Open Access Article
Daniele Malpicci
ab,
Daniele Marinotto
b,
Elena Lucenti
b,
Luca Zecchinello
ab,
Chiara Botta
c,
Alessandra Forni
*b and
Elena Cariati
*ab
aDepartment of Chemistry, Università degli Studi di Milano, via Golgi 19, 20133 Milano, Italy. E-mail: elena.cariati@unimi.it
bInstitute of Chemical Sciences and Technologies ‘‘Giulio Natta’’ (SCITEC) of CNR, via Golgi 19, 20133 Milano, Italy. E-mail: alessandra.forni@scitec.cnr.it
cInstitute of Chemical Sciences and Technologies “Giulio Natta” (SCITEC) of CNR, via Corti 12, 20133 Milano, Italy
First published on 28th January 2026
The photophysical behavior in solution, PMMA films and crystals of PhCHO mono-, TTPhCHO, and tri-substituted, TT(PhCHO)3, cyclic triimidazole (TT) derivatives is here reported and analyzed through computational and X-ray diffraction studies. The two compounds are poorly fluorescent in diluted solution but display aggregation-induced emissive, AIE, behavior in the solid state where molecular and multiple supramolecular long lived radiative deactivation channels are turned on. In particular, in PMMA films, TTPhCHO is mailny fluorescent while TT(PhCHO)3 is only phosphorescent. In crystals, both compounds show exclusively long lived components of molecular (HEP), CHO⋯TT (MEP) and π–π (TT-TT or TT-Ph, LEP) supramolecular origin as disclosed through theoretical and single crystal X-ray diffraction analysis. The different behaviour observed in PMMA films has been assigned to the easier S–T intersystem crossing associated with the three vs. one carbonyl substituents in TT(PhCHO)3 and TTPhCHO, respectively.
Typical strategies to obtain organic RTP materials include, on one side, incorporation of heavy halogen atoms or functional groups containing N, O or P atoms with lone-pair electrons, able to increase the intrinsically weak Spin–Orbit coupling (SOC) through accessible (n,π*) transitions. Particularly relevant in this regard are aromatic carbonyl-based materials, exhibiting highly efficient ultralong organic phosphorescence in the solid state due to the coupling of (n,π*) and (π,π*) excited-states.23 On the other side, structural rigidification at room temperature can reduce the molecular motions responsible for nonradiative decay and protect the triplet excitons from oxygen. To this aim, several approaches, such as crystallization-induced phosphorescence,24,25 host–guest system,26–30 H-aggregation,31,32 polymerization33 or doped polymers34–37 have been proposed resulting in the stabilization of the triplet state. Among these strategies, the incorporation of phosphorescent dopants in polymeric matrices has been pursued for the development of multifunctional RTP materials endowed with appealing processability. In this context, multicolor and mechano-responsive 3D printable materials with tunable properties have been prepared by integrating imidazole-based chromophores into different polymeric matrices.38 The rigid molecular environment provides RTP films suitable for applications in anticounterfeiting, afterglow display and as visual frequency indicators.39 Stretchable afterglow hydrogels for digital light processing 3D printing technology and anti-counterfeiting encryption have been realized by introducing afterglow microcrystals within various hydrogels.40
On the other side, crystal engineering through intermolecular interactions of the organic building block has been widely employed as a strategy for the design and realization of molecular architectures with RTP properties.41
Frequently, intermolecular interactions are able not only to stabilize emitting states but also to generate supramolecular structures characterized by their own radiative deactivation channels.42–45 Among others, it has been demonstrated46 that aromatic aldehydes and ketones can promote RTP through intermolecular electronic coupling (IEC) of the carbonyl and aromatic subunits thanks to their different (i.e. (n,π*) and (π,π*)) excited-state configurations. Such mechanism requires proximity and a specific geometrical arrangement of the two moieties within the crystal structure, with the carbonyl group of one molecule stacked approximately parallel to the aromatic group of a neighboring molecule.
Understanding the relationship between intermolecular interactions and RTP features of organic materials is therefore crucial to improve the material design process and optimize their performance in devices.
