Yves Jean - Molecular Orbitals of Transition Metal Complexes (793957), страница 3
Текст из файла (страница 3)
The last chapter contains a presentation of basic Group Theory, with applications to some ofthe complexes studied in the earlier chapters. This chapter is placed atthe end of the book so as not to disrupt the flow of the more chemicalaspects of the presentation, but the reader may consult it, if necessary,as and when reference is made to it in the book. Setting the sceneTransition metal complexes are molecules containing one or moremetallic centres (Ti, Fe, Ni, etc.) bound to a certain number of ‘ligands’.These latter may be atoms (H, O, Cl, etc.), molecular fragments (CR 3 ,NR 2 , SH, etc.), or molecules that are themselves stable in the absenceof any interaction with a metal (NR 3 , PR 3 , R2 C==CR 2 ), benzene, etc.).In this book, we shall study the electronic structure of these complexesby molecular orbital (MO) theory. We shall seek to establish the shape,the energetic ordering, and the electronic occupation of the MO; starting from this detailed description of the electronic structure, we shallconsider problems of geometry and reactivity.Certain important aspects of electronic structure can nevertheless beobtained from a far simpler description, which aims merely at providinga formal analysis of the electron distribution in the complex.
Althoughmuch simpler and more limited in its applications, this approach toelectronic structure turns out to be extremely useful, for at least tworeasons:1. It uses classical ideas and ‘language’ that are common to all chemists, such as electronegativity or Lewis structures for the ligands. Itprovides important information, such as the oxidation state (or number) of the metal in the complex, the number of electrons in theimmediate environment of the metal, and what one normally callsthe ‘electronic configuration’ of the complex.2.
In a way which can be a little surprising at first sight, it is veryuseful in the orbital approach when one wishes, for example, toknow the number of electrons that must be placed in the complex’snonbonding MO.There are two ways to obtain this formal distribution of the electrons(or electron count) in a complex.
The first, based on a ‘covalent’ modelof the metal–ligand bond, is mostly used in organometallic chemistry,that is, in complexes which possess one or more metal–carbon bonds.The second, based on an ‘ionic’ model of the metal–ligand bond inwhich the two electrons are automatically attributed to the ligand, ismore frequently employed for inorganic complexes. In fact, the choicebetween the two methods is largely a matter of taste, as they lead, as weshall see, to identical conclusions.Setting the scene1.1. Electron count in a complex: the covalent modelConsider a monometallic complex in which the transition metal M isbound to a certain number of ligands (Lig)i , that may be either atomsor molecules.
It is important to note that, in the covalent model, one alwaysconsiders the ligands in their neutral form (H, Cl, O, CO, CN, PR3 , CH3 ,etc.). Before making the formal electronic assignment for the complex,one must first categorize the ligands according to the nature of theirelectronic structure.1.1.1. Ligand classification (L or X)The main distinction is linked to the number of electrons that the ligandsupplies to the metal’s coordination sphere: if it supplies a pair of electrons, it is a ligand of type L, whereas a ligand that supplies just oneelectron is of type X.
However, some ligands can supply more than twoelectrons to the metal. This notation, introduced by M. L. H. Green, isgeneralized to yield ligands of type Lℓ Xx .1.1.1.1. L-type ligands1There is also a lone pair on the oxygenatom. We shall see later why CO bindspreferentially through the carbon atom(§ 1.5.2.4 and Chapter 3, § 3.2.2).2Complexes in which a dihydrogenmolecule is bound to a transition metalwere first characterized in the mid-1980s,and have since been extensively studied byboth experimental and theoreticalmethods (Chapter 4, § 4.1.4).MHCMHC1-2The simplest case concerns molecules which are coordinated to themetal through a lone pair located on one of their atoms (1-1).
Thesemolecules are L-type ligands, the metal–ligand bond being formed bythe two electrons supplied by the ligand. Examples include aminesNR 3 and phosphines PR 3 which contain a lone pair on the nitrogenor phosphorus atom, the water molecule or any ether (OR 2 ) which canbind to the metal through one of the lone pairs on the oxygen atom.Carbon monoxide is also an L-type ligand, due to the lone pair on thecarbon atom.1NRRRPRORRHHCO1-1There are other cases in which the two electrons supplied by theligand L form a bond between two atoms of that ligand, rather than alone pair. This can be a π-bond, as in the ethylene molecule, or, moresurprisingly, a σ -bond, as in the dihydrogen molecule (1-2).2In these examples, two atoms of the ligand are bound in anequivalent way to the metal centre.
