4-(Tris(4-methyl-1H-pyrazol-1-yl)methyl)aniline

Abstract

4-(tris(4-methyl-1H-pyrazol-1-yl)methyl)aniline was prepared in a 63% yield utilizing a C–F activation strategy from a mixture of 4-(trifluoromethyl)aniline, 4-methylpyrazole, and KOH in dimethylsulfoxide (DMSO). The identity of the product was confirmed by nuclear magnetic resonance spectroscopy, infrared spectroscopy, mass spectrometry, and single-crystal analysis. An analysis of crystals grown from the layering method (CH₂Cl₂/acetone/pentane) indicated two distinct polymorphs of the title compound. Moreover, density functional theory calculations utilizing the MN15L density functional and the def2-TZVP basis set indicated that 4-(tris(4-methyl-1H-pyrazol-1-yl)methyl)aniline forms with similar energetics to the previously reported unmethylated analog.

1. Introduction

Tris(pyrazolyl)methane (Tpm) ligands (Figure 1a) and their corresponding metal complexes have been exploited for a broad range of applications, including catalysis and biomedical chemistry [1,2,3]. For instance, the molecule [TpmMn(CO)₃]PF₆ has been investigated as a chemotherapeutic to treat colon cancer [4]. The neutral Tpm ligands are isoelectronic analogs of the anionic tris(pyrazolyl)borate (Tp) ligands [5] (Figure 1b), another class of tridentate ligands that have also been utilized in many applications [6,7,8]. Of note, Tp*Rh(CO)₂ (where Tp* = HB-Pz₃*, Pz* = 3,5-dimethylpyrazolyl) has been utilized to activate C-H bonds [9]. Both the Tp and Tpm ligands were reported by Trofimenko, with the preliminary study of Tp ligands demonstrating that these ligands could bind to both early and late transition metal complexes [10]. Later, Elguero et al. developed an improved synthesis of Tp ligands that exploited a phase transfer catalyst, tetrabutylammonium (TBA) bisulfate [11]. Current methods for generating Tp ligands still exploit phase transfer catalysts and have been utilized to synthesize a variety of Tp derivatives where substituents are added to the pyrazole ring, often providing desirable steric features for their corresponding metal complexes [3]. A disadvantage of this strategy is the long reaction time that is required. For example, a recent report describing the synthesis of the tris(1-pyrazolyl)methane ligand utilizing the TBABr phase transfer catalyst requires a 3-day heating period at 70 °C [12].

In some instances, only a short heating period is required to generate Tp ligands. A report by Liddle et al. describes a C–F activation strategy to synthesize Tp ligands (Figure 1c) that requires only a 1 h reflux in DMSO [13]. Our group has also exploited such a strategy to generate Tp derivatives, including a bitopic Tp ligand [14]. Inspired by Liddle’s work, we aim to utilize this C–F activation strategy to generate Tp ligands with functionalized pyrazole rings. Herein, we report the synthesis and characterization of 4-(tris(4-methyl-1H-pyrazol-1-yl)methyl)aniline (1), highlighting that the C–F activation strategy can be utilized with substituted pyrazoles.

2. Results and Discussion

2.1. Synthesis

4-(tris(4-methyl-1H-pyrazol-1-yl)methyl)aniline (1) was prepared by refluxing a mixture of KOH, 4-(trifluoromethyl)aniline (2), and 4-methylpyrazole (3) in DMSO for 1 h (Scheme 1). Multiple aqueous washes were required to remove the DMSO from the crude product mixture. Purification by column chromatography and subsequent Et₂O washes afforded 1 as determined by multiple analytical methods: nuclear magnetic resonance (NMR), infrared (IR) spectroscopy, mass spectrometry (MS), and X-ray crystallography.

2.2. Characterization

¹H-NMR spectroscopy provided initial evidence for the newly synthesized Tpm derivative (1). The symmetry of this species provided six unique hydrogen environments. The signature features of this compound include the downfield singlets at 7.50 and 7.25 ppm corresponding to the pyrazole hydrogens, the doublets at 6.88 and 6.62 ppm corresponding to the hydrogens on the aniline ring, a broad singlet at 3.85 ppm corresponding to the N-H protons, and the upfield singlet at 2.05 ppm corresponding to the methyl groups on the pyrazole rings. ¹³C-NMR indicated nine unique chemical environments, including four quaternary carbons, which is consistent with the symmetry of this species. HSQC and HMBC data were utilized to confirm our NMR assignments (e.g., no C-H correlation was observed for the broad singlet at 3.85 ppm). Specific assignments of the ¹H and ¹³C resonances are reported in the Supplemental Information section (Figure S4).

