (Eu‐OP‐C4), 74.1% (Eu‐OP‐C6), and 55.7% (Eu‐OP‐C8) (Figure 3f and Figure S17), closely mirroring the trend observed for RL intensity. Considering the comparable X‐ray absorption efficiencies of the Eu‐OP‐Cn series (Figure S10) and the nearly identical triplet energy levels of the OP‐Cn ligands, the reduction in PLQY and RL intensity is not primarily attributed to differences in ligand energy matching, sensitization efficiency or X‐ray absorption capability [24]. Instead, it is primarily governed by excited‐state relaxation dynamics [37, 38]. As shown in Figure 3g, the emission decay lifetime shortens progressively from 499 µs for Eu‐OP‐C2 to 389 µs for Eu‐OP‐C8. By combining the measured lifetime with the corresponding PLQY values, the calculated nonradiative decay rate constants (k_nr) are found to increase from 47 s⁻¹ for Eu‐OP‐C2 to 1136 s⁻¹ for Eu‐OP‐C8 (Table S3), indicating that elongation of the alkyl chains significantly enhances structural flexibility and introduces nonradiative relaxation pathways [1].

In addition, low‐temperature time‐resolved measurements on the corresponding Gd‐based analogues (Gd‐OP‐Cn, n = 2, 4, 6, 8) show that, whereas Gd(DBM)₃(H₂O)₂ exhibits a long emission decay lifetime of 185.86 µs, the incorporation of OP‐Cn ligands markedly shortens the DBM‐related decay to less than 35 µs (Figure S18), confirming the existence of an interligand coupling pathway between OP‐Cn and DBM [26, 34]. Specifically, the decay lifetimes of the Gd‐OP‐Cn series are 7.85, 9.40, 14.96, and 34.98 µs for Gd‐OP‐C2, ‐C4, ‐C6, and ‐C8, respectively, indicating that progressive elongation of the alkyl chain gradually weakens the spatial coupling between OP‐Cn and DBM. Crystallographic analysis provides direct structural evidence for the chain‐length–dependent nonradiative decay behavior. As summarized in Figure 3h,i and Table S4, both the intrachain and interchain Eu–Eu distances increase monotonically with increasing alkyl‐chain length. Specifically, the intrachain Eu–Eu separation expands from ≈9.4 Å in Eu‐OP‐C2 to ≈12.2 Å in Eu‐OP‐C8, while the interchain distance increases more moderately from ≈13.0 to ≈14.0 Å. The short‐chain derivatives adopt a more compact and tightly packed coordination arrangement, establishing a rigid and defect‐limited local environment that effectively suppresses vibrational relaxation and nonradiative decay. In contrast, elongation of the alkyl linkers leads to progressively expanded Eu–Eu separations, looser packing, and enhanced segmental mobility of the one‐dimensional coordination chains [1]. This structural softening facilitates additional nonradiative relaxation pathways, accounting for the gradual decrease in both PL and RL efficiencies across the Eu‐OP‐Cn series with increasing chain length [5].

2.3. Characteristics of 1D Ln‐CP Glasses

Although increasing alkyl‐chain flexibility partially compromises scintillation efficiency in crystalline Eu‐OP‐Cn, it simultaneously provides the essential structural prerequisite for polymeric segmental mobility and subsequent glass formation. Differential scanning calorimetry (DSC) reveals a monotonic decrease in the melting temperature (T_m) with increasing alkyl‐chain length, from 233°C for Eu‐OP‐C2 to 121°C for Eu‐OP‐C8 (Figure 4a). And distinct glass transition temperatures (T_g) emerge for the Eu‐OP‐C6 and Eu‐OP‐C8. Both satisfy the empirical criterion for stable glass formation (T_g/T_m > 2/3), while Eu‐OP‐C8 exhibits an exceptionally high T_g/T_m ratio of 0.883, among the highest reported for glass‐forming scintillators, confirming the formation of robust amorphous states under practical processing conditions [39, 40].

FIGURE 4.

(a) Differential scanning calorimetry (DSC) curves of Eu‐OP‐C2, Eu‐OP‐C4, Eu‐OP‐C6 and Eu‐OP‐C8. (b) XRD patterns of Eu‐OP‐C8G annealed at 100°C. (c) Fourier transform infrared (FTIR) spectra of Eu‐OP‐C6, Eu‐OP‐C6G, Eu‐OP‐C8 and Eu‐OP‐C8G. (d) RL spectra of Eu‐OP‐C6G, Eu‐OP‐C8G, measured under an X‐ray dose rate of 278 µGy s⁻¹ (50 kV, 79 mA), with BGO as reference. (e) Relative light yield of Eu‐OP‐C8G and the other lanthanide glassy scintillator. (f) Optical images of Eu‐OP‐C8G, Tb‐OP‐C8G, Sm‐OP‐C8G and Dy‐OP‐C8G.

