Novel luminescent β-ketoimine derivative of 2,1,3-benzothiadiazole: synthesis, complexation with Zn(II) and photophysical properties in comparison with related compounds

Abstract

A novel β-ketoimine-functionalized 2,1,3-benzothiadiazole (4-(2,1,3-benzothiadiazole-4-ylamino)pent-3-en-2-one, L2H) was synthesized and used as chelating ligand for the complexation with Zn(II) (complex ZnL2). The spectroscopic and photophysical properties of L2H and ZnL2 as well as those of previously described 4-amino-2,1,3-benzothiadiazole (L1) and its complex ZnL1 have been thoroughly studied in the crystalline state and solution. The UV-Vis absorption spectra of free ligands L1, L2H and its deprotonated form L2⁻, and complexes ZnL1 and ZnL2 were assigned on the basis of TD-DFT calculations. In contrast to ZnL2 having conventional single-band fluorescence, ZnL1 demonstrates a broad double-band purple-white fluorescence spanning the entire visible region. On the basis of careful spectroscopic studies the second band of this unusual fluorescence spectrum was attributed to the presence of traces of free ligand L1 in the crystals of complex ZnL1. The energy transfer from the excited state of ZnL1 to a free ligand L1 was revealed.

---

Introduction

2,1,3-Benzothiadiazole (BTD), 10π-electron hetaren (sulfur–nitrogen analogue of naphthalene) possessing positive electron affinity (i.e. being effective electron acceptor),¹ is one of the most popular building blocks in organic electronics.^2–5 The BTD derivatives are particularly interesting in the design and synthesis of organic light-emitting diodes (OLEDs) and low band gap polymers for photovoltaic applications.^6–14 Besides, they are promising as fluorophores in fluorescent thermometers^15,16 for intracellular temperature mapping in living systems¹⁷ and luminescent ligands^14,18 for metal complexes. This motivates further synthesis of various BTD derivatives and in-depth investigation of their photophysical properties. The special interest to the metal complexes containing luminescent ligands arises from the fact that fluorescence intensity may be greatly enhanced by coordination.^19–21 Zinc is environment-friendly metal, and its complexes are widely used as luminophores in science and technology due to their simple and low-cost synthesis as compared to Ru, Ir, Os and Pt complexes. At the same time, only several examples of Zn complexes with non-phtalocyanine BTD derivatives are known (for BTD-containing phtalocyanine metal complexes, see ref. 22 and 23 and references cited therein), and most of them are terpyridine multicomponent compounds, containing BTD derivatives as electron acceptor units.^24–27

Despite current progress of coordination chemistry of BTD and its derivatives, a number of metals involved and complexes structurally defined by X-ray diffraction (XRD) is rather limited.^28–30 Until recently, the complexation of BTD ligands with Zn has not been studied. In 2014, we reported on the preparation and single-crystal XRD characterization of the first zinc complex with 4-amino-substituted BTD (L1), namely, [Zn(L1)₂Cl₂] (ZnL1).²⁸ This paper presents our results on the synthesis and characterization of novel BTD derivative possessing β-ketoimine substituent (4-(2,1,3-benzothiadiazole-4-ylamino)pent-3-en-2-one, L2H) and its zinc complex [Zn(L2)₂] (ZnL2). Most attention has been paid to the in-depth study of spectral and photophysical properties of free L1 and L2H and complexes ZnL1 and ZnL2 exhibiting bright fluorescence in both the solid state and solution. All our experimental results were supported by time-dependent density functional theory (TD-DFT) calculations.

Results and discussion

Synthesis and characterization of L2H

Compound L2H was prepared in the isolated yield of 90% by condensation of compound L1 with excess of acetylacetone (Scheme 1). According to single-crystal XRD, from evaporated mother solution L2H crystallized in the form of two polymorphs (Fig. 1), α-L2H (major, m. p. = 122–124 °C) and β-L2H (minor, m. p. = 111–113 °C), which could be separated mechanically. In the α-polymorph, the neighboring molecules are connected by two shortened contacts S⋯N (3.255 Å), while in the β-polymorph there are no such contacts (Fig. 1). Furthermore, the structure of α-L2H consists of the stacks along the axis a featuring π–π interactions of the neighboring molecules with the distance between the centres of their benzene rings being 3.72 Å. In the case of β-L2H, there are no stacks, but the π–π interactions between pairs of neighboring molecules also exist with the distance between the centres of the hetero rings being 3.47 Å.

Scheme 1 Synthesis of L2H.

Fig. 1 ORTEP plot of L2H molecule (the same for both polymorphs) showing 50% probability ellipsoids (H atoms are shown as circles) (A) together with packing diagrams for polymorphs α-L2H (B) and β-L2H (C).

However the powder XRD data for the bulk sample indicated the presence of α-L2H only, and β-L2H was not detected giving minor contribution to the reaction product under conditions used.

According to the single-crystal XRD, the geometries of L2H molecules are almost the same in both polymorphs (Fig. 1). Moreover, the corresponding average bond lengths of BTD unit of L2H molecule are similar to that of the parent BTD.³¹ In the ketoimine fragment of L2H molecule, the bond distances C7–O1 and C5–C6 being ca. 1.24 and 1.36 Å, respectively, correspond to the double bonds while the distances C5–N3 and C6–C7 being ca. 1.36 and 1.43 Å correspond to the single bonds. Therefore, the acidic hydrogen atom is bonded to the atom N3 (Fig. 1).

