Yves Jean - Molecular Orbitals of Transition Metal Complexes (793957), страница 29
Текст из файла (страница 29)
However, these two interactionsare both bonding between the carbon atoms and the metal centre, sothey contribute to the formation of metal-carbon bonds.The question is thus open: is it appropriate to speak of an ethylenecomplex, in which the ligand is clearly identified as molecular ethylene(3-41a), or is it better to describe it as a metallacyclopropane, with aC−C single bond and two M−C bonds (3-41b)? These are in fact twomesomeric forms of the same complex; the metal oxidation state, inparticular, is different in these two forms, being higher by two in themetallacycle. The change from one limiting form to the other representsthe oxidative addition reaction of ethylene, which leads to the metallacyclopropane. The real complex is usually intermediate between thesetwo limiting forms.
Typically, the carbon–carbon bond length is around1.43 Å (instead of 1.32 or 1.54 Å for a standard double or single bond,respectively), and the ethylene ligand becomes non-planar: the CH2groups become pyramidal, in the direction opposite to the metal centre,though not as extensively as in a real three-membered ring.From the electronic structure point of view, there are similaritiesbetween the two SS and SA bonding molecular orbitals of the π complex(3-42 and 3-43, left-hand side) and the corresponding orbitals of themetallacyclopropane.
The latter can be represented schematically byconsidering the in-phase and out-of-phase combinations of two localizedorbitals, each of which characterizes a σMC bond (3-42 and 3-43, righthand side).10 The first combination corresponds to the orbital π + z2 ofthe molecular ethylene complex, the second to the orbital yz + π ∗ .SS3-42SA3-43π complexes: the example of ethylene3-44 (SS)3-45 (SA)Similarly, there is a resemblance between these orbitals and two ofthe three orbitals that characterize the C−C bonds in cyclopropane itself(3-44, symmetry SS and 3-45, symmetry SA).In fact, depending on the nature of the metallic fragment MLn whichinteracts with the ethylene ligand and on the substituents on this ligand,a whole range of interactions can be observed, ranging from weak interactions, which correspond to a complex of molecular ethylene, to stronginteractions which lead to a structure close to a metallacyclopropane.So the best answer to the general question posed at the beginning of thissection ‘a π complex of molecular ethylene or a metallacyclopropane?’seems in fact to be .
. . both!3.4.3. Metallocenes Cp2 M3.4.3.1. The π system of the cyclopentadienyl ligandThe shapes and the relative energies of the five π orbitals in cyclopentadiene (Cp) are presented in Figure 3.10, where the electronicoccupation corresponds to the anion Cp− (the ionic model). A symbolicrepresentation of the orbitals, which shows only the nodal positionsbetween the carbon atoms, is given on the right of the figure. As a consequence of the three occupied π MO, Cp− can be characterized as anL3 ligand.The donation and back-donation interactions of the Dewar–Chatt–Duncanson model can occur, thanks to the three occupied and the twoempty MO, respectively.3.4.3.2. Cp2 M complexes: the ferrocene exampleThe three occupied MO on each Cp− ligand combine in-phase (π (+) )and out-of-phase (π (−) ), leading to the formation of six occupied MO(+)that are delocalized onto the two ligands.
It is easy to verify that π1Figure 3.10. The π system of thecyclopentadienyl ligand (Cp− ).45231π -type interactionshas the proper symmetry to interact with the s orbital on the metallic(+)(−)centre (3-46a). Similarly, π1 can interact with pz (3-46b), π2 with(−)(−)(+)py (3-46c), π3 with px (3-46d), π2 with yz (3-46e) and π3 withxz (3-46(f )). In this way, six occupied bonding MO are formed thatcharacterize six bonds, as in an octahedral complex.pzs3-46a3-46byzpx3-46dyzxzz2xyx2–y2Figure 3.11. The d block of a Cp2 M complex,where the electronic occupation (d6 )corresponds, for example, to ferrocene[Cp2 Fe].py3-46cxz3-46(e)3-46(f )The participation of two d orbitals in the MO that describe the bondsleads to the presence of two antibonding orbitals in the d block (yz andxz), while the other d orbitals (z2 , x 2−y2 , and xy) make up a block ofthree nonbonding or nearly nonbonding orbitals.
