Novel alkoxide cluster topologies featuring rare seesaw geometry at transition metal centers

Article abstract of DOI:10.1002/chem.201302558

Caution! Chemists playing: Novel clusters of the form [M₂Li ₂Cl₂(OR)₄] featuring rare seesaw geometry at the transition metal centers were synthesized for M=Cr-Co. The use of sterically hindering alkoxide ligand

Full text of DOI:10.1002/chem.201302558

COMMUNICATION
DOI: 10.1002/chem.201302558
Novel Alkoxide Cluster Topologies Featuring Rare Seesaw Geometry
at Transition Metal Centers
James A. Bellow,[a] Dong Fang,[a] Natalija Kovacevic,[a] Philip D. Martin,[a]
Jason Shearer,*[b] G. Andrꢀs Cisneros,*[a] and Stanislav Groysman*[a]
Molecules with central atoms possessing a coordination
number of four are commonplace in coordination chemis-
try.[1] The majority of four-coordinate compounds display
either tetrahedral or square planar geometry. However, va-
lence shell electron pair repulsion (VSEPR) theory predicts
a third type of four-coordinate geometry—the seesaw, or
sawhorse. Two possible cases of seesaw geometry can be dis-
tinguished—the cis-divacant octahedron (angles of 1808 and
908) and the monovacant trigonal bipyramid (angles of 1808,
1208, and 908).[2] Examples of seesaw geometry can be found
in main-group compounds due to the repulsive presence of
lone pairs.[3] Seesaw transition metal complexes are of inter-
est, as they possess a vacant coordination site that enables
substrate coordination and activation.[2] However, seesaw
transition metal complexes are extremely rare, and the re-
quirements toward their attainment are poorly under-
stood.[4–9] Recent computational work by Alvarez and co-
workers supports the plausibility of seesaw geometry for
first-row transition elements.[10] In particular, it has been
proposed that seesaw geometry at transition metal centers
arises primarily for certain electron configurations, particu-
larly for d6 and closed-shell d10 metal centers.[2,11] Herein we
low-coordinate complexes containing a plethora of different
transition metals, many of which have shown small molecule
activation capabilities.[16–26] The majority of previously re-
ported bulky alkoxide ligands contain three identical sub-
stituents. We combined two tert-butyl groups with a phenyl
group, attempting to achieve directionality in the alkoxide
coordination, thus inducing higher crystallinity. The ligand
[OCACTHUNTRGNENG(U tBu)2Ph] ([OR] henceforth) was prepared by the addi-
tion of phenyllithium to hexamethylacetone. The structure
of the lithium salt of the ligand has been confirmed by X-
ray structure determination. The ligand crystallizes as the
trimer Li3(OR)3 (1; see the Supporting Information for the
crystal structure) from hexane, with two-coordinate geome-
try featured at the lithium centers.
Addition of two equivalents of 1 to a solution of iron(II)
chloride forms transiently the respective three-coordinate
bis
ACHTUNGTRENNUNG
structurally similar to a chromium bisACHTUNGTRENNNUG
previously reported by Power and co-workers.[27] Compound
2 is unstable in the absence of a coordinating solvent
(THF).[28] Dissolution in hexanes and solvent removal under
vacuum leads to the dissociation of THF ligands and subse-
quent dimerization, forming compound 3 (Scheme 1). Fol-
lowing similar protocols, the reactions of manganese(II)
chloride, cobalt(II) chloride, and chromium(II) chloride
with 1 form complexes 4–6, respectively. Compounds 3–6
were characterized by X-ray crystallography. X-ray quality
crystals of 3–6 were obtained upon recrystallization from
hexanes at ꢀ358C. The structures of compounds 3 and 4 are
depicted in Figure 1a and b; the structures of 2, 5, and 6 can
be found in the Supporting Information. Complexes 3–6 are
all isostructural clusters featuring an M2Li2Cl2(OR)4 core.[29]
Each chloride is bridging two transition metals. The lithium
centers are two-coordinate as in the structure of 1, whereas
the transition metals are four-coordinate. A Crystal Struc-
ture Database search demonstrates that such a topology is
unprecedented.[30] Further confirmation of the identical con-
nectivity pattern in these structures comes from the virtually
indistinguishable IR spectra of these complexes (Figure S8
in the Supporting Information).
demonstrate the first example in which the same seesaw ge-
II
ꢀ
ometry is observed for the series of transition metals (Cr
CoII) featuring varying electron configurations (d4–d7). This
geometry is enabled by the unique combination of the steric
bulk of the ancillary ligands, alkoxides, combined with the
inclusion of lithium chloride in the structure. Density func-
tional theory (DFT) calculations shed light on the bonding
interactions in these complexes and give insight into their
electronic structures.
