Carbon-oxygen bond cleavage by bis(imino)pyridine iron compounds: Catalyst deactivation pathways and observation of acyl C-O bond cleavage in esters

Article abstract of DOI:10.1021/om8005596

Investigations into the substrate scope of bis(imino)pyridine iron-catalyzed hydrogenation and [2π + 2π] diene cyclization reactions identified C-O bond cleavage as a principal deactivation pathway. Addition of diallyl or allyl ethyl ether to the bis(imino)pyridine iron dinitrogen complex, (^iPrPDI)Fe(N₂)₂ (^iPrPDI = 2,6-(2,6-ⁱPr₂-C₆H₃N=CMe) ₂C₅H₃N, 1-(N₂)₂), under a dinitrogen atmosphere resulted in facile cleavage of the C-O bond and yielded a mixture of the corresponding paramagnetic iron allyl and alkoxide complexes. For ethyl vinyl ether, clean and selective formation of the iron ethoxide was observed with concomitant loss of the vinyl fragment. In situ monitoring of the catalytic hydrogenation of trans-methyl cinnamate established ester C-O bond cleavage as a competing process. Stoichiometric reactions between 1-(N ₂)₂ and allyl and vinyl acetate also produced facile C-O oxidative addition. For the latter, a six coordinate diamagnetic bis(imino)pyridine acetatoxy iron vinyl compound was obtained and characterized by X-ray diffraction. Phenyl acetate undergoes exclusive acyl C-O bond cleavage, while alkyl-substituted esters such as ethyl, pentyl, benzyl, isopropyl, cyclohexyl, and tert-butyl acetate undergo competing ester and acyl C-O bond cleavage accompanied by iron-promoted decarbonylation. Deuterium labeling studies established that reversible C-H activation and chelate cyclometalation occur prior to, but are not a prerequisite for, carbon-oxygen bond oxidative addition of ethyl acetate. The molecular and electronic structures of the ether and ester C-O bond cleavage products have been established and demonstrate that ligand- rather than metal-based oxidation accompanies substrate activation.

Full text of DOI:10.1021/om8005596

6264
Organometallics 2008, 27, 6264–6278
Carbon-Oxygen Bond Cleavage by Bis(imino)pyridine Iron
Compounds: Catalyst Deactivation Pathways and Observation of
Acyl C-O Bond Cleavage in Esters
Ryan J. Trovitch, Emil Lobkovsky, Marco W. Bouwkamp, and Paul J. Chirik*
Department of Chemistry and Chemical Biology, Baker Laboratory, Cornell UniVersity,
Ithaca, New York 14853
ReceiVed June 16, 2008
Investigations into the substrate scope of bis(imino)pyridine iron-catalyzed hydrogenation and [2π +
2π] diene cyclization reactions identified C-O bond cleavage as a principal deactivation pathway. Addition
of diallyl or allyl ethyl ether to the bis(imino)pyridine iron dinitrogen complex, (iPrPDI)Fe(N2)2 (iPrPDI
) 2,6-(2,6-iPr2-C6H3NdCMe)2C5H3N, 1-(N2)2), under a dinitrogen atmosphere resulted in facile cleavage
of the C-O bond and yielded a mixture of the corresponding paramagnetic iron allyl and alkoxide
complexes. For ethyl vinyl ether, clean and selective formation of the iron ethoxide was observed with
concomitant loss of the vinyl fragment. In situ monitoring of the catalytic hydrogenation of trans-methyl
cinnamate established ester C-O bond cleavage as a competing process. Stoichiometric reactions between
1-(N2)2 and allyl and vinyl acetate also produced facile C-O oxidative addition. For the latter, a six
coordinate diamagnetic bis(imino)pyridine acetatoxy iron vinyl compound was obtained and characterized
by X-ray diffraction. Phenyl acetate undergoes exclusive acyl C-O bond cleavage, while alkyl-substituted
esters such as ethyl, pentyl, benzyl, isopropyl, cyclohexyl, and tert-butyl acetate undergo competing
ester and acyl C-O bond cleavage accompanied by iron-promoted decarbonylation. Deuterium labeling
studies established that reversible C-H activation and chelate cyclometalation occur prior to, but are not
a prerequisite for, carbon-oxygen bond oxidative addition of ethyl acetate. The molecular and electronic
structures of the ether and ester C-O bond cleavage products have been established and demonstrate
that ligand- rather than metal-based oxidation accompanies substrate activation.
and irreversible deactivation pathways.8 Introduction of redox-
Introduction
active ligands, those that can actively participate in reversible
Oxidative addition is a fundamental transformation in orga-
electron transfer chemistry with the metal,9,10 render oxidative
transformations more intriguing as formal electron loss may be
either metal- or ligand-based. Rationally designing compounds
that undergo well-understood electron transfer events coupled
to oxidative addition may ultimately prove valuable for dis-
covering base metal catalysts that mimic or surpass the reactivity
and selectivity often achieved with their second and third row
congeners.
nometallic chemistry and often constitutes a key bond activation
step in many stoichiometric reactions and catalytic processes.1
This two electron redox event typically requires accessible Mn
and Mn+2 oxidation states and is most common with electron
rich (e.g. d6 and d8), late transition metal complexes. For first
row ions, oxidative addition of alkyl halides is known to involve
radical processes2 and can result in a two electron oxidation at
a single metal center3 or occur as two one electron oxidations
involving two metals.4,5
Inspired by seminal reports on Fe(CO)5 promoted hydrogena-
tion catalysis,11,12 our laboratory has reported the synthesis of
aryl-substituted bis(imino)pyridine iron dinitrogen compounds,
As interest grows in replacing toxic and expensive precious
metal catalysts with more cost-effective and benign iron
compounds,6,7 so grows the need to understand the elementary
steps, such as oxidative addition, that comprise catalytic turnover
(
iPrPDI)Fe(N2)2 (iPrPDI ) 2,6-(2,6-iPr2-C6H3NdCR)2C5H3N, R
(8) For seminal studies on the oxidative addition of E-X bonds to
phosphine-ligated Fe(0) and related complexes see (a) Klein, H.-F. Angew.
Chem., Int. Ed. 1980, 19, 362. (b) Ittel, S. D.; Tolman, C. A.; English,
A. D.; Jesson, J. P. J. Am. Chem. Soc. 1976, 98, 6073. (c) Tolman, C. A.;
Ittel, S. D.; English, A. D.; Jesson, J. P. J. Am. Chem. Soc. 1978, 100,
4080. (d) Parker, S. F.; Peden, C. H. F.; Barrett, P. H.; Pearson, R. G.
J. Am. Chem. Soc. 1984, 106, 1304.
* Corresponding author. E-mail: pc92@cornell.edu.
(1) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke R. G. In
Principles and Applications of Organotransition Metal Chemistry; Univer-
sity Science Books: Mill Valley, CA, 1987.
(2) Rossi, R. A.; Pierini, A. B.; Penenory, A. B. Chem. ReV. 2003, 103,
71.
(9) Jørgensen, C. K. Coord. Chem. ReV. 1966, 1, 164.
(3) San Filippo, J., Jr.; Silbermann, J.; Fagan, P. J. J. Am. Chem. Soc.
1978, 100, 4834.
(10) (a) Pierpont, C. G Coord. Chem. ReV. 2001, 219-221, 415. (b)
Ward, M. D.; McCleverty, J. A. J. Chem. Soc., Dalton Trans. 2002, 275.
(c) Chaudhuri, P.; Verani, C. N.; Bill, E.; Bothe, E.; Weyhermu¨ller, T.;
Wieghardt, K. J. Am. Chem. Soc. 2001, 123, 2213.
(4) Spessard, G. O.; Miessler, G. L. Organometallic Chemistry; Prentice
Hall: Upper Saddle River, NJ, 1996; p 175.
(5) Blaser, H.-U.; Halpern, J. J. Am. Chem. Soc. 1980, 102, 1684.
(6) Bolm, C.; Legros, J.; Le Paih, J.; Zani, L. Chem. ReV. 2004, 104,
6217.
(7) Enthaler, S.; Junge, K.; Beller, M. Angew. Chem., Int. Ed. 2008,
47, 3317.
(11) (a) Schroeder, M. A.; Wrighton, M. S. J. Am. Chem. Soc. 1976,
98, 551. (b) Schroeder, M. A.; Wrighton, M. S. J. Organomet. Chem. 1977,
128, 345.
(12) Frankel, E. N.; Emken, E. A.; Peters, H. M.; Davison, V. L.;
Butterfield, R. O. J. Org. Chem. 1964, 29, 3292.
10.1021/om8005596 CCC: $40.75
2008 American Chemical Society
Publication on Web 10/30/2008

