and Mono(amidate) Complexes of Yttrium - American Chemical Society

3990

Organometallics 2009, 28, 3990–3998 DOI: 10.1021/om801190t

Bis- and Mono(amidate) Complexes of Yttrium: Synthesis, Characterization, and Use as Precatalysts for the Hydroamination of Aminoalkenes Louisa J. E. Stanlake and Laurel L. Schafer* Department of Chemistry, University of British Columbia, 2036 Main Mall, Vancouver, British Columbia, Canada, V6T 1Z1 Received December 16, 2008

A high-yielding synthetic route is disclosed for yttrium bis- and mono(amidate) complexes using the reaction of amide proligand and Y(N(SiMe3)2)3 starting materials. The structure, bonding, and solution phase characterization data for this new class of complexes are presented. The modular nature of the amidate ligand allows for easy addition of electron-withdrawing CF3 groups in the ligand backbone to tune electronic properties of the resulting precatalysts. The amidate ligands in the bis(amidate) complexes were found to be highly fluxional on the NMR time scale, while the mono(amidate) yttrium complex required heating to 110 °C before ligand redistribution was observed. Three bis(amidate) complexes with differing electronic properties and one mono(amidate) complex have been used as precatalysts for hydroamination using a wide range of aminoalkene substrates. Bis(amidate) complexes bearing the more electron-withdrawing amidate ligands were found to be the most active precatalysts for intramolecular alkene hydroamination. Introduction The catalytic synthesis of nitrogen-containing heterocycles is a much sought after process for potential application in the pharmaceutical and fine chemical industries. Hydroamination, which is the formal addition of nitrogen and hydrogen atoms across a carbon-carbon multiple bond, is an atomeconomical route to C-N bond formation that typically requires a catalyst.1,2a Metal-based catalysts for alkene hydroamination include examples from across the periodic table and have been recently reviewed.2 The rare-earth metals have proven to be very active for the intra-3 and intermolecular4 hydroamination of alkenes, but synthesis and handling of known rare-earth precatalysts can be challenging. An alternative and practical approach has focused on the in situ *Corresponding author. E-mail: [email protected]. (1) Severin, R.; Doye, S. Chem. Soc. Rev. 2007, 36, 1407. (2) (a) M€ uller, T. E.; Hultzsch, K. C.; Yus, M.; Foubelo, F.; Tada, M. Chem. Rev. 2008, 108, 3795. (b) Aillaud, I.; Collin, J.; Hannedouche, J.; Schulz, E. Dalton Trans. 2007, 5105. (c) Hultzsch, K. C. Adv. Synth. Catal. 2005, 347, 367. (d) Hong, S.; Marks, T. J. Acc. Chem. Res. 2004, 37, 673. (3) Representative examples include: (a) Yuen, H. F.; Marks, T. J. Organometallics 2008, 27, 155–158. (b) Yuan, Y.; Chen, Y.; Li, G.; Xia, W. Organometallics 2008, 27, 6307. (c) Ge, S.; Meetsma, A.; Hessen, B. Organometallics 2008, 27, 5339. (d) Aillaud, I.; Lyubov, D.; Collin, J.; Guillot, R.; Hannedouche, J.; Schulz, E.; Trifonov, A. Organometallics 2008, 27, 5929. (e) Zi, G. G.; Xiang, L.; Song, H. B. Organometallics 2008, 27, 1242. (f) Datta, S.; Gamer, M. T.; Roesky, P. W. Organometallics 2008, 27, 1207. (g) Bambirra, S.; Tsurugi, H.; van Leusen, D.; Hessen, B. Dalton Trans. 2006, 1157. (h) O’Shaughnessy, P. N.; Gillespie, K. M.; Knight, P. D.; Munslow, I. J.; Scott, P. Dalton Trans. 2004, 2251. (i) Lauterwasser, F.; Hayes, P. G.; Brase, S.; Piers, W. E.; Schafer, L. L. Organometallics 2004, 23, 2234. (j) Gribkov, D. V.; Hultzsch, K. C.; Hampel, F. Chem.;Eur. J. 2003, 9, 4796. (k) Kim, Y. K.; Livinghouse, T.; Horino, Y. J. Am. Chem. Soc. 2003, 125, 9560. (4) (a) Ryu, J. S.; Li, G. Y.; Marks, T. J. J. Am. Chem. Soc. 2003, 125, 12584. (b) Gribkov, D. V.; Hultzsch, K. C.; Hampel, F. J. Am. Chem. Soc. 2006, 128, 3748. pubs.acs.org/Organometallics

Published on Web 06/24/2009

assembly of rare-earth metal-based catalytic systems to achieve the cyclohydroamination of alkenes.5 While this facile approach is attractive for the synthesis of heterocycles, it provides limited insight into catalyst structure/reactivity patterns. Here we focus on the preparation and characterization of a new class of easily assembled bis- and mono(amidate) yttrium precatalysts for the cyclohydroamination of primary and secondary aminoalkenes. Research in our group has shown that bis(amidate)bis(amido) complexes of group 4 metals can be formed in high yield and are useful precatalysts for the hydroamination of aminoalkenes.6 The amidate ligand has steric and electronic properties that can be easily varied to tune the catalyst performance.7 Furthermore, we have shown that the direct synthesis of yttrium tris(amidate) complexes from commercially available Y(N(SiMe3)2)3 using a protonolysis reaction gives monomeric and thermally robust crystalline complexes.8 However, we have observed that such complexes are not reactive in catalytic hydroamination, presumably due to the lack of a reactive ligand such as -N(SiMe3)2 (vide infra). The synthesis (5) (a) Hannedouche, J.; Aillaud, I.; Collin, J.; Schulz, E.; Trifonov, A. Chem. Commun. 2008, 3552. (b) Quinet, C.; Ates, A.; Marko, I. E. Tetrahedron Lett. 2008, 49, 5032. (c) Kim, H.; Kim, Y. K.; Shim, J. H.; Kim, M.; Han, M. J.; Livinghouse, T.; Lee, P. H. Adv. Synth. Catal. 2006, 348, 2609. (d) Riegert, D.; Collin, J.; Meddour, A.; Schulz, E.; Trifonov, A. J. Org. Chem. 2006, 71, 2514. (e) Kim, J. Y.; Livinghouse, T. Org. Lett. 2005, 7, 1737. (f) Kim, J. Y.; Livinghouse, T. Org. Lett. 2005, 7, 4391. (g) Kim, Y. K.; Livinghouse, T., Angew. Chem., Int. Ed. 2002, 41, 3645.(h) Kim, Y. K.; Livinghouse, T.; Bercaw, J. E. Tetrahedron Lett. 2001, 42, 2933. (6) (a) Wood, M. C.; Leitch, D. C.; Yeung, C. S.; Kozak, J. A.; Schafer, L. L. Angew. Chem., Int. Ed. 2007, 46, 354. (b) Bexrud, J. A.; Li, C. Y.; Schafer, L. L. Organometallics 2007, 26, 6366. (c) Thomson, R. K.; Bexrud, J. A.; Schafer, L. L. Organometallics 2006, 25, 4069. (7) Lee, A. V.; Schafer, L. L. Eur. J. Inorg. Chem. 2007, 2243. (8) Stanlake, L. J. E.; Beard, J. D.; Schafer, L. L. Inorg. Chem. 2008, 47, 8062. r 2009 American Chemical Society