The properties and the reactivity of alkoxide-ligated
metal centers depend on the steric and electronic properties
of the alkoxide.[12–15] Bulky ligands have been employed in
[a] J. A. Bellow, D. Fang, N. Kovacevic, Dr. P. D. Martin,
Prof. G. A. Cisneros, Prof. S. Groysman
Wayne State University, Detroit, MI 48202 (USA)
E-mail: andres@chem.wayne.edu
groysman@chem.wayne.edu
[b] Prof. J. Shearer
Most significantly, dimeric complexes 3–6 all possess
seesaw geometry at transition metal centers. The space-fill-
ing model of 3 (Figure 1c) shows the bulky tert-butyl and
phenyl groups on the alkoxides forming a tight cage around
the complex. Thus, we propose that the seesaw geometry
Department of Chemistry, University of Nevada
Reno, NV 89557 (USA)
E-mail: shearer@unr.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201302558.
Chem. Eur. J. 2013, 19, 12225 – 12228
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12225

absorption spectroscopy. The
XANES region of the X-ray
absorption spectra (XAS) all
display pre-edge features cor-
responding to nominal 1s!3d
transitions. The pre-edge fea-
ture of 3 has a peak-area of
8.2(1) %eV relative to the
edge height (Figure 2a). Al-
though the geometry about the
FeII center is unprecedented,
we can think of the FeII center
as contained in a pseudotrigo-
nal bipyramidal structure. We
find that the area under this
feature compares well with
other 5-coordinate FeII com-
plexes, and is thus consistent
with the seesaw geometry.[32]
High resolution (k=2.0–
17.5 ꢁꢀ1) EXAFS analysis of
the XAS is also consistent with
the crystallographic structure
of 3. In addition to the two
Scheme 1. Synthesis of compounds 2–6.
ꢀ
ꢀ
Fe O (1.89 ꢁ), two Fe Cl
ꢀ
(2.48 ꢁ) and one Fe Fe
(3.60 ꢁ) vectors, there were
also
a plethora of multiple
scattering (MS) pathways that
needed to be accounted for to
properly solve the EXAFS
(Figure 2b). These MS path-
ways are strongly dependent
on the O-Fe-O and Cl-Fe-Cl
bond angles. A best fit to the
data yields an O-Fe-O bond
angle of 1748 and a Cl-Fe-Cl
bond angle of 868. Thus, the
results are consistent with the
crystallographic data. Analysis
of the XAS for complexes 4
and 5 yield similar fits to the
data (see the Supporting Infor-
mation).
Figure 1. a) Crystal structure of 3. b) Crystal structure of 4. c) Space-filling model of 3. d) Structure of seesaw
core with O-M-O and Cl-M-Cl angles. For the crystal structures of 3 and 4, 50% probability ellipsoids are
ꢀ
shown. Hydrogen atoms are omitted for clarity. Selected bond lengths [ꢁ] and angles [8] for 3: Fe1 O1
ꢀ
ꢀ
ꢀ
1.904(4), Fe1 O2 1.917(4), Fe1 Cl1 2.474(2), Fe1 Cl2 2.520(2), O1-Fe1-O2 173.8(2), Cl1-Fe1-Cl2 86.96(8)8.
arises in part due to the steric bulk of the alkoxide ligands.
The immediate coordination environment of the transition
metals is presented in Figure 1d. The O-M-O angles of the
complexes range from 1718 to 1768, and the Cl-M-Cl angles
Continuous symmetry measurements provide a useful
handle to evaluate the geometry of the metal center.[33,34]
The continuous symmetry values have been calculated using
the method detailed by Cicera, Alemany, and Alvarez in the
Shape2 software.[2] The continuous symmetry values for the
crystal (computationally optimized) structures (3–6) are 2.2
(2.6), 2.2 (2.4), 2.0 (2.9) and 2.1 (2.5), respectively. As the
continuous symmetry values range from 0 to 100 (0 means
perfect seesaw, whereas 100 means totally distorted), both
crystal and computational structures present geometry close
to that of a perfect seesaw.
ꢀ
are all within the 848–888 range. The M Cl distances in each
complex are similar to those in related complexes, confirm-
[22,27,31]
ꢀ
ing that there are M Cl bonds in these structures.
