Iron-catalyzed cross-coupling reactions involve the formation of a new carbon-carbon single bond via the coupling of a nucleophilic organometallic reagent with an electrophilic organohalide. Salts of iron(II) and iron(III) are highly active catalysts for the coupling of a variety of electrophiles; however, the scope of nucleophiles is generally limited relative to palladium- and nickel-catalyzed reactions.
When an electrophilic organohalide (or pseudohalide) and a nucleophilic organometallic reagent are mixed in the presence of a transition metal catalyst, coupling of the nucleophilic and electrophilic carbons may occur to furnish a product containing a new carbon-carbon bond. Although palladium is by far the most popular transition metal employed in cross-coupling reactions, its rising cost and limited effectiveness for some classes of substrates has led to the investigation of other transition metal catalysts for cross-coupling. Iron is an abundant, inexpensive, and highly active alternative to the Group 10 metals that exhibit complementary selectivity to these catalysts in cross-coupling reactions (Eq. 1). The most common iron sources employed are FeCl3, Fe(acac)3, and FeCl2.(1)
A cloudy mechanistic picture is currently associated with most iron-catalyzed cross-coupling reactions. The spin state of iron depends strongly on the nature of the ligands surrounding it and iron is associated with several accessible oxidation states. As a result, several different mechanisms may be operating simultaneously in these reactions. In addition, reactions that appear to involve similar conditions may in fact occur through very different mechanisms. Due in part to a lack of mechanistic understanding of the reaction, ligands for iron-catalyzed cross-couplings are very limited and no enantioselective variants of the reaction have been reported to date. In general, the scope of this reaction is limited relative to palladium- and nickel-catalyzed methods. Nonetheless, cross-couplings catalyzed by iron have been applied in a number of syntheses, particularly for the establishment of bonds between sp2- and sp3-hybridized carbons.
Mechanism and Stereochemistry
Mechanistic Studies and Proposals
Early studies by Kochi of reactions of alkyl and alkenyl bromides with alkylmagnesium reagents suggested a catalytic cycle analogous to that of palladium- and nickel-catalyzed reactions involving iron(I) and iron(III) species. After turnover-limiting oxidative addition of the organohalide to an iron(I) species, transmetalation of the Grignard reagent occurs followed by reductive elimination.
This straightforward picture was complicated by later experiments. For example, studies of the reaction of methyllithium with alkenyl bromides in the presence of catalytic amount of FeCl3 suggest that an iron(II) species is the low-valent form of iron in the catalytic cycle, and that an Fe(II)/Fe(IV) process is involved. These studies did not rule out a mechanism involving iron(III) and single-electron transfer processes, the involvement of which is supported by other experiments.
Studies of reactions of Grignard reagents containing β-hydrogens with alkyl halides support the presence of Fe(MgX)2 in the catalytic cycle, the "inorganic Grignard reagent." The active catalyst is generated via reduction of the iron salt by the Grignard reagent, and the catalytic cycle itself involves the typical sequence of oxidative addition, transmetalation, and reductive elimination steps (although the catalyst alternates between the 0 and -2 oxidation states).
A very detailed study involving isolation of iron-containing species in cross-coupling reactions as well as the investigation of a catalyst analogous to Fe(MgX)2 has provided validation for the overlapping cycles in Eq. 2 below.(2)
Reactions employing radical probes and labeled substrates suggest the involvement of radicals in iron-catalyzed cross-coupling reactions. For example, when 6-bromohexene and phenylmagnesium bromide are coupled in the presence of FeCl3 and a phosphine or phosphite ligand, the cyclized product predominates (Eq. 3). Radicals may be generated from Fe(II) complexes that "cross over" into the Fe(I)/Fe(III) catalytic cycle via homolytic cleavage of an iron-carbon bond (see Eq. 2 above).(3)
Alkenyl halides and alkenylmagnesium reagents typically react with retention of configuration. Reactions of chiral, enantiomerically enriched alkyl halides and alkylmagnesiums often yield racemic products, which provides evidence for the involvement of radical intermediates.