The crystal‐to‐glass transformation was unambiguously confirmed by powder X‐ray diffraction (PXRD). After melt‐quenching of Eu‐OP‐C6 and Eu‐OP‐C8 samples, the sharp Bragg reflections characteristic of the crystalline phase disappear and are replaced by broad amorphous halos (Figure S19), demonstrating the formation of homogeneous coordination polymer glasses (Eu‐OP‐C6G and Eu‐OP‐C8G) [41]. DSC measurements, performed with two successive heating scans, further reveal distinct glass‐state relaxation behaviors for Eu‐OP‐C6G and Eu‐OP‐C8G (Figure S20). Eu‐OP‐C8G exhibits a smaller discrepancy between the two scans and a smoother glass‐transition feature, suggesting that its amorphous framework is more stable and less structurally frustrated. This enhanced stability is further corroborated by variable‐temperature powder X‐ray diffraction (VT‐PXRD). As shown in Figure S21, Eu‐OP‐C8G retains its amorphous character up to higher temperatures without signs of cold crystallization, whereas Eu‐OP‐C6G begins to crystallize under identical thermal conditions, directly confirming the superior thermal stability of the Eu‐OP‐C8G glass. Besides, variable‐temperature polarized optical microscopy (POM) provides direct visual evidence of vitrification: the birefringence observed in the crystalline state vanishes upon heating above the melting temperature, and the isotropic dark field persists upon cooling, confirming the formation of uniform amorphous glasses (Figure S22) [42]. Notably, as shown in Figure 4b, residual crystalline order in Eu‐OP‐C8G is progressively eliminated upon annealing at 100°C for over 6 h. This annealing‐induced relaxation behavior, distinct from recrystallization commonly observed in 0D molecular glasses, provides compelling evidence for polymeric segmental dynamics [43, 44]. Enhanced chain‐mobility along the one‐dimensional coordination backbone disrupts local ordered domains, driving the system toward a thermodynamically stable, low‐energy amorphous state. This behavior closely resembles entanglement‐assisted relaxation in conventional polymer glasses, further substantiating the polymeric glassy nature of Eu‐OP‐C8G [45]. Furthermore, Fourier‐transform infrared spectra recorded before and after vitrification exhibit identical vibrational features (Figure 4c), indicating that no chemical decomposition or ligand rearrangement occurs during their thermal processing.

The scintillation performance of the resulting glassy materials was subsequently evaluated. As shown in Figure 4d,e, Eu‐OP‐C8G exhibits a markedly higher RL intensity than Eu‐OP‐C6G, reaching approximately 12.1 times that of the commercial BGO scintillator under identical irradiation conditions. The corresponding relative light yield of Eu‐OP‐C8G is calculated to be 16309 photons MeV⁻¹, placing it among the highest‐performing lanthanide‐based glassy scintillators reported to date. Moreover, Eu‐OP‐C8G retains over 97% of its initial RL intensity after a cumulative X‐ray dose of 333.6 mGy and more than 90% even after 100 Gy (Figures S23 and S24), demonstrating excellent irradiation stability. Importantly, this vitrification behavior is not limited to europium: analogous lanthanide systems, Ln‐OP‐C8 (Ln = Tb, Sm, Dy) also readily form glasses (Figure 4f), highlighting the generality and extensibility of this polymeric molecular design strategy for glass‐forming scintillators.

Interestingly, the RL intensity of Eu‐OP‐C8G reaches 460.5% of that of Eu‐OP‐C6G (Figure 4d), in sharp contrast to the trend observed in their crystalline counterparts, where shorter alkyl chains yield higher RL efficiencies (Figure 2a). To elucidate the origin of this inversion, detailed photophysical and local structural analyses were conducted. As shown in Figure 5a, the decay lifetime of Eu‐OP‐C6 decreases markedly from 451 µs in the crystalline state to 285 µs after vitrification, whereas Eu‐OP‐C8 remains nearly unchanged at ≈385 µs in both crystalline and glassy forms. Correspondingly, Eu‐OP‐C6G exhibits a low PLQY of 15.6% (21.1% retention relative to its crystalline state), while Eu‐OP‐C8G maintained a superior PLQY of 42.3% (75.9% retention) (Figure S25). Consistently, the calculated k_nr reveal that vitrification significantly enhances nonradiative quenching in Eu‐OP‐C6 (2961 s⁻¹), whereas Eu‐OP‐C8 (1502 s⁻¹) is only marginally affected (Table S3). These results indicate that the glass transition imposes a significantly smaller perturbation on the scintillation efficiency of Eu‐OP‐C8, highlighting its greater tolerance toward structural disorder.

