Requirements for functional models of the iron hydrogenase active site: D2/H2O exchange activity in {(μ-SMe)(μ-pdt)[Fe(CO)2(PMe3)] 2+}[BF4-]

Article abstract of DOI:10.1021/ic026005+

Hydrogen uptake in hydrogenase enzymes can be assayed by H/D exchange reactivity in H₂/D₂O or H₂/D₂/H₂O mixtures. Diiron(I) complexes that serve as structural models for the active site of iron hydrogenase are not active in such isotope scrambling but serve as precursors to Fe^IIFe^II complexes that are functional models of [Fe]H₂ase. Using the same experimental protocol as used previously for {(μ-H)(μ-pdt)[Fe(CO)₂(PMe₃)] ₂⁺}, 1-H⁺ (Zhao et al. J. Am. Chem. Soc. 2001, 123, 9710), we now report the results of studies of {(μ-SMe)(μ-pdt)[Fe(CO)₂(PMe₃)]₂ ⁺}, 1-SMe⁺, toward H/D exchange. The 1-SMe⁺ complex can take up H² and catalyze the H/D exchange reaction in D₂/H₂O mixtures under photolytic, CO-loss conditions. Unlike 1-H⁺, it does not catalyze H₂/D₂ scrambling under anhydrous conditions. The molecular structure of 1-SMe⁺ involves an elongated Fe...Fe separation, 3.11 A, relative to 2.58 A in 1-H⁺. It is proposed that the strong SMe^- bridging ligand results in catalytic activity localized on a single Fe^II center, a scenario that is also a prominent possibility for the enzyme active site. The single requirement is an open site on Fe^II available for binding of D₂ (or H₂), followed by deprotonation by the external base H₂O (or D₂O).

Full text of DOI:10.1021/ic026005+

Inorg. Chem. 2003, 42, 2489−2494
Requirements for Functional Models of the Iron Hydrogenase Active
Site: D /H O Exchange Activity in
2
2
+
{(µ-SMe)(µ-pdt)[Fe(CO)2(PMe3)]2 }[BF4-]
Irene P. Georgakaki, Matthew L. Miller, and Marcetta Y. Darensbourg*
Department of Chemistry, Texas A&M UniVersity, College Station, Texas 77843
Received September 5, 2002
Hydrogen uptake in hydrogenase enzymes can be assayed by H/D exchange reactivity in H /D O or H /D /H O
2
2
2
2
2
mixtures. Diiron(I) complexes that serve as structural models for the active site of iron hydrogenase are not active
II II
in such isotope scrambling but serve as precursors to Fe Fe complexes that are functional models of [Fe]H ase.
2
Using the same experimental protocol as used previously for {(µ-H)(µ-pdt)[Fe(CO)2(PMe3)]2+}, 1-H+ (Zhao et al.
J. Am. Chem. Soc. 2001, 123, 9710), we now report the results of studies of {(µ-SMe)(µ-pdt)[Fe(CO)2(PMe3)]2+},
1-SMe+, toward H/D exchange. The 1-SMe+ complex can take up H and catalyze the H/D exchange reaction in
2
D /H O mixtures under photolytic, CO-loss conditions. Unlike 1-H+, it does not catalyze H /D scrambling under
2
2
2
2
anhydrous conditions. The molecular structure of 1-SMe+ involves an elongated Fe‚‚‚Fe separation, 3.11 Å, relative
to 2.58 Å in 1-H+. It is proposed that the strong SMe- bridging ligand results in catalytic activity localized on a
II
single Fe center, a scenario that is also a prominent possibility for the enzyme active site. The single requirement
is an open site on Fe available for binding of D (or H ), followed by deprotonation by the external base H O (or
II
2
2
2
D O).
2
Introduction
of Fe-only hydrogenase, [Fe]H2ase, to easily synthesized and
well-known dinuclear FeIFeI organometallic complexes4
(Figure 1a and b, respectively) has inspired the preparation
of derivatives modified in such a way to better mimic the
features of the enzyme active site.5 Whereas the enzyme is
characterized by at least three different oxidation levels,
mixed-valent FeIFeII is a prominent candidate for the form
that is responsible for H2 uptake; heterolytic (H+/H-)
cleavage; and, by microscopic reversibility, H2 production
and evolution.6
The connection between enzymes that take up and activate
molecular hydrogen and noble metals that show similar
functions has intrigued scientists since the discovery of
hydrogenases in 1931.1 The recent wave of attention has been
fueled by protein crystal structures which provide striking
snapshots of binuclear active sites in both nickel-iron2 and
iron hydrogenases.3 The similarity of the active site structure
* Author to whom correspondence should be addressed. E-mail:
marcetta@mail.chem.tamu.edu.
