Structure and Conformational Mobility of OLED-Relevant 1,3,5-Triazine Derivatives

A series of OLED-relevant compounds, consisting of 1,3,5-triazine core linked to various aromatic arms by amino group, has been synthesized and characterized. The studied compounds exist in solution as a mixture of two conformers, a symmetric propeller and asymmetric conformer, in which one of the aromatic arms is rotated around the C-N bond. At temperatures below −40 °C, the VT NMR spectra in DMF-d7 are in a slow exchange regime, and the signals of two conformers can be elucidated. At temperatures above 100 °C, the VT NMR spectra in DMSO-d6 are in a fast exchange regime, and the averaged spectra can be measured. The ratio of symmetric and asymmetric conformers in DMF-d7 varies from 14:86 to 50:50 depending on the substituents. The rotational barriers of symmetric and asymmetric conformers in DMF-d7 were measured for all compounds and are in the interval from 11.7 to 14.7 kcal/mol. The ground-state energy landscapes of the studied compounds, obtained by DFT calculations, show good agreement with the experimental rotational barriers. The DFT calculations reveal that the observed chemical exchange occurs by the rotation around the C(1,3,5-triazine)-N bond. Although some of the compounds are potentially tautomeric, the measured absorption and emission spectra do not indicate proton transfer neither in the ground nor in the excited state.


Introduction
The design of new OLEDs is strongly dependent on the appropriate HOMO and LUMO energy levels. In this regard, a wide spectrum of different classes of organic compounds (exclusively aromatic) is tested experimentally. Although most of the OLEDs used in practice are polymers [1], significant efforts are also being made to develop monomeric analogues. A recent trend in their design is to replace the linear structure with a starshaped one [2,3]. A significant change in the properties of the material can be achieved by varying the "arms" and the core of the star-shaped architecture [4,5]. Until now, fluorene derivatives have mainly been reported as compounds, where the star-shaped structures are suitable for such applications [6,7]. Other studied star-shaped structures of particular interest are 1,3,5-triazines, as they possess desired optoelectronic and photovoltaic properties [8][9][10]. Moreover, a combination of 1,3,5-triazines and carbazole moieties in single molecules significantly improves their performance [11][12][13][14][15]. Different aspects of carbazole properties and their use in light-emitting applications have recently been reviewed [16]. In the study of Zassowski et al. [17], it was shown that star-shaped compounds of this type demonstrate even better optoelectronic properties than their linear analogues. Numerous studies have focused on devices based on polymers with an incorporated carbazole building block [18][19][20][21]. Carbazole derivatives are an effective host material for phosphorescent additives in PhOLEDs [15,[22][23][24][25][26][27][28], an active component of fluorescent OLEDs with thermally activated delayed fluorescence (TADF) [14,[29][30][31][32][33][34][35][36][37][38][39][40][41][42][43] or molecules with a hybridized

Potential Tautomerism
Some of the studied 1,3,5-triazine compounds (4, 11, 12 and 14) may exist in amino and imino tautomeric forms (Scheme 3), each of them exhibiting different dynamic isomerization. The first one is proton transfer by itself. The second one is E/Z isomerization between the imino tautomers, stabilized by conjugation with the aromatic substituents. The third one is conformational isomerism of amino tautomers due to restricted rotation about the amino triazine single bond. The overall situation supposes that up to nineteen possible isomers could be expected as a combination of all possible processes. In compounds 5, 6 and 13, the amino-imino tautomerism is not possible, while in 11, 12 and 14, an enol-keto proton transfer from the OH group in the phenyl ring to the triazine nitrogen cannot be additionally excluded.