FIGURE 5.

(a) Time‐resolved luminescence decay profiles for Eu‐OP‐C6G and Eu‐OP‐C8G. (b) Experimental k³χ (k) oscillation curves (scatters) and fitting results (solid lines) and (c) experimental FT (k³χ(k)) oscillation curves (scatters) and fitting results (solid lines) for Eu‐OP‐C6, Eu‐OP‐C6G, Eu‐OP‐C8 and Eu‐OP‐C8G.

To further elucidate the origin of the superior scintillation in Eu‐OP‐C8G, extended X‐ray absorption fine structure (EXAFS) analysis for Eu‐OP‐C6 (crystal and glassy state) and Eu‐OP‐C8 (crystal and glassy state) was performed. As shown in Figure 5b,c, the k ³‐weighted EXAFS oscillations and the corresponding Fourier transforms (R‐space) of Eu‐OP‐C6 and Eu‐OP‐C8 exhibit high similarity across both crystalline and glassy phases, suggesting that the primary coordination sphere of Eu³⁺ remains largely intact upon vitrification [46]. Eu L ₃‐edge XANES spectra further confirm that the Eu³⁺ oxidation state is preserved. Besides, quantitative EXAFS fitting (Table S5) reveals that although all samples exhibit high Eu‐O coordination numbers (CN = 10.4–11.0), their response to vitrification differs significantly [47]. In Eu‐OP‐C6, the CN drops from 10.86 to 10.38 (ΔCN = 0.48) during the crystal‐to‐glass transition, accompanied by a slight bond contraction. In contrast, Eu‐OP‐C8 exhibits superior coordination sphere integrity, with the CN only decreasing from 11.00 to 10.77 (ΔCN = 0.23). This suggests that the longer and more flexible alkyl chains in the Eu‐OP‐C8 system act as an effective conformational buffer, accommodating the long‐range disorder in the glassy matrix while minimizing the disruption to the local ligand‐to‐metal coordination [48]. The resulting higher coordination symmetry and stability in Eu‐OP‐C8G facilitate more efficient antenna‐sensitized energy transfer, thereby accounting for its remarkably higher RL intensity compared to Eu‐OP‐C6G.

2.4. Static and Dynamic X‐Ray Imaging Enabled by Eu‐OP‐C8G

As shown in Figure 6a, Eu‐OP‐C8 can be fabricated into transparent, smooth, and homogeneous centimeter‐scale scintillation screens (≈2 × 2 cm²) via a melt‐quenching process. Owing to its fully amorphous nature, Eu‐OP‐C8G effectively eliminates grain‐boundary scattering and refractive‐index mismatch, thereby overcoming the processing and device‐integration limitations inherent to conventional crystalline lanthanide scintillators [49]. This structural advantage provides a solid foundation for achieving high‐spatial‐resolution X‐ray imaging and videography. As shown in Figure 6b, the line‐pair phantom clearly resolves alternating bright and dark regions at a spatial frequency of 30.0 lp mm⁻¹. The modulation transfer function (MTF) value at this frequency was calculated to be 0.23. To further evaluate the spatial resolution beyond the maximum frequency of the line‐pair phantom, the edge method was employed [50], using a 1 mm‐thick tungsten plate as the test sample. As shown in Figure 6c, the spatial resolution reaches 36.8 lp mm⁻¹ at an MTF value of 0.2. In practical imaging demonstrations, Eu‐OP‐C8G enables clear visualization of a flexible printed circuit board and a hippocampal tissue specimen (Figure 6d) with sharp internal features, high contrast, and excellent spatial fidelity.

FIGURE 6.