Carbon-Oxygen Bond CleaVage
Organometallics, Vol. 27, No. 23, 2008 6265
Scheme 1
) Me, 1-(N2)2;13 R ) Ph, 2-(N2)2),14 that function as efficient
catalysts for the hydrogenation and hydrosilylation of olefins
and alkynes at low metal loadings. Subsequently, 1-(N2)2 was
also found to promote the catalytic [2π + 2π] cycloisomeriza-
tion of R, ω-dienes to yield substituted cyclobutanes.15 Spec-
troscopic and computational studies on 1-(N2)216 and on related
neutral ligand derivatives, (iPrPDI)FeLn (L ) CO, PR3, NH2R,
Nt CPh, etc.),17 have unequivocally established the redox-
activity of the bis(imino)pyridine chelate (Figure 1).18 Thus,
1-(N2)2 is best described as an intermediate spin, d6 ferrous
compound antiferromagnetically coupled to a bis(imino)pyridine
diradical dianion.16
During the course of our investigations into the substrate
scope of catalytic olefin hydrogenation and [2π + 2π] cycloi-
somerization reactions with 1-(N2)2,19 dramatic substituent
effects were observed (Scheme 1). Diallyl ether was readily
hydrogenated to dipropyl ether in the presence of 1-(N2)2 and
4 atm of dihydrogen. Performing the reaction under a dinitrogen
atmosphere, in an attempt to induce [2π + 2π] cycloaddition,
produced no turnover. Likewise, trans-methyl cinnamate and
dimethyl itaconate were both readily hydrogenated to the
corresponding acetatoxy-substituted alkane with 1-(N2)2 and 4
atm of H2, while allyl and vinyl acetate exhibited no turnover
under the same conditions (Scheme 1).19
Results
Carbon-Oxygen Bond Cleavage in Ethers. The divergent
reactivity of diallyl ether in catalytic hydrogenation (facile
turnover)19 versus [2π + 2π] cycloisomerization (no turnover)15
with 1-(N2)2 prompted further investigation into the substrate-iron
interaction. Treatment of a pentane solution of 1-(N2)2 with 0.5
equivalents of diallyl ether at 23 °C furnished two iron products,
the bis(imino)pyridine iron allyloxide and the iron allyl,
1-OCH2CHdCH2 and 1-Allyl, respectively (Scheme 2). Red,
paramagnetic 1-Allyl was also independently synthesized by
allylation of 1-Br with one equivalent of CH2dCHCH2MgBr.
While this method yielded 1-Allyl, the product was contami-
nated with small yet inseparable amounts of 1-Br even when
excess Grignard reagent was used.
1
The H NMR spectroscopic features of 1-Allyl are notably
different from S ) 3/2 bis(imino)pyridine iron alkyls, alkoxides
and halides. Typically, these compounds, (iPrPDI)Fe-X (X )
Br, 1-Br; Me, 1-Me; CH2CMe3, 1-Np),21,22 exhibit paramag-
netically shifted resonances over a 600 ppm chemical shift range.
Importantly, the hydrogen atoms located in the chelate plane
appear the most shifted from their diamagnetic reference values.
For example, the imine methyl groups of these complexes appear
upfield between -160 and -220 ppm, while the p-pyridine
hydrogen resonances are the most downfield, between 200 and
400 ppm, depending on the identity of the X-type ligand. The
m-pyridine peak generally appears around 65 ppm and varies
only slightly as a function of the X-type ligand. Additionally,
the isopropyl methine resonances are shifted significantly
upfield, to approximately -110 ppm, likely due to a through
space interaction with the metal center. In contrast, the para-
magnetically broadened resonances of 1-Allyl are dispersed over
a much narrower chemical shift range (∼175 ppm). The imine
methyl and m-pyridine resonances appear at -26.64 ppm and
47.64 ppm, respectively. The narrower chemical shift range is
likely due to an η3-, rather than η1-, allyl interaction in benzene
solution and a different degree of bis(imino)pyridine chelate
reduction. The observed C2V symmetry of the compound is a
result of an in-plane rotation of the η3-allyl or facile η3-, η1-
interconversion on the NMR time scale.
Here, we describe a systematic investigation into the interac-
tion of various ethers, esters, and carboxylated alkenes with
1-(N2)2. In addition to identifying important deactivation
pathways in catalytic olefin hydrogenation and [2π + 2π]
cycloisomerization processes, these studies provide fundamental
insight into oxidative addition chemistry at a reducing iron center
bearing a redox active ligand20 and demonstrate rare examples
of acyl C-O bond cleavage in alkyl-substituted esters.
(13) Bart, S. C.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc. 2004,
126, 13794.
(14) Archer, A. M.; Bouwkamp, M. W.; Cortez, M.-P.; Lobkovsky, E.;
Chirik, P. J. Organometallics 2006, 25, 4269.
(15) Bouwkamp, M. W.; Bowman, A. C.; Lobkovsky, E.; Chirik, P. J.
J. Am. Chem. Soc. 2006, 128, 13340.
(16) Bart, S. C.; Chlopek, K.; Bill, E.; Bouwkamp, M. W.; Lobkovsky,
E.; Neese, F.; Wieghardt, K.; Chirik, P. J. J. Am. Chem. Soc. 2006, 128,
13901.
(17) Bart, S. C.; Lobkovsky, E.; Bill, E.; Wieghardt, K.; Chirik, P. J.
Inorg. Chem. 2007, 46, 7055.
(18) (a) Knijnenburg, Q.; Gambarotta, S.; Budzelaar, P. H. M. Dalton
Trans. 2006, 5442. (b) Zhu, D.; Budzelaar, P. H. M. Organometallics 2008,
27, 2699.
(19) Trovitch, R. J.; Lobkovsky, E.; Bill, E.; Chirik, P. J. Organome-
tallics 2008, 27, 1470.
(20) (a) Bianchini, C.; Giambastiani, G.; Rios, I. G.; Mantovani, G.;
Meli, A.; Segarra, A. M. Coord. Chem. ReV. 2006, 250, 1391. (b) Small,
B. L.; Brookhart, M.; Bennett, A. M. A. J. Am. Chem. Soc. 1998, 120,
4049. (c) Britovsek, G. J. P.; Gibson, V. C.; Kimberley, B. S.; Maddox,
P. J.; McTavish, S. J.; Solan, G. A.; White, A. J. P.; Williams, D. J. Chem.
Commun. 1998, 849.
Figure 1. Bis(dinitrogen) iron complexes bearing redox active
bis(imino)pyridine ligands.

6266 Organometallics, Vol. 27, No. 23, 2008
TroVitch et al.
Scheme 2
The observation of facile diallyl ether C-O bond cleavage
by 1-(N2)2 under mild conditions prompted additional studies
into the scope of the transformation. Addition of 0.5
equivalents of allyl ethyl ether to a pentane solution of 1-(N2)2
at 23 °C yielded 1-Allyl and the bis(imino)pyridine iron
ethoxide compound, 1-OEt (Scheme 2). These two products
were formed exclusively, demonstrating selective and quan-
titative cleavage of the allylic C-O bond. The iron ethoxide
complex, 1-OEt, was independently synthesized by the
addition of one equivalent of anhydrous ethanol to 1-(N2)2.
time ethane23 and unidentified iron compounds were observed.
The H NMR resonances and half-life established from the
independent synthesis of 1-Vy clearly demonstrate that the iron
vinyl compound is not a product of ethyl vinyl ether cleavage.
1
1
Other unidentified, H NMR-silent paramagnetic iron com-
pounds are also formed from this synthetic route and likely
account for the fate of the hydrogen atom. These species were
detected by 2H NMR spectroscopy following the addition of
CH3CH2OD.
Selective C-O bond cleavage was also observed with ethyl
vinyl ether. Stirring a pentane solution of 1-(N2)2 with one
equivalent of CH2dCHOCH2CH3 for 24 h yielded 1-OEt as
1
the sole, H NMR observable iron product (Scheme 2). To
explore whether the putative bis(imino)pyridine iron vinyl
complex is sufficiently stable to observe by 1H NMR spectros-
copy, an independent synthesis was pursued. Our laboratory has
previously reported the oxidative addition of alkyl bromides to
1-(N2)2 as a convenient method to synthesize four-coordinate
bis(imino)pyridine iron alkyls.21 Extending this approach to sp2-
hybridized alkenyl halides may allow synthesis of the desired
vinyl compound.
To probe whether a 1H NMR-silent iron compound ac-
companied the formation of 1-OEt and accounted for the
missing vinyl fragment, 1-(N2)2 was treated with 0.5 equivalents
1
of ethyl vinyl ether. Monitoring this experiment by H NMR
spectroscopy established approximately 50% conversion to
1-OEt with the balance of the iron remaining as 1-(N2)2. While
accurate quantitation of the relative amounts of paramagnetic
versus diamagnetic iron compounds is complicated by the
different relaxation rates of the hydrogens, this experiment
demonstrated that no other iron compounds accompany 1-OEt
formation from ethyl vinyl ether cleavage. Because no other
organic products (ethane, acetylene, etc.) were observed, the
fate of the vinyl fragment remains unknown.
Addition of 0.5 equivalents of vinyl bromide to 1-(N2)2
yielded a mixture of the bis(imino)pyridine iron monobromide,
1-Br,22 and the desired iron vinyl complex, 1-Vy (eq 1).
Unfortunately, 1-Vy was formed in low (∼30%) yield. Attempts
to synthesize the compound by salt metathesis of 1-Br with
CH2dCHMgBr were unsuccessful. Under an inert atmosphere
in benzene-d6 at 23 °C, 1-Vy persists for over 48 h after which
Given the facility with which unsaturated ethers undergo
C-O bond cleavage with 1-(N2)2, similar chemistry was
(21) Trovitch, R. J.; Lobkovsky, E.; Chirik, P. J J. Am. Chem. Soc. 2008,
130, 11631.
(22) Bouwkamp, M. W.; Bart, S. C.; Hawrelak, E. J.; Trovitch, R. J.;
Lobkovsky, E.; Chirik, P. J. Chem. Commun. 2005, 27, 3406.
(23) Thorn, R. P.; Payne, W. A.; Stief, L. J.; Tardy, D. C. J. Phys. Chem.
1996, 100, 13594.