In this context, in the last decade we have explored an extended series of organic compounds based on triimidazo[1,2-a:1′,2′-c:1″,2″-e][1,3,5]triazine or cyclic triimidazole (from hereafter TT), characterized by a rich photophysical behavior comprising multiple fluorescence and phosphorescence of molecular and supramolecular origins, anti-Kasha emissions and excitation dependent photoluminescence.47 TT prototype itself displays aggregation-induced emissive, AIE, features including ultralong phosphorescence (up to 1 s) under ambient conditions associated with the presence of strong π–π stacking interactions in its crystalline structure.32 It has been demonstrated how the introduction of halogens, ethynyl, carboxyl and various chromophoric (such as pyrene, pyridines, carbazole and thiophene) groups on the TT scaffold allows not only to preserve, in most cases, RTP of π–π stacked origin, but also to activate molecular RTP and, in some cases, additional supramolecular short- or long-lived contributions associated with either hydrogen bonding, HB,43,48 or extrinsic and intrinsic heavy atom effects.49
Here, an investigation on the role of the carbonyl group on the photophysical behavior of TT functionalized with one or three benzaldehyde groups, namely 4-(triimidazo[1,2-a:1′,2′-c:1″,2″-e][1,3,5]triazin-3-yl)benzaldehyde, TTPhCHO, and 4,4′,4″-(triimidazo[1,2-a:1′,2′-c:1″,2″-e][1,3,5]triazine-3,7,11-triyl)tribenzaldehyde, TT(PhCHO)3, is reported. It is shown that these derivatives are characterized by intriguing AIE behavior. The two compounds show, in fact, weak fluorescence in solution but multiple molecularly and supramolecularly induced phosphorescences in the aggregated phases. In particular, π–π stacked aggregates and the related RTP contribution, together with IEC-based RTP, associated with the presence of short carbonyl⋯TT contacts, are observed for both compounds.
The two compounds display superimposable absorption and emission spectra in diluted DCM solution (1 × 10−5 M, see Fig. 1 and Table 1) characterized by an absorption at about 320 nm and a broad, weak fluorescence centered at 374 nm (Φ = 0.8 and 0.6%, τ = 1.6 and 2.0 ns, Fig. S13 and S21, for TTPhCHO and TT(PhCHO)3, respectively).
![]() | ||
| Fig. 1 Normalized absorption (dashed lines) and emission (full lines, λexc = 300 nm) of DCM (1 × 10−5 M) solution of TTPhCHO (black) and TT(PhCHO)3 (red). | ||
| 298 K | 77 K | |||||
|---|---|---|---|---|---|---|
| Φ (%) | λem (nm) | τav | λem (nm) | τav (ms) | ||
| TTPhCHO | DCM | 0.8 | 374 | 1.64 ns | ||
| PMMA | 0.7 | 360 | 1.35 ns | |||
| 434 | 488 µs | |||||
| 470 | ||||||
| 530 | 3.38 ms | |||||
| Crystals | 7 | 430 | 123.6 µs | 1.00 | ||
| 480 | 473 | |||||
| 560 | 6.01 ms | 534 | 47.86 | |||
| TT(PhCHO)3 | DCM | 0.6 | 374 | 2.02 ns | ||
| PMMA | 1.3 | 434 | 147.3 µs | |||
| 555 | 90.6 µs | |||||
| Crystals | 0.8 | 410 | 536.2 µs | 420 | 1.84 | |
| >430 | 3.92 ms | 465 | 11.8 | |||
| 575 | 10.5 ms | 570 | 65.7 | |||
To assess the molecular optical behavior in a rigid environment at RT, diluted (0.5 w/w%) spin-coated films were prepared using PMMA as matrix due to its effectiveness in restricting molecular motions and reducing non-radiative deactivation channels.38,47,50 Films of the two compounds display similar absorption features (λmax = 320 nm) but quite different steady-state excitation dependent photoluminescent behavior (Fig. 2). The similarity of absorption maxima in PMMA matrix and DCM solutions indicates that chromophores in films exist predominantly in a monodispersed state. However, minor contributions from aggregated species must be considered also at low loading, especially in the emissive patterns (here confirmed by the different DCM absorption and PMMA excitation spectra of both compounds, see Fig. S11 and S12). In fact, aggregation induced emissions (AIE) are observed for both compounds. At 300 nm excitation, the photoluminescence, PL, spectrum of both compounds in PMMA is characterized by a multicomponent emission. More in detail, PL of TTPhCHO is dominated by a narrow fluorescence centered at 360 nm (Φ = 0.7%, τ = 1.3 ns, Fig. S14), resembling the emission in DCM solution, together with additional low energy tails at about 435 and 500 nm (Fig. 2 upper panel, left). On the other hand, TT(PhCHO)3 shows a PL spectrum quite different from that collected in DCM comprising only one broad band with recognizable maxima at about 430 and 500 nm (Φ = 1.3%, Fig. 2 bottom panel, left). The different contributions can be better isolated by varying excitation wavelength. For TTPhCHO, the low energy components (at 430 and 500 nm) can be activated at the expense of the otherwise prevailing higher energy fluorescence by exciting at 360 and 430 nm (Fig. 2, upper panel, right), respectively. At the same excitation wavelengths, TT(PhCHO)3 shows better resolved spectra with emissions at 430 and 525 nm, respectively (Fig. 2, bottom panel, right). From lifetime measurements, all emissions except for the 360 nm one of TTPhCHO, can be classified as phoshporescences (see Table 1 and Fig. S15, S16, S27 and S28).
