Considering FeII/IV redox processes as mechanistically relevant to the catalytic hydrogenation of olefins by [PhBPiPr 3]Fe-Hx species

Article abstract of DOI:10.1021/ic0488583

Several coordinatively unsaturated pseudotetrahedral iron(II) precursors, [PhBP^iPr₃]Fe-R ([PhBP^iPr₃] = [PhB(CH₂PⁱPr₂)₃]^-; R = Me (2), R = CH₂Ph (3), R = CH₂CMe₃ (4)) have been prepared from [PhBP^iPr₃]FeCl (1) that serve as precatalysts for the room-temperature hydrogenation of unsaturated hydrocarbons (e.g., ethylene, styrene, 2-pentyne) under atmospheric H₂ pressure. The solid-state crystal structures of 2 and 3 are presented. To gain mechanistic insight into the nature of these hydrogenation reactions, a number of [PhBP^iPr₃]-supported iron hydrides were prepared and studied. Room-temperature hydrogenation of alkyls 2-4 in the presence of a trapping phosphine ligand affords the iron(IV) trihydride species [PhBP ^iPr₃]Fe(H)₃(PR₃) (PR₃ = PMe₃ (5); PR₃ = PEt₃ (6); PR₃ = PMePh₂ (7)). These spectroscopically well-defined trihydrides undergo hydrogen loss to varying degrees in solution, and for the case of 7, this process leads to the structurally identified Fe(II) hydride product [PhBP ^iPr₃]Fe(H)(PMePh₂) (9). Attempts to prepare 9 by addition of LiEt₃BH to 1 instead lead to the Fe(I) reduction product [PhBP^iPr₃]Fe(PMePh₂) (10). The independent preparations of the Fe(II) monohydride complex [PhBP ^iPr₃]Fe^II(H)(PMe₃) (11) and the Fe(I) phosphine adduct [PhBP^iPr₃]Fe(PMe₃) (8) are described. The solid-state crystal structures of trihydride 5, monohydride 11, and 8 are compared and demonstrate relatively little structural reorganization with respect to the P₃Fe-P′ core motif as a function of the iron center's formal oxidation state. Although paramagnetic 11 (S = 1) is quantitatively converted to the diamagnetic trihydride 5 under H ₂, the Fe(I) complex 8 (S = 3/2) is inert toward atmospheric H ₂. Complex 10 is likewise inert toward H₂. Trihydrides 5 and 6 also serve as hydrogenation precatalysts, albeit at slower rates than that for the benzyl complex 3 because of a rate-contributing phosphine dependence. That these hydrogenations appear to proceed via well-defined olefin insertion steps into an Fe-H linkage is indicated by the reaction between trihydride 5 and ethylene, which cleanly produces the ethyl complex [PhBP^iPr ₃]Fe(CH₂CH₃) (13) and an equivalent of ethane. Mechanistic issues concerning the overall reaction are described.

Full text of DOI:10.1021/ic0488583

Inorg. Chem. 2004, 43, 7474−7485
/
IV
Considering FeII Redox Processes as Mechanistically Relevant to the
Catalytic Hydrogenation of Olefins by [PhBPiPr3]Fe
−Hx Species
Erin J. Daida and Jonas C. Peters*
DiVision of Chemistry and Chemical Engineering, Arnold and Mabel Beckman Laboratories of
Chemical Synthesis, California Institute of Technology, Pasadena, California 91125
Received August 18, 2004
Several coordinatively unsaturated pseudotetrahedral iron(II) precursors, [PhBPiPr3]Fe
− ] )
R ([PhBPiPr [PhB(CH2PiPr2)3]-;
3
R
)
Me (2), R CH2Ph (3), R
)
)
CH2CMe3 (4)) have been prepared from [PhBPiPr3]FeCl (1) that serve as
precatalysts for the room-temperature hydrogenation of unsaturated hydrocarbons (e.g., ethylene, styrene, 2-pentyne)
under atmospheric H2 pressure. The solid-state crystal structures of 2 and 3 are presented. To gain mechanistic
insight into the nature of these hydrogenation reactions, a number of [PhBPiPr3]-supported iron hydrides were prepared
and studied. Room-temperature hydrogenation of alkyls 2
−
)
4 in the presence of a trapping phosphine ligand affords
the iron(IV) trihydride species [PhBPiPr3]Fe(H)3(PR3) (PR3
PMe3 (5); PR3 PEt3 (6); PR3 PMePh2 (7)). These
)
)
spectroscopically well-defined trihydrides undergo hydrogen loss to varying degrees in solution, and for the case
of 7, this process leads to the structurally identified Fe(II) hydride product [PhBPiPr3]Fe(H)(PMePh2) (9). Attempts
to prepare 9 by addition of LiEt3BH to 1 instead lead to the Fe(I) reduction product [PhBPiPr3]Fe(PMePh2) (10). The
independent preparations of the Fe(II) monohydride complex [PhBPiPr3]FeII(H)(PMe3) (11) and the Fe(I) phosphine
adduct [PhBPiPr3]Fe(PMe3) (8) are described. The solid-state crystal structures of trihydride 5, monohydride 11,
and 8 are compared and demonstrate relatively little structural reorganization with respect to the P3Fe
motif as a function of the iron center’s formal oxidation state. Although paramagnetic 11 (S 1) is quantitatively
converted to the diamagnetic trihydride 5 under H2, the Fe(I) complex 8 (S
3/2) is inert toward atmospheric H2.
−P′ core
)
)
Complex 10 is likewise inert toward H2. Trihydrides 5 and 6 also serve as hydrogenation precatalysts, albeit at
slower rates than that for the benzyl complex 3 because of a rate-contributing phosphine dependence. That these
hydrogenations appear to proceed via well-defined olefin insertion steps into an Fe−H linkage is indicated by the
reaction between trihydride 5 and ethylene, which cleanly produces the ethyl complex [PhBPiPr3]Fe(CH2CH3) (13)
and an equivalent of ethane. Mechanistic issues concerning the overall reaction are described.
Introduction
instance, hydridic iron sites may be important to the ability
of nitrogenase enzymes to reduce unsaturated organic
substrates (e.g., HCtCH, H2CdCH2)4 and have been
implicated as intermediates in hydrogenase enzymes.2,5 The
synthesis and study of well-defined synthetic iron complexes
Iron hydride complexes have played a prominent role in
fundamental organometallic studies,1 and their continued
examination is of increasing importance in light of their
possible function in various biocatalytic reactions.2,3 For
(3) (a) Hills, A.; Hughes, D. L.; Jiminez-Tenorio, M.; Leigh, G. J.; Rowley,
A. T. J. Chem. Soc., Dalton Trans. 1993, 3041. (b) Smith, J. M.;
Lachiotte, R. J.; Holland, P. L. J. Am. Chem. Soc. 2003, 125, 15752.
(c) Brown, S. D.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 4538.
(4) (a) Ashby, G. A.; Dilworth, M. J.; Thorneley, R. N. F. Biochem. J.
1987, 247, 547. (b) Burgess, B. K.; Lowe, D. J. Chem. ReV. 1996, 96,
2983. (c) Eady, R. R. Chem. ReV. 1996, 96, 3013.
(5) (a) Pavlov, M.; Seigbahn, P. E. M.; Blomberg, M. R. A.; Crabtree, R.
G. J. Am. Chem. Soc. 1998, 120, 548. (b) Gloaguen, F.; Lawrence, J.
D.; Rauchfuss, T. B. J. Am. Chem. Soc. 2001, 123, 9476. (c) Razavet,
M.; Borg, S. J.; George, S. J.; Best, S. P.; Fairhurst, S. A.; Pickett, C.
J. Chem. Commun. 2002, 700. (d) Adams, M. W. W.; Stiefel, E. I.
Curr. Opin. Chem. Biol. 2000, 4, 214.
* Author to whom correspondence should be addressed. E-mail:
jpeters@caltech.edu.
(1) (a) Morris, R. H.; Sawyer, J. F.; Shiralian, M.; Zubkowski, J. D. J.
Am. Chem. Soc. 1985, 107, 5581. (b) Bianchini, C.; Peruzzini, M.;
Zanobini, F. J. Organomet. Chem. 1988, 354, C19. (c) Field, L. D.;
Messerle, B. A.; Smernik, R. J. Inorg. Chem. 1997, 36, 5984. (d)
Stoppioni, P.; Mani, F.; Sacconi, L. Inorg. Chim. Acta 1974, 11, 227.
(2) (a) Rauchfuss, T. B. Inorg. Chem. 2004, 43, 14. (b) Darensbourg, M.
Y.; Lyon, E. J.; Zhao, X.; Georgakaki, I. P. Proc. Natl. Acad. Sci.
U.S.A. 2003, 100, 3683. (c) Darensbourg, M. Y.; Lyon, E. J.; Smee,
J. J. Coord. Chem. ReV. 2000, 206, 533. (d) Peters, J. W.; Lanzilotta,
W. N.; Lemon, B. J.; Seefeldt, L. C. Science 1998, 282, 1853.
7474 Inorganic Chemistry, Vol. 43, No. 23, 2004
10.1021/ic0488583 CCC: $27.50
© 2004 American Chemical Society
Published on Web 10/14/2004