Article

Organometallics, Vol. 28, No. 14, 2009

3991

Scheme 1. Synthesis of Bis(amidate) Yttrium Complexes

Figure 1. Aromatic region of the room-temperature 400 MHz 1 H NMR spectrum of crude product 4 in C6D6.

of discrete bis(amidate) complexes of yttrium was anticipated to present challenges for two main reasons: (1) mixed-ligand complexes of group 3 metals have a tendency to undergo ligand redistribution9 and (2) amidate ligands are known to promote the formation of metal aggregate species.10 Here we report high-yielding preparative methods for the synthesis of crystalline bis- and mono(amidate) complexes. These first examples of fully characterized bis- and mono(amidate) yttrium complexes have also been used as precatalysts for catalytic hydroamination. These investigations show that the modular nature of the flexible amidate ligand permits tunable catalytic activity to promote variable reactivity and selectivity in the cyclohydroamination of primary and secondary aminoalkenes.

Results and Discussion Complex Synthesis and Characterization. Amide proligands are synthesized in high yield from a facile reaction of acid chloride and primary amine starting materials. Proligands 1, 2,8 and 3 have been synthesized and can be used in the preparation of bis(amidate) complexes 4, 5, and 6, while proligand 1 can be used in the synthesis of mono(amidate) complex 7 (Scheme 1). The synthesis of amidate yttrium complexes is achieved by dissolving the appropriate amount of proligand and Y(N(SiMe3)2)3 in THF and stirring at 60 °C for 2 h. After filtration through a pipet plug of Celite to remove trace amounts of insoluble materials (a clear solution is observed; however, optimized yields of crystalline material are reproducibly obtained using this filtration protocol) before drying in vacuo to obtain the crude complex. As a representative example, the 1H NMR spectrum of crude complex 4 is presented in Figure 1. The signal corresponding to the ortho-proton on the naphthyl substituent is a very useful NMR handle, as there are two separate ortho-proton doublets in the 1H NMR spectrum (signals “a” and “b” in Figure 1) that indicate two discrete compounds in solution. There is no remaining unreacted proligand, as there are no N-H signals typically observed at approximately δ 6.6 (free proligand) or δ 11 (coordinated proligand) in the 1H NMR spectrum.8 Notably, the doublet at δ 9.26 matches the chemical shift for the ortho-proton on the naphthyl substituent in the previously characterized tris(amidate) complex,8 and the doublet at δ 9.09 (peak “b”) corresponds to the (9) Piers, W. E.; Emslie, D. J. H. Coord. Chem. Rev. 2002, 233, 131. (10) Zhou, X. G.; Zhu, M. J. Organomet. Chem. 2002, 647, 28.

desired bis(amidate) product, showing that a mixture of complexes results from this reaction. However, by selecting a higher boiling solvent such as toluene and subsequent heating to 90 °C for 2 h, the clean formation of bis(amidate) complex 4 is realized in high yield (82% recrystallized yield). This suggests that the bis(amidate) complex is the thermodynamic product. This facile synthetic approach can also be used in the formation of bis(amidate) complexes 5 and 6, in 80% and 84% yield, respectively (Scheme 1). Notably, attempts to prepare monomeric, characterizable amidate complexes in the absence of neutral donors such as THF or pyridine were not successful. The complexes are moisture sensitive and in the solid state can be stored at -30 °C for more than 4 months. Most importantly, these complexes are stable in solution at 110 °C for an extended period of time (C6D6 in a sealed J. Young NMR tube for up to 4 days), which renders them ideal candidates for application in catalysis. The 1H NMR spectra of all bis(amidate) complexes at room temperature are consistent with C2 symmetric compounds in solution. No line broadening is evident when an NMR sample of complex 4 is cooled to -50 °C, indicating no loss in symmetry at these lower temperatures. The bis(amidate) complexes all have one molecule of THF bound to the metal center, which is labile, as indicated by the broadness of the methylene proton signals and the shift of the signals in the 1H NMR spectra from free THF. The THF molecule remains bound even after exposing the bis(amidate) complexes to full vacuum overnight, as indicated by an unchanged integration value for the THF methylene protons before and after this treatment. Complex formation and electron delocalization through the amidate backbone is further confirmed by using the IR data for the bis(amidate) complexes. In all cases, the loss of the N-H stretch from the proligand and the weakening of the CdO stretch is evident (e.g., CdO stretch: 1643 to 1511 cm-1 for 1 to 4). The mass spectra of compounds 4, 5, and 6 all have molecular ion peaks associated with their respective complexes after loss of THF, while the fragmentation pattern shows initial loss of the -N(SiMe3)2 group. As a representative example, X-ray quality crystals of complex 4 were grown at -35 °C from hexanes with a few added drops of toluene. The solid-state molecular structure of 4 is shown in Figure 2, with selected bond lengths and angles presented in Table 1. Complex 4 is a six-coordinate, C1 symmetric, distorted pentagonal-pyramidal structure with the amido ligand in the axial position. There is electron delocalization throughout the κ2-amidate backbone, as indicated by the similar C-O and C-N bond lengths (average C-O, C-N for 4 is 1.291(4), 1. 312(5) A˚). The Y-O(amidate) and Y-N(amidate) average bond lengths are 2.282(3) and 2.444(3) A˚ for complex 4, showing that the binding of the amidate ligand is very asymmetric, with the Y-O bond length being shorter than the Y-N bond length.