In
ꢀ
addition, the Li Cl distances range from 2.5–3.0 ꢁ, which
are longer than those for Li Cl bonds in related struc-
ꢀ
tures.[22,27,31]
To confirm that the structures obtained by single crystal
diffraction analysis are representative of the bulk material,
complexes 3, 4 and 5 were subjected to metal K-edge X-ray
To shed light on the bonding and the electronic structure
of these unusual molecules, we conducted DFT calculations
12226
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 12225 – 12228

Novel Alkoxide Cluster Topologies
COMMUNICATION
Figure 3. NCI surfaces for complex 6, the isovalue is 0.5. a) The opti-
mized structure with tert-butyl was replaced by methyl (only the core is
shown), ꢀ0.07<sign(l2)1<0.07 a.u. b) NCI analysis for the complete
(unoptimized) crystal structure, ꢀ0.02<sign(l2)1<0.02 a.u. Regarding
the chosen arbitrary color code, red surface indicates strong repulsion;
blue surface indicates strong attraction; green surface indicates relatively
weak interactions. Different sign(l2)1 ranges may be used to distinguish
the strength of the interactions. Note that one can compare the relative
strength of the interactions based on the colors only when using the same
sign(l2)1 range.
Figure 2. a) XANES region of the Fe K-edge XAS for compound 3.
b) The k3(c) plots of the experimental data, best fit, deconvolved scatter-
ing pathways and residual for 3.
(see the Supporting Information for Computational Meth-
ods) on the simplified models of 3–6 (tBu groups replaced
by Me groups). The superposition of the calculated and crys-
tal structures shows good agreement between them (Fig-
ure S16 in the Supporting Information).[35] Combined elec-
tron localization function (ELF)[36]/noncovalent interaction
(NCI)[37,38] analysis[39–41] was performed to gain a better un-
derstanding of interatomic interactions (see the Supporting
Information for ELF results). NCI describes interatomic in-
teractions by peaks of the reduced electron density gradient
at low electron densities. Taking 6 as an example, the strong
the notable interactions include a weak attraction between
Cr and ortho-H (circled in red), and a weak attraction be-
tween Cl and H (tBu) atoms (circled in blue). Similar inter-
actions are also present in the optimized simplified struc-
tures bearing Me substituents (Figure S18 in the Supporting
Information). However, the full structure also demonstrates
a weak attraction between the tBu groups and neighboring
phenyl group (circled in violet). This interaction may be re-
sponsible for the formation of the cage-like structure (Fig-
ure 1c).
ꢀ
ꢀ
ꢀ
metal ligand (Cr O and Cr Cl) interaction is evidenced by
the blue NCI surfaces in Figure 3a. In addition, the surface
ꢀ
for Cr O (circled in black) has a darker blue color than the
ꢀ
surface for Cr Cl (circled in red), indicating that the inter-
action between Cr and Cl is slightly weaker than between
Cr and O. Moreover, the interaction between Li and Cl is
much weaker than the one between Cr and Cl as demon-
strated by the green surfaces (circled in yellow). Next, we
replaced the alkyl substituents R of the OR groups by hy-
drogen atoms to investigate their effect on the stability of
the seesaw geometry in the core. Although the optimized Cr
complex did not retain the seesaw geometry, the other
We carried out a preliminary investigation of the magnet-
ic properties of 3–6. To probe the solution state effective
magnetic moments of 3–6, the Evans method was conducted
at room temperature.[42] The largest effective magnetic
moment observed was that of 4, which was observed to be
11.6 mB. The expected spin-only magnetic moment for two
uncoupled MnII centers is 11.0 mB. The calculated and ob-
served values were also similar for complexes 3 (8.8 mB,
calcd 8.9 mB), 5 (7.8 mB, calcd 6.9 mB), and 6 (8.4 mB, calcd
8.9 mB). We have also conducted SQUID magnetometry on
compound 3, featuring the largest number of unpaired elec-
trons (Figure S5 in the Supporting Information). SQUID
data are fully consistent with solution measurements. At
room temperature, a magnetic moment of 11.6 mB (cT=
16.8 cm3 (molK)ꢀ1 is observed, indicating two uncoupled
ꢀ
(Mn Co) complexes were found to only undergo slight dis-
tortion (Figure S20 in the Supporting Information). These
ꢀ
results underscore the importance of lithium oxygen inter-
actions (circled in blue) in these structures for the stabiliza-
tion of the seesaw core. To determine the steric effect of the
[tBu2(Ph)CO] ligand, we carried out the NCI analysis on the
full crystal structure of 6 (unoptimized, Figure 3b). Some of
Chem. Eur. J. 2013, 19, 12225 – 12228
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12227

J. Shearer, G. A. Cisneros, S. Groysman et al.
[18] D. R. Neithamer, R. E. LaPointe, R. A. Wheeler, D. S. Richeson,
G. D. Van Duyne, P. T. Wolczanski, J. Am. Chem. Soc. 1989, 111,
9056–9072.