FeII/IV Redox and Hydrogenation of Olefins
that can facilitate such transformations therefore represent
exciting areas of research.6
and examined a series of [PhBPiPr3]Fe-Hx species that are
either Fe(II) monohydride or Fe(IV) trihydride derivatives.
These hydride precursors are themselves able to mediate the
catalytic hydrogenation of analogous alkene and alkyne
substrates under ambient conditions (room temperature, 1
atm H2). Direct olefin insertion into Fe(II)-H linkages
appears to be mechanistically important, and FeII/IV oxidative
addition/reductive elimination processes are plausibly in-
voked by reference to the apparent propensity for FeII/IV redox
processes within the “[PhBPiPr3]Fe” manifold. In certain
regards, the hydrogenation chemistry described therefore
appears to be mechanistically related to homogeneous
hydrogenation processes more typical of noble metal systems.
Concurrent with the present study, we note that Chirik
and co-workers have discovered structurally distinct iron
systems that also mediate catalytic olefin reduction under
ambient conditions.10 The systems Chirik and co-workers
describe are notably much more active and appear to be
mechanistically distinct as they are suggested to proceed via
Fe0/II redox steps. The [PhBPiPr3]Fe system described here
is attractive for fundamental studies in that it enables the
direct observation of several species that are plausible
intermediates or close relatives of such intermediates and
also enables the study of certain stoichiometric transforma-
tions that are likely relevant to the overall catalytic hydro-
genation cycle.
In this broad context, the respective groups of Darens-
bourg, Rauchfuss, and others have studied dithiolate-sup-
ported diiron µ-H systems capable of H2/D2 scrambling and
catalytic proton reduction.2a-c,5b,c,6b,c,e Under photolytic condi-
tions, some of these diiron systems can even mediate olefin/
D2 scrambling, where reversible olefin insertion into an
Fe-H linkage has been implicated.6c What is perhaps
surprising is that well-defined molecular iron hydrides that
catalytically reduce alkenes under ambient conditions (23
°C, 1 atm H2) in fact have little precedent. To the best of
our knowledge, Bianchini has contributed the sole report of
a thoroughly characterized molecular iron system that is
catalytically active toward the reduction of unsaturated
hydrocarbons under mild conditions, [[P(CH2CH2PPh2)3]Fe-
(H)(L)][BPh4] (L ) N2 or H2).6a This hydride catalyst system
reduces alkynes (e.g., PhCtCH) to their corresponding
alkenes under 1 atm of H2 at 20 °C (TOF ≈ 3 mol of
substrate per mol of catalyst per hour) but is incapable of
olefin reduction. Although zero-valent iron carbonyl systems
that do mediate olefin reduction were in fact reported several
decades ago,7 these systems typically require high temper-
atures and pressures, or photochemical methods,8 to achieve
measurable activity. Moreover, they are typically selective
for diene substrates because of instability of the catalysts in
the absence of a coordinating diene.7
The present study describes a family of tris(phosphino)-
borate-supported iron complexes within the context of olefin
hydrogenation chemistry. The study was motivated by our
recent observation that an elusive “[PhBP3]FeII-H” species
([PhBP3] ) [PhB(CH2PPh2)3]-), formed during the hydro-
genolytic cleavage of a low-spin imide [PhBP3]FeIII≡N-p-
tolyl, mediates the partial reduction of benzene via the
insertion of benzene into a reactive Fe-H linkage (eq 1).3c
This transformation suggested to us the possibility that a
catalytic olefin hydrogenation process based upon iron might
be realized if certain troublesome nuances of the [PhBP3]Fe
platform could be circumvented. We now report that 14-
electron pseudotetrahedral iron alkyl derivatives supported
by the second generation [PhBPiPr3] ligand ([PhBPiPr3] )
[PhB(CH2PiPr2)3]-)9 can serve as precatalysts for the room
temperature hydrogenation of various alkene and alkyne
substrates under atmospheric hydrogen at modest rates. To
ascertain whether hydride intermediates play a key role in
these reduction reactions, we have independently prepared
Results and Discussion
Preparation of S ) 2 [PhBPiPr3]FeII-R Systems. Room-
temperature hydrogenation of electron deficient, pseudo-
tetrahedral Fe(II) alkyls provides access to the hydrogenation
manifold of interest.11 The required [PhBPiPr3]Fe-R precur-
sors can be prepared in good yield via metathesis reactions
between [PhBPiPr3]FeCl (1) and suitable Grignard or alkyl-
lithium reagents (Scheme 1).9 For instance, the addition of
MeLi to 1 at low temperature in tetrahydrofuran (THF)
provides [PhBPiPr3]Fe-Me (2) as yellow crystals after
workup in 77% yield. A similar procedure using PhCH2-
MgCl or neopentyllithium affords red crystals of [PhBPiPr3]Fe-
CH2Ph (3) (83% isolated) and orange crystals of [PhBPiPr3]Fe-
1
CH2CMe3 (4) (55% isolated), respectively. The H NMR
spectra of these 14-electron alkyl species show reasonably
well-defined but paramagnetically shifted and broadened
resonances at 23 °C. The solution magnetism (Evans12)
(6) (a) Bianchini, C.; Meli, A.; Peruzzini, M.; Frediani, P.; Bohanna, C.;
Esteruelas, M. A.; Oro, L. A. Organometallics 1992, 11, 138. (b) Zhao,
X.; Georgakaki, I. P.; Miller, M. L.; Mejia-Rodriguez, R.; Chiang,
C.-Y.; Darensbourg, M. Y. Inorg. Chem. 2003, 41, 3917. (c) Zhao,
X.; Chiang, C.-Y.; Miller, M. L.; Rampersad, M. V.; Darensbourg,
M. Y. J. Am. Chem. Soc. 2003, 125, 518. (d) Kayal, A.; Rauchfuss,
T. B. Inorg. Chem. 2003, 42, 5046. (e) Rauchfuss, T. B. Inorg. Chem.
2004, 43, 14. (f) Bhugun, I.; Lexa, D.; Save´ant, J.-M. J. Am. Chem.
Soc. 1996, 118, 3982.
(7) (a) Frankel, E. N.; Emken, E. A.; Peters, H. M.; Davison, V. L.;
Butterfield, R. O. J. Org. Chem. 1964, 29, 3292. (b) Frankel, E. N.;
Emken, E. A.; Davison, V. L. J. Org. Chem. 1965, 30, 2739. (c) Cais,
M.; Maoz, N. J. Chem. Soc. A 1971, 11, 1811.
(10) Bart, S. C.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc., published
online Oct 5, http://dx.doi.org/10.1021/ja046753t.
(11) For other examples of low-coordinate, iron(II) alkyls see (a) Kisko,
J. L.; Hascall, T.; Parkin, G. J. Am. Chem. Soc. 1998, 120, 10561. (b)
Fryzuk, M. D.; Leznoff, D. B.; Ma, E. S. F.; Rettig, S. J.; Young, V.
G., Jr. Organometallics 1998, 17, 2313. (c) Smith, J. M.; Lachicotte,
R. J.; Holland, P. L. Organometallics 2002, 21, 4808. (d) Sciarone,
T. J. J.; Meetsma, A.; Hessen, B.; Teuben, J. H. Chem. Commun. 2002,
1580. (e) Bart, S. C.; Hawrelak, E. J.; Schmisseur, A. K.; Lobkovsky,
E.; Chirik, P. J. Organometallics 2004, 23, 237.
(8) Schroeder, M. A.; Wrighton, M. S. J. Am. Chem. Soc. 1976, 98, 551.
(9) Betley, T. A.; Peters, J. C. Inorg. Chem. 2003, 42, 5074.
(12) (a) Sur, S. K. J. Magn. Reson. 1989, 82, 169. (b) Evans, D. F. J.
Chem. Soc. 1959, 2003.
Inorganic Chemistry, Vol. 43, No. 23, 2004 7475