In THF solution, compound L2H was deprotonated by the action of a strong base, KN(SiMe₃)₂ or KO^tBu (but not Et₃N), into an anion L2⁻ (detected by UV-Vis spectroscopy) followed by its precipitation from the solution in the form of red salt KL2. The latter was dissolved in THF by adding equimolar amount of 18-crown-6, and the solution obtained had intense dark-violet color (Fig. S1, ESI†). Keeping the THF solution of [K(18-crown-6)][L2] in the air caused a gradual hydrolysis of the salt visualized by transformation of the dark-violet color of the initial solution into the yellow color of L2H.

Synthesis and characterization of [Zn(L2)₂]

The complex [Zn(L2)₂] (ZnL2) was prepared in THF by reaction of anhydrous ZnCl₂ with two equivalents of KL2 formed in situ by interaction of L2H with KN(SiMe₃)₂. In the solid state the complex ZnL2 is not moisture sensitive. It is also stable in the dry CHCl₃, according to the NMR data (Fig. S2, ESI†). Crystalline ZnL2 (Fig. 2) was obtained by cooling the THF solution. In the crystal, the Zn cation coordinates two anionic ligands L2⁻ with the β-ketoimine substituent acting as a chelating group. The BTD core is not coordinated by the zinc cation. Overall, the latter has a distorted tetrahedral environment. The double bonds of ketoimine unit of ligand L2⁻ are elongated by ∼0.05 Å and the single bonds are shortened by ∼0.03 Å as compared to the neutral L2H. In the crystal, all BTD planes are parallel with an interplanar separation of 3.67 Å within ZnL2 cores (the centroids of the hetero rings are shifted by 0.90 Å) and of 3.45 Å between neighboring ZnL2 cores (the centroids of the hetero rings are shifted by 2.10 Å). These distances imply π–π stacking interactions between BTD units (for π–π interactions, see ref. 32 and 33 and references cited therein) and one may think that the interactions form the crystal packing of the complex under discussion (Fig. 2). Note that DFT optimization of the structure of complex ZnL2 in gas phase and in solution also suggests almost parallel BTD planes (Fig. S8, ESI†).

Fig. 2 ORTEP plot of ZnL2 complex showing 50% probability ellipsoids (H atoms are not shown) (A). Packing diagram of ZnL2 (B).

The difference in the packing motifs of ZnL1 and ZnL2 complexes consists of the presence of the short S⋯N contacts between molecules in ZnL1 and the absence of such contacts in the case of ZnL2. Moreover, the ZnL2 crystal packing is formed by both the inter- and intramolecular π-stacking interactions; only intermolecular π-stacking interactions are revealed in ZnL1.²⁸

UV-Vis and fluorescence spectra and photophysical properties of free ligands L1 and L2H and complexes ZnL1 and ZnL2

Spectral properties of free ligands L1 and L2H and complex ZnL2 were studied in the solid state (polycrystalline samples) and dichloromethane solution, and those of ZnL1 complex only in the solid state due to its dissociation in solution (Fig. S3, ESI†).²⁸ UV-Vis and fluorescence (FL) spectra are presented in Fig. 3 and 4. All obtained spectroscopic and photophysical information is summarized in Table 1 and 2.

Fig. 3 (A) and (C) UV-Vis spectra of free L1 and L2H (black curves) and complexes ZnL1 and ZnL2 (red curves) in CH₂Cl₂ solutions (solid lines) and in polycrystalline samples in the form of Kubelka–Munk functions (dashed lines). Positions and relative oscillator strengths calculated at the TD-M06-HF/TZVP level of theory are depicted as solid bars. (B) and (D): the HOMO and LUMO involved in the lowest-energy transitions of L1 and L2H calculated at the same level. Corresponding lowest energy transitions and MOs involved for ZnL1 and ZnL2 are presented in Table S1 and S2 and Fig. S7 and S8, ESI.†

Fig. 4 FL spectra of L1 (A) and L2H (B) in the CH₂Cl₂ solutions and solid state and those of ZnL1 and ZnL2 in polycrystalline samples.

Table 1 Wavelengths of the UV-Vis (λ^abs_max) and FL (λ^Fl_max) band maxima, the extinction coefficients at maxima (ε_max), the FL lifetime (τ), the FL quantum yield (Φ), the radiative (k_R) and non-radiative (k_NR) rate constants for L1, L2H and ZnL2 in CH₂Cl₂ at the ambient temperature (standard error 10%)

Sample	λ^absmax, nm (εmax × 10^−4, l mol^−1 cm^−1)	λ^Flmax, nm	Φ, %	τ, ns	kR, s^−1	kNR, s^−1
L1	300 (0.67), 305 (0.67), 313 (0.81), 410 (0.24)	575	10	8.7	1.1 × 10^7	10.3 × 10^7
L2H	308 (2.1), 326 (1.6), 412 (1.2)	538	80	16.7	4.8 × 10^7	1.2 × 10^7
ZnL2	308 (4.4), 400 (1.6)	—	—	0.34	—	—

---

Table 2 Wavelengths of the UV-Vis (λ_abs) and FL (λ_em) band maxima, the characteristic time constants of the FL evolution (τ_i), and the FL quantum yield (Φ_Fl) for L1, ZnL1, L2H and ZnL2 in the solid state

Compound	λabs, nm	λem, nm	Φ, %	τ1, ns	τ2, ns	τ3, ns	τ4, ns
L1	308, 410	550	6.5	1.1	4.3	7.5	—
ZnL1	308, 355, 420	440/560	4.4	0.65	1.8	5.6	14.7
L2H	310, 395	593	8.3	0.17	0.66	3.9	7.1
ZnL2	310, 395	530	10.8	0.12	1.2	3.3	10.2

