Download THEORY AND IMPORTANCE OF OXYGEN

THEORY AND IMPORTANCE OF OXYGEN
BRIDGE-BONDING
B. JE0WSKA-TRZEBIATOWSKA
Institute of Chemistry, The University of Wroclaw, Wroclaw, Poland
ABSTRACT
An oxygen atom as a donor bridge causes the bonding of two atoms at lower
electronegativity. In a case of paramagnetic centres such as transition metal
ions the magnetic criterion occurs, i.e. spin quenching follows. The MO model
for the single oxygen bridge bonding in Ru4, W4, Mo5, Fe', Re4,
Cr', Os4 dimers of d4, d', d', d' and d5 electronic structure were calculated
using the SCCC method. This model gives a good explanation of magnetic
properties of all the known complexes with one linear or angular bridge. The
influence of other ligands on the ground term and on the complex properties
has been explained. The structures most convenient to linear bridge formation
are those of the d°—d4 electronic ones. At d5 the angular bridge is more energetic-
ally convenient. Double oxygen-bridge bonding was studied on the large group
of Mo—V dimers. The probable MO system explaining their magnetic properties
is given. The spectroscopic characteristic in the infra-red of single and double
oxygen bonding confirmed the bonding of the oxygen bridge.
The nature of the bonding of oxygen to other atoms, and the role of the
oxygen atoms as a bonding bridge have been two of the main problems with
which investigators of the structure of chemical compounds and condensed
phases have had to deal in recent years.
Mutual interaction of atoms through oxygen in solid lattices has been
known since the discovery of antiferromagnetism in metal oxide lattices.
This effect has been qualitatively explained in the theory of superexchange
proposed by Anderson', Kramers2 and Goodenough35. The turning
point in the development of views on the role of oxygen was the discovery
of molecular antiferromagnetism, that is, spin—spin quenching within one
molecule. Melor6, Dunitz and Orgel7 found in 1953 that the diamagnetic
ruthenium(iv) compound, the oxychlororuthenate(Iv) built of paramagnetic
ions, is a dimer containing a linear Ru—-O-—-Ru core. Later in 1954, the second
case of spin-quenching in metal ions by the oxygen bridge in a binuclear
complex of rhenium -oxobis(pentachlororhenate) (iv) was discovered by
B. Jeowska-Trzebiatowska and S. Wajda8. In 1962 Morrow9 demonstrated
by x-ray diffraction that the core is in fact, linear 177°, forming the oxobridged Re—O——Re dimer.
In the early 1950s, Pauling's model based upon the valence bond theory
was commonly employed to explain the magnetic properties and the electronic structure of complex compounds. This model explained the magnetic
properties of the ruthenium complex of a d4—d4 electron structure if the
89
B. JEOWSKA-TRZEBIAT0WSKA
oxygen atom was assumed to form two bonds and two it bonds. It failed,
however, to explain the properties of the rhenium complex with a d3—d3 electron structure. In the following years a number of metal ion systems have been
discovered involving molecular quenching of electron spins by the oxygen
bridge. In 1955 Mulay and Selwood'° reported such an interaction on
increasing hydrolysis of ferric perchlorate solution which provided evidence
that the spin-quenching is not a lattice effect. The studies of other authors
yielded similar results1 1• An analogous phenomenon was observed by Sacconi
and Cmi in 195412 during hydrolysis of molybdenum(v) complex solutions.
A step forward in the study of the quenching effect was the discovery of a
temperature singlet—triplet equilibrium between the states S = 1 and S = 0,
made by Earnshow and Lewis13 and later by Jeowska-Trzebiatowska and
Wojciechowski14 in chromium(iii) rhodochloride(j-oxohis [pentaamminechromium(in) chloride).
Many examples of electron spin quenching have been found in oxobridged iron(III) complexes in recent years. First, Schiff's base iron(iii)
compounds'5' 16 oxobridged binuclear iron(ni) complex with HEDTA'7' 18
(HEDTA = N-hydroxyethylenediaminetriacetate), phenanthroline oxobridged dimer'3' 16 and, of biological importance, oxobridged haemin
dimer19 should be mentioned.
The infra-red spectroscopic studies of the it bonds in inorganic compounds
have also shown some peculiar properties of the bonding of oxygen to other
atoms. Oxygen was found to be doubly or even triply bound, and the oxygen
bridges were found to contain, at least partly, it bonding. The role of the
oxygen bridge in electron transfer, hydrolysis and polymerization processes
stimulated further studies of the oxygen bonding. Such studies have been
taken in several world laboratories.
In our Institute the mechanism of oxygen bridge formation, the nature of
the oxygen bridge and the structure and properties of compounds with oxygen
bonding have been studied, using different methods, for many years. The
results of these studies induced me to put forward the idea that the oxygen
bridges are so important in binding atoms that the term oxygen bonding'
may be employed. Our results and consideration of the literature enabled
me to make certain generalizations regarding the nature of the oxygen
bond2022. The complexes containing the oxygen bonds could be considered as clusters—the compounds of various stabilities depending on the
electronic structure, the size of the bridge-bound atoms and the nature of
the other ligands. The oxygen bridge plays a similar role as a junction
between the atoms as a hydrogen bridge; in contrast to the acceptor hydrogen
bridge, however, it is a donor bridge.
The ions linked by the oxygen bridge form a core of the complex in
dimeric or polymeric clusters. Such a core may be linear or angular; the angle,
as a rule, exceeds 90°. The angular bridges appear most frequently in nontransition elements, in transition elements with d° electronic structure, and,
in some cases, in transition metal complexes as will be shown later for
definite d electronic structures. The double bridge-oxygen bonds are, of
course, also angular.
Such clusters, for instance, where the oxygen has a coordination number
of 3, are formed by chromium(ni) and iron(iii)23. The clusters containing
90
THEORY AND IMPORTANCE OF OXYGEN BRIDGE-BONDING
cores with a single or double oxygen bridge arise either as intermediates of
the salt hydrolysis prior to the formation of the final hydroxide or make the
core of the complex compound. Each of the cations is surrounded by ligands,
apart from the bridge oxygens, which complement its local symmetry up
to the distorted octahedron or tetrahedron. The nature of the bridge bonding
becomes most evident when the core consists of the paramagnetic cations
Figure 1. The core containing the d-electron metal ion.
Figure 2. Oxo-bridged dimer of the non-transitional ions.
