Aryl-Containing Pyridine-Imine and Azaallyl Chelates of Iron toward

364

Organometallics 2010, 29, 364–377 DOI: 10.1021/om900793c

Aryl-Containing Pyridine-Imine and Azaallyl Chelates of Iron toward Strong Field Coordination Compounds Emily C. Volpe, Peter T. Wolczanski,* and Emil B. Lobkovsky Department of Chemistry & Chemical Biology, Baker Laboratory, Cornell University, Ithaca, New York 14853 Received September 11, 2009

A series of aryl- and pyridyl-imine iron complexes have been prepared to explore the generation of strong fields provided by aryl chelates and the potential support of an azaallyl functionality. cis-(Me3P)4FeMe2 and imino-based ligand precursors afforded trans-{κ-C,N,N0 -(XAr-2-yl)CH2NdCH-2-py}(PMe3)2FeIICH3 (X = H, 1; p-OMe, 2; o-Cl, 3) and trans-{κ-C,N,N0 -(p-tBu-Ar-2-yl)CHdNCH2-2-py}(PMe3)2FeIICH3 (4) concomitant with 2 PMe3 and CH4 via ArH activation. Thermolysis of 4 þ PMe3 gave CH4 and py CH activation to provide mer-{κ-C,N,C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-py-3-yl)}(PMe3)3FeII (6). Derivatization of 6 with MeOTf produced [mer-{κ-C,N, C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-py-NCH3-3-yl)}(PMe3)3FeII]OTf (7), which could be deprotonated to give the zwitterionic azaallyl mer-{κ-C,N,C0 -(p-tBu-Ar-2-yl)CHNCH(2-py-NCH3-3-yl)}(PMe3)3FeII (8). Oxidation of 4 with AgOTf generated [trans-{κ-C,N,N0 -(p-tBu-Ar-2-yl)CHdNCH22-py}(PMe3)2FeIIICH3]OTf (9), and its deprotonation with KH produced azaallyl trans-{κ-C,N, N0 -(p-tBu-Ar-2-yl)CHNCH-2-py}(PMe3)2FeIIICH3 (10). Exposure of cis-(Me3P)4FeMe2 to diarylimines gave double arene CH activation products {mer-κ-C,N,C0 -(Ar-2-yl)CH2NdCH(Ar-2-yl-X)}Fe(PMe3)3 (X = H, 11; 4-CF3, 12; 3-OMe, 13) with loss of 2 CH4 and PMe3. Treatment of the parent diarylimine complex (11) with AgOTf afforded [mer-κ-C,N,C0 -{(Ar-2-yl)CH2NdCH(Ar-2-yl-X)}FeIII(PMe3)3]OTf (14), but attempts to deprotonate it to form an azaallyl failed. Structures of 7, 8, and 14 are detailed. Discussions regarding the feasibility of aryl chelates and utilization of the azaallyl functionality to support high oxidation states are provided, in addition to an interpretation of the UV-vis spectra of the azaallyls.

Introduction Previous work in these laboratories has introduced the concept of using metal-aryl frameworks to increase the field strengths about first-row transition metal species.1 According to the angular overlap model,2,3 as shown in Figure 1, C-based orbitals have better interaction energies than N- or O-based orbitals due to their relative proximity to appropriate metal orbitals. Aryl ligands also possess better orbital *Corresponding author. E-mail: [email protected]. (1) Frazier, B. A.; Wolczanski, P. T.; Lobkovsky, E. B. Inorg. Chem. 2009, 131, 11576–11585. (2) Figgis, B. N.; Hitchman, M. A. Ligand Field Theory and Its Applications; Wiley-VCH: New York, 2000. (3) Richardson, D. E. J. Chem. Educ. 1993, 70, 372. (4) (a) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of Organotransition Metal Chemistry; University Science Books: Mill Valley, CA, 1987. (b) Crabtree, R. H. The Organometallic Chemistry of the Transition Metals, 4th ed.; John Wiley & Sons: New York, 2005. (5) Bower, B. K.; Tennent, H. G. J. Am. Chem. Soc. 1972, 94, 2512– 2513. (6) Byrne, E. K.; Theopold, K. H. J. Am. Chem. Soc. 1989, 111, 3887– 3896. (7) Dimitrov, V.; Linden, A. Angew. Chem., Int. Ed. 2003, 42, 2631– 1633. (8) Carnes, M.; Buccella, D.; Chen, J. Y. C.; Ramirez, A. P.; Turro, N. J.; Nuckolls, C.; Steigerwald, M. Angew. Chem., Int. Ed. 2009, 48, 290–294. pubs.acs.org/Organometallics

Published on Web 12/16/2009

overlap (SC > SN > SO) than nitrogen and oxygen congeners with consequential metal orbitals, reinforcing the idea that the potential for imparting strong fields is substantial.4-8 Strong fields may be necessary for first-row transition metal complexes to undergo the 2e- changes so important for bond-making and -breaking steps in many catalytic processes.4 Often 1e- changes that are relatively common to the first row can disrupt the transformations desired in catalysis, and weaker field species are generally more prone to such events.9-11 A substantial fraction of current catalysis is carried out by second-row transition metal species that are expensive and typically pose a greater health hazard;as trace impurities in drugs, for example; than many first-row transition elements.4 Recent investigations point to a greater density of states (DOS) for secondrow species relative to those of the third row due in part to greater 6s/5d mixing in the latter.12 The greater DOS enables reactant to product paths in second-row (9) (a) Klinker, E. J.; Shaik, S.; Hirao, H.; Que, L. Angew. Chem., Int. Ed. 2009, 48, 1291–1295. (b) Dhuri, S. N.; Seo, M. S.; Lee, Y. M.; Hirao, H.; Wang, Y.; Nam, W.; Shaik, S. Angew. Chem., Int. Ed. 2008, 47, 3356–3359. (10) De Angelis, F.; Jin, N.; Car, R.; Groves, J. T. Inorg. Chem. 2006, 45, 4268–4276. (11) For elucidation of catalysis involving 1 e- changes, see: Poli, R. Angew. Chem., Int. Ed. 2006, 45, 5058–5070. (12) Hirsekorn, K. F.; Hulley, E. B.; Wolczanski, P. T.; Cundari, T. R. J. Am. Chem. Soc. 2008, 130, 1183–1196. r 2009 American Chemical Society

Article

Organometallics, Vol. 29, No. 2, 2010

365

higher oxidation states. During the course of investigating tetradentate chelates possessing Ar- and N-donors, an intriguing nitrogen-based ligand, 1,3-di-2-pyridyl-2-azaallyl (smif, Figure 2),27 was discovered via C-N bond activation.1 Since smif was shown to possess a reasonably strong field, replacement of one or both of the pyridines with aryl groups was considered to produce ligand frameworks potentially capable of supporting even stronger fields and higher oxidation states. Figure 2 illustrates the spate of smif analogues targeted in this report: monoanionic imine ligands ArCH2Impy and pyCH2Impy, which are potential precursors to the dianionic arylazaallylpyridine ArCHNCHpy, and dianionic imine ArCH2ImAr0 , which is a potential precursor to trianionic diarylazaallyl ArCHNCHAr0 . An investigation into the syntheses and properties of iron complexes28-31 containing some of these ligands is presented herein. Figure 1. Angular overlap arguments show that both the interaction energy and orbital overlap favor C-based over N- and Obased ligands in terms of field strength.

reactivity patterns that are energetically higher for thirdrow species.12-26 UV-visible spectroscopy clearly indicates that first-row transition metal species have a greater density of states (DOS) than related second-row species.2 The challenge is to generate first-row transition metal complexes more like the second row in terms of field strength and DOS, in the hope of bypassing detrimental reactivity. One way of probing stronger interactions is to seek covalent ligand frameworks that can support low-spin configurations and (13) (a) Kuiper, D. S.; Douthwaite, R. E.; Mayol, A.-R.; Wolczanski, P. T.; Lobkovsky, E. B.; Cundari, T. R.; Lam, O. P.; Meyer, K. Inorg. Chem. 2008, 47, 7139–7153. (b) Kuiper, D. S.; Wolczanski, P. T.; Lobkovsky, E. B.; Cundari, T. R. J. Am. Chem. Soc. 2008, 130, 12931–12943. (c) Kuiper, D. S.; Wolczanski, P. T.; Lobkovsky, E. B.; Cundari, T. R. Inorg. Chem. 2008, 47, 10542–10553. (14) Rosenfeld, D. C.; Wolczanski, P. T.; Barakat, K. A.; Buda, C.; Cundari, T. R.; Schroeder, F. C.; Lobkovsky, E. B. Inorg. Chem. 2007, 46, 9715–9735. (15) Veige, A. S.; Slaughter, L. M.; Lobkovsky, E. B.; Wolczanski, P. T.; Matsunaga, N.; Decker, S. A.; Cundari, T. R. Inorg. Chem. 2003, 42, 6204–6224. (16) Rosenfeld, D. C.; Kuiper, D. S.; Lobkovsky, E. B.; Wolczanski, P. T. Polyhedron 2006, 25, 251–258. (17) Wolczanski, P. T. Chem. Commun. 2009, 740–757. (18) (a) Poli, R. J. Organomet. Chem. 2004, 689, 4291–4304. (b) Poli, R. Acc. Chem. Res. 1997, 30, 1861–1866. (19) Matsunaga, N.; Koseki, S. Rev. Comput. Chem. 2004, 20, 101– 152. (20) Carreon-Macedo, J.; Harvey, J. N.; Poli, R. Eur. J. Inorg. Chem. 2005, 12, 2999–3008. (21) Poli, R.; Cacelli, I. Eur. J. Inorg. Chem. 2005, 12, 2324–2331. (22) (a) Petit, A.; Richard, P.; Cacelli, I.; Poli, R. Chem.;Eur. J. 2006, 12, 813–823. (b) Smith, K. M.; Poli, R.; Harvey, J. N. Chem.;Eur. J. 2001, 7, 1679–1690. (23) (a) Schrock, R. R.; Seidel, S. W.; Mosch-Zanetti, N. C.; Shih, K. Y.; O’Donoghue, M. B.; Davis, W. M.; Reiff, W. M. J. Am. Chem. Soc. 1997, 119, 11876–11893. (b) Schrock, R. R.; Seidel, S. W.; MoschZanetti, N. C.; Dobbs, D. A.; Shih, K. Y.; Davis, W. M. Organometallics 1997, 16, 5195–5208. (24) (a) Buccella, D.; Tanski, J. M.; Parkin, G. Organometallics 2007, 26, 3275–3278. (b) Buccella, D.; Parkin, G. J. Am. Chem. Soc. 2006, 128, 16358–16364. (c) Churchill, D. G.; Janak, K. E.; Wittenberg, J. S.; Parkin, G. J. Am. Chem. Soc. 2003, 125, 1403–1420. (25) (a) Fan, L.; Parkin, S.; Ozerov, O. V. J. Am. Chem. Soc. 2005, 127, 16772–16773. (b) Weng, W.; Guo, C. Y.; Moura, C.; Yang, L.; Foxman, B. M.; Ozerov, O. V. Organometallics 2005, 24, 3487–3499. (26) Soo, H. S.; Figueroa, J. S.; Cummins, C. C. J. Am. Chem. Soc. 2004, 126, 11370–11376.

