You are on page 1of 29
(Olprjolcaaes Quantum chemistry the chemical bond Introductory concepts 14.1 The classification of bonds 142 Potentia-energy curves Valence bond theory 143. Distomic molecules 144 Polyatorc molecules 4145 Promation and hybridization 148 Resonance ‘Molecular orbitals 14.7 Linear combinations of atoric orbitals “a vas 1410 40 Bonding and antibonding orbitals The structures of diatomic molecules Hydrogen and helurm molecules Period 2 diatomic molecules 1422 1433 Symmetry and overlap The electronic structures of homonuclear diatornic molecules 1434 1438 1416 Heteronuclear diatomic molecules The structures of polyatomic malecules ‘The Hilckel method Computational chemistry 14.17 Techniques 14:18 Graphical output 14.19 Appleations ‘CHECKLIST OF KEV IDEAS TABLE OF KEY EQUATIONS ‘QUESTIONS AND EXERCISES ‘The chemical bond, a link between atoms, is central toall aspects of chemistry. Reactions make them and break them, and the structures of solids and individual ‘molecules depend on them. The physical properties of individual molecules and of bulk samples of matter also stem in large part from the shifts in electron density that take place when atoms form bonds. The theory of the origin of the numbers, strengths, and three-dimensional arrangements of chemical bonds between atoms is called valence theory. (The name comes from a Latin word for strength.) Valence theory is an attempt to explain the proper- ties of molecules ranging from the smallest to the largest. For instance, it explains why N, is so inert that it dilutes the aggressive oxidizing power of atmospheric oxygen. At the other end of the scale, valence theory deals with the structural origins of the function of protein molecules and the molecular biology of DNA. The description of chemical bond- ing has become highly developed through the use of computers, and itis now possible to compute details of the electron distribution in molecules of almost any complexity. However, much can also be achieved in terms of simple qualitative understanding of bond formation, and thatis the initial focus of this chapter. ‘There are two major approaches to the calculation ‘of molecular structure, valence bond theory (VB theory) and molecular orbital theory (MO theory). Almost all modern computational work makes use of MO theory, and we concentrate on itin this chapter. Valence bond theory, though, has left its imprint on the language of chemistry, and itis important to know the significance of terms that chemists use every day. ‘The structure of this chapter is therefore as follows. First, we set outa few concepts common to al levels ‘of description. Then we present the concepts of VB theory that continue to be used in chemistry (such as hybridization and resonance). Next, we present the basic ideas of MO theory, and finally we see how computational techniques pervade all current discus- sions of molecular structure. Introductory concepts Certain ideas of valence theory will be well known, from introductory chemistry. This section reviews this background. 14.1 The classification of bonds “We distinguish between two types of bond An ionic bond is formed by the transfer of electrons from one stom to another and the consequent attraction between the ions so formed. A covalent bond is formed when two atoms share «pair of electrons, ‘The character of a covalent bond, on which we concentrate in this chapter, was identified by G. N. Lewis in 1916, before quantum mechanics was fully developed, We shall assume that Lewis's ideas are familiar, but for convenience they are reviewed in Appendix 4.2. In this chapter we develop the modern theory of chemical bond formation in terms of the quantum-mechanical properties of electrons and set. Lewis's ideas in a modern context, We shall see that ionic and covalent bonds are two extremes ofa single type of bond, However, because there are certain aspects of ionic solids that requie special attention, wwe treat them separately in Chapter 15. Lewis's original theory was unable to account for the shapes adopted by molecules. The most clement ary but qualitatively quite successful explanation of the shapes adopted by molecules is the valence-shell electron pair repulsion model (VSEPR model} in which we suppose that the shape of a molecule is determined by the repulsions between electron pairs in the valence shell. This model is fully discussed in Introductory chemistry texts, but we give a brief review of itin Appendix 4,3. Once again, the purpose of this chapter is to extend chese elementary argu ments and to indicate some of the contributions that quantum theory has made to understanding why a molecule adopts its characteristic shape. 14.2 Potential-energy curves All theories of molecular structure adopt the Born— Oppenheimer approximation. In this approximation, it is supposed thar the nuclei, being so much heavier A Internuclea distance, A» 141 molecule parental energy curve. The eaulnurs 'dlongth R, corresponds fo te energy min murs ft =), than an electron, move relatively slowly and may be treated as stationary while the electrons move around, them, We can therefore think of the nuclet as being. fixed at arbitrary locations, and then solve the Schrddinger equation for the electrons alone. The approximation is quite good for molecules in their electeonic ground states, for calculations suggest that {in classical terms) the nuclei in H move through only about 1 pm while the electron speeds theough 1000 pm, By invoking the Born~Oppenheimer approximation, ‘we can select an internuclear separation in a diatomic molecule and solve the Schrddinger equation for the clecttons for that nucleae separation. ‘Then we can choose a different sepazation and repeat the ealeula- tion, and so on. In this way we can explore how the ‘energy of the molecule varies with bond length and obtain a molecular potential energy curve, a graph showing how the molecular enezgy depends on che internuclear sepazation (Fig. 14.1). The geaph is called a potential enexgy curve because the nuclei ace stationary and contribute no kinetic energy. Once the curve has been calculated, we can identify the equilibrium bond length, X,, the internuclear separa tion at the minimum of the cuzve, and D,, the depth. of the minimum below the energy of the infinitely