Potential Tautomerism
Some of the studied 1,3,5-triazine compounds (4, 11, 12 and 14) may exist in amino and imino tautomeric forms (Scheme 3), each of them exhibiting different dynamic isomerization. The first one is proton transfer by itself. The second one is E/Z isomerization between the imino tautomers, stabilized by conjugation with the aromatic substituents. The third one is conformational isomerism of amino tautomers due to restricted rotation about the amino triazine single bond. The overall situation supposes that up to nineteen possible isomers could be expected as a combination of all possible processes. In compounds 5, 6 and 13, the amino-imino tautomerism is not possible, while in 11, 12 and 14, an enol-keto proton transfer from the OH group in the phenyl ring to the triazine nitrogen cannot be additionally excluded. Simple considerations based on the aromaticity of the triamino derivatives lost in the imino tautomers [53] and molecular mechanics calculations did not conclusively confirm the preference of the amino or imino form. Definitive proof was given by the dynamic NMR investigations and X-ray structural determination of Scheme 3. Border cases of the possible tautomers and corresponding isomers. Simple considerations based on the aromaticity of the triamino derivatives lost in the imino tautomers [53] and molecular mechanics calculations did not conclusively confirm the preference of the amino or imino form. Definitive proof was given by the dynamic NMR investigations and X-ray structural determination of 2-chloro-4,6-bis(pyrazolylamino)-1,3,5triazine [54], which, due to its low symmetry, exists in three different conformations of the amino tautomer.
In the current case, an a priori exclusion of some tautomers is not evident, but some logical consideration can help to distinguish the processes. Bearing in mind that both (amino and imino) tautomers have different conjugation systems and different polarity, they can easily be detected by their UV-Vis absorption spectra, which should be solventdependent [55]. The amino-imino proton exchange proceeds through a four-membered ring, which leads, without solvent assistance, to an increase of the transition state energy [56] and makes the process possible to be detected by NMR spectroscopy. Recently, the transition state through the four-membered ring in 1,3,5-triazine derivatives has been estimated by DFT calculation to be about 50 kcal/mol [57]. In addition, compounds 5 and 6, where tautomerism is not possible, can be used as model compounds of 4, while 13 is a reference compound for the series 11-14. The ground-state enol-keto proton transfer in the series 11-14 is also not likely due to a variety of reasons: the formed tautomer, originating from migration of the OH proton to the triazine nitrogen atom, is zwitterionic in nature; it is not stabilized by re-arrangement of the electronic density and proceeds through hydrogen bonding in the seven-membered ring, characterized by substantial nonplanarity. If observed in the ground state, it should be characterized by a red shift in the absorption spectra in polar solvents [58], while the relatively low barrier of proton exchange makes it not detectable by the NMR. The process is much more likely in the excited state and, if it occurs, could feature large Stokes shifts in the measured emission [59].
The absorption and emission spectra of the studied compounds are shown in Figure 1. The spectral data in acetonitrile are compared in Table 1 with the TD-DFT predicted absorption of the amine tautomers. Several conclusions can be made considering them. Considering series 4-6, the shapes of all three compounds are very similar and there are no solvent-dependent spectral changes going from dichloromethane to acetonitrile. This excludes the existence of tautomeric equilibrium at room temperature. The predicted absorbance of the amino tautomers very nicely corresponds, both in terms of position and intensity, to the experimentally measured ones. The observed weak emission (the maximal quantum yield is 5% in 6), attributed to the existence of amine nitrogen atoms, is also similar in shape. The values of the quantum yields are similar to those measured and reported in [17] for similar compounds. The largest Stokes shift is below 2500 cm −1 , which excludes excited state proton transfer. absorbance of the amino tautomers very nicely corresponds, both in terms of position and intensity, to the experimentally measured ones. The observed weak emission (the maximal quantum yield is 5% in 6), attributed to the existence of amine nitrogen atoms, is also similar in shape. The values of the quantum yields are similar to those measured and reported in [17] for similar compounds. The largest Stokes shift is below 2500 cm −1 , which excludes excited state proton transfer.  In the case of 11-14, as expected, the spectra of the studied compounds are similar in shape and, again, not solvent-dependent. This fact, along with the absence of emissions, clearly indicates that no proton transfer from the OH to the triazole nitrogen occurs. The steric hindrance in 13 and 14, hindering the O-H…N hydrogen bonding and the overall conjugation, leads to a slight blue shift in the long-wavelength absorption maxima. Our DFT calculations predict that the symmetric imino conformer is less stable than the symmetric amino conformer (Scheme 3) by ca. 20 kcal/mol. In the case of 11-14, as expected, the spectra of the studied compounds are similar in shape and, again, not solvent-dependent. This fact, along with the absence of emissions, clearly indicates that no proton transfer from the OH to the triazole nitrogen occurs. The steric hindrance in 13 and 14, hindering the O-H . . . N hydrogen bonding and the overall conjugation, leads to a slight blue shift in the long-wavelength absorption maxima. Our DFT calculations predict that the symmetric imino conformer is less stable than the symmetric amino conformer (Scheme 3) by ca. 20 kcal/mol.