The hapticity of the ligand is said to be2. This type of bond is indicated by the Greek letter η, the nomenclatureused being η2 -C2 H4 or η2 -H2 , respectively (1-2).Electron count in a complex: the covalent model1.1.1.2. X-type ligandsThese ligands supply only one electron to the metal’s coordinationsphere. As neutral entities, X-type ligands are radicals and the metal–ligand bond is formed by the unpaired electron of the ligand and a metalelectron.
Hydrogen (H) is an X-type ligand, as are the halogens (F, Cl,Br, I), alkyl radicals (CR 3 ), the amido (NR 2 ), alkoxy (OR), and cyano(CN) groups (1-3), etc.CClHRONRRRCRRN1-3It should be noted that in some of the examples given above, theradical centre also possesses one or more lone pairs, so that one mighthave considered it to be an L-type ligand. However, the use of a lone pairfor bond formation would lead to a complex with an unpaired electronon the metal (.L:—M).
This electronic structure is less stable than thatin which the unpaired electron and a metal electron are paired to formthe metal–ligand bond ( :X–:–M). It can be seen that in this case, all theelectrons are paired, either as bonding pairs or as lone pairs.1.1.1.3. Ligands of Lℓ Xx type3The ground-state electronicconfiguration for oxygen is 1s2 2s2 2p4 . In theelectronic ground state, two electrons arepaired in one p orbital, while the two other porbitals are singly occupied by electrons withparallel spin (a triplet, following Hund’s rules).For nitrogen (1s2 2s2 2p3 ), there are threeunpaired electrons, one in each p orbital.In a more general notation, ligands can be represented as Lℓ Xx whenthey use ℓ electron pairs and x unpaired electrons to bind to the metal.In the ground state, the oxygen atom possesses two unpaired electrons (1-4a).3 It is therefore a ligand of X2 type, which can bind to atransition metal to form an ‘oxo’ complex.
The sulfido (S) and imido(N-R) (1-4a) ligands behave similarly. Atomic nitrogen, with threeunpaired electrons, is an X3 ligand (1-4b), giving ‘nitrido’ complexes.In each case, one therefore considers all the unpaired electrons on theatom bound to the metal.OS1-4a (X2 )NRN1-4b (X3 )Conjugated polyenes constitute an important family of moleculeswhich are ligands of Lℓ Xx type; they form π-complexes with the metal,that is, complexes in which the π system of the ligand interacts withthe metal centre. Consider, for example, the cyclopentadienyl ligandSetting the scene1-5MM1-6 (η5 , L2 X)Fe1-7MM1-8 (η3 , LX)1-9 (η1 , X)M1-10(η4 , LC5 H5 (also represented Cp), whose Lewis structure (1-5) shows thatthe π system contains five electrons (two π bonds and one unpairedelectron).If this ligand is bound so that all five atoms are essentially at the samedistance from the metal centre (η5 -C5 H5 coordination), the five π electrons are involved in the metal–ligand bonds, so that cyclopentadienylis classified as an L2 X ligand.
Two graphical representations are therefore possible, depending on whether one gives a localized or delocalizeddescription of the ligand’s π system (1-6).Ferrocene is a particularly interesting π complex [Fe(η5 -C5 H5 )2 ](a ‘sandwich’ complex, in which an iron atom is placed between theplanes of two cyclopentadienyl ligands, 1-7). At first sight, one couldconsider it either as a complex with two ligands, [Fe(Lig)2 ] whereLig = C5 H5 , or as one with 10 ligands, [Fe(Lig)10 ], since the iron isbonded equivalently to all 10 carbon atoms.
However, the L/X ligandclassification shows us that each cyclopentadienyl ligand is of the L2 Xtype, so ferrocene is therefore an [FeL4 X2 ] complex in which the ironmust be considered as surrounded by six ligands, rather than two or ten.In fact, it is a pseudo-octahedral complex of [Fe(Lig)6 ] type!Two other coordination modes can be imagined for the cyclopentadienyl ligand, and they are indeed observed in some complexes.If only three π electrons (a double bond and the unpaired electron) aresupplied to the metal’s coordination sphere, C5 H5 acts as a ligand of LXtype. In this case, only three carbon atoms are bound to the metal, andthe coordination mode is η3 -C5 H5 (1-8).
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