[19] J. J. Curley, A. F. Cozzolino, C. C. Cummins, Dalton Trans. 2011, 40,
2429–2432.
metal centers. At low temperatures, cT values drop to about
0, indicating antiferromagnetic coupling.
In summary, we have synthesized novel clusters supported
by bulky alkoxide ligands featuring a M2Li2Cl2(OR)4 core
ꢀ
(M=Cr Co). These clusters exhibit the highly unusual
seesaw geometry at the transition metal centers. Our current
focus is on the synthesis of neutral two-coordinate bis-
[20] C. Schwarzmaier, A. Noor, G. Glatz, M. Zabel, A. Y. Timoshkin,
B. M. Cossairt, C. C. Cummins, R. Kempe, M. Scheer, Angew.
Chem. 2011, 123, 7421–7424; Angew. Chem. Int. Ed. 2011, 50, 7283–
7286.
[21] B. D. Murray, P. P. Power, J. Am. Chem. Soc. 1984, 106, 7011–7015.
[22] M. M. Olmstead, P. P. Power, G. Sigel, Inorg. Chem. 1986, 25, 1027–
1033.
[23] R. A. Bartlett, J. J. Ellison, P. P. Power, S. C. Shoner, Inorg. Chem.
1991, 30, 2888–2894.
[24] M. M. Rodriguez, E. Bill, W. W. Brennessel, P. L. Holland, Science
2011, 334, 780–783.
ACHTUNGTRENNUNG(alkoxide) complexes using these dimers as precursors and
the investigation of their reactivity.
CCDC-939862 (1), 939863 (2), 939864 (3), 939865 (4),
939866 (5), and 939867 (6) contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
[25] S. Groysman, D. Villagrꢂn, D. Freedman, D. G. Nocera, Chem.
Commun. 2011, 47, 10242–10244.
[26] M. B. Chambers, S. Groysman, D. Villagrꢂn, D. G. Nocera, Inorg.
Chem. 2013, 52, 3159–3169.
[27] J. Hvoslef, H. Hope, B. D. Murray, P. P. Power, J. Chem. Soc. Chem.
Commun. 1983, 1438–1439.
[28] We were able to obtain crystals of 2 from a mixture of hexanes and
a small amount of THF. It co-crystallizes with 3. The crystals are un-
stable, rapidly deteriorating even in the absence of air (covered by
Paratone oil).
Acknowledgements
We acknowledge Wayne State University for the start-up funding. The
computing time from Wayne State C&IT is gratefully acknowledged. We
thank Prof. D. Freedman and Michael Graham (Northwestern Universi-
ty) for the help with SQUID measurements. We also acknowledge Prof.
P. L. Holland (Yale University) for the insightful discussions.
[29] The complexes display transition metal/lithium disorder in the core.
The disorder has been satisfactorily modeled so that in a given con-
figuration the transition metals occupy positions trans to each other.
The remaining two positions are occupied by lithium atoms.
[30] Somewhat related topologies are observed, but none replicates all
of the structural features found in 3–6. For selected examples, see:
a) Y. K. Gun’ko, U. Cristmann, V. G. Kessler, Eur. J. Inorg. Chem.
2002, 1029–1031; b) C. E. Anson, W. Klopper, J.-S. Li, L. Ponikiew-
ski, A. Rothenberger, Chem. Eur. J. 2006, 12, 2032–2038; c) D. J.
Gallagher, K. W. Henderson, A. R. Kennedy, C. T. OꢃHara, R. E.
Mulvey, R. B. Rowlings, Chem. Commun. 2002, 376–377; d) K. J.
Drewette, K. W. Henderson, A. R. Kennedy, R. E. Mulvey, C. T.
OꢃHara, R. B. Rowlings, Chem. Commun. 2002, 1176–1177; e) A. R.
Kennedy, J. Klett, R. E. Mulvey, S. Newton, D. S. Wright, Chem.
Commun. 2008, 308–310; f) R. E. Mulvey, Dalton Trans. 2013, 42,
6676–6693.
Keywords: alkoxides
· cluster compounds · continuous
symmetry measures · seesaw geometry · transition metals
[1] N. N. Greenwood, A. Earnshaw, Chemistry of the Elements, 2nd ed.,
Pergamon, Oxford, 1997, pp. 912–914.
[2] J. Cicera, P. Alemany, S. Alvarez, Chem. Eur. J. 2004, 10, 190–207.