(1) Stephenson, M.; Stickland, L. H. Biochem. J. 1931, 25, 205.
(2) (a) Volbeda, A.; Charon, M.-H.; Piras, C.; Hatchikian, E. C.; Frey,
M.; Fontecilla-Camps, J. C. Nature 1995, 373, 580. (b) Volbeda, A.;
Garcin, E.; Piras, C.; De Lacey, A. L.; Fernandez, V. M.; Hatchikian,
E. C.; Frey, M.; Fontecilla-Camps, J. C. J. Am. Chem. Soc. 1996,
118, 12989. (c) Garcin, E.; Vernede, X.; Hatchikian, E. C.; Volbeda,
A.; Frey, M.; Fontecilla-Camps, J. C. Structure 1999, 7, 557. (d)
Higuchi, Y.; Yagi, T.; Yasuoka, N. Structure 1997, 5, 1671. (e)
Higuchi, Y.; Ogata, H.; Miki, K.; Yasuoka, N.; Yagi, T. Structure
1999, 7, 549.
(3) (a) Peters, J. W.; Lanzilotta, W. N.; Lemon, B. J.; Seefeldt, L. C.
Science 1998, 282, 1853. (b) Nicolet, Y.; Piras, C.; Legrand, P.;
Hatchikian, C. E.; Fontecilla-Camps, J. C. Structure 1999, 7, 13. (c)
Nicolet, Y.; De Lacey, A. L.; Verne´de, X.; Fernandez, V. M.;
Hatchikian, E. C.; Fontecilla-Camps, J. C. J. Am. Chem. Soc. 2001,
123, 1596. (d) Lemon, B. J.; Peters, J. W. Biochemistry 1999, 38,
12969.
(4) Seyferth, D.; Womack, G. B.; Gallagher, M. K.; Cowie, M.; Hames,
B. W.; Fackler, J. P., Jr.; Mazany, A. M. Organometallics 1987, 6,
283.
(5) (a) Lyon, E. J.; Georgakaki, I. P.; Reibenspies, J. H.; Darensbourg,
M. Y. Angew. Chem., Int. Ed. 1999, 38, 3178. (b) Schmidt, M.;
Contakes, S. M.; Rauchfuss, T. B. J. Am. Chem. Soc. 1999, 121, 9736.
(c) Le Cloirec, A.; Best, S. P.; Borg, S.; Davies, S. C.; Evans, D. J.;
Hughes, D. L.; Pickett, C. J. Chem. Commun. 1999, 2285. (d)
Lawrence, J. D.; Li, H.; Rauchfuss, T. B.; Be´nard, M.; Rohmer, M.-
M. Angew. Chem., Int. Ed. 2001, 40, 1768. (e) Razavet, M.; Davies,
S. C.; Hughes D. L.; Pickett, C. J. Chem. Commun. 2001, 847.
(6) (a) Popescu, C. V.; Mu¨nck, E. J. Am. Chem. Soc. 1999, 121, 7877.
(b) Pereira, A. S.; Tavares, P.; Moura, I.; Moura, J. J. G.; Huynh, B.
H. J. Am. Chem. Soc. 2001, 123, 2771. (c) De Lacey, A. L.; Stadler,
C.; Cavazza, C.; Hatchikian, E. C.; Fernandez, V. M. J. Am. Chem.
Soc. 2000, 122, 11232. (d) Bennett, B.; Lemon, B. J.; Peters, J. W.
Biochemistry 2000, 39, 7455.
10.1021/ic026005+ CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/11/2003
Inorganic Chemistry, Vol. 42, No. 8, 2003 2489

Georgakaki et al.