Carbon-Oxygen Bond CleaVage
Organometallics, Vol. 27, No. 23, 2008 6267
Scheme 3
explored with saturated compounds. No evidence for C-O bond
rupture was obtained upon heating 1-(N2)2 with a large excess
of diethyl ether or THF. For the THF experiments, the
diamagnetic coordination complex, 1-(THF)n (n ) 1, 2) was
observed by 1H NMR spectroscopy analogous to the previously
reported compound with the phenylated backbone.14 Addition
of excess anisole to a benzene-d6 solution of 1-(N2)2 produced
no change at 23 °C. Heating this solution in a sealed tube to
110 °C for three hours resulted in clean conversion to a new
diamagnetic, Cs symmetric iron compound with arene reso-
nances centered at 4.14 (para) and 5.46 (meta) ppm. These peaks
are diagnostic for an η6-coordinated aryl group (eq 2).14 The
product has therefore been assigned as 1-Aryl, similar to
the structurally characterized compound previously reported for
the bis(imino)pyridine iron complex with a phenyl-substituted
backbone.14 In a control experiment, a benzene-d6 solution of
1-(N2)2 was heated to 110 °C and yielded 1-Aryl, demonstrating
that the anisole plays little, if any, role in arene coordination.
of 5 mol % were required. Pure hydrocarbon substrates such as
cyclohexene or 1-hexene can be reduced with superior turnover
frequencies using only 0.3 mol % of 1-(N2)2,13 suggesting a
potential catalyst deactivation pathway with the functionalized
substrates.
The possibility of 1-(N2)2 deactivation by C-O bond cleavage
prompted a series of additional experiments whereby the fate
of the iron compound was directly studied by 1H NMR
spectroscopy. Performing the hydrogenation of trans-methyl
cinnamate with 10 mol % 1-(N2)2 established the formation of
two new paramagnetic iron compounds during the course of
catalytic turnover. Both compounds gradually accumulated over
time and were the exclusive iron products after hours of
hydrogenation. A subsequent series of experiments identified
these compounds as the bis(imino)pyridine iron cinnamate and
hydrocinnamate compounds, 1-CIN and 1-H2CIN, respectively,
arising from ester C-O bond cleavage (Scheme 3). Both
compounds were independently prepared by the addition of the
free carboxylic acid to 1-(N2)2. Isolating each compound and
subjecting it to catalytic hydrogenation conditions produced no
turnover, demonstrating that C-O bond cleavage is a competing
catalyst deactivation pathway. The hydrogenated compound,
1-H2CIN derives from C-O bond scission of the reduced
product formed during catalytic turnover, as attempts to
hydrogenate 1-CIN in the presence of 1-(N2)2 produced no
conversion.
Observation of ester C-O bond cleavage under catalytic
hydrogenation conditions prompted an additional series of
stoichiometric experiments. Addition of one-half of an equiva-
lent of trans-methyl cinnamate to a benzene-d6 solution of
1-(N2)2 at 23 °C under a dinitrogen atmosphere resulted in
immediate cleavage of the ester C-O bond to yield an equimolar
mixture of the bis(imino)pyridine iron cinnamate complex,
1-CIN, and the iron methyl compound, 1-Me22 (Scheme 3).
Exposure of this mixture of products and excess trans-methyl
cinnamate to 4 atm of H2 resulted in complete consumption of
Carbon-Oxygen Bond Cleavage in Carboxylated
Alkenes. As reported previously,19 1-(N2)2 is an effective
precatalyst for the hydrogenation of various carboxylated
alkenes. During the course of exploring the substrate scope in
this class of reductions, certain peculiarities were observed.
Substrates such as trans-methyl cinnamate and dimethylitaconate
were readily hydrogenated under 4 atm of H2 in the presence
of 1-(N2)2, while other compounds such as vinyl and allyl acetate
produced no turnover under the same conditions. For the suc-
cessful catalytic hydrogenations, relatively high catalyst loadings

6268 Organometallics, Vol. 27, No. 23, 2008
TroVitch et al.
Scheme 4
1-Me, forming methane, 1-CIN, and 1-H2CIN, and accounts
for the observations made during catalytic turnover.
mixes into the S ) 0 ground-state by spin orbit coupling.16
A
similar C-O bond cleavage reaction was observed following
the addition of methyl benzoate to 1-(N2)2; an equimolar mixture
of the bis(imino)pyridine iron benzoate compound, 1-OBz, and
1-Me, was identified (eq 3). In this case, the putative bis(imi-
no)pyridine iron methyl benzoate complex, 1-MeOBz, was not
observed by NMR spectroscopy.
The observation of H2 pressure dependent C-O bond
cleavage with trans-methyl cinnamate prompted a more detailed
investigation into the reaction chemistry of vinyl and allyl
acetate with 1-(N2)2. Recall that these substrates were not
hydrogenated using 5-10 mol % of 1-(N2)2 and 4 atm of
dihydrogen, suggesting rapid C-O bond cleavage (Scheme 1).19
Addition of one equivalent of vinyl acetate to a pentane solution
of 1-(N2)2 followed by recrystallization at -35 °C furnished
bright purple crystals identified as 1-(OAc)(Vinyl) (Scheme 4).
1-(OAc)(Vinyl) is diamagnetic and was readily characterized
by multinuclear (1H and 13C) and two-dimensional NMR
spectroscopy. The benzene-d6 1H NMR spectrum recorded at
23 °C exhibits the number of peaks consistent with overall Cs
molecular symmetry with the mirror plane equivalencing the
bis(imino)pyridine chelate plane. The ꢀ-hydrogens on the vinyl
ligand appear as two doublets centered at 4.66 and 1.87 ppm
for the cis and trans (with respect to the iron) hydrogens on the
terminal carbon. The methine hydrogen on the carbon directly
attached to the metal appears downfield at 10.07 ppm with a
{1H}13C NMR resonance for this position observed at 189.91
ppm. These assignments have been confirmed by COSY, HSQC,
and HMBC NMR spectroscopy.
Addition of allyl acetate to 1-(N2)2 at 23 °C also resulted in
rapid cleavage of the ester C-O bond; however, two products,
the bis(imino)pyridine iron acetate, 1-OAc, and allyl complex,
1-Allyl, were observed (Scheme 4). Small amounts (5-10%)
of free propene were also observed when the reaction was
conducted in a sealed NMR tube. The alkene likely arises from
decomposition of 1-Allyl, as more propene accumulated over
the course of one week at 23 °C and was observed from the
cleavage of diallyl ether.
C-O Bond Cleavage in Saturated and Phenyl-Substituted
Esters. The possibility of C-O bond cleavage in saturated esters
was also explored. Addition of methyl acetate to 1-(N2)2 at 23
°C resulted in the formation of the iron acetate complex, 1-OAc,
and the iron methyl compound, 1-Me (eq 3). Careful monitoring
of the reaction mixture by 1H NMR spectroscopy revealed that
the bis(imino)pyridine iron ester complex, 1-MeOAc, was
formed immediately after mixing. The imine methyl resonance
is shifted upfield to -4.88 ppm, while the m- and p-pyridine
resonances appear downfield at 11.67 and 8.63 ppm, respec-
tively. These chemical shifts are diagnostic of temperature
independent paramagnetism, whereby an S ) 1 excited-state
Addition of an isomer of methyl benzoate, phenyl acetate, to
1-(N2)2 reversed the selectivity of the C-O bond cleavage
reaction. Addition of one equivalent of phenyl acetate to a
benzene-d6 solution of 1-(N2)2 exclusively yielded the products
of acyl C-O bond cleavage: the iron phenoxide complex,
1-OPh, and 1-Me (eq 4). Careful inspection of the diamagnetic
region of the 1H NMR spectrum also revealed formation of small
quantities 1-(CO)2 accounting for the fate of the carbonyl
group.24
(24) Yamamoto, A.; Kakino, R.; Shimizu, I. HelV. Chim. Acta 2001,
84, 2996.
(25) Neese, F. Inorg. Chim. Acta 2002, 337, 181.
(26) Ferna´ndez, I.; Trovitch, R. J.; Lobkovsky, E.; Chirik, P. J.
Organometallics 2008, 27, 109.
(27) Furimsky, E. Appl. Catal., A 2000, 199, 147.
(28) Lin, Y. S.; Yamamoto, A. In Topics in Organometallic Chemistry;
Murai, S., Ed.; Springer-Verlag: Berlin, Germany, 1999; Vol 3, p 161.
(29) Trost, B, Verhoven, T. R. In ComprehensiVe Organometallic
Chemistry; Wilkinson, G., Stone, F. G. A., Abel, A. E., Eds.; Pergamon:
Oxford, England, 1982; Vol 8, p 799.
(30) Trost, B. M. Acc. Chem. Res. 1980, 13, 385.

Carbon-Oxygen Bond CleaVage
Organometallics, Vol. 27, No. 23, 2008 6269
repeated at 23 °C. The expected products, 1-OAc and 1-OEt,
were observed over the course of one week. No evidence for
the formation of either 1-Me or 1-Et was obtained by 1H NMR
spectroscopy. The inability to observe 1-Et is a result of the
relative fast rate (t1/2 ∼3 h) of decomposition of the iron alkyl21
compared to the slower rate (t1/2 ∼3 days) of the C-O bond
cleavage reaction. Because the bis(imino)pyridine iron acyl
complex, 1-C(O)Me, is a likely product following the oxidative
addition of the acyl C-O bond of ethyl acetate, attempts were
made to evaluate the kinetic stability of such an intermediate.
Addition of one atmosphere of carbon monoxide to a benzene-
d6 solution of 1-Me resulted in immediate formation of 1-(CO)2
along with small quantities of acetone and methane. In a related
experiment, addition of 0.5 equivalents of CH3C(O)Cl to a
benzene-d6 solution of 1-(N2)2 yielded 1-Cl along with 1-Me,
consistent with decarbonylation from the putative bis(imino)py-
ridine iron acyl.
As part of our studies into the substrate scope and selectivity
of C-O bond cleavage, the chemistry of ethyl benzoate was
explored. Addition of one equivalent of this ester to 1-(N2)2 at
23 °C resulted in rapid and exclusive cleavage of the ester C-O
bond over the course of two hours to furnish 1-Et and 1-OBz
(eq 6). No products derived from acyl C-O bond cleavage were
Because acyl bond cleavage, particularly in alkyl esters, is
relatively rare,24 the scope and selectivity of bis(imino)pyridine
iron C-O bond cleavage was examined in more detail. Addition
of one equivalent of ethyl acetate to a benzene-d6 solution of
1-(N2)2 at 23 °C immediately yielded the bis(imino)pyridine
iron ethyl acetate complex, 1-EtOAc. Gently warming a
benzene-d6 solution of 1-EtOAc to 65 °C for 18 h yielded an
equimolar mixture of 1-OAc and 1-OEt (eq 5).
1
detected by H NMR spectroscopy. This result demonstrates
that when ester C-O bond cleavage is sufficiently facile, the
alkyl products can be observed by 1H NMR spectroscopy. Thus,
it is likely that 1-Et is formed from ester C-O bond oxidative
addition of ethyl acetate but has a rate of decomposition faster
than substrate activation.
The formation of the iron acetate and ethoxide compounds
signals two competing C-O bond cleavage reactions. 1-OAc
is the product of ester C-O bond cleavage, while 1-OEt is the
persistent product derived from acyl C-O bond rupture. Ester
C-O bond cleavage would, in principle, also form the bis(imi-
no)pyridine iron ethyl complex, 1-Et. This compound was
previously synthesized by our laboratory and found to decom-
pose over the course of hours at 23 °C.21 Under the more forcing
conditions required for ethyl acetate cleavage (18 h at 65 °C),
1-Et would not persist. For acyl C-O bond cleavage, the
bis(imino)pyridine iron methyl complex, 1-Me, is the other
expected product. This compound was also not observed, but it
too is unstable under the conditions required for ethyl acetate
cleavage. Control experiments on isolated samples of pure 1-Me
revealed decomposition to an unidentified mixture of products
upon heating to 65 °C. Thus, 1-OAc and 1-OEt were the only
products thermally robust enough to persist following ethyl
acetate cleavage.
Having demonstrated rare examples of acyl C-O bond
cleavage with phenyl and ethyl acetate at ambient temperature,
1-(N2)2 was treated with other alkyl acetates to assay selectivity.
The results of these studies are summarized in Scheme 5.
Because the reactions were conducted at 65 °C, the correspond-
ing bis(imino)pyridine iron alkyl complexes were not observed.
Thus, only the kinetically persistent iron acetate and iron
1
alkoxide compounds were detected by H NMR spectroscopy.
The ratios of products are approximate because they were
determined by integration of paramagnetically broadened and
shifted isopropyl methyl resonances.
Observation of acyl C-O bond cleavage in esters suggested
that similar chemistry may be possible with formates. Treatment
of 1-(N2)2 with one equivalent of either ethyl, isopropyl, or
phenyl formate cleanly afforded the desired bis(imino)pyridine
iron alkoxide compounds, 1-OEt, 1-OiPr, and 1-OPh, respec-
tively (eq 7). As with acyl bond cleavage in esters, examination
In an attempt to observe the putative bis(imino)pyridine iron
alkyl products from ethyl acetate cleavage, the reaction was