Scope and Limitations
The scope of iron-catalyzed cross-coupling reactions is constrained by the limited number of ligands found to promote cross-coupling with iron. Tetramethylethylenediamine (TMEDA) is one of the most commonly used ligands; in many other cases, the iron salt is used directly without addition of a ligand. Iron catalysts are much more reactive in general than palladium or nickel catalysts, but palladium and nickel sometimes exhibit greater selectivity and functional-group tolerance than iron. Iron catalysts are very effective for sp2-sp2 and sp2-sp3 coupling reactions, but are not as effective as other metals at highly attractive sp3-sp3 couplings. In addition, no enantioselective cross-couplings catalyzed by iron have been reported to date, and reports of Heck-type reactivity for iron catalysts are limited.
Despite the limitations enumerated above, iron is useful for a wide variety of cross-couplings. Although the remainder of this section will focus on reactions of organomagnesiums, organozinc, organocopper, and organomanganese reagents have all been used with success in iron-catalyzed cross-coupling reactions.
N-Methylpyrrolidone is commonly used as a solvent or additive in coupling reactions of alkenyl halides with alkylmagnesium reagents to encourage stereospecificity and increase yield (Eq. 4).(4)
Acyl halides may also be employed as electrophiles alongside alkylmagnesium reagents (Eq. 5).(5)
In reactions with aryl halides or pseudohalides, secondary and tertary alkyl Grigard reagents typically do not perform as well as primary alkyl Grignard reagents. Here, the selectivity of iron catalysts complements that of palladium and nickel catalysts: chloride, triflate, and tosylate electrophiles afford the highest yields of product, while iodides and bromides react more sluggishly (Eq. 6).(6)
The reaction of allylic electrophiles with Grignard reagents often occurs in the absence of a catalyst, but the yield and selectivity of the reaction may be unacceptable. Allylic substitutions catalyzed by iron(-2) are rapid and tolerant of silyl groups, ketones, and esters in the substrate (Eq. 7). When allylic phosphates are employed as electrophiles, selectivity is high for direct substitution at the carbon bearing the phosphate group, a result complementary to the selectivity of copper-catalyzed allylic substitutions.(7)
Active methylene compounds may be allylated in the presence of an iron(0) catalyst. As for reactions of allylic phosphates with Grignard reagents, selectivity for direct substitution is high, even when the electrophilic carbon is highly substituted (Eq. 8).(8)
Alkyl halides react with arylmagnesium reagents in the presence of FeCl3 to afford products of sp2-sp3 coupling. TMEDA is a common additive in this reactions, for which slow addition of a slight excess of the Grignard reagent is essential (Eq. 9). Secondary halides generally perform better in these reactions that primary halides.(9)
Iron catalysts may also be used to promote oxidative couplings of arylmagnesium reagents and reductive couplings of organohalides. For example, arylmagnesium reagents undergo homocoupling in the presence of an iron catalyst and 1,2-dichloroethane as a stoichiometric oxidant. Oxygen may also be employed as an oxidant in conjunction with FeCl3 (Eq. 10).(10)
Wurtz- and Ullman-type reductive couplings of alkyl and aryl halides, respectively, are enabled by iron catalysts with magnesium as a terminal reductant. Cross-coupling of alkyl halides with aryl halides occurs in the presence of catalytic amounts of FeCl3, TMEDA, and magnesium metal.(11)
Iron has significant practical advantages over palladium and nickel for cross-couplings between sp2-hybridized electrophiles and alkyl nucleophiles, and this reaction has been applied in several syntheses. The chemoselectivity of these couplings can be remarkable. For example, in a synthesis of latrunculin B, coupling between an ester-containing enol triflate and an alkylmagnesium reagent containing a triple bond proceeded smoothly with no addition of the Grignard reagent to the ester group (Eq. 12). Cross-coupling of an acyl electrophile with methyllithium was used to prepare a separate fragment of the molecule.(12)
The chemoselectivity of iron-catalyzed cross-coupling was highlighted in a synthesis of the immunosuppressive agent FTY720 (Eq. 13). Coupling between the triflate and alkylmagnesium reagent occurred without addition to the ester group.(13)
Comparison to Other Methods
Iron-catalyzed cross-coupling methods face significant competition from reactions employing other metals, such as palladium and nickel. The primary advantages of iron are its low cost, abundance, and non-toxicity. In addition, the selectivity of iron catalysts can be complementary to that of other metals (see Eq. 6 above) and the selectivity of iron catalysts for unactivated alkyl halides makes them a very attractive choice for sp2-sp3 coupling reactions. Because iron catalysts are highly active, reaction times are often short and reaction conditions are mild. However, high catalyst loadings of iron are sometimes needed to counter rapid catalyst deactivation. Suitable general ligands for controlling the reactivity of iron catalysts have not yet been developed. As a consequence, no enantioselective cross-couplings catalyzed by iron are known at present.
