Investigation on the Protonation of a Trisubstituted [Fe 2(CO)3(PPh3)(κ2-phen)(μ- pdt)] Complex: Rotated versus Unrotated Intermediate Pathways

Article abstract of DOI:10.1021/ic100108h

The substitution of PPh₃ for a carbonyl group at the {Fe(CO)₃} moiety in [Fe₂(CO)₄(κ ²-phen)(μ-pdt)] results in the formation of the trisubstituted complex [Fe₂(CO)₃(PPh₃)(κ²- phen)(μ-pdt)] (2). Unlike its tetracarbonyl precursor, the protonation of 2 at low temperature does not afford any apparent transient terminal hydride species. Hydride formation for [Fe₂(CO)₃(L) (κ²-phen)(μ-pdt)] (L = PPh₃, CO) species is also studied by density functional theory calculations, which show that activation barriers to give terminal and bridging hydrides can be remarkably close for this class of organometallic compounds.

Full text of DOI:10.1021/ic100108h

Inorg. Chem. 2010, 49, 5003–5008 5003
DOI: 10.1021/ic100108h
Investigation on the Protonation of a Trisubstituted [Fe2(CO)3(PPh3)(K2-phen)(μ-pdt)]
Complex: Rotated versus Unrotated Intermediate Pathways
ꢀ ꢀ
ꢀ
Pierre-Yves Orain, Jean-Franc-ois Capon, Frederic Gloaguen, Franc-ois Y. Petillon, Philippe Schollhammer,* and
Jean Talarmin
ꢀ
ꢀ
ꢀ
Universite Europeenne de Bretagne, France, and CNRS, UMR 6521 “Chimie, Electrochimie Moleculaires
et Chimie Analytique”, ISSTB, CS 29238 Brest-Cedex 3, France
Giuseppe Zampella and Luca De Gioia*
Department of Biotechnology and Bioscience, University of Milano-Bicocca, Piazza della Scienza 1, 20126
Milan, Italy
Thierry Roisnel
ꢀ
ꢀ
Centre de Diffractometrie X, UMR CNRS 6226, Universite de Rennes 1, 35042 Rennes Cedex, France
Received January 18, 2010
The substitution of PPh3 for a carbonyl group at the {Fe(CO)3} moiety in [Fe2(CO)4(κ2-phen)(μ-pdt)] results in the
formation of the trisubstituted complex [Fe2(CO)3(PPh3)(κ2-phen)(μ-pdt)] (2). Unlike its tetracarbonyl precursor, the
protonation of 2 at low temperature does not afford any apparent transient terminal hydride species. Hydride formation
for [Fe2(CO)3(L)(κ2-phen)(μ-pdt)] (L = PPh3, CO) species is also studied by density functional theory calculations,
which show that activation barriers to give terminal and bridging hydrides can be remarkably close for this class of
organometallic compounds.
Introduction
mechanisms implied at the molecular level. Recently, the
use of chelating ligands allowed better control of the steric
and electronic properties of the artificial diiron framework.
New models of the general formula [Fe2(CO)4(κ2-L2)-
(μ-dithiolate)]2-7 and [Fe2(CO)2(κ2-L2)2(μ-dithiolate)]8 were
synthesized in order to stabilize terminal hydride species that
could act as possible key intermediates in the natural rever-
sible hydrogen activation processes. Barton and Rauchfuss8
demonstrated that two chelating diphosphines in diiron
molecules [Fe2(CO)2(κ2-L2)2(μ-dithiolate)] (dithiolate = pdt,
adt; L = dppv = cis-Ph2PCHdCHPPh2) inhibit the isomer-
ization of terminal hydride into bridging hydride because of
both the high basicity of the diiron center and the steric
crowding of the diphosphine, while monochelated compounds
[Fe2(CO)4(κ2-L2)(μ-dithiolate)] underwent fast isomerization
of terminal to bridging hydride even at temperature below
-30 °C.2,3,4a-c Considering the few studies dedicated to
Mimics of the diiron subsite of [FeFe]-hydrogenases
(Scheme 1) continue to be intensively developed because,
until now, few efficient electrocatalysts have been obtained
and the mechanism of the natural production/uptake of
hydrogen remains unclear.1
It is obvious that a fine-tuning of the structural and
electronic features of the coordination sphere of synthetic
diiron molecules is required to improve their activity toward
protons. Novel structural and chemical models are still
necessary to gain a better understanding of the protonation
*To whom correspondence should be addressed. E-mail: schollha@
univ-brest.fr (P.S.), luca.degioia@unimib.it (L.D.G.).