---

1. UV-Vis and FL spectra in solution. In CH₂Cl₂ solution at ambient temperature, the shapes of UV-Vis spectra of L1 and L2H are similar. However, comparing with L1 the spectrum of L2H exhibits a slight red shift, higher extinction coefficients at the maxima of two long-wavelength bands and broad and less resolved band around 300 nm (Fig. 3; S4, ESI†). UV-Vis spectra of L1 and L2H calculated by the TD-DFT approach are in agreement with experimental data (Fig. 3). The long-wavelength transitions in both spectra are mainly due to the electron promotion from the HOMO to the LUMO. In both compounds, the HOMOs are localized on the carbocycle and carbocyclic substituent, while the LUMOs – mainly on the heterocycle. Thus, these transitions are characterized by a charge-transfer (CT) from carbocyclic part of molecules onto heterocyclic one.

The FL spectra of the L1 and L2H solutions in CH₂Cl₂ exhibit the broad unresolved bands with maxima at 575 and 540 nm, respectively (Fig. 4; S4, ESI†). The FL excitation spectra resemble the corresponding UV-Vis spectra (Fig. S4, ESI†) that indicates the presence of only one type of emitting species in solution. Significant Stokes shift (Δν = 7120 and 5680 cm⁻¹ for L1 and L2H, respectively) is in agreement with a large CT between carbocyclic and heterocyclic parts of molecules in the lowest singlet excited state predicted by the calculations (Fig. 3). The nature of carbocyclic substituent strongly influences the FL properties: FL quantum yield (Φ) and FL lifetime (τ) increase from 10% and 8.7 ns for L1 to about 80% and 16.7 ns for L2H in CH₂Cl₂ solution. Almost a five-fold increase of the radiative rate constant (k_R = Φ/τ) on the β-ketoimine substitution (Table 1) is due to a similar increase of the extinction coefficient in the maximum of the long-wavelength band.

Complexation with zinc cation leads to a blue shift of the UV-Vis spectrum of L2H (Fig. 3A). At low concentration in solution, ZnL2 is very sensitive to traces of water provoking its fast hydrolysis (Fig. S5, ESI†).³⁴ Unfortunately, a partial decomposition of ZnL2 with formation of free L2H could not be prevented in CH₂Cl₂ even by the sample preparation under argon. As a result, in the steady state FL measurements in solution, a weak fluorescence from ZnL2 complex is hidden by the strong signal from L2H. Time-resolved data exhibit a presence of component decaying with characteristic time 340 ps on the background of long-living component corresponding to L2H (Fig. S5, ESI†). This short-living component was assigned to the FL decay of ZnL2. Based on the calculated oscillator strength, the k_R for ZnL2 was estimated as ∼2 × 10⁷ s⁻¹ and its FL quantum yield (Φ) in solution at the level of ≈ 2%. A significant decrease in the fluorescence intensity of ZnL2 could be originated from the intramolecular luminescence quenching due to the close proximity of two BTD units in the complex which is absent in free L2H.

2. UV-Vis spectra of polycrystalline samples. The UV-Vis spectra of polycrystalline L1, L2H and ZnL2 are close to those in solution (Fig. 3). For L1, the band maxima and band widths are almost the same in both states, only the characteristic structure of the band at 300 nm in solution is unresolved in the solid state. For L2H and ZnL2, the long wavelength bands around 400 nm are less pronounced in the solid-state spectra as compared with solution (Fig. 3C).

In the case of L1, complexation with zinc cation significantly changes the UV-Vis spectrum leading to a blue shift of the long-wavelength band of ZnL1 as compared to that of L1 (Fig. 3A). The UV-Vis spectra of both L1 and ZnL1 are characterized by the extended tails of the long-wavelength bands manifested itself by low-intensity shoulders at λ > 530 and λ > 440 nm, respectively (Fig. 3A). In order to clarify the origin of these tails, the purity of samples was examined by IR spectroscopy, powder XRD and thermogravimetry (Fig. S6, ESI†). The results obtained did not reveal any impurities. Thus, the presence of tails in the solid state UV-Vis spectra could be an indication of intermolecular interactions, unidentified instrumental problems or crystal defects discussed below. Note that the UV-Vis spectra of solid L2H and ZnL2 are also characterized by some tails comparing to the spectra of these compounds in solution (Fig. 3).

To explain a blue-shift on the complexation, the UV-Vis spectrum of ZnL1 was calculated at the same level of TD-DFT as that of L1 and found to be in a good agreement with the experiment (Fig. 3). In contrast to L1, a series of electronic transitions contribute to the long-wavelength band of ZnL1 with maximum at 355 nm. The main contributions to these transitions arise from six types of electron promotions, namely, from the HOMO, HOMO-1 and HOMO-2 to the LUMO and LUMO+1. The LUMO and LUMO+1 of ZnL1 are very close to the bonding and antibonding combinations of the LUMOs of two ligands L1 (Fig. S7, ESI†). However, the HOMO is localized most exclusively on the Cl⁻ ligands with small admixture of zinc d-AO. The HOMO-1 and HOMO-2 are mixtures of the p-AOs of Cl⁻ with the bonding and antibonding combinations of the HOMOs of two ligands L1. Thus, the band with maximum at 355 nm could be mainly assigned to the combination of intra- (L1) and inter-ligand (Cl⁻ → L1) CT.

