Homoleptic organolanthanide compounds supported by the bis(dimethylsilyl)benzyl ligand

Abstract

A β-SiH functionalized benzyl anion [C(SiHMe₂)₂Ph]⁻ is obtained by deprotonation of HC(SiHMe₂)₂Ph with KCH₂Ph or by reaction of KOtBu and (Me₂HSi)₃CPh; LnI₃(THF)_n and three equivalents of this carbanion combine to provide homoleptic tris(alkyl)lanthanide compounds Ln{C(SiHMe₂)₂Ph}₃ (Ln = La, Ce, Pr, Nd) containing secondary metal–ligand interactions.

---

Synthesis of homoleptic organolanthanide complexes, particularly those of the early trivalent lanthanides (La–Nd), is challenging due to the large radii of these elements, polar bonding, high charge, and high Lewis acidity.¹ Such homoleptic compounds should be valuable for the synthesis of new catalysts and new materials,² yet solvent- or donor-group-free, salt-free, and thermally robust organolanthanide compounds are not readily accessed for the larger metal centers. For example, the reaction of MeLi and LaCl₃ gives Li₃[LaMe₆] as a TMEDA adduct.³ Three THF molecules coordinate to the labile tris(benzyl)lanthanum allowing isolation of LaBn₃(THF)₃,^4,5 however, even this adduct eliminates toluene at room temperature with a half-life of ca. 2 h. The persistence of related compounds may be enhanced by chelating benzylic ligands, for example Ln(CH(NMe₂)Ph)₃⁶ or Ln(CH₂C₆H₄-2-NMe₂)₃.⁷ An alternative approach combines bulky β-SiMe₃ with the benzyl group in C(SiMe₃)₂Ph,⁸ exemplified by a bis(alkyl)calcium compound possessing metal–aryl π-interactions. Coordinative unsaturation is important to the reactivity of organolanthanides, and donors such as TMEDA or THF can diminish reactivity,⁹ facilitate alkane elimination,¹⁰ or react by C–O bond cleavage.¹¹ While donor-free lanthanum and cerium compounds such as Ln{CH(SiMe₃)₂}₃ are known, they are inconveniently accessed through multistep synthesis via Ln{O(2,6-C₆H₃tBu₂)}₃.^12,13

A strategy for stabilizing coordinatively unsaturated rare earth amides has involved the incorporation of SiH groups, which form labile secondary interactions with the lanthanide center.¹⁴ Furthermore, the SiH moiety provides a powerful signature in ¹H and ²⁹Si NMR and IR spectra. This β-SiH strategy may also be applied to alkyls, and the ligand C(SiHMe₂)₃ supports trivalent yttrium and divalent ytterbium and samarium homoleptic alkyls containing secondary Ln↼H–Si interactions.^15,16 Recently, we reported Ce{C(SiHMe₂)₃}₃ as a precursor to a zwitterionic hydrosilylation catalyst.¹⁷ New chemistry might be accessed with alkyl ligand variations that include both β-SiH and benzylic functionalities, and these groups could compete to enhance the homoleptic compounds' resistance to undesired ligand elimination pathways. Both SiH and benzyl groups may have significant charge delocalization and secondary interactions that might stabilize homoleptic compounds. A single ligand containing both elements, namely –C(SiHMe₂)₂Ph, would test these ideas. Here we report the synthesis of alkane precursors, two routes to potassium alkyl reagents, and isolation and characterization of a series of homoleptic organolanthanide complexes.

Reductive coupling of HCPhBr₂ and ClSiHMe₂ affords HC(SiHMe₂)₂Ph (1; eqn (1)) on preparative scale.

A diagnostic triplet in the ¹H NMR spectrum at 1.43 ppm (³J_HH = 4 Hz, 1 H) for the H on the central carbon is coupled to a signal at 4.34 ppm assigned to the SiH (2 H, ¹J_SiH = 186 Hz). The two SiHMe₂ groups are magnetically inequivalent giving a virtual octet for the SiH resonance (a M(AX₃Y₃)(AX₃Y₃)′ spin system). Compound 1 is also characterized by an intense ν_SiH absorbance at 2115 cm⁻¹ in its IR spectrum.