Results Imine Syntheses. The precursors to the aryliminopyridine and azaallyl ligands in Figure 2 are imines (Im1-Im9) that were synthesized via condensation of py-2-CH2NH2 with the appropriate benzaldehyde, through condensation of 2-pyridinecarboxaldehyde with the pertinent benzylamine or by condensation of the benzaldehyde and benzylamine via standard procedures.32-37 Metalations. 1. ArH and ArCl Bond Activation Attempts. Heterolytic ArH bond activation features the metathesis of an M-X linkage to afford M-Ar and HX, where X is typically a heteroatom-based anionic ligand or halide. Modest success with metal triflates enabled related chemistry with nickel,38 but similar efforts utilizing (o-Cl-ArH)CH2Impy (Im3) with various iron(II) complexes failed. When Im3 was added to FeX2(THF)2 (X = Cl, Br), color changes symptomatic of adduct formation occurred, but no heterolytic bond activation was noted, and attempts at subsequent reduction led to free imine. In the case of Fe{N(TMS)2}2(THF)39 and Im3, a simple color change (27) Frazier, B. A.; Wolczanski, P. T.; Lobkovsky, E. B.; Cundari, T. R. J. Am. Chem. Soc. 2009, 131, 3428–3429. (28) (a) de Meijere, A. Chem. Rev. 2000, 100, 2739–2740. (b) F€urstner, A. Angew. Chem., Int. Ed. 2009, 48, 1364–1367. (29) Ritter, S. K. Chem. Eng. News 2008, 86, 53–57. (30) (a) Sherry, B. D.; Furstner, A. Acc. Chem. Res. 2008, 11, 1500– 1511. (b) Furstner, A.; Martin, R. Chem. Lett. 2005, 34, 624–629. (c) Furstner, A.; Leitner, A.; Mendez, M.; Krause, H. J. Am. Chem. Soc. 2002, 124, 13856–13863. (31) Buchwald, S. L.; Bolm, C. Angew. Chem., Int. Ed. 2009, 48, 0000. (32) Im1: Ceder, R. M.; Muller, G.; Ordinas, M.; Ordinas, J. I. Dalton Trans. 2007, 1, 83-90. (33) Im2, Im5: Grigg, R.; Donegan, G.; Gunaratne, H. Q. N.; Kennedy, D. A.; Malone, J. F.; Sridharan, V.; Thianpatanagul, S. Tetrahedron 1989, 45, 1723-1746. (34) Im3: Kouznetsov, V. V.; Castro, J. R.; Puentes, C. O.; Stashenko, E. E.; Martinez, J. R.; Ochoa, C.; Pereira, D. M.; Ruiz, J. J. N.; Portillo, C. F.; Serrano, S. M.; Barrio, A. G.; Bahsas, A.; Amaro-Luis, J. Arch. Pharm. (Weinheim, Ger.) 2005, 338, 32-37. (35) Im7: Bowman, R. K.; Johnson, J. S. J. Org. Chem. 2004, 69, 8537-8540. (36) Im8: Kim, M.; Knettle, B. W.; Dahlen, A.; Hilmersson, G.; Flowers, R. A. Tetrahedron 2003, 59, 10397-10402. (37) Im9: Saha, M.; Chandrasekaran, S. Bangl. J. Sci. Ind. Res. 1999, 34, 120-123. (38) Volpe, E. C.; Chadeayne, A. R.; Wolczanski, P. T.; Lobkovsky, E. B. J. Organomet. Chem. 2007, 692, 4774–4783. (39) (a) Olmstead, M. M.; Power, P. P.; Shoner, S. C. Inorg. Chem. 1991, 30, 2547–2551. (b) Andersen, R. A.; Faegri, K.; Green, J. C.; Haaland, A.; Lappert, M. F.; Leung, W. P.; Rypdal, K. Inorg. Chem. 1988, 27, 1782– 1786.

366


Volpe et al.

Figure 2. Azaallyl ligand smif and its parent imine; monoanionic imine ligands ArCH2Impy and pyCH2Impy and dianionic imine ArCH2ImAr0 ; dianionic arylazaallylpyridine Araapy and trianionic diarylazaallyl AraaAr0 .

again implied adduct formation, but no tractable products ensued. Another ArH bond activation features the swapping of one M-Ar bond for another M-R. Since such activations are not heterolytic in character, they often fall under the aegis of σ-bond metathesis, even though the reactions are just as likely to involve formal oxidative addition/reductive elimination paths. When [Fe(mesityl)2]240 was treated with (o-Cl-ArH)CH2Impy (Im3), the solution changed color from brown to blue and the corresponding increase in intensity was typical of azaallyl formation,1,27 but severe decomposition was evident upon NMR spectral analysis. The choice of (o-Cl-ArH)CH2Impy (Im3) as the initial iminopyridine ligand was made due to the alternate possibility of preparing Fe-Ar bonds via ArCl oxidative addition.41 Previous studies had indicated that disproportionation of Fe(II) species;principally of Fe{N(TMS)2}2(THF);was a means toward “Fe(0)” (plus 2 Fe(III)) and oxidative addition.1 This methodology was sought to obviate the use of metal carbonyls as M(0) sources in order to avoid the complexity of metal acyl formation. No clear evidence of ArCl activation was obtained in the initial studies, and the intact iminopyridine was often identified in the 1H NMR of the degradation products. 2. Arylation via MeH Loss from cis-(Me3P)4FeMe2. Literature reports of aryl CH bond activation of aryl imines41-45 by Karsch’s cis-(Me3P)4FeMe2 complex46 prompted efforts to affect the related arylation of iminopyridines. As Scheme 1 (40) Klose, A.; Solari, E.; Ferguson, R.; Floriani, C. Organometallics 1993, 12, 2414–2416. (41) Shi, Y.; Li, M.; Hu, Q.; Li, X.; Sun, H. Organometallics 2009, 28, 2206–2210. (42) Klein, H.-F.; Camadanli, S.; Beck, R.; Fl€ orke, U. Chem. Commun. 2005, 381–382. (43) Klein, H.-F.; Camadanli, S.; Beck, R.; Leukel, D.; Fl€ orke, U. Angew. Chem., Int. Ed. 2005, 44, 975–977. (44) Beck, R.; Zheng, T.; Sun, H.; Li, X.; Fl€ orke, U.; Klein, H.-F. J. Organomet. Chem. 2008, 693, 3471–3478. (45) Camadanli, S.; Beck, R.; Fl€ orke, U.; Klein, H.-F. Organometallics 2009, 28, 2300–2310. (46) Karsch, H. H. Chem. Ber. 1977, 110, 2699–2711.

reveals, successful metalation of iron was achieved with several iminopyridines (Im1-3) and 2-pyridylmethyl-p-tertbutylbenzaldimine (Im4) via the loss of 2 equiv of PMe3 and 1 equiv MeH. While the aryl-activated iminopyridine products trans-{κ-C,N,N0 -(Ph-2-yl)CH2NdCH-2-py}(PMe3)2FeCH3 (1), trans-{κ-C,N,N0 -(p-MeO-Ar-2-yl)CH2NdCH-2py}(PMe3)2FeCH3 (2), and trans-{κ-C,N,N0 -(o-Cl-Ar-2-yl)CH2NdCH-2-py}(PMe3)2FeCH3 (3) were isolated as dark green microcrystals, the benzaldimine product, trans-{κ-C,N, N0 -(p-tBu-Ar-2-yl)CHdNCH2-2-py}(PMe3)2FeCH3 (4), was dark blue. Attempts to prepare sp3-activated products were made via treatment of cis-(Me3P)4FeMe2 with ethyliminopyridine47 and neopentyliminopyridine,48 but purple iminopyridine adducts trans,cis-(PMe3)2(CH3)2Fe{κ-N,N0 -RCH2Nd CH-2-py} (R = Me, 5-Et; tBu, 5-neoPe) formed intstead with trace byproducts. Subsequent thermolyses led to degradation, and no tractable sp3-activated products could be obtained; hence these species were not pursued nor fully characterized aside from routine NMR studies. Tables 1 and 2 contain the NMR spectral data for all new compounds and show that 31P{1H} NMR spectra enabled the stereochemical assignments for 1-4, which all manifested singlets consistent with a trans disposition of the phosphines. Reactivity of Aryl Imine and Iminopyridine Complexes. 1. Attempts at Ligand Exchange. Carbon-carbon bond-forming reactions can be envisaged with compounds 1-4, but it is readily apparent that some form of ligand exchange must occur for any catalytic process.49 Treatment of trans-{κ-C,N,N0 -(Ph-2-yl)CH2NdCH-2-py}(PMe3)2FeCH3 (1) with Im3 and the corresponding treatment of trans-{κ-C,N,N0 -(o-Cl-Ar-2-yl)CH2Nd CH-2-py}(PMe3)2FeCH3 (3) with Im1 failed to elicit any ligand exchange, even at elevated temperatures. No generation of {κ-C, (47) Vasconcellos-Dias, M.; Nunes, C. D.; Vaz, P. D.; Ferreira, P.; Brand~ao, P.; Felix, V.; Calhorda, M. J. J. Catal. 2008, 256, 301–311. (48) Zoet, R.; van Koten, G.; Vrieze, K. Inorg. Chim. Acta 1988, 148, 71–84. (49) Hull, K. L.; Lanni, E. L.; Sanford, M. S. J. Am. Chem. Soc. 2006, 128, 14047–14048.