Figure 1. Stick drawing structures of (a) active site of [Fe]H2ase,3 (b)
ground state of (µ-pdt)Fe2(CO)6, (c) calculated transition state of Fe(CO)3
unit rotation in (µ-pdt)Fe2(CO)6,8 and (d) a spectroscopically observed FeII-
FeI complex.7
The synthesis of mixed-valent complexes is not easily
achieved in the chemist’s laboratory. A recent spectroelec-
trochemical study provided evidence for such an FeIIFeI
species generated by one-electron oxidation of a diiron(I)
model.7 The mixed-valent product is a good spectroscopic
match for the CO-inhibited oxidized form of the enzyme,3d
resulting in the proposed structure shown in Figure 1d. The
bridging carbonyl of the FeIIFeI complex has not heretofore
been observed in (µ-SRS)[FeI(CO)2L]2 model complexes.
Nevertheless, DFT computations suggest an analogous
structure as a transition state for a rotation process that
equilibrates CO ligands in individual Fe(CO)3 units as seen
in 13C NMR spectra. This rotation accounts for the observed
fluxionality or intramolecular site exchange that interchanges
COapical with CObasal (Figure 1c).8
Figure 2. Stick drawing structures and infrared spectra of 1, 1-H+, and
1-SMe+.
As early as 1934, Farkas, Farkas, and Yudkin demonstrated
that hydrogenase from B. coli (Balantidium coli) catalyzed
the isotope exchange reaction between D2O and H2.11 This
reactivity, together with ortho/para H2 interconversion, the
mechanism of which was studied in detail by Krasna and
Rittenberg,12 has provided the basis for hydrogenase activity
assays that typically demonstrate H2 uptake as indicated by
H/D exchange in H2/D2O mixtures.13 Mixtures of H2/D2 have
also been reported to show scrambling,14 presumably via H2O
mediation. On the basis of such test reactions, our previous
studies indicated that the diiron(II) complex, 1-H+, serves
as a functional model of [Fe]H2ase in the catalytic isotopic
scrambling of D2/H2O and H2/D2 mixtures.10,15 During these
processes, all of which require photolysis, 1-H+ becomes
1-D+, indicating involvement of the bridging hydride in the
isotope exchange mechanism. Intermediates such as struc-
tures A and B, call upon the µ-H to shift to a terminal
position and serve as an internal base under anhydrous
conditions to deprotonate the proximate (η2-H2)FeII moiety.
A weak external base such as H2O can also deprotonate the
acidic (η2-H2)FeII intermediate.16
Although the electrochemically generated FeIIFeI species
has not, as yet, been isolated, homovalent FeIIFeII complexes
are readily obtained from FeIFeI precursors via the binuclear
oxidative addition of electrophiles such as H+ or SMe+.9
Reaction of (µ-pdt)[Fe(CO)2(PMe3)]2, complex 1 (pdt )
SCH2CH2CH2S), with H+ or SMe+ produces {(µ-H)(µ-pdt)-
+
[Fe(CO)2(PMe3)]2 }, 1-H+,10 or {(µ-SMe)(µ-pdt)[Fe(CO)2-
+
(PMe3)]2 }, 1-SMe+, respectively, with concomitant blue
shifts in the ν(CO) IR spectrum (Figure 2). The role of the
PMe3 ligands, analogues to cyanide in the enzyme active
site, is to both increase the electron density in the FeIFeI
bond and to stabilize the FeIIFeII oxidation level following
oxidative addition.
(7) Razavet, M.; Borg, S. J.; George, S. J.; Best, S. P.; Fairhurst, S. A.;
Pickett, C. J. Chem. Commun. 2002, 700.
(8) (a) Georgakaki, I. P.; Thomson, L. M.; Lyon, E. J.; Hall, M. B.;
Darensbourg, M. Y. Coord. Chem. ReV., manuscript submitted. (b)
Lyon, E. J.; Georgakaki, I. P.; Reibenspies, J. H.; Darensbourg, M.
Y. J. Am. Chem. Soc. 2001, 123, 3268.
(9) (a) Fauvel, K.; Mathieu, R.; Poilblanc, R. Inorg. Chem. 1976, 15, 976.
(b) Savariault, J.-M.; Bonnet, J.-J.; Mathieu, R.; Galy, J. C. R. Acad.
Sci. 1977, 284, C663. (c) Lyons, L. J.; Anderson, L. L.; Crane, R. A.;
Treichel, P. M. Organometallics 1991, 10, 587. (d) Treichel, P. M.;
Crane, R. A.; Matthews, R.; Bonnin, K. R.; Powell, D. J. Organomet.