(1) (a) Capon, J.-F.; Gloaguen, F.; Petillon, F. Y.; Schollhammer, P.;
Talarmin, J. Coord. Chem. Rev. 2009, 253, 1476–1494. (b) Tard, C.; Pickett,
C. J. Chem. Rev. 2009, 109, 2245–2274. (c) Gloaguen, F.; Rauchfuss, T. B.
Chem. Soc. Rev. 2009, 38, 100–108. (d) Rakowski DuBois, M.; DuBois, D. L.
Chem. Soc. Rev. 2009, 38, 62–72. (e) Felton, G. A. N.; Mebi, C. A.; Petro, B. J.;
Vannucci, A. K.; Evans, D. H.; Glass, R. S.; Lichtenberger, D. L. J. Organomet.
ꢀ
ꢀ
(2) Orain, P.-Y.; Capon, J.-F.; Kervarec, N.; Gloaguen, F.; Petillon, F.;
ꢀ
Chem. 2009, 694, 2681–2699. (f) Capon, J.-F.; Gloaguen, F.; Petillon, F. Y.;
Pichon, R.; Schollhammer, P.; Talarmin, J. Dalton Trans. 2007, 3754–3756.
(3) Morvan, D.; Capon, J.-F.; Gloaguen, F.; Le Goff, A.; Marchivie, M.;
Michaud, F.; Schollhammer, P.; Talarmin, J.; Yaouanc, J.-J. Organometal-
lics 2007, 26, 2042–2052.
Schollhammer, P.; Talarmin, J. C.R. Chimie 2008, 11, 842–851. (g) Capon, J.-F.;
Gloaguen, F.; Petillon, F. Y.; Schollhammer, P.; Talarmin, J. Eur. J. Inorg. Chem.
ꢀ
2008, 4671–4681.
r
2010 American Chemical Society
Published on Web 05/05/2010
pubs.acs.org/IC

5004 Inorganic Chemistry, Vol. 49, No. 11, 2010
Orain et al.
Scheme 1
Scheme 2
trisubstituted molecules of the general formula [Fe2(CO)3L3-
(μ-dithiolate)]5,9,10 that can be proposed as models mimicking
the dithiolate diiron tricarbonyl center of the natural diiron
[2Fe]H subsite, we decided to investigate the protonation of
[Fe2(CO)3L0(κ2-L2)(μ-dithiolate)] in order to verify that the
combination of a chelated ligand at one iron site with a
phosphine group located at the other metallic site could
provide sufficient basicity at the {Fe(CO)2L} moiety for
stabilizing a terminal hydride compound. Recently, Hogarth
and Richards reported the synthesis and protonation of the
triphosphine complex [Fe2(CO)3{μ,κ2-(Ph2PCH2CH2)2PPh}-
(μ-pdt)] at room temperature, giving a bridging hydride
compound.9 More recently, NMR and theoretical studies of
the protonation process of complexes [Fe2(CO)4-xLx(κ2-dppv)-
(μ-pdt)] (x = 0, 1; L = PMe3) were reported by De Gioia and
Rauchfuss.10 Their studies suggest a general pathway for the
protonation process of the diiron bridging dithiolate mole-
cules, implying a transient terminal hydride intermediate.
This prompts us to complete our ongoing works concerning
the chemical effect of the replacement of a carbonyl by a
phosphine at the {Fe(CO)3} moiety on the protonation
process in dissymmetrically substituted monochelated com-
pounds. To support this experimental work, we present also
a theoretical study of the protonation of [Fe2(CO)4-xLx-
(κ2-phen)(μ-pdt)] complexes (x = 0, 1; L = PPh3 or PMe3),
in which the electron-donating phenanthroline ligand pre-
vents the formation of basal-apical isomers, which limits the
analysis to a dibasal form.2
Results and Discussion
Synthesis, Characterization, and Protonation Study of
[Fe2(CO)3(PPh3)(K2-phen)(μ-pdt)] (2). The treatment of
[Fe2(CO)4(κ2-phen)(μ-pdt)] (1) with an excess of PPh3
in refluxing toluene in the presence of Me3NO 2H2O
3
afforded the expected PPh3-substituted complex 2 in low
yields after chromatography (Scheme 2). The 31P{1H}
NMR spectrum in CD2Cl2 displays a singlet at 59.3 ppm,
and the 1H NMR data (see the Experimental Section) are
in accordance with the replacement of one PPh3 for CO.