3992


Stanlake and Schafer

Figure 2. ORTEP diagram of the solid-state molecular structure of complex 4, with the probability ellipsoids drawn at the 50% level. The THF ring carbons, methyls of silyl substituents, and two molecules of toluene have been omitted for clarity. Table 1. Selected Bond Lengths and Angles for Complex 4 bond length (A˚) Y1-O1 Y1-N1 O1-C1 N1-C1 Y1-O2 Y1-N2 O2-C2 N2-C2 Y1-O3 Y1-N3

2.285(3) 2.450(3) 1.288(4) 1.313(5) 2.279(3) 2.437(3) 1.293(4) 1.310(5) 2.350(3) 2.255(3)

Scheme 2. Synthesis of Mono(amidate) Yttrium Complex 7

bond angle (deg) Y1-O1-C1 O1-C1-N1 C1-N1-Y1 N1-Y1-O1 Y1-O2-C2 O2-C2-N2 C2-N2-Y1 N2-Y1-O2 Y1-N3-Si1 Y1-N3-Si2 Si1-N3-Si2

97.6(2) 116.8(3) 89.4(2) 55.7(1) 97.6(2) 116.2(3) 89.9(2) 55.8(1) 124.6(2) 116.6(2) 118.8(2)

This is attributed to both the greater electronegativity of oxygen in combination with significant steric bulk about the nitrogen. The Y-N(SiMe3)2 bond length for 4 (2.255(3) A˚) is typical.11 The average bite angle of the amidate ligand (O-YN) is 55.7(1)°, in agreement with similar previously reported complexes;8,12 however it is significantly different than idealized angles for pentagonal-pyramidal structures. The average sum of the angles of the amidate metallacycle is 359.5° for 4, which indicates that the yttrium and amidate backbone are in the same plane. The bonding of the amidate ligand is best described as σ-bound through both the oxygen and nitrogen and is a four-electron donor to the Y3+ metal center. While an efficient route for the formation of bis(amidate) yttrium complexes has been determined, it is also of interest to investigate the synthesis of related mono(amidate) complexes. To this end, the reaction shown in Scheme 2 can be carried out, and after isolation of the crude product 7, the 1H NMR spectrum in C6D6 shows only one signal for the ortho-naphthyl signal at δ 9.16. This signal is different from the diagnostic δ 9.09 chemical shift for the bis(amidate) complex 4. Mono(amidate) complex 7 can be easily recrystallized by dissolving the crude product in a minimum amount of hexanes and cooling to -35 °C to give colorless plate-like crystals in high yield (77%). This compound is soluble in all common hydrocarbon solvents and is very moisture sensitive. The solid sample can be stored at -30 °C for greater than 4 months without decomposition; however, in solution phase it can (11) Hultzsch, K. C.; Spaniol, T. P.; Okuda, J. Organometallics 1997, 16, 4845. (12) An example of a related rare-earth ureate complex: Mao, L.; Shen, Q.; Xue, M. Organometallics 1997, 16, 3711.

undergo ligand redistribution at higher temperatures. For example, heating a solution of complex 7 to 65 °C results in no change of the ortho-naphthyl signal in the 1H NMR spectrum, but at 110 °C within an hour a new signal begins to appear at δ 9.09, matching that of bis(amidate) complex 4. The mono(amidate) complex 7 is similar to the bis(amidate) complexes in that there is one molecule of bound THF on the yttrium center, as indicated by the 1H NMR spectrum. The IR spectrum of the mono(amidate) complex 7 shows a disappearance of the N-H stretch, a weakening of the CdO stretch, and a new CdN stretch. The electronimpact mass spectrum of 7 gives a molecular ion peak that does not include the THF and a fragmentation pattern with (M+ - CH3) and (M+ - N(SiMe3)2) fragments. X-ray quality crystals of 7 can be obtained at -35 °C after recrystallization from a minimum amount of hexanes (Figure 3). The solid-state molecular structure of complex 7 shows a five-coordinate, C1 symmetric, distorted squarebased pyramidal structure with one amido ligand in the axial position. This low coordination number is rare for yttrium due to the previously discussed ligand redistribution pathways, although sterically bulky ligands such as -N(SiMe3)2 are well known to stabilize such species.13 Electron delocalization is evident through the amidate backbone since the C-O and C-N bond lengths are similar at 1.280(4) and 1.316(4) A˚ (Table 2). The Y-O(amidate) bond length is slightly shorter than those for the analogous bis(amidate) complexes (2.215(2) A˚), whereas the Y-N(amidate) bond length is much longer (2.519(3) A˚), likely due to the effects of the sterically demanding -N(SiMe3)2 group. The average Y-N(SiMe3)2 bond length for 7 (2.223(2) A˚) and amidate bite angle (55.52(8)°) are very similar to those of the bis(amidate) complex 4. (13) Bradley, D. C.; Ghotra, J. S.; Hart, F. A. J. Chem. Soc., Dalton Trans. 1973, 1021.

Article


3993

Scheme 3. σ-Bond Insertion Mechanism for Hydroamination of Aminoalkene 8 Using Rare-Earth Catalysts2a,14

Figure 3. ORTEP structure of complex 7 with the probability ellipsoids drawn at the 50% level. THF ring carbons are omitted for clarity. Table 2. Selected Bond Lengths and Bond Angles for Complex 7 bond length (A˚) Y1-O1 Y1-N1 O1-C1 N1-C1 Y1-O2 Y1-N2 Y1-N3

2.215(2) 2.519(3) 1.280(4) 1.316(4) 2.343(2) 2.214(2) 2.231(2)

bond angle (deg) Y1-O1-C1 O1-C1-N1 C1-N1-Y1 N1-Y1-O1 Y1-N2-Si1 Y1-N2-Si2 Si1-N2-Si2 Y1-N3-Si3 Y1-N3-Si4 Si3-N3-Si4

100.7(2) 117.6(3) 85.8(2) 55.52(8) 120.6(1) 119.9(2) 119.5(2) 112.4(1) 126.6(1) 120.8(2)