6270 Organometallics, Vol. 27, No. 23, 2008
TroVitch et al.
Scheme 5
Scheme 6
1
of the diamagnetic region of the H NMR spectrum revealed
the formation of small quantities of 1-(CO)2, consistent with
decarbonylation. To provide additional evidence for CO loss,
the cleavage of isopropyl formate was conducted with an excess
of 1-(N2)2. Under these conditions, the expected amount of
1-MeOAc-d6 cleanly converted to an approximately equimolar
mixture of 1-OAc-d3 and 1-CD3 (Scheme 6). Because the
methyl group of 1-CD3 (or isotopologues) has not been detected
by NMR spectroscopy, chemical degradation experiments were
used to determine the isotopic composition of the iron methyl
group. Hydrolysis of the product mixture and subsequent
analysis of the bis(imino)pyridine ligand by 2H NMR spectros-
copy established no deuterium incorporation into the isopropyl
methyl substituents. Collection of the liberated methane from
1
1-(CO)2 (25%) was detected by H NMR spectroscopy.
1
protonolysis of the iron methyl complex and analysis by H
NMR spectroscopy clearly demonstrated that CD3H was the sole
isotopologue of methane formed. A converse experiment was
also conducted whereby natural abundance methyl acetate was
added to the deuterium labeled bis(imino)pyridine iron dinitro-
gen complex, 1*-(N2)2, (* denotes deuterium labeling of the
isopropyl methyl substituents).13 In this case, 1*-OAc and 1*-
CH3 were identified as the sole products. Again, the isotopic
composition of the iron methyl group was assayed by hydrolysis,
1
and CH4 was the only methane isotopologue observed by H
NMR spectroscopy.
Analogous experiments were conducted with isotopologues
of ethyl acetate; preparation of 1,2-d2-ethyl acetate was ac-
complished by D2 addition to vinyl acetate in the presence of a
catalytic amount of palladium on carbon. Addition of the
isotopically labeled ester to 1-(N2)2 immediately furnished the
diamagnetic bis(imino)pyridine iron ethyl acetate complex,
Deuterium Labeling Experiments. A series of deuterium
labeling experiments were conducted to probe whether C-H(D)
activation was competitive with iron-promoted ester cleavage.
Addition of methyl acetate-d6 to a benzene or benzene-d6
solution of 1-(N2)2 at 23 °C resulted in observation of 1-MeOAc-
2
1-EtOAc-d2. As anticipated, the benzene solution H NMR
spectrum exhibited two peaks centered at 0.74 and 3.02 ppm
for the coordinated ester. Allowing the solution to stand at 23
°C for 20 min revealed isotopic exchange between the isopropyl
methyl group of the bis(imino)pyridine chelate and the meth-
1
2
d6, as judged by H and H NMR spectroscopy. Over time,

Carbon-Oxygen Bond CleaVage
Organometallics, Vol. 27, No. 23, 2008 6271
Scheme 7
Scheme 8
ylene position (Scheme 7). No evidence was obtained for
isotopic exchange involving the methyl group of the ester.
Warming the solution to 70 °C for 14 h yielded the bis(imi-
no)pyridine iron acetate and ethoxide compounds (Scheme 7).
Analysis of the product mixture by 2H NMR spectroscopy
revealed deuterium incorporation in the isopropyl methyl groups
of the bis(imino)pyridine ligands in both 1*-OEt and 1*-OAc
and in the methyl group of the iron ethoxide.
observation of reversible ester coordination is similar to the
behavior observed between 1-(N2)2 and amines or ketones.19
X-ray Structures of Selected C-O Bond Cleavage
Products. Several of the products of ether and ester C-O bond
cleavage were characterized by single crystal X-ray diffraction.
The bond lengths of the bis(imino)pyridine ligand determined
from high quality X-ray diffraction data are diagnostic for redox
activity of the chelate.18a While these parameters are undoubt-
edly useful, it should be noted that a complete and accurate
description of the electronic structure is only obtained from a
combination of magnetochemistry, structural analysis, and
Mo¨ssbauer spectroscopy augmented with open shell DFT
calculations.25 The relevant metrical parameters for the bis(imi-
no)pyridine chelates for all structurally characterized compounds
in this work are reported in Table 1. Features of each individual
structure will be described independently.
The converse experiment was conducted, whereby the
deuterium labeled iron dinitrogen compound, 1*-(N2)2, was
treated with a slight excess of natural abundance ethyl acetate
(Scheme 8). Over the course of four hours at 23 °C, deuterium
exchange was observed with the methylene position of both
the free and coordinated esters. No evidence was obtained for
isotopic exchange into the terminal methyl group. Warming the
solution to 65 °C resulted in C-O bond cleavage and yielded
the expected iron acetate and ethoxide compounds. Analysis of
the paramagnetic iron products by 2H NMR spectroscopy
established deuterium incorporation into the isopropyl methyl
groups of both compounds (Scheme 8). To establish the isotopic
composition of the acetate and ethoxide ligands, each bis(imi-
no)pyridine iron compound was treated with water and the
organic products: free bis(imino)pyridine, ethanol, and acetic
acid were analyzed by 2H NMR spectroscopy. While no
deuterium was detected in the acetic acid, the free ethanol
exhibited a peak consistent with isotopic incorporation exclu-
sively in the methylene position.
A representation of the solid state structure of 1-OCH2-
CHdCH2 is presented in Figure 2. An idealized square planar
iron center is observed with the sum of the angles around the
metal totaling 360°. The metrical parameters of the chelate
(Table 1) are consistent with one electron reduction.16
A
benzene-d6 solution magnetic moment of 4.0 µB was determined
by Evans method at 23 °C and is consistent with an S ) 3/2
iron compound. The magnetic data, in combination with the
metrical parameters, suggest that 1-OCH2CHdCH2 is best
described as a high spin iron(II) compound (SFe ) 2) antifer-
romagnetically coupled to a bis(imino)pyridine radical anion
(SL ) 1/2). Both electronic and molecular structures of this
compound are identical to that previously described for the
isoelectronic bis(imino)pyridine iron chloride, 1-Cl.16
In a related experiment, a benzene solution of 1-EtOAc was
exposed to 4 atm of D2 gas at 23 °C. Over the course of four
hours, isotopic exchange was observed in the isopropyl methyl
group of the bis(imino)pyridine chelate of 1-EtOAc as well as
in the methylene position of the ester (Scheme 8). When the
reaction is carried out with excess ethyl acetate, the deuteration
of the methylene is catalytic, demonstrating rapid C-H activa-
tion and ligand exchange prior to C-O bond cleavage. The
The solid state structures of 1-OAc and 1-OBz were also
determined and are presented in Figure 3. In both cases, the
iron is five coordinate with a κ2-carboxylate, where the
O(1)-C(34)-O(2) and iron chelate planes are essentially
orthogonal. The central carboxylate carbons, C(34), are nearly