The IR spectrum, recorded in toluene, exhibits three
bands in the νCO region at 1950, 1897, and 1880 cm-1
,
which are shifted to lower energies by ca. 39 cm-1 relative
to those of 1, which is consistent with the electron-
donating effect of the phosphine ligand. Several attempts
to perform X-ray analysis on different samples of crystals
of 2 cannot be considered satisfactory (see Figure 1 in the
Supporting Information). The most that can be said is
that the X-ray results are consistent with the chemical and
spectroscopic evidence, which indicates the formation of
2 with a dibasal binding mode of the phenanthroline
group and an apical position of the PPh3 ligand, as has
been shown for complexes [Fe2(μ-pdt)(dppv)(PMe3)-
(CO)3] and (Et4N)[Fe2(μ-pdt)(dppv)(CN)(CO)3].5
The treatment of 2with an excess of HBF4 OEt2 (5 equiv)
3
at 183 K led to the straightforward formation of a bridging
hydride species [Fe2(CO)3(PPh3)(κ2-phen)(μ-pdt)(μ-H)]þ
(3; Scheme 3). No intermediate with a terminal hydride
was detected in this reaction.
ꢀ
(4) (a) Ezzaher, S.; Capon, J.-F.; Gloaguen, F.; Petillon, F. Y.; Schollhammer,
P.; Talarmin, J.; Pichon, R.; Kervarec, N. Inorg. Chem. 2007, 46, 3426–3428.
(b) Adam, F. I.; Hogarth, G.; Kabir, S. E.; Richards, I. C.R. Chimie 2008, 11, 890–
The 1H NMR spectrum of 3 in CD2Cl2 at 183 K exhibits
two doublets at -7.69 and -9.37 ppm with typical cou-
pling constants11 JPH of 25.6 and 13.0 Hz, respectively
(Figure 2 in the Supporting Information). These signals
can be confidently assigned to basal (3ba) and apical (3ap)
isomers, respectively.10b The warming of the solution at
room temperature revealed a typical apical-basal isomer-
ization of 3ap into 3ba (Scheme 3).4
ꢀ
905. (c) Ezzaher, S.; Capon, J.-F.; Gloaguen, F.; Petillon, F. Y.; Schollhammer, P.;
Talarmin, J.; Pichon, R.; Kervarec, N. C.R. Chimie 2008, 11, 906–914. (d) Ezzaher,
ꢀ
S.; Capon, J.-F.; Gloaguen, F.; Petillon, F. Y.; Schollhammer, P.; Talarmin, J. Inorg.
Chem. 2007, 46, 9863–9872. (e) Wang, N.; Wang, M.; Liu, T.; Zhang, T.;
Darensbourg, M. Y.; Sun, L. Inorg. Chem. 2008, 47, 6948–6955. (f) Adam, F. I.;
Hogarth, G.; Richards, I.; Sanchez, B. E. Dalton Trans. 2007, 2495–2498.
(g) Duan, L.; Wang, M.; Li, P.; Na, Y.; Wang, N.; Sun, L. Dalton Trans. 2007,
ꢀ
1277–1283. (h) Morvan, D.; Capon, J.-F.; Gloaguen, F.; Petillon, F. Y.; Schollhammer,
P.; Talarmin, J.; Yaouanc, J.-J; Michaud, F.; Kervarec, N. J. Organomet. Chem. 2009,
ꢀ
694, 2801–2807. (i) Ezzaher, S.; Capon, J.-F.; Dumontet, N.; Gloaguen, F.; Petillon,
The IR spectrum of a CH2Cl2 solution of 3, obtained at
room temperature by the addition of 5 equiv of HBF4
F. Y.; Schollhammer, P.; Talarmin, J. J. Electroanal. Chem. 2009, 626, 161–170.
(j) Song, L.-C.; Wang, T.-H.; Ge, J.-H.; Mei, S.-Z.; Gao, J.; Wang, L.-X.; Gai, B.; Zhao,
L.-Q.; Yan, J.; Wang, Y.-Z. Organometallics 2008, 27, 1409–1416. (k) Harb, M. K.;
Windhager, J.; Daraosheh, A.; Gorls, H.; Lockett, L. T.; Okumura, N.; Evans, D. H.;
Glass, R. S.; Lichtenberger, D. L.; El-Khateeb, M.; Weigand, W. Eur. J. Inorg. Chem.
