A pseudo-Jahn-Teller distortion in an Mo2(μ2-O)2 ring having the shortest MoIV-MoIV double bond

Article abstract of DOI:10.1021/ja025713r

Four structures of edge-sharing bioctahedral compounds of the type Mo₂(μ₂-DArF)₂(η²-L-L)₂(μ₂-O)₂, where DArF is an anion of an N,N-diarylformamidine and L-L is a chelating acetate or DArF group, are reported here. The cores of the ring formed by the Mo₂(μ₂-O)₂ are very similar with very short Mo-Mo distances of 2.306[2] A. These are consistent with the presence of a Mo=Mo double bond of the type σ²π². As expected for these electronic configurations, the compounds are diamagnetic. The most striking characteristic is the distortion of the Mo₂(μ₂-O)₂ ring where a set of two Mo-O distances are significantly shorter then the other set (by ca. 0.05 A). This D_2h → C_2h distortion is explained on the basis of a pseudo-Jahn-Teller effect. Copyright

Full text of DOI:10.1021/ja025713r

Published on Web 02/27/2002
A Pseudo-Jahn-Teller Distortion in an Mo2(µ2-O)2 Ring Having the Shortest
MoIV-MoIV Double Bond
F. Albert Cotton,*,† Lee M. Daniels,† Carlos A. Murillo,*,†,‡ and J. Garrett Slaton†
Department of Chemistry and the Laboratory for Molecular Structure and Bonding, P.O. Box 30012, Texas A&M
UniVersity, College Station, Texas 77842-3012, and the Department of Chemistry, UniVersity of Costa Rica,
Ciudad UniVersitaria, Costa Rica
Received January 25, 2002
The literature on molecules with edge-sharing bioctahedral
structures (ESBOs) (I) is vast. The metal-to-metal distances in such
structures are governed in somewhat obvious ways by the size of
the bridging atoms, X, and the oxidation states of the metal atoms,
but in a more interesting way by the electron configurations of the
metal atoms and their varying abilities to form metal-metal bonds.1
The basic relationships were developed2 and structural data
reviewed3 some years ago, mainly for molecules where X ) Cl.
Since that time there have been abundant new developments, but
there is no more recent review.
It has previously been recognized that metal-metal bond
formation may occur by overlap of d orbitals according to the
Figure 1. One of the two crystallographically independent and centrosym-
general pattern shown as II. Thus, with two d1 metal atoms, a single
metric molecules in Mo2(DXylF2,6)2(O2CCH3)2(µ2-O)2. 1a, with displace-
bond, σ2, is formed and with d2 metal atoms a double bond, σ2π2,
ment ellipsoids shown at the 50% probability level. The other molecule in
is formed. With d3 metal atoms, the simple picture II would incline
1a and that in 1b are similiar. In 2, the two acetate anions are substituted
one to believe that there would be a σ2π2δ2 triple bond. In fact,
by the chelating formamidinate anions DAniF (the anion of N,N′-di-p-
anisylformamidine).
this is not a safe prediction because of the role that pπ orbitals on
the X atoms may play.2a,d It is possible that the actual result may
tion of chloroform from solutions containing 1 afforded a mono-
be a σ2π2δ*2 configuration, because of an interaction of the filled
pπ orbitals of the X atoms on the dδ orbitals of the metal atoms.