6272 Organometallics, Vol. 27, No. 23, 2008
TroVitch et al.
Table 1. Selected Bond Lengths (Å) and Angles (Deg) for Compounds Crystallographically Characterized in This Work
1-OCH2CHdCH2
1-OAc
1-OBz
1-Allyl
1-(OAc)(Vinyl)
Fe(1)-N(1)
Fe(1)-N(2)
Fe(1)-N(3)
Fe(1)-O(1)
Fe(1)-O(2)
Fe(1)-C(34)
2.1242(17)
1.9983(17)
2.1467(18)
1.8486(16)
2.1085(15)
1.9949(17)
2.1615(15)
2.0837(16)
2.1271(14)
2.430(2)
2.1166(13)
1.9842(15)
2.1269(15)
2.116(13)
2.1033(13)
2.437(2)
1.9934(11)
1.8418(11)
2.0017(12)
2.074(3)a
2.151(3)b
2.155(4)
1.9540(9)
1.7976(9)
1.9504(9)
2.0771(9)
2.0135(8)
1.9942(14)
N(1)-C(2)
N(3)-C(8)
N(2)-C(3)
N(2)-C(7)
1.312(3)
1.306(2)
1.370(2)
1.374(3)
1.316(2)
1.302(3)
1.376(2)
1.371(2)
1.305(2)
1.308(2)
1.372(2)
1.375(2)
1.3273(18)
1.3369(18)
1.3800(18)
1.3799(18)
1.3127(14)
1.3103(15)
1.3685(13)
1.3655(14)
C(2)-C(3)
C(7)-C(8)
1.459(3)
1.460(3)
1.444(3)
1.461(3)
1.450(3)
1.455(2)
1.420(2)
1.410(2)
1.4402(16)
1.4444(15)
O(1)-C(34)
O(2)-C(34)
1.351(3)
1.275(3)
1.256(3)
1.272(2)
1.268(2)
N(1)-Fe(1)-N(2)
N(1)-Fe(1)-N(3)
N(2)-Fe(1)-N(3)
75.75(7)
150.80(7)
75.05(7)
75.94(6)
149.75(7)
75.06(6)
75.26(6)
150.27(6)
75.99(6)
79.40(5)
152.10(5)
79.46(5)
81.00(4)
159.67(4)
81.15(4)
a Fe(1)-C(35). b Fe(1)-C(36).
redox activity. Somewhat surprising, however, are the
Cimine-Cpyridine distances of 1.420(2) and 1.410(2) Å, contrac-
tions consistent with two rather than one electron reduction.16
Likewise, the Nimine-Cimine distances are elongated to 1.3273(18)
and 1.3369(18) Å, also consistent with a doubly reduced
bis(imino)pyridine.
Notably, the iron-nitrogen bond distances in 1-Allyl are
statistically shorter than other previously characterized (iPrP-
DI)Fe-X complexes. Typically, this class of compounds exhibits
an overall S ) 3/2 ground-state comprising a high spin (SFe
)
2) iron center antiferromagnetically coupled to a bis(imino)py-
ridine radical anion. The contraction of the iron-nitrogen bonds
in the solid state structure of 1-Allyl suggests an intermediate-
spin ferric center (SFe ) 3/2) that avoids population of the σ*
dx2-y2 orbital, antiferromagnetically coupled to a bis(imino)py-
ridine diradical dianion. More detailed spectroscopic, magnetic,
and computational studies are required to fully elucidate the
electronic structure of this compound.
Figure 2. Molecular structure of 1-OCH2CHdCH2 at 30%
probability ellipsoids. Hydrogen atoms are omitted for clarity.
The solid state structure of 1-(OAc)(Vinyl) was also deter-
mined by X-ray diffraction (Figure 5) and represents a rare
example of a neutral six-coordinate bis(imino)pyridine iron
compound.20a The N(1)-C(2) and N(3)-C(8) bond lengths
(Table 1) of 1.3127(14) and 1.3103(15) Å are comparable to
those observed for compounds with single electron reduction
of the bis(imino)pyridine chelate.16 Similarly, the C(2)-C(3)
and C(7)-C(8) distances of 1.4402(16) and 1.4444(15) Å are
contracted compared to the values of 1.487(3) Å in free iPrPDI
and also support one electron reduction of the bis(imino)pyridine
chelate.16 Two descriptions of the electronic structure accom-
modate the observed diamagnetic ground state. One possibility
is a low spin, d6 ferrous complex with a neutral bis(imino)py-
ridine chelate, while the other is a low spin, d5 ferric compound
antiferromagnetically coupled to a iPrPDI radical anion. While
the metrical data seems to favor the latter description, care must
be taken in over interpreting crystallographic data as ligand field
strength, and other factors may also influence the bond lengths
of the coordinated bis(imino)pyridine.
symmetrically disposed about the metal. The metrical parameters
of the chelate (Table 1) are consistent with one electron
reduction.16 The Fe(1)-O(1) bond lengths of 2.0837(16) and
2.116(13) Å for 1-OAc and 1-OBz are elongated from typical
iron-alkoxide complexes (vide infra). Similar Fe(1)-O(2)
distances of 2.1271(14) and 2.1033(13) Å were observed for
1-OAc and 1-OBz, respectively.
One particularly interesting compound is 1-Allyl. Bis(imi-
no)pyridine iron monoalkyl compounds are relatively rare,21,22,26
and to our knowledge, an allyl derivative has yet to be prepared.
The solid state structure of 1-Allyl is presented in Figure 4.
The allyl ligand is η3-coordinated with the three carbon plane
oriented nearly perpendicular to the iron chelate plane and the
methine carbon, C(35), directed toward an imine nitrogen.
Notably, this carbon is not in the iron chelate plane, providing
a molecule with a more idealized square pyramidal rather than
trigonal bipyramidal geometry. The allyl ligand is positionally
disordered with nearly equal populations of the rotamer where
the methine carbon is directed toward N(1) and the opposite
rotamer where it is directed toward N(3). Although successfully
modeled, the disorder compromises discussion of the metrical
parameters of the allyl ligand. The distortions observed in the
bond distances of the bis(imino)pyridine chelate clearly establish
Discussion
Carbon-oxygen bond oxidative addition reactions are of
interest because of their potential application in hydrodeoxy-
genation (HDO) reactions27 as well as various cross coupling

Carbon-Oxygen Bond CleaVage
Organometallics, Vol. 27, No. 23, 2008 6273
Figure 3. Molecular structure of 1-OAc (top) and 1-OBz (bottom) at 30% probability ellipsoids. View of the core of the molecules (right).
Hydrogen atoms are omitted for clarity.
methodologies used in organic synthesis.28 Oxidative addition
of allylic ethers, carboxylates, carbonates, and halides has
enjoyed special utility in the synthesis of metal allyl complexes,
which can undergo subsequent nucleophilic attack29,30 or cross-
coupling chemistry.31 C-O bond cleavage chemistry is well
known for second and third row transition metals including
palladium,24,32 rhodium,33 iridium,34 ruthenium,35 zirconium,36
molybdenum,37 and tantalum.38,39 For first row metal ions, C-O
oxidative addition chemistry of both ethers40,41 and esters42,43
to Ni(0) has been well-studied and known for some time.
Examples with nickel(II) precursors have been reported more
recently.44
Examples of C-O bond cleavage with well-defined iron
compounds are more limited. In a seminal example, Ganem and
Small reported the application of ferric chloride to the cleavage
of ethereal C-O bonds.45 A more recent example, described
by Plietker and co-workers, applies nitrosylated variants of
Collman’s reagent46 for the transesterification of vinyl acetates,
phenyl carboxylates, and electron poor esters with various
alcohols.47 Iron acyl complexes that do not undergo decarbo-
nylation are proposed as key intermediates in the catalytic cycle.
Direct oxidative addition of carbon-oxygen bonds to electron
rich iron(0) compounds has also been reported by Ittel and co-
workers. Addition of anisole or methyl benzoate to a transient
bis(diphosphine)iron(0) yielded the products of C-O cleavage.48,49
(31) Waetzig, S. R.; Rayabharapu, D. K.; Weaver, J. D.; Tunge, J. A.
Angew. Chem., Int. Ed. 2006, 45, 4977.
(32) Tsuji, J. In Palladium Reagents and Catalysts; Wiley: Chichester,
U. K., 1995; p 290.
(33) van der Boom, M. E.; Liou, S.-Y.; Ben-David, Y.; Shimon, L. J. W.;
Milstein, D. J. Am. Chem. Soc. 1998, 120, 6531.
(34) Lara, P.; Paneque, M.; Poveda, M. L.; Salazar, V.; Santos, L. L.;
Carmona, E. J. Am. Chem. Soc. 2006, 128, 3512.
(41) Eisch, J. J.; Im, K. R. J. Organomet. Chem. 1977, 139, C45.
(42) Yamamoto, T.; Ishizu, J.; Kohara, T.; Komiya, S.; Yamamoto, A.
J. Am. Chem. Soc. 1980, 102, 3758.
(43) Yamamoto, T.; Ishizu, J.; Yamamoto, A. J. Am. Chem. Soc. 1981,
103, 6863.
(44) van der Boom, M. E.; Liou, S.-Y.; Shimon, L. J. W.; Ben-David,
Y.; Milstein, D. Inorg. Chim. Acta 2004, 357, 4015.
(45) Ganem, B.; Small, V. R., Jr. J. Org. Chem. 1974, 39, 3728.
(46) Collman, J. P. Acc. Chem. Res. 1975, 8, 342.
(47) Magens, S.; Ertelt, M.; Jatsch, A.; Plietker, B. Org. Lett. 2008, 10,
53.
(35) Ueno, S.; Mizushima, E.; Chatani, N.; Kakiuchi, F. J. Am. Chem.
Soc. 2006, 128, 16516.
(36) Bradley, C. A.; Veiros, L. F.; Pun, D.; Lobkovsky, E.; Keresztes,
I.; Chirik, P. J. J. Am. Chem. Soc. 2006, 128, 16600.
(37) Faller, J. W.; Linebarrier, D. Organometallics 1988, 7, 1670.
(38) Bonanno, J. B.; Henry, T. P.; Neithamer, D. R.; Wolczanski, P. T.;
Lobkovsky, E. B. J. Am. Chem. Soc. 1996, 118, 5132.
(39) Kawaguchi, H.; Matsuo, T. J. Am. Chem. Soc. 2003, 125, 14254.
(40) Yamamoto, T.; Ishizu, J.; Yamamoto, A. Chem. Lett. 1979, 1385.
(48) Tolman, C. A.; Ittel, S. D.; English, A. D.; Jesson, J. P. J. Am.
Chem. Soc. 1979, 101, 1742.
(49) Ittel, S. D.; Tolman, C. A.; English, A. D.; Jesson, J. P. J. Am.
Chem. Soc. 1978, 100, 7577.

