Functionalized Tris(anilido)triazacyclononanes as Hexadentate Ligands for the Encapsulation of U(III), U(IV) and La(III) Cations

Abstract

Tripodal multidentate ligands have become increasingly popular in f-element chemistry for stabilizing unusual bonding motifs and supporting small molecule activation processes. The steric and electronic effects of ligand donor atom substituents have proved crucial in both of these applications. In this study we functionalized the previously reported tris-anilide ligand {tacn(SiMe₂NPh)₃} (tacn = 1,3,7-triazacyclononane) to incorporate substituted aromatic rings, with the aim of modifying f-element complex solubility and ligand steric effects. We report the synthesis of two proligands, {tacn(SiMe₂NHAr)₃} (Ar = C₆H₃Me₂-3,5 or C₆H₄Me-4), and their respective group 1 transfer agents—{tacn(SiMe₂NKAr)₃}, M(III) complexes [M{tacn(SiMe₂NAr)₃}] for M = La and U, and U(IV) complexes [M{tacn(SiMe₂NAr)₃}(Cl)]. These compounds were characterized by multinuclear NMR and FTIR spectroscopy and elemental analysis. The paramagnetic uranium complexes were also characterized by solid state magnetic measurements and UV/Vis/NIR spectroscopy. U(III) complexes were additionally studied by EPR spectroscopy. The solid state structures of all f-block complexes were authenticated by single-crystal X-ray diffraction (XRD), together with a minor byproduct [U{tacn(SiMe₂NC₆H₄Me-4)₃}(I)]. Comparisons of the characterization data of our f-element complexes with similar literature examples containing the {tacn(SiMe₂NPh)₃} ligand set showed minor changes in physicochemical properties resulting from the different aromatic ring substitution patterns we investigated.

1. Introduction

Coordinatively unsaturated metal complexes with well-defined reactive sites are ideal candidates for systematic reactivity studies, and as with the rest of the periodic table, such systems can be accessed for f-elements through the use of judiciously selected ligands [1]. Given the predominantly electrostatic bonding regimes and large ionic radii of f-elements, sterically demanding ligands with hard donor atoms provide the kinetic and electronic stabilization required to give robust complexes [2]. Multidentate and macrocyclic ligands form a privileged subset over their monodentate analogues in f-element coordination chemistry, as they can impart additional thermodynamic stability [1,2]. Tripodal examples, which exhibit approximate C₃ symmetry in solution, have proved particularly effective in controlling the coordination sphere of uranium by encapsulating the metal ion in a single well-defined “steric pocket” along the C₃ axis. Such motifs have supported unusual uranium oxidation states and bonding regimes, and coordinatively unsaturated U(III) centers that support rich small molecule activation chemistry [3,4,5,6,7].

Over the last 25 years, the tripodal ligands that have proved most popular for generating landmark uranium complexes include the anchored tris-aryloxides {N(CH₂OAr)₃} (Ar = substituted aryl), {tacn(CH₂OAr)₃} (tacn = 1,3,7-triazacyclononane) and {Mes(CH₂OAr)₃} (Mes = C₆H₂Me₃-2,4,6); and the tris-amides {N(CH₂CH₂NSiR₃)₃} (SiR₃ = SiMe₃, Si^tBuMe_2, SiⁱPr₃) [3,4,5,6,7]. Several related macrocyclic ligand systems have more recently been applied in f-element chemistry [8,9,10]. f-Element complexes of tripodal ligands that impose approximate C₃ symmetry have been applied in the electrocatalytic reduction of water [11,12,13], and it has been shown that the steric and electronic effects of substituents on donor atoms of these ligands are crucial for dictating the physicochemical properties of resultant complexes. For example, the reaction of CO₂ with [U{tacn(CH₂Ar^tBu,tBuO)₃}] (Ar^tBu,tBuO = 3,5-di-tert-butyl-2-oxybenzyl) gives the dinuclear U(IV) bridging oxo complex [{U[tacn(CH₂Ar^tBu,tBuO)₃]}₂(µ-O)] [14], whereas the more sterically demanding [U{tacn(CH₂Ar^Ad,AdO)₃}] (Ar^Ad,AdO = 3-adamantyl-5-tert-butyl-2-oxybenzyl) reacts with CO₂ to yield the remarkable terminal CO₂ U(IV) complex [U{tacn(Ar^Ad,AdO)₃}(CO₂)] [15]. Those examples showcase how the outcomes of small molecule activation reactions for tripodal U(III) complexes are dependent upon the size of the axial reactivity pocket [3,4,5,6,7].

By contrast, there have only been a handful of reports to date of the synthesis and physicochemical properties of uranium complexes of the related tris-anilido ligand {tacn(SiMe₂NPh)₃} [16,17,18,19,20]. However, it has already been shown that the SiMe₂ linkers engender considerable flexibility compared to the more rigid substituted aryl groups in [U{tacn(CH₂ArO)₃}] [3,4,5,6,7]. For example, the nearly trigonal prismatic arrangement of N-donor atoms in the U(III) complex [U{tacn(SiMe₂NPh)₃}] [16] is lost upon oxidation by elemental sulfur to give the distorted bicapped trigonal bipyramidal dinuclear U(IV) complex [{U[tacn(SiMe₂NPh)₃]}₂(µ-S)] [17], whilst approximate C₃ symmetry is typically retained when the U(III) complexes [U{tacn(CH₂ArO)₃}] are converted to U(IV/V/VI) products [3,4,5,6,7].

Herein we report the synthesis of two novel tris-anilido ligands with substituted aryl rings, {tacn(SiMe₂Nar)₃} (Ar = C₆H₃Me₂-3,5; C₆H₄Me-4). We utilized group 1 transfer agents of these ligands to synthesize U(III), La(III) and U(IV) complexes; and we characterized these complexes by a variety of techniques in order to compare these data with respective literature examples of lanthanum and uranium complexes of the unsubstituted ligand {tacn(SiMe₂NPh)₃}.

2. Results

2.1. Synthesis

The ligand precursors {tacn(SiMe₂NHAr)₃} (Ar = C₆H₄Me-4, 1-H₃; C₆H₃Me₂-3,5, 2-H₃) were prepared by modifications of the reported synthesis of {tacn(SiMe₂NHPh)₃} [21] (Scheme 1). The substituted anilines NH₂C₆H₄Me-4 and NH₂C₆H₃Me₂-3,5 were separately reacted with excess NaH in THF to provide the sodium anilides NaNHC₆H₄Me-4 and NaNHC₆H₃Me₂-3,5. These salts were filtered from the remaining NaH and were separately added dropwise to toluene solutions of {tacn(SiMe₂Cl)₃} [21] to give 1-H₃ and 2-H₃, respectively, as pale yellow oils in very good yields after removal of volatiles under reduced pressure. The proligands were both reacted with excess KH in THF without further purification to afford the group 1 transfer agents {tacn(SiMe₂NKC₆H₄Me-4)₃}·2THF (1-K₃) and {tacn(SiMe₂NKC₆H₃Me₂-3,5)₃}·THF (2-K₃), respectively, in fair–good yields as off-white powders (Scheme 1). Despite repeated attempts, crystals of 1-K₃ and 2-K₃ suitable for analysis by single-crystal X-ray diffraction could not be obtained, but elemental analysis values obtained from powders dried in vacuo for 1 h indicated that bound THF molecules remained, as is the case for the Li and Na salts of {tacn(SiMe₂NPh)₃} [21].