Numerous structural results for M(µ2-Cl)2M cases have confirmed
this.2,3
clinic form of 1‚1.93CHCl3, 1b. We therefore have three crystal-
lographically independent structure determinations for molecule 1;
in each case the molecule lies on a crystallographic inversion
center.5 The first remarkable feature of this compound is that it
contains the shortest bond between two molybdenum(IV) atoms
ever reported. The range and the average of Mo-Mo distances in
these three is 2.3048(4) to 2.3077(6) Å and 2.306[2] Å. As a result
of the strong and short ModMo bond, the O-Mo-O and Mo-
O-Mo angles of the independent molecules in structures 1a, 1b,
and 2 are all 107° ( 1° and 73° ( 1°, respectively. DFT
calculations6 have confirmed conclusively that, as expected from
II, there is a σ2π2 double bond between the metal atoms. The
exceptional shortness of this bond is presumably due to the
combined effects of the smallness of the oxygen atoms, the high
basicity and small bite of the bridging formamidinate ions, and the
relatively weak binding of the acetate ions that are approximately
trans to the µ2-O atoms.7
We now report some important results concerning a new ESBO
molecule, 1, shown in Figure 1. The violet compound was easily
prepared by air-oxidation of cis-Mo2(µ2-O2CCH3)2(µ2-DXylF2,6
)
2
(DXylF2,6 ) N,N′-di(2,6-xylylformamidine)) according to eq 1:4
The mean of all the (µ2-O)-Mo distances is 1.938[4] Å, but
this average is far less important than the fact that there are
systematic inequalities within each set of four. Instead of D2h
symmetry, the Mo(µ2-O)2Mo units have only C2h symmetry, as
shown in Table 1.8 While packing distortions can occur in crystals,
it seems very unlikely that the same distortion would occur four
times in crystallographically independent molecules. We propose
that these results, which are summarized in III, are the systematic
result of a pseudo-Jahn-Teller effect,9,10 as will now be explained.
We employ as the local coordinate system for each M and X
atom that shown in I. For a rigorously rhomboidal M2(µ2-O)2 group
(D2h) the pair of pz orbitals on the oxygen atoms can be combined
into linear combinations b1u (p1 + p2) and b3g (p1 - p2). The dxz
cis-Mo2(µ2-O2CCH3)2(µ2-DXylF2,6)2 + O2 f
Mo2(η2-O2CCH3)2(µ2-DXylF2,6)2(µ2-O)2 (1)
Crystals of 1a, obtained by diffusion of hexanes into a dichlo-
romethane solution of 1, crystallized in space group P1h with two
chemically similar but crystallographically independent molecules
(molecules 1 and 2). Alternatively, crystallization by slow evapora-
* To whom correspondence should be addressed. E-mail: cotton@tamu.edu
(F.A.C.). murillo@tamu.edu (C.A.M.).
† Texas A&M University.
‡ University of Costa Rica.
9
2878 VOL. 124, NO. 12, 2002 J. AM. CHEM. SOC.
10.1021/ja025713r CCC: $22.00 © 2002 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Mo-(µ2-O) Distances, Å
Supporting Information Available: Synthetic and crystallographic
procedures, fully labeled drawings of each independent molecule, and
mass spectroscopic data for 1a (simulation and actual data) (PDF) and
crystal data for 1a, 1b, and 2 (CIF). This material is available free of
charge via the Internet at http://pubs.acs.org.
compd
longer
shorter
difference
1a, molecule 1
1a, molecule 2
1b
2
1.960(2)
1.965(2)
1.970(3)
1.996(2)
1.965a
1.913(2)
1.907(2)
1.913(2)
1.948(2)
1.911a
0.047(3)
0.058(3)
0.057(4)
0.048(3)
0.053b
average
References
(1) Cotton, F. A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M. AdVanced
Inorganic Chemistry, 6th ed.; John Wiley & Sons: New York, 1999; p
647.
a Values for 2 not included. b Value for 2 included.
orbitals form the combinations b2g (d1 + d2) and b1u (d1 - d2). The
latter, by itself, provides the essential part of the dπ-dπ bond
between the metal atoms, if an electron pair is available to fill it.
When it is recalled that the interaction of the b3g combination of
the Xpz orbitals with the b3g combination of the dyz orbitals leads
to the inversion of the δ and δ* orbitals,2a,d it is clear that we have
the π-bonding orbital as the HOMO, of b1u symmetry, and the δ*
(au) orbital consisting largely of metal dyz orbitals as the LUMO.
(2) (a) Shaik, S.; Hoffmann, R.; Fisel, C. R.; Summerville, R. H. J. Am. Chem.
Soc. 1980, 102, 4555. (b) Anderson, L. B.; Cotton, F. A.; DeMarco, D.;
Fang, A.; Ilsley, W. H.; Kolthammer, B. W. S.; Walton, R. A. J. Am.