2009, 3414–3420.
(5) Justice, A. K.; Zampella, G.; De Gioia, L.; Rauchfuss, T. B.; van der Vlugt,
J. I.; Wilson, S. R. Inorg. Chem. 2007, 46, 1655–1664.
(6) Ezzaher, S.; Orain, P.-Y.; Capon, J.-F.; Gloaguen, F.; Petillon, F. Y.;
3
OEt2 to 2, exhibits three strong bands in the νCO region
at 2036, 1984, and 1967 cm-1, which are similar to those
observed for the related bridging hydride compound
[Fe2(CO)3{μ,κ2-(Ph2PCH2CH2)2PPh}(μ-pdt) (μ-H)]þ 9
(Figure 3 in the Supporting Information). Unfortunately,
like its carbonyl analogue [Fe2(CO)4(κ2-phen)(μ-pdt)(μ-
H)]þ (4),2 3 was unstable at room temperature and no
reliable elemental analysis and single crystal could be
obtained. This led us to try to crystallize compound 4 at
low temperature. As a matter of fact, single crystals of 4
were successfully grown by the slow diffusion of pentane
€
ꢀ
Roisnel, T.; Schollhammer, P.; Talarmin, J. Chem. Commun. 2008, 2547–
2549.
ꢀ
(7) (a) Ezzaher, S.; Capon, J.-F.; Gloaguen, F.; Petillon, F. Y.; Schollhammer,
P.; Talarmin, J.; Kervarec, N. Inorg. Chem. 2009, 48, 2–4. (b) Wang, N.; Wang, M.;
Zhang, T.; Li, P.; Liu, J.; Sun, L. Chem. Commun. 2008, 5800–5802.
(8) Barton, B. E.; Rauchfuss, T. B. Inorg. Chem. 2008, 47, 2261–2263.
(9) Hogarth, G.; Richards, I. Inorg. Chem. Commun. 2007, 10, 66–70.
(10) (a) Zampella, G.; Fantucci, P.; De Gioia, L. J. Am. Chem. Soc. 2009,
131, 10909–10917. (b) Barton, B. E.; Zampella, G.; Justice, A. K.; De Gioia, L.;
Rauchfuss, T. B.; Wilson, S. R. Dalton Trans. 2010, 3011–3019.
(11) van der Vlugt, J. I.; Rauchfuss, T. B.; Whaley, C. M.; Wilson, S. R.
J. Am. Chem. Soc. 2005, 127, 16002–16013.

Article
Inorganic Chemistry, Vol. 49, No. 11, 2010 5005
Scheme 4. Computed Energy Profile for the Formation of Terminal
Hydride and μ-Hydride Derivatives for Complex 2a
Figure 1. Molecularstructureofcompound[Fe2(CO)4(κ2-phen)(μ-pdt)-
(μ-H)](BF4), CH2Cl2, and H2O BH3 (4) with thermal ellipsoids at 30%
3
probability.
a Energies are in kcal/mol. All structures can be drawn showing the
central carbon atom of the dithiolate ligand directed either at the open
coordination site or away from that site. Because the two forms were
computed to be very close in energy, we arbitrarily decided to draw the
structures in the first fashion.
Scheme 3
Scheme 5. Computed Energy Profile for the Formation of Terminal
Hydride and μ-Hydride Derivatives for Complex 2, When F3CSO3H
(Triflic Acid) Is Employed as a Proton-Releasing Systema
˚
Table 1. Selected Bond Lengths (A) and Angles (deg) for 4
Fe1-Fe2
Fe1-N1
Fe1-N2
Fe1-S1
Fe1-S2
Fe1-H
Fe1-C1
Fe2-C2
Fe2-C3
Fe2-C4
Fe2-S1
2.5990(5)
1.998(2)
1.997(2)
2.2252(7)
2.2422(7)
1.77(3)
1.780(3)
1.815(3)
1.817(3)
1.825(3)
2.2842(7)
Fe2-S2
Fe2-H
C1-O1
C2-O2
C3-O3
C4-O4
2.2867(7)
1.59(3)
1.144(3)
1.135(3)
1.134(3)
1.134(3)
a Energies in kcal/mol.