Article


367

Scheme 1

N,N0 -(Ar-2-yl)CH2NdCH-2-py}2Fe was noted during these studies, indicating that the remaining two trimethylphosphines and the iron methyl functionality were not easily lost during thermolyses. The addition of ArCH2NH2 to 1 did not incur a change with the benzylamine portion of the chelate; hence exchanges of this type were not considered readily viable. 2. Thermal Activity. Although imine ligand exchange was not observed, thermolyses of trans-{κ-C,N,N0 -(o-Cl-Ar-2-yl)CH2NdCH-2-py}(PMe3)2FeCH3 (3) and trans-{κ-C,N,N0 ( p-tBu-Ar-2-yl)CHdNCH2-2-py}(PMe3)2FeCH3 (4) provided conflicting and interesting results, as shown in Scheme 2. The o-Cl derivative 3 was chosen for study because if Ar-Me reductive elimination occurred, the Fe(0) center generated could be trapped by oxidative addition of the Ar-Cl bond.41 When heated with or without the presence of PMe3, 3 degraded to a myriad of products. However, the corresponding thermolysis (80 °C, C6D6) of 4, in which the pyridine is attached to the methylene rather than the imine linkage, led to the formation of mer-{κ-C,N,C0 ( p-tBu-Ar-2-yl)CHdNCH2(2-py-3-yl)}(PMe3)3Fe (6) along with some degradation products. The addition of PMe3 to 4 suppressed byproduct formation, and scale-up of the process afforded 6, in which the 3-position of the pyridine contains a second Fe-Ar bond, as yellow-brown crystals in 67% yield. COSY and HMBC50 NMR spectroscopic correlations were used to assign the 1H and 13C resonances listed in Table 1. The 31 P{1H} NMR spectrum revealed an A2B pattern51 centered at δ 24.77 (Pax) and δ 19.60 (2Ptrans) with JPP = 61 Hz, consistent with the expected mer-configuration of the pseudooctahedral structure resulting from elimination of CH4, which was observed in the sealed NMR tube thermolysis. It can only (50) Balci, M. Basic 1H-13C-NMR Spectroscopy; Elsevier: New York, 2005. (51) Abraham, R. J.; Fisher, J.; Loftus, P. Introduction to NMR Spectroscopy; John Wiley & Sons, Ltd.: New York, 1988.

be speculated that the additional flexibility accorded the pyridylmethyl linkage enables it to achieve a configuration where the 3-position of the pyridine can be activated to provide a hydrogen for methane elimination. Apparently the rigidity of the iminopyridine moiety in 3 renders pyridine activation uncompetitive with degradation. While it is conceivable that the o-Cl of 3 and/or its configuration promoted Ar-Me reductive elimination, no evidence of Ar-Me bond formation38 was observed in the products resulting from thermolysis. 3. 3-Pyridyl Methylation. Although the spectral assignments of mer-{κ-C,N,C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-py-3yl)}(PMe3)3Fe (6) seemed definitive, X-ray quality crystals of 6 could not be obtained for corroboration. A derivative was sought that could potentially lead to the desired azaallyl functionality (Scheme 2). Alkylation of the 3-pyridyl-nitrogen of 6 was accomplished via MeOTf in benzene, producing the brown crystalline cationic complex [mer-{κ-C, N,C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-py-NCH3-3-yl)}(PMe3)3Fe]OTf (7) in 77% isolated yield. Observation of the A2B pattern in the 31P{1H} NMR spectrum (δ 23.02 (Pax), 18.82 (2Ptrans), JPP = 61 Hz)50,51 supported the retention of the merpseudo-octahedral structure of 7, which was verified by X-ray crystallography. 4. Zwitterionic Azaallyl Formation. The cation [mer-{κ-C, N,C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-py-NCH3-3-yl)}(PMe3)3Fe]OTf (7) contains relatively acidic backbone methylene protons that were deemed susceptible to deprotonation. As Scheme 2 illustrates, exposure of 7 to excess KH in THF at -78 °C produced the neutral, dark green zwitterion mer-{κ-C, N,C0 -(p-tBu-Ar-2-yl)CHNCH(2-py-NCH3-3-yl)}(PMe3)3Fe (8), in 61% yield. 1H NMR spectra of 8 revealed that the singlet due to the methylene hydrogens of 7 at δ 5.47 was replaced by a 1 H signal at δ 5.84, while the 31P{1H} NMR spectrum revealed another A2B pattern at δ 27.33 (Pax) and 20.74 (JPP = 59 Hz),50,51 once again suggesting that

368


Volpe et al.

Table 1. 1H and 31P{1H} NMR Spectral Assignments (δ (J(Hz), assnmt)a,b for Imine and Imine-Derived Complexes

1

9.01 (s, a), 7.04 (m, c), 7.00 (t, 5, d), 6.87 (m, e), 8.63 (d, 5, f), 7.21 (d, 5, i), 7.04 (m, j), 7.10 (t, 8, k), 6.92 (d, 5, l), 4.91 (s, n), 0.38 (s, o), 1.52 (t, 12, p); 6.37 (s, PMe3, o)

2

9.00 (s, a), 7.01 (m, c), 6.66 (d, 8, d), 6.58 (dd, 3, 6, e), 8.62 (d, 5, f), 7.01 (m, i), 7.13 (d, 8, k), 6.93 (d, 7, l), 4.94 (t, 4, n), 0.39 (t, 4, o), 1.47 (t, 14, p), 3.59 (s, OMe, q); 6.68 (s, PMe3, o)

3

8.93 (s, a), 7.04 (t, 5, c), 6.80 (t, 8, d), 6.72 (m, e), 8.46 (d, 5, f), 6.96 (m, i), 6.87 (m, j), 6.96 (m, k), 5.24 (t, 5, n), 0.33 (t, 5, o), 1.36 (t, 15, p); 8.44 (s, PMe3, o)

4

4.42 (s, a), 6.14 (m, c), 6.59 (t, 8, d), 6.14 (m, e), 8.05 (d, 5, f), 8.75 (s, i), 7.11 (d, 10, k), 7.57 (d, 10, l), 7.90 (s, n), 0.69 (s, o), -0.23 (t, 10, p), 1.52 (s, q); 22.84 (s, PMe3, o)

5-Et

9.18 (s, a), 7.05 (d, 8, c), 7.12 (t, 7, d), 7.20 (t, 6, e), 9.23 (d, 6, f), 0.49 (s, o), 1.29 (t, 13, p), -0.01 (t, 11, s), 3.96 (t, 7 t), 2.16 (“sx”, 8, u(CH2)), 1.11 (t, 7, u(CH3)), -0.59 (s, PMe3, o)

5-neoPec9.31 (a), 7.03 (c), 7.13 (d), 7.20 (e), 9.46 (f), 0.51 (s, o), 1.12 (p), 0.13 (s), 3.80 (t), 1.28 (u); -1.43 (s, PMe3, o) 6

5.03 (br s, a), 8.40 (d, 2, d), 6.86 (dd, 8, 2, e), 7.91 (d, 8, f), 8.09 (s, i), 7.03 (d, 8, k), 7.34 (d, 8, l), 7.86 (br s, n), 0.55 (t, 4, o), 1.32 (d, 8, p), 1.43 (s, q); 24.77 (“t”,d 61, PMe3, p), 19.60 (d, 61, PMe3, o)

7e

5.47 (s, a), 7.87 (d, 6, d), 7.01 (t, 7, e), 8.34 (d, 7, f), 8.02 (s, i), 6.92 (d, 8, k), 7.40 (d, 8, l), 8.68 (br s, n), 0.80 (t, 3, o), 1.58 (d, 5, p), 1.33 (s, q), 4.14 (s, r); 23.02 (“t”,d 61, PMe3, p), 18.82 (d, 61, PMe3, o)

8

5.84 (s, a), 6.73 (d, 6, d), 5.51 (t, 6, e), 5.80 (d, 6, f), 8.06 (s, i), 7.03 (d, 8, k), 7.18 (d, 8, l), 7.37 (s, n), 0.99 (s, o), 1.33 (d, 6, p), 1.50 (s, q), 2.38 (s, r); 27.33 (“t”,d 59, PMe3, p), 20.74 (d, 59, PMe3, o)

9e,f

40.80 (1H), 28.75 (1H), 19.88 (1H), 16.92 (1H), 13.35 (2H, a), 1.29 (1H), 1.16 (9H, q), -0.46 (1H), -5.22 (3H), -15.30 (18H, o), -55.30 (1H)

10e,f

30.12 (1H), 22.03 (1H), 13.76 (1H), 12.73 (1H), 1.91 (9H, q), 1.77 (3H), -1.71 (1H), -1.97 (1H), -12.36 (1H), -14.39 (18H, o), -27.00 (1H), -47.97 (1H)

11g

4.79 (s, a), 7.03 (d, 8, c), 7.11 (t, 7, d), 7.12 (t, 7, e), 7.81 (d, 7 f), 8.13 (d, 8, i), 7.18 (m, j), 7.18 (m, k), 7.46 (d, 8 l), 8.02 (br s, n), 0.59 (s, o), 1.33 (d, 6, p); 23.73 (“t”,d 61, PMe3, p), 19.43 (d, 61, PMe3, o)