With 1-K₃ and 2-K₃ in hand, we adapted the syntheses of [La{tacn(SiMe₂NPh)₃}(THF)] and [U{tacn(SiMe₂NPh)₃}] [16] to perform separate salt metathesis reactions with [La(I)₃(THF)₄] [22] and [U(I)₃(THF)₄] [23] in THF to give [M{tacn(SiMe₂NC₆H₄Me-4)₃}] (M = La, 3-La, U, 3-U) and [M{tacn(SiMe₂NC₆H₃Me₂-3,5)₃}] (M = La, 4-La, U, 4-U) in fair yields following work-up and recrystallization (Scheme 2). We first performed the lanthanum reactions, as these have no radiological hazard and La(III) is diamagnetic, providing NMR spectra that are often more easily interpreted than 5f³ U(III) [3,4,5,6,7]. Additionally, La(III) is a reasonable surrogate for U(III) due to their comparable ionic radii [6-coordinate, La(III) = 1.032 Å, U(III) = 1.025 Å] [24]. It is noteworthy that on one occasion during the synthesis of 3-U, we observed several crystals of the U(IV) complex [U{tacn(SiMe₂NC₆H₄Me-4)₃}(I)] (5), which we attribute to the presence of traces of UI₄, but the low yield precluded solution phase characterization (see below). Instead of optimizing conditions for the synthesis of 5, we decided to target analogous U(IV) chloride complexes, as straightforward salt metathesis routes to [U{tacn(SiMe₂NPh)₃}(Cl)] have been reported previously [10]. By adapting these procedures, the separate reactions of UCl₄ [25,26] with 1-K₃ and 2-K₃ gave the U(IV) complexes [U{tacn(SiMe₂NC₆H₄Me-4)₃}(Cl)] (6) and [U{tacn(SiMe₂NC₆H₃Me₂-3,5)₃}(Cl)] (7) in poor yields, following work-up and recrystallization from THF or DME, respectively (Scheme 2). We found that 6 and 7 could also be synthesized by the respective oxidation of 3-U or 4-U with excess ^tBuCl. This oxidative route gave improved yields of 6 and 7 over salt metathesis from UCl₄. We anticipate that the U(III) complexes 3-U and 4-U could give rich reactivity with small molecules and unsaturated organic compounds, as previously seen for similar tripodal U(III) complexes [3,4,5,6,7]; however, we were unable to isolate any uranium-containing products from the separate 1:1 reactions of 4-U with either benzophenone or 4,4′-bipyridine in toluene. Since [U{tacn(SiMe₂NPh)₃}] has been shown to react with elemental sulfur [17], and the 2e^– oxidation reactions of other tripodal U(III) complexes have been shown to give U(V) products [3,4,5,6,7], we attempted the synthesis of terminal U(V) oxo complexes by the separate reactions of 3-U with pyridine N-oxide in toluene and 4-U with 4-methyl morpholine N-oxide in THF; no products could be identified from either of these reaction mixtures.

2.2. NMR and IR Spectroscopy

NMR spectra are presented in the ESI Figures S13–S42. The ¹H and ¹³C{¹H} NMR spectra of 1-H₃ in C₆D₆ indicate that approximate C₃ symmetry was present on the NMR timescale at 298 K. One signal was observed for each proton environment in the ¹H NMR spectrum apart from the para-Me resonances, where two signals were observed in a 3:6 ratio (δ_H: 2.11 and 2.20 ppm). The three NH protons were obscured by the broad multiplets of the tacn protons (δ_H: 2.98–3.19 ppm). Similar features were observed for 2-H₃, though with the meta-Me resonances in a 12:6 ratio (δ_H: 2.23 and 2.24 ppm) in the ¹H NMR spectrum. In both 1-H₃ and 2-H₃ there are also more Ar-H resonances than would be expected from ideal C₃ symmetry, and this deviation is tentatively attributed to conformational rigidity of the tacn ring in solution rendering the CH₂ protons diastereotopic. The low solubility of 1-K₃ and 2-K₃ in benzene and toluene necessitated the employment of deuterated THF for NMR spectroscopy, precluding accurate integration of the proton THF resonances to validate elemental analysis data. The addition of more polar solvents to 1-K₃ and 2-K₃, such as pyridine, lead to decomposition. The doubling of the number of resonances in the ¹³C and ²⁹Si NMR spectra of 1-K₃ (δ_Si: −18.48 and −15.33 ppm) and 2-K₃ (δ_Si: −17.88 and −14.74 ppm) compared to the parent anilines 1-H₃ (δ_Si: −8.02 ppm) and 2-H₃ (δ_Si: −7.97 ppm) indicates that either coordinated THF reduced these systems to approximate C_s symmetry or that two different approximately C₃ symmetric complexes formed in solution. The former interpretation is the most plausible, as coordinating solvents should disfavor oligomerization.

The most noteworthy difference between the ¹H NMR spectra of 3-M and 4-M and their parent proligands is that the tacn methylene group signals split into two multiplets with AA′BB′ patterns and a relative intensity of 1:1 upon complexation. This pattern of signals is due to a significant deviation from ideal C₃ symmetry and fluxional behavior, and was previously seen for [M{tacn(SiMe₂NPh)₃}] (M = Y, Eu, Yb, U) and [La{tacn(SiMe₂NPh)₃}(THF)] by Marques and co-workers [16]. Following variable temperature ¹H NMR spectroscopy experiments on those literature complexes, the authors postulated that the two N₃ vertices of an approximate trigonal prism twist with respect to each other with a concomitant inversion of the tacn chelate rings; thus, we infer that similar processes are in operation for 3-M and 4-M. As expected, the ¹H NMR spectra of 3-U and 4-U are paramagnetically shifted, with signals from −30 to +10 ppm that show significant line broadening [27]. The ¹H NMR spectra of 6 and 7 displayed similar features to those of 3-U and 4-U, but the signals were broadened to a greater extent, with some FWHM values being as high as 1200 Hz for 6, and the aromatic ring protons of 7 were not observed. The bulk features of these spectra are similar to those previously seen for [U{tacn(SiMe₂NPh)₃}(Cl)] [16]. Only one resonance was seen for the methylene groups in the ¹³C{¹H} NMR spectra of 3-La and 4-La, but no signals could be observed in the ¹³C{¹H} NMR spectra of 3-U, 4-U, 6 and 7 due to paramagnetism. The ²⁹Si{¹H} NMR spectra of 3-La (δ_Si: −5.10 ppm) and 4-La (δ_Si: −5.18 ppm) are deshielded compared to 1-H₃ and 2-H₃; whilst no signals were seen for the U(IV) complexes 6 and 7, the U(III) complexes exhibited remarkably shielded resonances (δ_Si: −263.80 ppm, 3-U; −270.90 ppm, 4-U) that to the best of our knowledge are the most negative values reported for any U(III) complex to date [28].