Dynamic NMR Spectroscopy, Conformational Analysis and Molecular Geometry
The structures of the newly synthesized compounds were confirmed by 1D and 2D NMR spectra. Conformational exchanges occurred in solution at rates that were intermediate on the NMR time scale. The possible conformers (propeller and asymmetric) of the studied compounds are presented in Figure 2.

Dynamic NMR Spectroscopy, Conformational Analysis and Molecular Geometry
The structures of the newly synthesized compounds were confirmed by 1D and 2D NMR spectra. Conformational exchanges occurred in solution at rates that were intermediate on the NMR time scale. The possible conformers (propeller and asymmetric) of the studied compounds are presented in Figure 2.  Table 2.
The population of symmetric and asymmetric conformers of the studied compounds in DMF-d7 vary from 14%/86% in compound 6 to 50%/50% in compound 4 ( Table 2). These ratios are measured from 1 H NMR spectra at the lowest measured temperature (223 or 233 K).   Table 2.   Table 2.
The population of symmetric and asymmetric conformers of the studied compounds in DMF-d7 vary from 14%/86% in compound 6 to 50%/50% in compound 4 ( Table 2). These ratios are measured from 1 H NMR spectra at the lowest measured temperature (223 or 233 K).   Table 2.
The population of symmetric and asymmetric conformers of the studied compounds in DMF-d7 vary from 14%/86% in compound 6 to 50%/50% in compound 4 ( Table 2). These ratios are measured from 1 H NMR spectra at the lowest measured temperature (223 or 233 K).   Table 2.
The population of symmetric and asymmetric conformers of the studied compounds in DMF-d7 vary from 14%/86% in compound 6 to 50%/50% in compound 4 ( Table 2). These ratios are measured from 1 H NMR spectra at the lowest measured temperature (223 or 233 K). The population of symmetric and asymmetric conformers of the studied compounds in DMF-d7 vary from 14%/86% in compound 6 to 50%/50% in compound 4 ( Table 2). These ratios are measured from 1 H NMR spectra at the lowest measured temperature (223 or 233 K). The measured rotational barriers between the conformers, presented in Table 3 together with the theoretically predicted values, show that there is a good correspondence between the theoretical and experimental data. The differences almost fall within the expected deviation of theoretical values at this level of theory. The measured rotational barriers between the conformers, presented in Table 3 together with the theoretically predicted values, show that there is a good correspondence between the theoretical and experimental data. The differences almost fall within the expected deviation of theoretical values at this level of theory. The measured rotational barriers between the conformers, presented in Table 3 together with the theoretically predicted values, show that there is a good correspondence between the theoretical and experimental data. The differences almost fall within the expected deviation of theoretical values at this level of theory. The population of symmetric and asymmetric conformers of the studied compounds in DMF-d7 vary from 14%/86% in compound 6 to 50%/50% in compound 4 ( Table 2). These ratios are measured from 1 H NMR spectra at the lowest measured temperature (223 or 233 K).
The measured rotational barriers between the conformers, presented in Table 3 together with the theoretically predicted values, show that there is a good correspondence between the theoretical and experimental data. The differences almost fall within the expected deviation of theoretical values at this level of theory. The measured rotational barriers between the conformers, pres gether with the theoretically predicted values, show that there is a go between the theoretical and experimental data. The differences alm expected deviation of theoretical values at this level of theory.  The measured rotational barriers between the conformers, presente gether with the theoretically predicted values, show that there is a good between the theoretical and experimental data. The differences almost expected deviation of theoretical values at this level of theory. The measured rotational barriers between the conformers, pres gether with the theoretically predicted values, show that there is a go between the theoretical and experimental data. The differences alm expected deviation of theoretical values at this level of theory. The measured rotational barriers between the conformers, pres gether with the theoretically predicted values, show that there is a go between the theoretical and experimental data. The differences alm expected deviation of theoretical values at this level of theory.