12

4.74 (br s, a), 7.00 (d, c), 7.11 (t, d), 7.14 (m, e), 7.73 (d, f), 8.51 (s, i), 7.28 (m, k), 7.28 (m, l), 7.94 (br s, n), 0.47 (t, o), 1.30 (d, p); 22.28 (“t”,d 63, PMe3, p), 18.01 (d, 63, PMe3, o)

13

4.84 (br s, a), 6.99 (d, 8, c), 7.09 (t, 7, d), 7.13 (m, e), 7.77 (d, 8, f), 7.93 (d, 8, j), 7.18 (m, k), 7.02 (d, 8, l), 8.03 (br s, n), 0.61 (t, 3, o), 1.36 (d, 6, p); 24.16 (“t”,d 62, PMe3, p), 19.51 (d, 62, PMe3, o)

14h

143.78 (2H, n), 37.27 (1H), 28.61 (1H), 17.42 (1H), 17.36 (1H), -14.21 (18H, o), -15.99 (9H, p), -22.59 (1H), -34.30 (1H), -36.32 (1H), -39.96 (1H), -56.33 (1H)

a Benzene-d6 unless otherwise noted. b Assignments for 7 were made on the basis of HMBC and NOESY; assignments for the remaining compounds were made analogously, by comparison to literature species or via COSY (12). c Signals broad; coupling not resolved. d Actually appears as a non-firstorder dd in A2B spin system; shifts and JPP determined from simulation. e THF-d8. f Paramagnetic spectra were assigned only on the basis of integrated intensity. g Assignments are for structure with NdC(a). h CD2Cl2.

the mer-configuration was retained. Like all azaallyl derivatives,1,27 the UV-vis spectrum of 8 was highlighted by intense intraligand (IL) bands (417 nm, ε ∼21 100 M-1 cm-1; 613 nm, ε ∼7900 M-1 cm-1). Structures of Fe(II) Iminoaryl and Azaallyl Complexes. 1. [mer-{K-C,N,C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-pyNCH3-3-yl)}(PMe3)3Fe]OTf (7). Data collection and refinement details can be found in Table 3, and Table 4 contains metric parameters that can be viewed in comparison to related complexes, such as 8. As the molecular view of the cation given in Figure 3 illustrates, the pseudo-octahedral geometry of 7 and the mer-disposition of the PMe3 groups are substantiated, and the methylene unit of the chelate backbone links the methylated pyridine and the imino nitrogen. The average iron-phosphorus distance of 2.243(3) A˚, the iron-aza distance of 1.9347(11) A˚, and the iron-aryl and -pyridinium-3-yl

distances of 2.0287(12) and 1.9940(13) A˚ are unexceptional, except that the latter is slightly shorter due to the inductively withdrawing effect of the pyridinium cation. Note that the differences in aza-methylene (1.4567(18) A˚) and aza-methine (1.2885(15) A˚) clearly differentiate CH2 from dCH-. The core angles reveal the constraints of the aryliminopyridyl ligand, as the C(py)-Fe-N and C(Ar)-Fe-N angles are 80.24(5)° and 81.82(5)°, respectively. Note also that both the aryl and pyridyl rings are asymmetrically bound, with the “outer” Fe-C1-C2 and Fe-C13-C12 angles (∼134°) being roughly 20° larger than the “inner” Fe-C1-C6 and Fe-C13-C9 angles (∼114°). The corresponding “in-plane” C(Ar)-Fe-P2 and C(py)-Fe-P2 angles are 103.14(4)° and 94.88(4)°, respectively, reflecting a modest asymmetry that is probably attributable to a more constrained ring that contains the imine residue.

Article


369

Table 2. 13C{1H} NMR Spectral Assignments (δ, JPC(Hz), assnmt)a,b,c for Imine and Imine-Derived Complexes

1 153.23 (a), 161.93 (b), 124.21 (t, 4, c), 126.93 (d), 120.61 (t, 1, e), 152.99 (f), 178.48 (t, 37, h), 140.31 (t, 4, i), 120.83 (t, 3, j), 121.49 (k), 117.94 (t, 3, l), 152.17 (m), 70.26 (n), 10.77 (t, 10, o), -4.61 (t, 33, p) 2 157.15 (a), 161.89 (b), 125.39 (t, 3, c), 126.98 (d), 120.52 (t, 3, e), 152.98 (f), 181.67 (t, 37, h), 145.58 (t, 1, i), 106.42 (j), 121.51 (k), 117.76 (t, 3, l), 152.12 (m), 69.67 (n), 10.84 (t, 11, o), -4.34 (t, 33, p), 54.83 (q) 3 149.43 (a), 161.70 (b), 125.54 (t, 4, c), 127.53 (d), 120.40 (t, 3, e), 152.61 (f), 185.33 (h), 138.52 (i), 120.77 (j), 121.65 (k), 125.22 (l), 152.60 (m), 70.25 (n), 10.68 (o), -4.38 (t, 33, p) 4 61.08 (a), 165.02 (b), 122.05 (c), 128.78 (d), 116.36 (e), 151.59 (f), 190.17 (h), 141.76 (i), 146.88 (j), 116.60 (k), 123.46 (l), 152.04 (m), 163.51 (n), 11.35 (o), -10.00 (p), 35.41 (CMe3, q), 32.37 (C(CH3)3, q) 6 66.23 (a), 172.55 (b), 141.97 (d), 120.57 (e), 150.09 (f), 178.00 (g), 201.77 (h), 142.78 (i), 148.34 (j), 117.12 (k), 126.22 (l), 151.51 (m), 170.09 (n), 17.59 (o), 23.84 (p), 35.12 (CMe3, q), 32.42 (C(CH3)3, q) 7d 62.62 (a), 166.37 (b), 155.57 (d), 122.11 (e), 135.28 (f), 182.50 (g), 196.53 (h), 143.00 (i), 149.91 (j), 118.40 (k), 127.05 (l), 152.07 (m), 173.66 (n), 17.42 (t, 11, o), 23.26 (d, 16, p), 35.48 (CMe3, q), 32.27 (C(CH3)3, q), 46.60 (r) 8 108.95 (a), 160.77 (b), 145.20 (d), 121.20 (e), 135.80 (t, 3, f), 190.15 (td, 13, 19, g), 195.10 (td, 10, 25 (h), 142.26 (i), 148.24 (d, 3, j), 117.56 (k), 127.31 (l), 154.22 (d, 5, m), 106.78 (n), 17.44 (td, 2, 10, o), 23.73 (d, 15, p), 34.88 (CMe3, q), 32.59 (C(CH3)3, q), 41.39 (r) 11e168.70 (t, 4, a), 151.86 (b), 125.87 (c), 116.13 (t, 4, d), 140.17 (e), 143.99 (f), 206.41 (m, g), 183.21 (m, h), 118.57 (i), 120.98 (t, 3, j), 141.44 (k), 124.71 (t, 3, l), 157.37 (m), 66.48 (n), 23.76 (d, 15, o), 17.63 (t, 10, p) 12f 66.48 (a), 157.37 (b), 124.71 (t, 3, c), 141.44 (d), 120.98 (t, 3, e), 118.57 (t, 2, f), 183.21 (m, g), 206.41 (m, h), 143.99 (i), 140.17 (j), 116.13 (t, 4, k), 125.87 (l), 151,86 (m), 168.70 (t, 4, n), 23.34 (td, 3, 16, o), 17.30 (td, 3, 11, p) a Benzene-d6 unless otherwise noted. b Assignments for 7 were made on the basis of HMBC and NOESY; assignments for the remaining compounds were made analogously or by comparison to literature species. c JPC are given when the resolution and signal-to-noise permitted an unambiguous assessment. d THF-d8. e Assignments are for structure with NdC(a). f Signal for CF3 (q) not located.

Scheme 2

2. Zwitterion mer-{K-C,N,C0 -(p-tBu-Ar-2-yl)CHNCH(2py-NCH3-3-yl)}(PMe3)3Fe (8). Select data collection and refinement parameters for 8 are also listed in Table 2, and

comparative bond distances and angles for the zwitterion are given in Table 3. Figure 4 illustrates a molecular view of 8 with a related orientation as 7 in Figure 3. Since the

370


Volpe et al.

Table 3. Selected Crystallographic and Refinement Data for [mer-{K-C,N,C0 -(p-tBu-Ar-2-yl)CHdN-CH2(2-py-NCH3-3-yl)}(PMe3)3Fe]OTf (7), mer-{κ-C,N,C0 -(p-tBu-Ar-2-yl)CHNCH(2-py-NCH3-3-yl)}(PMe3)3Fe (8), and [mer-κ-C,N,C0 {(Ph-2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3]OTf (14) 7

8

14

formula fw space group Z a, A˚ b, A˚ c, A˚ R, deg β, deg γ, deg V, A˚3 Fcalc, g cm-3 μ, mm-1 temp, K λ, A˚ R indices [I > 2σ(I)]a,b

C28H48N2O3F3P3SFe C27H47N2P3Fe C24H38NO3F3P3SFe 698.50 548.43 626.37 P21/c P21/c P1 2 4 8 9.1821(3) 16.9889(5) 24.1590(6) 13.7066(4) 9.3446(2) 12.6587(3) 14.5326(5) 19.4216(6) 18.8974(4) 97.4860(10) 90 90 105.8730(10) 105.4310(10) 93.2550(10) 102.5040(10) 90 90 1682.20(9) 2972.12(14) 5769.9(2) 1.379 1.226 1.442 0.701 0.686 0.807 173(2) 173(2) 173(2) 0.71073 0.71073 0.71073 R1 = 0.0419 R1 = 0.0404 R1 = 0.0418 wR2 = 0.0920 wR2 = 0.0916 wR2 = 0.0960 a,b R1 = 0.0623 R1 = 0.0628 R1 = 0.0730 R indices (all data) wR2 = 0.1012 wR2 = 0.1041 wR2 = 0.1043 1.059 1.038 1.037 GOFc P P P P P a R1 = ||Fo| - |Fc||/ |Fo|. b wR2 = [ w(|Fo| - |Fc|)2/ wFo2]1/2. c GOF (all data) = [ w(|Fo| - |Fc|)2/(n - p)]1/2, n = number of independent reflections, p = number of parameters.