Thus, by using a simple protonolysis route from commercially available yttrium starting material, both the bis(amidate) and mono(amidate) complexes can be formed in high yield by simply varying the reaction stoichiometry. The thermal stability of these complexes makes them attractive, potential hydroamination precatalysts. Furthermore, catalyst performance can be tuned by taking advantage of the easily modified amidate backbone. Here, the bis- and mono(amidate) precatalysts 4, 5, 6, and 7 have been explored as new catalysts for cyclohydroamination. Hydroamination. Rare-earth catalysts have been shown to mediate cyclohydroamination via a σ-bond insertion mechanism (Scheme 3).2a,14 In many well-investigated examples, the alkene insertion step is understood to be the turnover-limiting step.2a,14 Ligand design can be used to advantage for enhancing rate, substrate scope, and regioand stereoselectivity of the reaction.2 Initially, to test the reactivity of the bis(amidate) complexes, the hydroamination of 2,2-diphenyl-4-pentenylamine (8) was performed.15 The precatalyst, standard (1,3,5-trimethoxybenzene), and substrate were weighed out separately in an inert atmosphere glovebox and dissolved in approximately 0.7 g of deuterated benzene. The reaction was monitored until >99% conversion was noted in the (14) Gagne, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 275. (15) Hong, S. W.; Tian, S.; Metz, M. V.; Marks, T. J. J. Am. Chem. Soc. 2003, 125, 14768.

1

H NMR spectrum, as indicated by complete depletion of substrate alkene proton multiplets at δ 5.73 and 5.10. After full conversion, NMR yields were obtained by comparing the integration of 1,3,5-trimethoxybenzene standard proton signals (aryl and methyl protons) with the diagnostic proton signals at δ 2.47 (-CH2CH(CH3)NH-) and 1.09 (-CH2CH(CH3)NH-) for compound 9 in the 1H NMR spectrum. The bis(amidate) complexes were highly efficient in the conversion of compound 8 to heterocycle 9 (entry 1, Table 3) at room temperature. As an illustration of the practical application of these catalysts in the synthesis of substituted pyrrolidine products, isolated yields > 93% were obtained with catalysts 4-6. These yields are comparable to the 1H NMR yields determined using internal standard. In order to determine the substrate scope of the bis(amidate) precatalyst, a variety of aminoalkenes were tested at various reaction conditions. Entries 2-5 contain substrates that have a decrease in geminal substituent size, from a cyclohexyl group (10) to no gem-disubstituents at all (16). The gem-disubstituent effect is evident when comparing entries 2-5, as an increase in reaction time is noted, and in the case of entry 5, an increase in temperature is also needed to achieve full conversion. For the hydroamination of substrate 10, all bis(amidate) complexes give high yield in less than 15 min. When the size of the ring in the gem-position on the substrate is reduced to a cyclopentyl group (12), reaction time is increased to at least an hour, but still provides high yields. This reaction also illustrates the reactivity difference between the amidate complexes, with the CF3-substituted bis(amidate) complexes 5 and 6 giving the fastest times (1 h). The ring substituents in substrates 10 and 12 promote transition-state formation (shown in Scheme 3), even more than the methyl substituents of 14, because of the enforced gauche interactions of the reactive amine and alkene functionalities of the substrate. This is evidenced by the increase in reaction time (entry 4, Table 3) for the cyclization of substrate 14. Substrate 16, containing no substituents, requires the use of higher temperatures (110 °C) in order to achieve full conversion (entry 5, Table 3). For hydroamination of aminoalkenes, five-membered ring formation is facile, and these precatalysts can effect the formation of six- and even seven-membered rings (substrates 18 and 20). Importantly, the formation of

3994



Table 3. Hydroamination of Various Aminoalkenes Using 10 mol % Bis(amidate) Complexes 4, 5, and 6

a Time for >99% conversion in C6D6. bYield determined by 1H NMR spectroscopy using 1,3,5-trimethoxybenzene as an internal standard. cIsolated yield. dPercent conversion.

seven-membered rings by cyclohydroamination is known to be challenging and has been rarely observed.14,16 Throughout all of these experiments, the CF3-substituted complexes display enhanced reactivity, as evidenced by reduced reaction times required for full conversion. Interestingly, paraCF3-substituted precatalyst 5 and meta-CF3-disubstituted precatalyst 6 have very similar reactivities. To test if there is a difference in catalyst performance between preformed precatalysts and those assembled in situ, the hydroamination reaction in entry 6, Table 3, was also carried out with an in situ catalyst preparation. First, 20 mol % of the ligand, N-20 ,60 -diisopropylphenyl(naphthyl)amide, and 10 mol % of yttrium tris(bis(trimethylsilyl)amide) were combined in approximately 0.7 g of deuterated benzene. The attempted observation of the in situ-formed catalytic species by NMR spectroscopy yielded only broad, complex spectra, consistent with fluxional species in solution. However, upon addition of substrate 18, a potential neutral donor, a time zero NMR spectrum of the catalytic reaction mixture revealed spectra consistent with the formation of previously characterized bis(amidate) complex 4. After 1.8 h the reaction was completed with high conversion (>95%), as evidenced by the lack of alkene signals at δ 5.72 and 4.92, and the growth of product 19 proton signals was observed

in the 1H NMR spectrum. This result correlates well to the case in which isolated precatalyst 4 was used. This demonstrates that the reaction can be effected by a facile in situ catalyst formation using commercially available Y(N(SiMe3)2)3 and an easily prepared amide, with no loss of reactivity. This shows the potential ease of use of these catalytic systems in organic synthesis. Another challenging reaction is the hydroamination of internal alkenes. Entries 1 and 2 in Table 4 show that all bis(amidate) yttrium precatalysts promote the cyclohydroamination of internal alkenes. For the substrate with a phenyl substituent on the terminus (22, an activated alkene), reaction rates are very fast for the CF3 -containing bis(amidate) complexes 5 and 6. However, much longer reaction times and higher temperatures are needed when the phenyl terminus is replaced with a methyl substituent (compound 24). High yields are obtained for this reaction, and all amidate complexes give similar reaction times for this challenging reaction. Hydroamination using rare-earth metals, which catalyze the reaction through a σ-bond insertion mechanism, can also react with secondary amine substrates. This is in contrast to neutral bis(amidate) group 4 catalyzed hydroamination reactions, which require a primary amine to form the proposed imido intermediate.6c,17 In a preliminary screen using

(16) Bexrud, J. A.; Eisenberger, P.; Leitch, D. C.; Payne, P. R.; Schafer, L. L., J. Am. Chem. Soc. 2009, 131, 2116.

(17) Bexrud, J. A.; Beard, J. D.; Leitch, D. C.; Schafer, L. L. Org. Lett. 2005, 7, 1959.