The FTIR spectra of all complexes containing {tacn(SiMe₂NAr)₃} scaffolds exhibit strong absorptions at $\tilde{\nu }$ ~ 1600 cm⁻¹ due to the substituted anilide groups, as has been seen by other authors for f-block complexes containing {tacn(SiMe₂Nar)₃} [16,17,18,19,20]. In addition, 1-H₃ ($\tilde{\nu }$ = 3406 cm⁻¹) and 2-H₃ ($\tilde{\nu }$ = 3404 and 3377 cm⁻¹) exhibit N–H stretching absorptions, as does {tacn(SiMe₂HNHPh)₃} ($\tilde{\nu }$ = 3381 cm⁻¹) [16].

2.3. NIR/Vis/UV Spectroscopy

The electronic absorption spectra of 3-U, 4-U, 6 and 7 were recorded as 0.5 mM toluene solutions (Figure 1). The similar spectra of the pale green U(IV) complexes 6 and 7 are dominated by intense charge transfer bands in the UV region that tail into the visible region, and are otherwise essentially featureless. The charge transfer bands are less prominent at higher energy for red 3-U and 4-U solutions, as they both exhibit a series of strong absorptions with shoulders in the visible region ($\tilde{\nu }$_max, cm⁻¹ (ε, M⁻¹ cm⁻¹): ~17,000 (~628, 3-U; ~692, 4-U); ~20,000 (~1,444, 3-U; ~1,758, 4-U); ~23,000 (~1,308, 3-U; ~1,612, 4-U)) that are mainly assigned as 5f²6d¹ ← 5f³ transitions, consistent with U(III) centers [6]. In the NIR region, 3-U and 4-U exhibit nearly identical series of weak absorptions (ε < 250 M⁻¹ cm⁻¹) that are assigned as 5f ← 5f transitions with intensities that are typical of U(III) complexes [29].

2.4. Structural Characterization

The molecular structures of 3-M·0.5C₇H₈, 4-M, 5, 6·THF and 7 were determined by single-crystal X-ray diffraction (selected bond lengths and angles are listed in

Table 1. Selected bond lengths (Å) and angles (°) for 3-M·0.5C₇H₈, 4-M, 5, 6·THF and 7 (M = La, U; X = Cl, I). Symmetry operations to generate equivalent atoms: i 1–x, +y, +z; ii = +x, ½–y, +z.

Complex	M–Namine	M–Nanilide	N–M–Namine	N–M–Nanilide	M–X
3-La·0.5C7H8	2.705(3)	2.395(3)	66.21(8)	120.06(9)
	2.705(3)	2.414(3)	68.00(9)	113.46(8)	-
	2.654(3)	2.436(3)	67.54(7)	115.46(10)
3-U·0.5C7H8	2.637(9)	2.381(8)	68.2(2)	119.0(3)
	2.659(8)	2.364(7)	67.4(3)	116.6(3)	-
	2.658(6)	2.396(10)	67.5(3)	110.8(3)
4-La	2.674(2)	2.421(3)	67.80(7)	115.74(9)
	2.676(3)	2.426(2)	67.09(8)	116.64(9)	-
	2.707(3)	2.400(3)	68.02(8)	115.99(8)
4-U	2.616(12)	2.348(10)	68.2(4)	117.9(6)
	2.646(10)	2.31(2)	68.3(4)	115.6(4)	-
	2.63(2)	2.418(14)	68.6(5)	112.9(5)
5	2.74(4)	2.36(2)	66.0(1)	96.6(4)
	2.67(2)	2.32(2)	64.8.(8)	154.5(7)	3.143(2)
	2.59(4)	2.32(2) i	67.7(9)	96.6(4) i
6·THF	2.659(8)	2.285(7)	66.6(3)	95.4(3)
	2.683(7)	2.295(8)	67.1(2)	160.3(3)	2.706(3)
	2.635(7)	2.327(8)	66.0(2)	94.1(3)
7	2.69(2)	2.305(5)	66.8(5)	95.77(10)
	2.667(7)	2.286(5)	67.8(5)	158.2(3)(10)	2.687(2)
	2.62(2)	2.305(5) ii	65.4(6)	95.77(10) ii

, and supporting crystallographic information is compiled in the ESI Tables S1 and S2). The structures of 3-U·0.5C₇H₈, 4-U, 6·THF and 7 are shown in Figure 2 and Figure 3. As 3-La·0.5C₇H₈ and 4-La are structurally analogous with 3-U·0.5C₇H₈ and 4-U, respectively, these are discussed together with U(III) congeners for brevity; and their structures are depicted in the ESI together with 5, which shows similar features to 6·THF (Figures S1–S3).

There are two independent complex molecules in the unit cells of 3-M·0.5C₇H₈. This was seen previously for [U{tacn(SiMe₂NPh)₃}]·0.5C₇H₈ and was attributed to the presence of two isomers [16]. As the metrical parameters of the isomers are similar, we describe one of the molecules of 3-M here. In each case, the metal centers exhibit approximate trigonal prismatic geometries defined by approximately parallel and near-eclipsed amine (N₁-N₃) and anilide (N₄-N₆) planes, and lie much closer to the anilide [M···N₄-N₆ (Å): 0.468(2) (3-La); 0.513(4) (3-U)] than the amine [M···N₁-N₃ (Å): 2.064(2) (3-La); 2.030(4) (3-U)] planes. Despite differences in 6-coordinate ionic radii (La(III) = 1.032 Å, U(III) = 1.025 Å) [18], significant disorder in the dataset for 3-U led to poor resolution of M–N distances, to the extent that these were within statistical error of the corresponding distances in 3-La. The bulk features of 4-M are similar to those of their 3-M analogs; for example, in 4-U the position of the uranium center within the N₆ prism [U···N₄-N₆: 0.509(8) Å; U···N₁-N₃: 2.002(8) Å; mean U–N_amine: 2.63(2) Å; U–N_anilide: 2.36(2) Å] is almost identical to those seen for [U{tacn(SiMe₂NPh)₃}] [16] and 3-U (see Figure S4 for space-filling representations of [U{tacn(SiMe₂NPh)₃}], 3-U and 4-U viewed along the approximate C₃ axes above the N-aryl rings).