Chem. 1991, 402, 233.
(11) Farkas, A.; Farkas, L.; Yudkin, J. Proc. R. Soc. (London) 1934, B115,
373.
(12) Krasna, A. I.; Rittenberg, D. J. Am. Chem. Soc. 1954, 76, 3015.
(13) (a) Adams, M. W. W.; Mortenson, L. E.; Chen, J.-S. Biochim. Biophys.
Acta 1981, 594, 105. (b) Albracht, S. P. J. Biochim. Biophys. Acta
1994, 1188, 167.
(14) Yagi, T. J. Biochem. 1970, 68, 649.
(15) Zhao, X.; Georgakaki, I. P.; Miller, M. L.; Mejia-Rodriguez, R.;
Chiang, C.-Y.; Darensbourg, M. Y. Inorg. Chem. 2002, 41, 3917.
(16) Landau, S. E.; Morris, R. H.; Lough, A. J. Inorg. Chem. 1999, 38,
6060.
(10) Zhao, X.; Georgakaki, I. P.; Miller, M. L.; Yarbrough, J. C.;
Darensbourg, M. Y. J. Am. Chem. Soc. 2001, 123, 9710.
2490 Inorganic Chemistry, Vol. 42, No. 8, 2003

Functional Models of the Iron Hydrogenase ActiWe Site
with saturated aqueous solution of NH4PF6 in MeOH. The solid
was collected, washed with H2O and Et2O, and dried in air (0.6 g;
yield, 44%). 31P NMR, acetone-d6: 20.95 (PMe3) and -142.7 ppm
(PF6-). Elemental analysis, calculated for Fe2C14H27S3O4P3F6
(found)%: C, 24.92 (24.66); H, 4.00 (4.03).
Test for Catalytic Formation of HD in H2/D2 Mixture. A
medium-pressure NMR tube (Wilmad, 528-PV-7) was charged with
a solution containing 20 mg of {(µ-SMe)(µ-pdt)[Fe(CO)2(PMe3)]2+}-
[PF6-] in 1 g of CD2Cl2. The tube was then pressurized with 5 bar
H2 and with D2 to a total pressure of 10 bar and exposed to sunlight.
1H NMR spectra were taken daily over the course of 8 days to
check the formation of HD. None was observed.
Because an open site on Fe can, under photolysis, be
created by a hydride shift or by CO loss, we questioned
whether the hydride was a requirement for the heterolytic
H2/H2O or D2/H2O cleavage reaction. Hence, the following
study was designed to assess the requirement of a H- ligand
in candidates for H2ase functional models.
H/D Exchange in D2/H2O Mixture. A 0.8-mL portion of a
solution made from 0.15 g of [1-SMe+][PF6-] in 3 mL of CH2Cl2
was put in a medium-pressure NMR tube together with 2 µL of
H2O. The tube was pressurized with 10 bar D2 and exposed to
2
sunlight. H NMR spectra were taken at time intervals to follow
Experimental Section
the formation of HOD.
Materials and Techniques. All manipulations were performed
using standard Schlenk-line and syringe/rubber-septa techniques
under N2 or in an argon atmosphere glovebox. Solvents were of
reagent grade and purified as follows: Dichloromethane was
distilled over P2O5 under N2. Acetonitrile was distilled once from
CaH2, distilled once from P2O5, and freshly distilled from CaH2
immediately before use. Diethyl ether was distilled from sodium/
benzophenone under N2. The following materials were of reagent
grade and used as received: Fe3(CO)12, 1,3-propanedithiol, MeSS-
Me, Me3OBF4 and NH4PF6 (Aldrich Chemical Co.); deuterated
solvents and D2 (Cambridge Isotope Laboratories).
Infrared spectra were recorded on a Mattson 6021 FTIR
spectrometer with DTGS and MCT detectors. 1H, 13C, and 31P NMR
(85% H3PO4 was used as an external reference) spectra were
recorded on a Unity+ 300-MHz superconducting NMR instrument
operating at 299.9, 75.43, and 121.43 MHz, respectively. 2H NMR
spectra were recorded on a Unity Inova-400 NMR instrument with
a 5-mm autoswitchable probe operating at 61.35 MHz and on a
VXR-300 NMR instrument operating at 46.05 MHz.