Article


3995

Table 4. Hydroamination of Aminoalkenes with Terminal Substituents Using 10 mol % Bis(amidate) Complexes 4, 5, and 6

a Time for >99% conversion in C6D6. bYield determined by 1H NMR spectroscopy using 1,3,5-trimethoxybenzene as an internal standard. cIsolated yield. dPercent conversion, 1:1 mixture of diastereomers.

substrate 28 and precatalyst 4 (eq 1), reactivity of a secondary amine was confirmed by the appearance of product methyl proton signals at δ 2.07 (N(CH3)) and 1.11 (CH(CH3)) in the 1H NMR spectrum.

This reactivity observed for secondary amines can then be extended to a tandem reaction using the substrate 1-amino2,2-diallylpropane (26), which contains two alkene groups (entry 3, Table 4).18 Here rare-earth-catalyzed hydroamination results in a tandem cyclization to form first the allylsubstituted secondary amine, which can react further to give the tertiary amine 27. The bis(amidate) complexes 4, 5, and 6 were used in the hydroamination of 26 initially at 65 °C and then 110 °C. Using precatalysts 4, 5, and 6 these reactions proceed to 95%, 85%, and 77% conversion, respectively, with the observed formation of an approximately 1:1 diastereomeric mixture of products (endo,exo: exo,exo).18 After longer reaction times at 110 °C no further reaction was noted; however, if more precatalyst was added to the reaction mixture, the reaction would go to completion. This reaction “stalling” may be due to product inhibition or catalyst decomposition.3h Catalyst decomposition was ruled out by adding a second aliquot of substrate to a stalled experiment, and further conversion of newly added substrate to product was observed upon heating to 110 °C. In order to test the product inhibition hypothesis, 30 mol % of product 27 was added to the hydroamination reaction in one NMR tube, and this reaction was run side-by-side with another reaction containing no product. Both reactions were run at 110 °C concurrently for 1 h. Within that time the hydroamination reaction with only precatalyst 4 and substrate 26 resulted in a 96% conversion, as noted by 1H NMR spectroscopy. The reaction that had product 27 added gave a conversion of 86% in the same amount of time. Furthermore, reaction monitoring in the synthesis of (18) Hultzsch, K. C.; Hampel, F.; Wagner, T. Organometallics 2004, 23, 2601.

Table 5. Hydroamination Using Complex 7 as Precatalyst (10 mol %) entry

aminoalkene

product

temp [°C]

timea [h]

yieldb (isolated)c[%]

1 2 3 4 5 6 7 8 9 10

8 10 12 14 16 18 20 22 24 26

9 11 13 15 17 19 21 23 25 27

25 25 25 25 110 65 110 25 65 110

95d

a Time for >99% conversion in C6D6. b Yield determined by 1H NMR spectroscopy using 1,3,5-trimethoxybenzene as an internal standard. c Isolated yield. d Percent conversion.

both tertiary amines 27 and 29 showed sluggish reactivity as the reaction neared completion. These observations suggest that product inhibition in the synthesis of tertiary amines is a concern for these catalyst systems. Not surprisingly, when comparing the different complexes, product inhibition seems to be more problematic for the more Lewis acidic precatalysts that contain the electronwithdrawing CF3 groups on the amidate backbone (complexes 5 and 6), as evidenced by the maximum conversions of 85% and 77% observed, respectively, even with prolonged heating. The mono(amidate) complex 7 was also screened in hydroamination experiments, and the results are shown in Table 5. The less sterically congested mono(amidate) complex 7 had comparable rates and yields to complexes 5 and 6, as shown in entries 1-7 (Table 5). For the hydroamination of internal alkene substrate 22, mono(amidate) complex 7 effected the transformation much faster than the analgous bis(amidate) complex 4, but slightly slower than bis(amidate) complexes 5 and 6 (entry 8, Table 5). Notably, the corresponding, previously reported tris(amidate) complex8 gives no reaction in the hydroamination of reactive aminoalkene 8, even with elevated reaction temperatures (up to 110 °C) and prolonged reaction times (>24 h). When comparing the tandem cyclization reaction (entry 10, Table 5) between the amidate

3996


Scheme 4. Hydroamination of 1-Methyl-4-pentenylamine (30)a

Yield determined by 1H NMR spectroscopy using 1,3,5-trimethoxybenzene as an internal standard and comparison of known values14,19 for the HNCHMe 1H signal in the NMR spectrum of 31 and 32. a

complexes, the mono(amidate) complex 7 promotes high conversion to product 27, as does the bis(amidate) complex 4. Both the bis- and mono(amidate) complexes are efficient precatalysts for the cyclohydroamination of primary and secondary aminoalkenes. The mono(amidate) complex 7 has relative reaction rates comparable to the more active CF3containing bis(amidate) complexes 5 and 6. This suggests that the increased steric accessibility of the mono(amidate) yttrium complex is favorable for enhanced reactivity. A good scope of reactivity has been observed for these systems, and these complexes can be used in the efficient formation of five-, six-, and even seven-membered ring products. Although the precatalysts are well-characterized in all cases, the exact nature of the catalytically active species remains an area of investigation. Considering that ligand redistribution has been observed to occur at elevated temperatures (vide supra), it is possible that the formation of a mixture of catalytically active complexes can arise using these reaction conditions. However, neither the formation of such complexes arising from ligand redistribution nor evidence of proligand can be observed by NMR spectroscopy while monitoring catalytic reactions. While these amidate complexes are promising precatalysts for hydroamination, thus far, attempts to mediate intermolecular hydroamination with these systems have not been successful. The hydroamination of an R-monosubstituted aminoalkene (30) to give diastereomeric products (Scheme 4) can be carried out. In this case the use of bis- (4) and mono(amidate) (7) precatalysts allows a direct comparison of the effect of the number of auxiliary ligands on catalytic activity and provides insight into catalyst features required to enhance diastereoselectivity. The bis(amidate) complex 4 results in a better diastereomeric ratio (dr) after 99% conversion (over 3 h), supporting the fact that steric congestion imposed by the amidate ligands can dramatically improve diastereoselectivity. Monitoring of the reaction by NMR spectroscopy showed that the diastereomeric ratio remains constant throughout both reactions, suggesting that no significant change in catalytic structure is occurring over 3 h at 65 °C. These results point toward the potential application of chiral bis(amidate) group 3 precatalysts for asymmetric cyclohydroamination.