Although 5–7 all crystallize in different space groups and 6·THF additionally incorporates a molecule of THF in the unit cell, the metrical parameters of 5–7 are similar to those of [U{tacn(SiMe₂NPh)₃}(Cl)]. Whilst [U{tacn(SiMe₂NPh)₃}(I)] was also synthesized previously by oxidation of [U{tacn(SiMe₂NPh)₃}] with iodine, its solid state structure was not reported [16]. As with [U{tacn(SiMe₂NPh)₃}(Cl)] [16], the positions of the donor atoms in 5–7 can be described as bicapped trigonal bipyramids with one anilide and two amine donors forming the equatorial plane; the halide and remaining amine at the axial positions [N(1)–U(1)–X(1): 169.7(5)° (5); 171.7(2)° (6); 171.7(2)° (7)]; and the other two anilides capping two faces near the halide apex. The uranium centers are situated above the N₃-tacn plane [U···N₁-N₃ (Å): 1.04(2) (5); 1.021(5) (6); 1.011(6) (7)] such that for 6 and 7 the apical U–N distances [2.683(7) Å (6); 2.667(9) Å (7)] are similar to the U–Cl distances [2.706(3) Å (6); 2.687(2) Å (7)].

2.5. Magnetism and EPR Spectroscopy

Solution magnetic susceptibility measurements of U(III) and U(IV) complexes, measured in C₆D₆ at 298 K using the Evans method [30], gave similar χT values: 3-U (1.05 cm³ K mol⁻¹), 4-U (0.95 cm³ K mol⁻¹), 6 (0.98 cm³ K mol⁻¹) and 7 (1.13 cm³ K mol⁻¹). Whilst these molar susceptibilities are all lower than the free ion values [U(III) 5f^{3 4}I_9/2 = 1.70 cm³ K mol⁻¹; U(IV) 5f^{2 3}H₄ = 1.60 cm³ K mol⁻¹], this was expected, as not all crystal field levels are thermally occupied at this temperature [31]. Indeed, the values obtained are around the middle of the previously reported ranges for monometallic U(III) (0.38–1.81 cm³ K mol⁻¹) and U(IV) (0.48–1.34 cm³ K mol⁻¹) complexes [32]. Variable temperature solid state magnetic susceptibility measurements for powdered samples of 3-U, 4-U, 6 and 7 suspended in eicosane were recorded using a SQUID from 300–2 K (ESI Figures S5–S9). At 300 K, the magnetic susceptibility temperature products [χT (cm³ K mol⁻¹): 0.84 (3-U); 0.70 (4-U); 0.60 (6); 0.80 (7)] were consistently lower than solution values by 0.2–0.4 cm³ K mol⁻¹, but such discrepancies should be expected from changes in phase, together with variable errors associated with sample masses and diamagnetic corrections. At 298 K, the solid state χT value for 3-U is similar to that previously reported for [U{tacn(SiMe₂NPh)₃}] (0.94 cm³ K mol⁻¹) [19]. U(IV) complexes 6 and 7 showed χT temperature dependence that correlates with metal oxidation state, tending towards zero at low temperatures due to a non-magnetic singlet ground state [χT (cm³ K mol⁻¹) at 2 K: 0.07 (6); 0.10 (7)]. The U(III) complexes, 3-U and 4-U, reached values that are higher than those of 6 and 7 [χT (cm³ K mol⁻¹) at 2 K: 0.20 (3-U); 0.15 (4-U)], but did not reach their low temperature limit, which suggests occupation of low lying excited states [33,34,35]. This is consistent with the magnetization data (see below).

The U(III) formulations of 3-U and 4-U were confirmed by X- and Q-band EPR spectroscopy on powdered samples at 5 K (X-band spectrum of 4-U shown in Figure 4a; see ESI Figures S10–S12 for other spectra). The spectra of 4-U are characteristic of an effective spin ½, consistent with a Kramers doublet, with g_eff = 2.80, 2.37 and 1.17. Complex 3-U shows greater rhombicity in the g_eff values, but they are complicated by what appears to be an impurity, and we were unable to model these spectra. As U(III) complexes can exhibit single-molecule magnet (SMM) behavior in small applied magnetic fields [33,34,35], we investigated the dynamic magnetic properties of 3-U and 4-U. Both complexes exhibited hysteresis at 1.8 K, but this was more pronounced for 3-U (Figure 4b) than for 4-U (see ESI Figure S6d); in both cases, magnetic saturation was not achieved at 7 T, the physical limit of the magnetometer. Ac susceptibility measurements were performed on 3-U and 4-U under an applied dc field of 600 G at 1.8 K. Whilst 3-U did not exhibit in-phase or out-of-phase components, 4-U displayed clear frequency dependent behavior (Figure 4c,d). An Arrhenius plot of the data gave an effective energy barrier to magnetic reversal, U_eff, of 16.2 (±0.8) K (see ESI Figure S9).

3. Discussion

Comparisons of the multinuclear NMR spectra of 3-M and 4-M with each other and analogous M(III) {tacn(SiMe₂NPh)₃} complexes [16] indicate that minor variations in anilide substitution patterns do not considerably affect dynamic solution behavior. The solubility and solution stability of complexes also did not appear to vary significantly upon ligand substitution. In the solid state, the single-crystal XRD data show that most of the metrical parameters of 3-U (e.g., mean U–N_amine: 2.651(13) Å; U–N_anilide: 2.38(2) Å) are nearly identical to those reported previously for [U{tacn(SiMe₂NPh)₃}] [U···N₁–N₃: 0.52 Å; U···N₄–N₆: 2.02 Å; mean U–N_amine: 2.66(3) Å; U–N_anilide: 2.35(3) Å] [10]. However, the ranges of N_anilide–U–N_anilide [3-U: 110.8(3)–119.0(3)°; 4-U: 112.9(5)–117.9(6)°] and U–N_anilide–C_ipso [3-U: 115.8(6)–127.0(7)°; 4-U: 117.7(10)–125.9(9)°] angles are greater for 3-U than for 4-U, consistent with nearer-axial EPR spectra of 4-U, and showing that substitution of the N-aryl groups can influence both the shape and size of the apical channel in the solid state. Structural differences are even more pronounced for the La(III) homologs 3-La and 4-La, which do not contain bound THF molecules, in contrast with the 7-coordinate La(III) unsubstituted analog [La{tacn(SiMe₂NPh)₃}(THF)] [16]. This change in coordination number leads to more significant changes in metrical parameters; for example, the mean La–N_amine [2.685(5) Å] and La–N_anilide [2.415(5) Å] distances for 3-La differ markedly from [La{tacn(SiMe₂NPh)₃}(THF)] [mean La–N_amine: 2.751(8) Å; La–N_anilide: 2.453(8) Å] [16].

The main differences between the solid state structures of the U(IV) and U(III) complexes reported herein are that the approximate C₃ symmetry is broken upon oxidation of the metal center, with the seventh coordination site being occupied by a halide ion. Despite the increase in metal oxidation state, the respective mean U–N_anilide and U–N_amine distances of 5–7 were similar to those seen for 3-U and 4-U, which we attribute to the considerable rearrangement of metal coordination spheres. This facile reorganization is a distinguishing feature of the flexible SiMe₂ linkers in {tacn(SiMe₂NAr)₃} frameworks [16,17,18,19,20] compared to the more rigid tethered aryloxides in {tacn(CH₂ArO)₃)}, where oxidation of U(III) starting materials to U(IV/V/VI) products has been shown to proceed with retention of approximate C₃ symmetry [3,4,5,6,7]. As {tacn(SiMe₂NAr)₃} scaffolds readily reorganize to accommodate an additional donor atom, we posit that THF may bind to 3-M and 4-M in solution, but that this molecule is readily displaced upon exposure to vacuum during work-up and recrystallization. A detailed reactivity study would be required to determine if the fluxionality of {tacn(SiMe₂NAr)₃} frameworks in solution reduces the steric effect of anilide substituents compared to analogous R-group variation in more rigid tripodal ligands.