Mass spectrometry data supported product formation with peaks at m/z 370.0 and 717.2 corresponding to sodium (Na⁺) adducts of our complex (respectively, [C₁₉H₂₁N₇ + Na⁺] and [(C₁₉H₂₁N₇)₂ + Na⁺]). The base peak in the mass spectrum (m/z 266.0) corresponded to the loss of a pyrazole ligand from the complex [M − 81]⁺ (Figure 2, compound 4). The loss of one of the pyrazoles during electrospray is plausible considering the capillary and cone voltages and the fact that resonance from the aniline would stabilize the resulting carbocation (Figure 2).

IR spectroscopy confirmed the presence of aromatic rings (1626 cm⁻¹) and the aniline N-H stretches (3328 cm⁻¹ and 3448 cm⁻¹). These assignments were confirmed by performing DFT calculations. Optimization and frequency calculations were performed at the MN15L/def2-TZVP level on 4-(tris(4-methyl-1H-pyrazol-1-yl)methyl)aniline (1). The visualization of the calculated frequencies of 1628 cm⁻¹ to 1639 cm⁻¹ indicated they were due to the aromatic rings. The calculated frequencies of 3291 cm⁻¹ and 3533 cm⁻¹ corresponded to the N-H bond stretches. The calculations are consistent with the experimental IR data.

2.3. Solid-State Structure and Polymorphism

Single crystals of 4-(tris(4-methyl-1H-pyrazol-1-yl)methyl)aniline (1) were grown by layering a CH₂Cl₂ solution of the title compound with acetone followed by pentane. Growth in a standard NMR tube yielded two distinct morphologies—blocklike crystals (polymorph I) and tabular crystals (polymorph II). Single-crystal X-ray diffraction indeed revealed these to be two different polymorphs of the title compound (Figure 3). Polymorph I crystallizes in space group P2₁/n, while polymorph II crystallizes in space group C2/c (Supporting Information, Table S1). The polymorphs may be considered to have a rotamer relationship with one another, where fixing the orientation of the aniline substituent leads to differing orientations of the N-N bonds and methyl groups of the methylpyrazolyl substituents (Figure 3, overlay). The mean-plane-to-mean-plane angles between the substituent rings are summarized in

Table 1. Mean-plane angles (degrees) between substituent rings in polymorphs I and II of compound 1. Rings are defined according to the atom labeling in Figure 3: A: N1–C4, B: N3–C8, C: N5–C12, D: C14–C19.

Rings	Polymorph I	Polymorph II
A–B	87.96 (5)	73.20 (7)
A–C	22.66 (10)	57.19 (11)
B–C	83.37 (4)	79.06 (10)
A–D	78.16 (5)	56.80 (10)
B–D	80.60 (5)	81.78 (10)
C–D	79.78 (5)	79.77 (10)

. In polymorph I, two of the methylpyrazolyl rings begin to approach coplanarity (22.66 (10)°), which is not the case in polymorph II, where the same methylpyrazolyl rings are more significantly inclined to one another (57.19 (11)°). In this way, polymorph II more closely resembles the structure of 4-(tri(1H-pyrazol-1-yl)methyl)aniline (CSD refcode HIXLIK), where the shallowest dihedral angle between pyrazolyl rings is 44.37° [13].

Both polymorphs feature intermolecular N-H···N interactions between the NH₂ group of the aniline substituent and the N6 acceptor atom of a neighboring molecule (

Table 2. Selected intermolecular interactions in polymorphs of the title compounds.

Polymorph	D-H (Å)	H···A (Å)	D···A (Å)	D-H···A (°)
I
N7-H7N2···N6 a	0.92 (2)	2.33 (2)	3.2463 (19)	178.2 (18)
C10-H10···N4 b	0.95	2.65	3.5450 (19)	157.0
II
N7-H7N2···N6 c	0.94 (3)	2.33 (3)	3.199 (3)	152 (2)

Symmetry codes: ^a −x + 2, −y + 1, −z + 1; ^b x + 1, y, z; ^c x + 1/2, y − 1/2, z.