(a) Optical images of Eu‐OP‐C8G. (b) The corresponding pixel intensity profile across positions and X‐ray contrast images of the line‐pair card (type 39B; 1.5–30 lp mm⁻¹). (c) MTF versus spatial frequency, derived from edge image analysis. (d) Comparison images of flexible circuit board under visible light and X‐ray irradiation, hippocampus comparison image. (e) A schematic illustration of the underwater X‐ray imaging setup. (f) Real‐time underwater X‐ray videography (60 fps, 16.7 ms interval) was performed on the packaged chip, demonstrating motion tracking capabilities without afterglow artifacts. (g) Time‐course X‐ray imaging of a submerged shadow mask. Successive screenshots from real‐time X‐ray videography show the mask after 0, 12, and 24 h of immersion in water.

In addition, Eu‐OP‐C8G exhibits outstanding water stability, retaining strong luminescence after immersion in water for 90 days (Figure S26). The characteristic Eu³⁺ radioluminescence profile remains essentially unchanged, with no obvious alteration in emission shape, and the RL intensity retains 97.5% of its initial value, indicating negligible scintillation degradation after prolonged water exposure (Figure S27). Besides, the XPS spectra of Eu‐OP‐C8G remain essentially unchanged after water immersion (Figure S28), further confirming the excellent water resistance of this glassy material. The pH‐dependent stability of Eu‐OP‐C8G was further evaluated by immersing the glass samples in HCl solution (pH = 4) and NaOH solution (pH = 10) for 7 days (Figure S29). After alkaline treatment, Eu‐OP‐C8G retained its glassy appearance and 99% of its initial RL intensity, demonstrating excellent alkaline stability, highlighting its potential for underwater dynamic imaging. To evaluate this capability, a top‐irradiation X‐ray imaging setup was constructed, in which the X‐ray source was positioned above a water tank, the test object submerged underwater, and the Eu‐OP‐C8G film placed between the object and a high‐sensitivity CCD camera (Figure 6e). Using an electronic chip and a school emblem as imaging targets, the system enabled real‐time visualization of internal structures while simultaneously capturing object motion (Figure 6f and Video S1). Moreover, long‐term monitoring confirms that high‐quality underwater dynamic imaging is maintained after continuous immersion for 12 and 24 h (Figure 6g and Video S2). The internal structure was clearly visible during horizontal movement, with no ghosting or artifact shadows. Collectively, these results demonstrate that glassy lanthanide coordination scintillators offer a promising platform for underwater nondestructive testing and real‐time bioimaging applications.

3. Conclusions

In summary, we present a rigid‐node flexible‐linker molecular design strategy that resolves the long‐standing incompatibility between scintillation efficiency and processability in lanthanide coordination scintillators. By coupling rigid lanthanide‐based emissive nodes with alkylated bis(phosphine oxide) linkers of tunable flexibility, a family of polymeric 1D Ln‐CPs was constructed in which scintillation performance and glass‐forming ability can be systematically regulated by linker chain length. Among them, short‐chain derivatives retain highly rigid coordination environments, exemplified by crystalline Eu‐OP‐C2, which delivers an ultrahigh relative light yield of 70379 photons MeV⁻¹. In contrast, elongation of the alkyl linkers induces segmental mobility in the one‐dimensional chains, enabling Eu‐OP‐C6 and Eu‐OP‐C8 to undergo stable melt vitrification while preserving favorable local coordination symmetry in the glassy state. Notably, the Eu‐OP‐C8 glass scintillator integrates melt processability, long‐term environmental and water stability, and robust radioluminescence, achieving a relative light yield of 16309 photons MeV⁻¹ with excellent irradiation durability. Benefiting from its homogeneous amorphous structure and high optical transparency, Eu‐OP‐C8G enables high‐resolution static X‐ray imaging with a spatial resolution of up to 30 lp mm⁻¹ @ MTF = 0.23, as well as nearly afterglow‐free real‐time underwater X‐ray videography at 60 fps. These results underscore the unique advantages of 1D Ln‐CP glasses for advanced X‐ray imaging under complex and extreme environments. Moreover, the successful extension of this strategy to Tb³⁺‐, Sm³⁺‐, and Dy³⁺‐based systems highlights its generality and versatility. It provides a broadly applicable molecular design paradigm for transforming coordination polymers into functional glassy radiation‐detection materials, opening new opportunities for advanced X‐ray imaging applications across diverse scenarios.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: adma73911‐sup‐0001‐SuppMat.docx.

Supporting File 2: adma73911‐sup‐0002‐VideoS1.mp4.

Supporting File 3: adma73911‐sup‐0003‐VideoS2.mp4.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.