In the case of ZnL2, the ligands are formally anions L2⁻. The UV-Vis spectrum of L2⁻ is characterized by intense bands at about 470 and 630 nm (Fig. S1, ESI†). Thus, complexation of L2⁻ also leads to significant blue-shift of its UV-Vis spectrum. Similar to ZnL1, a series of electronic transitions contribute to the long-wavelength band of ZnL2 at about 400 nm. The main contributions to these transitions arise from four types of electron promotions, namely, from the HOMO and HOMO-1 to the LUMO and LUMO+1 (Table S2 and Fig. S8, ESI†).

3. FL dynamics of polycrystalline samples. The steady state FL spectrum of solid L1 represents a broad band with maximum at 550 nm (Φ = 6.5%), which is blue-shifted with respect to the solution spectrum (575 nm, Fig. 4A). Time profiles of the FL decay recorded over the whole emission spectrum after the excitation at 375 nm revealed a deceleration of the decay from blue to red side of the spectrum (Fig. 5B). All kinetic curves are well reproduced in global analysis by a sum of three exponential functions convolved with instrument response function (IRF). The resulting time constants τ_i (i = 1–4) are listed in Table 2 and the best fits are presented in Fig. 5A as smooth lines.

Fig. 5 (A) The normalized time profiles recorded with polycrystalline L1 at different emission wavelengths after the excitation at 375 nm. Inset: The same data at shorter time scale. (B) Intensity-normalized TRES of polycrystalline L1.

Reconstructed time resolved emission spectra (TRES) demonstrate a monotonic red shift of the maximum and a broadening of the FL band during the whole decay of fluorescence (Fig. 5). This indicates a complex dynamics on the nanosecond time scale for polycrystalline sample of L1. It should be noted that a shift of the emission spectrum to low-energy side is a characteristic feature of the relaxation dynamics in the solution.³⁵ This relaxation dynamics reflects environment adjustment to the changes in the molecular geometry and electronic density distribution after the electronic excitation. Most likely, a relaxation dynamics contributes to the observed evolution of non-exponential FL decay of L1 (Fig. 5A). However in solution, significantly more mobile medium, this relaxation is a fast process occurring usually on a picosecond time scale.^35,36

Coordination of L1 with Zn(II) drastically changes not only electronic absorption but also emission properties. The complex ZnL1 exhibits FL spectrum with two broad bands with maxima at ca. 440 and 560 nm (Fig. 4A) and the total FL quantum yield Φ = 4.4%, resulting in bright purple-white emission spanning the entire visible region (Fig. S12, ESI†). The double-band FL spectrum of ZnL1 is independent of the excitation wavelength in the 300–390 nm region (Fig. S10A, ESI†). In turn, the excitation at λ ≥ 430 nm results in a FL spectrum with a single band centered at 560 nm which is slightly red-shifted relative to the FL spectrum of polycrystalline L1 and close to that in the solution (Fig. 4A; S9, ESI†). This indicates the presence of at least two emitting species with different spectral properties. The FL excitation spectra support this hypothesis (Fig. S10B, ESI†). As mentioned above, the available data (Fig. S6, ESI†) did not reveal any traces of impurities in ZnL1. Nevertheless, taking into account all spectroscopic information one can propose that both the pronounced tailing in the UV-Vis spectrum of polycrystalline ZnL1 with a shoulder at about 440 nm (Fig. 3A) and the FL band at 570 nm can be attributed to a trace amount of free ligand L1. The latter could originate from the defects in ZnL1 crystals formed during their formation and growth. Indeed, the ratio of the emission bands at 440 and 570 nm could significantly differ from crystal to crystal (Fig. S10D, ESI†) indicating different contribution from free ligand. The independence of the double-band FL spectrum on the excitation wavelength in the range 300–390 nm is most likely due to the energy transfer (ET) from the excited state of ZnL1 to a free ligand L1 presenting in trace amount.

For ZnL1, time profiles of the FL decay recorded over the whole emission band after the excitation at 375 nm revealed that the early dynamics includes fast signal decay at the blue side and a corresponding rise at the red side of the emission spectrum (Fig. 6A). Later signal evolution demonstrates independent decay of the blue and red bands of the FL spectrum with different rates. The obtained data are well reproduced by a sum of four exponential functions (global fit). The resulting time constants τ_i (i = 1–4) are listed in Table 2 and best fits are presented in Fig. 6A as smooth lines. The reconstructed emission spectra (Fig. 6B) show the formation of intense band at 440 nm immediately after the laser pulse. At the early stages of FL dynamics the band at 570 nm is hidden by intense band at 440 nm but it becomes clearly visible at later stages. TRES after normalization (Fig. 6C) additionally illustrates the observed dynamics.

Fig. 6 (A) Time profiles of FL decay recorded at different wavelengths after the excitation of polycrystalline ZnL1 at 375 nm. Smooth lines present the best four exponential global fit. Inset: the early dynamics at the same wavelengths. (B) TRES obtained from the global fit of fluorescence time profiles. (C) TRES in B after normalization. (D) DAS from the same fit.

Decay Associated Spectra (DAS) of amplitudes corresponding to four different time constants A_i(λ) (i = 1–4) are displayed in Fig. 6D; positive values correspond to a decay and negative to a rise of the signal intensity. The fastest component, A₁(λ), exhibits both positive and negative bands indicating a decay on the high-energy side and a rise on the low-energy side of the emission band. This component could be attributed to a relaxation dynamics in polycrystalline sample or to an ET from the excited molecules of ZnL1 to those of free ligand L1 (through preceding migration of the energy over ZnL1). All other components demonstrate exclusively positive values of A_i(λ) (i = 2–4) that indicates the decay dynamics. The DAS A₂(λ) mainly consists of the emission band with maximum at 440 nm which should be attributed to ZnL1. The DAS A₃(λ) and A₄(λ) almost coincide with the steady-state emission band at 560 nm (Fig. 4A) and are similar to the DAS A₂(λ) and A₃(λ) detected for the polycrystalline L1 (Fig. S9, ESI†). Note that the three-exponential decay of the FL band of ZnL1 at 560 nm has the time constants (τ₂–τ₄) similar to those observed for the FL decay of the polycrystalline L1 (τ₁–τ₃, Table 2). This is also in agreement with the assumption that the 560 nm band is due to the fluorescence of free ligand.