). The differing rotamer orientation of the nitrogen atoms in the polymorphs, however, leads to different long-range packing patterns from these interactions (Figure 4). In polymorph I, the N-H···N interactions are coupled to create dimers of neighboring molecules. These dimers are then extended into a one-dimensional long-range structure via C-H···N interactions, where the chains of dimers propagate along the a-axis (Supporting Information, Figure S9). In polymorph II, the N-H···N interactions form the one-dimensional long-range structure without additional assistance. These chains propagate along the [−1 1 0] crystallographic direction. The N-H···N chain arrangement in polymorph II is again more closely aligned with what occurs in 4-(tri(1H-pyrazol-1-yl)methyl)aniline (Figure 5, compound 5), though the significant corrugation in the chains of 5 does provide a notable distinction with the straight chains of polymorph II.

2.4. Computational Studies

DFT calculations using the MN15-L functional and the def2-TZVP basis set were performed to compare the reaction thermodynamics of the methylpyrazole (3) and the parent unsubstituted pyrazole with 4-(trifluoromethyl)aniline (2) (Figure 5). Optimization and frequency calculations were performed on 4-(tris(4-methyl-1H-pyrazol-1-yl)methyl)aniline (1), 4-(trifluoromethyl)aniline (2), 4-methylpyrazole (3), 4-(tri(1H-pyrazol-1-yl)methyl)aniline (5), pyrazole (6), and hydrogen fluoride. The Gibbs free energy of the reaction to produce 1 (Figure 5, Reaction 1) is calculated to be 7.1 kcal/mol, compared to the Gibbs free energy of the reaction to produce 5 (Figure 5, Reaction 2), which is 7.6 kcal/mol. For the calculated enthalpies of reaction, Reaction 1 has an enthalpy of 9.6 kcal/mol, whereas Reaction 2 has an enthalpy of 10.4 kcal/mol. The calculated Gibbs free energy and enthalpy values indicate that Reaction 1 is more favorable compared to Reaction 2. This small increase in favorability for the methylated compound is attributable to the incorporation of a methyl group into the pyrazole reactant, making it a better nucleophile.

Using the same level of theory, transition states were calculated (Figure 6) for the reactions between 4-(trifluoromethyl)aniline (2) and the two pyrazole derivatives (3 and 6). For this calculation, SMD solvation was included to model the DMSO solvent [15]. Based on pKas and the fact that KOH is present in the reaction conditions, the deprotonated forms of the pyrazole derivatives were selected as the nucleophiles in the transition state calculations [16,17]. As suggested by the computational results, a possible mechanism for product formation involves an S_N2-type transition state where the deprotonated pyrazole species replaces a fluoride leaving group. The activation energy is slightly lower when deprotonated 4-methylpyrazole (3) is used as a nucleophile, which is again reflective of this species being a better nucleophile. The high activation barriers are consistent with the need for refluxing DMSO (bp = 189 °C).

3. Materials and Methods

All reagents were purchased from commercial suppliers and used as received: 4-(trifluoromethyl)aniline and 4-methylpyrazole (Ambeed, Inc., Arlington Heights, IL, USA); KOH and DMSO (Fisher Scientific, Hampton, NH, USA). NMR spectra were obtained with a JEOL (Peabody, MA, USA) 400 MHz YH. ¹H NMR resonances were referenced against tetramethylsilane using the residual proton signal, and ¹³C NMR resonances were referenced against the ¹³C resonance of CDCl₃ (77.16 ppm) [18]. IR spectra were obtained with a Bruker (Ettlingen, Germany) ALPHA-P. IR data analysis was performed with OPUS version 8.5.

The purified sample was dissolved in LC-MS-grade acetonitrile, and MS data were obtained via direct infusion into a Waters (Milford, MA, USA) Xevo TQD operated in scan mode using the first quadrupole. Electrospray ionization in the positive mode was used with a capillary voltage of 2.5 kV, a cone voltage of 25 V, a desolvation temperature of 250 °C, a desolvation flow of 500 L N₂/h, and a cone flow of 20 L N₂/h. The mass spectrum was scanned from m/z 100–900 at 3000 amu/sec. Data analysis was performed with MassLynx version 4.2.

The DFT calculations were performed using Gaussian 16 computation software [19] with the MN15L density functional [20] and def2-TZVP basis set [21]. The MN15-L functional was chosen because of its accuracy in modeling organic molecules with good agreement with experimental trends [22], and the def2-TZVP basis set is reasonably accurate while not being as computationally expensive as its larger relatives [23]. For the transition state calculation, SMD solvation was used to model DMSO [15]. Optimization and frequency calculations were performed on 4-(tris(4-methyl-1H-pyrazol-1-yl)methyl)aniline (1), 4-(trifluoromethyl)aniline (2), 4-methylpyrazole (3), 4-(tri(1H-pyrazol-1-yl)methyl)aniline (5), pyrazole (6), and hydrogen fluoride. Each molecule was benchmarked for accuracy by performing a Root Mean Squared Deviation test for $p\le 0.05$ and an analysis of the Pearson Correlation Coefficient value x such that $0.90\le x \le 1.00$. These tests were performed on the bond lengths between the computed molecular models and the real molecule’s data using R Statistics Software version 4.3.3 [24]. The Gibbs free energy and enthalpy values of the molecules were used to determine the favorability of the reaction between the methyl-substituted aniline product (1) and the unsubstituted aniline compound product (5).