Figure 3. Angular oxygen bonding found in dinuclear complexes with one or two bridges.
of transition metals. It may then be found experimentally that the oxygen
bond makes mutual interaction of the electron spins possible, as in the case
of a direct metal—metal bond. The magnetic criterion of the chemical bond
appears. Such complexes are, therefore, most helpful in elucidating the nature
of the oxygen bond. In dinuclear complexes consisting of the paramagnetic
transition metal ions, these ions have, as a rule, a local distorted octahedral
symmetry. The present paper will be, first of all, devoted to these types
91
B. JEOWSKA-TRZEBIATOWSKA
of cluster structures. The conclusions obtained for the clusters consisting of
paramagnetic ions may be generalized to cover compounds consisting
of diamagnetic ions.
The common occurrence and the role of oxygen-bridge bonds result from
the fact that terrestrial media, both air and water, are rich in oxygen. The air
Figure 4. The dimer with two oxygen bridges. The oxygen atom can also form a bridge between
the three other atoms.
Figure 5. Oxygen bridge of coordination number three.
is a reservoir of 02 molecules which readily become negative ions with
acceptance of electrons and water, a molecule of which is a simple uncharged
HOH core which may either be changed into a trinuclear complex H3O
or may dissociate, yielding 0H ions. Now, if we know what happens, the
next question is why? Why should the oxygen atom be different from other
bridge-forming donor atoms like sulphur, nitrogen or the halogens? This
92
THEORY AND TMPORTANCE OF OXYGEN BRIDGE-BONDING
preferential position of oxygen results from a number of properties, such as
its electronic structure, clectronegativity and the resulting ability of the
oxygen ion to delocalize electrons and form a small covalent radius. Sulphur
and nitrogen are also capable of bridge formation but could not compete
with oxygen in this respect. While the electronegativity of sulphur with
respect to the first electron is lower than that of oxygen, it is higher with
respect to the second electron and hence its tendency to delocalize electrons
is lower than that of oxygen.
Table 1. Electronegativity of bridge elements24
Element
Valence state
Orbital
Calculated electronegativity
(eV)
H
s
Li
C
Sp3
N
O
S'
s2P3
s2p5
s2p4
s2p3
s2p5
s2p4
Cl
sp5
0
01
F
Cl
Br
I
s2p4
s2p5
s21)5
6.92
2.92
s
s
s
13.70
6.52
8.18
—2.46
p
p
p
p
p
p
p
p
p
p
p
9.50
24.58
11.34
7.16
8.88
19.75
8.23
7.49
* Ref. 25.
In addition, its larger size is a negative factor. The difference between
nitrogen and oxygen is still more significant if we compare the electronegativities of the corresponding electronic structures: N and O, 8.18 and
24.58 eV respectively. This, and the fact that the electronegativity of nitrogen with respect to two electrons is negative, results in an excessive ability
of the nitrogen to delocalize electrons. Nitrogen as a bridge should be therefore stabilized by means of a third acceptor, for instance by H which gives
more stable NH or NH2 bridges. Only in exceptional cases can the nitrogen
bridge be stabilized by metal ions of relatively high electronegativity.
The halogen atoms have unfavourable properties with respect to single
bridge formation. Comparison of the respective electronic structures 0 to
F shows the difference in electronegativity of oxygen and fluorine, the radii
of which are approximately equaL
In our studies on oxygen bridges we have developed the magnetic method
for investigation of the mechanism and determination of the equilibrium
constants for the hydrolysis of some mononuclear complexes of
rhenium(iv)26, osmium(iv)27, and molybdenum(v)283° in order to demonstrate that spin-quenching is a molecular phenomenon and not a lattice-type
effect.
2[ReC16] 2± H20 ± [Re2OCl10] ± 2H ± 2C1
paramagnetic
diamagnetic
2[OsCl6] 2- + H20 [Os2OCl10] + 2H + 2Cl
paramagnetic
diamagnetic
93
B. JE2OWSKA-TRZEBIATOWSKA
2[MoOBr5] 2- ± 4H20 [Mo2O4Br(H20)2] 2 + 6Br + 4W
paramagnetic
diamagnetic
4[MoOBr5]2 + 4H20 ± [Mo4O4Br,2(OH)4] ' + 8Br ± 4W
diamagnetic
paramagnetic
Correlation of these data with spectroscopic results in solution and results
obtained for the complexes in the solid phase enables us to define the structure
of the polymerized molecules in solution. These studies have shown that in
all cases, except for the osmium complex, the diamagnetic dimeric products
of hydrolysis, and in the case of molybdenum the tetramers also, are formed
in solution. The osmium dimer is unstable, and in fact exists only in the
solid state.
The magnetic properties have been studied in the solid state in the temperature range 77°—350°K. The dimeric linear or quasilinear complexes of
rhenium8' 3 137 ruthenium38, osmium27, chromium13'14 and molybdenum39—42 obtained by us have been investigated (Table 3).
These complexes are diamagnetic or exhibit a small temperature-independent paramagnetism. The chromium complex which shows a singlet—
triplet temperature equilibrium is an exception. We have also taken into
account the results obtained by other authors on iron'3" 5—19 and tungsten
complexes43.
In our opinion, the dimer of tungsten44 should be treated like the
tungsten(Iv) complex with a d2—d2 electron structure. Now, a theoretical
justification of the magnetic anomalies should have been found. We have
attempted, as has Martin45, to apply the theory of superexchange in order
to explain the diamagnetism of the dinuclear complexes. Any transposition
of this qualitative method, however, from a collective spin system in the solid
into a molecular system cannot give satisfactory results. In such cases this
method may be considered as a prophetic one, but the qualitative picture
it provides does not always remain in agreement with the experimental facts.
For d3—d3 and d'—d' electronic structures of dimers, this theory would
predict a weak paramagnetism which is true for the chromium complex
only. In the case of the rhenium(iv) and molybdenum(v) complexes which
exhibit only a low temperature independent pararnagnetism, this is inconsistent with the experimental results. The method of superexchange does
not fully justify the magnetic properties and yields no information about
the spectroscopic properties and the nature of the chemical bond.
The electronic structure of the MOM and MM cores is most conveniently
described by the molecular orbital method, which successfully explains the
structure and properties of molecules forming the species of a definite nature.