Chem. Soc. 1981, 103, 5078. (c) Cotton, F. A.; Diebold, M. P.; O’Connor,
C. J.; Powell, G. L. J. Am. Chem. Soc. 1985, 107, 7438. (d) Chakravarty,
A. R.; Cotton, F. A.; Diebold, M. P.; Lewis, D. B.; Roth, W. J. J. Am.
Chem. Soc. 1986, 108, 971. (e) Canich, J. A. M.; Cotton, F. A.; Daniels,
L. M.; Lewis, D. B. Inorg. Chem. 1987, 26, 4046.
(3) Cotton, F. A. Polyhedron 1987, 6, 667.
(4) Mo2IV(η2-O2CCH3)2(µ2-DXylF2, 6)2(µ2-O)2, la, was prepared by dissolving
0.080 g (0.10 mmol) of trans-Mo2(O2CCH3)2(DXylF2,6
) in 70 mL of
2
distilled, degassed acetone and sparging with dry dioxygen. The oxygen
was added via cannula until no further color change from the initial yellow
to a deep red occurred. Upon standing, a violet solid developed. After
isolation by filtration; this solid was rinsed with 3 × 10 mL of acetone
until the washings became colorless. The solid was then washed with 3
× 10 mL portions each of methanol and hexanes, dried under vacuum,
and then extracted with 5 × 30 mL of dichloromethane. The saturated
solution was layered with hexanes and after a few days crystals suitable
for X-ray structural analysis developed. 1H NMR (δ, ppm in CDCl3) 7.41
(s, 2 H, N-CH-N), 6.94 (s, 12 H, aromatic C-H), 2.21 (s, 24 H, -CH3),
1.03 (s, 6 H, -CH3). ESI MS (monoisotopic m/z): 849.2 [M + 1]+.
Absorption spectrum (CH2Cl2): λ/nm (ꢀM) 394 (13 300). IR (KBr): 3468
(s), 2960 (s), 1946 (w), 1515 (s), 1492 (s), 1464 (s), 1303 (s), 1260 (s),
1191 (m), 819 (s). The yield was 0.051 g (61%)
(5) Crystal data for the following: (a) Mo2IV(η2-O2CCH3)2(µ2-DXylF2,6)2(µ2-
O)2, 1a, violet crystal, C38H44Mo2N4O6, M ) 844.7, triclinic, space group
P1, a ) 10.9617(6) Å, b ) 11.7084(7) Å, c ) 15.7064(9) Å, R ) 90.159-
(1)°, â ) 109.117(1)° γ ) 98.745(1)°, V )1878.8(2) Å3, Z ) 2, Dc )
1.495 g cm-3, λ ) 0.71073 Å, µ(Mo KR) 0.717 mm-1. R1 ) 0.034, wR2
) 0.072 for all data. (b) Mo2IV(η2-O2CCH3)2(µ2-DXylF2,6)2(µ2-O)2‚
1.93CHCl3, 1b, violet crystal, C39.93H45.93Cl5.79Mo2N4O6, M ) 1075.24,
monoclinic, space group C2/c, a ) 16.823(l) Å, b ) 14.517(l) Å, c )
20.321(2) Å, â ) 110.245(l)°, V ) 4656.2(7) Å3, Z ) 4, Dc ) 1.534 g
cm-3, λ ) 0.71073 Å, µ(Mo KR) 0.918 mm-1, R1 ) 0.062, wR2 ) 0.127
for all data.
A necessary condition for the occurrence of a pseudo-Jahn-
Teller effect to carry a more symmetrical structure into a less
symmetrical one (while preserving a center of inversion) is that
the HOMO and the LUMO be coupled by a normal vibration.11
This will cause a lowering of the energy of the HOMO thereby
stabilizing the molecule in the structure of lower symmetry. In this
case the b1u HOMO and the au LUMO can interact via the b1g in-
plane vibration of the Mo(µ2-O)Mo rhombus (b1u × au × b1g ) ag
in D2h). This vibration is depicted schematically as IV, and it is
evident that it corresponds to exactly the type of (D2h f C2h)
distortion observed in the MoIV-(µ2-O)2MoIV molecules reported
here.