In fact, x 2−y2 and xyare stabilized by bonding interactions with the π ∗ orbitals of appropriatesymmetry on the Cp rings (Figure 3.11).The d block of a Cp2 M complex therefore has the same characteristics as that of an octahedral complex. In ferrocene, [Cp2 Fe], whoseelectronic configuration is (d6 ), the three nonbonding d orbitals areoccupied and this complex can be described as pseudo-octahedral, with18 electrons.3.4.4. Cp2 MLn complexesThere are many complexes of the type [Cp2 MLn ] (n = 1, 3), whereM is a transition metal towards the left of the periodic classification(M = Ti, V, Zr, Hf, Mo, for example). In these complexes, the Cp2 Mfragment is bent, rather than linear as in the metallocenes.
The otherligands are all located in the plane that is perpendicular to that definedby the M atom and the centres of the two Cp rings (3-47 and 3-48).If n = 3, this arrangement imposes very small L−M−L angles (aboutπ complexes: the example of ethyleneHHNb60◦ in Cp2 Nb(H)3 , 3-47). Since each Cp ligand forms three bonds withthe metal, the coordination number of these complexes is seen to bebetween 7(n = 1) and 9(n = 3), which is quite rare for the metals ofthe first three transition series.H3.4.4.1. The bent Cp2 M fragment3-47ClMoCl3-48zyx2a 12a 1Consider the three low-energy d orbitals previously established for alinear Cp2 M complex (Figure 3.11, the z2 , x 2−y2 , and xy orbitals).
In theC2v point group, these orbitals have a1 (z2 and x 2−y2 ) and b2 (xy) symmetries. When the Cp2 M fragment is bent (Figure 3.12), the b2 orbitalis destabilized, since the stabilizing interactions with the π ∗ orbitalson the Cp rings decrease and the repulsive interactions between theM−Cp bonds increase. The same behaviour could be expected for the1a1 orbital (x 2−y2 ), but it mixes with the 2a1 orbital (z2 ), with theresult that its energy stays roughly constant, whereas the 2a1 orbitalis strongly destabilized (an interaction between two orbitals of thesame symmetry). This mixing between z2 and x 2−y2 which accompanies the bending of the fragment leads to a polarization of x 2−y2along the x-axis, and of z2 in the xy plane, as shown to the right ofFigure 3.12.It will be useful to consider another representation of these orbitals,which shows their amplitudes and nodal properties in the symmetryplane xy which interchanges the two Cp ligands (3-49).
The other ligands(L) in [Cp2 MLn ] complexes lie in this plane.b2b21a 1y1a 1xFigure 3.12. Energy changes for the threelowest-energy orbitals of the d block for aCp2 M complex passing from a linear to abent arrangement of the Cp ligands.1a 1b22a 13-493.4.4.2. Cp2 MLn complexes (n = 2, 3)We consider first the complex [Cp2 MoCl2 ] (3-48). It can be decomposedinto a metallic fragment, [Cp2 Mo]2+ , and two chloride ligands, Cl− (theionic model).
The metallic fragment has a d2 electronic configuration:the 1a1 orbital described above is occupied and the two other d orbitals(b2 and 2a1 ) are empty (Figure 3.13, left-hand side). The two Cl− ligandsπ -type interactions2a1b21a1b2a1[Cp2Mo]2+–[Cp2MoCl2]2 ClFigure 3.13. Construction of the MO of thecomplex [Cp2 MoCl2 ] from those on the[Cp2 Mo]2+ and 2 Cl− fragments.supply four electrons that occupy two MO, which are in-phase (a1 ) andout-of-phase (b2 ) combinations of two p orbitals (Figure 3.13, righthand side).