Studies based on the MO scheme were reported for the first time by Dunitz
and Orgel7 in 1953 as well as by Jeowska-Trzebiatowska and Wojciechowski in 196346. The MOM core was treated as an isolated species, and only
the interactions between the bridging oxygen and the acceptor ions linked
to it were taken into account. The orbital schemes obtained in this way
had no general significance because they did not explain all the magnetic
anomalies found in dimeric compounds. Therefore, we have, with Natkaniec
and Kozlowski47 once again proceeded to develop the orbital model, bearing
in mind that in such complex systems as the dinuclear complexes in which
94
THEORY AND IMPORTANCE OF OXYGEN BRIDGE-BONDING
each central ion of the core is surrounded, apart from the bridge-oxygen,
by five other ligands of a local distorted symmetry, strong mutual interactions exist. Omission of these interactions, as in the previous orbital
models, results in too serious approximations. Now we have taken into
account three types of essential interaction:
1. Interaction between oxygen and both metal nuclei which is essential
for the formation of oxygen bonding. The interaction follows the formation of
a stable MOM core and is chiefly responsible for magnetic and spectroscopic properties of a new chemical species formed in this way.
2. Interaction between the metal ions and their local ligand environment
excluding the bridging oxygen. The interaction should be considered as
similar to that in mononuclear complexes.
3. Interaction between the hgands belonging to the two nuclei, in particular between the ligands placed in the xy planes perpendicular to the
bridge axis.
It should be emphasized, that the two latter types of interaction as was shown
by our present studies, affect also the electronic structure of the core and its
related physical properties.
4z
IT
li-v
pz
-li-v
7Th'
Figure 6. Diagram illustrating molecular interaction.
95
B. JE2OWSKA-TRZEBIATOWSKA
Let us consider a type L5—M----O—M—-L5 complex with a linear oxygen
bridge of D4h symmetry. This type of structure appears most frequently in
the transition metals: rhenium, ruthenium, osmium, chromium and molybdenum.
Figure 7. Linear cluster.
The MO calculations have been performed by us for the complex molecule
as a whole by the self-consistent charge and configuration (SCCC) method48.
The d orbitals of the central ion as well as the s and p orbitals of oxygen are
transformed as bases of the following irreducible representations. Table 2
shows that the metal
d. orbitals of e symmetry interact with the p,
p, orbitals of the same symmetry, giving the it-bonding and it-anti-bond
molecular orbitals. On the other hand, the combined -d2 orbitals of both
metals with symmetries aiq and a2 overlap each other with the oxygen s
and-p orbitals respectively, leading to the formation of molecular orbitals.
Among the group of molecular orbitals discussed, one should distinguish
the bonding and antibonding a2 and e orbitals as bridging ones. These
orbitals are characterized by the considerable extent of electron delocalization. This is caused by a large overlap and similar energies of the atomic
orbitals of oxygen and of metals of which these MOs consist. The aig orbitals
consist of the oxygen and metal orbitals overlapping one another to a large
Table 2. Orbital transformation in D4h symmetry
Representations
b28
b1
eq
e
Metal d orbitals
Oxygen
sp orbitals
(1/,J2)(d3. + d,)
—
(1/J2)(d + d)
(1/J2)(d + d)
(1/J2)(d — d)
(1//2) (dy: —
d)
p
P
big
(1/j2)(d2_2 ± dci_yi)
b2
(1/j2) (dXz_),i — d2_,2)
ajg
(1/j2)fd ±d2)
S
a2
(1/J2)(d. —d2)
-p
96
THEORY AND IMPORTANCE OF OXYGEN BRIDGE-BONDING
extent. However, due to the large difference in energies of the oxygen s
orbital and metal d orbital, the extent of electron delocalization in these
orbitals is insignificant.
We have also taken into account the and it interactions of other ligands
with metal ions.
it orbitals of ligands with b1g and b2 symmetries overlap
with the metal orbitals of suitable symmetries, that is, with combinations
x
Figure 8 (a) and (b). Oxygen—metal p —
d,,
interaction.
of the d2_2 yielding MOs. As a result, the energy ordering of molecular
rbitals has been obtained (Figure 10).
Three orbitals aq,
and e, consisting mainly of the oxygen orbitals,
are occupied by the oxygen valence electrons. The a2 and e orbitals act as
the main 'binder' in oxygen bond formation. The other orbitals of this
ordering consist mainly of the metal ion d orbitals. The metal electrons
are placed in these orbitals in the energy ordering. Thus, the oxygen bonding
makes possible the pairing of electrons stemming from both metal atoms in
the same molecular orbital.
97
P.A.C—27/1_--E
B. ThZOWSKA-TRZEBIATOWSKA
x
z
x
Figure 9. The overlap between the s, p orbital of oxygen and the d2 orbitals of metal atoms.
The orbitals; 1%, e and ht are close to each other in their energies
and their ordering is determined by interactions between the ligands of both
nuclei. The sequence of these orbitals determines their magnetic properties.
The energy interval of the next e orbital is higher by one order. The b2
and biq as well as a2 and aiq orbitals are close to each other in energy. The
b29, e0 b1 orbital ordering is shown in the scheme (Figure 10) corresponding
to the medium or even weak ic-bonding halogen, thiocyanate and other
aig
// ___
2u
/7
(nl)d
___
N
lu
—----
\
_tac LJk;
(p.py)O
\\\N\N
\N
\ \ (Pz)
\
N
\PsO
\s-o
\Ulg
AO of ligonds
AO Of metal
Figure 10. Molecular orbital diagram for a linear dinuclear transition metal complex with
oxygen bridge of D4h symmetry.
98
THEORY AND IMPORTANCE OF OXYGEN BRIDGE-BONDING
ligands. For the a-only bonding ligands, this ordering is altered. For instance,
in the case of the NH3 ligands appearing in p-oxo-bis[pentaamminechromium(iii)] salt, (NH3)5(CrOCrNH)54, there are in fact no interactions
between the ligands of two nuclei coordinated in the plane perpendicular
to the line connecting the two nuclei.
As a result, the b2g and b1 orbitals become very close in energy, causing
a change in the sequence of orbitals to give the ordering in the RHS of
Figure 11, which determines the magnetic properties. The orbital scheme
obtained by us made it possible for the magnetic properties of all known
dinuclear complexes with oxygen bridge of d electron metals with D4h or
similar symmetries to be explained (Table 3).