Summary Bis- and mono(amidate) complexes of yttrium are easily synthesized in high yield to give crystalline compounds. The bis- and mono(amidate) complexes all contain one molecule of THF and are six- and five-coordinate, respectively. The amidate bonding is most symmetric for the bis(amidate) complex 4, as evidenced by the similar Y-O and Y-N bonds. For the mono(amidate) complex 7, the Y-O bond


is shorter and the Y-N bond is longer than in complex 4. The solution and thermal stability of these complexes make them very attractive candidates as hydroamination precatalysts. Overall amidate complexes of yttrium have been shown to be good precatalysts for the hydroamination of aminoalkenes, although desirable and challenging low-temperature and intermolecular reactivity that has been observed for other group 3 systems has yet to be realized.4 The trends in reaction times indicate that the CF3-containing bis(amidate) complexes (complexes 5 and 6) are more reactive than the bis(amidate) 4. These observations show that advantageous tuning of the amidate ligand can result in enhanced catalyst performance. While the mono(amidate) complex 7 had comparable reactivity to complexes 5 and 6, the bis(amidate) analogue (4) proved to give better diastereoselectivity than the mono(amidate) complex 7. It was shown that in situ catalyst preparation gives comparable results to the reactions using isolated precatalyst, rendering these systems wellsuited for general application in synthesis. Most importantly, the ease of modification and synthesis of the amidate backbone makes it an attractive ligand set for the development of new, more reactive and selective catalyst systems.

Experimental Section General Procedures. All operations were performed under an inert atmosphere of nitrogen using standard Schlenk-line or glovebox techniques. THF, toluene, pentane, and hexanes were all purified by passage through an alumina column and sparged with nitrogen. Y(N(SiMe3)2)3 was synthesized as described in the literature13 or purchased from Aldrich and recrystallized from hexanes before use. The compounds 2,2-diphenyl-4-pentenylamine,15 4-pentenylamine,14 1-methyl-4-pentenylamine,19 C-(1-allylcyclohexyl)methylamine,20 C-(1-allylcyclopentyl)methylamine,5d 2,2-dimethylpent-4-enylamine,21 2,2-diphenyl-5-hexenylamine,22 2,2-diphenyl-6-septenylamine,16 2,2,5-triphenyl-4-pentenylamine,22 5-methyl-2,2-diphenyl-4-pentenylamine, and 1-amino-2, 2-diallylpropane23 were made according to previously reported procedures and purified by distillation and storage over molecular sieves. All other chemicals were commercially available and used as received unless otherwise stated. 1H and 13C NMR spectra were recorded on Bruker AV300, AV400, and AV600 spectrometers. Elemental analyses and mass spectra were performed by the microanalytical laboratory of the Department of Chemistry at the University of British Columbia. Some elemental analyses gave low carbon content, possibly due to carbide formation.25 X-ray crystallography was conducted at the University of British Columbia by Dr. Brian Patrick, Dr. Rob Thomson, or Neal Yonson. Syntheses of compounds 1 and 2 were previously reported.8 Synthesis of N-(2,6-Diisopropylphenyl)(3,5-bis(trifluoromethyl))phenylamide (3). To a 250 mL round-bottom flask were added 2,6-diisopropylaniline (3.50 mL, 18.5 mmol) and 125 mL of dichloromethane. The reaction mixture was cooled to 0 °C using an ice bath, and triethylamine (3.20 mL, 23.0 mmol) was added dropwise by syringe. The resulting solution was (19) Harding, K. E.; Burks, S. R. J. Org. Chem. 1981, 46 (19), 3920– 3922. (20) Bender, C. F.; Widenhoefer, R. A. J. Am. Chem. Soc. 2005, 127, 1070. (21) Tamaru, Y.; Hojo, M.; Higashimura, H.; Yoshida, Z. J. Am. Chem. Soc. 1988, 110, 3994. (22) Kondo, T.; Okada, T.; Mitsudo, T. J. Am. Chem. Soc. 2002, 124, 186. (23) Quinet, C.; Jourdain, P.; Hermans, C.; Ates, A.; Lucas, I.; Mark o, I. E. Tetrahedron 2008, 64, 1077. (25) Vitanova, D. V.; Hampel, F.; Hultzsch, K. C. Dalton Trans. 2005, 1565.