Fewer nucleophiles are amenable to iron catalysis than palladium catalysis: although organozincs and organomagnesiums react well, few examples of iron-catalyzed reactions involving silicon, boron, and tin reagents have been reported to date. In addition, Heck-type reactivity of iron has only been observed in one case and the one reported example of sp3-sp3 coupling catalyzed by iron is low yielding and exhibits limited scope.
Experimental Conditions and Procedure
Iron-catalyzed cross-coupling reactions must be carried out under an inert atmosphere to prevent deactivation of the catalyst and undesirable side reactions, such as homocoupling of the nucleophile. Ethereal solvents are most commonly employed. Among sources of iron, iron(III) tri(acetylacetonate) is affordable, highly active, and soluble in ethereal solvents. Iron(III) chloride is also commonly used as a catalyst precursor, although its hygroscopic nature makes handling difficult. Use of iron(II) chloride as the precatalyst can alleviate issues with reductive side reactions that consume the substrates, but this salt is less soluble in ethereal solvents than FeCl3.
The most common additives used in iron-catalyzed cross-couplings are NMP and TMEDA. NMP improves the yield of couplings between alkenyl halides and alkylmagnesiums, while TMEDA is commonly added to reactions of alkyl halides, a historically difficult class of electrophile. The precise roles of these additives are unclear, and in general ligands are rarely added to iron-catalyzed reactions.
Grignard reagents (organomagnesium halides) are a very popular class of nucleophiles employed in this reaction, although their highly nucleophilic and basic nature precludes the presence of base-sensitive functionality in the nucleophile (however, base-sensitive groups in the electrophile may be tolerated if the catalyst promotes chemoselectivity). Organozinc reagents are milder and exhibit wider functional-group tolerance, and these are also widely employed. Silicon-, boron-, and tin-containing nucleophiles are rarely employed with success.
n-Butylmagnesium chloride (1.2 M solution in THF, 22.9 mL, 27.5 mmol) was added dropwise over 10 min, between –5 and 0 °C, to a solution of (E)-8-acetoxy-1-chlorooct-1-ene (5.11 g, 25.0 mmol) and Fe(acac)3 (0.088 g, 0.25 mmol) in a mixture of THF (30 mL) and NMP (25 mL). Stirring was continued for 15 min and then the reaction mixture was hydrolyzed at –10 °C with aqueous HCl solution (1 M, 80 mL). After decanting, the aqueous layer was extracted with Et2O and the combined organic phases were washed with saturated, aqueous NaHCO3 solution, water, and dried with MgSO4. The solvents were removed in vacuo and the product was isolated by distillation (4.53 g, 80%): bp 120–123 °C/5 mm Hg; IR (neat) 1742, 968, 724 cm–1; 1H NMR (CDCl3) δ 0.95 (t, J = 5.6 Hz, 3H), 1.25–1.50 (m, 8H), 1.66–1.68 (m, 2H), 2.03–2.11 (m, 6H), 2.11 (s, 3H), 4.10–4.15 (m, 2H), 5.43–5.45 (m, 2H); 13C NMR (CDCl3) δ 13.5, 20.4, 21.8, 25.4, 28.3, 28.4, 29.1, 31.5, 31.9, 32.1, 64.1, 129.7, 130.1, 137.1.
- ↑ Nakamura, E.; Hatakeyama, T.; Ito, S.; Ishizuka, K.; Ilies, L.; Nakamura, M. Org. React. 2014, 83, 1.
- ↑ Metal-Catalyzed Cross-Coupling Reactions, 3rd ed.; de Meijere, A., Diederich, F., Eds.; Wiley-VCH: New York, 2013.