[CI5ReOReCI5]4
[(NH3)5CrOCr (NH3 )5]4
e
E
b1
b1
'2g
eg
b
Figure 11. Diagrams showing MO' sequence dependence on the character of the ligands.
The dimers of rhenium, molybdenum, ruthenium and osmium have a
diamagnetic singlet ground term 1A1 which results from the completely
filled orbitals. The bridge complex of tungsten has a triplet ground state
3A2g resulting from incomplete filling of the eg orbitals. Small differences
in energies of the b2g, eg and b1 orbitals and a possibility of electron transition
results in an increase of the moment from 1.72 B.M. at 76°K up to 2.40B.M.
at 2300K28 per tungsten atom.
The dimer of chromium(in) exhibits a singlet—triplet transition. This
complex is paramagnetic and its moment is 1.85 B.M. at 293°K while at
low temperatures it decreases, almost to zero at — 125°K. This is due to the
fact that the energy gap between b2g and b1 is very small and the distance
between the corresponding terms is therefore comparable to k T The singlet
ground term 'A1 changes into the triplet state 3A2 with simultaneous
filling of the bI orbital.
I would like to draw attention to the temperature-independent
paramagnetism which appears in the dimers possessing a diamagnetic
singlet ground state. This temperature-independent paramagnetism depends,
of course, on the distance between the interacting levels, that is, the distance
between the nearest excited state and the ground state, and on the shape of
99
B. JEOWSKA-TRZEBIATOWSKA
Table 3. Magnetic properties of dinuclear complexes of transition metals with oxygen bridge
Electronic structure
Compound
T(°K)
Ueff
XM )< 106
average values
K4[Re2OC110]
(b2g)2 (e9)4 (b1)°
K4[Re2OC110 H20]
K4[Re20(C204)2(OH)6]
K4[Re20(C204)4(OH)2]
(b2)2 (eq)4 (b1)°
(b29)2 (e)4 (b1)°
450
220
diam. (8)
diam. (26)
diam. (32)
diam. (32)
1.73 (43)
2.40
diam. (38)
diam. (27)
diam. (13,14)
1.8 (13,14)
(a) diam(13)
(b2g)2 (eq)2 (b1)°
K4[W20C110]
(b2q)2 (eq)2 (b1j°
(b2q)1 (e9)2 (b1)1
K4[Ru2OC110]
(b2g)2 (e9)4 (b1)2
K4[Os2OCl1o]
(b2g)2 (e9)4 (b1)2
[Cr20(NH3)10]C14
(e6)4 (b2q)2 (b1)°
(e9)4 (b2g)1 (b1)'
(a) [Fe20(Ph)4(H20)2X4] (b1)2 (a2)2 (b2)2 (a2)2
(a1)2 (b1)° (I)
(b1)2 (a2)2 (b2)2 (a2)2
(b) (Fe Selen)20P3,2
(a1)1 (b1)1 (II)
(139°)
equilibrium 1—11
80—300
80—300
—300
—540
76.5
293
200
— 160
126
293
276
(b)diam(15,16)
(a) 1.76(13)
(b) 1.87(15,16)
(a) 0.77 (13)
(b) 0.58 (15,16)
80
their wavefunctions which enter into the matrix element of Zeeman interaction.
In the linear bridges, where the d'—d1 and d3—d3 electron structures are
involved as molybdenum(vi) and rhenium(iv) complexes, the distance between
the filled orbital and the excited orbital is of the order io cm For this
.
reason these complexes show rather normal temperature-independent
paramagnetism.
a1
a
————
_______
* ____________
2u
____________
b2
big
____
bf
b1
eg
Sg
D4h
a1
/
______
— — ——
a24 +
b2tH,
b14
f.
C2j,
Figure 12. Orbital splitting in dinuclear Fe(nI) complexes with the bent bridge.
100
THEORY AND IMPORTANCE OF OXYGEN BRIDGE-BONDING
In the d4—d4 ruthenium and osmium dimers the energy separation between
the strong n-antibonding bridging e orbital and the completely filled b,
iO cm '.Therefore, the temperature-independent
orbital is large, range
paramagnetism is negligible, viz, less than 200 x 106 3s
The d5-configuration appearing in the iron(IIi) dimers exhibits peculiar
properties. On the basis of the orbital scheme, two electrons should be
placed into the strong antibonding bridging e orbital. The structure of a
bent bridge will therefore be more favourable, corresponding to the altered
hybridization with the increased p-orbital contribution to the bonds. This
results in weaker d—p interactions which cause the splitting of e orbital
into two orbitals.
Meanwhile, the symmetry is changed into C2,. At the small bend of the
Fe—O——Fe bridge'5'9 the splitting is also small and the distance between
the a and bT orbitals is small; the singlet—triplet equilibrium and tempera-
ture dependence of the moment are observed. The diamagnetic singlet
ground state 'A1 of the a, b configuration is changed, on increasing the
temperature, into the triplet state 'B, of the
b,° configuration.
The interpretation of magnetic properties based upon our orbital scheme
gives clear evidence that the linear MOM bridges may be found only for
d°, d1, d2, d3 and d4 configurations in complexes. In the case of the d5 con-
figuration an angular structure of the bridge becomes more convenient.
Figure 13. Model of angular cluster.
With the increase of electron filling to d6, d7 and d8 configurations, the formation of a bridge becomes less probable since the successive electrons should
have been placed in the strong antibonding bridging e orbital and then, in
the antibonding orbitals. Such a system would, of course, be very unstable.
Now it is necessary to discuss the properties and structure of dimeric
clusters with two oxygen bridges, taking as an example the large group of
molybdenum(v) complexes synthesized and studied by Rudolf and me. The
d'—d' electron structure seemed to be most convenient for the examination of
the properties of these more complex systems. These compounds have been
obtained as hydrolysis products of the monomeric oxyhalogenomolybdates.
Magnetic studies carried out in solution and in the solid phase have shown
that dimers or higher polymerized species are formed in which the molybdenum atoms are linked by double oxygen bridges3942. We have isolated
about twelve dinuclear di-t-oxoha1ogenomolybdates(v), and a new type of
molybdenum complex: tetranuclear molybdenum(v) clusters. I shall deal
only with dimers possessing twO oxygen bridges.
