Synthesis, crystal structure and protonation of the asymmetric iron-only hydrogenase model [Fe2(CO)3(μ-pdt){μ,η2-Ph2PCH2CH2P(Ph)CH2CH2PPh2}] (pdt = SCH2CH2CH2S)

Article abstract of DOI:10.1016/j.inoche.2006.09.005

Heating [Fe₂(CO)₆(μ-pdt)] (pdt = SCH₂CH₂CH₂S) and bis(2-diphenylphosphinoethyl) phenylphosphine (triphos) in toluene gives [Fe₂(CO)₃(μ-pdt){μ,η²-Ph₂PCH₂CH₂P(Ph)CH₂CH₂PPh₂}], the major form of which has been characterised crystallographically. The complex is highly asymmetric; while one iron centre has the expected square-based pyramidal coordination environment, the second is nearer a trigonal bipyramid. In solution at room temperature, four isomers of the major form interconvert via two processes which can be frozen out at low temperature, while protonation affords a bridging hydride complex which appears to be static at all temperatures.

Full text of DOI:10.1016/j.inoche.2006.09.005

Inorganic Chemistry Communications 10 (2007) 66–70
www.elsevier.com/locate/inoche
Synthesis, crystal structure and protonation of the asymmetric iron-only
hydrogenase model [Fe2(CO)3(l-pdt){l,g2-Ph2PCH2CH2P(Ph)
CH2CH2PPh2}] (pdt = SCH2CH2CH2S)
*
Graeme Hogarth , Idris Richards
Department of Chemistry, University College London, 20 Gordon Street, London, WC1H 0AJ, UK
Received 2 August 2006; accepted 10 September 2006
Available online 23 September 2006
Abstract
Heating [Fe2(CO)6(l-pdt)] (pdt = SCH2CH2CH2S) and bis(2-diphenylphosphinoethyl) phenylphosphine (triphos) in toluene gives
[Fe2(CO)3(l-pdt){l,g2-Ph2PCH2CH2P(Ph)CH2CH2PPh2}], the major form of which has been characterised crystallographically. The
complex is highly asymmetric; while one iron centre has the expected square-based pyramidal coordination environment, the second
is nearer a trigonal bipyramid. In solution at room temperature, four isomers of the major form interconvert via two processes which
can be frozen out at low temperature, while protonation affords a bridging hydride complex which appears to be static at all
temperatures.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Iron-only hydrogenase; Dithiolate; Diiron; Triphos; Fluxionality; Hydride
There has been a recent resurgence in interest in the
chemistry of dithiolate-bridged diiron complexes,
[Fe2(CO)6(l-SR)2], due to the realisation that they closely
resemble the two-iron unit of the H-cluster active site of
iron-only hydrogenases [1–5]. In the enzyme a three atom
unit bridges the two sulphur atoms and widely studied
model complexes have been based on complexes of the type
[Fe2(CO)6(l-SCH2XCH2S)] (X = O, NH, CH2) [6–10].
Despite significant advances in our understanding of this
system gleaned from the study of these model complexes,
a number of major limitations still need to be addressed
as identified in a recent theoretical study of the system by
Tye et al. [11]. Notably, a key role of the two-iron unit
of the H-active cluster is to promote the reduction of pro-
tons to dihydrogen, which almost certainly involves proton
binding to the diiron unit. This probably occurs at a single
iron atom although to date no model complex has been
able to replicate this. Hexacarbonyl, [Fe2(CO)6(l-pdt)]
(pdt = SCH2CH2CH2S), is not basic enough to bind a pro-
ton but this shortcoming can be overcome by substituting
two carbonyls for more basic phosphine or cyanide ligands
[12–20]. In almost all cases, however, substitution occurs in
a symmetrical fashion, with each iron atom bearing a single
phosphine or cyanide ligand. From their recent theoretical
work, Hall and co-workers, came to the conclusion that
‘‘asymmetric substitution of strong donor ligands is the
most viable method of making better synthetic diiron com-
plexes that will serve as both structural and functional
models’’ of the active site of iron-only hydrogenase [11].