Fe1-S1-Fe2
Fe1-S2-Fe2
70.37(2)
70.03(2)
process affording the bridging hydride diiron derivative
operates systematically through an initial formation of a
transient terminal hydride species. Moreover, no terminal
1
hydride species was detected at 188 K in the H NMR
into dichloromethane at 213 K, which allowed us to describe
the structure of one member of the series of complexes [Fe2-
(CO)3L(κ2-phen)(μ-pdt)(μ-H)]þ (Figure 1 and Table 1). The
main feature of this structure is the unsymmetrical binding
spectrum of the tetracarbonyl edt analogue of 1, [Fe2-
(CO)4(κ2-phen)(μ-edt)], in the presence of a weaker acid
than HBF4,2,4a such as CF3CO2H, but the straightforward
formation of a bridging hydride derivative was observed.
The protonation of 1 with HBF4 in the same reaction
conditions affords a terminal hydride species that iso-
merizes at higher temperature. Similar results have been
obtained in the reaction of the related diphosphine com-
pound [Fe2(CO)4(κ2-dppe)(μ-pdt)] with CF3CO2H. This
might suggest that the protonation at the inverted iron
center in the rotated transition state requires the use of a
sufficiently strong acid; otherwise, it is the unrotated form
that is protonated at the bridging site.
˚
˚
mode of the hydride [Fe1-H, 1.77(3) A;Fe2-H, 1.59(3) A],
which is similar to that observed in the bis-N-heterocyclic
carbene analogue [Fe2(CO)4(η2-IMeCH2IMe)(μ-pdt)(μ-H)]þ
(whereIMeCH2IMe = 1,3-dimethylimidazol-2-ylidene; Fe-H,
3
˚
1.710 and 1.562 A).
The substitution of PPh3 for a carbonyl ligand at the
{Fe(CO)3} moiety in the trisubstituted complex 2 does
not stabilize, even at low temperature, any transient
terminal hydride species. This led us to wonder if the

5006 Inorganic Chemistry, Vol. 49, No. 11, 2010
Orain et al.
Scheme 6. Rearrangement of the (CO)2L Ligands Leading to [Fe2(CO)3(L)(κ2-phen)(μ-pdt)] (L = PPh3, CO) Forms Featuring a Terminal Vacant
Coordination Position
Scheme 7. Computed Energy Profile for the Formation of Terminal
Hydride and μ-Hydride Derivatives for the [Fe2(CO)3(PMe3)(κ2-phen)-
(μ-pdt)] Complexa
Scheme 8. Computed Energy Profile for the Formation of Terminal
Hydride and μ-Hydride Derivatives for Complex 2, When F3CSO3H Is
Employed as the Proton-Releasing Systema
a Energies are in kcal/mol.
a Energies are in kcal/mol.
Remarkably, when CF3SO3H is the acid, it turns out that
μ protonation is the most kinetically favored pathway,
even though the reaction energy barriers for terminal
protonation and μ-protonation are still very similar (16.3
and 17.5kcal/mol forμ-protonation and terminal protona-
tion, respectively). In this context, it must be noted that
previous DFT investigations of terminal protonation ver-
sus μ-protonation in diiron complexes containing phos-
phine ligands10a showed that terminal protonation is
generally clearly favored relative to μ protonation. This
latter observation prompted us to further investigate the
protonation mechanism of 2, with the aim of better high-
lighting the electronic and steric factors responsible of its
peculiar behavior. First we have taken into account the
possibility that terminal hydride formation takes place on
going through a stepwise mechanism, which implies an
initial rotation of the Fe(CO)2(PPh3) moiety, leading to a
species featuring a vacant coordination site suited to react
with the acid and form the terminal hydride (Scheme 6).
Indeed, it turned out that the rotated species 2 does not
correspond to a stable species but to a transition state,
excluding the possibility of a stepwise protonation me-
chanism. Nevertheless, it is worth noting that the com-
puted energy barrier for the rotation of Fe(CO)2(PPh3) in 2
is very low (4.6 kcal/mol) and similar to the corresponding
value computed for the rotation of the Fe(CO)3 moiety in
1, clearly showing how the stereoelectronic properties of
the PPh3 group do not hinder the rotation process. This
result suggests that the kinetically hindered event during
the terminal protonation of 2 is not the rearrangement of
the ligands coordinated to the iron atom but the proton
To strengthen the understanding of the protonation
processes discussed above, we used density functional
theory (DFT) calculations to dissect the reaction between
[Fe2(CO)3(L)(κ2-phen)(μ-pdt)] (L = PPh3, PMe3, CO)
and typical acids such as HBF4 OEt2 or CF3SO3H.