NMR Structure Determination
The 1 H NMR spectra of compound 11 at 293 K showed broad NH, H-3 and H-4 protons ( Figure S10). The NMR spectra were DMSO-d6 and DMF-d7 solutions at different temperature intervals. A the signals were split, and four signals were observed for every pr high intensity and three similar signals of medium intensity were ob could be assigned to two isomeric compounds, one symmetrical asymmetrical conformer ( Figure S10). A similar behavior was observ spectra. The NMR spectra of compound 11 at 293 K in DMSO and DM broad signals that coalesce to give signals for a single compound Similarly, for compounds 4-6 and 12-14 at 223 K (DMF), two isomeri be detected: one symmetric and one asymmetrical conformer. Upon to 343-403 K, the signals coalesced to show the presence of one single ( 1 H and 13 C spectra in the supporting information). 12

NMR Structure Determination
The 1 H NMR spectra of compound 11 at 293 K showed broad NH, H-3 and H-4 protons ( Figure S10). The NMR spectra were r DMSO-d6 and DMF-d7 solutions at different temperature intervals. A the signals were split, and four signals were observed for every pr high intensity and three similar signals of medium intensity were obs could be assigned to two isomeric compounds, one symmetrical c asymmetrical conformer ( Figure S10). A similar behavior was observ spectra. The NMR spectra of compound 11 at 293 K in DMSO and DM broad signals that coalesce to give signals for a single compound a Similarly, for compounds 4-6 and 12-14 at 223 K (DMF), two isomeri be detected: one symmetric and one asymmetrical conformer. Upon h to 343-403 K, the signals coalesced to show the presence of one single ( 1 H and 13 C spectra in the supporting information).

NMR Structure Determination
The 1 H NMR spectra of compound 11 at 293 K showed broad NH, H-3 and H-4 protons ( Figure S10). The NMR spectra were r DMSO-d6 and DMF-d7 solutions at different temperature intervals. A the signals were split, and four signals were observed for every pr high intensity and three similar signals of medium intensity were obs could be assigned to two isomeric compounds, one symmetrical c asymmetrical conformer ( Figure S10). A similar behavior was observ spectra. The NMR spectra of compound 11 at 293 K in DMSO and DM broad signals that coalesce to give signals for a single compound a Similarly, for compounds 4-6 and 12-14 at 223 K (DMF), two isomeri be detected: one symmetric and one asymmetrical conformer. Upon h to 343-403 K, the signals coalesced to show the presence of one single ( 1 H and 13 C spectra in the supporting information). 13

NMR Structure Determination
The 1 H NMR spectra of compound 11 at 293 K showed broad NH, H-3 and H-4 protons ( Figure S10). The NMR spectra were DMSO-d6 and DMF-d7 solutions at different temperature intervals. A the signals were split, and four signals were observed for every pr high intensity and three similar signals of medium intensity were ob could be assigned to two isomeric compounds, one symmetrical asymmetrical conformer ( Figure S10). A similar behavior was observ spectra. The NMR spectra of compound 11 at 293 K in DMSO and DM broad signals that coalesce to give signals for a single compound Similarly, for compounds 4-6 and 12-14 at 223 K (DMF), two isomeri be detected: one symmetric and one asymmetrical conformer. Upon to 343-403 K, the signals coalesced to show the presence of one single ( 1 H and 13 C spectra in the supporting information).

NMR Structure Determination
The 1 H NMR spectra of compound 11 at 293 K showed broad signals for the OH, NH, H-3 and H-4 protons ( Figure S10). The NMR spectra were recorded in CDCl 3 , DMSO-d6 and DMF-d7 solutions at different temperature intervals. At 223 K in DMF-d7, the signals were split, and four signals were observed for every proton. One signal of high intensity and three similar signals of medium intensity were observed. This pattern could be assigned to two isomeric compounds, one symmetrical conformer and one asymmetrical conformer ( Figure S10). A similar behavior was observed in the 13 C NMR spectra. The NMR spectra of compound 11 at 293 K in DMSO and DMF solutions showed broad signals that coalesce to give signals for a single compound at 353 K in DMSO. Similarly, for compounds 4-6 and 12-14 at 223 K (DMF), two isomeric compounds could be detected: one symmetric and one asymmetrical conformer. Upon heating the samples to 343-403 K, the signals coalesced to show the presence of one single molecule in DMSO ( 1 H and 13 C spectra in the supporting information).
Assignment of the NMR signals to the two conformers was performed at a low temperature in DMF-d7 solutions. At a high temperature in DMSO-d6, due to rapid exchange of the NMR signals, they were also assigned. Table 2 shows the ratio of isomers determined by 1 H-NMR spectroscopy.