To date the only reported phosphine complex that fulfils
this criteria is [Fe2(CO)4(l-pdt)(g2-dppm)] (dppm = Ph2-
PCH2PPh2), prepared by ourselves as a minor component
of the thermal reaction of dppm with [Fe2(CO)6(l-pdt)]
[21]. A second major limitation of all current iron-only
hydrogenase models is the unrotated nature of the two
square-planar iron centres. Thus, in models the two
Fe(CO)2L units roughly eclipse one another (unrotated)
while in contrast in the enzyme they are staggered with
*
Corresponding author.
E-mail address: g.hogarth@ucl.ac.uk (G. Hogarth).
1387-7003/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.inoche.2006.09.005

G. Hogarth, I. Richards / Inorganic Chemistry Communications 10 (2007) 66–70
67
one of the carbonyls residing in the area between the two
iron centres (rotated) [11].
two phosphorus atoms bind to one apical and one basal
site, while the across the iron–iron vector a transoid
basal–basal conformation is adopted. The latter is analo-
gous to that found in [Fe2(CO)4(l-pdt)(l-dppm)], but the
chelate unit differs from the basal–basal conformation in
[Fe2(CO)4(l-pdt)(g2-dppm)] [21]. The three iron–phospho-
rus vectors vary only slightly [Fe(1)–P(2) 2.1953(7), Fe(1)–
Our efforts in the area of modelling the active site of the
iron-only hydrogenenase enzyme have been focused on
coordinating bidentate ligands [21,22] in an attempt to
perturb both the steric and electronic nature of the diiron
core. As an extension to this work we identified tridentate
phosphines as potentially useful ligands since they would
provide a more basic iron centre, being capable of strongly
binding a proton, while also leading to an asymmetric
diiron environment. This could possibly lead to significant
structural perturbations perhaps leading to the formation
of a rotated model, while also providing electronic asym-
metry whereby one of the iron atoms is significantly more
basic, potentially leading to the formation of a terminal
hydride. We present here our first efforts in this area utilis-
ing the well-known triphosphine ligand, bis(2-diphenyl-
phosphinoethyl) phenylphosphine (triphos).
˚
P(1) 2.2016(7), Fe(2)–P(3) 2.2109(7) A], suggesting that
there is not a great deal of strain in the binding mode
adopted. Coordination of the triphos ligand renders the
two sulfur atoms of the dithiolate-bridge inequivalent, thus
completely breaking the symmetry of the system and gener-
ating a highly asymmetric diiron centre. Unfortunately,
this does not perturb the relative geometry of the two iron
atoms significantly which are still essentially in an unro-
tated form [11] (Fig. 2). However, while the coordination
geometry around Fe(1) is in line with that of previously
characterised complexes of this type, being best described
as square-based pyramidal, the basal plane being defined
by trans angles of 162.54(3)ꢁ and 157.61(3)ꢁ between
S(1)–C(1) and S(2)–P(2) respectively, that around Fe(2) is
significantly perturbed. Here the geometry is best described
as trigonal bipyramidal with S(2) and P(3) occupying axial
sites [S(2)–Fe(2)–P(3) 164.92(3)ꢁ] and S(1), C(2) and C(3)
lying in the equatorial plane. Indeed, the sum of the angles
at Fe(2) subtended by these three atoms totals 359.98ꢁ.