13CO Exchange Experiment in [(µ-SMe)(µ-pdt)(Fe(CO)2-
PMe3)2+][PF6-]. A 0.7-mL portion of a solution made from 23
mg of [1-SMe+][PF6-] in 1.6 mL of d6-acetone was transferred
into a medium-pressure NMR tube. The tube was lightly degassed,
filled with 20 psi 13CO, and exposed to sunlight. The 13C NMR
spectrum after 1 day showed two doublets (208.95 and 208.07 ppm,
JC-P ) 19.5 and 14.6 Hz, respectively) in the CO region, indicating
the incorporation of 13CO into the complex. The ν(CO) infrared
spectrum in CH2Cl2 showed a shift of the stretching frequencies to
lower wavenumbers consistent with 13CO/12CO exchange in
1-SMe+. Infrared spectrum, ν(CO) region: ν(CO), 1-SMe+
2038, 2024, 1981; ν(CO), 13CO-enriched 1-SMe+ 2020, 1981, 1947
cm-1
.
X-ray Structure Determination. A single crystal was mounted
on a glass fiber at 110 K. The X-ray data were collected on a Bruker
Smart 1000 CCD diffractometer and covered a hemisphere of
reciprocal space by a combination of three sets of exposures. The
space group was determined on the basis of systematic absences
and intensity statistics. Crystal data for [1-SMe+][BF4-]: C14H27-
O4P2S3BF4Fe2, M ) 616.01, orthorhombic, space group Pnma, a
) 14.79(3) Å, b ) 12.54(2) Å, c ) 12.98(2) Å, R ) 90°, â ) 90°,
γ ) 90°, V ) 2408(7) Å3, Z ) 4, Dc) 1.699 g cm-3. The structure
was solved by direct methods. Hydrogen atoms were placed at
idealized positions and refined with fixed isotropic displacement
parameters. The pdt bridge was refined with “envelope-flap”
disorder. Anisotropic refinement for all non-hydrogen atoms was
done by a full-matrix least-squares method with R1 ) 0.0592 and
wR2 ) 0.0956. Programs used include SMART18 for data collection
and cell refinement, SAINTPLUS19 for data reduction, SHELXS-
86 (Sheldrick)20 for structure solution, SHELXL-97 (Sheldrick)21
for structure refinement, and SHELXTL-Plus, version 5.1 or later
(Bruker),22 for molecular graphics and preparation of material for
publication.
Preparations. The neutral dinuclear iron compounds were
prepared according to literature procedures.15
[Me2SSMe+][BF4-]. Following the published procedure,17
a
solution containing 0.74 mL of methyl disulfide (8.06 mmol) in
∼7 mL of CH3CN was added dropwise to an equimolar amount of
-
Me3O+BF4 (1.04 g, 8.06 mmol) dissolved in ∼8 mL of CH3CN
at 0 °C. After the mixture had been stirred for ∼2 h at 0 °C, dry
ether was added to precipitate dimethylthiomethylsulfonium fluo-
roborate ([Me2SSMe+][BF4-]) as a white solid that was stored in
the glovebox at -36 °C (1.1 g; yield, 69%).
Synthesis of {(µ-SMe)(µ-pdt)[Fe(CO)2(PMe3)]2+}[BF4-]. A red
solution of 0.97 g of (µ-pdt)[Fe(CO)2(PMe3)]2 (2 mmol) in ∼50
mL of dry CH2Cl2 was transferred via cannula into a Schlenk flask
containing 0.41 g (2 mmol) of [Me2SSMe+][BF4-]. Following
overnight stirring at 22 °C, the IR spectrum [ν(CO) region] showed
the presence of a mixture of the neutral precursor and the cationic
product in the brown solution. After the mixture had been fitered
through Celite under Ar and concentrated in a vacuum, dry Et2O
was added to precipitate the product and remove the unreacted
neutral complex, 1. Crystals suitable for an X-ray crystal structure
determination were grown from CH2Cl2 solutions layered with Et2O
at -5 °C. Infrared spectrum, ν(CO), CH2Cl2: 2038(m), 2024(s),
1981(s) cm-1. Because of the difficulty in isolating the product in
Results and Discussion
Reaction of (µ-pdt)[Fe(CO)2(PMe3)]2, 1, with the SMe+
-
synthon, [Me2SSMe+][BF4 ], in CH2Cl2 results in the
(18) SMART 1000 CCD; Bruker Analytical X-ray Systems: Madison, WI,
1999.