The FL spectrum of polycrystalline L2H is red-shifted by 60 nm as compared to that in CH₂Cl₂ (Fig. 4B). This could be due to the intermolecular interactions including π⋯π interactions in the former case (Fig. 1) which are known to play an essential role in decreasing the HOMO–LUMO gaps.^37,38L2H and ZnL2 exhibit similar FL quantum yields in the solid state i.e. Φ_Fl is 8.3% for L2H and 10.8% for ZnL2.

Time resolved FL dynamics for both L2H and ZnL2 exhibits a fast decay on the blue side and the corresponding rise on the red side of the emission bands (Fig. S11, ESI†). Subsequent FL decay can be well described for both compounds by the sum of four exponential functions; the resulting time constants τ_i (i = 1–4) from global fit are listed in Table 2. The TRES analysis shows a monotonic red shift combined with a broadening of the emission band (Fig. S11, ESI†) similarly to that observed for L1 (Fig. 5B; S9, ESI†). ZnL2 demonstrates the largest red shift and broadening dynamics as compared with ligands.

The fastest components observed for ZnL1, L2H and ZnL2 possess the τ₁ values lying in the range 0.1–0.7 ns (Table 2) and exhibit similar shapes of A₁(λ) (Fig. 6D and S9 of ESI†). The absence of second emission band in the case of both L2H and ZnL2 indicates this component as a relaxation dynamics in polycrystalline sample after the optical excitation. Basing on the similarity of τ₁ values and A₁(λ) shape, the same assignment should be proposed for ZnL1. Most probably ET, observed by wavelength-independence of the double emission spectrum of ZnL1 (Fig. S10†), occurs at the time scale <50 ps and could not be resolved by the used equipment.

Conclusions

The synthesis and XRD characterization of the first derivative of 2,1,3-benzothiadizole (BTD) bearing chelating β-ketoimine substituent in the position 4, i.e. L2H, and its zinc complex ZnL2 were performed. With XRD, two polymorphs of L2H were observed. The UV-Vis and FL spectroscopy, including time-resolved fluorescence, have been applied to both the dissolved in organic solvent and polycrystalline L2H and previously synthesized 4-aminosubstituted BTD (L1), as well as to their zinc complexes ZnL2 and ZnL1. Results of spectroscopic studies supported by TD-DFT calculations led to a deeper understanding of the electronic properties of new compounds. Comparison of spectral and photophysical properties of L2H and ZnL2 with those obtained in this work for related ligand L1 and complex ZnL1 described previously revealed interesting both similarities and differences. It should be noted that ZnL1 and ZnL2 are characterized by fundamentally different thermodynamics of formation and bonding situation. The ZnL1 is very weakly-bonded complex of ZnCl₂ core with neutral amino-substituted BTD, which dissociates easily in solution in agreement with TD-DFT calculations. In the ZnL2, two L2⁻ anions act as bidentate chelating ligands leading to a significantly more stable complex.

All compounds possess bright fluorescence in both the solid state and solution, except for ZnL1 undergoing dissociation in solution. In the polycrystalline state, a FL quantum yield of ZnL2 (7.8%) is similar to that of L2H (10.8%). In contrast to ZnL2 having conventional single-band FL spectrum, ZnL1 demonstrates a broad double-band fluorescence spanning the entire visible region which results in the purple-white emission. On the basis of careful spectroscopic studies the long-wavelength FL band was assigned to the minor traces of free ligand L1 in the crystalline ZnL1. Moreover, based on the independence of the double-band FL spectrum on the excitation wavelength we proposed the ET from the excited state of ZnL1 to a free ligand L1 presenting in trace amount. The reason for this is most likely the easiness of defect (i.e. free ligand) formation in crystalline ZnL1 due to a very weak bonding in this complex. In this context, one can think that working with weakly-bonded complexes or/and doping metal complexes by traces of their free ligands might create under conditions of ET transfer from complexes to ligands additional possibilities in the design and synthesis of new luminescent materials, having broad “white light” fluorescence. Overall, functionalized BTD derivatives L1 and L2 and, especially, their complexes ZnL1 and ZnL2 can be considered as promising candidate building blocks in the design and synthesis of new OLEDs.

Experimental and computational details

General

Solvents and starting materials were used as received from suppliers. Synthesis of ZnL2 was performed in vacuum using Teflon Schlenk technique and its isolation from the reaction mixture was carried out under argon. THF used for the synthesis of ZnL2 was dried by Na–K alloy. Ligand L1³⁹ and complex ZnL1²⁸ were synthesized by the known methods.

Elemental analysis was performed with a Eurovector EuroEA3000 analyzer. ¹H NMR spectra (500.13 MHz) were taken with a Bruker DRX-500 spectrometer in CDCl₃, the solvent signal (δ¹H = 7.24) was used as internal standard. IR spectra were recorded in KBr pellets with a FT-801 spectrometer.