Heating a toluene solution of [Fe2(CO)6(l-pdt)] [23] and
triphos in toluene for 16 h results in the clean formation of
a single product identified as [Fe2(CO)3(l-pdt){l,g2-Ph2
PCH2CH2P(Ph)CH2CH2PPh2}] (1) [24]. The molecular
structure [25] of the major component 1a (Fig. 1) shows
that the triphos ligand binds in a l,g2-fashion to the diiron
˚
unit, bridging the diiron centre [Fe(1)–Fe(2) 2.5279(5) A]
and chelating to one of the iron atoms, Fe(1). Here the
Fig. 1. Molecular structure of [Fe2(CO)3(l-pdt){l,g2-Ph2PCH2 CH2P(Ph)CH2CH2PPh2}] (1a) with selected bond lengths (A) and angle (ꢁ): Fe(1)–C(1)
1.750(3), Fe(1)–P(2) 2.1953(7), Fe(1)–P(1) 2.2016(7), Fe(1)–S(1) 2.2589(7), Fe(1)–S(2) 2.2644(7), Fe(1)–Fe(2) 2.5279(5), Fe(2)–C(3) 1.760(3), Fe(2)–C(2)
1.761(3), Fe(2)–P(3) 2.2109(7), Fe(2)–S(1) 2.2718(7), Fe(2)–S(2) 2.2918(7), P(1)–Fe(1)–Fe(2) 156.05(2), P(2)–Fe(1)–Fe(2) 101.66(2), P(3)–Fe(2)–S(2)
164.92(3), P(3)–Fe(2)–Fe(1) 109.52(2), C(1)–Fe(1)–Fe(2) 106.46(8), C(2)–Fe(2)–Fe(1) 88.00(8), C(3)–Fe(2)–Fe(1) 154.24(8), Fe(1)–S(1)–Fe(2) 67.83(2),
Fe(1)–S(2)–Fe(2) 67.39(2), P(2)–Fe(1)–P(1) 88.37(3), S(1)–Fe(1)–C(1) 162.54(8), S(1)–Fe(2)–C(2) 140.62(8), S(2)–Fe(1)–P(2) 157.61(3).
˚

68
G. Hogarth, I. Richards / Inorganic Chemistry Communications 10 (2007) 66–70
Ph2
Ph2
P
S
S
S
S
CH2 flip
P
OC
OC
Fe
Fe
Fe
Fe
CO
CO
Ph2P
Ph2P
P
P
CO
CO
Ph
Ph
X-ray
1a
trigonal
rotation
Ph2
Ph2
P
S
S
S
S
P
OC
OC
Ph2P
OC
OC
Fe
Fe
Fe
Fe
CO
CO
P
P
Ph2P
Ph
Ph
Fig. 3. Suggested structures for the interconverting forms of 1a.
more contentious. We suggest a structure in which the two
phosphorus atoms at Fe(1) bind in a basal–basal fashion.
The temperature independent nature of the 31P NMR spec-
trum of 1b suggests that the fluxional processes proposed
to occur in 1a are either significantly faster or slower in this
isomer. The reduced steric interactions between the phos-
phine and pdt ligand in this form could account for a lower
activation barrier to the methylene flip process, while the
trigonal twist at the Fe(CO)2L centre would be expected
to have a higher activation barrier as this would lead to
two PPh2 end groups becoming co-planar. We have been
unable to separate 1a and 1b and note that upon dissolu-
tion of crystals prepared for X-ray crystallography both
isomers were present in approximately the same ratio as
seen for the uncrystallised material.
Fig. 2. Molecular structure of [Fe2(CO)3(l-pdt){l,g2- Ph2PCH2CH2P-
(Ph)CH2CH2PPh2}] (1a) looking along the iron–iron vector with selected
atom labeling.
In solution the structure of 1 is more complex. At room
temperature the 31P NMR spectrum in CD2Cl2 shows two
sets of three resonances in an approximate 3:1 ratio
assigned to isomers 1a and 1b. The minor set, 1b, are all
sharp and consist of two doublets at 89.7 (J 17.2 Hz) and
61.6 ppm (J 9.2 Hz) and a doublet of doublets at
86.3 ppm. The major set, 1a, which are observed at 83.3,
81.8 and 63.9 ppm are all broad at this temperature and
individual coupling constants cannot be extracted. In each
isomer, we assign the low-field resonances to the phospho-
rus bound to the iron dicarbonyl unit, the intermediate res-
onance to the central phosphorus atom of the triphos
ligand and the high-field resonance to the chelate end of
the triphosphine. Upon cooling resonances associated with
1b do not vary significantly, however, those associated with
1a broaden and collapse until at 183 K they are replaced by
twelve new sharp resonances associated with four separate
isomeric forms [24]. In recent work we have shown that
[Fe2(CO)4(l-pdt)(l-dppm)] undergoes two separate flux-
ional processes; a flipping of the central methylene group
of the pdt ligand which renders the two iron atoms equiv-
alent and a concerted trigonal twist of each Fe(CO)2L unit
which equivalences the thiolate bridges [21]. We propose
that similar processes occur in the major isomer of 1a,
which would lead to the formation of four interconverting
isomers as shown (Fig. 3).