Daida and Peters
Scheme 1
recorded for each of these complexes in benzene-d6 indicates
four unpaired electrons (S ) 2) (µeff ) 5.48 µB (2); µeff
)
5.04 µB (3); µeff ) 4.86 µB (4)). The precursor complex 1
and the recently reported amide [PhBPiPr3]Fe(NPh2) are
likewise S ) 2 spin systems.9,13 Worth noting is that access
to tetrahedral iron alkyl derivatives supported by the [PhBP3]
ligand (i.e., [PhBP3]Fe-R species) is synthetically trouble-
some because of problematic side-reactions.
The cyclic voltammetry of 2-4 was examined in THF
([TBA][PF6], 100 mV/s) and revealed in each case a
reversible wave at low potential that is assigned as an Fe-
(II/I) redox process. This event is centered at -2.3 V for 2,
-2.2 V for 3, and -2.4 V for 4 versus ferrocene as the
internal reference. Attempts to affect the reduction of 2-4
chemically, for example with Na/Hg, Na0, or KC8, led only
to poorly defined product mixtures. Though stable σ-hydro-
carbyl complexes of iron(I) are extremely uncommon,
Bianchini has shown that σ-alkynyl iron(I) derivatives can
be electrochemically generated and isolated when neutral
tetrapodal polyphosphine ligands are used to stabilize such
species (e.g., {P(CH2CH2PPh2)3}Fe(CtCR)).14 The Fe(I)
oxidation state is accessible at much higher potential for the
case of these neutral phosphine systems.
Pseudotetrahedral L3FeII-alkyl derivatives are structurally
uncommon, and the closest relative to the alkyl complexes
reported herein is Parkin’s substituted tris(pyrazolyl)borate
methyl complex [PhTptBu]FeII-CH3.11a The solid-state crystal
structures of 2 and 3 were determined, and their core P3Fe
atoms, along with their alkyl substituents, are shown in
Figure 1. The structures are unremarkable by comparison to
other [PhBPiPr3]Fe-X complexes that have been character-
ized previously.9,15 The Fe-P bond distances in 2 (avg )
Figure 1. Solid-state molecular structures of 2 (left) and 3 (right) showing
50% displacement ellipsoids for the P3Fe-R cores. Selected bond distances
(Å) and angles (deg) for 2: Fe-C28, 2.013(3); Fe-P1, 2.4343(9); Fe-P2,
2.4395(10); Fe-P3, 2.4327(10); C28-Fe-P1, 122.47(10); C28-Fe-P2,
124.65(12); C28-Fe-P3, 123.37(11); P1-Fe-P2, 92.74(3); P1-Fe-P3,
93.87(3); P2-Fe-P3, 90.82(3). For 3: Fe-C28, 2.0683(17); Fe-P1,
2.4633(5); Fe-P2, 2.4525(5); Fe-P3, 2.4570(5); C28-Fe-P1, 129.64(6);
C28-Fe-P2, 114.26(6); C28-Fe-P3, 125.20(6); P1-Fe-P2, 92.588(16);
P1-Fe-P3, 91.908(16); P2-Fe-P3, 94.104(16).
2.43 Å) are similar to those of [PhBPiPr3]FeCl (avg ) 2.42
Å). Complex 3 is somewhat less symmetric in nature than 2
because of the steric demands of its benzyl substituent, which
also seem to give rise to slightly longer Fe-P bond distances.
The Fe-C bond distance for 3 (2.0683(17) Å) is ca. 0.05 Å
longer than for 2 (2.013(3) Å), and for further comparison
the Fe-C bond distance in Parkin’s [PhTptBu]FeII-CH3
complex is 2.079(3) Å.
Generation of [PhBPiPr3]FeHx Derivatives. Alkyl com-
plexes 2-4 can each be hydrogenated over a period of hours
at room temperature to liberate RH in the presence of PMe3,
PEt3, or PMePh2 to afford a series of diamagnetic iron(IV)
trihydrides [PhBPiPr3]FeIV(H)3PR3 (PMe3 (5); PEt3 (6); or
PMePh2 (7)). This reactivity contrasts with that of the
previously described [PhBP3]Fe system ([PhBP3] ) [PhB(CH2-
PPh2)3]), which invariably generates the benzene reduction
product [PhBP3]Fe(η5-cyclohexadienyl) whenever an “FeII-
H” species is generated in benzene solution.3c As noted
elsewhere, the reactivity patterns of [PhBP3]Fe and [PhBPiPr3]-
Fe systems can be remarkably different.21
(13) Betley, T. A., Peters, J. C. J. Am. Chem. Soc. 2004, 108, 6252.
(14) Bianchini, C.; Laschi, F.; Masi, D.; Ottaviani, F. M.; Pastor, A.;
Peruzzini, M.; Zanello, P.; Zanobini, F. J. Am. Chem. Soc. 1993, 115,
2723.
(15) Turculet, L.; Feldman, J. D.; Tilley, T. D. Organometallics 2003, 22,
4672.
7476 Inorganic Chemistry, Vol. 43, No. 23, 2004

FeII/IV Redox and Hydrogenation of Olefins
Reliably established iron(IV) hydride species are rare but
do benefit from literature precedent. Thoroughly character-
ized examples of such species include [Cp*Fe(dppe)(H)2]+
(dppe ) bis(diphenylphosphino)ethane),16 (η6-arene)Fe(H)2-
(SiCl3)2,17 and [Fe(H)3(PEt3)4]+.18 Complexes 5-7 are unique
both with respect to their method of preparation (directly
from H2) and with respect to their reactivity properties (vide
infra). The closest structural relative, [Fe(H)3(PEt3)4]+,18 was
reported several years ago by Berke and co-workers and was
prepared by the protonation of Fe(H)2(PEt3)4. Clean isolation
of the dihydride precursor from the starting materials FeCl2,
NaBH4, and PEt3 requires a rather laborious synthetic
protocol. The ease with which presently described trihydrides
can be generated is therefore worth emphasizing.
Hydrogenation of 2 by D2 rather than H2 selectively
generated CH3D and the trideuteride [PhBPiPr3]Fe(D)3PMe3
(5-D3). The intense Fe-H vibration discernible for 5 (1907
cm-1) vanishes in the IR spectrum of 5-D3, and the expected
Fe-D vibration is ill-resolved because of overlap with other
low energy vibrations. That CH3D is the sole methane
byproduct of deuteration implies that methane dissociation
from the system is rapid relative to any reversible C-H(D)
activation processes that can be envisaged and moreover that
the H-atom delivered to the departing methyl group is derived
exclusively from hydrogen. An alternative ligand metalation
process that would transfer an H-atom, for example, from
an isopropyl substituent to the departing alkyl ligand can
thus be kinetically discounted. The hydrogenation of the
tetrahedral alkyl precursors in the absence of added phos-
phine is distinct and is discussed separately below.
The solution stability of the trihydrides 5-7 seems to
depend on the relative donor strength or the cone angle of
the capping phosphine ligand. For instance, complex 5 is
stable to vacuum over a period of hours and exhibits only
slow deuterium incorporation when stored under D2 at 23
°C. Complex 6 is reasonably stable but invariably shows a
trace amount of free H2 in its 1H NMR spectrum, even upon
dissolution of recrystallized solid samples in benzene-d6. This
observation is suggestive of the equilibrium shown in Scheme
1. In further support of this scenario, H2 loss occurs quite
readily for the case of the PMePh2 derivative 7. Satisfactory
NMR spectra of 7 are in fact only obtained when a hydrogen
atmosphere is preserved above the NMR sample solution.
Furthermore, an attempt to crystallize 7 by storing it as a
cold ethereal solution under N2 afforded crystals of the
paramagnetic monohydride complex [PhBPiPr3]Fe(H)(PMePh2)
(9). The independent preparation of 9 has proven difficult.
For example, the addition of LiEt3BH to [PhBPiPr3]FeCl (1)
in the presence of PMePh2 instead generates the iron(I)
reduction product [PhBPiPr3]Fe(PMePh2) (10) (43% crystal-
lized yield). Both 9 and 10 have been structurally character-
ized and can be viewed as distorted trigonal bipyramidal and
pseudotetrahedral complexes, respectively (see Supporting
Information). The hydride ligand occupies an axial position
trans to a [PhBPiPr3] donor arm in 9. The complex [PhBPiPr3]-
Fe(H)(PMe3) (11), which is structurally related to 9 (vide
infra), can nonetheless be readily prepared from the metath-
esis of KEt3BH with 1 in the presence of PMe3 (Scheme 1).
As might be expected, the addition of atmospheric hydrogen
to a benzene solution of 11 quantitatively generates trihydride
5 upon mixing.
Assignment of the hydrogenation products as bona fide
iron(IV) trihydrides is based upon comparative analysis of
their spectroscopic signatures with those of the related
species18 and also a high-resolution crystal structure obtained
1
for complex 5 (vide infra). For instance, the H NMR
spectrum of 5 features a single resonance in the hydride
region at δ -13.6 ppm that is split into a doublet of quartets
2
due to JP-H coupling to the PMe3 and [PhBPiPr3] ligands
(61.2 and 32.0 Hz, respectively). Consistent with this
formulation is a 31P{1H} NMR spectrum showing two
resonances, one for the [PhBPiPr3] phosphines (δ 70.8 ppm)
and one for the PMe3 ligand (δ 28.8 ppm). Both resonances
are well-resolved at all temperatures examined (-100 °C to
2
23 °C), though no JP-P coupling is discernible over this
temperature range. T1min measurements were performed for
the hydride resonances of 5 and 6 and provided values of
140 ms and 145 ms, respectively (-50 °C in THF-d8), fully
consistent with the terminal iron trihydride assignment.19 For
comparison, Berke’s [FeIV(H)3(PEt3)4]+ system provides a
T1min value of 177 ms, whereas the structurally related
dihydrogen adduct [FeII(H2)(H)(PMe3)4]+ exhibits much
faster relaxation due to a direct H-H interaction (T1min
)
13.5 ms).20 The fact that the hydride resonance and the 31P
NMR resonances for 5 and 6 change very little between room
temperature and -100 °C is consistent with their assign-
ments as trihydrides in solution. Upon the basis of Berke’s
results, we would expect some temperature dependence of
the spectrum were there an equilibrium distribution of
[PhBPiPr3]Fe(H)3PR3 and its nonclassical isomer [PhBPiPr3]-
Fe(H)(H2)PR3 in solution. We generated [PhBPiPr3]Fe(H)-
(D)2(PMe3) by the addition of D2 to an NMR tube sample
of [PhBPiPr3]Fe(H)(PMe3) (11, vide infra). No H-D coupling
was resolvable by 1H NMR spectroscopy (300 MHz),
presumably because of H-Fe-H angles that are close to
90°. Structural data obtained for 5 is presented below and
supports this point.
Conceptual removal of each of the three H atoms from 5
begets the iron(I) core structure [PhBPiPr3]Fe(PMe3) (8), a
derivative that can be independently prepared by Na/Hg
reduction of [PhBPiPr3]FeCl in the presence of PMe3 or
alternatively by the addition of excess PMe3 to {[PhBPiPr3]-
Fe}2(µ-N2).21 Complex 8 features a magnetic moment of
4.39 µB (Evans), implying three unpaired electrons and
(16) Hamon, P.; Toupet, L.; Hamon, J.-R.; Lapinte, C. Organometallics
1992, 11, 1429.
(17) Yao, Z.; Klabunde, K. J.; Asirvatham, A. S. Inorg. Chem. 1995, 34,
5289.
(18) Gusev, D. G.; Hu¨bener, R.; Burger, P.; Orama, O.; Berke, H. J. Am.
Chem. Soc. 1997, 119, 3716.
(19) Crabtree, R. H.; Lavin, M.; Bonneviot, L. J. Am. Chem. Soc. 1986,
108, 4032.
3
consistent with the S ) /2 ground state attributed to the
structurally related derivative [PhBP3]Fe(PPh3).22 Compound
(20) Jessop, P. G.; Morris, R. H. Coord. Chem. ReV. 1992, 21, 155.
(21) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2003, 125, 10782.
(22) Brown, S. D.; Peters, J. C. J. Am. Chem. Soc. 2003, 125, 322.
Inorganic Chemistry, Vol. 43, No. 23, 2004 7477