Table 4. Selected Distances (A˚) and Angles (deg) for [mer-{K-C,N,C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-py-NCH3-3-yl)}(PMe3)3Fe]OTf (7), mer-{κ-C,N,C0 -(p-tBu-Ar-2-yl)CHNCH(2-py-NCH3-3-yl)} (PMe3)3Fe (8), and [mer-κ-C,N,C0 -{(Ph-2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3]OTf (14)

Fe-P1 Fe-P2 Fe-P3 Fe-N1 Fe-C1 Fe-C13(C14) N1-C7 N1-C8 C6-C7 C8-C9 P-C(av) N1-Fe-C1 N1-Fe-C13(14) N1-Fe-P1 N1-Fe-P2 N1-Fe-P3 C1-Fe-C13(14) C1-Fe-P1 C1-Fe-P2 C1-Fe-P3 C13(14)-Fe-P1 C13(14)-Fe-P2 C13(14)-Fe-P3 P1-Fe-P2 P1-Fe-P3 P2-Fe-P3 Fe-N1-C7 Fe-N1-C8 Fe-C1-C2 Fe-C1-C6 Fe-C13(14)-C9 Fe-C13(14)-C12 Fe-P-C(av) a

14a

7

8

2.2407(4) 2.2470(4) 2.2424(4) 1.9347(11) 2.0287(12) 1.9940(13) 1.2885(15) 1.4567(18) 1.4296(19) 1.4958(19) 1.833(7)

2.2252(4) 2.2122(4) 2.2295(4) 1.9622(12) 2.0220(14) 2.0163(15) 1.3028(19) 1.380(2) 1.432(2) 1.375(2) 1.838(6)

80.24(5) 81.82(5) 86.52(4) 175.30(4) 89.95(4) 161.95(6) 87.35(4) 103.14(4) 84.72(4) 93.52(4) 94.88(4) 93.33(4) 90.354(16) 171.773(14) 93.603(16) 119.55(10) 120.10(8) 134.83(9) 113.77(11) 113.23(9) 134.15(12) 118.5(19)

80.66(5) 81.76(6) 88.73(3) 174.66(4) 85.49(3) 162.04(6) 85.24(4) 103.74(4) 87.38(4) 90.76(4) 94.02(4) 94.86(4) 94.615(16) 171.288(17) 91.636(16) 118.03(11) 116.93(10) 134.30(10) 111.72(11) 110.08(10) 135.70(12) 118.9(22)

2.2982(6) 2.2959(6) 2.3133(6) 1.9372(16) 2.033(2) 2.036(2) 1.292(3) 1.464(3) 1.439(3) 1.494(3) 1.825(8) 80.30(8) 81.82(8) 89.26(5) 175.66(5) 86.48(5) 161.70(8) 82.88(6) 103.70(6) 85.07(6) 93.06(6) 94.30(6) 97.70(6) 92.90(2) 167.73(2) 92.14(2) 119.55(14) 119.51(14) 134.26(16) 111.36(14) 112.93(15) 132.55(17) 117.5(21)

2.3023(6) 2.2728(6) 2.3179(6) 1.9406(15) 2.0521(18) 2.0444(18) 1.294(2) 1.452(2) 1.436(3) 1.489(3)

80.25(7) 81.64(7) 90.12(5) 174.16(5) 85.75(5) 160.04(8) 82.35(5) 105.09(6) 84.31(5) 89.37(5) 93.41(6) 102.67(5) 92.93(2) 166.53(2) 92.37(2) 119.30(13) 119.11(12) 133.80(15) 110.86(13) 111.93(13) 132.71(15)

First values correspond to Figure 5; second values correspond to equivalent molecular distances and angles in second independent molecule.

neutralization reaction that afforded 8 is purely ligandbased, the core distances and angles in the zwitterionic azaallyl were expected to be very similar to that of the

cationic precursor 7, and that is indeed the case. The Fe-P (2.222(9) A˚ av), Fe-N (1.9622(12) A˚), Fe-C(Ar) (2.0220(14) A˚), and Fe-C(py) (2.0163(15) A˚) distances have

Article


371

Figure 3. Molecular view of the cation of [mer-{κ-C,N,C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-py-NCH3-3-yl)}(PMe3)3Fe]OTf (7); the triflate anion is not shown.

Figure 4. Molecular view of mer-{κ-C,N,C0 -(p-tBu-Ar-2-yl)CHNCH(2-py-NCH3-3-yl)}(PMe3)3Fe (8).

hardly changed. With the exception of a slightly greater asymmetry of the phosphine ligands, and subtle angular changes probably derived from the slightly increased (∼0.03 A˚) Fe-N distance, all the core angles are very similar. The only distance changes of interest are the N1-C7 (1.3028(19) A˚) and N1-C8 (1.380(2) A˚), which now both show double-bond character in accordance with the azaallyl formulation. As expected, the latter distance is longer due to the distribution of charge into the pyridinium ring, which is reflected in the shortening of C8-C9 (1.375(2) A˚) relative to C7-C6 (1.432(2) A˚). Iminoaryl and Azaallyl Fe(III) Derivatives. In order to test the concept of stabilizing higher oxidation states with the aryl-containing chelates, trans-{κ-C,N,N0 -(p-tBu-Ar-2-yl)CHdNCH2-2-py}(PMe3)2FeCH3 (4) was chosen as a candidate for exploration given its solubility and previous reactivity. As Scheme 3 illustrates, treatment of 4 with AgOTf in THF at -78 °C afforded the red-orange iminoaryl cationic

complex [trans-{κ-C,N,N0 -(p-tBu-Ar-2-yl)CHdNCH2-2-py}(PMe3)2FeCH3]OTf (9) in 64% isolated yield upon crystallization from THF/Et2O. Fortunately, a relatively clear 1H NMR spectrum of paramagnetic 9 could be obtained (Table 1), and a μeff of 1.51 μB was measured by Gouy balance; both are consistent with a low-spin d5 center, with S = 1/2. Once the Fe(III) center was established, the means to azaallyl formation were available via simple deprotonation. Exposure of [trans-{κ-C,N,N0 -(p-tBu-Ar-2-yl)CHdNCH2-2py}(PMe3)2FeCH3]OTf (9) to excess KH in THF provided dihydrogen and the neutral Fe(III) complex trans-{κ-C,N, N0 -(p-tBu-Ar-2- yl)CHNCH-2-py}(PMe3)2FeCH3 (10) in 75% isolated yield as blue-green microcrystals. Further evidence of the azaallyl unit was provided by its UV-vis spectrum, which manifested intense IL bands at 417 nm (ε ∼29 200 M-1 cm-1) and 646 (ε ∼8300 M-1 cm-1) in addition to other CT bands. Evans’ method measurements

372


Volpe et al. Scheme 3

Scheme 4

Scheme 5

gave a μeff of 1.90 μB, which is consistent with a low-spin d5, S = 1/2 iron center. Diarylimino Fe(II/III) Derivatives. The successful preparation of Fe(III) iminoaryl and azaallyl species prompted a change to diarylimino Fe(II) ligands, in the hope that greater arylation might prove beneficial toward further oxidation toward Fe(III) and even perhaps Fe(IV). As Scheme 4 reveals, treatment of cis-(Me3P)4FeMe2 with diarylimines Im7, Im8, and Im9 afforded double arene CH activation products {mer-κ-C,N,C0 -(Ar-2-yl)CH2NdCH(Ar-2-yl-X)}Fe(PMe3)3

(X = H, 11; 4-CF3, 12; 3-OMe, 13) with concomitant loss of 2 equiv of methane and 1 equiv of PMe3. The substituents caused subtle changes in color as the rust-red of 11 changed to a dark pink for the p-CF3 species (12), and the m-OCH3 derivative (13) was red. The parent compound, 11, was chosen for oxidation studies, and treatment with AgOTf in THF afforded the Fe(III) cationic complex [mer-κ-C,N,C0 {(Ar-2-yl)CH2NdCH(Ar-2-yl-X)}Fe(PMe3)3]OTf (14) in 52% yield as purple crystals. The μeff of 14 was 1.91 μB according to Gouy balance measurements, again indicative