A comparison of the characterization data for the U(III) complexes 3-U, 4-U and [U{tacn(SiMe₂NPh)₃}] [16,19] indicates that minor changes in the anilide substituents can lead to subtle changes in the physicochemical properties of complexes. The UV/Vis/NIR electronic absorption spectrum of 4-U shows more intense absorption intensities than that of 3-U. To the best of our knowledge, the corresponding data for [U{tacn(SiMe₂NPh)₃}] have not been published. Similarly, as we were only able to model the EPR spectra of 4-U (g_eff = 2.80, 2.37 and 1.17) and we could not find literature EPR data for [U{tacn(SiMe₂NPh)₃}], we cannot make a detailed comparison of structurally similar complexes, though we note that g_eff =3.54(5), 2.042(4) and 1.66(5) were determined for [U{tacn(SiMe₂NPh)₃}(OPPh₃)], where the phosphine oxide occupies the apical position [20]. We previously reported the EPR spectra of the planar U(III) tris-amide [U{N(SiBuMe₂)₂}₃] [36]. They gave g_eff (=3.55, 2.97, 0.55) that approach those expected for the |m_J| = 1/2 doublet of a ⁴I_9/2 term (3.65, 3.65, 0.73), which is stabilized by the in-plane crystal field. For 4-U, the uranium center is close to the anilide N₃ plane, which likewise would stabilize the |m_J| = 1/2 state, but the out-of-plane crystal field arising from the three amine donors leads to significant mixing. Assuming three-fold symmetry, the m_J = ±1/2 state can mix with both the ±5/2 and ±7/2 states (assuming a ⁴I_9/2 term) leading to very different g-values. A similar result was concluded for the parent [U{tacn(SiMe₂NPh)₃}], where crystal field calculations gave a very mixed ground state comprising almost equal fractions of |m_J| = 5/2 and 1/2 (EPR spectra were not reported) [19]. Finally, we note that the SMM behavior of U(III) {tacn(SiMe₂NAr)₃} complexes appears to be quite sensitive to anilide substituent variation. No effective barrier to magnetic reversal was seen for [U{tacn(SiMe₂NPh)₃}] [19] and 3-U, whereas U_eff = 16.2 (±0.8) K for 4-U is similar to values reported previously for U(III) complexes [33,34,35]; e.g., [U{tacn(SiMe₂NPh)₃}(OPPh₃)], U_eff = 21.9(7) K [20].

4. Materials and Methods

General Information

Caution—uranium-238 (t_1/2 = 4.47 × 10⁹ years) is a weak α-emitter; therefore, all manipulations should be performed in suitable laboratories that have been designated for radiochemical use, and α-counting equipment should be available. All manipulations were carried out using standard Schlenk and glove box techniques under an atmosphere of dry argon. THF, toluene, n-hexane and DME were dried by refluxing over potassium. All solvents were stored over K mirrors (with the exception of THF and DME, which were stored over activated 4 Å molecular sieves), and were degassed prior to use. Deuterated solvents were dried over K, degassed by three freeze-pump-thaw cycles and stored under Ar. tacn(SiMe₂Cl)₃ [21], [La(I)₃(THF)₄] [22], [U(I)₃(THF)₄] [23] and UCl₄ [25,26] were prepared according to literature procedures, and all other reagents were purchased. p-Toluidine was dried for 4 h under vacuum before use, whilst 3,5-dimethylaniline was refluxed over CaH₂ and distilled. KH and NaH were obtained as suspensions in mineral oil and were washed three times with n-hexane before use.

¹H, ¹³C{¹H} and ²⁹Si{¹H} NMR spectra were recorded on a Bruker Avance III 400 MHz spectrometer operating at 400.1, 100.6 and 79.5 MHz, respectively; chemical shifts are quoted in ppm and are relative to TMS. FTIR spectra were recorded as Nujol mulls in KBr discs on a Perkin Elmer Spectrum RX1 spectrometer (Perkin Elmer, Waltham, MA, USA). Elemental microanalyses were carried out by Stephen Boyer at the Microanalysis Service, London Metropolitan University, or by Martin Jennings and Anne Davies at The University of Manchester. Low carbon values were consistently obtained in elemental analyses of 2-K₃, 3-La and 4-La. We attribute this observation to silicon carbide formation, as <5% protic impurities were observed by ¹H NMR spectroscopy, and we note that low C values were intermittently obtained in microanalysis experiments for f-element complexes of the related ligand {tacn(SiMe₂NPh)₃} [16]. UV-Vis-NIR spectroscopy was performed on samples in Youngs tap appended 10 mm pathlength quartz cuvettes on an Agilent Technologies Cary Series UV-Vis-NIR Spectrophotometer (Agilent Technologies, Santa Clara, CA, USA) from 175–3300 nm. X- and Q-band EPR spectroscopy was performed on powdered samples in quartz tubes at 5 K sealed under vacuum. Magnetic measurements were made using a Quantum Design MPMS-XL7 SQUID magnetometer on ground crystalline samples suspended in eicosane in borosilicate tubes sealed under vacuum.

Crystals of 3-M·0.5C₇H₈, 4-M (M = La, U) and 5–8 were examined on a Rigaku Oxford Diffraction SuperNova CCD area detector diffractometer (Rigaku, Tokyo, Japan) using mirror-monochromated Mo Kα radiation (λ = 0.71073 Å). Intensities were integrated from data recorded on 1° frames by ω rotation. Cell parameters were refined from the observed positions of all strong reflections in each data set. A Gaussian grid face-indexed absorption correction with a beam profile correction was applied [37]. The structures were solved variously by direct and heavy atom methods using SHELXS [38] or SHELXT [39] and were refined by full-matrix least-squares on all unique F² values [38], with anisotropic displacement parameters for all non-hydrogen atoms, and with constrained riding hydrogen geometries; U_iso(H) was set at 1.2 (1.5 for methyl groups) times U_eq of the parent atom. The largest features in final difference syntheses were close to those of heavy atoms and were of no chemical significance. CrysAlisPRO [37] was used for control and integration, SHELX [38] and was employed through OLEX2 [40] for structure solution and refinement and ORTEP-3 [41] and POVRAY [42] were used for molecular graphics.