X-ray structure determination

All single-crystal XRD data for α-L2H, β-L2H and ZnL2 (Table S3, ESI†) were collected using the graphite-monochromated MoKα-radiation (λ = 0.71073 Å) on a Bruker Nonius X8Apex diffractometer equipped with a 4k CCD area detector at 293 K (α-L2H) and 150 K (β-L2H and ZnL2). The ϕ- and ω-scan techniques were employed to measure intensities. Absorption corrections were applied using the SADABS program.⁴⁰ The crystal structures were solved by direct methods, and refined by the full-matrix least squares techniques with the use of the SHELXTL package.⁴¹ Atomic thermal parameters for non-hydrogen atoms were refined anisotropically. The positions of hydrogen atoms were calculated corresponding to their geometrical conditions and refined using the riding model.

Spectral and photophysical properties

UV-Vis spectra of solutions were measured with an Ultrospec3300 pro UV/Visible spectrophotometer, and solid-state diffuse reflectance spectra with a Shimadzu UV-3101 spectrophotometer. Samples for the diffuse reflectance measurements were prepared by a thorough grinding of a mixture of the compounds under study (ca. 0.02 mole fraction) with BaSO₄, which was used also as a standard. Spectral dependence of the diffuse reflectance was converted into a spectrum of a Kubelka–Munk function.⁴²

FL excitation and emission spectra of samples in CH₂Cl₂ solution and in the solid state were recorded with FLSP920 spectrofluorometer (Edinburgh Instruments, UK). The absolute quantum yields of the compounds in the solid state were measured with integrating sphere combined with the Fluorolog-3 FL3-22 spectrophotometer (Horiba Jobin Yvon), for all compounds the excitation wavelength was 375 nm. All FL spectra were corrected for the wavelength-dependent sensitivity of the detection. Liquid samples with the absorption below 0.1 at the absorption maximum were placed in a 10 × 10 mm² quartz cell and the solid samples were held in a 0.5 × 10 mm² quartz cell. The FL quantum yields of samples in solutions were determined relative to the perylene in ethanol, Φ_F = 92%.⁴³ The FL dynamics was recorded using a time-correlated single-photon counting (TCSPC) unit of a FLSP920 spectrofluorometer. A pulsed diode laser (EPL-375, λ = 375 nm, FWHM ≈ 80 ps) was used for the irradiation of both liquid and solid samples. The time profiles were recorded at the maxima of FL bands for the liquid samples and over the whole FL band for the polycrystalline samples. The obtained emission time profiles were globally analyzed using the numerical convolution of instrument response function (IRF) with the sum of exponential functions. To build time-resolved emission spectra (TRES) the best fits were scaled on the factors

where S(λ) is the steady-state FL intensity and a_i(λ) is an amplitude corresponding to an exponential function with a characteristic time constant τ_i. The TRES represent the intensities of F(λ)-corrected best fits taken at different time delays after the excitation pulse.

Quantum chemical calculations

Positions and oscillator strengths of the electronic transitions in UV-Vis spectra of L1, L2H, ZnL1 and ZnL2 were calculated using TD-DFT⁴⁴ at the M06-HF/def2-TZVP level of theory^45–47 using corresponding XRD geometries. This method is known to correctly reproduce positions of the CT transitions.^28,47 In the case of L2⁻, calculations of the electronic transitions were performed at both the XRD geometry of anion in ZnL2 and geometry optimized at the B97-D3 level^48,49 with def2-TZVP basis set.⁴⁷ The Becke-Johnson damping function was used in all dispersion-corrected calculations.⁵⁰ To estimate thermodynamics of the complex formation between ZnCl₂ and L1 or L2H, the following calculations were performed: thermochemical corrections were obtained from the gas phase optimization/frequencies calculations at the B97-D3/def2-TZVP level of theory. The Grimme geometrical counterpoise (gCP) correction scheme was used to semi-empirical treatment of the BSSE effects.⁵¹ Then the single point calculations were performed at the same level for the THF solution using COSMO solvation model,⁵² and energies after outlying charge correction were used to calculate enthalpies and Gibbs free energies. For TD–DFT calculations, the GAUSSIAN09 suit of programs⁵³ has been employed. All other quantum chemical calculations were performed using ORCA suit of programs (version 3.0.0).⁵⁴ The resolution-of-the identity approximation (RI) was used to speed up the GGA computations.⁵⁵

Syntheses

4-(2,1,3-Benzothiadiazole-4-ylamino)pent-3-en-2-one L2H. At ambient temperature, acetylacetone (1.35 mL) was added to a mixture of L1 (1.00 g, 3.32 mmol) and toluene (10 mL) and the solution obtained was refluxed for 2 days using Dean–Stark apparatus. Then the solution was concentrated to 3 mL volume and n-hexane (10 mL) was added. The resulting orange crystalline product was filtered off, washed with several portions of n-hexane and dried in vacuo. Yield 1.40 g (90%). M.p. = 122–124 °C (α-L2H). C₁₁H₁₁N₃SO (233.3): calcd C 56.6, H 4.8, N 18.0, S 13.7; found C 56.6, H 5.0, N 17.4, S 13.0. ¹H NMR, δ (CDCl₃): 7.68 (d, 1H, Ar), 7.51 (t, 1H, Ar), 7.22 (d, 1H, Ar), 5.36 (s, 1H, C(O)CHC(N)), 2.27 (d, 3H, CH₃), 2.16 (d, 3H, CH₃).

Slow evaporation of the mother liquor affords crystallization of additional portion of L2H as two polymorphs: α-L2H (major) and β-L2H (minor, few crystals, m.p. = 111–113 °C), the polymorphs were separated manually.