[3] P. Atkins, T. Overton, J. Rourke, M. Weller, F. Armstrong, Inorganic
Chemistry, 4th ed., W. H. Freeman and Company, New York, 2006,
pp. 45–48.
[4] W. Baratta, E. Hertweck, P. Rigo, Angew. Chem. 1999, 111, 1733–
1735; Angew. Chem. Int. Ed. 1999, 38, 1629–1631.
[5] D. Huang, W. E. Streib, J. C. Bollinger, K. C. Caulton, R. F. Winter,
T. Scheiring, J. Am. Chem. Soc. 1999, 121, 8087–8097.
[6] F. Bonhomme, K. Yvon, G. Triscone, K. Jansen, G. Auffermann, P.
Mueller, W. Brçnger, P. Fischer, J. Alloys Compd. 1992, 178, 161–
166.
[7] D. M. Lunder, E. B. Lobkovsky, W. E. Streib, K. C. Caulton, J. Am.
Chem. Soc. 1991, 113, 1837–1838.
[8] T. J. Hubin, N. W. Alcock, D. H. Busch, Acta Crystallogr. Sect. C
2000, 56, 37–39.
[9] R. K. Minhas, J. J. H. Edema, S. Gambarotta, A. Meetsma, J. Am.
Chem. Soc. 1993, 115, 6710–6717.
[10] J. Cicera, E. Ruiz, S. Alvarez, Inorg. Chem. 2008, 47, 2871–2889.
[11] M. Olivꢂn, E. Clot, O. Eisenstein, K. C. Caulton, Organometallics
1998, 17, 3091–3100.
[12] P. T. Wolczanski, Polyhedron 1995, 14, 3335–3362.
[13] P. T. Wolczanski, Chem. Commun. 2009, 740–757.
[14] P. P. Power, J. Organomet. Chem. 2004, 689, 3904–3919.
[15] S. A. Cantalupo, J. S. Lum, M. C. Buzzeo, C. Moore, A. G. DiPas-
quale, A. L. Rheingold, L. H. Doerrer, Dalton Trans. 2010, 39, 374–
383.
[16] M. Bochmann, G. Wilkinson, G. B. Young, M. B. Hursthouse,
K. M. A. Malik, J. Chem. Soc. Dalton Trans. 1980, 1863–1871.
[17] R. E. LaPointe, C. P. Schaller, P. T. Wolczanski, J. F. Mitchell, J. Am.
Chem. Soc. 1986, 108, 6382–6384.
[31] N. A. Eckert, J. M. Smith, R. J. Lachicotte, P. L. Holland, Inorg.
Chem. 2004, 43, 3306–3321.
[32] C. R. Randall, L. Shu, Y-.M. Chiou, K. S. Hagen, M. Ito, N. Kitaji-
ma, R. J. Lachicotte, Y. Zang, L. Que, Jr., Inorg. Chem. 1995, 34,
1036–1039.
[33] H. Zabrodsky, S. Peleg, D. Avnir, J. Am. Chem. Soc. 1992, 114,
7843–7851.
[34] M. Pinsky, D. Avnir, Inorg. Chem. 1998, 37, 5575–5582.
[35] Root mean squared deviation (RMSD) with respect to the crystal
structures for 3–6 are 0.34, 0.15, 018 and 0.22 ꢁ, respectively.
[36] B. Silvi, A. Savin, Nature 1994, 371, 683–686.
[37] E. R. Johnson, S. Keinan, P. Mori-Sꢂnchez, J. Contreras-Garcꢄa, A. J.
Cohen, W. Yang, J. Am. Chem. Soc. 2010, 132, 6498–6506.
[38] J. Contreras-Garcꢄa, E. R. Johnson, S. Keinan, R. Chaudret, J.-P. Pi-
quemal, D. N. Beratan, W. Yang, J. Chem. Theory Comput. 2011, 7,
625–632.
[39] N. Gillet, R. Chaudret, J. Contreras-Garcꢄa, W. Yang, B. Silvi, J.-P.
Piquemal, J. Chem. Theory Comput. 2012, 8, 3993–3997.
[40] D. Fang, R. Chaudret, J.-P. Piquemal, G. A. Cisneros, J. Chem.
Theory Comput. 2013, 9, 2156–2160.
[41] D. Fang, R. L. Lord, G. A. Cisneros, J. Phys. Chem. B 2013, 117,
6410–6420.
[42] D. F. Evans, J. Chem. Soc. 1959, 2003–2005.
Received: July 2, 2013
Published online: August 9, 2013
12228
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 12225 – 12228