Crystallographic data were obtained at 100 K using a Bruker (Madison, WI, USA) D8 Quest diffractometer with an Incoatec (Geesthacht, Germany) microfocus Mo Kα (λ = 0.71073 Å) source and a Photon III CMOS detector. The APEX4 (2022.10-0) software package [25] was used for instrument control and data processing (SAINT integration and SADABS multiscan absorption correction). The structure was solved by intrinsic phasing (SHELXT) [26] and refined using least-squares techniques (SHELXL) [27]. All non-hydrogen atoms were refined anisotropically. Hydrogen atoms attached to carbon atoms were treated using riding models. Hydrogen atoms attached to nitrogen atoms were identified from the difference in electron density map, and their positions were refined without restraints. Renderings of the crystal structure were generated using Mercury (2023.3.1) [28]. Crystallographic data are summarized in the Supporting Information, Table S1. Data may be obtained in CIF form from the Cambridge Crystallographic Data Centre via CCDC 2344377–2344378.

4-(tris(4-methyl-1H-pyrazol-1-yl)methyl)aniline: A mixture of KOH (1.69 g, 30.1 mmol), 4-(trifluoromethyl)aniline (1.62 g, 10.1 mmol), and 4-methylpyrazole (2.47 g, 30.1 mmol) in 10 mL of DMSO was heated to reflux under N₂. The yellow heterogeneous mixture changed to a red homogeneous solution before finally becoming yellow and homogeneous during the reflux period. After 1 h, the reaction mixture was diluted with CH₂Cl₂ (100 mL) and washed with 5 × 20 mL of NaCl (sat, aq.). The organic layer was dried over MgSO₄, and the solvent was removed in vacuo. The resulting yellow semi-solid was dissolved in minimal CH₂Cl₂, purified by column chromatography on silica gel (eluent = Et₂O, R_f = 0.43), and subsequently washed with Et₂O (3 × 20 mL) to afford the desired product as a white solid (2.18 g, 6.29 mmol, 63%). Note: Because the title compound is insoluble in Et₂O, while collecting fractions, the product occasionally precipitates out of solution. If this occurs, CH₂Cl₂ can be used to redissolve the solid. ¹H-NMR (CDCl₃, 400 MHz), δ (ppm): 7.50 (s, 3H), 7.25 (s, 3H), 6.88 (d, J = 8.7 Hz, 2H), 6.62 (d, J = 8.7 Hz, 2H), 3.85 (s, 2H), 2.05 (s, 9H). ¹³C-NMR (CDCl₃, 101 MHz), δ (ppm): 148.1, 141.9, 130.6, 130.1, 127.3, 116.5, 114.2, 92.9, 9.1. IR ν_max (cm⁻¹): 3448, 3328, and 1626.

4. Conclusions

We describe the synthesis and characterization of 4-(tris(4-methyl-1H-pyrazol-1-yl)methyl)aniline (1). The synthesis and purification of this species can be performed in under a day because of the short reaction time needed (1 h) to synthesize this Tp derivative, which contrasts with the lengthy heating periods often required for the synthesis of these ligands (vide supra). Crystal growth of 1 fortuitously led to the formation of two polymorphs, differing primarily in the rotational orientation of the pyrazolyl substituents and the resulting N-H···N intermolecular interactions. Specifically, the formation of N-H···N dimers versus N-H···N chains distinguishes polymorphs I and II, respectively. Because the addition of methyl groups is often utilized to increase the solubility of species in organic solvents [29,30], future work will investigate the solubility differences between metal compounds of our title compound and 4-(tri(1H-pyrazol-1-yl)methyl)aniline (5). Moreover, because other 4-sustituted pyrazole ligands (e.g., 4-fluoropyrazole) are commercially available, this strategy may be useful for generating Tp derivatives with varying electronic properties. The generation of novel Tp ligands is advantageous because of their use in catalysis and medicinal chemistry [1,2,3,4], and Tp ligands have received less attention compared to their Tpm analogs [3,31].