Synthesis of {tacn(SiMe₂NHC₆H₄Me-4)₃} (1-H₃): p-toluidine (5.36 g, 50.0 mmol) was added to a slurry of NaH (2.16 g, 90.0 mmol) in THF (30 mL) and was heated to 50 °C for 18 h. The resultant brown suspension was allowed to cool to room temperature and filtered. The dark brown supernatant was added dropwise to a solution of tacn(SiMe₂Cl)₃ (6.30 g, 15.5 mmol) in toluene (50 mL) at −78 °C, allowed to slowly warm to room temperature and stirred for 18 h. Volatiles were removed under reduced pressure and the resultant yellow oil was extracted with hexane (3 $\times$ 50 mL). The solvent was removed under reduced pressure to give crude 1-H₃ as a viscous yellow oil (7.56 g, 79%), which was used without further purification. Anal. Calcd. for C₃₃H₅₄N₆Si₃: C, 64.02; H, 8.79; N, 13.58. Found: C, 64.22; H, 8.94; N, 13.44. ¹H NMR (C₆D₆, 298 K): δ = 0.21 (18H, s, Si(CH₃)₂), 2.11 (3H, s, Ar-CH₃), 2.20 (6H, s, Ar-CH₃), 2.98–3.19 (15H, br m, NCH₂ and NH), 6.61 (d, 4H, J_HH = 8.0 Hz, Ar-H), 7.01 (m, 6H, J_HH = 8.0 Hz, Ar-H), 7.13 (2H, br m, J_HH = 8.0 Hz, Ar-H). ¹³C{¹H} NMR (C₆D₆, 298 K): δ = –0.69 (Si(CH₃)₂), 20.99 (Ar-CH₃), 50.96 (NCH₂), 117.25 (p-Ar-CH), 127.25 (Ar-CH), 130.45 (Ar-CH), 145.33 (ipso-Ar-C). ²⁹Si{¹H} NMR (C₆D₆, 298 K): δ = –8.02 (SiMe₂). FTIR (Nujol, $\tilde{\nu }$/cm⁻¹): 3406 (m), 1614 (s), 1512 (w), 1285 (s), 1167 (m), 1148 (m), 1109 (m), 995 (s), 968 (s), 895 (s), 692 (m), 638 (m).

Synthesis of {tacn(SiMe₂NKC₆H₄Me-4)₃}·2THF (1-K₃): 1-H₃ (7.56 g, 12.2 mmol) was dissolved in THF (20 mL) and added dropwise to a suspension of KH (2.79 g, 70 mmol) in THF (10 mL) at −78 °C. The reaction mixture was allowed to warm to room temperature, heated to 50 °C overnight, allowed to cool and filtered. Volatiles were removed under reduced pressure and the resultant solid was washed with hexane (20 mL) and dried in vacuo to yield 1-K₃ as an off-white powder (6.10 g, 62%). Anal. Calcd. for C₄₁H₆₇K₃N₆O₂Si₃: C, 56.11; H, 7.70; N, 9.52. Found: C, 55.95; H, 7.80; N, 9.47. ¹H NMR (C₄D₈O, 298 K): δ = 0.01 (s, 12H, Si(CH₃)₂), 0.11 (s, 6H, Si(CH₃)₂), 1.78 (m, 4H, OCH₂CH₂), 2.05 (s, 9H, Ar-CH₃), 2.72–2.91 (m, 12H, NCH₂), 3.62 (m, 4H, OCH₂CH₂), 6.23 (d, 4H, J_HH = 8.0 Hz, Ar-H), 6.56 (d, 6H, J_HH = 8.0 Hz, Ar-H), 6.63 (d, 2H, J_HH = 8.0 Hz, Ar-H). ¹³C{¹H} NMR (C₄D₈O, 298 K): δ = 0.22 (Si(CH₃)₂), 0.39 (Si(CH₃)₂), 20.74 (Ar-CH₃), 20.83 (Ar-CH₃), 26.55 (OCH₂CH₂), 49.19 (NCH₂), 49.32 (NCH₂), 68.39 (OCH₂CH₂), 117.46 (p-Ar-C), 118.82 (p-Ar-C), 122.23 (Ar-CH), 122.82 (Ar-CH), 130.66 (Ar-CH), 130.90 (Ar-CH), 160.09 (ipso-Ar-C), 160.94 (ipso-Ar-C). ²⁹Si DEPT NMR (C₄D₈O, 298 K): δ = –18.48, –15.33 (SiMe₂). FTIR (Nujol, $\tilde{\nu }$/cm⁻¹): 1595 (s), 1172 (m), 1057 (m), 995 (s), 955 (m), 893 (m), 870 (m), 756 (m), 683 (w), 635 (w).

Synthesis of {tacn(SiMe₂NC₆H₃Me-3,5)₃} (2-H₃): 3,5-dimethylaniline (7.23 g, 59.7 mmol) was added to a slurry of NaH (3.46 g, 144.0 mmol) in THF (30 mL) and heated to 50 °C for 18 h. The resultant brown suspension was allowed to cool to room temperature and filtered. The dark brown supernatant was added dropwise to a solution of tacn(SiMe₂Cl)₃ (7.51 g, 18.5 mmol) in toluene (50 mL) at –78 °C, allowed to slowly warm to room temperature and stirred for 18 h. Volatiles were removed under reduced pressure and the resultant yellow oil was extracted with hexane (3 $\times$ 50 mL). The solvent was removed under reduced pressure to give crude 2-H₃ as a viscous yellow oil, which was used without further purification (10.40 g, 85%). Anal. Calcd. for C₃₆H₆₀N₆Si₃: C, 65.40; H, 9.15; N, 12.71. Found: C, 64.94; H, 9.14; N, 12.38. ¹H NMR (C₆D₆, 298 K): δ = 0.23 (s, 12H, s, Si(CH₃)₂), 0.27 (s, 6H, s, Si(CH₃)₂), 2.23 (s, 12H, Ar-CH₃), 2.24 (6H, s, Ar-CH₃), 3.02–3.25 (15H, br m, NCH₂ and NH), 6.32 (s, 4H, Ar-H), 6.35 (s, 2H, Ar-H), 6.45 (s, 3H, Ar-H). ¹³C{¹H} NMR (C₆D₆, 298 K): δ = 0.68 (Si(CH₃)₂), 22.08 (Ar-CH₃), 51.03 (NCH₂), 115.35 (Ar-CH), 120.62 (Ar-CH), 138.93 (m-Ar-C), 147.70 (ipso-Ar-C). ²⁹Si{¹H} NMR (C₆D₆, 298 K): δ = −7.97 (SiMe₂). FTIR (Nujol, $\tilde{\nu }$/cm⁻¹): 3404 (m), 3377 (m), 1599 (s), 1406 (m), 1325 (m), 11776 (s), 1120 (s), 1109 (m), 1057 (s), 966 (m), 821 (br, s), 691 (s).