Dynamic NMR Studies of Compounds 4-6 and 11-14
Using dynamic NMR spectroscopy (DNMR), we studied the restricted rotation of triazines 4-6 and 11-14 (Scheme 1). The most frequently used methods for deriving the exchange rate constants are line shape analysis and magnetization transfer [60,61]. Their application becomes inefficient, difficult and even impossible when the number of exchangeable forms increases or the difference in chemical shifts decreases. In such multi-site systems or when the dispersion of the signals is small, the use of 2D exchange spectroscopy (2D EXSY) is preferred. In 2D exchange spectroscopy, interference between processes is avoided and the rate constants for chemical exchange of each process can be estimated from the corresponding 2D matrix of integral intensities using the relationship between the integral intensities, mixing time and the rate constants [62]. Further, 2D EXSY has been applied to resolve numerous complex kinetic processes in many fields, including organometallic chemistry, metallotropy, fluxional behavior and conformational studies [63][64][65][66][67][68][69]. In addition, 1D EXSY techniques have also been introduced and successfully applied [70,71]. We applied the 2D Rotating-frame Overhauser Enhancement SpectroscopY (2D ROESY) [72] to study the chemical exchange because in ROESY experiments, the signals generated by chemical exchange differ from NOE signals [73], unlike in the NOESY experiment, where these effects may produce magnetization in the same direction. The ROESY experiment is affected by some possible complications [72], such as COSY-type crosspeaks between J-coupled spins, TOCSY transfers within J-coupled spin systems, offset dependencies of spin-lock fields, and zero-quantum (ZQ) artifacts, which may influence the quantitative applications of crossrelaxation. In order to avoid these artifacts, a T-ROESY [74] NMR experiment was developed. The offset dependencies of the spinlock field have to be considered for quantitative applications [75]. Thiele et al. [76] published the 'EASY ROESY' experiment, which makes the offset dependency of cross-peak volumes negligible and reduces TOCSY transfers. An efficient ZQ suppression scheme [77] was recently incorporated in NOESY experiments. Recently, Pure Shift NOESY was introduced [78], and gradient-selected pure-shift EASY-ROESY techniques have become quantitative [79]. In our case, the exchange is between spin systems located in different substituents. Therefore, we cannot expect TOCSY or COSY artefacts that will affect the quantitative measurements of exchange rates.
The assignment of exchanging conformers in VT NMR spectra of complex systems is difficult without the application of an integrated approach, which combines methods of dynamic NMR study and DFT calculation. Dynamic NMR study provides information about the exchanging sites and the corresponding rate constants. The DFT calculations can help in conformer assignment by predicting their populations and/or NMR chemical shifts. The comparison of experimental and DFT calculated barriers can reveal the exchange mechanism. This protocol, including the combination of dynamic NMR and DFT calculation, was recently successfully applied for studying the structure and exchange mechanism of palladium complexes [65,66,69], ortho-diphenylphosphinobenzenecarboxamide ligands [67], galantamine derivatives [68], and atropisomers of 2,20-diaryl-1,10-binaphthalenes containing three stereogenic axes [80].
The experimentally determined rotational barriers of the studied compounds in DMF-d7 are within the interval from 11.7 to 14.7 kcal/mol (Table 3). It is known that arylamino-1,3,5-triazines show hindered rotation about the partially double exocyclic C-N bond, due to the n-πconjugation of the amine lone pair with the triazine ring [81]. The substituents that enhance the double character of the C-N bond increase the barrier to hindered rotation. On the other hand, the steric effects of substituents have to be taken into account. An additional factor that will influence the rotational barriers is the solvent and the possible interactions with the solvent.
In order to better understand the role of the substituents, the solvent and the possible intramolecular hydrogen bonding on the rotational barrier height, we perform DFT calculations in the studied compounds.

DFT Calculations
In order to understand the trends in the populations of conformers and in the rotational barriers, we carried out a DFT conformational search in the ground state, which reveals the stable structures (GS) and transition state (TS) of all studied compounds (Figure 3). In order to better understand the role of the substituents, the solvent and the possible intramolecular hydrogen bonding on the rotational barrier height, we perform DFT calculations in the studied compounds.