3
According to the DFT results, the μ-hydride species 3 is
lower in energy than the corresponding terminal hydride
isomer (Scheme 4). In addition, the basal form 3ba is
computed to be thermodynamically more stable than the
corresponding apical 3ap isomer (not shown), in agreement
with experimental observations. The reaction energy bar-
riers that were calculated for the protonation of 2 by
HBF4 OEt2 are quite large (27.4 and 30.6 kcal/mol;
3
Scheme 4), and they differ only by 3.2 kcal/mol. In parti-
cular, DFT calculations would suggest that the terminal
protonation of 2 is kinetically slightly preferred to μ pro-
tonation, in contradiction with experimental evidence.
However, it must be noted that the difference between the
reaction energy barriers is close to the expected error of the
method for such kinds of calculations and, most impor-
tantly, the description within the DFT representation of the
acid moiety HBF4 OEt2 might be nonoptimal. In fact, it is
3
known that the proper description of the chemical nature of
the HBF4 OEt2 species in solution can be difficult.12
3
To try to overcome modeling limitations, we have also
computed the reaction energy profile for the protonation
of 2 using CF3SO3H as the acid moiety (Scheme 5).
(12) Greenwood, N. N.; Earnshaw, A. Chemistry of Elements; 2nd ed.;
Butterworth Heinemannn (Elsevier Science): Linacre House, Jordan Hill, Oxford,
1997; p 199.

Article
Inorganic Chemistry, Vol. 49, No. 11, 2010 5007
Scheme 9. Computed Energy Profile for the Formation of Terminal and μ-Hydride Derivatives for Complex 1a
a Energies are in kcal/mol.
transfer from the acid to the diiron complex. Therefore, to
further shed light on the factors affecting the terminal
protonation in 2, we have used again DFT methods to
study the protonation of the [Fe2(CO)3(PMe3)(κ2-phen)-
(μ-pdt)] complex, which differs from 2 by the replacement
of the PPh3 ligand by the sterically less bulky PMe3 group.
is more hindered in 2 by the bulky PPh3 ligand. The latter
observations emphasize the critical role of the metal acces-
sibility to implement the kinetics of the protonation in
diiron complexes featuring phosphine and phenanthroline
ligands. They also fit well with recently reported results
that highlighted the crucial role of pendant acid/basic
groups (such as an amine) within the diiron complex
scaffold to increase the rate of protonation of bioinspired
diiron complexes.13
We used both HBF4 OEt2 and CF3SO3H as proton
sources (Schemes 7 and 8). Considering the reaction with
3
HBF4 OEt2, it turns outthat the energybarriers calculated
3
for μ-protonation in [Fe2(CO)3(PMe3)(κ2-phen)(μ-pdt)]
and 2 are very similar (29.0 and 30.6 kcal/mol, respec-
tively), while that calculated for the terminal protonation
in [Fe2(CO)3(PMe3)(κ2-phen)(μ-pdt)] is somewhat lower
(24.2 kcal/mol) than that in 2 (27.4 kcal/mol). The kinetic
preference for terminal protonation versus μ-protonation
is even more evident when CF3SO3H is the acid. Indeed, in
this case, the energy barriers for μ protonation of [Fe2-
(CO)3(PMe3)(κ2-phen)(μ-pdt)] and 2 are still similar (14.7
and 17.5 kcal/mol, respectively), whereas the corresponding
energy barriers for terminal protonation differ by 10 kcal/
mol (7.5 and 17.5 kcal/mol). The larger energy difference
between the reaction energy barriers leading to terminal
hydride and μ-hydride computed for [Fe2(CO)3(PMe3)-
(κ2-phen)(μ-pdt)], relative to that computed for 2, can be
rationalized by considering that steric accessibility to the
Fe-Fe bond is essentially identical in the two diiron
complexes, whereas that to the terminal coordination site
The pathways for the formation of the μ-hydride and
terminal hydride species in the reaction between 1 and
HBF4 OEt2 have also been explored by DFT calculations
3
(Scheme 9). The computed energy profile shows that in
this case the route leading to the terminal hydride species
is clearly kinetically favored because the corresponding
activation energy is 4.7 kcal/mol lower than that calcu-
lated for the formation of the bridging hydride isomer.
