The Crystal Structure of N,N′-bis(3,5-di-tert-Butylsalicylidene)propane-1,2-diamine)-oxidovanadium(IV) and Its Parent Ligand

Abstract

The asymmetric salen ligand N,N′-bis(3,5-di-tert-butylsalicylidene)propane-1,2-diamine (1) with a single chiral center and its vanadyl complex (2) have been prepared and characterized by single-crystal X-ray diffraction.

1. Introduction

First reported in 1889, salen ligands are formed through the 2:1 condensation reaction between a salicylaldehyde (sal) derivative with a primary diamine (en). Upon deprotonation of the phenols, a tetradentate ligand with a square planar N₂O₂ coordination core is formed, which can readily chelate to a metal center (Scheme 1) [1,2]. A vast amount of work has been carried out in the field of metal–salen chemistry, as evidenced by the many reviews that have been published regularly since the 1960s [3,4,5,6,7].

Salen ligands can be functionalized by utilizing substituted salicylaldehydes and substituted diamines. A very well-known use of salen complexes is in asymmetric catalysis, for example in the expoxidation of alkenes using Jacobsen’s catalyst. The ligand in this complex features a chiral backbone due to the inclusion of trans-1,2-diaminocyclohexane as the starting diamine, and bulky tert-butyl groups on the phenol ring, and is coordinated to a Mn(III) center [8,9] (Figure 1). Salen–metal complexes have also been used in many other catalytic transformations such as the oxidation of sulfides [10], hydroxylation [11], aziridination [12], and cyclopropanation [13], to name but a few. Many metals have been utilized in salen-ligand-based catalysis, including transition metals such as titanium [14,15], vanadium [16,17], iron [18], cobalt [19], palladium [20], and iridium [21], p-block metals such as aluminum [22], indium [23], and tin [24], and even s-block and f-block metals [25]. Accordingly, metal–salen complexes have been studied for wider applications including biosensors [26], artificial enzymes [27], and DNA modification [28].

The chemistry of vanadium-based salen complexes is mainly dominated by vanadium in the +III, +IV and +V oxidation states with uses such as electrochemical reductions [17,29,30] and oxygen transfer reactions [30,31]. The oxidovanadium(IV), or vandyl, ion (VO²⁺) is well known for its stability in both organic and aqueous conditions [32] and, as such, is the core feature in many vanadium–salen complexes. The incorporation of an asymmetric methyl(ethylene) spacer into a salen ligand provides a simple and effective way to break the inherent C₂ symmetry of the classical ethylenediamine backbone, thereby introducing a well-defined chiral environment around the metal center. This asymmetry can promote enhanced diastereoselective substrate binding and improve control over catalytic pathways by creating inequivalent coordination sites. Additionally, the substituted backbone increases conformational rigidity while subtly tuning electronic properties [33].

Herein we report the crystal structures of the asymmetric salen ligand N,N′-bis(3,5-di-tert-butylsalicylidene)propane-1,2-diamine (1) and its vanadyl complex (2) (Scheme 2).

2. Results and Discussion

The asymmetric ligand 1 with a single chiral center was first utilized in several electrochemical and catalysis studies of iron [34] and copper [35] complexes. Complex 2 has been prepared previously as part of a study into the reactivity of vanadyl–salen complexes with bulky substituents [36], but neither have been characterized by single crystal X-ray diffraction.

The salen ligand 1 was prepared by the condensation of racemic 1,2-diaminopropane and 3,5-di-tert-butyl-2-hydroxybenzaldehyde in ethanol under reflux, affording 1 as a bright yellow powder in excellent yields (91%) (Scheme 2). This was characterized by ¹H NMR spectroscopy, which shows the asymmetry in the molecule, owing to the chiral center, through minor differences in the chemical shifts (Figures S1 and S2), for example, the imine hydrogen atoms with one at δ_H 8.39 ppm and the other at δ_H 8.35 ppm. The effect is not as pronounced with the aromatic hydrogen atoms but still noticeable, with the position meta to the tert-butyl groups showing signals at δ_H 7.07 and 7.06 ppm. This is also prevalent in the ¹³C DEPTQ NMR spectrum, with the imine carbon atoms having signals at δ_C 167.6 and 165.8 ppm (see Supplementary Materials Figure S4). The phenolic hydrogen atoms cannot be observed in CDCl₃ due to dynamic isotope exchange but are observed at δ_H 13.99 ppm in d₆-DMSO (Figure S3).