Daida and Peters
Figure 2. Solid-state molecular structures of (A) 8, (B) 11, and (C) 5 showing 50% displacement ellipsoids. Hydrogen atoms other than hydrides are
omitted for clarity. Selected bond distances (Å) and angles (deg) are shown.
8 is curiously resistant to hydrogen uptake (1 atm H2, 60
°C). An NMR sample of a THF-d8 solution of 8 under an
atmosphere of hydrogen fails to show any broadening of the
H2 resonance that might be expected were a reversible
equilibrium to be present between 8 and either an iron(I)
adduct complex “[PhBPiPr3]FeI(PMe3)(H2)” or an iron(III)
dihydride species “[PhBPiPr3]FeIII(PMe3)(H)2”. The direct
interconversion between 5 and 8 appears to be kinetically
inaccessible.23
Structural Data. We were fortunate that X-ray quality
crystals of each of the PMe3 adduct derivatives 5, and 8,
and 11 could be obtained for comparative study. High-
resolution X-ray diffraction analysis of a single crystal of
each species provided the three solid-state structures shown
in Figure 2. Important to note is that all three hydride
positions were located in the difference Fourier map and
refined for complex 5, and a peak consistent with a hydride
ligand was also located for the complex 11, and its position
was refined. As can be gleaned from the figure, each complex
is monomeric and is characterized by one tripodal phosphine
ligand and a PMe3 ligand that is coordinated to the opposite
face. This situation lends approximate 3-fold symmetry to
each species. From the perspective of the P3Fe-P′ core
atoms, each structure can be qualitatively viewed as pseudo-
tetrahedral. The iron(I) complex [PhBPiPr3]Fe(PMe3) 8 is
typified by the longest Fe-P bond distances (Fe-Pavg ) 2.33
Å, Fe-PMe3 ) 2.32 Å) and is also the most symmetric of
the three species. The core atoms of the iron(II) monohydride
complex 11 are nearly isostructural with those of 8, and 11
accommodates its single hydride ligand at a site trans to one
of its [PhBPiPr3] donor arms. Thus, another limiting structural
description for the case of 11 is to regard it as trigonal
bipyramidal with three of the phosphine ligands (P2, P3, and
P4) puckered toward the site occupied by the sterically less
demanding hydride ligand. It is interesting to note that there
is a negligible, if any, discernible trans influence that results
from the hydride ligand trans to P1 in 11. This is also true
in the structure of 9 (see Supporting Information) and is in
contrast to the situation in 5. The addition of two more
hydride ligands at the sites opposite to the other two
[PhBPiPr3] donor arms provides the structure observed for
trihydride 5. Interestingly, the average of the Fe-P bond
distance of divalent 11 is shorter than that for monovalent
8, and the average of the Fe-P bond distances of tetravalent
5 is the shortest among the three complexes. Of particular
signficance is the apical Fe-P4 bond in 5, which is
appreciably shorter than the three other Fe-P interactions
of the [PhBPiPr3] ligand. This difference is consistent with
the presence of trans influencing hydrides for each Fe-
P([PhBPiPr3]) linkage, whereas no such influence is present
for the apical Fe-P4 linkage. It is apparent that in the present
system the structural impact of a strongly trans influencing
(23) Carreo´n-Macedo, J.-L.; Harvey, J. N. J. Am. Chem. Soc. 2004, 126,
5789.
7478 Inorganic Chemistry, Vol. 43, No. 23, 2004

FeII/IV Redox and Hydrogenation of Olefins
hydride ligand is more easily discernible for a closed-shell
species, where the metal-ligand bond distances are on
average shorter, than that for an expanded open-shell
complex. These various structural data collectively demon-
strate that the [PhBPiPr3]Fe-PMe3 platform can sample three
different iron oxidation states while exhibiting rather little
structural reorganization as one or three hydride ligands are
added to or removed from the system.
Trihydride 5 serves as a zwitterionic complement to the
cationic complex [FeH3(PEt3)4]+ mentioned above18 and as
such warrants a brief structural comparison. Crystallographic
analysis of [FeH3(PEt3)4]+ showed that its overall geometry
is quite similar to that of 5. The three P-Fe-P angles that
make up its P3Fe-P′ core (102.1, 101.5, 96.5°) are noticeably
larger than the three related angles of 5 that are constrained
by the tripodal borate auxiliary (92.7, 92.5, 92.1°). As a
consequence, the P′-Fe-P angles in 5 (122, 126, 123°) are
on average larger than those of [FeH3(PEt3)4]+ (113.8, 122.2.
117.4°). Finally, the apical Fe-P bond distance reported for
[FeH3(PEt3)4]+ is slightly shorter than the Fe-P distances
to its other three phosphine ligands, as it is for the case of
5.
Figure 3. 1H NMR data (300 MHz, C6D6) for the hydride region during
the reaction between 2 and 1 atm hydrogen. The time course shows the
hydride resonances assigned to the intermediate [PhBPiPr3]Fe(H)3 and its
predominant degradation byproduct 12 at (a) ) 15 min, (b) ) 30 min, and
(c) ) 180 min.
Hydrogenation of Alkyls 2-4 in the Absence of Added
Phosphine. The addition of atmospheric hydrogen to the
alkyl derivatives at room temperature in benzene also releases
alkane to generate reactive hydride species. For instance,
when an NMR tube sample of the paramagnetic methyl
complex 2 dissolved in benzene-d6 is charged with hydrogen,
CH4 is produced over a period of hours, and two diamagnetic
hydride reaction products can be observed at early stages in
the reaction. The same reaction profile occurs in tetrahy-
drofuran solution and also occurs when the alkyl complexes
3 and 4 are used in place of 2.
Scheme 2
The first diamagnetic hydride that is produced, which itself
decays to the second hydride species, features C3 symmetry,
a single hydride resonance (quartet, -12.4 ppm in C6D6,
2JP-H ) 7.3 Hz, see Figure 3) and a single resonance in its
31P{1H} NMR spectrum (77.0 ppm). This species is reason-
ably assigned as the trihydride [PhBPiPr3]Fe(H)3 by analogy
to our assignments of the isolable trihydrides 5-7. The
absence of a capping phosphine donor renders [PhBPiPr3]-
Fe(H)3 too reactive to be isolated. Its complete decay occurs
over a period of hours at room temperature. The presence
of an appreciable quantity of the secondary hydride byprod-
uct throughout the reaction course, in addition to the requisite
atmosphere of hydrogen that is necessary to observe [PhBPiPr3]-
Fe(H)3, prevents the acquisition of meaningful T1min data for
this species. The secondary hydride species, which is the
dominant degradation product of [PhBPiPr3]Fe(H)3, is also
unstable but decays only slowly when stored under a
hydrogen atmosphere. At room temperature, this second
species exhibits a doublet of triplets resonance (2JP-H ) 22.8
and 35.6 Hz) centered at -12.7 ppm in the hydride region
of its 1H NMR spectrum. Its 31P{1H} NMR spectrum
suggests that the C3 symmetry of the system has been lost,
and two resonances, one doublet at 104.7 ppm (2P) and one
triplet at 77.9 ppm (1P) are now observed. If one records a
31P NMR spectrum with selective decoupling of the aliphatic
protons, then a quintet of triplets signal is observed for the
2
resonance centered at 77.9 ppm (2JP-P ) 27.5 Hz; JH-P
)
35.6 Hz), and a doublet of quintets is observed for the
resonance at 104.7 ppm (2JP-P ) 27.5 Hz; 2JH-P ) 22.8 Hz).
The spectrum was satisfactorily simulated using gNMR
software.24 The nature of these two multiplets confirms the
presence of four equivalent H atoms, presumably hydrides,
coordinated to iron. Although more stable than [PhBPiPr3]-
Fe(H)3, this species is relatively unstable even under a
hydrogen atmosphere, and the gradual production of PiPr2-
Me becomes increasingly evident over a period of 24 h by
31P NMR spectroscopy. Attempts to isolate the complex are
thwarted by its apparent tendency to lose H2 and degrade
further. We resorted to a combination of spectroscopic NMR
data, collected under both H2 and D2, to firmly establish the
identity of the second hydride byproduct as the (phosphino)-
borane-supported species {PhB(CH2PiPr2)2}Fe(H)4(PiPr2Me)
(12) (Scheme 2). It is spectroscopically still ambiguous as
to whether this species is better formulated as a divalent
dihydride dihydrogen complex (i.e., {PhB(CH2PiPr2)2}Fe-
(H)2(H2)(PiPr2Me)). In addition to the 1H and 31P{1H} NMR
(24) gNMR, version 4.0.1; Cherwell Scientific Publishing: Oxford, U.K.,
1998.
Inorganic Chemistry, Vol. 43, No. 23, 2004 7479