Bis(4-(2,1,3-benzothiadiazole-4-ylamino)pent-3-en-2-onate)zinc ZnL2. At ambient temperature, L2H (0.056 g, 0.24 mmol) and KN(SiMe₃)₂ (0.045 g, 0.23 mmol) were dissolved in THF (5 mL) with stirring. Red precipitate of KL2 was gradually formed. After 0.5 h, the solution of ZnCl₂ (0.016 g, 0.12 mmol) in THF (5 mL) was added which led to dissolution of KL2 and formation of white precipitate of KCl. The reaction mixture was stirred overnight and filtered. Evaporation of the mother liquor to 4 mL followed by cooling to 2 °C gave crystalline ZnL2. Yield 0.032 g (50%). C₂₂H₂₀N₆O₂S₂Zn (530.0): calcd C 49.8, H 3.8, N 15.8, S 12.1; found C 50.0, H 3.8, N 15.8, S 12.0. ¹H NMR, δ (CDCl₃): 7.48 (d, 1H, Ar), 7.28 (t, 1H, Ar), 6.80 (d, 1H, Ar), 5.08 (s, 1H, C(O)CHC(N)), 2.09 (d, 3H, CH₃), 1.60 (d, 3H, CH₃).

Acknowledgements

The authors are grateful to the Russian Foundation for Basic Research (RFBR), the Presidium of the Russian Academy of Sciences (program no. 35) and the Ministry of Education and Science of the Russian Federation (project of Joint Laboratories of Siberian Branch of the Russian Academy of Sciences and National Research Universities) for financial support. Syntheses of the compounds were supported by RFBR projects 16-33-00305 and 16-03-00637. Time-resolved FL measurements were performed with the financial support from the RFBR project 14-14-00056. Computational part was supported by the RFBR project 15-03-03242.