While HC(SiHMe₂)₃¹⁸ reacts readily with lithium diisopropylamide,¹⁹ deprotonation of HC(SiHMe₂)₂Ph is more challenging. Attempts to synthesize [C(SiHMe₂)₂Ph]⁻ using LiN(SiMe₃)₂, nBuLi, KH, or KC(SiHMe₂)₃ as bases returned HC(SiHMe₂)₂Ph. Potassium benzyl (KBn) gives Me₂SiBn₂ as the major product in its reaction with 1 at room temperature. Fortunately, reactions with KBn performed at −78 °C yielded a mixture now dominated by KC(SiHMe₂)₂Ph (2), assigned to a doublet at 0.47 ppm in the ¹H NMR spectrum. This signal was affected by addition of TMEDA, which gave a new doublet at 0.57 ppm. In preparative scale reactions, the desired potassium alkyl is crystallized from pentane at −30 °C to provide Ph(Me₂HSi)₂CK(TMEDA) (2·TMEDA) as dark red crystals (eqn (2)), albeit in low isolated yield.

The ¹H NMR spectrum of isolated 2·TMEDA contained a septet at 4.78 ppm (¹J_SiH = 162 Hz). This one-bond coupling constant was reduced compared to HC(SiHMe₂)₂Ph (186 Hz). In 2·TMEDA, the SiMe₂ groups appeared as one doublet (³J_HH = 3.6 Hz), unlike the diastereotopic methyls in HC(SiHMe₂)₂Ph noted above. In addition, the IR spectrum of 2·TMEDA revealed two ν_SiH bands at 2115 and 1995 cm⁻¹.

A single-crystal X-ray diffraction study revealed a polymeric structure for 2·TMEDA,‡ with each K cation interacting with two C(SiHMe₂)₂Ph groups (Fig. 1) through the H1s (2.82(4) Å), the C6 and C11 (from a phenyl group) of one ligand, and the C9 and C10 from a phenyl of the second. Notably, the K1–C1 distance (3.565(4) Å; i.e., to the presumed carbanionic center) is exceedingly long and outside expected bonding range. For comparison, the K–C distance (3.030(5) Å) is much shorter in dimeric {(Me₂HSi)₃CK(TMEDA)}₂ than in 2·TMEDA.¹⁶ The SiH-free potassium alkyl {KC(SiMe₃)₂Ph}_n (3.007(2) Å)⁸ and the compound {KC(SiMe₂Ph)₂(SiHMe₂)}_n (3.167(8) Å)²⁰ also have polymeric structures with π-coordinated arenes. While the central, carbanionic carbon adopts distorted, nearly planar geometries in these three examples (Σ_angles = 358.8°, 357.4° and 356.5°),^8,16,20 C1 in 2·TMEDA is perfectly planar (Σ_angles = 360.0(5)°).

Fig. 1 Thermal ellipsoid plot of Ph(Me₂HSi)₂CK(TMEDA) (2·TMEDA) at 50% probability. H1s and H2s were located objectively in the Fourier difference map and refined. H atoms bonded to C are not illustrated for clarity. Selected interatomic distances (Å): K1–C1, 3.565(4); K1–H1s, 2.82(4); K1–Si1, 3.844(2); K1–C6, 3.049(4); K1–C7, 3.389(5); K1–C8#, 3.359(5); K1–C9#, 3.085(5); K1–C10#, 3.038(4); K1–C11, 3.091(5); C1–C6, 1.446(6); C6–C7, 1.434(7); C7–C8, 1.376(7); C8–C9, 1.392(7); C9–C10, 1.383(7); C10–C11, 1.374(6); C11–C6, 1.422(6). Selected interatomic angles (°): K1–H1s–Si1, 124(2); C6–C1–Si2, 119.8(3); Si2–C1–Si1, 121.6(2); Si1–C1–C6, 118.6(3).