101
B. JE2OWSKA-TRZEBIATOWSKA
Table 4. Magnetic susceptibilities of Mo(v) dinuclear cornplexes
XM x 106
Complex
(PyrH)2[Mo2O4C14(H20)2]
(Met4N)2[Mo2O4C14(H20)2]
(PyrH)2[Mo204C14(H20)2]H20
(NH4)2[Mo2O4Cl2(OH)2(H20)2]
(PyrH)4[Mo204C12(OH)4]
(PyrH)2[Mo2O4Br4(H20)2]
(PyrH)4[Mo204(SCN)6]
K2[Mo204(C204)2(H20)2] 3H20
Ba[Mo204(C204)2(H20)2] 31120
207
328
165
157
260
185
448
115
107
The magnetic studies on the molybdenum(v) dimers have shown that
their electron spins are almost completely quenched. Only a small, temperature-independent paramagnetism remains3942. The dipyridyl complex
[Mo2O4C12(dipy)2], showing a typical dependence of magnetic susceptibility of temperature and a magnetic moment of about 0.9 B.M., is an exception49.
Figure 14. Model of cluster with two oxygen bridges.
Comparison between the systems with a 'double oxygen bonding' and the
system with a 'single oxygen bonding' suggests that the presence of two
bridging oxygens and also of strong it bonding ligands as terminal oxygens
results in the weakening of the bridge it bonds. There is then a simultaneous
increase in the metal—oxygen interactions and a change in the geometry
from D4h into C2h which results in a further weakening of the metal—oxygen
it bonds. Due to the decrease in the molybdenum—molybdenum distance a
direct metal—metal interaction appears.
The x-ray data obtained by Cotton5° for K2{[Mo02(C204)H20]20}
indicate that this distance is small, about 2.4 A. Due to the presence of two
bridging oxygens, an additional oxygen—oxygen interaction also appears.
Let us now consider the molecular model for a dimeric molybdenum
cluster of C21, symmetry, taking into account all these interactions.
The terminal ligands oxygen, water and halogens are also of significance
and disturb the electronic structure of the core. In such a system as shown
in Figure 15, the d orbitals of the Mo(v) ions and the sp orbitals of the bridge
102
THEORY AND IMPORTANCE OF OXYGEN BRIDGE-BONDING
tz
1z
CL
Figure 15. The structure and stereochemistry of [Mo2O4C14(H2O)2]2.
oxygens make up the basis of irreducible representations of the point
group C2h.
The combinations of the d..2_)2 orbitals of that group form the type
bonds along the z and x, y axes, respectively. While the other d orbitals
contribute to the molecular it orbitals, the orbitals in the bridge plane and
orbitals lie in the planes perpendicular to them. Because of the
the
short metal—metal distance, a c Mo—Mo interaction through the dX), orbitals
is possible here. If the it interaction in the bridge plane is assumed to be
weaker, owing to the angular character of the bridges, as shown above, the
following energy ordering of the d orbitals should be expected.
The magnetic properties of such compounds are determined by the
sequence of molecular it orbitals consisting mainly of the metal d orbitals.
Table 5. The linear combination of atomic orbitals for the C2h point
group in dinuclear complex
ag
(1/\/2)(dX.
+ d)
(1/2)(d. + d. —
b
a
bg
('/2)( — p,.+ p —
(1/j2)(dz + d2)
—
(1/2)(s + s')
d;2)
(l/2)(p + p + p' + ji,)
(1/,j2)(d2 — d2)
(1/.j2)(d,, — di,,)
(1/\/2Xp, + i"2)
(1/2)(d2 + d. + d. + d2)
(l/2)(p — — p' + i';)
1j2(d — d,. + d'- — d) (1/J2)(s — s')
(t/2)(.p ± py Pc —
(1/j2)(d2_2 +d_y2)
(1i2xd — — d ± d;4 (1/'2Xp. —
(1/,,J2Xd22_2 — d2_2)
103
B. JEOWSKA-TRZEBIATOWSKA
The energy intervals between the ag(xy) and bjxy) orbitals in the above
molybdenum(v) complex are sufficiently large, due to the c Mo—Mo interactions, for the compound to be diamagnetic (ground term 1Ag). If, however,
instead of oxygen, another ligand of a it acceptor nature or one that does not
give any it bonds, is introduced into the molecule, then the separation of the
d.) would be very small and the compound
lowest orbitals ag(dxy, dr..),
would exhibit a temperature-dependent paramagnetism (ground term 3Am).
2
cxOg(dz)
comb SaOb O3H2O oCt
abu(dz)
comb POb Ot 0H20,UCL
aa(d2-y2) comb P6Qt,
crbg(di-y2)
comb P00b
0C1
0CL
E
lTbu(dxzdyz) comb
zObTOt
7TCI
fdxzdyz) comb
1TXObtOt
rCt
comb
7r0
ITCL
rag(dxzdyz) comb
rO
7TCL
lTbu(xy) 1
comb
ITfrY)Ob
irCt
Tfry)Ob
ITCI
QM0-Mo
lrag(xy) J comb
Figure 16. Molecular orbital scheme for dinuclear molybdenum complex with two oxygen
bridges.
The paramagnetic properties of the dipyridyl complexes may be explained
by the appearance of the 3A term. Lowering of the magnetic moment with
respect to one electron spin per molybdenum atom may result from the
equilibrium existing between the paramagnetic and diamagnetic isomers.
Vibrational spectra make it possible to draw decisive conclusions on the
bridge structure. There is a choice between the linear and angular structures
and it is now possible to determine the bond order and bond strength, and
hence, the it interactions between the bridge oxygen and the atoms linked
to it. The extensive investigations by infrared spectroscopy of different compounds with the oxygen-bridge, carried out by Hanuza and me and obtained
from the literature5 054 allow further extension of the oxygen bridge theory.
The linear and angular structures of the bridge described as a 'three body
system differ in the number of infra-red and Raman active frequencies.
The normal modes of vibration given in Table 6 for the linear and angular
MOM systems show that for the complexes with angular bridges, all three
types of frequencies are both infra-red and Raman active while the linear
bridge complexes should not exhibit any symmetric stretching frequencies
in the infra-red. The dimers with two oxygen bonds treated as a tetratomic
104
THEORY AND IMPORTANCE OF OXYGEN BRIDGE-BONDING
0
()
Dipy
Figure 17(a) and (b). Geometrical isomers of [Mo2O4C12(dipy)2].