The fact that the reaction energy barriers computed for 1
are larger than those for 2 (28.9 and 33.6 kcal/mol versus
27.4 and 30.6 kcal/mol) can be explained by the replace-
ment of CO with a PPh3 group, which increases the
basicity of the complex. Such an effect is even more
evident when considering the relative thermodynamic
stability of the hydride species. In fact, the terminal
(13) (a) Barton, B. E.; Olsen, M. T.; Rauchfuss, T. B. J. Am. Chem. Soc.
2008, 130, 16834–16835. (b) Wang, N.; Wang, M.; Liu, J.; Chen, L.; Sun, L.
Inorg. Chem. 2009, 48, 11551–11558.

5008 Inorganic Chemistry, Vol. 49, No. 11, 2010
Orain et al.
hydride [HFe2(CO)4(κ2-phen)(μ-pdt)] is significantly higher
in energy than the corresponding bridging hydride isomer
(Scheme 9).
Table 2. Crystallographic Data for Complex 4
empirical formula
fw
C20H19 B2 Cl2F7Fe2N2O5S2
768.71
100(2)
triclinic
P1
11.3463(7)
11.4918(7)
12.4419(8)
88.214(3)
64.190(3)
81.344(3)
1442.75(16)
2
temperature (K)
cryst syst
space group
Conclusion
˚
a (A)
The replacement of a carbonyl by a phosphine at the
{Fe(CO)3} moiety in dissymmetrically substituted monoche-
lated diiron compounds does not allow characterization of
˚
b (A)
˚
c (A)
R (deg)
β (deg)
γ (deg)
any terminal hydride species upon reaction with HBF4 OEt2,
3
and only the corresponding bridging hydride isomer is ob-
served. DFT calculations on the hydride formation for
species 1 indicate that terminal hydride is the initial product,
in agreement with the experimental observations, whereas for
2, activation barriers for the formation of terminal and
bridging hydrides are remarkably close and therefore the
possibility of the direct formation of a bridging hydride
cannot be excluded.
3
˚
V (A )
Z
F
calcd (Mg/mm3)
1.770
1.416
μ (mm-1
)
cryst size (mm)
range of θ (deg)
reflns measd
Rint
0.54 ꢀ 0.34 ꢀ 0.27
3.53-27.46
24 721
0.0427
unique data/param
R1 [I > 2σ(I)]
R1 (all data)
wR2 (all data)
GOF on F2
6559/389
0.0355
0.0534
0.0876
1.030
Experimental Section
Methods and Materials. All of the experiments were carried
out under an inert atmosphere, using Schlenk techniques.
[Fe2(CO)4(κ2-phen)(μ-pdt)] (1) was prepared according to the
reported procedure.2 The NMR spectra (1H and 31P) were
recorded at room temperature in a CD2Cl2 solution with a
Bruker AMX 400, DRX 500, or AC300 spectrometer and were
referenced to SiMe4 (1H NMR) and H3PO4 (31P NMR). The IR
spectra were recorded on a Nicolet Nexus Fourier transform
spectrometer. Chemical analyses were made by the Service de
Microanalyses ICSN, Gif sur Yvette, France.
3
˚
ΔFmax, ΔFmin (e/A )
0.854, -0.931
B-P86,16 in conjunction with a valence triple-ζ basis set with
polarization on all atoms, a level of theory that has been shown
to be suited to reliably investigate [FeFe]-hydrogenase models.17
Stationary points of the energy hypersurface have been located
by means of energy gradient techniques, and a full vibrational
analysis has been carried out to further characterize each stationary
point. Optimization of the transition state structures has been
carried out according to a procedure based on a pseudo-Newton-
Raphson method. Initially, geometry optimization of a guessed
transition state structure is carried out, constraining the distance
corresponding to the reaction coordinate. Vibrational analysis at
the B-P86/TZVP level of the constrained minimum-energy struc-
tures is then carried out, and if one negative eigenmode corres-
ponding to the reaction coordinate is found, the curvature deter-
mined at such a point is used as the starting point in the transition
state search. The location of the transition state structure is carried
out using an eigenvector-following search: the eigenvectors in the
Hessian are sorted in ascending order, with the first one being that
associated with the negative eigenvalue. After the first step, how-
ever, the search is performed by choosing the critical eigenvector
with a maximum overlap criterion, which is based on the dot
product with the eigenvector followed in the previous step.