The formation of Bn₂SiMe₂ in these reactions implies nucleophilic attack by KBn on a SiHMe₂ group and suggests an alternative route to the desired KC(SiHMe₂)₂Ph (2) via Si–C cleavage. A related Si–Si bond cleavage provides MSi(SiMe₃)₃ from Si(SiMe₃)₄ and LiMe or KOtBu.²¹ This idea was tested by the reaction of (Me₂HSi)₃CPh and KOtBu to give the desired KC(SiHMe₂)₂Ph (2) in excellent yield (eqn (3)).

The spectroscopic features of the SiH group in 2 and 2·TMEDA are similar, including the IR stretching frequency, the chemical shift, and the one-bond coupling constant.

Reactions of three equiv. of 2 or 2·TMEDA and LaI₃(THF)₄, CeI₃(THF)₄, PrI₃(THF)₃, or NdI₃(THF)₃ provide Ln{C(SiHMe₂)₂Ph}₃ (Ln = La (3), Ce (4), Pr (5), Nd (6)) in excellent yields (eqn (4)).

The series of compounds provide pale yellow, orange, yellow and green crystalline materials, respectively. IR spectra for 3, 5, and 6 (KBr) each contained a sharp, higher energy ν_SiH band (2109 ± 5 cm⁻¹) assigned to non-bridging SiH and a broad, lower energy band (1866 ± 6 cm⁻¹) attributed to a ν_SiH of the Ln↼H–Si. In contrast, the cerium compound 4 showed only one ν_SiH band, which appeared at 2115 cm⁻¹. The solution IR spectra similarly showed two SiH absorbances for 3, 5 and 6. Although two ν_SiH bands were obtained for 4 in solution, the bands were low intensity and a number of the IR spectra were dominated by HC(SiHMe₂)₂Ph. We attributed these observations to labile secondary Ln↼H–Si interactions present both in solution and solid state.

The ¹H NMR spectrum of diamagnetic 3 revealed signals at 4.24, 0.43, and 0.32 ppm attributed to equivalent SiHMe₂ groups with diastereotopic methyl moieties. A doublet resonance at 5.84 ppm, assigned to an ortho-C₆H₅, appeared upfield compared to its chemical shift in the alkane starting material (6.97 ppm), suggesting a multihapto benzyl-Ln coordination. Evidence for La↼H–Si interactions were provided by the ¹J_SiH of 144 Hz. The equivalence of the SiHMe₂ in the room temperature NMR spectrum contrasts the two types of SiH groups observed in the IR spectra, suggesting fluxional process(es). A ¹H NMR spectrum collected at −73 °C in toluene-d₈ revealed that two SiHMe₂ groups were inequivalent: signals at 4.67 (¹J_SiH ∼ 180 Hz) and 3.82 (¹J_SiH ∼ 120 Hz) ppm were assigned to nonbridging SiH and bridging Ln↼H–Si moieties, respectively. These resonances correlated in a COSY experiment to signals at 0.35 and 0.08 ppm (with the downfield SiH) and 0.77 and 0.67 ppm (with the upfield SiH) of the now inequivalent methyl groups. Notably, one of the ortho-C₆H₅, whose resonance appeared unusually upfield at 4.15 ppm, was even more shielded than the nonbridging SiH. Moreover, all five H in the C₆H₅ were inequivalent. Thus, the alkyl ligands are equivalent in the low temperature solution-phase structure, with each ligand containing one La↼H–Si and a π-coordinated aryl group.

Single crystal X-ray diffraction reveals that the molecular structure of 3 (P3̄) contains three, crystallographically related C(SiHMe₂)₂Ph ligands, each of which interacts with the lanthanum center through the central carbon (C1), through a benzylic-type coordination of the C6, and also through one La↼H–Si (Fig. 2).§ The three ligands are arranged in a trigonal geometry around the lanthanum center (Σ_{C1–La1–C1} = 357.15(6)°). This structure is consistent with the low temperature NMR and IR spectroscopic data. A few of the notable structural features include the sharp La1–C1–Si1 angle (93.0(1)°), short La1⋯Si1 and La1⋯H1s distances (3.3141(9) and 2.69(4) Å), and an unusually long La1–C1 distance (2.674(3) Å). The corresponding La–C distances in six-coordinate tris(benzyl)lanthanum compounds, e.g., La(CH₂Ph)₃(THF)₃ (2.648(2) Å),⁵ are shorter, and the distance in La(CH(SiMe₃)₂)₃ (2.515(9) Å) is much shorter.¹² These trends extend to the comparisons of structures of 4–6 to the analogous benzyllanthanide species. Moreover, the close contacts in the series (i.e., Ln–C, Ln⋯Si, and Ln⋯H) follow the expected trend based on ionic radius (La > Ce > Pr > Nd).