As eluded to in the introduction, a key feature of
iron-only hydrogenase models is their ability to bind a
proton, preferably in a terminal fashion at a single iron
atom and we hoped that the asymmetry in 1 may allow
the latter. Protonation of 1 by HBF4 Æ Et2O in dichloro-
methane is clean and rapid, being easily followed by IR
spectroscopy. A red solution of 1 turned cloudy and pink
upon addition of acid initial carbonyl absorptions at
1947s and 1889vs cmÀ1 being replaced by new peaks at
2039vs, 1986s and 1964s cmÀ1 assigned to 2 [26]; the
average shift to higher wavenumbers of 88 cmÀ1 being
consistent with protonation at the diiron centre. The salt
produced was stable in air, no further change being
noted even after sitting for 3 d. In the 31P NMR spec-
trum of 2, a single set of three resonances was observed
at 88.9, 85.2 and 46.4 ppm suggesting that both 1a and
1b protonate to afford the same product, while upon
1
cooling no significant changes were observed. In the H
NMR spectrum a doublet of triplets at d-13.49 (J 29.2,
22.4 Hz) is assigned to the new hydride. Both the posi-
tion of the resonance and the observation that it couples
to all three phosphorus nuclei suggest that it bridges the
diiron centre. The significant high-field shift (ca.
The nature of the minor component 1b, which shows an
essentially temperature independent 31P NMR spectrum, is

G. Hogarth, I. Richards / Inorganic Chemistry Communications 10 (2007) 66–70
69
X
X
Cys
X
S
S
OC
CO
CO
S
S
S
S
[4Fe4S]
S
Fe
Fe
Fe
Fe
Fe
Fe
OC
OC
OC
OC
CO
CO
CO
CO
OC
NC
CN
CO
CO
CO
unrotated
rotated
S
Ph2
S
S
S
S
CO
PPh2
OC
S
triphos, 110oC
- 3 CO
OC
CO
CO
P
OC
Ph2P
Fe
Fe
Fe
Fe
CO
Fe
CO
Fe
+
OC
OC
Ph2P
CO
P
CO
P
Ph
Ph
1a
1b
Ph2
P
Ph2
S
S
S
S
P
OC
OC
HBF4.Et2O
Fe
Fe
Fe
CO
Fe
CO
Ph
CO
H
Ph2P
Ph2P
P
P
CO
Ph
1a-b
2
17.5 ppm) of the resonance associated with the chelate
end of the triphosphine suggests that at least one end
of the triphos ligand has been significantly perturbed,
while in contrast the other two phosphorus atoms change
only slightly. Unfortunately as yet we have been unable
to grow crystals suitable for X-ray diffraction and thus
the precise structural arrangement of the ligands in this
species remains unclear.
html, or from the Cambridge Crystallographic Data Cen-
tre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44
1223 336 033; or e-mail: deposit@ccdc.cam.ac.uk. Supple-
mentary data associated with this article can be found, in
the online version, at doi:10.1016/j.inoche. 2006.09.005.
References
In conclusion, the facile preparation of [Fe2(CO)3(l-pdt)
{l,g2-Ph2PCH2CH2P(Ph)CH2CH2PPh2}] suggests that
asymmetric diiron l-pdt complexes are readily accessible
and potentially serve as more realistic structural and func-
tional models of the active site of iron-only hydrogenase.