Article

of a low-spin d5, S = 1/2 center. Paramagnetic shifts in the 1 H NMR spectrum of 14 were also consistent with this formulation. Attempts to generate azaallyl ligands in conjunction with the diaryl platform were considered along two lines: deprotonation of the Fe(III) cation and H-atom abstraction of the neutral diarylimino derivatives. Efforts to deprotonate [mer-κ-C,N,C0 -{(Ph-2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3]OTf (14) using a plethora of common bases failed to elicit the desired azaallyl product, as Scheme 5 indicates. Instead, reduction of 14 occurred to regenerate the neutral Fe(II) complex {κ-C,N, C0 -(Ph-2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3 (11) as the major observable in all cases. It is possible that {κ-C,N,C0 -(Ph-2-yl)CHNCH(Ph-2-yl)}Fe(PMe3)3 (15) was generated, but immediately abstracted an H atom from THF solvent, remaining reagent, or byproduct. In support of this mechanism, evidence for the conjugate acids of the bases employed in the deprotonation attempts was obtained in each case (1H NMR). Efforts to oxidize the Fe(II) species 11 with H-atom abstraction reagents that were compatible with the system failed to achieve azaallyl formation as well. From studies in the smif system, it was thought that the thermodynamics of atom transfer appeared reasonable, but the lack of hydroquinone, diaminobenzene, and Ph3CH in the reaction mixtures at 23 °C renders this assumption questionable. Some quantities of H-atom transfer products (i.e., 1,4-MesNHC6H4, Ph3CH) were obtained when the reaction mixtures were heated, but thermolyses of the reagents in the absence of 11 also generated the same products. Structure of [mer-K-C,N,C0 -{(Ph-2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3]OTf (14). Table 2 provides selected data collection and refinement information pertaining to [mer-κ-C,N,C0 -{(Ph2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3]OTf (14), Table 3 lists pertinent bond distances and angles, and Figure 5 reveals one of the two crystallographically independent pseudo-octahedral complex cations of 14. The distances and angles bear a remarkable similarity to those of [mer-{κ-C,N,C0 -(p-tBu-Ar2-yl)CHdNCH2(2-py-NCH3-3-yl)}(PMe3)3Fe]OTf (7) and mer-{κ-C,N,C0 -(p-tBu-Ar-2-yl)CHNCH(2-py-NCH3-3-yl)}(PMe3)3Fe (8), as Table 3 indicates. Minor deviations included 0.05-0.09 A˚ greater Fe-P distances for 14 versus 7 and 8 and the slightly more irregular P-Fe-C angles in the core for 14. Since 14 is Fe(III), it is plausible that the modest contraction of the 3d orbitals relative to the Fe(II) cases elongates the nonchelate Fe-P distances to a greater extent as their overlap with the phosphorus P orbitals declines. The remaining Fe-C distances are also slightly longer, but barely outside the limits of 3 standard deviations. Perhaps the flat, charged chelate is not as susceptible to effects of 3d contraction, as ionic effects help compensate, and steric factors are minimal. The C1-Fe-P2 angle of ∼104° is again greater than the C14-Fe-P2 angle (∼94°), in accord with the constraints of the imine-containing FeN1C7C6C1 ring. Note that the imine is again clearly differentiated from the N-CH2-Ar linkage, as N1-C7 is 1.294(2) A˚ compared to the 1.452(2) A˚ of N1-C8. A survey of the literature revealed that crystallographic characterizations of Fe(III) aryl derivatives are uncommon and are (52) Goedken, V. L.; Peng, S.-M.; Park, Y. J. Am. Chem. Soc. 1974, 96, 284–285. (53) Doppelt, P. Inorg. Chem. 1984, 23, 4009–4011. (54) Balch, A. L.; Olmstead, M. M.; Safari, N.; St. Claire, T. N. Inorg. Chem. 1994, 33, 2815–2822. (55) Kadish, K. M.; Tabard, A.; Van Caemelbecke, E.; Aukauloo, A. M.; Richard, P.; Guilard, R. Inorg. Chem. 1998, 37, 6168–6175.


373

Figure 5. Molecular view of the cation of [mer-κ-C,N,C0 -{(Ph2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3]OTf (14); the triflate counterion is not shown.

limited to (por)FeIIIAr type species52-56 and the tetraarylanion [Fe(C6Cl5)4][Li(THF)4].57 Many other Fe(III) alkyls and aryls are unusual,58-60 such as Nast’s [Fe(CCH)6]3-58 and the proposed [(1-norbornyl)4Fe]- anion en route to (1-norbornyl)4Fe,5,6 while others are commonly invoked in catalysis29,30,59 or transiently generated via oxidation.60 UV-Vis Spectra of Azaallyls. Figure 6 provides the UV-vis spectra of two azaallyls, the zwitterionic Fe(II) diaryl complex mer-{κ-C,N,C0 -(p-tBu-Ar-2-yl)CHNCH(2-py-NCH3-3-yl)}(PMe3)3Fe (8), and the Fe(III) pyridine-aryl derivative trans-{κ-C,N,N0 -(p-tBu-Ar-2-yl)CHNCH-2-py}(PMe3)2FeCH3 (10). The intensity of their intraligand (IL) transitions contrasts greatly with the precursor imine cations [mer-{κ-C,N,C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-py-NCH3-3-yl)}(PMe3)3Fe]OTf (7) and [trans-{κ-C,N,N0 -(p-tBu-Ar-2-yl)CHd NCH2-2-py}(PMe3)2FeCH3]OTf (9), respectively, which possess absorption spectra that are relatively featureless.

Discussion Aryl Chelates and Strong Fields. The premise of using aryl ligation in the form of a chelate to help impart strong fields capable of sustaining catalysis involving 2 e- changes has yet to be tested, but there are some indications that the concept is plausible. The Fe(III) aryliminopyridine species [trans-{κ-C, N,N0 -(p-tBu-Ar-2-yl)CHdNCH2-2-py}(PMe3)2FeCH3]OTf (9) and the related diarylimine Fe(III) complex [mer-κ-C,N, C0 -{(Ph-2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3]OTf (14) are both stable compounds, as is the neutral azaallyl Fe(III) (56) Bill, E.; Sch€ unemann, V.; Trautwein, A. X.; Weiss, R.; Fisher, J.; Tabard, D.; Guilard, R. Inorg. Chim. Acta 2002, 339, 420–426. (57) Alonso, P. J.; Arauzo, A. B.; Fornies, J.; Garcı´ a-Monforte, M. A.; Martı´ n, A.; Martı´ nez, J. I.; Menj on, B.; Rillo, C.; Saiz-Garitaonandia, J. J. Angew. Chem., Int. Ed. 2006, 45, 6707–6711. (58) Nast, R.; Urban, F. Z. Anorg. Allg. Chem. 1956, 287, 17–23. (59) O’Reilly, R. K.; Gibson, V. C.; White, A. J. P.; Williams, D. J. Polyhedron 2004, 23, 2921–2928. (60) Lau, W.; Huffman, J. C.; Kochi, J. K. Organometallics 1982, 1, 155–169.

374


Volpe et al.

Figure 6. UV-vis spectra of the two azaallyl complexes, Fe(II) 8 (dark green) and Fe(III) 10 (black), relative to precursor imines, Fe(II) 7 (brown) and Fe(III) 9 (orange), that highlight the dramatic intensities and vibrational progressions associated with the intraligand bands that transfer charge from the CNCnb backbone to the pyridine π* orbitals.

Figure 7

derivative derived from 9, trans-{κ-C,N,N0 -(p-tBu-Ar-2-yl)CHNCH-2-py}(PMe3)2FeCH3 (10). Clearly, aryl chelates are capable of sustaining covalent Fe(III) centers, although in each complex strong-field PMe3 ligands are also present. While recent success in Fe catalysis has been achieved via “redox-active” ligand frameworks that permit the metal to maintain a lower formal oxidation state while undergoing changes typical of oxidation,61,62 many cross-coupling schemes favor Fe(III) intermediates.29-31 Furthermore, the stability of these Fe(III) aryls suggests that Fe(II/IV) catalytic schemes remain a plausible future goal. Azaallyl as a Strong Field Complement to Aryl. A significant component of this work was the exploration of the azaallyl ligand as a complement to aryl ligation via chelation, especially due to the similarity between it and NHC-carbenes.1,63,64 While it was hoped that the functionality would be stable with respect to changes in oxidation state, the (61) Tondreau, A. M.; Darmon, J. M.; Wile, B. M.; Floyd, S. K.; Lobkovsky, E.; Chirik, P. J. Organometallics 2009, 28, 3928–3940. (62) Sylvester, K. T.; Chirik, P. J. J. Am. Chem. Soc. 2009, 131, 8772– 8774. (63) Jacobsen, H.; Correa, A.; Poater, A.; Costabile, C.; Cavallo, L. Coord. Chem. Rev. 2009, 253, 687–703. (64) de Fremont, P.; Marion, N.; Nolan, S. P. Coord. Chem. Rev. 2009, 253, 862–892.

results have been mixed. When pyridine is part of the chelate, the azaallyl functionality supports the Fe(III) oxidation state, but when supporting two aryl components in the tridentate chelate, there is reasonable evidence that the azaallyl is unstable, as the inability to convert [mer-κ-C,N, C0 -{(Ph-2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3]OTf (14) or {κ-C,N,C0 -(Ph-2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3 (11) into {κ-C,N,C0 -(Ph-2-yl)CHNCH(Ph-2-yl)}Fe(PMe3)3 (15) implies. Figure 7 illustrates the problem in an oxidation context by using a simple molecular orbital construct. The azaallyl backbone contains an MO that is CNCnb as its highest filled orbital. If this orbital resides above the “t2g” set in an Fe(II) precursor, an oxidation is likely to produce an organic radical in the ligand framework that may lead to degradation. For example, (smif)2Fe (smif, see Figure 1)27 cannot be oxidized to a stable Fe(III) derivative, and calculations indicate that the CNCnb MOs in this system are the HOMO and HOMO-1 orbitals. If, on the other hand, the CNCnb orbital resides below the “t2g” set in energy, then the oxidation is metal-localized, and the Fe(III) derivative is likely to be stable. A related logic would explain the inability to convert the Fe(III) imine (e.g., 14) into a plausible azaallyl (e.g., 15).

Article


Assuming Figure 7 is applicable to the compounds investigated herein, the py groups must interact to stabilize the CNCnb orbital to enough of an extent that it resides energetically lower than dxz, dyz, and dxy, or at least two of the three. Note that the symmetry of dxy and that of CNCnb is the same, and while the overlap between the two orbitals would appear to be marginal, their relative energies may be crucial to the overall stability of an azaallyl in this system. UV-Vis Spectral Features. Inspection of the azaallyl spectra in Figure 6 reveals the presence of vibrational progressions in both IL transitions,65,66 which are more obvious in the lower energy band as a consequence of plotting wavelength. In pyridine-aryl Fe(III) 10, the presumed GS(v=0)fES(v=0) transition is lower in energy (646 nm) than that of diaryl 8 (613 nm), and the progression is roughly 1100 cm-1. The IR spectrum of 10 manifests several absorptions in the region from 1105 to 1170 cm-1, which is a region expected for bends of the CNC azaallyl unit; the excited state is likely to have related features of the appropriate symmetry. A rough assessment of the inflections to the high energy side on the λmax = 417 nm band of 10 suggests that either the same vibration or one slightly higher in energy (∼1200 cm-1) may be the primary contributor, and again there are related IR absorptions in the 11971298 cm-1 region of the GS. Diaryl 8 shows clear λmax at 613 and 570 nm that reveal a vibronic component of about 1200 cm-1 and shoulders on the λmax = 417 nm absorption at ∼431 and ∼398 nm that indicate a coupled vibration of ∼1100 cm-1; its IR spectrum reveals ample possibilities in the window from 1048 to 1292 cm-1. Given the low symmetry of both systems, it is likely either symmetric or antisymmetric in-plane bends could be coupled to the transfer of charge from the CNCnb orbital to the pyridine or aryl rings. An alternative interpretation of the lower energy band(s) of 10 is that the low-energy feature at 646 nm may actually be an MLCT band (i.e., the GS(v=0)fES(v=0) of the IL band would then be at 606 nm). It is likely that at least one of the “t2g” d orbitals, most probably the dxy given its symmetry (Figure 7), is above the CNCnb orbital in energy because the d5 configuration is stable. As such, a lower energy MLCT transition is clearly plausible. The related transition should be to the higher energy side of the first IL band in 8, and while one is not obvious, it could be of lesser intensity or masked by the IL band.