Fig. 2 Thermal ellipsoid plot of La(C(SiHMe₂)₂Ph)₃ (3). La1 is located on a crystallographic 3-fold axis. H atoms bonded to Si are located in the Fourier difference map, their positions are refined and are illustrated. All other H atoms and a disordered pentane molecule (0.5) are not shown for clarity. Short La–C (green) and Ln↼H–Si (black) distances are highlighted with dashed lines. Selected interatomic distances (Å): La1–C1, 2.674(3); La1–C6, 2.822(2); La1–Si1, 3.3141(9); La1–H1s, 2.69(4); C1–Si1, 1.821(3); Si1–H1s, 1.37(4); C1–Si2, 1.853(3); Si2–H2s, 1.45(4); C1–C6, 1.483(4); C6–C7, 1.415(5); C6–C11, 1.409(4). Selected interatomic angles (°): C1–La1–C1, 119.05(2); La1–C1–C6, 80.0(1); La1–C1–Si1, 93.0(1); La1–C1–Si2, 128.3(1).

Interestingly, isomorphous cerium 4, praseodymium 5 and neodymium 6 compounds' structures (P2₁/c) are inequivalent with that of La 3. The two molecules in the unit cells for 4, 5 and 6 have inequivalent configurations, with one of the molecules containing only two Ln↼H–Si bridging moieties (Fig. 3).¶ The Ce–C distances for the five ligands that contain bridging Ce↼H–Si interactions average 2.65 ± 0.02 Å, whereas the η²-benzyl-only ligand (Ce2–C56, 2.587(3) Å) distance is shorter. This distinction is also apparent in compounds 5 (Pr–C_ave, 2.63 ± 0.02; Pr2–C56, 2.556(5) Å) and 6 (Nd–C_ave, 2.61 ± 0.02; Nd2–C56, 2.541(4) Å).

Fig. 3 Thermal ellipsoid plot of one of two crystallographically distinct molecules of Ce(C(SiHMe₂)₂Ph)₃ (4). H atoms bonded to Si are located in the Fourier difference map, their positions are refined and are illustrated. All other H atoms are not shown for clarity. Short Ce–C (green) and Ln↼H–Si (black) distances are highlighted with dashed lines. Selected interatomic distances (Å): Ce2–C34, 2.613(3); Ce2–C45, 2.671(2); Ce2–C56, 2.587(3); Ce2–Si7, 3.1947(9); Ce2–H7s, 2.47(2); C34–Si7, 1.829(2); Si7–H7s, 1.48(3); Ce2–Si9, 3.2379(9); Ce2–H9s, 2.46(3); C45–Si9, 1.829(3); Si9–H9s, 1.48(3); Selected interatomic angles (°): C34–Ce2–C45, 118.57(9); Ce2–C34–C39, 83.8(2); Ce2–C34–Si7, 90.1(1).