Article stirred for 5 min with subsequent dropwise addition of 3,5-bis(trifluoromethyl)benzoyl chloride (5.00 mL, 18.0 mmol), which fumed upon addition. The reaction was stirred overnight, in which time a white solid precipitated from the solution. The solid was isolated by filtration and recrystallized by dissolving in warm dichloromethane, with subsequent addition of hexanes and then cooling to 0 °C. A white fibrous solid was isolated by filtration. Yield: 5.29 g, 70%. 1H NMR (C6D6, 600 MHz, 293 K): δ 8.14 (s, 2H, aryl-H), 7.73 (s, 1H, aryl-H), 7.23 (t, J = 6 Hz, 1H, aryl-H), 7.11 (d, J = 6 Hz, 2H, aryl-H), 6.49 (s, 1H, N-H), 2.95 (septet, J = 6 Hz, 2H, CH(CH3)2), 1.18 (d, J = 6 Hz, 12H, CH(CH3)2). 13C NMR (C6D6, 375 MHz, 293 K): δ 162.59 (CdO), 146.0, 136.3, 131.6 (q, J = 86 Hz, CCF3), 130.4, 128.5, 126.9, 124.5, 123.2 (aryl C’s), 28.6 (CH(CH3)2), 23.1 (CH(CH3)2). IR data (KBr, cm-1): 3306 (br), 2967 (s), 2930 (s) 1647 (s), 1589 (w), 1525 (s), 1465 (s), 1374 (m), 1333 (w), 1274 (s), 1193 (s), 1140 (s), 911 (w), 797 (w). EIMS (m/z): 417 [M]+. Anal. Found (calcd for C21H21F6NO): C 60.52 (60.43), N 3.60 (3.36), H 5.28 (5.07). Synthesis of Bis(N-20 ,60 -diisopropylphenyl(naphthyl)amidate) mono(trimethylsilylamido)yttrium mono(tetrahydrofuran) (4). Inside a glovebox, a parallel synthetic apparatus tube was charged with yttrium tris(bis(trimethylsilyl)amide) (0.345 g, 0.605 mmol), 5 mL of tetrahydrofuran, and a stirbar. The reaction mixture was stirred until all solid was dissolved, and N-(diisopropylphenyl)naphthyl amide (1) (0.401 g, 1.21 mmol) dissolved in 5 mL of tetrahydrofuran was added dropwise. The solution was stirred within the glovebox for 2 h at 60 °C and then filtered through a pipet plug of Celite and concentrated under reduced pressure to a pale yellow solid. This solid was redissolved in toluene and stirred at 90 °C for a subsequent 2 h. The product was then concentrated again to a pale yellow solid and recrystallized by dissolving in hexanes with a few drops of toluene to dissolve all solid and then left at -30 °C to give a white crystalline solid. Yield: 0.450 g, 82%. 1H NMR (300 MHz, C6D6): δ 9.09 (d, J = 9 Hz, 2H, aryl-H), 7.49 (m, 4H, aryl-H), 7.36 (d, J = 8 Hz, 2H, aryl-H), 7.24 (m, 2H, aryl-H), 7.14 (m, 2H, aryl-H), 7.03 (m, 2H, aryl-H), 6.96 (m, 4H, aryl-H), 6.81 (t, J = 8 Hz, 2H, aryl-H), 3.94 (br t, J = 6 Hz, 4H, O-CH2), 3.67 (br septet, 4H, J = 7 Hz, CH(CH3)2), 1.23 (overlapping t and d, 16H, O-CH2CH2 and CH(CH3)2), 0.65 (d, J = 6 Hz, 12H, CH (CH3)2), 0.53 (s, 18H, N(Si(CH3)3)2). 13C NMR (100.6 MHz, C6D6): δ 179.6 (d, J = 2 Hz, CdO), 141.9, 141.4, 137.5, 134.3, 131.9, 131.5, 130.4, 128.9, 128.4, 126.8, 126.3, 125.5, 125.3, 124.5, 123.8, 123.6 (aryl-C’s), 69.9 (O-CH2), 28.0 (CH(CH3)2), 25.4 (O-CH2CH2), 24.9 (CH(CH3)2), 23.5 (CH(CH3)2), 4.6 (N(Si(CH3)3)2). IR data (KBr, cm-1): 2962 (w), 1511 (s), 1496 (s), 1400 (s), 1382 (s), 1245 (s), 964 (w), 842 (w), 828 (w), 779 (w) cm-1. EIMS (m/z): 909 [M+], 749 [M+ - N(Si(CH3)3)2], 331 [naphthyl [O,N]Dipp]. Anal. Found (calcd for C56H74N3O3Si2Y): C 68.15 (68.47), N 4.65 (4.28), H 7.97 (7.59). Synthesis of Bis(N-20 ,60 -diisopropylphenyl(p-(trifluoromethylphenyl)amidate) mono(trimethylsilylamido)yttrium mono(tetrahydrofuran) (5). The experimental method described for 4 was used in the preparation of 5 using 2 (0.400 g, 1.15 mmol) and yttrium tris(bis(trimethylsilyl)amide) (0.326 g, 0.572 mmol) to give a pale yellow solid. The product was recrystallized by dissolving in a minimum amount of hexanes, with a few drops of toluene, and then left at -30 °C to give a white crystalline solid. Yield: 0.432 g, 80%. 1H NMR (400 MHz, C6D6): δ 7.48 (d, J = 8 Hz, 4H, aryl-H), 7.10 (m, 6H, aryl-H), 7.01 (d, 4H, arylH), 3.94 (br s, 4H, O-CH2), 3.40 (septet, J = 7 Hz, 4H, CH(CH3)2), 1.18 (d, J = 7 Hz, 12H, CH(CH3)2), 1.15 (m, 4H, O-CH2CH2), 0.81 (d, J = 7 Hz, 12H, CH(CH3)2), 0.47 (s, 18H, N(Si(CH3)3)2). 13C NMR (100.6 MHz, C6D6): δ 175.0 (CdO), 141.4, 140.4, 136.5, 131.8 (q, J = 32 Hz, C(CF3)), 129.8, 125.2, 124.1, 123.7 (aryl-C’s), 69.9 (O-CH2), 27.8 (CH(CH3)2), 24.4 (O-CH2CH2), 24.0 (CH(CH3)2), 23.4 (CH(CH3)2). IR data (KBr, cm-1): 2964 (w), 1626 (s), 1528 (s), 1503 (s), 1410 (s), 1325 (s), 1170 (w), 1132 (s), 1067 (s), 1016 (w), 857 (s), 786 (w), 764 (w) cm-1.