105
B. JE2OWSKA-TRZEBIATOWSKA
Table 6. Normal modes of vibration of MOM core
Assignments
Dh symmetry-linear bridge
Normal vibration
Vs(MOM)
M
M
0
0
OM
0
0-' 4-0
0
M
0
O—O
M
4-0
Vas(MOM)
M
4-0
(MOM)
Representation Activity
Frequency
range (cm 1)
Ra
200—350
i.r.
750—900
i.r.
50—100
9
Bent bridge C2 symmetry
Normal vibration
Vs(MOM)
VaS(MOM)
MOzM
ö(MOM)
Representation Activity
Frequency
range (cm ')
A1
i.r., Ra
400—600
B2
i.r., Ra
750—900
A1
jr., Ra
100—220
system with a D2h symmetry possess six types of normal modes of vibration.
Among them, three are infra-red active: B1, B2, B3. Two strong bands
corresponding to the stretching frequencies of the double bridge lie in the
ranges 450—670 cm1 and 560—830 cmt (see Table 9).
Arguments based upon the molecular orbital scheme suggest that linear
bridges are formed, above all, by the d° to d4 electronic structure of metal ions
with local distorted octahedron symmetry and a total symmetry of the
complex ion D4h. Table 7 contains these types of dinuclear clusters; the values
obtained from infra-red studies of various compounds with different outer
ions are shown as frequency ranges.
The appearance of some infra-red inactive symmetric frequencies over the
range 200—250 cm' follows from small distortions of the linear bridge.
This is particularly evident for the osmium and ruthenium complexes where
the introduction of mixed ligands results in a certain change in symmetry.
With the bent bridge, the symmetric frequencies become active in the
infra-red according to group theory with simultaneous considerable increase
in vibrational energy. The symmetric frequencies are situated, however, in
the same range as those of the linear bridges.
106
THEORY AND IMPORTANCE OF OXYGEN BRIDGE-BONDING
v(MO)Ag
ring deform.B10
v(MM)Ag
v(MO)Big
v(MO)820
v(MO)830
A
Figure 18. Norma! modes of vibration of M202 core.
The increase of the symmetry frequencies follows from the decrease of
the bridge bond order due to the reduced interactions between the oxygen
p orbitals and the p or d5 orbitals of the central ions. This is because the
hybridization of oxygen varies with changing angle from 180° to 120° from
the diagonal sp into the trigonal sp2 and at 109° into the tetrahedral sp3. As
the oxygen bridge bond angle diminishes, the it-interaction in the bridge
decreases.
Reduction of the angle leads to the formation of a double-bridge bonding,
that is to the two oxygen bridges, which appear in the di-t-oxocomplexes.
Table 7. Vibrational frequencies of linear oxygen bridge MOM (angle 175—180°)
Compound
[Re2OC110]4
[Re2O3en2Cl4]
[Re2O3py4Cl4]
Vas(M0M)
Vs(M0M)
850—855
730
700
752
200*
K6[Re203(CN)8]
[Os2OCl10:14
[0s20Br10]4
[Os2OI10]4
838—850
843—845
835—841
[Os2OCl6Br4]4
[Os2OCl8I2]4
846—848
830—832
[Ru2OC110]4'
[Ru2OBr1 ]4_
[Ru2OCI8Br2]4
[Cr20(NH3)10]C14
(NMe)4[Zr2OC110]
[Mo203(OX)4]Y2
880—885
C74H42N18Mn'O•2py
200*
200*
200*
205
205
200
204*
213_215*
750
230
230
245*
215
213
860
820
—
858—86 1
852
873
Refs
a,b,c,d,e
a
a
a,b,d,e
a,b
a
a,h
a
a.h,c,d,e
abc,
a
c,d
g
6
ii,
B. Jezowska-Trzebiatowska. J. Hanuza and M. Baluka. 4cta Phys. Polonica, in press; Specsrochimica Acta, in press.
W. P. Griffith, J. Chem. Soc. A. 211 (1969).
R. M. Wing and K. P. Callahan, borg. Chess. 8,871 (1969).
D. J. Hewkin and W. P. Griffith, J. Chem. Soc. A, 472 (1966).
F. A. Cotton and R. M. Wing, borg. Chess. 4. 867 (1965).
A. Feltz, Zeit. Anorg. Ailgem. C/tern. 358, 21(1968).
P. C. M. Mitchell, J. Chess. Soc. A, 146 (1969).
J. A. Elvidge and A. B. P. Lever, Proc. Chern. Soc. 195 (1959).
L. H. Vogt, ir, A. Zalkin and D. H. Templeton, Science, 151, 569 (1966).
* observed in jr. spectra.
107
B. JE±OWSKA-TRZEBIATOWSKA
Table 8. Vibrational frequencies of bent oxygen bridge
Compound
VSSIM0M)
Ge2OFf
Sn2OF
Sn20(CH3)6
As2OF
Sb2OF'
P2O
As2O'
SOS
Vs(M0M)
8(MOM)
B2
A5
A1
858
820—750
465—490
—
—
—
a
a
200—210
c
458
737
415
889—899
859—883
890—980
733—780
760—820
530—538
435—473
682—730
530—560
260—350
558
503
Refs
b
200—300
240—245
125—160
d
—
—
—
g
h
Cr2O
V2O'
772
810
790
Se2O
Si2OCI6
557—670
491—516
170—250
1132
421
100
'
c
<200
556
e
f
I
j
k
B. Je±owska-Trzebiasowslca, J. Hanuza and W. Wojciechowski. Spear. Ada, 23A, 2631 (1967).
H. Kriegsmann and S. Pischtschan, Z. Anorg. Allgem. Chess. 315.283(1962).
B. Jeiowska-Trzebiatowska and J. Hanuza, Spectr. Acm. in press.
E. Steger and G. Leukroth, Z. Assorg. Aligem. Chess. 303. 169 (1960).
B. Je±owska-Trzebiasowska and 3. Hanuza, in preparation.
R. J. Gillespie and F. A. Robinson, Spear. Acme, 19, 741 (1963).
H. Trammreich, D. Bassi, 0. Sala and H. Siebert, Spear. Acme, 13, 192 (1958).
W. P. Griffith and T. 0. Wiekins, J. Chem. Soc. A, 1087 (1966).
R. M. Wing and K. P. Gallahan, Inorg. Chess. 8, 871 (1969).
A. Simon and R. Paetzold, Z. Anorg. Allgern. Chess. 303, 39 (1960); 353, 53 (1967).
J. F. Griftiths and D. F. Sturman, Specer. .4cma, 25k, 1415 (1969).
Table 9. Vibrational frequencies of double oxygen bridge M
Compound
[As2O2F8]2
V5(M2o2)
V6(M2o2)
A19
B39
—
—
[Sb2O2F8]2'
[Mo204(EDTA)]2
759
[Mo2O4C14(H2O)2]2
729
[Mo2O4Br4(H20)2]2
[Mo202(OH)2Br2(OH)4]2
[Mo2O4CI2(OH)4]4"
[Mo202(OH)2C14(OH)2]
[Mo204ox2(H2O)2]4
K4[Os2O6(N02)4]
K4[0s206ox3]
[Ru2O6(NH3)4]
[W204(NCS)6]4
[1208(OH)2]4
590—605
560—570
736
705
—
—
720
739
510
520
745
—
—
—
654
V5(M2o2)
B29
N0Z
V4(M2o2)
470—490 291—315
450—470
635
675
490
567
Rels.