6274 Organometallics, Vol. 27, No. 23, 2008
TroVitch et al.
Figure 4. Molecular structure of 1-Allyl at 30% probability ellipsoids. Hydrogen atoms are omitted for clarity. Core of the molecule
presented on the right.
Figure 5. Molecular structure of 1-(OAc)(Vinyl) at 30% probability ellipsoids (left). Side view of the core of the molecule (right). Hydrogen
atoms are omitted for clarity.
Scheme 9
For methyl benzoate addition, exclusive scission of the ester,
not the acyl C-O bond, was observed.
yields two iron products where formal electron loss occurs at
the bis(imino)pyridine chelate and maintains the ferrous oxida-
tion state (Scheme 9).21 Several examples of C-O bond
cleavage reported in this work follow this general reaction
paradigm: oxidative addition to yield two iron complexes with
formal oxidation of the ligand not the metal.
Oxidative Addition in Reduced Bis(imino)pyridine Iron
Compounds. Oxidative addition reactions to reduced iron
complexes bearing electronically non-innocent bis(imino)pyri-
dine chelates raise the interesting possibility of having ligand-
rather than metal-based redox events. Related ligand-based
oxidative chemistry has been reported by Heyduk and co-
workers with group 4 transition metals.50 Previous studies from
our laboratory have established that oxidative addition of alkyl
halides (e.g., CH3CH2Br, (CH3)2CHCH2Br, etc.) to 1-(N2)2
(50) (a) Blackmore, K. J.; Lal, N.; Ziller, J. W.; Heyduk, A. F. J. Am.
Chem. Soc. 2008, 130, 2728. (b) Ketterer, N. A.; Fan, H.; Blackmore, K. J.;
Yang, X.; Ziller, J. W.; Baik, M.-H.; Heyduk, A. F. J. Am. Chem. Soc.
2008, 130, 4364.

Carbon-Oxygen Bond CleaVage
Organometallics, Vol. 27, No. 23, 2008 6275
Scheme 10
Scheme 11
Scheme 12
The oxidative addition of vinyl acetate to 1-(N2)2 provides
additional insight into the nature of the bond cleavage event at
the reducing iron center. Observation of the six-coordinate
1-(OAc)(Vinyl) compound suggests that such monometallic
oxidative addition products may be intermediates common to
all C-O bond cleavage reactions. In cases where this species
is not observed, it is likely that Fe-C bond homolysis occurs
from the initial five- or six-coordinate oxidative addition product
followed by radical capture to yield the observed products.
Several observations support this assertion. Attempts to prepare
five-coordinate bis(imino)pyridine iron bis(neopentyl) or neo-
pentyl chloride complexes by ligand substitution reactions
resulted in the isolation of four-coordinate (iPrPDI)FeX com-
pounds arising from Fe-C bond homolysis.26 Likewise, oxida-
tive addition of 5-hexenyl-1-bromide to 1-(N2)2 yielded 1-Br
along with the cyclized iron alkyl product,21 consistent with

6276 Organometallics, Vol. 27, No. 23, 2008
TroVitch et al.
ejection of the 5-hexenyl radical, which is known to undergo
rapid cyclization.51,52 In addition, hydrogenation of the vinyl
ligand of 1-(OAc)(Vinyl) in the presence of palladium on carbon
resulted in conversion to 1-OAc, supporting radical ejection
upon conversion to the saturated iron-ethyl intermediate. It
should be noted, however, that the difference in bond strengths
between sp2- and sp3-hybridized C-O bonds may result in a
change in mechanism, and in certain cases, radical ejection may
occur without prior formation of an iron-carbon bond.
Relative Rates and Selectivity for C-O Cleavage by
1-(N2)2. When examining the relative rates of C-O bond
cleavage within classes of ethers and esters promoted by 1-(N2)2,
several salient trends emerge. Within the family of ester
substrates, benzoate esters generally undergo more rapid cleav-
age than the corresponding acetates. Specifically, ethyl benzoate
(exclusive ester cleavage) converted to products over the course
of hours at 23 °C, while ethyl acetate (both acyl and ester
cleavage) required nearly a week under the same conditions.
This effect is likely electronic in origin as the electron
withdrawing phenyl group renders the substrate more electro-
philic and prone to oxidative cleavage. It is also possible that
differences in one electron reduction potentials account for the
observed differences.
C-O oxidative addition.58 Using pyridine directing groups, a
Ru3(CO)12-catalyzed reductive decarbonylation method has also
been developed and has been extended to alkyl-substituted
substrates.59
To our knowledge, acyl C-O bond cleavage in alkyl esters
lacking a directing group has not been reported. Thus, the facile
C-O bond cleavage of alkyl acetates with 1-(N2)2 at 65 °C
highlights the reducing potential of a formally iron(II) center
with electrons stored in the bis(imino)pyridine chelate. For the
alkyl acetates presented in Scheme 5, a modest trend in
selectivity emerges. The smallest member in the series, methyl
acetate, undergoes rapid and selective ester C-O bond cleavage
likely due to the steric accessibility of the ester C-O bond.
Lengthening the alkyl chain resulted in slower rates as ethyl,
pentyl, benzyl, isopropyl, cyclohexyl, and tert-butyl acetate all
require heating to 65 °C or extended reaction times at 23 °C to
reach conversion. With the exception of tert-butyl acetate, the
selectivity of acyl bond cleavage increases as the steric
protection of the ester C-O bond increases. Thus, for alkyl-
substituted acetates, it appears that in the absence of overriding
steric effects, ester C-O bond cleavage is preferred over acyl.
Why tert-butyl acetate reverses the kinetic selectivity for C-O
bond cleavage is not known.
Mechanistic Considerations. Proposed mechanisms for
ether, ester, and acyl C-O bond cleavage are presented in
Schemes 10 (ethers) and 11 (ester, acyl) and are inspired by
pathways previously reported by Yamamoto for electron rich
Ni(0) compounds.24,40,42,43 Because ether cleavage was only
observed with allyl and vinyl-substituted ethers, initial coordina-
tion of the alkene of the substrate is proposed.43 For diallyl and
allyl ethyl ethers, SN2′-type substitution is also plausible and
may be preferred accounting for the observed rapid rates of
reaction. Oxidative addition of the C-O bond yields a five-
coordinate iron allyl alkoxide intermediate, analogous to the
six-coordinate iron vinyl acetate complex, 1-(OAc)(Vinyl), that
was isolated and structurally characterized. Ejection of allyl
radical and capture by a reduced iron species, either 1-(N2)2 or
the iron olefin compound, yields the observed products. To have
a significant concentration of reduced iron compound available
for allyl radical capture, C-O bond cleavage must be the rate-
determining step. Circumstantial experimental data support this
assertion. Addition of one equivalent of diallyl ether to 1-(N2)2
yielded an equimolar mixture of 1-Allyl and 1-OCH2CHdCH2
with unreacted substrate remaining. This suggests that the
products do not participate in radical capture. For the case of
ethyl vinyl ether cleavage, the olefin compound was observed
by 1H NMR spectroscopy immediately after mixing the reagents;
C-O bond cleavage occurred over the course of 24 h at 23 °C.
In general, when the C-O bond cleavage reactions are fast, all
of the products of radical capture are observed. When these
reactions are slow, as in the case of ethyl vinyl ether, the fate
of the ejected organic radical fragment has not been determined.
Similar pathways are proposed in Scheme 11 for ester and
acyl C-O bond cleavage. Ethyl acetate is presented as a
representative substrate because both ester and acyl scissions
were observed. For ester C-O bond cleavage, oxidative addition
to yield the six-coordinate iron ethyl acetate complex followed
by iron-carbon bond homolysis yields the observed 1-OAc
product. A six-coordinate iron intermediate is proposed on the
basis of the analogy to isolated 1-(OAc)(Vinyl). Ejection of
the ethyl radical from this compound seems plausible given the
Comparison of various substituted acetates also establishes
several important trends. The unsaturated esters, vinyl and allyl
acetate, undergo extremely rapid C-O bond cleavage; a reaction
so fast that no catalytic hydrogenation is observed in the
presence of 1-(N2)2 and 4 atm of H2. In contrast, C-O bond
cleavage in trans-methyl cinnamate, while relatively facile under
an N2 atmosphere, is sufficiently slow such that catalytic olefin
hydrogenation can be achieved with 5 mol % of 1-(N2)2 and 4
atm of dihydrogen. Eventually, however, all of the iron is
converted to catalytically inactive iron carboxylate compounds,
establishing C-O bond activation as the principal catalyst
deactivation pathway for this class of substrates.
Carbon-oxygen bond cleavage in various allylated substrates
also demonstrates the importance of leaving group effects. For
diallyl and allyl ethyl ether, the rate of C-O bond cleavage is
sufficiently slow such that iron-catalyzed olefin hydrogenation
is observed. As mentioned previously, allyl acetate has not been
hydrogenated using our experimental conditions because of the
extremely fast rate of C-O bond oxidative addition. This trend
is similar to those observed in nickel(0) and palladium(0)
chemistry, where allyl acetates undergo more facile oxidative
addition than the corresponding allylated ethers.33,42,43
It is noteworthy that 1-(N2)2 promotes acyl C-O bond
cleavage in alkyl esters. Electron rich Ni(0) compounds42
undergo selective acyl C-O bond cleavage with phenyl acetate
but are unreactive toward alkyl-substituted esters. Aryl acetate
cleavage reactions are also known with (dppe)2Mo(N2)2 (dppe
) 1,2-diphenylphosphinoethane),53 (Ph3P)3RhH,54 (η5-C5Me5)-
RhCl(mdmpp-κP, -κO),55 (Ph3P)3Ru(CO)2,56 and Ru3(CO)12.57
In one case, an acyl rhodium complex was isolated following
(51) Lal, D.; Griller, D.; Husband, S.; Ingold, K. U. J. Am. Chem. Soc.
1974, 96, 6355.
(52) Williams, G. M.; Schwartz, J. J. Am. Chem. Soc. 1982, 104, 1122.
(53) Tatsumi, T.; Tominaga, H.; Hidai, M.; Uchida, Y. J. Organomet.
Chem. 1981, 218, 177.
(54) Yamamoto, T.; Miyashita, S.; Naito, Y.; Komiya, S.; Ito, T.;
Yamamoto, A. Organometallics 1982, 1, 808.
(55) Yamamoto, Y.; Han, X.-H.; Ma, J.-F. Angew. Chem., Int. Ed. 2000,
39, 1965.
(56) Hiraki, K.; Kira, S.-I.; Kawano, H. Bull. Chem. Soc. Jpn. 1997,
70, 1583.
(57) Tatamidani, H.; Yokota, K.; Kakiuchi, F.; Chatani, N. J. Org. Chem.
2004, 69, 5615.
(58) Grotjahn, D. B.; Joubran, C. Organometallics 1995, 14, 5171.
(59) Chatani, N.; Tatamidani, H.; Ie, Y.; Kakiuchi, F.; Murai, S. J. Am.
Chem. Soc. 2001, 123, 4849.