Daida and Peters
Table 1. X-ray Diffraction Experimental Details for
data reported above, data in support of our assignment of
12 follows. 11B{1H} NMR analysis of the sample shows a
broad resonance centered at 77 ppm, which is consistent with
a three-coordinate borane unit. Tetrahedral borate centers in
(phosphino)borates give rise to signature 11B{1H} NMR
resonances upfield of 0 ppm, typically in the range between
-10 and -15 ppm,9,25 and the resonance at 77 ppm
unambiguously establishes that the borate unit has been
converted to a borane. The methylene resonance of the
borane ligand of 12 is centered at 1.63 ppm and integrates
to 4 H atoms, as does the hydride resonance centered at
[PhBPiPr3]Fe(PMe3) (8), [PhBPiPr3]Fe(H)(PMe3) (11), and
[PhBPiPr3]Fe(H)3(PMe3) (5)
8
11
5
chemical formula
fw
T (°C)
λ (Å)
a (Å)
C30H62BFeP4
613.34
-173
0.71073
9.7111(3)
21.5823(7)
17.0048(6)
90
93.855(2)
90
3555.9(2)
P2(1)/c
4
C30H63BFeP4
614.34
-173
0.71073
9.7306(5)
21.4172(10)
16.9869(8)
90
93.2110(10)
90
3534.5(3)
P2(1)/c
4
C30H65BFeP4
616.36
-173
0.71073
9.5930(5)
21.5351(13)
17.0091(10)
90
93.022(2)
90
3509.0(3)
P2(1)/c
4
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
space group
Z
i
-12.7 ppm. The Pr2PMe ligand resonances of 12 are well
1
resolved in the H and 13C{1H} spectra, as they are in the
31P{1H} spectrum (vide supra). When D2 is used in place of
H2, the methylene resonance of the bis(phosphino)borane
ligand vanishes (1H), implying complete deuterium incor-
poration into the methylene position. The hydride resonance
at -12.7 ppm also decays, though residual proton signal
persists even after a period hours. When D2 is used to
hydrogenate 2 in C6D6, both CH3D and CH4 are liberated,
and H2 and HD are clearly visible in the sealed NMR tube
sample. The formation of CH4, H2, and HD byproducts is
consistent with ligand activation processes being competitive
with deuteriolysis of the alkyl group, distinct from the
deuteriolysis of 2 in C6D6 in the presence of added phosphine,
which liberates only CH3D (vide supra). The rate of decay
of 2 is slightly attenuated under one atmosphere of D2 by
comparison to H2 (krel(H2/D2) ) 1.1).
D
calcd (g/cm3)
µ(mm-1
1.146
0.621
1.154
0.625
1.167
0.629
)
R1, wR2a (I > 2σ(I)) 0.0582, 0.0943 0.0444, 0.0762 0.0476, 0.0937
a R1 ) ∑||Fo| - |Fc||/∑|Fo|, wR2 ) {∑[w(Fo2 - Fc )2]/∑[w(Fo )2]}1/2
.
2
2
Table 2. Catalytic Olefin Hydrogenation Reactions with 2, 3, 5, and 6
entry precatalysta
substrate
H2 (atm) time (min)c TOFd
1
2
3
4
5
6
7
8
9
3
3
2
3
3
2
3
3
5b
6
styrene
styrene
styrene
ethylene
1-hexene
1-hexene
cyclooctene
2-pentyne
styrene
1
4
4
4
1
1
1
1
1
1
410
100
78
1.5
6.0
7.7
24.0
4.6
5.2
10.9
1.6
0.5
25
130
115
55
370
1260
1010
10
styrene
0.6
Hydrogenation Studies. When trihydride 5 was subjected
to an atmosphere of ethylene, a clean transformation ensued
and the new iron(II) alkyl, [PhBPiPr3]Fe(CH2CH3) (13), was
produced along with an equivalent of ethane and free PMe3
(eq 2). The ethyl complex could be isolated from the reaction
solution as a crystalline orange solid in 63% yield. Its
analytical data are analogous to those of complexes 2-4.
The production of a stoichiometric equivalent of ethane
suggests that in situ hydrogenation of an Fe-Et bond,
presumably formed via insertion of ethylene into an Fe-H
linkage, is relatively facile. This observation intimated that
a catalytic hydrogenation process might be realized. Com-
plexes 2, 3, 5, and 6 were therefore each examined for their
potential to hydrogenate styrene in benzene solution. Expo-
sure of 5 to 10 equiv of styrene under 1 atm of H2 produced
ethylbenzene quantitatively. The reaction was complete
within 21 h at 50 °C. The more labile triethylphosphine
derivative 6 completed the hydrogenation within 17 h at 23
°C. The most active precatalysts are the Fe(II) alkyls
themselves. For example, 10 equiv of styrene were hydro-
genated using 3 as a precatalytst within 7 h at 23 °C. When
the reaction was instead carried out under 4 atm of H2, 10
equiv of styrene were hydrogenated by 3 within 2 h. Other
olefins reacted similarly. For instance, using precatalyst 3
dissolved in C6D6 (10 mol %), the conversion of 1-hexene
and cyclooctene to their corresponding alkanes under 1 atm
of H2 was complete within 3 and 1 h, respectively, at 23 °C.
Ethylene reduction was complete within 30 min under
a Conditions: 10 mol % precatalyst, [olefin] ) 0.2 M in C6D6, 23 °C.
b Reaction conducted at 50 °C. c Time to >95% conversion to the
corresponding alkane as ascertained by 1H NMR spectroscopy. d TOF
expressed as moles of substrate per moles of iron catalyst per hour,
calculated on the basis of the times given.
analogous conditions using 4 atm of H2. For the case of
1-hexene, initial isomerization to internal olefins was ob-
1
served (as evident by H NMR) prior to the quantitative
production of n-hexane, a result that suggests â-hydride
elimination processes compete kinetically with the Fe-alkyl
hydrogenolysis step. The methyl complex 2 proved to be a
slightly more efficient precatalyst for the hydrogenation of
styrene and 1-hexene than the benzyl complex 3, perhaps
because 2 is hydrogenated more rapidly than 3, affording
more efficient entry into the catalytic cycle. For the three
alkyl complexes described thus far, the relative rates of
reaction with H2 follow the trend 2 > 3 > 4. These hydro-
genation data are summarized in Table 2.
Certain alkyne substrates can be effectively hydrogenated
in the presence of precatalyst 3. For example, 2-pentyne is
quantitatively converted to pentane, with a small amount of
1
the intermediate cis-2-pentene observed by H NMR spec-
troscopy during the course of the reaction. Certain alkynes
proved problematic. For instance, an attempt to hydrogenate
(25) Thomas, J. C.; Peters, J. C. Inorg. Chem. 2003, 42, 5055.
7480 Inorganic Chemistry, Vol. 43, No. 23, 2004

FeII/IV Redox and Hydrogenation of Olefins
Scheme 3
phenylacetylene under standard conditions resulted in the
formation of only small amounts of styrene and ethylbenzene.
Products with molecular weights with suggestive of reductive
dimerization and trimerization, as well as cyclotrimerization,
were also observed by GC/MS. A substantial amount of an
intractable bright red/orange material was also produced,
which was likely the product of higher degrees of oligo-
merization or polymerization. Attempts to carry out the
hydrogenation of acetylene were similarly problematic, and
a large amount of a dark, hydrocarbon insoluble material
was rapidly produced which once again indicated oligomer-
ization/polymerization processes. These insoluble materials
have not been further characterized.
An obvious issue concerns whether these hydrogenation
reactions arise from homogeneous solution processes. While
it is inevitably difficult to rule out a catalytic role for
heterogeneous components, we note that the addition of
mercury to styrene hydrogenation reactions did not impact
the overall reaction profile, or the approximate reaction rate.26
Moreover, light does not appear to play a role. Parallel NMR
tube hydrogenations of styrene facilitated by 3 in the presence
and absence of ambient room light proceeded at equal rates.
As an additional control experiment, we examined the
hydrogenation of norbornylene by D2 in benzene-d6. The
exclusive product detected by 1H NMR spectroscopy proved
to be exo,exo-2,3-d2-norbornane, consistent with cis addition
of D2.27
Mechanistic Considerations. Although numerous mecha-
nistic questions regarding the hydrogenation reactions de-
scribed herein remain, our collective observations allow us
to sketch a plausible mechanistic scenario at this stage. Such
an outline is shown in Scheme 3.
Iron precursors of different ground spin-states (S ) 0
[PhBPiPr3]Fe(H)3(PR3), S ) 1 [PhBPiPr3]Fe(H)(PR3), and S
) 2 [PhBPiPr3]Fe(R)) can each serve as a precatalyst to
the hydrogenation manifold. From an S ) 0 trihydride
derivative, dissociation of PR3 exposes a reactive “FeH3”
source that would most likely exist as either a classical
trihydride [PhBPiPr3]FeIV(H)3 or a dihydrogen hydride com-
plex “[PhBPiPr3]FeII(H)(H2)”. Both species, and their inter-
conversion, have ample literature precedent.6a,18 Upon the
basis of the NMR data presented above, we can assign the
trihydride [PhBPiPr3]Fe(H)3 as a detectable solution species.
Recall that the addition of hydrogen to 2 in the absence of
added phosphine gives rise to two diamagnetic hydride
species, one assigned as the transient [PhBPiPr3]Fe(H)3, and
the second assigned as its degradation product 12. We have
no spectroscopic evidence for the species “[PhBPiPr3]FeII-
(H)(H2)” but cannot discount it as a reactive, equilibrium
isomer of [PhBPiPr3]Fe(H)3. Loss of H2 and concomitant
olefin binding from [PhBPiPr3]Fe(H)3 (or [PhBPiPr3]FeII(H)-
(H2)) can be envisioned to occur associatively or dissocia-
tively via the four-coordinate iron(II) hydride [PhBPiPr3]Fe(H)
as shown in Scheme 3. Upon the basis of our unsuccessful
efforts to generate and directly detect four-coordinate
[PhBPiPr3]Fe(H) (and [PhBP3]Fe(H)3c), we expect it to be
far more reactive than its alkyl analogues if formed. Regard-
less, olefin uptake to generate [PhBPiPr3]Fe(H)(olefin) would
set the stage for a subsequent insertion step to generate a
relatively stable S ) 2 iron alkyl product. Indirect evidence
for such a sequence is provided by the addition of ethylene
to trihydride 5, which produces the ethyl complex 13 and
PMe3 and ethane as stoichiometric byproducts. The iron alkyl
then reacts with hydrogen to release alkane, a step that can
be studied explicitly. This step regenerates the iron hydride
source, perhaps initially as [PhBPiPr3]Fe(H) which rapidly
converts to [PhBPiPr3]Fe(H)3 under hydrogen, or [PhBPiPr3]-
Fe(H)3(PR3) in the presence of added phosphine.
For the phosphine-capped trihydride precatalysts [PhBPiPr3]-
Fe(H)3(PR3), a distribution of five-coordinate Fe(II) species
(including [PhBPiPr3]Fe(H)(H2), [PhBPiPr3]Fe(H)(olefin), and
[PhBPiPr3]Fe(H)(PR3)) are likely to be present in solution
throughout the course of the reaction, and it is interesting to
consider what their respective spin states might be. Whereas
we have definitively identified [PhBPiPr3]Fe(H)(PMe3) (11)
as a ground-state triplet species, we have reported elsewhere
that the complex [PhBPiPr3]FeCl(CO) exhibits a singlet
ground state.9 It seems likely that [PhBPiPr3]Fe(H)(H2) and
[PhBPiPr3]Fe(H)(olefin), which both feature ligands that are
relatively weak sigma donors and also π acceptors, would
feature low spin (S ) 0) ground states.
(26) Whitesides, G. M.; Hackett, M.; Brainard, R. L.; Lavalleye, J.-P. P.
M.; Sowinski, A. F.; Izumi, A. N.; Moore, S. S.; Brown, D. W.; Staudt,
E. M. Organometallics 1985, 4, 1819.
Several salient observations concerning the styrene hy-
drogenation reactions allow us to speculate as to the catalyst
(27) Marchand, A. P.; Marchand, N. W. Tetrahedron Lett. 1971, 18, 621.
Inorganic Chemistry, Vol. 43, No. 23, 2004 7481