B5
—
314
296
—
—
—
a
a
b
b, e
e
e
e
e
730
742
722
829
480
799
830
503
243
240
490
200
C
483
335
332
d
d
—
757
498
612
530
613
B. Je±owska-Trzebiatowska and J. Hanuza, Spectr. Acme, in press.
R. M. Wing and K. P. Callahan, lnarg. Cheiss. 8, 873 (1969).
B. Je±owska-Trzebiatowska, 3. Hanuza and M. Batsika, Acme Phys. Polonica, in press.
W. P. Griffith, J. Chess. Soc. A, 211(1969).
B. Je±owska'Trzebiasowska and M. Rudolf, to be published.
108
70NM
—
—
b, e
c
c
M X4
Vus(M0M)
0
*
V(M_oH)
VB3(M202)
VB2(M202)
VS(MOM)
X or Y can be CI, Br, I, ox/2, py, en2, NO2.
M(OH)X5
4X M
,/\
0
VUS(M02)
X5MOMX5
VS(M02)
M02X4
trans
VUS(M03)
V503)
M03X3
CiS
VMO
of compound vibrationalmode
MOX5
MOX3Y2
Type
order
Bond
fr
cm 1
—
550—570
1.0—1.2
800—830
503—530
204—215
2.82—2.99 448—503
3.30—3.87
830-848
775835
720—855
200—205
952
915
889—905
6.06—6.83
799—843
6.96
890—912
890
1.2—1.3
1.3—1.6
2.0—2.1
2.0—2.3
1
j
Osmium
—
cm1
Jr
Ruthenium
1.84
2.6—3.25
3.90-4.15
—
—
528—584
2.40—2.92
2.8—3.0
720—760
490—670
490
235
830
3.05—3.30
5.2—5.3
6.0—6.71
850—860
220—230
982—910
824—839
820—830
852—885
2.8—3.3
Jr
930—1020 7.20—7.8
cm'
Molybdenum
6.09-6.70 807—822 5.12—5.38 770-780
912—993 7.91—8.55 952—960 8.01
cm
Rhenium
Correlation between the metal—oxygenfrequencies, bond order and force constants
[Md/A]
2.2—2.5
Table 10.
—
7.0—8.0
J
520—560
860—870
210—220
2.4—2.9
3.2—3.6
980—1020 7.1—7.4
—
950—960
cm
Chromium
C
rn
Q
rn
0
0
rj
rn
z
(1
0
0
rn
B. JEOWSKA-TRZEBIATOWSKA
The angle may approach 900 (Table 9). The bond order in such bridges decreases and the bond order has values intermediate between those characteristic of metal—oxygen c bonding and that existing in the linear bridge MOM.
Infra-red spectroscopy is the technique which makes possible quantitative
studies of the effect of ic-interaction in all oxycompounds. In particular, a
correlation between the bond order and the force constants calculated from
the frequencies for a number of oxycompounds with various functions of the
oxygen is a good illustration for the varying contribution of it-interactions
in the bonding.
Table 10 contains the observed frequencies and the force constants calculated from them for various types of oxycompounds formed by the transition
metals with different ligands which have been discussed above, with respect
to their magnetic properties and the theoretical interpretation of their
electronic structure on the basis of the MO scheme. A wide range of force
constants for oxygen bridges in rhenium complexes results from taking into
account the clusters with strong ic-bonding ligands beyond the oxygen
bridge. Higher values of force constants are associated with the weak
it-bonding ligands such as the halogens. The force constant values obtained
are a linear function of the bond order.
It follows from these investigations that the contribution of the ic-interaction is largest for the mononuclear mono-oxycomplexes; this decreases,
however, with the increasing number of oxygen atoms in the molecule. The
it-interaction is, in turn, much lower in the oxygen bridge and extremely low
in a double oxygen bridge. It diminishes in the following order in the series
of oxycompounds:
MOX5 >
/o\
MOX6> MOM > M /M > MOH > MOH2
It seems that the correlation of studies by means of several methods,
principally magnetic and infra-red spectroscopy, with simultaneous theoretical interpretation by the molecular orbital method gives the clearest idea of
the structure and character of oxygen bonding.