To better investigate the long-lived components, we collected for both compounds delayed emission spectra of their PMMA films. Intriguingly, for TTPhCHO three different long-lived emissions are recognized when emission is collected at different delay times and integrated in different time windows (see Fig. 3, left): (i) at short delay (50 µs) a high energy phosphorescence (HEP) superimposable to the 434 nm emission observed in steady-state experiments; (ii) at longer delay (200 µs) a medium energy phosphorescence (MEP) centered at 470 nm, which was not revealed through steady-state spectra; (iii) by extending the time window up to 5 ms a low energy phosphorescence (LEP) is recognized at 530 nm. This latter corresponds to the 500 nm one observed in steady-state spectrum with maximum position affected by the presence of the higher energy contributions. In the case of TT(PhCHO)3, analogously to steady-state measurements, two broad phosphorescences are evidenced in the delayed spectra (Fig. 3, right): a vibronically resolved HEP with maximum at 434 nm, closely matching the 430 nm PL emission, and LEP centered at 555 nm. This latter again shifted to lower energy with respect to its PL analogue where the presence of the higher energy component affects the emission wavelength. Despite measurements in various experimental conditions (excitation wavelength, time delays and time windows) the presence of a medium energy phosphorescence, similar to TTPhCHO, was not evidenced. However, this contribution could be hidden by the broad HEP probably having similar lifetimes.
The photoluminescent behaviour of the two compounds was further investigated on crystalline samples (see Fig. 4 and 5). A notable result in this respect is the absence of fluorescent emission for both crystalline systems regardless of excitation wavelength. More in details, in the 298 K PL spectrum of TTPhCHO crystals, HEP (at 430 nm, τ = 123.6 µs, Fig. S17) and LEP (at 525 nm, τ = 6.01 ms, Fig. S18) are observed, both having clear contribution from MEP, by exciting at 340 and 430 nm, respectively (Fig. 4 left, solid lines). HEP, MEP and LEP (at 430, 480 and 560 nm, respectively) are better resolved in delayed emission spectra by varying the time delay and integration time window (Fig. 4 left, dashed lines). On the contrary, for TT(PhCHO)3 the presence or absence of MEP cannot, again, be unambiguously established by room-temperature measurements. More specifically, HEP at about 410 nm (τ = 536.2 µs, Fig. S29) is clearly recognized in both steady state and delayed spectra (Fig. 4 right) together with an additional lower energy phosphorescence. This latter is centered at 515 nm (τ = 3.92 ms, Fig. S30) in the steady state spectrum and at 575 nm in the delayed one suggesting the presence of different contributions in the broad 515 nm PL emission. For this reason, we have monitored lifetimes at 575 nm obtaining a significantly longer τ (τ = 10.5 ms, Fig. S31) unambiguously indicating concomitant MEP and LEP in the 515 nm broad emission.
To better disclose different contributions in the emission spectra, we have performed variable temperature spectral analysis. Intriguingly, at 77 K and 300 nm excitation MEP of TTPhCHO crystals, hardly visible in the 298 K PL spectra, becomes dominant (at 473 nm, τ = 1.00 ms, Figure S19) over HEP and LEP, this latter resolved (at 534 nm, τ = 47.86 ms, Figure S20) at 430 nm excitation (Fig. 5 left). The different lifetimes of MEP and LEP are evidenced in the LT delayed spectra by varying time delays and integration windows. In fact, at 77 K for short integration times (<150 µs) MEP is by far the dominant emission; LEP is recognized when integration windows are extended to 1–2 ms, being the main contribution at delays larger than 500 µs (Fig. S22). The temperature dependence of delayed spectra is also indicative of the different origin of the phosphorescences. At short times, MEP dominates the spectrum at any temperature below RT (see Fig. S22–S24). At intermediate times, LEP appears at 140 K by cooling from RT (Fig. S23). For longer times (integration windows of 2 ms), LEP is the main phosphorescence at any temperature (Fig. S22–S26).