Conclusions A series of aryl CH-bond-activated iminopyridine67 and related diaryl iron complexes have been prepared in ferrous (primarily) and ferric oxidation states. The concept of generating strong fields imparted by metal-aryl bonds is still in question, especially since the aryl chelates reported herein are accompanied by strong-field phosphines in every case. However, there is enough evidence via the stabilization of Fe(III) to proceed with different but related ligand systems. It appears that the azaallyl functionality has a limited utility (65) (a) Mack, J.; Stillman, M. J.; Kobayashi, N. Coord. Chem. Rev. 2007, 251, 429–453. (b) Mack, J.; Stillman, M. J. Coord. Chem. Rev. 2001, 219-221, 993–1032. (66) (a) G€ udel, H. U.; Zilian, A. Coord. Chem. Rev. 1991, 111, 33–38. (b) Colombo, M. G.; Brunold, T. C.; Riedener, T.; G€udel, H. U.; F€ortsch, M.; B€ urgi, H.-B. Inorg. Chem. 1994, 33, 545–550. (67) For a recent insightful study on Ir, see: Li, L.; Brennessel, W. W.; Jones, W. D. Organometallics 2009, 28, 3492–3500.

375

in the construction of these systems, because the low-lying CNCnb orbital that is critical to its unique optical characteristics is too susceptible to oxidation and subsequent degradation.

Experimental Section General Considerations. All manipulations were performed using either glovebox or high-vacuum-line techniques. Hydrocarbon solvents, containing 1-2 mL of added tetraglyme, and ethereal solvents were distilled under nitrogen from purple sodium benzophenone ketyl and vacuum transferred from the same prior to use. Benzene-d6 and toluene-d8 were dried over sodium, vacuum transferred, and stored under N2. THF-d8 was dried over sodium benzophenone ketyl. Methylene chloride-d2 was dried over CaH2, vacuum transferred, and stored over activated 4 A˚ molecular sieves. Fe(PMe3)4Me2 was prepared according to literature procedures.46 Imines were also prepared via literature procedures or as described.32-37 All other chemicals were commercially available and used as received. All glassware was oven-dried. NMR spectra were obtained using Mercury-300, INOVA400, and Inova-500 spectrometers. Chemical shifts are reported relative to benzene-d6 (1H δ 7.16; 13C{1H} δ 128.39), THF-d8 (1H δ 3.58; 13C{1H} δ 67.57), and CD2Cl2 (1H δ 5.32; 13C{1H} δ 54.00). Infrared spectra were recorded on a Nicolet Avatar 370 DTGX spectrophotometer interfaced to an IBM PC (OMNIC software). UV-vis spectra were obtained on an Ocean Optics USB2000 spectrometer. Solution magnetic measurements were conducted via Evans’ method in benzene-d6.68 Solid-state magnetic measurements were performed using a Johnson Matthey magnetic susceptibility balance calibrated with HgCo(SCN)4.69 Elemental analyses were performed by Robertson Microlit Laboratories, Madison, NJ. Procedures. 1. General Procedure for Synthesis of Imines Im1-Im9. To a suspension of MgSO4 (5-8 equiv) in CH2Cl2 were added 1.5 mmol of aldehyde and 1.5 mmol of amine. After stirring for 2 h, the mixture was filtered and concentrated to yield a clear to pale yellow oil in >98% purity (by 1H NMR). Spectra and syntheses for Im1,32 Im2,33 Im5,33 and Im735 are in their respective literature references. Im3: 1H NMR (C6D6, 300 MHz): 8.53 (s, a), 8.10 (d, 8, c), 7.02 (t, 8, d), 6.62 (dd, 5, 8, e), 8.47 (d, 5, f), 7.18 (d, 8, h), 6.91 (t, 8, i), 6.79 (t, 8, j), 7.29 (d, 8, k), 4.71 (s, n). 13C NMR (C6D6, 300 MHz): 164.65 (a), 155.82 (b), 124.96 (c), 136.35 (d), 127.32 (e), 149.96 (f), 130.46 (h), 129.86 (j), 121.34 (k), 134.16 (l), 137.84 (m), 62.21 (n). *One signal obscured by solvent peak. Im4: 1H NMR (C6D6, 300 MHz): 5.00 (s, a), 7.39 (d, 8, c), 7.12 (d, 6, d), 6.65 (t, 6, e), 8.53 (d, 5, f), 7.78 (d, 8, h,l), 7.28 (d, 8, i,k), 8.10 (s, n), 1.17 (s, r). 13 C NMR (C6D6, 300 MHz): 67.55 (a), 160.97 (b), 122.50 (c), 134.90 (d), 122.13 (e), 149.86 (f), 126.09 (i), 154.30 (j), 126.09 (k), 136.45 (m), 162.86 (n), 35.15 (q), 31.58 (r). *Two signals obscured by solvent peak. Im6: 1H NMR (C6D6, 300 MHz): 5.00 (s, a), 7.34 (d, 8, c), 7.08 (t, 8, d), 6.64 (dd, 6, 8, e), 8.52 (d, 5, f), 7.55 (s, h), 6.86 (dd, 3, 8, j), 7.12 (t, 8, k), 7.26 (d, 8, l), 8.03 (s, n), 3.28 (s, q). 13C NMR (C6D6, 300 MHz): 67.42 (a), 160.77 (b), 122.59 (c), 136.51 (d), 122.22 (e), 149.87 (f), 112.59 (h), 160.59 (i), 118.08 (j), 130.08 (k), 122.29 (l), 138.75 (m), 163.02 (n), 55.14 (q). Im8: 1 H NMR (C6D6, 300 MHz): 4.57 (s, a), 7.28-7.21 (m, c,d,f,g), 7.14 (m, e), 7.32 (d, 8, h,l), 7.52 (d, 8, i,k), 7.80 (s, n). 13C NMR (C6D6, 300 MHz): 65.48 (a), 140.18 (b), 128.61 (c), 129.04 (d), 127.69 (e), 129.04 (f), 128.61 (g), 129.13 (h,l), 125.96 (q, 6, I,k), 139.95 (m), 160.21 (n). *Two signals not observed. Im9: 1H NMR (C6D6, 300 MHz): 4.61 (s, a), 7.32 (d, 8, c,g), 7.20 (t, 7, d,f), 7.11 (t, 8, e), 7.59 (s, h), 6.87 (d, 8, j), 7.08 (t, 7, k), 7.27 (d, 7, l), 8.01 (s, n), 3.27 (s, q). 13C NMR (C6D6, 300 MHz): 65.54 (a), (68) (a) Evans, D. F. J. Chem. Soc. 1959, 2003–2005. (b) Schubert, E. M. J. Chem. Educ. 1992, 69, 62. (69) Carlin, R. L. Magnetochemistry; Springer-Verlag: Berlin, 1986.

376


140.51 (b), 128.69 (c,g), 129.05 (d,f), 127.47 (e), 112.42 (h), 140.51 (i), 118.13 (j), 130.08 (k), 122.30 (l), 138.83 (m), 161.77 (n), 55.12 (q). 2. Synthesis of the Imine Complexes. A 100 mL bomb reactor was charged with Fe(PMe3)4Me2 (100-300 mg, 0.2560.641 mmol) and imine (1 equiv). A 15 mL amount of benzene was transferred at -78 °C, and the solution was allowed to warm to 23 °C and stir for 24 h. Upon removal of solvent, the crude mixture was dissolved in Et2O, filtered, and washed (4 10 mL of Et2O). Crystallization from hexanes at -78 °C afforded product. a. trans-{K-C,N,N0 -(Ph-2-yl)CH2NdCH-2-py}(PMe3)2FeCH3 (1). Dark green microcrystals (85 mg) were obtained in 79% yield. Anal. Calcd for C20H32N2P2Fe: C, 57.43; H, 7.71; N, 6.70. Found: C, 56.45; H, 7.36, N, 6.37. b. trans-{K-C,N,N0 -(pMeO-Ar-2-yl)CH2NdCH-2-py}(PMe3)2 FeCH3 (2). Dark green microcrystals (102 mg from 0.383 mmol) were obtained in 59% yield. Anal. Calcd for C21H34N2OP2Fe: C, 56.26; H, 7.64; N, 6.25. Found: C, 55.98; H, 7.71, N, 5.96. c. trans-{K-C,N,N0 -(o-Cl-Ar-2yl)CH2NdCH-2-py}(PMe3)2FeCH3 (3). Dark green crystals (150 mg from 0.512 mmol) were obtained in 65% yield. Anal. Calcd for C20H31N2P2ClFe: C, 53.06; H, 6.90; N, 6.19. Found: C, 52.84; H, 6.62, N, 6.04. d. trans-{K-C,N,N0 -(pt Bu-Ar-2-yl)CHdNCH2-2-py}(PMe3)2FeCH3 (4). Midnight blue crystals (222 mg, 0.641 mmol) were obtained in 73% yield. Anal. Calcd for C24H40N2P2Fe: C, 60.76; H, 8.50; N, 5.91. Found: C, 60.48; H, 8.23, N, 5.68. UV-vis: 266 nm (12 100 M-1 cm-1), 417 (4900), 506 (4600), 613 (8900). e. {mer-K-C,N,C0 -(Ph-2-yl)CH2-