We are currently working towards the preparation of vari-
ants using asymmetric bi- and tridentate phosphine ligands
with the aim of perturbing both the steric and electronic
asymmetry of the diiron centre to an even greater extent
with the aim of preparing a terminal hydride complex.
These results will be reported in due course.
[1] Y. Nicolet, C. Piras, P. Legrand, C.E. Hatchikian, J.C. Fontecilla-
Camps, Structure 7 (1999) 13.
[2] J.W. Peters, W.N. Lanzilotta, B. Lemon, L.C. Seefeldt, Science 282
(1998) 1853.
[3] B.J. Lemon, J.W. Peters, Biochemistry 38 (1999) 12969.
[4] Y. Nicolet, A.L. De Lacy, X. Verne´de, V.M. Fernandez, E.C.
Hatchikian, J.C. Fontecilla-Camps, J. Am. Chem. Soc. 123 (2001)
1596.
[5] H.-J. Fan, M.B. Hall, J. Am. Chem. Soc. 123 (2001) 3828.
[6] I.P. Georgakaki, L.M. Thomson, E.J. Lyon, M.B. Hall, M.Y.
Darensbourg, Coord. Chem. Rev. 238–239 (2003) 255.
[7] D.J. Evans, C.J. Pickett, Chem. Soc. Rev. 32 (2003) 268.
[8] T.B. Rauchfuss, Inorg. Chem. 43 (2004) 14.
˚
[9] L. Sun, B. Akermark, S. Ott, Coord. Chem. Rev. 249 (2005) 1653.
Acknowledgement
[10] X. Liu, S.K. Ibrahim, C. Tard, C.J. Pickett, Coord. Chem. Rev. 249
(2005) 1641.
[11] J.W. Tye, M.Y. Darensbourg, M.B. Hall, J. Am. Chem. Soc. 45
(2006) 552.
[12] E.J. Lyon, I.P. Georgakaki, J.H. Reibenspies, M.Y. Darensbourg,
Angew. Chem., Int. Ed. Engl. 38 (1999) 3178.
We thank Dr Abil Aliev for help with variable temper-
ature 31P{1H} NMR experiments.
[13] E.J. Lyon, I.P. Georgakaki, J.H. Reibenspies, M.Y. Darensbourg, J.
Am. Chem. Soc. 123 (2001) 3268.
Appendix A. Supplementary material
[14] X. Zhao, I.P. Georgakaki, M.L. Miller, J.C. Yarbrough, M.Y.
Darensbourg, J. Am. Chem. Soc. 123 (2001) 9710.
[15] X. Zhao, I.P. Georgakaki, M.L. Miller, R. Mejia-Rodriguez, C.-Y.
Chiang, M.Y. Darensbourg, Inorg. Chem. 41 (2002) 3917.
CCDC 614638 contains the supplementary crystallo-
graphic data for 1a. These data can be obtained free of
charge via htpp://www.ccdc.cam.ac.uk/conts/retrieving.

70
G. Hogarth, I. Richards / Inorganic Chemistry Communications 10 (2007) 66–70
[16] D. Chong, I.P. Georgakaki, R. Mejia-Rodriguez, J. Sanabria-
Chinchilla, M.P. Soriaga, M.Y. Darensbourg, Dalton Trans. (2003)
4158.
[17] R. Mejia-Rodriguez, D. Chong, J.H. Reibenspies, M.P. Soriaga,
M.Y. Darensbourg, J. Am. Chem. Soc. 126 (2004) 12004.
[18] F. Gloaguen, J.D. Lawrence, T.B. Rauchfuss, J. Am. Chem. Soc. 123
(2001) 9476.
[19] F. Gloaguen, J.D. Lawrence, T.B. Rauchfuss, M. Be´nard, M.-M.
Rohmer, Inorg. Chem. 41 (2002) 6573.
[20] J. Nehring, D.M. Heinekey, Inorg. Chem. 42 (2003) 4288.
[21] G. Hogarth, I. Richards, J. Organomet. Chem, submitted for
publication.