For TT(PhCHO)3, HEP (at 420 nm, τ = 1.84 ms, Fig. S32) and LEP (at 500 nm, τ = 11.8 ms, Fig. S33) are evidenced at 77 K by exciting at 300 and 420 nm, respectively (Fig. 5 right).
In time delayed experiments at low temperature, all three phosphorescences are concomitantly observed at short times (delay 20 µs, window 200 µs), while LEP is by far the dominant emission for delays larger than 200 µs (Fig. 5 and Fig. S36). The delayed spectra recorded at short times by lowering the temperature allow to recognize a spectral evolution from HEP (RT) to HEP + MEP (250 K) to HEP + MEP + LEP (150 K), see Fig. S37.
To better interpret the photophysical behavior of the two compounds, we performed a systematic theoretical study and a comparative analysis of their crystal structures, aimed at elucidating molecular and supramolecular features involved in the emissive processes.
The DFT optimized ‘gas-phase’ TTPhCHO and TT(PhCHO)3 molecular structures are weakly or scarcely polar (µ = 3.14 and 1.46 D, respectively). Their TDDFT excitation spectra are almost overlapped (Fig. S40–S41 and Tables S2–S3) and consist of one low energy strong band, which corresponds to states S2 and S2′ for TTPhCHO and TT(PhCHO)3, respectively. The prime refers to the fact that, in the triply substituted derivative, levels have threefold degeneration, so S1′ refers to {S1, S2, S3}, S2′ to {S4, S5, S6}, and so on. S1 and S1′ have (σ, π*) character, with almost vanishing oscillator strength and therefore not observed in the UV-vis spectra. They mainly involve inner HOMO(s) and the LUMO(s), both of them localized on the PhCHO group(s). S2 and S2′, instead, correspond to strong (π,π*) transitions mainly involving HOMO(s), delocalized on the whole molecule, and LUMO(s) localized on the PhCHO group(s). For both compounds, calculations on the ground state geometry provide a triplet state (T3 and T3′) with (π,π*) character, located below S1 and S1′ by ΔES–T = 0.14 and 0.02 eV for TTPhCHO and TT(PhCHO)3, respectively, suggesting easy singlet-to-triplet intersystem crossing (ISC) based on both the different character of the two states (El-Sayed rule) and their small energy difference. Optimization of S1 converges in both cases to a geometry very similar to the ground state one, maintaining the same (σ, π*) character of the Franck–Condon state. Including SOC in TDDFT calculations provides triplet states T2 which are located below S1 by 0.37 and 0.34 eV for TTPhCHO and TT(PhCHO)3, respectively (Fig. 6). Though similar S1–T2 SOC elements are obtained for the two compounds (24 and 17 cm−1, respectively), it should be considered that, in TT(PhCHO)3 three degenerate relaxed S1 states (rather than the single one in TTPhCHO) contribute to determine the singlet-to-triplet ISC, therefore indicating an easier population of triplet exciton with respect to the mono-substituted compound.
![]() | ||
| Fig. 6 Computed energy diagram and spin orbit coupling matrix elements (ξ) for TTPhCHO (left) and TT(PhCHO)3 (right). | ||
Based on these results, it is expected that excitation at 300 nm populates both S1 and S2, and the weak fluorescence observed in solution, with relatively small (0.66 eV) Stokes shift, is imputable to emission from S1 (eventually after internal conversion, IC, from S2), having very low oscillator strength. In PMMA and crystals, the highly favorable S1–T2 ISC becomes competitive with respect to radiative decay from S1, thanks to rigidification effects (through the matrix or intermolecular interactions) which inhibit non-radiative decay paths, explaining observation of HEP for both compounds. This long-lived contribution can, in fact, be originated by radiative decay from T1 (computed at 418 nm for both compounds) after IC from T2. This attribution agrees with its predominance and coincident position (435 nm) in PMMA films of both compounds. Moreover, S1–T2 ISC is expected to be particularly efficient in TT(PhCHO)3 owing to the presence of three carbonyl groups with associated three-fold degeneration of S1, thus explaining the lack of fluorescence in its solid phases (PMMA and crystals).
