Iron-catalyzed ole ﬁ n hydrogenation at 1 bar H 2 with a FeCl 3 – LiAlH 4 catalyst †

The scope and mechanism of a practical protocol for the iron-catalyzed hydrogenation of alkenes and alkynes at 1 bar H 2 pressure were studied. The catalyst is formed from cheap chemicals (5 mol% FeCl 3 – LiAlH 4 , THF). A homogeneous mechanism operates at early stages of the reaction while active nanoparticles form upon ageing of the catalyst solution

Catalytic hydrogenations of olefins constitute one of the strongholds of transition metal catalysis within organic synthesis and technical processes. 1The majority of methods involve noble metal catalysts based on Pd, Pt, Rh, Ir or toxic metals such as Ni or Co. Iron-catalyzed hydrogenations of olefins have only recently attracted great interest due to their expedient economic and environmental qualities. 2Homogeneous iron catalysts were mostly reported with phosphine and pyridyl-2,6-diimine ligands, sometimes requiring high pressures of H 2 . 3,4Nanoparticle Fe catalysts could be prepared by reduction of iron salts with Grignard reagents in the absence of a suitable ligand or by decomposition of iron carbonyls. 5Fe-catalyzed reductions of olefins were recently reported with cheap ferrous salt pre-catalysts FeX 2 in the presence of an excess of lithium N,N-dimethylaminoborohydride (10 equiv.)or sodium triethylborate (4 equiv.)and required a high catalyst loading or the addition of tetra-dentate ligands. 6eductions of alkenes and alkynes with LiAlH 4 in the presence of various transition metal halides (NiCl 2 , TiCl 2 , CoCl 2 , FeCl 3 ) were already reported in the 1960s and postulated to involve metal hydride species that engage in formal hydrometalations of the olefin. 7Here, we wish to present a synthetic and mechanistic study on a hydrogenation protocol using catalytic amounts of a cheap Fe salt and catalytic amounts of lithium aluminiumhydride (LiAlH 4 ) as catalyst activator under an atmosphere of 1 bar H 2 as stoichiometric hydrogen source (Scheme 1).7e This method allows the use of standard (ambient pressure) equipment.H 2 is an abundant raw material; LiAlH 4 is an easy-to-handle reductant with numerous applications. 8

Reaction conditions and substrate scope
Initial experiments with the model substrate allylbenzene (1) aimed at the identification of a suitable catalytic reductant which assists the formation of a low-valent iron catalyst (with dark brown colour) from the commercial pre-catalyst FeCl 3 (Table 1). 9LiAlH 4 displayed excellent selectivity which exceeded that of earlier protocols with Grignard reagents. 5Isomerization of the terminal double bond into conjugationwhich occurred in the related EtMgCl-mediated protocols (entries 2, 4)was effectively suppressed. 10NaBH 4 was far less active even at elevated temperature and pressure (entries 6, 7).Interestingly, low ratios of LiAlH 4 -FeCl 3 (1/1 to 2/1) fared optimal in the hydrogenation of 1 at 1 bar H 2 .When employing a larger excess of LiAlH 4 (>2/1), the catalytic activity collapsed.7e This stoichiometry differs from literature reports where large excess amounts of hydride reagents effected clean hydrogenations of olefins. 6,7a-c At 60 °C, the FeCl 3 -LiAlH 4 catalyst decomposed upon decolorization.The catalyst system Scheme 1 Iron-catalyzed reductions of olefins: hydride vs. hydrogen methods.† Electronic supplementary information (ESI) available.CCDC 1034372.
For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c4gc02368d a Institute of Organic Chemistry, University of Regensburg, 93040 Regensburg, Germany.E-mail: axel.jacobi@ur.decomprises of cheap and easy-to-handle reagents (FeCl 3 or FeCl 2 , LiAlH 4 , THF); the reaction operates under ambient conditions (1 bar H 2 , 20 °C), which make the general protocol practical for every-day use in standard synthesis laboratories.The optimized conditions were applied to functionalized allylbenzenes and styrenes (Tables 2 and 3). 9llylbenzenes underwent only minimal olefin isomerization. 10Styrenes exhibited low propensity to undergo polymerization (entry 13, Table 3).The general protocol is compatible with several functional groups including F, Cl, Br, allyl and benzyl ethers, esters, carboxamides, pyridines and anilines.