NdCH(Ph-2-yl)}Fe(PMe3)3 (11). The brick-red microcrystalline solid (342 mg from 1.02 mmol) was obtained in 70% yield. Anal. Calcd for C23H38NP3Fe: C, 57.87; H, 8.02; N, 2.93. Found: C, 57.82; H, 8.09, N, 2.81. UV-vis: 239 nm (2100 M-1 cm-1), 332 (3900), 375 (1700), 445 (2400), 501 (1000). f. {mer-K-C,N,C0 -(Ph-2yl)CH2NdCH(Ar-2-yl-4-CF3)}Fe (PMe3)3 (12). The dark pink microcrystalline solid (107 mg from 0.256 mmol) was obtained in 77% yield. Anal. Calcd for C24H37NP3F3Fe: C, 52.86; H, 6.84; N, 2.57. Found: C, 54.30; H, 7.03; N, 2.69. 3. Oxidation to Give the Iron(III) Triflate Complexes. A 25 mL round-bottom flask was charged with either trans-{κ-C, N,N0 -(p-tBu-Ar-2-yl)CHdNCH2-2-py}(PMe3)2FeCH3 (4) or {κC,N,C0 -(Ph-2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3 (11) (typically ∼100 mg) and AgOTf (1 equiv). Then 10 mL of THF was vacuum transferred onto the solids at -78 °C. The solution was allowed to warm to 23 °C and stir for 12 h (9) or 48 h (14). The solvent was stripped and the resulting residue dissolved in 5 mL of THF. After filtration through Celite, the solution was layered with Et2O and cooled to -30 °C to yield a crystalline solid. a. [trans-{K-C,N, N 0 -(p-tBu-Ar-2-yl)CHdNCH2-2-py}(PMe3)2FeCH3]OTf (9). Red-orange crystals (116 mg from 0.291 mmol) were obtained in 64% yield. Anal. Calcd for C25H40N2O3P2F3SFe: C, 48.16; H, 6.47; N, 4.49. Found: C, 47.92; H, 6.27, N, 4.34. UV-vis: 270 nm (18 500 M-1 cm-1), 396 (4300), 419 (5300), 490 (2100), 565 (1000), 607 (1100), 646 (1000). b. [mer-K-C,N,C0 -{(Ph-2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3]OTf (14). Purple crystals (136 mg from 0.418 mmol) were obtained in 52% yield. Anal. Calcd for C24H38NO3P3F3SFe: C, 46.02; H, 6.11; N, 2.24. Found: C, 45.94; H, 6.00; N, 2.11. 4. Synthesis of Iron Azaallyl Complexes. Iron complex of [mer-{κ-C,N,C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-py-NCH3-3-yl)}(PMe3)3Fe]OTf (7) or [trans-{κ-C,N,N0 -(p-tBu-Ar-2-yl)CHd NCH2-2-py} (PMe3)2FeCH3]OTf (9) (typically 0.160 mmol) and excess KH (26 mg, 0.642 mmol) were weighed into a 25 mL round-bottom flask. THF (10 mL) was vacuum transferred at -78 °C. The emerald green solution was placed under argon and allowed to warm slowly to room temperature overnight. The solvent was removed and the mixture filtered and washed with Et2O. Subsequent recrystallization from cold hexanes afforded clean product. a. mer-{K-C,N,C0 -(p-tBu-Ar-2-yl)CHNCH(2-pyNCH3-3-yl)}(PMe3)3Fe (8). Dark green microcrystals (62 mg

Volpe et al. from 0.215 mmol) were obtained in 61% yield. Anal. Calcd for C27H47N2P3Fe: C, 59.13; H, 8.64; N, 5.11. Found: C, 58.88; H, 8.38, N, 4.89. UV-vis: 358 nm (13 400 M-1 cm-1), 417 (21 100), 570 (6000), 613 (7900). b. trans-{K-C,N,N0 -(p-tBu-Ar-2-yl)CHNCH-2-py}(PMe3)2FeCH3 (10). Dark blue-green crystals (60 mg) were obtained in 75% yield. Anal. Calcd for C24H39N2P2Fe: C, 60.89; H, 8.30; N, 5.92. Found: C, 58.85; H, 8.27, N, 5.53. UV-vis: 338 nm (5800 M-1 cm-1), 377 (10 000), 417 (29 200), 485 (4500), 566 (5600), 606 (8600), 646 (8300). 5. mer-{K-C,N,C0 -(p- tBu-Ar-2-yl)CHdNCH 2(2-py-3-yl)}(PMe 3)3Fe (6). To a 200 mL bomb charged with trans-{κ-C, N,N 0 -(p-MeO-Ar-2-yl)CH 2NdCH-2-py}(PMe3)2FeCH 3 (2, 600 mg, 1.26 mmol) was vacuum transferred 30 mL of benzene. Then 1.5 equiv of PMe3 (0.20 mL, 1.90 mmol) was transferred via gas bulb. The reaction mixture was heated at 80 °C for 18 h. After cooling, the solvent was stripped and the remainder dissolved in Et 2O. The solution was filtered through a frit, washed three times with Et 2O, and concentrated. The crude solid was recrystallized from cold pentane/ PMe3 (5 mL/0.5 mL) to yield 453 mg of golden brown solid (67%). Anal. Calcd for C26H 45N 2P3Fe: C, 58.43; H, 8.49; N, 5.24. Found: C, 58.17; H, 8.21; N, 5.13. UV-vis: 333 nm (5300 M-1 cm-1), 369 (3400), 417 (3800), 430 (3500), 496 (1100), 604 (500), 671 (300), 738 (100). 6. [mer-{K-C,N,C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-py-NCH3-3yl)}(PMe3)3Fe]OTf (7). Into a solution of 3 (200 mg, 0.374 mmol) in benzene (20 mL) was syringed methyl triflate (42 μL, 0.374 mmol) under argon. The solution was stirred for 4 h, then allowed to sit for 12 h. The resulting mixture was filtered to yield brown crystals, which were subsequently washed with hexanes (200 mg, 77%). Anal. Calcd for C28H48N2O3P3F3SFe: C, 48.14; H, 6.93; N, 4.01. Found: C, 48.01; H, 7.08; N, 3.88. UV-vis: 336 nm (6900 M-1 cm-1), 375 (5300), 437 (4000), 495 (2300), 569 (1000). 7. NMR Tube Reactions. A 20 mg amount of Fe(PMe3)4Me2 (0.051 mmol) was placed into a flame-dried NMR tube, which was sealed to a 14/20 joint and attached to a needle valve. A solution of imine (0.051 mmol) in benzene (0.7 mL) was added to the tube, at which point a color change was observed. The tube was degassed via freeze-pump-thaw and sealed under active vacuum. Loss of starting material was typically complete after 24 h. a. trans, cis-(PMe3)2(CH3)2Fe{K-N,N0 -EtCH2NdCH-2-py} (5-Et). Purple solution. b. trans,cis-(PMe3)2(CH3)2Fe{K-N,N0 -neoPeCH2NdCH2-py} (5-neoPe). Purple solution. c. {mer-K-C,N,C0 -(Ph-2-yl)CH2NdCH(Ar-2-yl-3-OMe)}Fe(PMe3)3 (13). Red solution. Single-Crystal X-ray Diffraction Studies. Upon isolation, the crystals were covered in polyisobutenes and placed under a 173 K N2 stream on the goniometer head of a Siemens P4 SMART CCD area detector (graphite-monochromated Mo KR radiation, λ = 0.71073 A˚). The structures were solved by direct methods (SHELXS). All non-hydrogen atoms were refined anisotropically unless stated, and hydrogen atoms were treated as idealized contributions (riding model). 8. [mer-{K-C,N,C0 -(p-tBu-Ar-2-yl)CHdNCH2(2-py-NCH33-yl)}(PMe3)3Fe]OTf (7). A red block (0.40 0.30 0.25 mm) was obtained from benzene. A total of 41 235 reflections were collected with 10 134 determined to be symmetry independent (Rint = 0.0346), and 7579 were greater than 2σ(I). A semiempirical absorption correction from equivalents was applied, and the refinement utilized w-1 = σ2(Fo2) þ (0.0468p)2 þ 0.4618p, where p = (Fo2 þ 2Fc2)/3. 9. mer-{K-C,N,C0 -(p-tBu-Ar-2-yl)CHNCH(2-py-NCH3-3-yl)}(PMe3)3Fe (8). A dark green block (0.40 0.30 0.15 mm) was obtained from hexanes. A total of 23 971 reflections were collected with 8674 determined to be symmetry independent (Rint = 0.0413), and 6343 were greater than 2σ(I). A semiempirical absorption correction from equivalents was applied, and the refinement utilized w-1 = σ2(Fo2) þ (0.0458p)2 þ 0.0000p, where p = (Fo2 þ 2Fc2)/3. 10. [mer-K-C,N,C0 -{(Ph-2-yl)CH2NdCH(Ph-2-yl)}Fe(PMe3)3]OTf (14). A dark purple block (0.50 0.25 0.10 mm) was obtained from THF/Et2O. A total of 48 646 reflections were

Article collected with 14 326 determined to be symmetry independent (Rint = 0.0568), and 9881 were greater than 2σ(I). A semiempirical absorption correction from equivalents was applied, and the refinement utilized w-1 = σ2(Fo2) þ (0.0426p)2 þ 0.3446p, where p = (Fo2 þ 2Fc2)/3.

Acknowledgment. We thank the National Science Foundation (CHE-0718030, P.T.W.) and Cornell University for


377

financial support. We thank Dr. Ivan Keresztes and Prof. David B. Zax for aid in interpreting NMR spectra, and Dr. Jay Winkler for helpful discussions. Supporting Information Available: Where the EAs were significantly off in C (1, 10, and 12), pertinent NMR spectra are provided; a CIF file pertaining to 7, 8, and 14. This material is available free of charge via the Internet at http://pubs.acs.org.

Aryl-Containing Pyridine-Imine and Azaallyl Chelates of Iron toward

Recommend Documents