To interpret the additional long-lived, lower-energy contributions, supramolecular emissive species should be invoked and to do so we have deeply analyzed the single crystal structure of both compounds.
TTPhCHO crystallizes in the P2/n space group with two molecules per asymmetric unit (Table S1, Fig. 7 and Fig. S38). The two independent molecules, labelled with A and B, display the same distortion of the phenyl ring with respect to the TT unit (the dihedral angle between the two units measuring 50.9°), but opposite orientation of the CHO group, which is otherwise coplanar to the phenyl. DFT geometry optimization of A and B provides the same stability for the two conformations. In the crystal structure, the two molecules form, along the crystallographic direction b, segregate columnar aggregates …AAA… and …BBB… with identical interlayer separation, held together through π–π stacking interactions and further stabilized by sheared parallel CO⋯CO ones (Fig. 7, left).51 The distance between triazinic geometrical centroids measures 3.877 Å, denoting quite reduced slippage along the stack. A and B are related to one another by a pseudo-inversion center and are connected, on the TT side, through short C–H⋯N hydrogen bonds, HBs, (rH5A⋯N5B = 2.38 Å and rH5B⋯N5A = 2.51 Å), with TT units lying on almost parallel planes separated by about 0.6 Å. On the PhCHO side, both molecules are in turn linked to their respective centrosymmetry-related partner through CO⋯CO interactions according to a slightly sheared antiparallel motif (rC16A⋯O1A′ = 3.190(3) Å and rC16B⋯O1A″ = 3.383(3) Å).51 As a result, infinite non-parallel ⋯A⋯A⋯B⋯B⋯A⋯A⋯ chains are formed, laterally interacting with each other through C–H⋯N, C–H⋯O and C–H⋯π HBs (Fig. 7, right). Moreover, a short contact between the carbonyl group of A and the TT moiety of B is present (rO1A⋯C1B = 3.129(3) Å), with the two subunits lying on almost parallel planes (the dihedral angle through their least-squares planes is equal to 4.95°, see Fig. 7, left).
TT(PhCHO)3 crystallizes in the C2/n space group with one molecule and a co-crystallized DCM one per asymmetric unit (Table S1, Fig. 8 and Fig. S39). Differently from TTPhCHO, this structure is highly disordered, essentially due to the possibility of each CHO group to assume both orientations observed in TTPhCHO. More precisely, only two of them, i.e. those bearing atoms O1 and O2, statistically occupy two positions (A and B) independently of each other (the latter requiring as well to split the bonded phenyl) resulting in four possible conformations (in Fig. 8 only the major components are shown). They display comparable intermolecular interactions with adjacent chromophores (C–H⋯N and C–H⋯O HBs) or co-crystallized DCM (C–H⋯Cl HBs). The third CHO assumes only one position, coinciding with that of A in TTPhCHO. The phenyl rings are rotated with respect to the TT plane by 50.0, 43.2 (A) or 39.4 (B), and 50.0°, again showing slightly lower distortion than the isolated conformation as given by DFT calculations (optimized values are 52.0, 50.3 and 53.5°). Molecules of TT(PhCHO)3 form centrosymmetric dimeric units through relatively short HBs involving two carbonyl groups (those bearing O1A or O1B and O3), which generate relatively large cavities hosting DCM disordered chains (Fig. 8, left). Moreover, molecules stack along b by overlapping TT with phenyl units (Fig. 8, right), while neither TT⋯TT nor CO⋯CO interactions are present, denoting looser packing with respect to TTPhCHO. Finally, a close carbonyl⋯TT contact is detected also in TT(PhCHO)3, with shorter distances with respect to TTPhCHO (rO3⋯N4 = 2.979(3) Å, rO3⋯C3 = 2.984(3) Å) but greater angle (37.56°) between carbonyl and TT least-squares planes (see Fig. 8, right).