Daida and Peters
rate dependence of styrene hydrogenation on hydrogen
pressure and initial concentration of precatalyst 3 was
investigated. These studies revealed a first-order rate depen-
dence on both hydrogen pressure and total iron concentration
(Figures 4b and 4c), consistent with a rate-determining step
involving the reaction of H2 with an iron alkyl species. The
nonzero intercept observed for the rate dependence on iron
concentration (Figure 4c) is likely due to the competitive
decomposition of the catalytically active iron hydride species
to give 12 and possibly other degradation products. In accord,
when an initial precatalyst concentration of 0.005 M 3 was
used under standard conditions (0.2 M styrene, C6D6, 1 atm
H2), only three turnovers were measured, and decomposition
i
to give some Pr2PMe was clearly evident by 31P NMR
spectroscopy.
When trihydride 6 is used as a precatalyst for the
hydrogenation of styrene, it is clearly present in significant
concentration throughout the course of the reaction (1H NMR,
31P NMR). Additionally, monitoring the rate of styrene
consumption in the presence of precatalyst 6 under a constant
pressure of H2 (1 atm) results in a reaction profile that
displays a first-order dependence on styrene concentration
(Figure 4a). In this latter case, the resting state for the catalyst
system is likely trihydride 6 itself, and exchange of the
coordinated phosphine for styrene is the most plausible
turnover-limiting step.
Because H2 coordination to generate [PhBPiPr3]FeII(R)(H2)
is a requisite step that precedes release of alkane from
[PhBPiPr3]Fe(R), it needs to be underscored that the alkyls
2-4 are resistant to the uptake of other two-electron donor
ligands such as ethylene and PMe3. These latter ligands may,
however, exhibit reversible association with the iron center.
That this is likely for the specific case of PMe3 is evident
from the observation of appreciable line-broadening of
the 31P NMR resonance for free PMe3 added to a solution
of 3 (width at half-maximum ≈ 100 Hz). H2 appears to be
unique as a ligand within this system, however, because of
its propensity to react irreversibly to release RH, driving
the system forward. The fact that iron(IV) is so readily
accessible within the [PhBPiPr3]Fe system suggests that
a sequence to consider would involve a stepwise oxidative
addition/reductive elimination process (e.g., [PhBPiPr3]-
Fe-R + H2 f [PhBPiPr3]FeII(R)(H2) f [PhBPiPr3]FeIV(R)-
(H)2 f [PhBPiPr3]FeII(RH)(H) f [PhBPiPr3]FeII(H) + RH).
The small, positive primary kinetic isotope effect observed
for the reaction of alkyl 2 with H2 (vida supra) is consistent
with oxidative addition of H2 to 2.28 An alternative sequence
that circumvents redox chemistry is one that occurs via a
σ-bond methathesis step to release alkane and generate
[PhBPiPr3]Fe(H) T [PhBPiPr3]Fe(H)3. We favor the FeII/IV
redox scenario because such redox processes appear to be
particularly prevalent within the [PhBPiPr3]Fe manifold. The
addition of hydrogen to 11 to generate 5 ([PhBPiPr3]FeII(H)-
(PMe3) + H2 f [PhBPiPr3]FeIV(H)3(PMe3)) underscores this
Figure 4. (a) Hydrogenation of styrene (0.20 M) at 23 °C in C6D6; 9 )
10 mol % 3, 4 atm H2; [ ) 10 mol % 3, 1 atm H2; O ) 10 mol % 6, 1
atm H2. (b) Hydrogenation of styrene (0.20 M) at 23 °C in C6D6 with
precatalyst 3 as a function of H2 pressure. (c) Hydrogenation of styrene
(0.20 M) at 23 °C in C6D6 under 1 atm H2 as a function of precatalyst 3
concentration.
resting state and the turnover-limiting step when either the
alkyl complex 3 or the trihydride species 6 is used as the
precatalyst. During the hydrogenation of styrene by 3, a
1
single intermediate can be observed by H NMR spectros-
copy at room temperature. This intermediate species exhibits
paramagnetically shifted [PhBPiPr3] resonances with chemi-
cal shifts quite similar to those of the paramagnetic iron(II)
alkyl complexes we have already described and is assigned
as the complex [PhBPiPr3]Fe(CH2CH2Ph). This species can
be independently generated and observed by 1H NMR
spectroscopy upon the addition of an equivalent of styrene
to trihydride 6. Also, we have monitored the rate of styrene
hydrogenation by 3 under a constant pressure of H2 (1 atm)
and note that the rate of styrene consumption does not change
during the course of the reaction, implying a zero-order
dependence on styrene concentration (Figure 4a). This
observation is consistent with a scenario in which an iron
alkyl species, in the present case [PhBPiPr3]Fe(CH2CH2Ph),
is the resting state of the catalytic cycle. The fact that we
can observe [PhBPiPr3]Fe(CH2CH2Ph) as an intermediate
suggests that the reaction between H2 and an iron alkyl is
rate-determining. To further investigate this hypothesis, the
(28) (a) Chock, P. W.; Halpern, J. J. Am. Chem. Soc. 1966, 88, 3511. (b)
Zhou, P.; Vitale, A. A.; San Filippo, Jr., J.; Saunders: Jr., W. H. J.
Am. Chem. Soc. 1985, 107, 8049.
7482 Inorganic Chemistry, Vol. 43, No. 23, 2004