3997

EIMS (m/z): 945 [M+], 784 [M+ - N(SiMe3)2], 639 [M+ - N(SiMe3)2 - pCF3phenyl[O,N]Dipp], 349 [pCF3phenyl[O,N]Dipp]. Anal. Found (calcd for C50H68F6N3O3Si2Y): C 58.33 (58.98), N 4.00 (4.13), H 6.56 (6.73). Synthesis of Bis(N-20 ,60 -diisopropylphenyl((3,5-bis(trifluoromethyl)phenyl)amidate) mono(trimethylsilylamido)yttrium mono(tetrahydrofuran) (6). The experimental method described for 4 was used in the preparation of 6 using 3 (0.201 g, 0.481 mmol) and yttrium tris(bis(trimethylsilyl)amide) (0.138 g, 0.242 mmol) to give a pale yellow solid. The product was recrystallized by dissolving in a minimum amount of hexanes, with a few drops of toluene, and then left at -30 °C to give a white crystalline solid. Yield: 0.219 g, 84%. 1H NMR (600 MHz, C6D6): δ 8.00 (s, 4H, aryl-H), 7.60 (s, 2H, aryl-H), 7.06 (m, 2H, arylH), 7.02 (d, J = 6 Hz, 4H, aryl-H), 3.91 (br s, 4H, O-CH2), 3.38 (br septet, 4H, J = 6 Hz, CH(CH3)2), 1.16 (d, J = 6 Hz, 12H, CH(CH3)2), 1.05 (br s, 4H, O-CH2CH2) 0.78 (d, 12H, J = 6 Hz, CH(CH3)2), 0.40 (s, 18H, N(Si(CH3)3)2). 13C NMR (150.9 MHz, C6D6): δ 173.0 (CdO), 140.7, 140.3, 135.3, 130.8 (q, J = 33 Hz, C(CF3)), 129.7, 127.6, 127.2, 125.6, 124.0, 123.5, 122.7 (q, J = 270 Hz, C(CF3)) (aryl-C’s), 69.9 (O-CH2), 27.7 (CH(CH3)2), 24.2 (O-CH2CH2), 23.9 (CH(CH3)2), 23.2 (CH(CH3)2), 4.1 (N(Si(CH3)3)2). IR data (KBr, cm-1): 2963 (w), 1623 (w), 1529 (s), 1463 (w), 1400 (s), 1349 (s), 1279 (s), 1185 (s), 1136 (s), 961 (w), 908 (w), 839 (w), 801 (w), 702 (w), 682 (w) cm-1. EIMS (m/z): 1082 [M+], 1066 [M+ - CH3], 921 [M+ - N(Si(CH3)3)2], 665 [M+ - N(Si(CH3)3)2 and 3,5-bisCF3(phenyl)[O,N] Dipp], 416 [3,5-bisCF3[O,N]Dipp]. Anal. Found (calcd for C52H66F12N3O3Si2Y): C 54.50 (54.11), N 3.75 (3.64), H 5.81 (5.76). Synthesis of Mono(N-20 ,60 -diisopropylphenyl(naphthyl)amidate)bis(trimethylsilylamido)yttrium mono(tetrahydrofuran) (7). Inside a glovebox, a parallel synthetic apparatus tube was charged with yttrium tris(bis(trimethylsilyl)amide) (0.401 g, 0.701 mmol), 10 mL of tetrahydrofuran, and a stirbar. The reaction mixture was stirred until all solid was dissolved, and 1 (0.401 g, 1.21 mmol) dissolved in 10 mL of tetrahydrofuran was added very slowly (approximately over 10 min) to the stirring solution of yttrium tris(bis(trimethylsilyl)amide) at room temperature. The solution was stirred within the glovebox for 2 h and then filtered through a pipet plug of Celite and concentrated under reduced pressure to a white solid. The product was recrystallized by dissolving in a minimum amount of hexanes and then left at -30 °C to give colorless plates. Yield: 0.443 g, 77%. 1H NMR (600 MHz, C6D6): δ 9.16 (d, J = 6 Hz, 1H, aryl-H), 7.54 (t, J = 6 Hz, 1H, aryl-H), 7.48 (d, J = 6 Hz, 1H, aryl-H), 7.33 (t, J = 6 Hz, 2H, aryl-H), 7.23 (t, 1H, J = 6 Hz, aryl-H), 6.97 (m, 3H, aryl-H), 6.75 (t, J = 6 Hz, 1H, aryl-H), 3.81 (br t, J = 6 Hz, 4H, O-CH2), 3.48 (septet, 4H, J = 6 Hz, CH(CH3)2), 1.20 (d, 6H, J = 6 Hz, CH(CH3)2), 1.13 (br t, J = 6 Hz, 4H, O-CH2CH2), 0.70 (d, J = 6 Hz, 6H, CH(CH3)2), 0.49 (s, 32H, N(Si(CH3)3)2). 13C NMR (150.9 MHz, C6D6): δ 179.9 (CdO), 142.8, 141.9, 135.1, 132.7, 132.0, 131.9, 131.2, 129.2, 128.7, 128.6, 127.6, 127.3, 126.4, 125.5, 125.0, 124.2 (aryl-C’s), 72.4 (O-CH2), 28.3 (CH(CH3)2), 26.2 (O-CH2CH2), 25.2 (CH(CH3)2), 24.5 (CH(CH3)2), 6.1 (N(Si(CH3)3)2). IR data (KBr, cm-1): 2963 (w), 1516 (s), 1497 (s), 1399 (s), 1379 (s), 1245 (s), 956 (w), 863 (w), 844 (w), 668 (w) cm-1. EIMS (m/z): 739 [M+], 724 [M+ - CH3], 578 [M+ - N(Si(CH3)3)2]. Anal. Found (calcd for C39H68N3O2Si4Y): C 57.82 (57.67%), N 5.46 (5.17), H 8.38 (8.44). Crystallography. Clear, colorless crystals of complex 4 suitable for X-ray analysis were obtained by cooling a concentrated hexanes solution (with a few added drops of toluene) to -30 °C. A concentrated hexanes solution of compound 7 was cooled to -30 °C to obtain clear, colorless X-ray quality crystals. All data were collected on a Bruker X8 APEX II area detector and are summarized in Table S-1 in the Supporting Information. Typical Procedure for Hydroamination Using Amidate Complexes (for entry 1, Table 3). Inside an inert-atmosphere glovebox, complex 4 (24.11 mg, 0.025 mmol, 10 mol %), 1,3,5-trimethoxybenzene (14.2 mg, 0.084 mmol), and 2,2-diphenyl-4-pentenylamine (8) (60.0 mg, 0.253 mmol) were weighed out in separate vials,

3998


dissolved, and combined in approximately 0.7 g of deuterated benzene inside a J-Young sealable NMR tube. The 1H NMR spectrum was immediately obtained (approximately 15 min delay) to monitor conversion (>99%). By comparison of the integration values for the proton signals of 1,3,5-trimethoxybenzene (δ 6.25 and 3.32 for aryl-H and -O(CH3) protons, respectively) to the newly formed CH(CH3) signal (2.47 ppm) in the 1H NMR spectrum, the NMR yield is obtained (>95%). The reaction was exposed to air and quenched with dichloromethane. The residual material was added to a small frit silica column and initially flushed with 200 mL of a 1:1 solution of hexanes and ethyl acetate (to remove residual ligand). The column was then flushed with 200 mL of 10% methanol and 1% isopropylamine in dichloromethane, which was collected and dried in vacuo to give the

Stanlake and Schafer product 9 (57.0 mg, 95% yield). NMR yield is typically larger than isolated yield, due to loss of product during the isolation procedure.

Acknowledgment. The authors thank UBC, CFI, BCKDF, and DuPont for financial support of this work. L.J.E.S. thanks UBC and MEC for scholarships, as well as N. Yonson, R. K. Thomson, and B. O. Patrick for X-ray crystallographic analysis. Supporting Information Available: Crystallographic information files (CIF) of complexes 4 and 7. This material is available free of charge via the Internet at http://pubs.acs.org.

and Mono(amidate) Complexes of Yttrium - American Chemical Society

Recommend Documents