(19) SAINT-Plus, version 6.02; Bruker: Madison, WI, 1999.
(20) Sheldrick, G. SHELXS-86: Program for Crystal Structure Solution;
Institu¨t fu¨r Anorganische Chemie, Universita¨t Gottingen: Gottingen,
Germany, 1986.
(21) Sheldrick, G. SHELXL-97: Program for Crystal Structure Refinement;
Institu¨t fu¨r Anorganische Chemie, Universita¨t Gottingen: Gottingen,
Germany, 1997.
-
-
the solid form as its BF4 salt, the PF6 salt was prepared by ion
exchange reaction of {(µ-SMe)(µ-pdt)[Fe(CO)2(PMe3)]2+}[BF4
]
-
(17) Smallcombe, S. H.; Caserio, M. C.J. Am. Chem. Soc. 1971, 93, 5826.
(22) SHELXTL, version 5.1 or later; Bruker: Madison, WI, 1998.
Inorganic Chemistry, Vol. 42, No. 8, 2003 2491

Georgakaki et al.
flattening of the Fe2(pdt-S)2 core without significant change
in the S‚‚‚S cross-ring distance. The FeS2C3 metallodithio-
cyclohexane rings are maintained in the typical chair/boat
configuration with normal distances. An additional conse-
quence of the four-electron-donor µ-SMe- vs the two-
electron-donor µ-H- bridge is the increased octahedral
character of the FeS3(CO)2PMe3 coordination spheres. As
measured by the displacement of the Fe from the (pdt-S2)-
(CO)(P) plane, the value is ∼0.38 Å in complex 1, which
has the two electrons of the Fe-Fe bond in the sixth
coordination position; it decreases to 0.23 Å in 1-H+ and
decreases further in 1-SMe+ to 0.15 Å.
Activity of 1-SMe+ as an H/D Exchange Catalyst.
Reactivity studies of 1-SMe+ were based on the experimental
protocol used to explore the H2 activation and H/D exchange
capabilities of 1-H+ and analogues.10,15 To establish whether
the µ-SMe moiety might exchange with D- derived from
D2, a medium-pressure NMR tube containing [1-SMe+]-
[PF6-] in CH2Cl2 was pressurized with 10 bar D2 and exposed
to sunlight as described in the Experimental Section. The
2H NMR monitor found no (µ-D)Fe2+ or MeSD formed (eq
1). In the same time period, significant amounts of H/D
exchange was observed for 1-H+/D2 mixtures.10,15 Similarly
to the experiment with 1-H+, H2/D2 gaseous mixtures were
introduced into an anhydrous CD2Cl2 solution of 1-SMe+.
In the absence of light, as well as with extended periods
of photolysis (sunlight for 8 days or more), no HD was
Figure 3. Thermal ellipsoid representation (30% probability) of the
molecular structure of 1-SMe+.
Table 1. Selected Metric Data for 1, 1-H+, and 1-SMe+
1
1-H+
1-SMe+
Fe‚‚‚Fe
2.555(2)
3.026
2.5784(8)
3.064
3.109
2.976
S‚‚‚S
Fe-P
2.234(3)
1.772(9)
1.742(10)
2.254(2)
N/A
69.06(8)
84.34(11)
0.376
2.2523(12)
1.779(4)
1.778(4)
2.2717(11)
N/A
69.15(4)
84.77(4)
0.231
2.257(4)
1.774(11)
1.764(10)
2.324(4)
2.304(4)
84.0(2)
79.50(14)
0.153
Fe-COap
Fe-COba
Fe-S(pdt-br)a
Fe-SMe
Fe-S-Fea
S-Fe-Sa
Fe dspb
a Average of all equivalent bonds and angles. b Fe dsp ) average
displacement of the Fe atoms out of the best planes defined by the two S’s
of the pdt bridge, C of basal CO, and P atoms.