1b:1aI:1aII:1aIII:1aIV ca 1:1:0.7:0.3:0.5; Anal. calc. for Fe2P3-
S2O3C40H39: C, 57.42, H, 4.67; Found C, 57.54, H, 3.86.
[25] Crystallographic data for 1a Æ CH2Cl2 – Single crystals were mounted
on a glass fibre and all geometric and intensity data were taken from
these samples using a Bruker SMART APEX CCD diffractometer
˚
with graphite-monochromated Mo-Ka radiation (k = 0.71073 A) at
150 2 K. Data reduction was carried out with SAINT PLUS and
absorption correction applied using the programme SADABS.
Structures were solved by direct methods and developed using
alternating cycles of least-squares refinement and difference-Fourier
synthesis. All non-hydrogen atoms were refined anisotropically and
hydrogen atoms were generally placed in calculated positions (riding
model). Structure solution used SHELXTL PLUS V6.10 program
package. Orange block, dimensions 0.16 · 0.15 · 0.04 mm, triclinic,
[22] G. Hogarth, I. Richards, unpublished results.
[23] D. Seyferth, G.B. Womack, M.K. Gallagher, M. Cowie, B.W.
Hames, J.P. Fackler, A.M. Mazany, Organometallics 6 (1987) 283.
ꢀ
˚
˚
˚
3
space group P1, a = 9.4439(8) A, b = 9.6258(9) A, c = 22.437(2) A,
1
[24] IR m(CO) (CH2Cl2): 1947s, 1889vs cmÀ1; H NMR (CD2Cl2, 298 K):
a = 99.783(1)ꢁ, b = 95.331(2)ꢁ, c = 94.278(2)ꢁ, V = 1992.8(3) A ,
˚
d 8.03–7.16 (m, 25H, Ph), 3.04–2.41 (m, 8H, PCH2), 2.36 (br, 2H,
SCH2), 2.29 (br, 2H, SCH2), 2.14 (br, 2H, CH2); 31P{1H} NMR
(CD2Cl2, 298 K): d 89.7 (d, J 17.2, 1b), 86.3 (dd, J 17.2, 9.2, 1b), 83.3
(br, 1a), 81.8 (brd, J 23.4, 1a), 63.9 (br, 1a), 61.6 (d, J 9.2, 1b); (183 K)
d 91.74 (d, J 29.1, 1aI), 88.34 (d, J 28.1, 1aII), 87.78 (d, J 16.3, 1b),
86.11 (dd, J 16.3, 9.2, 1b), 83.52 (dd, 27.7, 9.4, 1aI), 80.70 (dd, m,
1aII+III), 77.48 (d, J 25.0, 1aIII), 76.90 (dd, J 25.3, 6.5, 1aIV), 74.42 (d, J
6.6, 1aIV), 71.99 (d, J 6.0, 1aIII), 70.64 (d, J 25.0, 1aIV), 64.88 (d, J 9.2,
Z = 2, F(000) 948, dcalc = 1.536 g cmÀ3 l = 1.127 mmÀ1
, . 17384
reflections were collected, 9087 unique [R(int) = 0.0228] of which
7520 were observed [I > 2.0r(I)]. At convergence, R1 = 0.0405,
wR2 = 0.0950 [I > 2.0r(I)] and R1 = 0. 0510, wR2 = 0.0998 (all data),
for 478 parameters.
[26] IR m(CO) (CH2Cl2): 2039vs, 1986s, 1964s cmÀ1 1H NMR (CD2Cl2,
;
298 K): d 8.01–7.22 (m, 25H, Ph), 3.07–2.53 (m, 8H, PCH2), 2.31 (br,
4H, SCH2), 2.01 (br, 2H, SCH2), À13.49 (dt, J 29.2, 22.4, 1H, l-H);
31P{1H} NMR (CD2Cl2, 298 K): d 88.9 (s), 85.3 (s), 46.4 (s).
1b), 61.44 (d,
J
9.8, 1aII), 57.66 (d,
J
9.8, 1aI), ratio