The energies reported in the present contribution are pure
electronic energies (E). In fact, the computation of classical
corrections by means of evaluating the rotovibrational partition
function, in order to have an estimate of the Gibbs free energy,
did not afford, in the present case, significant variations to the
reported energy profiles. Furthermore, COSMO corrections
have also been tested by including the ɛ value corresponding
to the experimentally adopted solvent (dichloromethane, ɛ =
9.1 at standard conditions of T and P). Because no appreciable
changes to the whole energy profiles were observed, only in
vacuo values are reported. In light of the available experimental
data and considering the chemical nature of the ligands, only
low-spin species have been investigated.
Crystal data (Table 2) for 4 were collected at T = 100 K on a
Bruker-AXS APEXII diffractometer, equipped with a jet cooler
device and graphite-monochromated Mo KR radiation (λ =
˚
0.710 73 A). The structure was solved by direct methods using
the SIR97 program14a and refined by least-squares methods
based on F2 with SHELXL.14b
Synthesis and Spectroscopic Data of [Fe2(CO)3(PPh3)-
(K2-phen)(μ-pdt)] (2). A total of 1.040 g (2.0 mmol) of 1, 1.604 g
(6.1 mmol) of PPh3, and 0.227 g (2.0 mmol) of Me3NO 2H2O
3
were stirred in refluxing toluene overnight. After filtration, the
solvent was removed in vacuo. The product was purified by
column chromatography on silica gel with CH2Cl2/hexane
(60:40) as the eluent and washed with pentane. Yield: 0.175 g,
11.5%. IR (toluene, cm-1): νCO 1950(s), 1897(s), 1880(s). 1H
NMR (CD2Cl2, 25 °C): δ 9.04 (dd, 2H, JHH = 0.9 Hz, JHH
=
5.4 Hz, phen), 8.15 (d, 2H, JHH = 0.9 Hz, JHH = 7.9 Hz, phen),
7.86 (s, 2H, phen), 7.51 (dd, 2H, JHH = 5.5 Hz, JHH = 7.9 Hz,
phen), 7.60 (m, 6H, PPh3), 7.32 (m, 9H, PPh3), 1.60 (m, 3H, pdt),
1.44 (m, 2H, pdt), 0.55 (m, 1H, pdt). 31P{1H} (CD2Cl2, 25 °C): δ
59.3. Anal. Calcd for C36H29Fe2N2O3PS2,CH2Cl2: C, 53.58; H,
3.77; N, 3.78. Found: C, 53.31; H, 3.77; N, 3.23.
DFT Calculations. DFT calculations have been carried out with
the TURBOMOLE 5.7 suite.15 Geometry optimizations and tran-
sition-state searches have been carried out using the pure functional
(14) (a) SIR97: Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna,
R. J. Appl. Crystallogr. 1999, 32, 115–119. (b) Sheldrick, G. M. Acta Crystal-
logr., Sect. A 2008, 64, 112–122.
(15) (a) Ahlrichs, R.; Bar, M.; Haser, M.; Horn, H.; Kolmel, C. Chem.
Phys. Lett. 1989, 162, 165–169. (b) Eichkornn, K.; Weigend, F.; Treutler, O.;
Ahlrichs, R. Theor. Chem. Acc. 1997, 97, 119–124.
(16) (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098–3100. (b) Perdew, J. P.
Phys. Rev. B 1986, 33, 8822–8824.
Acknowledgment. The authors thank CNRS, ANR “CatH2”,
UBO, and University of Milano-Bicocca for financial support.
P.-Y.O. thanks ADEME for funding.
(17) (a) Bertini, L.; Bruschi, M.; De Gioia, L.; Fantucci, P.; Greco, C.;
Zampella, G. At. Approaches Mod. Biol.: Quantum Chem. Mol. Simul. 2007,
268, 1–46. (b) Bruschi, M.; Zampella, G.; Fantucci, P.; De Gioia, L. Coord.
Chem. Rev. 2005, 249, 1620–1640. (c) Bruschi, M.; Greco, C.; Zampella, G.;
Ryde, U.; Pickett, C. J.; De Gioia, L. C.R. Chimie 2008, 11, 834–841.
Supporting Information Available: Crystallographic data for
2 and 4, NMR figures, tables giving xyz coordinates of opti-
mized models, and a CIF file for compound 4. This material is
available free of charge via the Internet at http://pubs.acs.org.