1
observed in the H NMR spectrum (eq 2). Under the same
conditions, 1-H+ showed extensive scrambling of isotopes
(formation of HD from H2 and D2 as well as formation of
1-D+).10,15
+
-
formation of {(µ-SMe)(µ-pdt)[Fe(CO)2(PMe3)]2 }[BF4 ],
-
1-SMe+. Crystals of [1-SMe+][BF4 ] suitable for X-ray
structure determination were grown in CH2Cl2 solution
layered with Et2O at -5 °C. The descriptions of 1-SMe+,
as a stick drawing Figure 2 and a thermal ellipsoid plot in
Figure 3, display the molecular structure as a bioctahedron,
face-bridged by three thiolates.
For the neutral precursor 1 and other (µ-SRS)[FeI(CO)2L]2
complexes, which are described as edge-bridged bisquare
pyramids, the positions of the L ligands are defined as apical,
basal, cisoid, and transoid. Although the oxidative addition
products 1-H+ and 1-SMe+ are face-bridged bioctahedra, we
have retained the designations for the L positions from the
neutral precursor. As defined by crystallography, the repo-
sitioning of the PMe3 ligands from the basal/basal transoid
conformation in 1 and 1-H+10 to basal/basal cisoid in 1-SMe+
minimizes steric interactions of the PMe3 ligands with the
SMe bridge. The shift in the ν(CO) IR spectral values as 1
is converted to 1-SMe+ is similar to that for 1-H+, i.e.,
approximately +70 cm-1 (Figure 2). The change in pattern,
i.e., the splitting of the high-frequency band into two, is
evidence of the change in symmetry described above.
Key structural metric data for 1, 1-H+, and 1-SMe+ are
listed in Table 1. The dramatic increase in the Fe‚‚‚Fe
distance of ca. 0.5 Å with the four-electron donor SMe- as
the bridge has several consequences. The average of the Fe-
S-Fe angle increases to 84.0(2)° in 1-SMe+, from the
average of 69° in the parent 1 and 1-H+, resulting in a
In contrast, mixtures of D2 and H2O show substantial H/D
exchange with both 1-SMe+ and 1-H+ as catalysts (eq 3).
A sample containing 1-SMe+ and 2 µL of H2O in CH2Cl2
was pressurized with 10 bar D2 and exposed to sunlight. To
follow the formation of HOD, 2H NMR spectra were
recorded after 1, 2, and 4 h of exposure (Figure 4). After 1
h of photolysis, the spectrum showed a resonance at 1.75
ppm corresponding to the dissolved D-enriched water. Its
intensity was 0.75 relative to the natural-abundance deute-
rium in the solvent, CH2Cl2, which appears at 5.32 ppm.
Within 4 h of exposure to sunlight, this resonance shifted
slightly to 1.80 ppm, and its intensity increased to 8.53 times
that of the solvent peak, indicating significant exchange
between D2 and H2O in the presence of 1-SMe+.
Conclusions
The conclusions and mechanistic implications of the above
study are summarized as follows. The bioctahedral FeIIFeII
complex, 1-SMe+, face-bridged by three thiolates, (µ-SMe)-
2492 Inorganic Chemistry, Vol. 42, No. 8, 2003

Functional Models of the Iron Hydrogenase ActiWe Site
Scheme 1
SR)2FeIIL4 complexes, where the FeII‚‚‚FeII distance expands
to ∼3.5 Å.23
For both 1-H+ and 1-SMe+, an open site is required, as
indicated by the need of photolysis for H2 activation by the
external base, H2O. The creation of an open site in these
complexes is consistent with the lability of the CO’s indicated
by 13CO/CO exchange under photolysis by sunlight for both
1-H+ and 1-SMe+. A compatible mechanism is presented
in Scheme 1, where the open site is depicted trans to the
µ-SMe, in analogy with the open site in the H2ase binuclear
active site. Formation of the (η2-H2)-FeII interaction is well-
precedented, as is the enhanced acidity of the metal-bound
H2.16
Thus, the two binuclear model complexes 1-H+ and
1-SMe+ mimic the H/D exchange activity of the [Fe]H2ase
enzyme in that both demonstrate H2 (or D2) uptake and
heterolytic cleavage by D2O (or H2O) under photolytic
conditions. The singular requirement for these processes is
an open site on an FeII center. A built-in hydride is not needed
for H/D exchange reactivity in D2/H2O or in H2/D2 mixtures
in the presence of water; it is needed for anhydrous
conditions where the mechanism of H2 activation calls upon
cooperation between the two metal sites in true binuclearity.