Mechanistic studies
The distinction between homogeneous and heterogeneous catalysts is a challenging task. 17However, kinetic experiments with selective poisons can provide valuable information on the topicity of the catalyst species.We have performed two sets of poisoning experiments which appear to support geneous mechanism.Dibenzo[a,e]cycloocta-tetraene (dct) is a selective ligand for homogeneous metal species due to its rigid tub-like structure and π-acceptor properties. 18Upon addition of 30 mol% dct (6 equiv.per [Fe]) to the hydrogenation of isopropenylbenzene at 1 bar H 2 after 30 min, the catalyst activity was significantly inhibited (Scheme 2, top). 9,19A similar conclusion can be derived from a poisoning experiment with 3 equiv.Hg (60 equiv.Hg per [Fe]).A potential amalgam formation 20 was not observed and no significant change of the catalyst activity was observed in comparison with the control reaction (Scheme 2, bottom). 9These results suggest the operation of a homogeneous catalyst species during the early stage of the catalytic hydrogenation.Previous studies showed that the reaction of FeCl 3 with an excess of LiAlH 4 ultimately leads to the formation of iron metal and AlH 3 via the intermediate formation of a thermally unstable iron(II) compound with the composition Fe(AlH 4 ) 2 . 21,22In an attempt to gain deeper insight into the catalyst species operating in homogeneous solution, we treated [FeCl 2 (tmeda)] 2 (tmeda = N,N,N′,N′-tetramethylethylenediamine) with LiAlH 4 at −70 °C and obtained dark red crystals of the oligohydride compound [Li(thf ) 2 -{Fe(tmeda)} 2 (AlH 5 )(Al 2 H 9 )] (4, Scheme 3). 9 The hexa-metallic macrocyclic cage contains 14 bridging hydrido ligands and two Fe atoms with distorted octahedral coordination geometries.Unfortunately, the thermal instability prevented further spectroscopic characterization.However, complex 4 showed no activity in hydrogenations of styrenes (1-10 bar H 2 , −10 °C) and maintained its red colour throughout the reaction.Above −10 °C, the complex rapidly decomposed upon H 2 evolution to give a brown paramagnetic species which afforded good yields in hydrogenations at 20 °C and 4 bar H 2 .The crystallographic characterization of 4 documents that this or similar oligonuclear Fe(II) aluminohydride complexes may be intermediates en route to the formation of catalytically active low-valent iron species. 23he initially homogeneous dark-brown catalyst species ( possibly in the oxidation states 0 and/or +1) 23 experience rapid ageing and particle formation after appr. 1 h under the reductive conditions.Several methods of synthesis and characterization techniques of naked Fe(0) nanoparticles ( prepared by reduction of ferric and ferrous halides) have been reported. 5,7,23,24DLS measurements (dynamic light scattering) of freshly prepared catalyst solutions (5 mol% FeCl 3 -LiAlH 4 , THF, r.t., 10 min, then 100 nm nanofiltration) documented the presence of poly-disperse particles of 250-1500 nm size after 30 min of ageing under anaerobic conditions in the absence of substrates.The aged species are much less catalytically active than their homogeneous counterparts.Catalyst solutions (FeCl 3 -LiAlH 4 (1/1) in THF) stored at 0 °C under argon for 6 h, 24 h, and 48 h afforded 42%, 12%, and 5% conversion of α-methylstyrene under standard conditions (see entry 16 in Table 2), respectively.
We postulate a homogeneous mechanism of soluble, lowvalent iron catalyst in the initial stage of the hydrogenation reactions (Scheme 4).Such species form by reduction of FeCl 3 (or L n FeCl 2 ) with LiAlH 4 at above 0 °C and are typically characterized by the dark brown colour.The absence of suitable ligands leads to the formation of Fe(0) nanoclusters 5,22,24 which require higher H 2 pressures than the homogeneous species to maintain catalytic activity.Deuterium incorporation was observed at higher catalyst concentrations (30 mol% FeCl 3 -LiAlD 4 ) in the absence of H 2 which gave ∼55% hydrogenation product (Scheme 5, center). 9uch H 2 -free conditions can effect H/D scrambling in the starting material and product (via reversible hydroferration) and the formation of radical intermediates (with participation of THF as H donor). 9 However, the radical mechanism is very unlikely to operate under hydrogenation conditions in the presence of H 2 gas (Scheme 5): 9 reaction work-up with deuterium oxide (D 2 O) and employment of lithium aluminiumdeuteride (LiAlD 4 ) showed no deuterium incorporation into the products, respectively (Scheme 5, top right).Further, the intermediacy of free C-radicals is unlikely: employment of the Scheme 3 Synthesis of the soluble LiAlFe-oligohydride complex 4. Scheme 4 Proposed formation and catalysis of low-valent iron species.
Scheme 5 Mechanistic studies with deuterated reagents (top), in the absence of H 2 (center), and with radical probe (bottom).
General procedure A 10 mL vial was charged with a freshly prepared solution of FeCl 3 in THF (0.50 mL, 0.05 M) and an aliquot of a vigorously stirred suspension of LiAlH 4 in THF (0.50 mL, 0.1 M) under an argon atmosphere.After stirring the dark brown mixture for 10 min (during which H 2 evolution can be observed), the alkene (0.50 mmol) was added and the vial purged with dihydrogen gas (1 min).For reactions under higher H 2 pressures, the vial was transferred to a high pressure reactor (Parr™), the reactor purged with H 2 (1 min), sealed, and the internal pressure adjusted to 1 bar.After 3-20 h at room temperature, the reaction was quenched with saturated aqueous NaHCO 3 (1 mL) and extracted with ethyl acetate (2 × 2 mL).The organic phases were dried (Na 2 SO 4 ) and subjected to flash chromatography (SiO 2 , pentane/ethyl acetate) or analyzed by quantitative GC-FID analysis vs. pentadecane as internal reference.

Conclusions
In summary, we have studied the iron-catalyzed hydrogenation of various styrenes, alkenes, and alkynes under an atmosphere of 1 bar H 2 .This method uses cheap and easy-to-handle reagents (FeCl 3 , LiAlH 4 , THF, H 2 ) which allow facile implementation in standard synthesis labs.Alkynes underwent Z-selective semi-hydrogenation.Sterically hindered and functionalized olefins showed higher conversions at elevated H 2 pressures.Mechanistic studies support the notion of a homogeneous catalyst species at the outset of the hydrogenation reactions (<1 h) while catalyst ageing results in the formation of particles which exhibited somewhat lower catalytic activity.The crystallographically characterized homogeneous Fe(II) oligohydride complex 4 can serve as starting point for further model catalyst preparations.

Table 1
Selected optimization experiments a

Table 2
Hydrogenation of allylbenzenes at 1 bar H 2