The vanadyl complex of 1 was prepared by heating a mixture of 1 and VO(acac)₂ under reflux in ethanol. The dark green precipitate was collected in good yield (76%). The HRMS confirmed the identity of 2 owing to the M+Na peak present at 594.2994 (calcd. 594.3002 for C₃₃H₄₈N₂O₃VNa) (Figures S5 and S6). The complex is dark green in the solid state and in solution. There is a medium absorption band at λ_max 630 nm with ε = 199 dm³ mol⁻¹ cm⁻¹, indicating a d−d transition (Figure S6).

Crystals of both 1 and 2, suitable for single crystal X-ray diffraction, were grown from acetonitrile. The structure of 1 shows that the entire diimine backbone (C15=N16−C17(−C21)−C18−N19=C20) was disordered over a symmetry element and that there is minor disorder in the orientation of the phenolic hydrogen atoms, which is related to this (Figure 2 and Figure S9). The structure is centrosymmetric, with both aromatic rings being crystallographically identical. All bond lengths and angles in 1 are all within the expected ranges (

Table 1. Selected bond lengths (Å) and angled (°) from 2. Values in square parentheses are from the second molecule in the asymmetric unit.

	2
C1−O1	1.316(4) [1.317(4)]
C15−N16	1.285(5) [1.290(5)]
C17−N16	1.516(6) [1.502(5)]
C18−N19	1.464(5) [1.471(5)]
C20−N19	1.278(5) [1.282(5)]
V1−N16	2.062(4) [2.061(3)]
V1−N19	2.047(3) [2.042(3)]
V1−O1	1.935(3) [1.937(3)]
V1−O21	1.911(3) [1.916(3)]
V1−O11	1.593(3) [1.594(3)]
O1−V1−O21	77.9(1) [78.0(1)]
N6−V1−N19	90.2(1) [89.8(1)]
Out-of-plane displacement	0.5878(16) [0.5877(16)]

).

The vanadyl complex 2 was also characterized by single crystal X-ray diffraction. Two molecules of 2 are present in the asymmetric unit with very minor differences in bond lengths and angles (Figure 3 and Table 1), leading to a root mean square deviation across the whole molecule (major component of disorder) of 0.03, as well as a maximum deviation in atomic positions of 0.09 Å.

Both enantiomers are present in the solid, with the methyl(ethylene) backbone of both independent units in 2 showing substitutional disorder with the methyl swapped between the two ethyl carbons (Figure S10). In both independent molecules, the same enantiomer predominated at >90% for one and ~58% for the other. The vanadium adopts a square pyramidal geometry, with the V atom sitting 0.5878(16) Å above the mean N₂O₂ plane. The V=O bond is of characteristic length at 1.593(3) [1.594(3)] Å and can also be identified from the strong stretch at 976 cm⁻¹ in the IR spectrum [37,38]. The aromatic rings are no longer coplanar and sit with an angle of 15.03(12)–15.41(12)° between them with the inclusion of vanadium.

3. Materials and Methods

Glassware was dried in an oven (ca. 110 °C) prior to use. All chemicals were used as provided from the laboratory inventory without further purification. The IR spectrum was recorded on a Perkin Elmer Spectrum Two instrument (High Wycombe, UK) with a Deuterated Triglycine Sulfate (DTGS) detector and diamond Attenuated Total Reflectance (ATR) attachment (Bruker, Billerica, MA, USA). The High-Resolution Mass Spectrometry (HMRS) data were acquired from the University of St Andrews Mass Spectrometry Service using a ThermoScientific Orbitrap Fusion Lumos with Ultimate 3000 Nano LC (Oxford, UK) and Vanquish standard LC (ThermoFisher, Oxford, UK). All NMR spectra were recorded using a Bruker Avance III 400 MHz spectrometer at 20 °C (Bruker, Billerica, MA, USA). The ¹³C spectrum was recorded using the DEPTQ-135 pulse sequence with broadband proton decoupling. Tetramethylsilane was used as external standard for ¹H and ¹³C NMR (δ_H/δ_C 0.00 ppm). Where possible, the residual solvent signal was used as a secondary reference (CDCl₃, δ_H 7.260, δ_C 77.160 ppm; d₆-DMSO, δ_H 2.500 ppm). Chemical shifts (δ) are given in parts per million (ppm) relative to the solvent peaks. Coupling constants (J) are given in Hertz (Hz). NMR Spectra were analyzed using the MestReNova software package (Santiago de Compostela, Spain) (version 14). The NMR numbering scheme for 1 is shown in Figure 4.