DFT Calculations
In order to understand the trends in the populations of conformers and in the rotational barriers, we carried out a DFT conformational search in the ground state, which reveals the stable structures (GS) and transition state (TS) of all studied compounds ( Figure 3). Dynamic NMR study using 2D EXSY spectra provides information about the structure of exchanging conformers (one symmetric and one asymmetric conformer) and about the exchanging routes. DFT calculations of GS structures provide information about the thermodynamic stability of possible conformers. In most cases of exchange between complex structures or many conformers, the calculation of populations or prediction of chemical shifts helps in the conformer's assignment. This is redundant here, as Dynamic NMR study using 2D EXSY spectra provides information about the structure of exchanging conformers (one symmetric and one asymmetric conformer) and about the exchanging routes. DFT calculations of GS structures provide information about the thermodynamic stability of possible conformers. In most cases of exchange between complex structures or many conformers, the calculation of populations or prediction of chemical shifts helps in the conformer's assignment. This is redundant here, as it is clear from the experiment who is who; moreover, even the application of broad base sets and modern DFT methods does not predict the experimental populations accurately enough ( Table 2). The comparison of experimental and DFT calculated barriers can reveal the exchange mechanism. This integrated approach, which combines dynamic NMR spectroscopy and computational chemistry, has recently been successfully applied to study the structure and mechanism of exchange [63][64][65][66][67][68][69].
We begin the discussion with the results of the DFT calculations of the GS structures of our simplest compound. There are at least six possible GS conformations. GS1 is the most stable, and it represents the structure of the symmetric conformer. GS2 is obtained by rotation around the C triazine -N bond, and this structure is the asymmetric conformer. Further study of this structure led to the GS2_3H structure with three hydrogen bonds ( Figure S24). This structure is predicted to be preferable by SMD(DMF)//M062X/TZVP calculations. Both structures (GS2 and GS2_3H) are consistent with the measured 15 N NMR spectra of 1 at 243 K ( Figures S22 and S23). Unfortunately, no correlation was observed in 1H, 15 N-HSQC NMR spectra between hydroxyl hydrogens and 1,3,5-triazine nitrogens. The GS3 is obtained by rotation around the C Ph -N bond.
The TS structures are two, syn and anti TS, with respect to the orientation of oxygen atoms. In syn TS, the oxygens are on the same side, while in anti TS, one of the oxygen molecules is on the other side of the 1,3,5-triazine plane ( Figure S25). In a similar manner, there are two TS structures regarding rotation around the N-Ph bond: syn and anti TS ( Figure S26). The calculated theoretical thermodynamic parameters of restricted rotation around the 1,3,5-triazine-N bond are in good agreement with the experimental values (Table S23 and Table 3). The calculated theoretical thermodynamic parameters of restricted rotation around the N-Ph bond correspond to free rotation around a single bond (Table S24).
The same protocol was applied for all studied compounds: calculation of two syn and anti GS structures and two syn and anti TS structures. The results from the calculations of rotation around the triazine-N bond are given in Table 3. The obtained theoretical thermodynamic parameters of restricted rotation around the triazine-N bond reproduce the experimental values well. The observed differences can be reduced by the inclusion of specific solvent interaction in DMF.

General
Reagents were of commercial grade and used without further purification. Tetrahydrofurane (THF) and dioxane were distilled over sodium/benzophenone. Commercial anhydrous acetic acid was used for the reactions. Thin-layer chromatography (TLC) was performed on aluminum sheets pre-coated with Merck Kieselgel 60 F 254 0.25 mm (Merck, Darmstadt, Germany). Flash column chromatography was carried out using Silica Gel 60 230-400 mesh (Fluka, Buchs, Switzerland). Melting points were determined in a capillary tube on the SRS MPA100 OptiMelt (Sunnyvale, CA, USA) automated melting point system (uncorrected). The NMR spectra were recorded on a Bruker Avance II+ 600 (600.13 for 1 H NMR and 150.92 MHz for 13 C NMR) spectrometer with TMS as the internal standard for chemical shifts (δ, ppm). 1 H and 13 C NMR data are reported as follows: chemical shift, multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, br = broad, and m = multiplet), coupling constants J (Hz), integration, and identification. The assignment of the 1 H and 13 C NMR spectra was made on the basis of DEPT, COSY, HSQC, HMBC and NOESY experiments. Mass spectra (MS) were recorded on a Shimadzu Liquid Chromatograph Mass Spectrometer LCMS-2020. Direct MS-regime was applied; each compound was dissolved in MeOH with the following: concentration 1 mg/mL; injection volume 3-5 µL; mobile phase-1% formic acid in MeOH; and MS detector with electrospray ionization (ESI). MS spectra are reported as fragmentation in m/z with relative intensities (%). IR spectra were recorded using a Bruker Tensor 27 FTIR spectrometer.