Therefore, both structures share strong carbonyl⋯TT interaction which can be recognized at the origin of IEC, resulting in phosphorescence in crystalline samples.46 It can be proposed that such interactions are responsible for MEP in both TTPhCHO and TT(PhCHO)3, while LEP can be assigned to π–π (TT-TT and TT-Ph, respectively) interactions. In fact, stacking interactions are associated with low energy phosphorescence in various TTs characterized by multiple emissions (a selection of TTs with non chromophoric substituents is reported for comparison in Table S4). This interpretation is supported by the strong spectral similarity of TTPhCHO LEP with that of TTs sharing analogous distances between triazinic centroids (about 3.8 Å), namely TT itself,32 polymorph-H of TT-2py (TT-2pyH),52 3-(2-fluoropyridin-4-yl)triimidazo[1,2-a:1′,2′-c:1″,2″-e][1,3,5]triazine (TT-4PyF)53 and TT co-crystallized as guest in [Zn3(CH3COO)6(H2O)2](TT)2.54 All these compounds, in fact, display a vibronically resolved, narrow phosphorescence with peaks covering the 490-620 nm interval (room temperature spectra of crystals of TTPhCHO, TT, TT-2pyH and [Zn3(CH3COO)6(H2O)2](TT)2 are reported in Fig. 9).
1H and 13C NMR spectra were recorded on a Bruker AVANCE-400 instrument (400 MHz). Chemical shifts are reported in parts per million (ppm) and are referenced to the residual solvent peak (CH2Cl2, 1H 5.32 ppm, 13C 54.0 ppm; DMSO, 1H 2.50 ppm, 13C 39.5 ppm). Coupling constants (J) are given in hertz (Hz) and are quoted to the nearest 0.5 Hz. Peak multiplicities are described in the following way: s, singlet; d, doublet; m, multiplet. Purity of the compounds has been assessed by 1H-NMR (1024–2048 scans) spectra recorded with 13C decoupling to suppress 13C satellite signals.
Polymethylmethacrylate (PMMA) films were prepared by spin coating (2000 rpm, 60s) a dichloromethane (DCM) solution of the chromophore on a quartz substrate (chromophore/PMMA = 0.5 wt%; PMMA = 10 wt% with respect to the solvent).
Photoluminescence quantum yields have been measured using a C11347 Quantaurus–Absolute Photoluminescence Quantum Yield Spectrometer (Hamamatsu Photonics K.K), equipped with a 150 W Xenon lamp, an integrating sphere and a multichannel detector. Steady state emission and excitation spectra and photoluminescence lifetimes have been obtained using a FLS 980 (Edinburg Instrument Ltd) spectrofluorimeter. The steady state measurements have been recorded by a 450 W Xenon arc lamp. Photoluminescence lifetime measurements have been performed using a EPLED-300 (Edinburg Instrument Ltd) and microsecond flash Xe-lamp (60W, 0.1–100 Hz) with data acquisition devices time correlated single-photon counting (TCSPC) and multi-channel scaling (MCS) methods, respectively.
Average lifetimes are obtained
from bi-exponential or three-exponential fits.
Phosphorescence spectra are obtained with a NanoLog composed by a iH320 spectrograph, with PPD-850 single photon detector module and Time-Gated Separation by exciting with a pulsed Xe lamp. The spectra are corrected for the instrument response. Low temperature measurements have been performed by immersion of the sample in a liquid N2 quartz dewar (steady state measurements) or by using an Oxford DN1704 cryostat (Phosphorescence measurements).
NMR data for TTPhCHO (9.4 T, CD2Cl2, 298 K, δ, ppm): 1H NMR 10.09 (s, 1H), 7.96 (m, 4H), 7.84 (d, J = 1.6 Hz, 1H), 7.82 (d, J = 1.6 Hz, 1H), 7.33 (s, 1H), 7.30 (d, J = 1.6 Hz, 1H), 7.18 (d, J = 1.6 Hz, 1H). 13C NMR: 192,1 (CHO), 137,3 (C), 136,8 (C), 136,1 (C), 135,9 (C), 134,3 (C), 130,9 (CH), 129,8 (CH), 129,8 (CH), 129,6 (CH), 129,1(CH), 128,5 (C), 112,3 (CH), 111,9 (CH) (Fig. S1-S5).
Single crystals suitable for XRD analysis were obtained by slowly cooling to room temperature a boiling CH3CN solution of TT(PhCHO)3.
NMR data for TT(PhCHO)3 (9.4 T, DMSO-d6, 298 K, δ, ppm): 1H NMR 10.09 (s, 3H), 7.98 (m, 12H), 7.42 (s, 3H). 13C NMR: 192.7 (CHO), 137.5 (C), 135.5 (C), 133.9 (C), 130.0 (CH), 128.9 (CH), 128.8 (CH), 126.6 (C) (Fig. S6–S10).
CCDC 2500777 and 2500778 contain the supplementary crystallographic data for this paper.61a,b
| This journal is © the Partner Organisations 2026 |