- ↑ Smith, R. S.; Kochi, J. K. J. Org. Chem. 1976, 41, 502.
- ↑ Tamura, M.; Kochi, J. K. J. Organomet. Chem. 1971, 31, 289.
- ↑ Kauffmann, T. Angew. Chem., Int. Ed. Engl. 1996, 35, 386.
- ↑ a b Fürstner, A.; Martin, R.; Krause, H.; Seidel, G.; Goddard, R.; Lehmann, C. W. J. Am. Chem. Soc. 2008, 130, 8773.
- ↑ Fürstner, A.; Leitner, A.; Méndez, M.; Krause, H. J. Am. Chem. Soc. 2002, 124, 13856.
- ↑ Hill, D. H.; Parvez, M. A.; Sen, A. J. Am. Chem. Soc. 1994, 116, 2889.
- ↑ Bedford, R. B.; Betham, M.; Bruce, D. W.; Danopoulos, A. A.; Frost, R. M.; Hird, M. J. Org. Chem. 2006, 71, 1104.
- ↑ Tamura, M.; Kochi, J. K. J. Am. Chem. Soc. 1971, 93, 1487.
- ↑ Tamura, M.; Kochi, J. K. Synthesis 1971, 93, 303.
- ↑ Cahiez, G.; Habiak, V.; Duplais, C.; Moyeux, A. Org. Lett. 2007, 9, 3253.
- ↑ Nakamura, M.; Matsuo, K.; Ito, S.; Nakamura, E. J. Am. Chem. Soc. 2004, 126, 3686.
- ↑ Hölzer, B.; Hoffmann, R. W. Chem. Commun. 2003, 732.
- ↑ Hatakeyama, T.; Nakagawa, N.; Nakamura, M. Org. Lett. 2009, 11, 4496.
- ↑ Knochel, P.; Dunet, G. Synlett 2006, 407.
- ↑ Cahiez, G.; Marquais, S. Pure Appl. Chem. 1996, 68, 53.
- ↑ a b Cahiez, G.; Avedissian, H. Synthesis 1998, 1199.
- ↑ Alvarez, E.; Cuvigny, T.; Hervé du Penhoat, C.; Julia, M. Tetrahedron 1988, 44, 111.
- ↑ Fürstner, A.; Leitner, A. Angew. Chem., Int. Ed. 2002, 41, 609.
- ↑ Martin, R.; Fürstner, A. Angew. Chem., Int. Ed. 2004, 43, 3955.
- ↑ Yanagisawa, A.; Nomura, N.; Yamamoto, H. Synlett 1991, 513.
- ↑ Xu, Y.; Zhou, B. J. Org. Chem. 1987, 52, 974.
- ↑ Nakamura, M.; Matsuo, K.; Ito, S.; Nakamura, E. J. Am. Chem. Soc. 2004, 126, 3686.
- ↑ Nagano, T.; Hayashi, T. Org. Lett. 2005, 7, 491.
- ↑ Cahiez, G.; Moyeux, A.; Buendia, J.; Duplais, C. J. Am. Chem. Soc. 2007, 129, 13788.
- ↑ Xu, X.; Cheng, D.; Pei, W. J. Org. Chem. 2006, 71, 6637.
- ↑ Hassan, J.; Sévignon, M.; Gozzi, C.; Schulz, E.; Lemaire, M. Chem. Rev. 2002, 102, 1359.
- ↑ Czaplik, W. M.; Mayer, M.; von Wangelin, A. J. Angew. Chem., Int. Ed. 2009, 48, 607.
- ↑ Fürstner, A.; De Souza, D.; Parra-Rapado, L.; Jensen, J. T. Angew. Chem., Int. Ed. 2003, 42, 5358.
- ↑ Fürstner, A.; Laurich, D.; Seidel, G. J. Org. Chem. 2004, 69, 3950.
- ↑ Loska, R.; Volla, C. M. R.; Vogel, P. Adv. Synth. Catal. 2008, 350, 2859.
- ↑ Dongol, K. G.; Koh, H.; Sau, M.; Chai, C. L. L. Adv. Synth. Catal. 2007, 349, 1015.