The formation of the MO of the complex from those onthe fragments therefore involves an interaction between b2 and a1 pairsof orbitals. For the latter, the dominant interaction involves the 2a1orbital on the metallic fragment. In fact, the a1 orbital on the chloridesis concentrated in the nodal planes of the 1a1 orbital on the metallicfragment, so the overlap between these two orbitals is very weak.There is thus only one approximately nonbonding orbital in thed block, and it is doubly occupied.
With a d2 electronic configuration,and taking account of the eight metal–ligand bonds, [Cp2 MoCl2 ] istherefore an 18-electron complex. 16-electron d0 complexes are alsoknown in this family (n = 2), such as [Cp2 ZrCl2 ]; the nonbonding dorbital is empty in them.We now turn to a complex from the Cp2 ML3 family, such asCp2 NbH3 (3-47). This can be decomposed into a metallic fragment,Cp2 Nb3+ , whose electronic configuration is d0 , and a fragment thatgroups together the three hydrides, H33− (the ionic model), with sixelectrons in three orbitals. The symmetries of these three orbitals arewell adapted to interact with the three d orbitals of the metallic fragment(3-50). In particular, there is a large overlap between the 1a1 orbital onthe metallic fragment and the 2a1 orbital of H33− .2a1Nb1a 1b2Nbb21a1Nb2a13-50π interactions and electron countingAfter the interaction, three doubly occupied bonding MO are formedthat are mainly concentrated on the hydrides (3-51a).
They characterizethe three Nb−H bonds. The three antibonding combinations, (3–51b)which belong to the d block, are empty.CpCpCpCpCpCp3-51aCpCpCpCpCpCp3-51bThere are therefore no nonbonding d orbitals in this complex.With its d0 electronic configuration and nine bonds around the metal,[Cp2 NbH3 ] is an 18-electron complex.3.5. π interactions and electron counting11Notice, however, that when thesubstituents R are very electronegative, theσ ∗ orbitals associated with the P−R bonds aresubstantially lowered in energy, and thephosphine PR 3 can behave as a double-faceπ -acceptor.
In extreme cases, such as PF3 , itsπ-acceptor strength is even close to that of acarbonyl ligand.The formal electron count, as it is usually performed for transition metalcomplexes, (Chapter 1, § 1.1), takes account of the electrons involved inthe σ interactions and the n nonbonding electrons on the metal whichdo not participate in these interactions (dn electronic configuration). Butit ignores the consequences of the π interactions, even though these areaccompanied by transfers of electron density involving some d-blockorbitals.As a first illustration of this point, consider [W(PR 3 )6 ] and[W(CO)6 ]; both are described as octahedral complexes whose electronic configuration is d6 , with six electrons in the three t2g orbitals ofthe d block of the octahedron.
A phosphine ligand, PR 3 , may, in a firstapproximation, be considered as a simple σ -donor.11 From this viewpoint, the occupied t2g orbitals of the [W(PR 3 )6 ] complex are thereforepure atomic d orbitals, entirely localized on the metal. However, sincethe carbonyl ligand is a π acceptor, the t2g orbitals of the [W(CO)6 ]complex are partially delocalized on to the ligands (3-34), leading to asubstantial transfer of electron density from the metal to the ligands.Although both formally have a d6 electronic configuration, the shape ofπ -type interactions12We noted, in Chapter 2, that complexeswith a d0 electronic configuration can exist ifthe ligands possess lone pairs (see § 2.1.3.2,2.3.3.2, and 2.4.2.2)the orbitals reflects the fact that the electron density at the metal centreis higher in the [W(PR 3 )6 ] complex than in [W(CO)6 ].Complexes with π-donor ligands that can transfer electron densityto the metal centre are even more interesting.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