Synthesis of {tacn(SiMe₂NKC₆H₃Me₂-3,5)₃}·THF (2-K₃): 2-H₃ (10.40 g, 15.7 mmol) was dissolved in THF (20 mL) and added dropwise to a suspension of KH (2.16 g, 54.0 mmol) in THF (10 mL) at −78 °C. The reaction mixture was allowed to warm to room temperature and heated to 50 °C overnight, allowed to cool and filtered. Volatiles were removed under reduced pressure, and the resultant solid was washed with hexane (20 mL) and dried in vacuo to yield 2-K₃ as an off-white powder (12.50 g, 86%). Anal. Calcd. for C₄₀H₆₅K₃N₆OSi₃: C, 58.69; H, 7.73; N, 9.92. Found: C, 54.96; H, 7.72; N, 9.95. ¹H NMR (C₄D₈O, 298 K): δ = 0.04 (s, 12H, s, Si(CH₃)₂), 0.13 (s, 6H, s, Si(CH₃)₂), 1.78 (m, OCH₂CH₂), 2.00 (s, 12H, Ar-CH₃), 2.06 (6H, s, Ar-CH₃), 2.72-2.93 (12H, br m, NCH₂), 3.62 (m, OCH₂CH₂), 5.70 (s, 2H, Ar-H), 5.78 (s, 1H, p-Ar-H), 6.00 (s, 4H, Ar-H), 6.29 (s, 2H, Ar-H). ¹³C{¹H} NMR (C₄D₈O, 298 K): δ = 0.35 (Si(CH₃)₂), 0.43 (Si(CH₃)₂), 22.31 (Ar-CH₃), 22.36 (Ar-CH₃), 26.55 (OCH₂CH₂), 49.16 (NCH₂), 49.39 (NCH₂), 68.39 (OCH₂CH₂), 112.52 (Ar-CH), 113.67 (Ar-CH), 120.32 (Ar-CH), 120.91 (Ar-CH), 138.34 (m-Ar-C), 138.74 (m-Ar-C), 162.82 (ipso-Ar-C), 163.25 (ipso-Ar-C). ²⁹Si DEPT NMR (C₄D₈O, 298 K): δ = –17.88, –14.74 (SiMe₂). FTIR (Nujol, $\tilde{\nu }$/cm⁻¹): 1557 (s), 1352 (m), 1240 (m), 1194 (m), 1173 (m), 1117 (m), 1057 (m), 978 (w), 916 (m), 893 (m), 843 (m), 760 (m), 654 (m).

Synthesis of [La{tacn(SiMe₂NC₆H₄Me-4)₃}] (3-La): THF (20 mL) was added to a pre-cooled (−78 °C) mixture of [La(I)₃(THF)₄] (0.40 g, 0.5 mmol) and 1-K₃ (0.44 g, 0.5 mmol). The pale yellow reaction mixture was allowed to warm to room temperature and stirred for 16 h. Volatiles were removed under reduced pressure and the residue was extracted with toluene (2 $\times$ 10 mL), reduced in volume to ca. 2 mL and stored at −30 °C overnight to give colorless crystals of 3-La·0.5C₇H₈ (0.13 g, 32%). Anal. Calcd. for C₃₃H₅₁LaN₆Si₃·C₇H₈: C, 56.71; H, 7.02; N, 9.92. Found: C, 51.64; H, 6.93; N, 9.54. ¹H NMR (C₆D₆, 298 K): δ = 0.27 (s, 18H, Si(CH₃)₂), 2.21 (s, 9H, Ar-CH₃), 2.23-2.34 (m, 6H, NCH₂), 2.81-2.89 (m, 6H, NCH₂), 6.64 (d, 6H, J_HH = 8.0 Hz, Ar-H), 6.98 (d, 6H, J_HH = 8.0 Hz, Ar-H). ¹³C{¹H} NMR (C₆D₆, 298 K): δ = 0.80 (Si(CH₃)₂), 21.13 (Ar-CH₃), 48.65 (NCH₂), 121.04 (Ar-CH), 126.14 (p-Ar-C), 131.33 (Ar-CH), 151.10 (ipso-Ar-C). ²⁹Si{¹H} NMR (C₆D₆, 298 K): δ = –5.10 (s, SiMe₂). FTIR (Nujol, $\tilde{\nu }$/cm⁻¹): 1606 (s), 1503 (s), 1297 (m), 1178 (w), 1041 (m), 1017 (m), 958 (m), 894 (m), 816 (s), 771 (s).

Synthesis of [U{tacn(SiMe₂NC₆H₄Me-4)₃}] (3-U) and [U{tacn(SiMe₂NC₆H₄Me-4)₃}(I)] (5): THF (20 mL) was added to a pre-cooled (−78 °C) mixture of [U(I)₃(THF)₄] (0.45 g, 0.5 mmol) and 1-K₃ (0.44 g, 0.5 mmol). The dark red reaction mixture was allowed to warm to room temperature and stirred for 16 h. The volatiles were removed in vacuo and the residue was extracted with toluene (2 $\times$10 mL) and filtered. The solution was reduced in volume to ca. 2 mL and stored at −30 °C overnight to give dark red crystals of 3-U∙0.5C₇H₈ (0.25 g, 56%). On one occasion, several crystals of 5 formed. The solid state structure of 5 was determined, but no other characterization data could be obtained due to the low yield. Data for 3-U: Anal. Calcd. for C₃₃H₅₁N₆Si₃U·0.5C₇H₈: C, 48.70; H, 6.16; N, 9.34. Found: C, 48.53; H, 6.09; N, 9.47. µ_eff = 2.97 µ_B (Evans method). ¹H NMR (C₆D₆, 298 K): δ = −30.29 (br, ν½ ~ 150 Hz, 6H, NCH₂), −7.50 (br, ν½ ~ 150 Hz, 6H, NCH₂), −0.65 (br, ν½ ~ 20 Hz, 6H, Ar-CH), −0.40 (br, ν½ ~ 30 Hz, 9H, Ar-CH₃), 2.67 (br, ν½ ~ 30 Hz, 6H, Ar-CH), 7.74 (br, ν½ ~ 50 Hz, 18H, Si(CH₃)₂). ¹³C{¹H} NMR (C₆D₆, 298 K): not observed. ²⁹Si{¹H} NMR (C₆D₆, 298 K): δ = −263.80 (SiMe₂). FTIR (Nujol, $\tilde{\nu }$/cm⁻¹): 1605 (s), 1506 (m), 1290 (s), 1260 (m), 1177 (m), 1034 (m), 951 (s), 912 (w), 895 (m), 862 (m), 816 (s), 770 (s).