FeII/IV Redox and Hydrogenation of Olefins
point, as does the N-atom transfer reaction from [PhBPiPr3]-
FeII(dbabh) to generate [PhBPiPr3]FeIVtN and anthracene.13
Moreover, spectroscopically observable [PhBPiPr3]FeIV(H)3
is structurally and electronically quite similar to the alkyl
dihydride intermediate that needs to be invoked (i.e.,
[PhBPiPr3]FeIV(R)(H)2; see Scheme 3) via the oxidative
addition path.
The FeII/IV hydrogenation scheme proposed is intriguing
in that it emphasizes a catalytically relevant role for
concerted, multielectron FeII/IV redox couples in the context
of homogeneous catalysis. Efforts to probe this mechanistic
scenario in greater detail are underway.
of toluene, filtered, and stored at -25 °C, giving 2 as yellow crystals
(0.0559 g, 77%). H NMR (C6D6, 300 MHz): δ 47.44 (s), 22.06
(s), 20.37 (s), -3.62 (br, s), -23.27 (br, s), -46.81 (br, s). UV-
vis (C6H6) λmax, nm (ꢀ, M-1 cm-1): 369 (1400). Evans Method
(C6D6): 5.48 µB. Anal. Calcd for C28H56BFeP3: C, 60.89; H, 10.22.
Found: C, 60.50; H, 10.41.
1
Synthesis of [PhBPiPr3]Fe-CH2Ph, 3. A 390 µL (0.390 mmol)
aliquot of 1.0 M benzylmagnesiumchloride in diethyl ether was
added to a frozen THF solution (5 mL) of 1 (0.2010 g, 0.351 mmol).
The solution was allowed to thaw to room temperature and stirred
for 1 h, giving a dark-red solution. The THF was removed in vacuo,
and the resulting red solid was extracted with petroleum ether (60
mL) and filtered through Celite. Removal of the petroleum ether
in vacuo gave a red solid that was dissolved in a minimal amount
of diethyl ether, filtered, and stored at -25 °C, giving 3 as dark-
red crystals (0.1832 g, 83%). 1H NMR (C6D6, 300 MHz): δ 104.52
(s), 54.91 (s), 42.92 (s), 20.29 (s), 18.93 (s), -2.26 (br, s), -17.89
(br, s), -37.60 (br, s). UV-vis (C6H6) λmax, nm (ꢀ, M-1 cm-1):
340 (3680), 470 (2340). Evans Method (C6D6): 5.04 µB. Anal.
Calcd for C34H60BFeP3: C, 64.98; H, 9.62. Found: C, 64.88; H,
9.90.
Experimental Section
All manipulations were carried out using standard Schlenk or
glovebox techniques under a dinitrogen atmosphere. Unless oth-
erwise noted, solvents were deoxygenated and dried by thorough
sparging with N2 gas followed by passage through an activated
alumina column. Nonhalogenated solvents were tested with a
standard purple solution of sodium benzophenone ketyl in tetrahy-
drofuran in order to confirm effective oxygen and moisture removal.
All reagents were purchased from commercial vendors and used
without further purification unless otherwise stated. [PhBPiPr3]FeCl
(1),9 {[PhBPiPr3]Fe}2(µ-N2),21 and neopentyllithium29 were prepared
according to literature procedures. Elemental analyses were per-
formed by Desert Analytics, Tucson, AZ. Deuterated benzene was
purchased from Cambridge Isotope Laboratories, Inc. and degassed
and dried over activated 3-Å molecular sieves prior to use.
Deuterated tetrahydrofuran was purchased from Cambridge Isotope
Laboratories, Inc. and dried over activated alumina prior to use. A
Varian Mercury-300 NMR spectrometer was used to record 1H and
31P NMR spectra at ambient temperature. Variable temperature 1H
T1 measurements were performed on a Varian Inova-500 NMR
Synthesis of [PhBPiPr3]Fe-CH2CMe3, 4. Solid neopentyllithium
(0.0203 g, 0.260 mmol) was added to a frozen THF solution (3
mL) of 1 (0.0950 g, 0.167 mmol). The solution was allowed to
thaw to room temperature and stirred for 30 min, giving an orange/
brown solution. The THF was removed in vacuo, and the resulting
solid was extracted with petroleum ether, filtered, and dried in
vacuo. The resulting orange solid was dissolved in a minimal
amount of pentane, filtered, and stored at -25 °C, giving 4 as
1
orange crystals (0.0556 g, 55%) H NMR (C6D6, 300 MHz): δ
85.37 (br, s), 44.13 (s), 21.07 (br, s), 19.47 (br, s), -5.98 (br, s),
-19.09 (br, s), -39.81 (br, s). UV-vis (C6H6) λmax, nm (ꢀ, M-1
cm-1): 375 (1350), ∼400 (shoulder). Evans Method (C6D6): 4.86
µB. Anal. Calcd for C32H64BFeP3: C, 63.17; H, 10.60. Found: C,
62.94; H, 10.31.
1
spectrometer using the inversion recovery method. H chemical
Synthesis of [PhBPiPr3]Fe(H)3(PMe3), 5. A 32 µL (0.309 mmol)
aliquot of PMe3 was added to 0.1129 g (0.204 mmol) of 2 dissolved
in 5 mL of THF in a 25 mL reaction bomb. The solution was
degassed by three freeze/pump/thaw cycles, and the bomb was
pressurized with 1 atm of H2. The solution was stirred vigorously
for 24 h and then evaporated to dryness in vacuo. The resulting
yellow solid was dissolved in a minimal amount of diethyl ether,
filtered, and stored at -25 °C under an N2 atmosphere, giving 5 as
yellow crystals (0.0873 g, 69%). 1H{31P} NMR (C6D6, 300 MHz):
δ 8.06 (m, 2H), 7.61 (t, 2H), 7.34 (t, 1H), 1.67 (septet, 6H), 1.44
(s, 9H), 1.18 (d, 18H), 1.15 (d, 18H), 0.88 (br, 6H), -13.72 (s,
3H). 31P{1H} NMR (C6D6, 121.4 MHz): δ 70.81 (s, 3P), 28.75 (s,
1P). Anal. Calcd for C30H65BFeP4: C, 58.46; H, 10.63. Found: C,
58.45; H, 10.30. 5-D3 was prepared analogously using D2. The
IR spectrum (KBr, C6H6) of 5 showed an Fe-H stretching vibration
at 1907 cm-1 that was absent from the IR spectrum of 5-D3. The
isotopically shifted Fe-D stretch (calcd 1363 cm-1) could not be
observed because of the presence of other vibrational modes in
this region of the spectrum.
shifts were referenced to residual solvent. 31P chemical shifts were
externally referenced to 85% H3PO4. 11B chemical shifts were
referenced to neat BF3‚OEt2. Broad (br) resonances in the 1H NMR
spectra typically refer to resonances greater than 50 Hz width at
half-maximum. UV-vis measurements were taken on a Cary 50
scanning spectrometer using a quartz cell with a Teflon cap. IR
measurements were obtained with a KBr solution cell using a Bio-
Rad Excalibur FTS 3000 spectrometer controlled by Bio-Rad Merlin
Software (v. 2.97) set at 4 cm-1 resolution. X-ray diffraction studies
were carried out in the Beckman Institute Crystallographic Facility
on a Bruker Smart 1000 CCD diffractometer.
Electrochemistry. Electrochemical measurements were carried
out in a glovebox under a dinitrogen atmosphere in a one-
compartment cell using a BAS model 100/W electrochemical
analyzer. A glassy carbon electrode and platinum wire were used
as the working and auxiliary electrodes, respectively. The reference
electrode was Ag/AgNO3 in THF. Solutions (THF) of electrolyte
(0.4 M tetra-n-butylammonium hexafluorophosphate) and analyte
(2 mM) were also prepared in a glovebox.
Synthesis of [PhBPiPr3]Fe(H)3(PEt3), 6. A 30 µL (0.203 mmol)
aliquot of PEt3 was added to 0.0950 g (0.172 mmol) of 2 dissolved
in 3 mL of THF in a 25 mL reaction bomb. The solution was
degassed by three feeze/pump/thaw cycles, and the bomb was
pressurized with 1 atm of H2. The solution was stirred vigorously
for 12 h and then evaporated to dryness in vacuo. The resulting
beige solid was washed with 2 mL of petroleum ether. The solid
was dissolved in a minimal amount of diethyl ether and stored at
-25 °C under an N2 atmosphere, giving 6 as yellow crystals (0.045
Synthesis of [PhBPiPr3]Fe-Me, 2. A 95 µL aliquot of 1.6 M
methyllitium in diethyl ether (0.152 mmol) was added to a frozen
THF solution (2 mL) of 1 (0.0751 g, 0.131 mmol). The solution
was allowed to thaw to room temperature and stirred for 1 h.
Removal of THF in vacuo gave a yellow solid that was extracted
with benzene and filtered. The benzene extract was dried in vacuo,
and the resulting yellow solid was dissolved in a minimal amount
(29) Schrock, R. R.; Fellmann, J. D. J. Am. Chem. Soc. 1978, 100, 3359.
Inorganic Chemistry, Vol. 43, No. 23, 2004 7483