3.1. Synthesis and Characterization of 1

A solution of 3,5-di-tert-butyl-2-hydroxybenzaldehyde (5.86 g, 25 mmol) was added to ethanol (30 mL) and heated gently to dissolve. To this, a solution of racemic 1,2-diaminopropane (930 mg, 12.5 mmol) in ethanol (10 mL) was added dropwise over 1–2 min. The solution was heated under reflux for two hours. Upon cooling, a bright yellow precipitate was formed. This was isolated via vacuum filtration and washed with ice-cold ethanol (3 × 10 mL) before being dried in vacuo overnight (5.79 g, 91%). Crystals of 1 of suitable quality for SCXRD were grown by slow evaporation of a saturated solution in acetonitrile. ¹H NMR (400.3 MHz, CDCl₃) δ_H 8.39 (1H, s, H-7′), 8.35 (1H, s, H-7), 7.37 (1H, d, ⁴J_HH 2.5 Hz, H-4/4′), 7.36 (1H, d, ⁴J_HH 2.5 Hz, H-4/4′), 7.07 (2H, 2×d ~t, ⁴J_HH 2.5 Hz, H-6,6′), 3.88–3.80 (1H, m, H-14/14′), 3.77–3.67 (2H, m, H-13, H-14/14′), 1.44 (9H, s, H-10/10′), 1.43 (9H, s, H-10/10′), 1.42 (3H, d, ³J_HH 6.2 Hz, H-12), 1.28 (18H, s, H-11,11′). ¹H NMR (400.3 MHz, d₆-DMSO) δ_H 13.99 (2H, s, OH), 8.57 (1H, s, H-7′), 8.55 (1H, s, H-7), 7.28 (2H, 2×d ~ t, ⁴J_HH 2.7 Hz, H-4/4′), 7.21 (2H, d, ⁴J_HH 2.7 Hz, H-6,6′), 3.87–3.81 (1H, m, H-14/14′), 3.79–3.69 (2H, m, H-13, H-14/14′), 1.36 (9H, s, H-10/10′), 1.35 (3H, s, H-12), 1.34 (9H, s, H-10/10′), 1.24 (18H, s, H-11,11′). ¹³C NMR (100.7 MHz, CDCl₃) δ_C 167.6 (s, C-7′), 165.8 (s, C-7), 158.2 (s, C-2/2′), 158.0 (s, C-2/2′), 140.23 (s, C-5/5′), 140.19 (s, C-5/5′), 136.8 (s, C-3/3′), 136.7 (s, C-3/3′), 127.2 (s, C-4/4′), 127.1 (s, C-4/4′), 126.22 (s, C-6/6′), 126.18 (s, C-6/6′), 117.9 (s, C-1,1′), 65.9 (s, C-14), 65.1 (s, C-13), 35.1 (s, C-8,8′), 64.2 (s, C-9,9′), 31.6 (s, C-11,11′), 29.59 (s, C-11/11′), 25.56 (s, C-11/11′), 20.7 (s, C-12).

3.2. Synthesis and Characterization of 2

A solution of 1 (1.12 g, 2.2 mmol) and oxidobis(acetylacetonato)vanadium(IV) (580 mg, 2.2 mmol) were combined as solids under an atmosphere of dry nitrogen. Dry ethanol (40 mL) was added, and the solution heated under reflux for two hours. The mixture was cooled to ambient temperature and the dark-green precipitate collected via vacuum filtration and dried in vacuo for 30 min (680 mg, 76%). Crystals of 2 of suitable quality for SCXRD were grown by slow evaporation of a saturated solution in acetonitrile.