(6) Slaton, J. G.; Thomson, L. M. Unpublished results.
(7) We have also prepared and obtained the structure of 2, an analogous
molecule in which there are µ2-DAniF (DAniF ) N,N′-di-p-anisylforma-
midinate) ligands and η2-DAniF ligands (instead of acetate ions) in the
central plane. Here, because of the stronger bonding of the latter ligands,
the ModMo bond length increases by 0.04 Å to 2.345(5) Å. In this
centrosymmetric molecule the Mo-(µ2-O) distances are 1.996(2) and
1.948(2) Å, the difference being 0.048(3) Å.
(8) Here we have also included data for Mo2IV(η2-DAniF)2(µ2-DAniF)2(µ2-
O)2‚4CH2Cl2, 2. Compound 2 was prepared by oxidation of Mo2(DAniF)4
and crystallized analogously to 1a. The core is similar to that in 1, but
the chelating acetate groups have been substituted by formamidinate
ligands. Crystal data are the following: black crystal, C62H64Cl4Mo2N8O10,
M ) 1414.89, monoclinic, space group P21/n, a ) 13.208(7) Å, b )
10.3291(5) Å, c ) 22.958(1) Å, γ ) 92.136(1)°, V ) 3147.1(3) Å3, Z )
2, Dc ) 1.493 g cm-3, λ ) 0.71073 Å, µ(Mo KR) 0.632 mm-1, R1 )
0.062, wR2 ) 0.126 for all data.
(9) (a) Englman, R. The Jahn-Teller Effect in Molecules and Crystals; Wiley-
Interscience: London, 1972. (b) Pearson, R. G. Proc. Natl. Acad. Sci.
U.S.A. 1975, 72, 2104. (c) Bersuker, I. B. The Jahn-Teller Effect: A
Bibliographic ReView; IFI/Plenum: New York, 1984. (d) Bersuker, I. B.
The Jahn-Teller Effect and Vibronic Interactions in Modern Chemistry;
Plenum Press: New York, 1984. (e) Bersuker, I. B. Electronic Structure
and Properties of Transition Metal Compounds; Wiley-Interscience: New
York, 1996. (f) Barckholtz, T. A.; Miller, T. A. Int. ReV. Phys. Chem.
1998, 17, 435.
(10) Most of the Jahn-Teller effects discussed in ref 9 are centered at a single
metal atom, but there is a precedent for a pseudo-Jahn-Teller effect in a
metal atom cluster compound, namely, W4(OC2H5)16. See: Cotton, F. A.;
Fang, A. J. Am. Chem. Soc. 1982, 104, 113.
It is logical to ask whether the pseudo-Jahn-Teller effect
reported here has previously been seen. It appears that it has not,
but that, of course, raises the question of why not. First, it must be
recognized that the prediction that a certain stereoelectronic system
is susceptible to a pseudo-Jahn-Teller effect (or even a first-order
one, for that matter) is not a prediction that it will be obserVed.
Observability depends on the related questions of how large the
distortion is and whether experimental data are precise enough to
see it.12 In practically all previously determined ESBO structures
such as those with halide bridges the M-M, X-X, and M-X
distances are long, thus lowering the magnitude of the effect.
Moreover, few ESBO structure determinations have replicated with
the exactitude of those reported here.
(11) In principle, coupling of the HOMO with any higher empty orbital of
approximate symmetry can cause a distortion. However, because of an
inverse dependence on the square of the energy difference, the LUMO is
by far the most important.
(12) The difficulty in observing expected Jahn-Teller distortions has been
encountered earlier. An example is the series of face-sharing bioctahedral
anions of the type W2X93-. These typically have nondistorted D3h
symmetry. Only once a slight canting of two planes that lowered the
3-
symmetry to C2V was measured in the W2Br9 anion. See: Templeton,
Acknowledgment. We thank the Robert A. Welch foundation
(Grant No. A494) for support and Mr. Namal de Silva for assistance
in the preparation and characterization of 2.
J. L.; Jacobson, R. A.; McCarley, R. E. Inorg. Chem. 1977, 16, 3320.
JA025713R
9
J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002 2879