This work extends the idea that the β-SiH group supports large, coordinatively unsaturated rare earth centers in homoleptic, solvent-free compounds to include a new mixed benzyl dimethylsilyl ligand. Tris(alkyl) lanthanides Ln{C(SiHMe₂)₂Ph}₃ are synthesized in good yields, and the secondary interactions involving β-SiH and aryl moieties are likely important to the facile isolation of these compounds. The structural parameters (e.g., Ln–C, Si–H, Si–C distances) for moieties involved in secondary Ln↼H–Si interactions are different than those with nonbridging SiH groups, and the Ln–C distances are also affected by the presence or lack of secondary Ln↼H–Si interactions. The ligand itself is synthesized by deprotonation of the new alkane HC(SiHMe₂)₂Ph with KBn, but Me₂SiBn₂ and other side products in reactions of HC(SiHMe₂)₂Ph and KBn suggested a competing reaction involving nucleophilic attack on a Si center to cleave the C–Si bond. Therefore we developed an alternative route to KC(SiHMe₂)₂Ph by reacting PhC(SiHMe₂)₃ with KOtBu that affords the desired product in excellent yield. This straightforward two-step synthesis to homoleptic organolanthanides may allow their application in the preparation of heteroleptic lanthanide complexes and as precursors for new catalytic chemistry.

This research was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences. The Ames Laboratory is operated for the U.S. Department of Energy by Iowa State University under Contract No. DE-AC02-07CH11358.

References

1. M. Zimmermann and R. Anwander, Chem. Rev., 2010, 110, 6194–6259.
2. A. M. Kawaoka, M. R. Douglass and T. J. Marks, Organometallics, 2003, 22, 4630–4632; H. F. Yuen and T. J. Marks, Organometallics, 2008, 27, 155–158; H. F. Yuen and T. J. Marks, Organometallics, 2009, 28, 2423–2440; M. P. Conley, G. Lapadula, K. Sanders, D. Gajan, A. Lesage, I. del Rosal, L. Maron, W. W. Lukens, C. Copéret and R. A. Andersen, J. Am. Chem. Soc., 2016, 138, 3831–3843; M. Zimmermann, K. W. Törnroos and R. Anwander, Organometallics, 2006, 25, 3593–3598; H. Martin Dietrich, G. Raudaschl-Sieber and R. Anwander, Angew. Chem., Int. Ed., 2005, 44, 5303–5306; R. Taube, S. Maiwald and J. Sieler, J. Organomet. Chem., 2001, 621, 327–336; S. R. Daly, D. Y. Kim, Y. Yang, J. R. Abelson and G. S. Girolami, J. Am. Chem. Soc., 2010, 132, 2106–2107.
3. H. Schumann, J. Muller, N. Bruncks, H. Lauke, J. Pickardt, H. Schwarz and K. Eckart, Organometallics, 1984, 3, 69–74.
4. A. J. Wooles, D. P. Mills, W. Lewis, A. J. Blake and S. T. Liddle, Dalton Trans., 2010, 39, 500–510.
5. S. Bambirra, A. Meetsma and B. Hessen, Organometallics, 2006, 25, 3454–3462.
6. A. C. Behrle and J. A. R. Schmidt, Organometallics, 2011, 30, 3915–3918.
7. S. Harder, Organometallics, 2005, 24, 373–379.
8. F. Feil and S. Harder, Organometallics, 2000, 19, 5010–5015.
9. P. L. Watson, J. Am. Chem. Soc., 1982, 104, 337–339.
10. H. Van der Heijden, C. J. Schaverien and A. G. Orpen, Organometallics, 1989, 8, 255–258.
11. P. L. Watson, J. Chem. Soc., Chem. Commun., 1983, 276–277.
12. P. B. Hitchcock, M. F. Lappert, R. G. Smith, R. A. Bartlett and P. P. Power, J. Chem. Soc., Chem. Commun., 1988, 1007–1009.
13. A. G. Avent, C. F. Caro, P. B. Hitchcock, M. F. Lappert, Z. N. Li and X. H. Wei, J. Chem. Soc., Dalton Trans., 2004, 1567–1577.
14. W. A. Herrmann, J. Eppinger, M. Spiegler, O. Runte and R. Anwander, Organometallics, 1997, 16, 1813–1815; J. William, S. Rees, O. Just, H. Schumann and R. Weimann, Angew. Chem., Int. Ed. Engl., 1996, 35, 419–422.
15. K. Yan, A. V. Pawlikowski, C. Ebert and A. D. Sadow, Chem. Commun., 2009, 656–658; K. Yan, B. M. Upton, A. Ellern and A. D. Sadow, J. Am. Chem. Soc., 2009, 131, 15110–15111; A. Pindwal, A. Ellern and A. D. Sadow, Organometallics, 2016, 35, 1674–1683.
16. K. Yan, G. Schoendorff, B. M. Upton, A. Ellern, T. L. Windus and A. D. Sadow, Organometallics, 2013, 32, 1300–1316.
17. A. Pindwal, S. Patnaik, W. C. Everett, A. Ellern, T. L. Windus and A. D. Sadow, Angew. Chem., Int. Ed., 2016 DOI:10.1002/anie.201610263.
18. C. Eaborn, P. B. Hitchcock and P. D. Lickiss, J. Organomet. Chem., 1983, 252, 281–288.
19. E. J. Hawrelak, F. T. Ladipo, D. Sata and J. Braddock-Wilking, Organometallics, 1999, 18, 1804–1807.
20. A. Asadi, A. G. Avent, M. P. Coles, C. Eaborn, P. B. Hitchcock and J. D. Smith, J. Organomet. Chem., 2004, 689, 1238–1248.
21. C. Marschner, Eur. J. Inorg. Chem., 1998, 221–226; G. Gutekunst and A. G. Brook, J. Organomet. Chem., 1982, 225, 1–3.