IR: νmax (ATR)/cm⁻¹ 2956w (ν_CH), 2867w (ν_CH), 1634s (ν_CN), 1252s, 976s (ν_VO), 835s (CH bend), 543s, 488m (ν_VN). UV–Vis: λ_max (CH₂Cl₂)/nm 630 (ε/dm³ mol⁻¹ cm⁻¹ 199). HRMS (ES+): m/z (%) Calcd. for C₃₃H₄₈N₂O₃VNa: 594.3002, found 594.2994 [M + Na⁺] (100).

3.3. X-Ray Crystallography

X-ray diffraction data for compounds 1 and 2 were collected using a Rigaku FR-X Ultrahigh Brilliance Microfocus RA generator/confocal optics with an XtaLAB P200 diffractometer (Tokyo, Japan) [Mo Kα radiation (λ = 0.71073 Å)]. Data for both compounds were collected (using a calculated strategy) and processed (including correction for Lorentz, polarization, and absorption) using CrysAlisPro [39]. The structures were solved by dual-space methods (SHELXT [40]) and refined by full-matrix least-squares against F² (SHELXL-2025/1 [41]). Non-hydrogen atoms were refined anisotropically, and hydrogen atoms were refined using a riding model except for the two hydrogen atoms on O1 and the hydrogen on C20 in 1, which were located from the difference Fourier map and refined with U_iso riding on their parent atom, occupancy fixed at 0.5, and either subject to a distance restraint (O1) or constrained distance (C20). The diimine backbone in 1 was disordered over an inversion center and all seven atoms were modeled in one complete section (PART-1), with occupancy fixed at 0.5, and tight geometric and thermal restraints. Electron-density peaks in the refinement of 1 suggested multiple additional disordered positions of the atoms in the diimine backbone were possible, and that the disorder might extend into the aromatic rings, all of which would be at low occupancy. Attempts were made to incorporate several of these additional atomic positions into the disorder model, including transforming the structure into the space group P1 to remove the inversion center. None of these resulted in a complete stable model of the disorder, although several of these partial additional models showed improved values of R₁ and wR₂. This suggests that the somewhat elevated values of the R-factors arise due to the additional unmodeled disorder. The methyl(ethylene) backbone of both independent units in 2 showed substitutional disorder, with the methyl swapped between the two ethyl carbons and showing a different apparent enantiomer. In one independent unit, the occupancy of the minor site was low (<10%) and refinement was unstable when this was modeled. In the other independent molecule, the minor and major components had a more equal ratio of occupancy, and the methyl carbon (C35) was modeled in two parts without restraints. All calculations were performed using the Olex2 interface [42]. Selected crystallographic data are presented in

Table 2. Selected crystallographic data for 1 and 2.

Title 1	1	2
formula	C33H50N2O2	C33H48N2O3V
fw	506.75	571.67
temperature [K]	173	100
crystal description	pale yellow prism	green prism
crystal size [mm3]	0.42 × 0.06 × 0.02	0.09 × 0.07 × 0.02
space group	P$\overline{1}$	P21
a [Å]	5.9532(6)	6.59478(14)
b [Å]	9.8368(11)	18.8177(4)
c [Å]	14.0955(12)	25.7633(6)
α [°]	77.493(8)
β [°]	81.404(8)	90.4155(19)
γ [°]	73.661(9)
vol [Å]3	769.85(14)	3197.11(12)
Z	1	4
ρ (calc) [g/cm3]	1.093	1.188
μ [mm−1]	0.067	0.344
F(000)	278	1228
reflections collected	16,583	70,905
independent reflections (Rint)	3565 (0.0642)	14,688 (0.0504)
parameters, restraints	205, 97	740, 7
GoF on F2	1.069	1.015
R1 [I > 2σ(I)]	0.1062	0.0489
wR2 (all data)	0.3327	0.1150
largest diff. peak/hole [e/Å3]	0.819/−0.756	0.980/−0.318
Flack parameter	−	−0.012(7)

.