Computational Details
All calculations were performed by means of quantum chemical calculations at the density functional theory (DFT) level using Gaussian09 Rev. D.01 program package [85].
The geometries of all compounds have been fully optimized, and the corresponding transition states were localized using the B3LYP [86] functional with the 6-31+G(d,p) basis set [87] or using the M06-2X [88,89] functional with the TZVP [90] basis set. Solvent effect was included implicitly to the optimizations via the SMD [91] model with the built-in parameters for solvents (DMF). The nature of all critical points was confirmed by means of vibrational analysis.
The ∆H, ∆S and ∆G values were calculated at T = 298.15 K at the same level of theory including zero-point energy in the particular solvent environment (represented by relative permittivity) and vibrational, rotational and translational thermal energy corrections.
For prediction of the UV-Vis spectral data, the M06-2X [88,89] functional with the TZVP [90] basis set was used for the structure optimizations in the ground state. Solvent effect was included implicitly to the optimizations via the SMD [91] model with the builtin parameters for the solvent (CH 3 CN). In the excited state, CAM-B3LYP [92] with the same basis set was applied [93]. The TD-DFT method [94,95] was used for singlet excitedstate optimizations, again without restrictions, applying tight optimization criteria and an ultrafine grid in the computation of two-electron integrals and their derivatives. Bearing in mind that M06-2X systematically underestimates the absorption band positions [96], the UV-Vis spectral data were predicted by the B3LYP [97] functional using the M06-2X optimized ground-state geometries.

Dynamic NMR Studies
1 H and 13 C spectra were recorded on a Bruker Avance II+ 600 spectrometer (Bruker BioSpin GmbH, Ettlingen, Germany) using BBO probe at 600.13 for 1 H NMR and 150.92 MHz for 13 C NMR with TMS as the internal standard for chemical shifts (δ, ppm). The spectra were recorded in steps of 5 K between 223 and 323 K (0.05M in 600 µL DMF-d7). Temperature calibration was performed with the builtin B-VT 3000 unit (it was checked and calibrated with methanol and ethylene glycol reference samples). 1 H NMR spectra were acquired using a spectral width of 10 kHz, an acquisition time of 3.4 s, and 32 scans, zerofilled to 64k datapoints (0.15 Hz per point) and processed without apodization. 1 H EXSY spectra (noesygpphzs Bruker pulse program) were recorded on a BBO probe typically in steps of 5 K between 223 and 258 K. The spectra were acquired using a spectral width of 4.2 kHz, 2048 × 256 complex time domain datapoints, mixing times in the range of 0.02 to 1.5 s and 2 scans in about 45 min. The spectra were zerofilled to 4096 × 4096 datapoints and processed with a shifted square sine bell apodization in both dimensions. The measurement details and processing of studied compounds can be found in the Supporting Information. Exchange rates were calculated from diagonal and crosspeak integrals using EXSYCalc (MestreLab Research S.L., Santiago de Compostela, Spain).
Error analysis: Usually, the presented errors in the activation parameters are the statistical errors based on scattering of the data points around the Eyring straight line only. The errors in this analysis are due to inaccuracies in both the calculated rate constants, k, and the measured temperatures, T, and are computed according to the error propagation equations of Binsch [97] and Heinzer and Oth [98]. The absolute error in temperature is assumed to be no more than ±0.5 K. The relative errors in k are estimated to be no more than ±10% at all temperatures according to the precision of the volume integration of peaks. The errors analysis was performed using a self-made computer program using the cited equations.

Conclusions
Seven OLED-relevant compounds (two known and five new) consisting of a 1,3,5triazine core linked to various aromatic arms by amino group were synthesized and characterized. Conformational studies with respect to hindered rotation around the C(1,3,5triazine)-N bond and C(Ph)-N bond were performed. The conformations of the studied compounds were elucidated on the basis of NMR and DFT studies. In the solution, two conformers were revealed and proved to originate from restricted rotation around the C(1,3,5-triazine)-N bond. The absorption and emission spectra of the studied compounds, as well as the HOMO-LUMO energy gap, limited the selection of some of these compounds for OLED applications. Further studies will show the applicability of these compounds in novel multilayer devices.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/molecules28031248/s1, Analytical data for synthesized compounds; 1 H and 13 C NMR spectra of the studies compounds; Dynamic NMR spectra (Figures S1-S21: VT 1 H NMR spectra, 1 H ROESY spectra and Eyring plots of rate constants; Tables S1-S21: Rate constants and thermodynamic parameters of exchange process); Figure