Notes and references

1. A. V. Lonchakov, O. A. Rakitin, N. P. Gritsan and A. V. Zibarev, Molecules, 2013, 18, 9850–9900.
2. C. B. Nielsen, A. J. P. White and I. McCulloch, J. Org. Chem., 2015, 80, 5045–5048.
3. X. Zhang, H. Bronstein, A. J. Kronemeijer, J. Smith, Y. Kim, R. J. Kline, L. J. Richter, T. D. Anthopoulos, H. Sirringhaus, K. Song, M. Heeney, W. Zhang, I. McCulloch and D. M. DeLongchamp, Nat. Commun., 2013, 4, 2238.
4. H. N. Tsao, D. M. Cho, I. Park, M. R. Hansen, A. Mavrinskiy, D. Y. Yoon, R. Graf, W. Pisula, H. W. Spiess and K. Müllen, J. Am. Chem. Soc., 2011, 133, 2605–2612.
5. C. B. Nielsen, B. C. Schroeder, A. Hadipour, B. P. Rand, S. E. Watkins and I. McCulloch, J. Mater. Chem., 2011, 21, 17642–17645.
6. F. S. Mancilha, L. Barloy, F. S. Rodembusch, J. Dupont and M. Pfeffer, Dalton Trans., 2011, 40, 10535–10544.
7. P.-L. T. Boudreault, A. Najari and M. Leclerc, Chem. Mater., 2010, 23, 456–469.
8. T. Linder, E. Badiola, T. Baumgartner and T. C. Sutherland, Org. Lett., 2010, 12, 4520–4523.
9. K. R. Justin Thomas, J. T. Lin, M. Velusamy, Y. T. Tao and C. H. Chuen, Adv. Funct. Mater., 2004, 14, 83–90.
10. X. Zhang, H. Gorohmaru, M. Kadowaki, T. Kobayashi, T. Ishi-i, T. Thiemann and S. Mataka, J. Mater. Chem., 2004, 14, 1901–1904.
11. Q. Hou, Q. Zhou, Y. Zhang, W. Yang, R. Yang and Y. Cao, Macromolecules, 2004, 37, 6299–6305.
12. S. Ellinger, K. R. Graham, P. Shi, R. T. Farley, T. T. Steckler, R. N. Brookins, P. Taranekar, J. Mei, L. A. Padilha, T. R. Ensley, H. Hu, S. Webster, D. J. Hagan, E. W. Van Stryland, K. S. Schanze and J. R. Reynolds, Chem. Mater., 2011, 23, 3805–3817.
13. J. Liu, Q. G. Zhou, Y. X. Cheng, Y. H. Geng, L. X. Wang, D. G. Ma, X. B. Jing and F. S. Wang, Adv. Funct. Mater., 2006, 16, 957–965.
14. B. A. D. Neto, A. A. M. Lapis, E. N. da Silva Júnior and J. Dupont, Eur. J. Org. Chem., 2013, 228–255.
15. S. Uchiyama, T. Tsuji, K. Ikado, A. Yoshida, K. Kawamoto, T. Hayashi and N. Inada, Analyst, 2015, 140, 4498–4506.
16. S. Uchiyama, K. Kimura, C. Gota, K. Okabe, K. Kawamoto, N. Inada, T. Yoshihara and S. Tobita, Chem.–Eur. J., 2012, 18, 9552–9563.
17. K. Okabe, N. Inada, C. Gota, Y. Harada, T. Funatsu and S. Uchiyama, Nat. Commun., 2012, 3, 705.
18. B. A. D. Neto, P. H. P. R. Carvalho and J. R. Correa, Acc. Chem. Res., 2015, 48, 1560–1569.
19. E. Guzman-Percastegui, M. Vonlanthen, B. Quiroz-Garcia, M. Flores-Alamo, E. Rivera and I. Castillo, Dalton Trans., 2015, 44, 15966–15975.
20. Z.-P. Zheng, Q. Wei, W.-X. Yin, L.-T. Wan, X. Huang, Y. Yu and Y.-P. Cai, RSC Adv., 2015, 5, 27682–27689.
21. H. Mandal, S. Chakrabartty and D. Ray, RSC Adv., 2014, 4, 65044–65055.
22. J. Svec, P. Zimcik, L. Novakova, O. A. Rakitin, S. A. Amelichev, P. A. Stuzhin and V. Novakova, Eur. J. Org. Chem., 2015, 596–604.
23. P. A. Stuzhin, M. S. Mikhailov, E. S. Yurina, M. I. Bazanov, O. I. Koifman, G. L. Pakhomov, V. V. Travkin and A. A. Sinelshchikova, Chem. Commun., 2012, 48, 10135–10137.
24. N. Wei, Y.-R. Zhang and Z.-B. Han, CrystEngComm, 2013, 15, 8883–8886.
25. T. He, D. Rajwar, L. Ma, Y. Wang, Z. Bang Lim, A. C. Grimsdale and H. Sun, Appl. Phys. Lett., 2012, 101, 213302.
26. Z. B. Lim, H. Li, S. Sun, J. Y. Lek, A. Trewin, Y. M. Lam and A. C. Grimsdale, J. Mater. Chem., 2012, 22, 6218–6231.
27. A. Wild, A. Teichler, C.-L. Ho, X.-Z. Wang, H. Zhan, F. Schlutter, A. Winter, M. D. Hager, W.-Y. Wong and U. S. Schubert, J. Mater. Chem. C, 2013, 1, 1812–1822.
28. D. A. Bashirov, T. S. Sukhikh, N. V. Kuratieva, E. A. Chulanova, I. V. Yushina, N. P. Gritsan, S. N. Konchenko and A. V. Zibarev, RSC Adv., 2014, 4, 28309–28316.
29. T. S. Sukhikh, D. A. Bashirov, N. V. Kuratieva, A. I. Smolentsev, A. S. Bogomyakov, V. A. Burilov, A. R. Mustafina, A. V. Zibarev and S. N. Konchenko, Dalton Trans., 2015, 44, 5727–5734.
30. D. A. Bashirov, T. S. Sukhikh, N. V. Kuratieva, D. Y. Naumov, S. N. Konchenko, N. A. Semenov and A. V. Zibarev, Polyhedron, 2012, 42, 168–174.
31. T. Suzuki, T. Tsuji, T. Okubo, A. Okada, Y. Obana, T. Fukushima, T. Miyashi and Y. Yamashita, J. Org. Chem., 2001, 66, 8954–8960.
32. W. Wang, Y. Zhao, Y. Zhang and Y.-b. Wang, Comput. Theor. Chem., 2014, 1030, 1–8.
33. C. Janiak, J. Chem. Soc., Dalton Trans., 2000, 3885–3896.
34. An addition of a drop of water (ca. 10 μL) to the sample of ZnL2 (3 mL, 10⁻⁵ M) results in the change of solution UV-Vis spectrum of the complex to that of free L2H.
35. M. L. Horng, J. A. Gardecki, A. Papazyan and M. Maroncelli, J. Phys. Chem., 1995, 99, 17311–17337.
36. O. Braem, T. J. Penfold, A. Cannizzo and M. Chergui, Phys. Chem. Chem. Phys., 2012, 14, 3513–3519.
37. A. Kobayashi, H. Tanaka and H. Kobayashi, J. Mater. Chem., 2001, 11, 2078–2088.
38. S.-L. Zheng, J.-P. Zhang, W.-T. Wong and X.-M. Chen, J. Am. Chem. Soc., 2003, 125, 6882–6883.
39. Y. Liu, Y. Lu, M. Prashad, O. Repič and T. J. Blacklock, Adv. Synth. Catal., 2005, 347, 217–219.
40. APEX2 (Version 2.0), SAINT (Version 8.18c), and SADABS (Version 2.11), Bruker Advanced X-ray Solutions, Bruker AXS Inc., Madison, Wisconsin, USA, 2000–2012.
41. G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008, 64, 112–122.
42. P. Kubelka, J. Opt. Soc. Am., 1948, 38, 448–457.
43. G. A. Crosby and J. N. Demas, J. Phys. Chem., 1971, 75, 991–1024.
44. A. Dreuw and M. Head-Gordon, Chem. Rev., 2005, 105, 4009–4037.
45. Y. Zhao and D. G. Truhlar, J. Phys. Chem. A, 2006, 110, 13126–13130.
46. Y. Zhao and D. G. Truhlar, Acc. Chem. Res., 2008, 41, 157–167.
47. F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005, 7, 3297–3305.
48. S. Grimme, J. Comput. Chem., 2006, 27, 1787–1799.
49. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 15404.
50. S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem., 2011, 32, 1456–1465.
51. H. Kruse and S. Grimme, J. Chem. Phys., 2012, 136, 154101.
52. S. Sinnecker, A. Rajendran, A. Klamt, M. Diedenhofen and F. Neese, J. Phys. Chem. A, 2006, 110, 2235–2245.
53. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Revision A.01, Gaussian, Inc., Wallingford, CT, USA, 2009.
54. F. Neese, WIREs Computational Molecular Science, 2012, 2, 73–78.
55. F. Neese, J. Comput. Chem., 2003, 24, 1740–1747.

---

Footnote

† Electronic supplementary information (ESI) available: Crystallographic data and additional figures. CCDC 1457754 (α-L2H), 1457755 (β-L2H), 1457756 (ZnL2). For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6ra06547c

---