REFERENCES
P. W. Anderson, Phys. Rev. 79, 350 (1950); 79, 705 (1950); 86, 964 (1953).
H. Kramers, Physica, 1, 182 (1930); 18, 101 (1952).
J. B. Goodenough and A. L. Loeb, Phys. Rev. 98, 391 (1955).
J. B. Goodenough, Phys. Rev. 100, 564 (1955).
J. B. Goodenough, Magnetism and Chemical Bond, Wiley: New York (1963).
6 D.
P. Mellor, J. Proc. Roy Soc. Wales, 47, 145 (1944).
J. D. Dunitz and L. E. Orgel, J. Chem. Soc. 2594 (1953).
8 B. Jeowska-Trzebiatowska and S. Wajda, Bull. Acad. Polon. Sci. 2, 249 (1956).
J. C. Morrow, Acta Cryst. 15. 851 (1962).
10
L. N. Mulay and P. Seiwood, J. Am. Chem. Soc. 77, 2693 (1955).
B. Werbel, V. H. Dibeler and W. C. Vosburgh, J. Am. Chem. Soc. 81, 1033 (1959).
12 L. Sacconi and R. Ci, J. Am. Chein. Soc. 76, 4239 (1954).
13 A. Earnshow and .1. Lewis, J. Chem. Soc. 396 (1961).
14 B. Jeowska-Trzebiatowska and W. Wojciechowski, J. Inorg. NucI. Chem. 25, 1977 (1963).
2
110
THEORY AND IMPORTANCE OF OXYGEN BRIDGE-BONDING
C. Scheringer, K. Hinkler and M. V. Stackelberg, J. Anorg. Chem. 306, 35 (1960).
J. Lewis, F. E. Mebbs and A. Richards, Nature, London, 207, 855 (1965);
M. Gerloch, J. Lewis, F. E. Mebbs and A. Richards, Nature, London, 212, 809 (1966);
J. Lewis, F. E. Mebbs and A. Richards, J. Chem. Soc. A, 1014 (1967);
M. Gerloch and F. F. Mobbs, J. Chem Soc. A, 1598 (1967);
M. Gerloch and F. E. Mobbs,J. Chem. Soc. A, 1900 (1967);
M. Gerloch, J. Lewis, F. E. Mobbs and A. Richards, J. Chem. Soc. A, 112 (1968);
K. S. Murray, A. von der Bergen, M. J. Connor, N. Rehak and B. 0. West, Inorg. NucI. Chem.
Letters, 4, 87 (1968).
16 G. M. Bancroft, A. C. Maddock and R. R. Pandl, J. Chem. Soc. A, 2939 (1968).
17 H. Schugar, C. Walling, R. B. Jones and H. B. Gray, J. Am. Chem. Soc. 89, 3712 (1967).
18 s J.
Lippard, H. Schugar and C. Walling, Inorq. Chem. 6, 1825 (1967).
19 S. B. Brown, P. Jones and I. R. Lantzke, Nature, London, 223, 1960 (1969).
20 B.
Jelowska-Trzebiatowska, Theory and Structure of Complex Compounds, p 1. Pergamon
Press: London; WNT: Warszawa (1964).
21 B.
Je±owska-Trzebiatowska, Exper. Suppl., 9, 128 (1964).
B. Jeowska-Trzebiatowska, J. Chim. Phys. 765 (1964).
22 B. Jelowska-Trzebiatowska, Coord. Chem. Rev. 3, 255 (1968).
23 B. M. Figgis and C. B. Robertson, Nature, London, 205, 694 (1965).
24 G.
Klopman, J. Chem. Phys. 43, 5124 (1965).
25
Hinze and H. H. Jaffé, J. Am. Chem. Soc. 84, 540 (1962).
26 B.
27
28
29
30
31
Jeowska-Trzebiatowska, J. Mroziñski and W. Wojciechowski, Roan. Chem, 39, 1187
(1965).
B. Jeowska-Trzebiatowska, J. Hanuza and W. Wojciechowski, J. Inorg. Chem. 28,2701(1966).
B. Jeowska-Trzebiatowska and M. Rudolf, Roczn. Chem. 41, 453 (1967).
B. Je±owska-Trzebiatowska and M. Rudolf. Roan. Chem. 41. 1879 (1967).
B. Je±owska-Trzcbiatowska and M. Rudolf. Roczn. Chem. 42, 1227 (1968).
B. Je±owska-Trzebiatowska and S. Wajda, Mémoires présentés a Ia Section de Chimie,
XVI IUPAC Congress, Paris (1957).
32 B. Je±owska-Trzebiatowska and S. Wajda, Bull. Acad. Polon. Sc!., Cl. III, 6, 217 (1958).
B. Je±owska-Trezbiatowska and S. Wajda, Bull. Acad. Pol. Sc!., Ser. Sc!. Ch!m. 9, 57 (1961).
B. Je±owskaTrzebiaeowska, S. Wajda and W. Wojciechowski, Bull. Polon. Sc!. Ser. Sc!.
Ch!m. 9, 65 (1961).
B. Jelowska-Trzebiatowska, S. Wajda and W. Wojciechowski, Bull. Acad. Polon. Sc!. Ser.
36
Sc!. Ch!m. 9, 767 (1961).
B. Je±owska-Trzebiatowska and W. Wojciechowski, Bull. Acad. Polon. Sc!. 9, 685 (1961).
S. Wajda and F. Pruchnik, Roczn. Chem., 41, 1169 (1967).
38 B. Je±owska-Trzebiatowska, R. Grobelny and W. Wojciechowski, Bull. Acad. Polon. Sc!.
"
Ser. Sc!. Ch!m. 12, 827 (1964).
B. Je±owska-Trzebiatowska, W. Wojciechowski and M. Rudolf, Chem. Zvest!, 19, 229 (1968).
40 B.
Je2owska-Trzebiatowska and M. Rudolf, Bull. Acad. Polon. Sc!., Ser. Sc!. Chim. 17, 419
(1969).
41 B. Je±owska-Trzebiatowska and M. Rudolf, Roczn. Chem. 44 (1969).
42 B. Je±owska-Trzebiatowska and M. Rudolf, Roczn. Chem. 44 (1969).
'
'
R. Colton and G. G. Rose, Austral. J. Chem. 21, 883 (1968).
E. König, Inory, Chem. 8, 1278 (1969).
R. Martin, 'New Pathways !n Inorgan!c Chemistry', p 175. Cambridge University Press:
London (1968).
46 B. Je±owska-Trzebiatowska and W. Wojciechowski, Zh. Strukt. Kh!m. 4, 872 (1963).
B. Je±owska-Trzebiatowska, L. Natkaniec and H. Kozlowski, Bull. Acad. Polon. Sc!., in press.
48 C. J.
50
52
55
Ballhausen and H. B. Gray, Jnorg. Chem. 1, 111(1962).
B. Je±owska-Trzebiatowska and M. Rudolf, unpublished results.
F. A. Cotton, S. M. Morehouse and J. S. Wood, Jnorg. Chem. 3, 1603 (1964).
B. Je2owska-Trzebiatowska, J. Hanuza and W. Wojciechowski, Spectroch!m. Acta, 23A, 2631
(1967).
B. Je±owska-Trzebiatowska and J. Hanuza, Spectroch!m!ca Acta, in press.
B. Je±owska-Trzebiatowska, J. Hanuza and M. Baluka, Acta Phys. Polon. 38, 70 (1970).
B. Jetowska-Trzebiatowska, J. Hanuza and M. Baluka, Acta Phys. Polon. in press.
H. J. Schugar, A. T. Hubbard, F. C. Auson and H. B. Gray, J. Am. Chem. Soc. 31, 71(1969).
111