---

Footnotes

† Electronic supplementary information (ESI) available: Experimental procedures, and characterization for compounds 1, 2, 2·TMEDA, 3–6. CCDC 1517682–1517686. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6cc09304c

‡ X-ray data for 2·TMEDA (CCDC 1517682): C₁₇H₃₅KN₂Si₂; FW 362.75; monoclinic; a: 8.7354(6), b: 23.003(1), c: 11.6433(7), β: 100.207(4), volume: 2302.5(3); P121/c1; Z = 4; temp. 173 K; reflections: collected, 20 907; independent, 3284; R_int 0.1552; 2217 data I > 2σ(I): R₁ 0.0705, wR₂ 0.1845; R_all: R₁ 0.0996, wR₂ 0.2145.

§ X-ray data for 3 (CCDC: 1517683): C₃₃H₅₇LaSi₆·(C₅H₁₂)_1/2; FW 797.31; trigonal; a: 12.4380(6), c: 16.157(1), volume, 2164.7(3); P3̄; Z = 2; temp. 173 K; reflections: collected, 23 107; independent, 3715; R_int 0.0345; 3439 data I > 2σ(I): R₁ 0.0309, wR₂ 0.0828; R_all: R₁ 0.0351, wR₂ 0.0928.

¶ X-ray data for 4 (CCDC: 1517684): C₃₃H₅₇CeSi₆; FW 762.44; monoclinic; a: 22.324(2), b: 19.841(1), c: 19.921(1), β: 116.076(1), volume: 7925.5(9); P121/c1; Z = 8; temp. 173 K; reflections: collected, 79 198; independent, 18 762; R_int 0.0485; 14 549 data I > 2σ(I): R₁ 0.0315, wR₂ 0.0698; R_all: R₁ 0.0515, wR₂ 0.0836; X-ray data for 5 (CCDC 1517686): C₃₃H₅₇PrSi₆; FW 763.23; monoclinic; a: 22.3310(6), b: 19.8399(5), c: 19.8802(5), β: 116.047(1), volume: 7913.2(4); P121/c1; Z = 8; temp. 173 K; reflections: collected, 79 681; independent, 15 456; R_int 0.0653; 14 828 data I > 2σ(I): R₁ 0.0695, wR₂ 0.1909; R_all: R₁ 0.0710, wR₂ 0.1931; X-ray data for 6 (CCDC: 1517685): C₃₃H₅₇NdSi₆; monoclinic; a: 22.335(2), b: 19.843(2), c: 19.855(2), β: 116.017(1), volume: 7908(1); P121/c1; Z = 8; temp. 173 K; reflections: collected, 113 017; independent, 20 627; R_int 0.0711; 15 075 data I > 2σ(I): R₁ 0.0346, wR₂ 0.0623; R_all: R₁ 0.0624, wR₂ 0.0768.

---