Daida and Peters
1
g, 40%). H{31P} NMR (C6D6, 300 MHz): δ 8.07 (m, 2H), 7.61
and stored at -25 °C, giving 10 as yellow/orange crystals (0.1120
g, 43%). 1H NMR (THF-d8, 300 MHz): δ 86.83 (br, s), 77.84 (br,
s), 14.74 (s), 10.26 (s), 9.25 (br, s), 7.87 (s), 6.75 (s), -1.04 (s),
-2.09 (br, s), -8.92 (br, s). UV-vis (THF) λmax, nm (ꢀ, M-1 cm-1):
915 (1015). Evans Method (THF-d8): 4.36 µB. Anal. Calcd for
C40H66BFeP4: C, 65.14; H, 9.02. Found: C, 65.12; H, 9.07.
(t, 2H), 7.34 (t, 1H), 1.80 (septet, 6H), 1.70 (q, 6H), 1.23 (d, 18H),
1.18 (d, 18H), 0.97 (t, 9H), 0.92 (br, 6H), -14.08 (s, 3H). 31P{1H}
NMR (C6D6, 121.4 MHz): δ 68.34 (s, 3P), 66.71 (s, 1P). Anal.
Calcd for C33H71BFeP4: C, 60.19; H, 10.87. Found: C, 60.36; H,
10.51.
Generation of [PhBPiPr3]Fe(H)3(PMePh2), 7. A 30 µL (0.161
mmol) aliquot of PMePh2 was added to 0.0102 g (0.016 mmol) of
3 dissolved in 0.5 mL of C6D6 in a J. Young tube. The solution
was degassed by three freeze/pump/thaw cycles, and the tube was
filled with 1 atm of H2. After 24 h, only 7, toluene from the
hydrogenation of 3, and unreacted PMePh2 were observed by NMR.
1H{31P} NMR (C6D6, 300 MHz): δ 8.06 (m, 2H), 7.62 (t, 2H),
7.59 (m, 4H), 7.29 (t, 1H), 7.08 (m, 6H), 2.33 (s, 3H), 1.72 (septet,
6H), 1.15 (d, 18H), 1.06 (d, 18H), 0.92 (m, 6H), -13.16 (s, 3H).
31P{1H} NMR (C6D6, 121.4 MHz): δ 68.52 (s, 3P), 60.08 (s, 1P).
Synthesis of [PhBPiPr3]Fe(PMe3), 8. A 50 µL (0.483 mmol)
aliquot of PMe3 was added to 0.0512 g (0.046 mmol) of {[PhBPiPr3]-
Fe}2(µ-N2) dissolved in 5 mL of THF. After being stirred for 3
days, a nearly colorless solution was obtained. The THF was
removed in vacuo, and the resulting solid was triturated three times
with petroleum ether, giving 8 as an analytically pure off-white
solid (0.0496 g, 87%). Alternatively, a 0.5 wt % Na/Hg amalgam
(0.0053 g, 0.230 mmol of sodium dissolved in 1.0901 g mercury)
was stirred in THF (5 mL) with 80 µL (0.773 mmol) of PMe3 for
several minutes. A solution of 1 (0.1093 g, 0.191 mmol) in THF
(1 mL) was added to the amalgam at room temperature, and the
solution was vigorously stirred for 1.5 h. The resulting nearly
colorless solution was filtered and evaporated to dryness. The
resulting pale solid was extracted with diethyl ether, filtered, and
evaporated to dryness. Storing a concentrated ethereal solution of
the resulting pale solid at -25 °C gave 8 as pale-green blades
Synthesis of [PhBPiPr3]Fe(H)(PMe3), 11. A 183 µL (0.183
mmol) aliquot of 1.0 M KHBEt3 in THF was added to a frozen
THF (5 mL) solution of 1 (0.1047 g, 0.183 mmol) and PMe3 (38
µL, 0.367 mmol). The solution was allowed to thaw to room
temperature and stirred for 15 min. The resulting cloudy yellow
solution was evaporated to dryness in vacuo. Extraction with diethyl
ether, filtration, and evaporation of the ether in vacuo gave a yellow
solid, which was washed with petroleum ether, dissolved in a
minimal amount of diethyl ether, and stored at -25 °C, giving 11
as yellow crystals (0.0579 g, 52%). 1H NMR (C6D6, 300 MHz): δ
62.93 (br, s), 52.15 (br, s), 9.95 (s), 8.16 (s), 7.51 (s), 7.49 (s),
4.76 (br, s), -2.34 (br, s). UV-vis (C6H6) λmax, nm (ꢀ, M-1 cm-1):
420 (1780). Evans Method (C6D6): 2.91 µB. Anal. Calcd for
C30H63BFeP4: C, 58.65; H, 10.34. Found: C, 58.49; H, 9.97.
Generation of {PhB(CH2PiPr2)2}Fe(H)4(PiPr2Me), 12. A 0.0150
g (0.027 mmol) amount of 2 dissolved in 0.7 mL of C6D6
was placed in a J. Young tube. The sample was degassed by three
freeze/pump/thaw cycles, and the tube was filled with 1 atm H2.
After 3 h, all 2 had been consumed and 12 was observable by NMR
1
spectroscopy. H{31P} NMR (C6D6, 300 MHz): δ 7.93 (m, 2H,
BPh), 7.31 (m, 3H, BPh), 1.86 (septet, 4H, PhB(CH2P(CHMe2)2)2),
1.72 (septet, 2H, PMe(CHMe2)2), 1.63 (s, 4H, PhB(CH2PiPr2)2),
1.23 (d, 6H, P(CH(CH3)2)2), 1.21 (d, 12H, P(CH(CH3)2)2), 1.10
(d, 6H, P(CH(CH3)2)2), 1.08 (s, 3H, PiPr2CH3), 1.02 (d, 12H, P(CH-
(CH3)2)2), -12.70 (s, 4H, Fe(H)4). 13C{1H} NMR (C6D6, 75.4
Hz): δ 143.01 (br, BPh), 134.64 (BPh), 132.63 (BPh), 128.24
1
(0.0917 g, 78%). H NMR (THF-d8, 300 MHz): δ 79.52 (br, s),
(BPh), 32.45 (d, JP-C ) 27.5 Hz, P(CHMe2)2), 31.73 (d, JP-C
)
11.45 (br, s), 8.21 (s), 7.09 (s), -4.03 (br, s). UV-vis (THF) λmax
,
33.3 Hz, P(CHMe2)2), 31.50 (d, JP-C ) 21.8 Hz, P(CHMe2)2), 25.20
nm (ꢀ, M-1 cm-1): 950 (1050). Evans Method (THF-d8): 4.39
µB. Anal. Calcd for C30H62BFeP4: C, 58.75; H, 10.19. Found: C,
59.06; H, 9.82.
2
(br, PhB(CH2PiPr2)2), 19.89 (d, JP-C ) 2.3 Hz, P(CH(CH3)2)2),
19.65 (P(CH(CH3)2)2), 19.07 (P(CH(CH3)2)2), 18.90 (P(CH-
(CH3)2)2), 15.89 (PiPr2CH3). 31P{1H} NMR (C6D6, 121.4 MHz):
δ 104.71 (d, 2P), 78.01 (t, 1P). 11B{1H} NMR (C6D6, 160.3
Synthesis of [PhBPiPr3]Fe(H)(PMePh2), 9. A 160 µL (0.860
mmol) aliquot of PMePh2 was added to 0.0945 g (0.165 mmol) of
2 dissolved in 5 mL of THF in a 25 mL reaction bomb. The solution
was degassed by three freeze/pump/thaw cycles, and the vessel was
filled with 1 atm of H2 and allowed to stir for 24 h. The solution
was evaporated to dryness in vacuo, and the resulting solids were
washed with 5 mL of petroleum ether, dried, and stored at -25 °C
as a diethyl ether solution, giving yellow crystals. An X-ray
diffraction experiment performed on a selected crystal showed it
1
MHz): δ 77 (br). Assignments of H and 13C NMR resonances
are based on heteronuclear multiple quantum coherence (HMQC)
and distortion enhancement by polatization transfer (DEPT) analy-
ses.
Synthesis of [PhBPiPr3]Fe(CH2CH3), 13. A 30 µL (0.290 mmol)
aliquot of PMe3 was added to 0.1107 g (0.200 mmol) of 2 dissolved
in 4 mL of THF in a 25 mL reaction bomb. The solution was
degassed by three freeze/pump/thaw cycles, filled with 1 atm of
H2, and allowed to stir for 12 h. The resulting yellow solution of
5 was evaporated to dryness. The resulting yellow powder was
dissolved in 7 mL of diethyl ether, filtered, and evaporated to
dryness, giving 5 as a yellow solid, which was redissolved in 5
mL of THF and returned to the 25 mL reaction bomb. The solution
was degassed by three freeze/pump/thaw cycles, and the vessel was
filled with 1 atm of ethylene and allowed to stir overnight. The
resulting orange solution was evaporated to dryness, giving an
orange solid that was dissolved in diethyl ether and filtered.
Removal of the diethyl ether in vacuo gave an orange solid that
was dissolved in a minimal amount of diethyl ether and stored at
1
to be the iron(II) monohydride species, 9. H NMR (C6D6, 300
MHz) and 31P NMR (C6D6, 121.4 MHz) revealed that the crystals
were a mixture of the trihydride species, 7, and the paramagnetic
species, 9, based on resonances observed for a paramagnetic species
that were paramagnetically shifted in a manner similar to those of
the analogous iron(II) monohydride species, 11.
Synthesis of [PhBPiPr3]Fe(PMePh2), 10. A 365 µL (0.365
mmol) aliquot of 1.0 M LiHBEt3 in THF was added to a frozen
THF solution (5 mL) of 1 (0.2007 g, 0.350 mmol) and PMePh2
(130 µL, 0.699 mmol). The solution was allowed to thaw and stirred
for 20 min. The THF was removed in vacuo, giving an orange/
brown oil that was triturated twice with petroleum ether. The
resulting residue was extracted with diethyl ether and filtered. The
ethereal solution was evaporated to dryness, giving an orange/brown
residue that was triturated twice more with petroleum ether. The
resulting orange solid was washed with 5 mL of petroleum ether,
dried in vacuo, then dissolved in a minimal amount of diethyl ether,
1
-25 °C, giving 13 as orange crystals (0.0711 g, 63%). H NMR
(C6D6, 300 MHz): δ 46.40 (s), 21.73 (s), 20.17 (s), -3.95 (br, s),
-22.01 (br, s), -44.87 (br, s). UV-vis (C6H6) λmax, nm (ꢀ, M-1
cm-1): 380 (1620), ∼400 (shoulder). Evans Method (C6D6): 5.44
µB. Repeated attempts to obtain satisfactory combustion analysis
7484 Inorganic Chemistry, Vol. 43, No. 23, 2004

FeII/IV Redox and Hydrogenation of Olefins
of crystalline samples were consistently low in %C and %H. Anal.
Calcd for C29H58BFeP3: C, 61.50; H, 10.32. Found: C, 60.04; H,
9.94.
corresponding alkane. For hydrogenations using 3 as the precatalyst,
linear fits to the consumption of starting material all gave R2 values
greater than 0.99, which were used to calculate turnover frequencies;
for the hydrogenation of styrene using 6 as a precatalyst, an
exponential fit to the consumption of styrene gave R2 > 0.98, which
was used to calculate the turnover frequency.
Hydrogenation Reactions (General Procedure): A 0.0379 g
(0.060 mmol) amount of 3 and 0.0110 g (0.059 mmol) of ferrocene
were dissolved in 3.00 mL of C6D6, giving a stock solution of 0.02
1
M 3 and 0.02 M ferrocene. Ferrocene was used as an internal H
Acknowledgment. This work was supported by the NSF
(CHE-01232216) and the DOE (PECASE). E.J.D. is grateful
for the Laszlo Zechmeister Fellowship (Caltech). Lawrence
Henling and Theodore A. Betley are acknowledged for
crystallographic assistance. We thank Paul J. Chirik for
sharing results prior to publication.
NMR integration standard and did not affect the rates of hydro-
genation. A 0.14 mmol amount of each substrate was added to 0.70
mL of the stock solution, giving a 0.20 M solution of the substrate,
and the C6D6 solution was placed in a J. Young tube (∼3.4 mL
capacity). The solution was degassed by three freeze/pump/thaw
cycles, and the tube was refilled with 1 atm H2 (∼2.7 mL, 0.11
mmol). For all reactions conducted at room temperature, the tube
was continuously inverted (12 min-1) to increase mass transfer of
H2 into the solution. The tube was periodically refilled to maintain
Supporting Information Available: ORTEP diagrams, tables
of bond lengths and angles, details of structure solution for 2, 3, 5,
and 8-11, and crystallographic data in CIF format. This material
is available free of charge via the Internet at http://pubs.acs.org.
1
an H2 pressure of 1 atm, and the reaction was monitored by H
NMR until completion. All reactions proceeded cleanly, resulting
in complete conversion of the alkene or alkyne substrate to the
IC0488583
Inorganic Chemistry, Vol. 43, No. 23, 2004 7485