Carbon-Oxygen Bond CleaVage
Organometallics, Vol. 27, No. 23, 2008 6277
previous studies from our laboratory that have demonstrated
the propensity of five-coordinate bis(imino)pyridine iron dialkyls
to undergo Fe-C homolysis.26 For acyl C-O bond cleavage,
oxidative addition of the C-O bond forms a five-coordinate
intermediate analogous to 1-(OAc)(Vinyl). Decarbonylation
followed by Fe-CH3 homolysis yields the observed product.
Observation of 1-Me and 1-Cl following treatment of 1-(N2)2
with CH3C(O)Cl provides a precedent of decarbonylation in
bis(imino)pyridine iron compounds.
addition of the C-O bond to the reduced bis(imino)pyridine
iron complex is proposed. In all but one case, vinyl acetate,
homolysis of the iron-carbon bond from the product of
oxidative addition yields the observed products. Deuterium
labeling studies with isotopologues of the iron dinitrogen
complex and ethyl acetate establish competing cyclometalation
of the 2,6-diisopropyl aryl substituents. Taken together, these
studies once again highlight both the chemical and redox-activity
of bis(imino)pyridine ligands in reduced iron chemistry.
Deuterium labeling studies with ethyl acetate also established
a competing C-H activation pathway. A proposed mechanism
for this transformation, not required for C-O bond cleavage,
is presented in Scheme 12. Recall that this reaction is selective:
isotopic exchange occurs only between the isopropyl aryl methyl
groups on the bis(imino)pyridine chelate and the methylene
position on the ethyl chain adjacent to the oxygen. The process
begins with reversible dissociation-coordination of ethyl acetate
to facilitate C-H activation of either the methylene position of
the ester or cyclometalation of an isopropyl methyl group of
the bis(imino)pyridine chelate. Following the formation of the
alkyl deuteride, isotopic exchange can occur through cyclom-
etalation, likely by σ-bond metathesis, with either the iron-
hydride (deuteride) or alkyl. If the reaction occurs with the iron
hydride (deuteride), loss of H-D occurs. Subsequent productive
recapture places deuterium in the isopropyl group and hydrogen
in the ester, completing isotopic exchange. An alternative σ-bond
metathesis pathway involves the reaction of the isopropyl methyl
group with the iron alkyl, liberating free ethyl acetate, which
following reductive elimination of the iron cyclometalated
deuteride and recapture, completes isotopic exchange. Pathways
involving formal oxidative addition and formation of iron(IV)
intermediates are also possible but are deemed less plausible.
The selective isotopic exchange between the methylene
position of the ester and the isopropyl methyl groups of the
bis(imino)pyridine chelate raises the question of why this process
was not observed with the corresponding iron methyl acetate
complex. The lack of a competing C-H activation process for
the methyl acetate derivative is accommodated by the relative
rates of C-O bond cleavage for the two substrates. Whereas
C-O bond cleavage in ethyl acetate occurs over one hour at
65 °C, methyl acetate cleavage occurs within minutes at 23 °C.
The lack of deuterium exchange with methyl acetate clearly
demonstrates that C-H activation can be competitive with C-O
cleavage, but is by no means a prerequisite.
Experimental Section60
Preparation of (iPrPDI)Fe(O2CMe)(CHdCH2) (1-(OAc)(Vinyl)).
A 20 mL scintillation vial was charged with 0.200 g (0.337 mmol)
of 1-(N2)2 and approximately 10 mL of pentane. A separate solution
of 0.029 g (32 µL, 0.337 mmol) of vinyl acetate in ∼5 mL of
pentane was slowly added dropwise while stirring, resulting in
immediate evolution of N2 along with a change in color to bright
purple. After 2 h, the solution was filtered through Celite, and the
solvent was removed in Vacuo to yield a purple residue. Recrys-
tallization from pentane yielded 0.131 g (62%) of a bright purple
solid identified as 1-(OAc)(Vinyl). Analysis for C37H49FeN3O2:
Calcd C, 71.26; H, 7.92; N, 6.74. Found: C, 71.11; H, 7.74; N,
6.61. 1H NMR (benzene-d6): δ ) 10.07 (dd, 16.0 Hz, 7.5 Hz, 1H,
CH ) CH2), 7.43 (d, 7.5 Hz, 2H, m-pyr), 7.25 (t, 7.5 Hz, 1H, p-pyr),
6.96 (d, 6.0 Hz, 1H, m-aryl), 6.95 (d, 6.0 Hz, 1H, m-aryl), 4.66 (d,
7.5 Hz, 1H, CHdCH2), 3.12 (sept., 7.0 Hz, 2H, CH(CH3)2), 1.94
(sept., 7.0 Hz, 2H, CH(CH3)2), 1.87 (d, 16.0 Hz, 1H, CHdCH2),
1.78 (m, 12H, CH(CH3)2 and C(CH3)), 1.23 (d, 7.0 Hz, 6H,
CH(CH3)2), 1.18 (s, 3H, CO2CH3), 1.07 (d, 7.0 Hz, 6H, CH(CH3)2),
0.80 (d, 7.0 Hz, 6H, CH(CH3)2), p-aryl peak not located. 13C{1H}
NMR (benzene-d6): δ ) 189.91 (CHdCH2), 181.48 (CO2Me),
163.67 (C(CH3)), 144.44 (aryl), 143.20, 142.26, 126.89 (p-aryl),
124.28 (m-aryl), 123.99 (m-aryl), 120.79 (m-py), 118.39 (p-py),
117.14 (CHdCH2), 28.54 (CH(CH3)2), 28.25 (CH(CH3)2), 25.59
(CH(CH3)2), 25.07 (CH(CH3)2), 24.81 (CH(CH3)2), 24.51
(CH(CH3)2), 22.42 (CO2CH3), 19.20 (C(CH3)).
Preparation of (iPrPDI)Fe(OCH2CHdCH2) (1-OCH2CHdCH2).
A 20 mL scintillation vial was charged with 0.050 g (0.084 mmol)
of 1-(N2)2 and approximately 15 mL of pentane. Using a microsy-
ringe, 0.005 g (6 µL, 0.084 mmol) of allyl alcohol was added to 5
mL of pentane, and this mixture was added slowly to the stirring
1-(N2)2 solution. After one hour, the resulting solution was filtered
through Celite, and the solvent was removed in vacuo to yield 0.026
g (52%) of a dark brown solid identified as 1-OCH2CHdCH2.
This complex was additionally observed upon the addition of allyl
ether to 1-(N2)2. Analysis for C35H48FeN3O: Calcd C, 72.15; H,
8.30; N, 7.21. Found: C, 71.94; H, 8.30; N, 6.99. Magnetic
susceptibility: µeff ) 4.0 µB (benzene-d6, 20 °C). 1H NMR (benzene-
d6, 20 °C): δ ) 112.09 (190 Hz, 1H), 92.59 (298 Hz, 1H), 73.61
(73 Hz, 2H), 54.76 (52 Hz, 1H), -8.90 (28 Hz, 2H, m-aryl), -14.82
(24 Hz, 1H, p-aryl), -24.26 (29 Hz, 12H, CH(CH3)2), -38.47 (149
Hz, 12H, CH(CH3)2), -118.79 (394 Hz, 4H, CH(CH3)2), -219.87
(131 Hz, 6H, C(CH3)), one peak not located.
Concluding Remarks
Investigation into the substrate scope of catalytic olefin
hydrogenation reactions promoted by the bis(imino)pyridine iron
complex, 1-(N2)2, established that oxidative addition of
carbon-oxygen bonds of ethers and esters is a principal catalyst
deactivation pathway. For substrates such as diallyl ether, vinyl
ethyl ether, and trans-methyl cinnamate, catalytic hydrogenation
is competitive with C-O cleavage, and conversion to alkane
was observed. For molecules such as allyl and vinyl acetate,
C-O cleavage is more rapid than hydrogenation (up to 4 atm
of H2), and no turnover was observed. Exploration of the scope
of ester C-O bond cleavage established competing ester and
acyl C-O bond cleavage reactions, depending on the specific
substrate. Methyl acetate and trans-methyl cinnamate underwent
exclusive ester C-O bond cleavage, while phenyl acetate
yielded products from selective acyl C-O bond scission. In
alkyl-substituted esters such as ethyl, pentyl, isopropyl, cyclo-
hexyl, and tert-butyl acetate, competing ester and acyl C-O
bond cleavage occurred. In both ethers and esters, oxidative
Preparation of (iPrPDI)Fe(CH2CHdCH2) (1-Allyl). This com-
pound was directly synthesized by the slow addition of 0.053 g of
allylmagnesium bromide (362 µL of a 1.0 M solution in ether, 0.362
mmol) to a cold stirring pentane solution containing 0.175 g (0.283
mmol) of 1-Br. Approximately 1 mL of 1,4-dioxane was added to
the reaction mixture to aid MgBr2 precipitation. After stirring for
1 h, the reddish-brown solution was filtered through Celite, and
the solvent was removed in vacuo to yield a reddish-brown solid.
This residue was washed 5 times with approximately 2 mL of
pentane, and the resulting concentrated pentane solution was chilled
to -35 °C for days. Recrystallization yielded 0.035 g (21%) of a
brown solid identified as 1-Allyl. This complex was also observed
upon the addition of substoichiometric amounts of allyl ether, ethyl
allyl ether, or allyl acetate to 1-(N2)2. Analysis for C36H48FeN3:

6278 Organometallics, Vol. 27, No. 23, 2008
TroVitch et al.
Calcd C, 74.73; H, 8.36; N, 7.26. Found: C, 74.66; H, 8.49; N,
6.90. Magnetic susceptibility: µeff ) 2.4 µB (benzene-d6, 20 °C).
1H NMR (benzene-d6, 20 °C): δ ) 148.64 (770 Hz, 2H, allyl),
98.47 (795 Hz, 1H, allyl), 72.80 (619 Hz, 2H, allyl), 47.64 (42 Hz,
2H, m-pyr), 8.07 (39 Hz, 2H), 2.99 (59 Hz, 4H, CH(CH3)2), 0.09
(48 Hz, 12H, CH(CH3)2), -0.46 (50 Hz, 12H, CH(CH3)2), -26.64
(184 Hz, 6H, C(CH3)), two peaks not located.
13C{1H} (benzene-d6): δ ) 170.3, 155.6, 147.0, 146.7, 144.2, 143.3,
141.5, 129.1 (m-aryl), 125.9, 124.2, 122.4, 116.7 (p-aryl), 79.3 (η6-
m-aryl), 78.4 (η6-p-aryl), 28.2 (CH(CH3)2), 27.6 (CH(CH3)2), 27.0
(C(CH3)), 24.9 (CH(CH3)2), 24.7 (CH(CH3)2), 23.4 (CH(CH3)2),
22.4 (CH(CH3)2), 16.3 (C(CH3)).
Preparation of (iPrPDI)Fe(O2CCHdCH(Ph)) (1-CIN). A 20
mL scintillation vial was charged with 0.050 g (0.084 mmol) of
1-(N2)2 and approximately 10 mL of pentane. To the solution, 0.013
g (0.084 mmol) of trans-cinnamic acid was added, and dinitrogen
evolution was observed along with a color change from green to
brown. The solution was allowed to stir for 2 h and was then filtered
through Celite. The solvent was removed in Vacuo, and the residue
was recrystallized from pentane at -35 °C to afford 0.036 g (62%)
of a brown solid identified as 1-CIN. Analysis for C42H50FeN3O:
Calcd C, 75.03; H, 7.50; N, 6.25. Found: C, 75.24; H, 7.90; N,
Preparation of (iPrPDIFe)(CH3O2CCH2CH3) (1-EtOAc). A 20
mL scintillation vial was charged with 0.80 g (0.135 mmol) of
1-(N2)2 and approximately 10 mL of diethyl ether. With stirring,
0.012 g (13 µL, 0.135 mmol) of ethyl acetate was added via
microsyringe resulting in immediate evolution of N2along with a
change in color to reddish-brown. After 30 min, the solution was
filtered though a Celite fitted frit, and the solvent was removed in
vacuo to yield 0.070 g (83%) of a dark brown solid identified as
1-EtOAc. Analysis for C37H51FeN3O2: Calcd C, 71.03; H, 8.22;
1
6.32. H NMR (benzene-d6, 20 °C): δ ) -283.51 (336 Hz, 6H,
1
N, 6.72. Found: C, 70.74; H, 7.86; N, 6.97. H NMR (toluene-d8,
C(CH3)), -113.05 (524 Hz, 4H, CH(CH3)2), -30.12 (135 Hz, 12H,
CH(CH3)2), -19.56 (50 Hz, 12H, CH(CH3)2), -16.66 (42 Hz, 1H,
p-aryl), -3.14 (47 Hz, 2H, m-aryl), 11.82 (30 Hz, 1H, p-phenyl),
19.01 (16 Hz, 2H, m-phenyl), 27.99 (45 Hz, 2H, o-phenyl), 58.80
(191 Hz, 1H, CH(C6H5)), 119.25 (294 Hz, 2H, m-pyr), 159.36 (275
Hz, 1H, COCH), 372.72 (219 Hz, 1H, p-pyr).
-60 °C): δ ) 11.61 (d, 7.5 Hz, 2H, m-pyr), 8.66 (t, 7.0 Hz, 1H,
p-pyr), 7.50 (t, 8.0 Hz, 2H, p-aryl), 7.09 (d, 7.0 Hz, 4H, m-aryl),
2.92 (q, 6.0 Hz, 2H, CH2CH3), 2.52 (sept., 5.5 Hz, 4H, CH(CH3)2),
1.24 (d, 5.5 Hz, 12H, CH(CH3)2), 0.81 (t, 6.0 Hz, 3H, CH2CH3),
0.58 (s, 3H, COCH3), -0.12 (d, 5.5 Hz, 12H, CH(CH3)2), -4.67
(s, 6H, C(CH3)). 13C {1H} NMR (toluene-d8, -60 °C): δ ) 183.41,
181.65, 163.30, 160.00, 137.62, 136.80 (p-pyr), 124.53 (p-aryl),
123.69 (m-aryl), 103.83 (m-pyr), 64.22 (CH2CH3), 35.39 (C(CH3)),
28.29 (CH(CH3)2), 23.73 (CH(CH3)2), 22.86 (CH(CH3)2), 22.33
(COCH3), 13.79 (CH2CH3).
Preparation of (iPrPDI)Fe(O2CMe) (1-OAc). Method A: A
solution of 1-(EtOAc) was prepared in a thick-walled glass vessel
by adding 21 µL (0.202 mmol) of ethyl acetate to 10 mL of a
pentane solution containing 0.120 g (0.202 mmol) of 1-(N2)2. The
vessel was sealed, and the solution was heated to 65 °C in an oil
bath for 16 h. The resulting brown solution was filtered though
Celite, and the solvent was removed in vacuo to yield a dark brown
solid. Recrystallization from pentane yielded 0.044 g (36%) of
1-OAc. Method B: A thick-walled glass vessel was charged with
0.100 g (0.170 mmol) of 1-(N2)2 and approximately 10 mL of
pentane. Upon addition of 0.015 g (16 µL, 0.189 mmol) of methyl
acetate, the vessel was sealed, submerged in liquid nitrogen, and
evacuated on a high vacuum line. After adding 1 atm of dihydrogen,
the solution was thawed and stirred for 4 h at ambient temperature.
The volatiles were removed, and the resulting reddish-brown residue
was washed through a Celite fitted frit with diethyl ether. The
solvent was removed in vacuo, and recrystallization from pentane
at -35 °C afforded spectroscopically pure 1-OAc. Analysis for
C35H46FeN3O2: Calcd C, 70.46; H, 7.77; N, 7.04. Found: C, 70.25;
H, 8.04; N, 6.79. 1H NMR (benzene-d6, 20 °C): δ ) -284.76 (280
Hz, 6H, C(CH3)), -114.95 (525 Hz, 4H, CH(CH3)2), -30.37 (170
Hz, 12H, CH(CH3)2), -19.78 (75 Hz, 12H, CH(CH3)2), -16.50
(66 Hz, 1H, p-aryl), -3.20 (73 Hz, 2H, m-aryl), 119.59 (163 Hz,
2H, m-pyr), 187.25 (370 Hz, 3H, CO2CH3), 372.47 (18 Hz, 1H,
p-pyr).
Preparation of (iPrPDI)Fe(CHdCH2) (1-Vy). Using a cali-
brated gas bulb, 0.021 mmol of vinyl bromide was transferred to a
J. Young NMR tube containing 0.025 g (0.042 mmol) of 1-(N2)2
and approximately 0.7 mL of benzene-d6. Upon standing at ambient
temperature for 20 min, analysis of the reaction mixture by 1H NMR
spectroscopy revealed the formation of a mixture of 1-Br and 1-Vy.
1H NMR (benzene-d6, 20 °C): δ ) 127.07 (221 Hz, 1H, p-pyr),
52.84 (45 Hz, 2H, m-pyr), -7.37 (32 Hz, 12H, CH(CH3)2), -12.74
(136 Hz, 12H, CH(CH3)2), -55.29 (271 Hz, 4H, CH(CH3)2),
-141.77 (506 Hz, 6H, C(CH3)), aryl and Vinyl resonances not
located.
Preparation of 1-Aryl. To a solution of 1-(N2)2 (0.033 g, 0.056
mmol) in benzene-d6 (0.7 mL) was added (12 µL, 0.057 mmol) of
hexamethyldisiloxane. The dark green solution was transferred to
an NMR tube fitted with a J. Young adapter and placed in a 95 °C
oil bath for 16 h. The resulting burgundy solution was cooled to
room temperature and filtered, and solvent and other volatiles were
removed in vacuo, yielding a pink solid identified as 1-Aryl (0.023
g, 0.044 mmol, 78%). Similaily, 1-Aryl was effectively prepared
in the presence of a stoichiometric amount of anisole, instead of
hexamethyldisiloxane, or in the absence of ether. Analysis for
C33H43N3Fe: Calcd: C, 73.73; H, 8.06; N, 7.82. Found: C, 73.69;
Acknowledgment. We thank the Packard Foundation
(Fellowship in Science and Engineering to P.J.C.) for
financial support. Dr. Bradley Wile is acknowledged for the
characterization of 1-Aryl.
1
H, 7.81; N, 7.95. H NMR (benzene-d6): δ ) 7.37-7.16 (m, 3H,
pyr), 6.61 (d, 3JHH ) 8.0 Hz, 2H, m-aryl), 6.56 (t, 3JHH ) 8.0 Hz,
3
3
1H, p-aryl), 5.46 (d, JHH ) 5.6 Hz, 2H, η6-m-aryl), 4.12 (t, JHH
) 5.6 Hz, 1H, η6-p-aryl), 3.36-3.22 (m, 4H, CH(CH3)2), 2.83 (s,
3H, C(CH3)), 1.46 (d, 3JHH ) 6.8 Hz, 6H, CH(CH3)2), 1.39 (s, 3H,
Supporting Information Available: Complete experimental
details, crystallographic data for 1-(OAc)(vinyl), 1-OBz, 1-OAc,
1-Allyl, and 1-OCH2CHdCH2 in cif format. This material is
available free of charge via the Internet at http://pubs.acs.org.
C(CH3)), 1.26 (d, 3JHH ) 6.8 Hz, 6H, CH(CH3)2), 1.06 (d, 3JHH
6.8 Hz, 6H, CH(CH3)2), 0.97 (d, JHH ) 6.8 Hz, 6H, CH(CH3)2).
)
3
(60) General considerations and additional experimental procedures can
be found in the Supporting Information.
OM8005596