We propose that the 1-SMe+ model localizes reactivity on
one Fe center and engages the second iron only for thiolate
ligand modification, maintaining the low-spin FeII, d6 con-
figuration conducive to η2-H2 binding. Thus, the reactivity
scenario and structural design that is reasonable for the
enzyme active site is an attractive archetype for the mech-
anism for the model complexes.
Both [Fe]H2ase and [NiFe]H2ase appear to have been
designed, protein engineered, to have an open site on iron
in their catalytic active sites. Whether both classes of
metalloenzymes follow the same mechanism of H2 uptake
on FeII and cleavage or H2 formation and evolution from an
(η2-H2)FeII species is not at all clear. Convincing experiments
by Sellmann, Geipel, and Moll24 demonstrated the possibility
of D2 activation at a nickel(II) thiolate under anhydrous
conditions, resulting in D+/D- cleavage and formation of
Ni-D and nickel-bound RSD in a mononuclear nickel
Figure 4. 2H NMR spectra showing the formation of HOD (δ ) 1.70-
-
1.80 ppm) in a CH2Cl2 solution containing 1-SMe+ as the PF6 salt, 10
bar D2, and 2 µL of H2O: (a) before exposure to sunlight, (b) after 2 h of
photolysis, and (c) after 4 h of photolysis. Relative ratio of (natural
abundance, δ ) 5.32 ppm) CHDCl2 and HOD in parentheses.
(µ-pdt), maintains the H/D exchange capability in D2/H2O
mixtures that was detected for the analogous complex 1-H+,
face-bridged by a hydride and two thiolates. However,
1-SMe+ loses the capability for catalysis of H/D exchange
in H2/D2 mixtures in anhydrous solution as was demonstrated
for 1-H+.10,15 Our mechanistic proposal for the latter called
upon an open site created by H- shift or CO loss, as
discussed earlier, and deprotonation of (η2-H2)-FeII by the
internal H- base. Whether the η2-H2 and the terminal Fe-H
exist on the same FeII or on adjacent atoms (binuclear
activation) depends on whether a CO loss event is concomi-
tant with µ-H f t-H conversion. However, such bridge
breakage is either excluded or nonproductive in the case of
1-SMe+, as conversion of µ-SMe to µ-H was not observed
when 1-SMe+ was pressurized with H2. This is consistent
with the fact that µ-SMe- is a stronger bridging ligand than
µ-H-, suggesting that the open site needed for H2 binding
in 1-SMe+ comes from CO loss. Furthermore the elongated
Fe‚‚‚Fe distance of ∼3.1 Å in 1-SMe+ relative to 2.6 Å in
1-H+ would prevent involvement of both metals in the H2
activation process. The rigidity of the bidentate chelating
propanedithiolate bridge is expected to prohibit further
elongation of the Fe‚‚‚Fe distance and flattening of the
butterfly Fe2S2 core to a diamond shape, as seen in L4FeII(µ-
(23) Liaw, W.-F.; Lee, N.-H.; Chen, C.-H.; Lee, C.-M.; Lee, G.-H.; Peng,
S.-M. J. Am. Chem. Soc. 2000, 122, 488.
(24) Sellmann, D.; Geipel, F.; Moll, M. Angew. Chem., Int. Ed. 2000, 39,
561.
Inorganic Chemistry, Vol. 42, No. 8, 2003 2493

Georgakaki et al.
complex. Would an adjacent FeII center positioned to trap
and hold H2 in close proximity to Ni-SR enhance this
activity? Such a reaction scenario, first expressed by Fon-
tecilla-Camps and co-workers,3c is an attractive possibility
in the enzyme active site.
diffractometer and crystallographic computing system) and
contributions from the R. A. Welch Foundation. We also
thank Dr. Joseph Reibenspies for his assistance in the X-ray
crystal structure analysis.
Supporting Information Available: Molecular structure and
X-ray crystallographic tables for 1-SMe+. This material is available
free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. We acknowledge financial support
from the National Science Foundation (Grants CHE-
9812355, CHE-0111629, and CHE 85-13273 for the X-ray
IC026005+
2494 Inorganic Chemistry, Vol. 42, No. 8, 2003