Synthesis of [La{tacn(SiMe₂NC₆H₃Me₂-3,5)₃}] (4-La): THF (20 mL) was added to a pre-cooled (−78 °C) mixture of [La(I)₃(THF)₄] (0.40 g, 0.5 mmol) and 2-K₃ (0.46 g, 0.5 mmol). The pale yellow reaction mixture was allowed to warm to room temperature and stirred for 16 h. Volatiles were removed under reduced pressure and the residue was extracted with toluene (2 $\times$ 10 mL), reduced in volume to ca. 1 mL and stored at −30 °C overnight to give colorless crystals of 4-La (0.27 g, 68%). Anal. Calcd. for C₃₆H₅₇LaN₆Si₃: C, 54.25; H, 7.21; N, 10.54. Found: C, 53.06; H, 7.23; N, 10.03. ¹H NMR (C₆D₆, 298 K): δ = 0.30 (s, 18H, s, Si(CH₃)₂), 2.13 (s, 18H, Ar-CH₃), 2.23-2.33 (br m, 6H, NCH₂), 2.79-2.87 (br m, 6H, NCH₂), 6.39 (s, 6H, o-Ar-H), 6.42 (s, 3H, p-Ar-H). ¹³C{¹H} NMR (C₆D₆, 298 K): δ = 0.82 (Si(CH₃)₂), 21.84 (Ar-CH₃), 48.66 (NCH₂), 118.08 (Ar-CH), 119.95 (Ar-CH), 140.01 (m-Ar-C), 153.33 (ipso-Ar-C). ²⁹Si{¹H} NMR (C₆D₆, 298 K): δ = –5.18 (s, SiMe₂). FTIR (Nujol, $\tilde{\nu }$/cm⁻¹): 1585 (s), 1350 (m), 1246 (m), 1196 (s), 1070 (m), 1041 (m), 966 (m), 988 (m), 896 (s), 866 (s), 769 (s).

Synthesis of [U{tacn(SiMe₂NC₆H₃Me-3,5)₃}] (4-U): THF (20 mL) was added to a pre-cooled (−78 °C) mixture of [U(I)₃(THF)₄] (0.45 g, 0.5 mmol) and 2-K₃ (0.46 g, 0.5 mmol). The dark red reaction mixture was allowed to warm to room temperature and stirred for 16 h. Volatiles were removed under reduced pressure and the residue was extracted with toluene (2 $\times$ 10 mL) and filtered. The solution was reduced in volume to ca. 2 mL and stored at −30 °C overnight to give dark red crystals of 4-U (0.29 g, 65%). Anal. Calcd. for C₃₆H₅₇N₆Si₃U: C, 48.25; H, 6.64; N, 9.38. Found: C, 47.74; H, 5.96; N, 9.53. µ_eff = 2.61 µ_B (Evans method). ¹H NMR (C₆D₆, 298 K): δ = −30.84 (br, ν½ ~ 60 Hz, 6H, NCH₂), −9.53 (br, ν½ ~ 60 Hz, 6H, NCH₂), –3.64 (br, ν½ ~ 20 Hz, 18H, Ar-CH₃), −0.75 (br, ν½ ~ 40 Hz, 6H, o-Ar-H), 0.51 (s, 3H, p-Ar-H), 8.09 (br, ν½ ~ 40 Hz, 18H, Si(CH₃)₂). ¹³C{¹H} NMR (C₆D₆, 298 K): not observed. ²⁹Si{¹H} NMR (C₆D₆, 298 K): δ = −270.90 (SiMe₂). FTIR (Nujol, $\tilde{\nu }$/cm⁻¹): 1584 (s), 1344 (s), 1289 (w), 1069 (m), 1038 (m), 989 (w), 966 (m), 912 (m), 897 (s), 866 (s), 772 (s), 673 (m).

Synthesis of [U{tacn(SiMe₂NC₆H₃Me-4)₃}(Cl)] (6): Method 1: THF (20 mL) was added to a pre-cooled (−78 °C) mixture of UCl₄ (0.38 g, 1 mmol) and 1-K₃ (0.88 g, 1 mmol). The dark green reaction mixture was allowed to warm to room temperature and stirred for 16 h. Volatiles were removed in vacuo and the residue was extracted with THF (2 $\times$ 30 mL) and filtered. The solution was reduced in volume to ca. 2 mL and stored at −30 °C overnight to give green crystals of 6·THF (0.16 g, 16%). Method 2: An excess of ^tBuCl (0.023 g, 0.25 mmol) was added to a pre-cooled (−10 °C) solution of 3-U (0.050 g, 0.056 mmol) in toluene. Upon warming to room temperature, a color change to green was observed. Volatiles were removed under reduced pressure to yield 6 in quantitative crude yield as a green solid. Anal. Calcd. for C₃₃H₅₁ClN₆Si₃U: C, 44.56; H, 5.78; N, 9.45. Found: C, 44.56; H, 5.73; N, 9.33. µ_eff = 2.76 µ_B (Evans method). ¹H NMR (C₄D₈O, 298 K): δ = −46.31 (br, ν½ ~ 70 Hz, 6H, Ar-CH), −15.17 (br, ν½ ~ 1,200 Hz, 6H, NCH₂), 5.86 (br, ν½ ~ 100 Hz, 9H, Ar-CH₃), 12.34 (br, ν½ ~ 150 Hz, 6H, Ar-CH), 16.08 (br, ν½ ~ 600 Hz, 18H, Si(CH₃)₂), 23.82 (br, ν½ ~ 1,000 Hz, 6H, NCH₂). ¹³C{¹H} and ²⁹Si{¹H} NMR (C₄D₈O, 298 K): not observed. FTIR (Nujol, $\tilde{\nu }$/cm⁻¹): 1605 (s), 1501 (m), 1247 (m), 1176 (w), 1109 (w), 1085 (m), 1045 (s), 1013 (s), 945 (m), 921 (m), 905 (m), 831 (s), 776 (m), 730 (s), 708 (m).

Synthesis of [U{tacn(SiMe₂NC₆H₃Me₂-3,5)₃}(Cl)] (7): Method 1: THF (20 mL) was added to a pre-cooled (−78 °C) mixture of UCl₄ (0.38 g, 0.5 mmol) and 2-K₃ (0.92 g, 1 mmol). The dark green reaction mixture was allowed to warm to room temperature and stirred for 16 h. Volatiles were removed in vacuo and the residue was extracted with DME (2 $\times$ 20 mL) and filtered. The solution was reduced in volume to ca. 2 mL and stored at −30 °C overnight to give green crystals of 7 (0.09 g, 10%). Method 2: An excess of ^tBuCl (0.079 g, 0.85 mmol) was added to a pre-cooled (−10 °C) solution of 4-U (0.150 g, 0.16 mmol) in toluene. Upon warming to room temperature, a color change to green was observed. Volatiles were removed under reduced pressure and the green solid was dissolved in DME (1 mL) and stored −30 °C for 8 h to give green needles of 7 (0.063 g, 42%). Anal. Calcd. for C₃₆ClH₅₇N₆Si₃U: C, 46.41; H, 6.17; N, 9.02. Found: C, 46.34; H, 6.30; N, 8.93. µ_eff = 3.20 µ_B (Evans method). ¹H NMR (C₆D₆, 298 K): δ = −50.42 (br, ν½ ~ 200 Hz, 2H, Ar-CH), 1.15 (br, ν½ ~ 300 Hz, 18H, Si(CH₃)₂), 12.67 (br, ν½ ~ 1,000 Hz, 12H, NCH₂), most Ar-CH and Ar-CH₃ not observed. ¹³C{¹H} and ²⁹Si{¹H} NMR (C₆D₆, 298 K): not observed. FTIR (Nujol, $\tilde{\nu }$/cm⁻¹): 1587 (s), 1317 (s), 1184 (s), 1042 (m), 962 (w), 920 (m), 903 (m), 889 (m), 679 (m).