widely separated atoms. In Chapter 19 we shall also see that the narrowness of the potential well is an ndication of the stiffness of the bond, Similar con- siderations apply to polyatomic molecules, where bond angles as well as bond lengths may be varied. Valence bond theory In valence bond theory, a bond is regarded as forming ‘when an electron in an atomic orbital on one atom pairs its spin with that of an electron in an atomic 4324 CHAPTER 14; QUANTUM CHEMISTRY: THE CHEMICAL BOND oe @ 1a) o 142 In the valence bond theory, a 6 bond is formed shen two electrons in orbitals on naighoourng atoms, a3 in (a), pair and the orbitals merge to fr a eyndcal elcvron ‘lous, 35 orbital on another atom (Fig. 14.2). To understand why this pairing leads to bonding, we have to exam- ine the wavefunction for the two electrons that form the bond. 14.3 Diatomic molecules ‘We begin by considering the simplest possible chem- ical bond, the one in molecular hydrogen, H—H. When the two ground-state H atoms are far apart, swecan be confident that electron 1 isin the 1s orbital ‘of atom A that we denote y(t), and slectron 2s the Isorbital of atom B, which we denote y4(2)..We saw in Section 13.1 that it is a general rale in quantum rmechanies that the wavefunction for several non- interacting particles is the product of the wave- functions for each particle (this is the separation of variables argument) so providing we can ignore the interactions between the electrons we can write yi1,2) = val1iyg(2) ‘When the two atoms are at their bonding distance, it may still be trae that electron 1 ison A and electron, 2 is on B. However, an equally likely arrangement is for electron 1 to escape from A and be found on B and for electron 2 to be on A. In this case the wave- function is y(1,2) = wal2}yQ(1) ‘Whenever two outcomes are equally likely, the rules ‘of quantum mechanics tell us to add together, formally 10 superimpose, the two corresponding wavefunc- tions. Therefore, the (unnormalized) wavefunction for the two electrons in a hydrogen molecule is Vir-r(L2)= va(liva(2) + vaC2)¥—(1) (14.1) ‘This expression is the VB wavefunction for the bond jn molecular hydrogen. It expresses the idea that wwe cannot keep track of either electron and their dis- tributions blend together. The wavefunction is only an approximation, because when the two atoms are close together it is not true that the electrons do not interact. However, this approximate wavefunction is a reasonable starting point for all discussions of the VB theory of bonding. ‘We show in Derivation 14.1 that for technical reasons related to the Pauli exclusion principle, the wavefunction in eqn 14.1 can exist only if the two clectrons it describes have opposite spins. It follows that the merging of orbitals that gives rise toa bond is accompanied by the pairing of the two electrons that contribute to it. Bonds do not form because elec trons tend to pair: bonds are allowed to form by the clectrons pairing their spins, Derivation 14.1 ‘The Pauli principle and bond formation ‘The VB wavefunction in eqn 14.1 does not change sign wynen the lates? ard 2 are interchanged, To formulate 2 wavefunction thot obeys the Poul principle (Further information 13.1) and does change sign when the lobels 1 end? are incerchangea, we must combine this syonme- ‘vc spatial wavefunction yt g(2) with the antisymmetic spin function a 912) ~ BIT} and write Wsi2.0) =iyg( tly 2+ wal2IyaK1 fat BthaaD ‘This isthe ony permitted combination of space and sein ‘functions, Beesuse a6} ~ fITja2} represents spin- pated state of the two electrons, we see thatthe Paul principle roqures the to electrons i tho bond to be pated (1), Because y is built from the merging of Ills orbitals, we can expect the overall distribution of the electrons in the molecule to be sausage-shaped (as in Fig. 14.2). A VB wavefunction with eylindrical sym- metry around the internuclear axis s called a @ bond. ‘The bondis so called because, when viewed along the internuclear axis it resembles a pair of electrons in an s orbital (and o, sigma, isthe Greek equivalent of 5. All VB wavefunctions are constructed in a similar way, by using the atomic orbitals available on the participating atoms. In general, therefore, the (un- normalized) VB wavefunction for an A~B bond is Ya-all2)= valtve(2) + val2lyg(t) (14.2) To calculate the energy of a molecule for a series of internuclear separations R, we substitute the VB wavefunction into the Schrodinger equation for the molecule and carry out the necessary mathematical manipulations to calculate the corresponding values of the energy. When this energy is plotted against R, Fig. 143 The normalized) electron density in H, according to the valence-bond model of the chemical Bond and the electron densities cortesponding to the contributing amie title. The nucio\ ae denaze by large dots on the herzon- talline. Note the sceumulation of electron censity in the internuclear gion we get a curve like that shown in Fig. 14.1, As R decreases from infinity, the energy falls below that of ‘wo separated H atoms as each electron becomes free to migrate to the other atom. This decrease in energy as the outcome of several effect: ‘+ Asthe two atoms approach each other, there is an accumulation of electron density berween the two nuclei (Fig. 14.3). The electrons attract the two nuclei, and the potential energy is lowered. ‘+ This accumulation between the nuclei is at the expense of removing electron density from close to the nuclei, which contributes an increase in potential energy. ‘+ The freedom of the electrons to migrate between, the atoms is like the transfer of an electron from. a small box to a bigger box, which (as we saw in the discussion of a particle in a box) results in a lowering of their kinetic energy. In Hy the last is the dominant effect, but the relative importance of changes in potential and kinetic energy is still unclear in more complex molecules ‘The overall decrease in energy due to the redistribu- sion of electrons is counteracted by an increase in energy from the Coulombic repulsion berween the two positively charged nuclei of charges Ze and Ze, which has the form ZaZae2 (143) “4 (For II,, Zy = Zp = 1.) This positive contribution to the energy becomes large as R becomes small (and the decrease in electronic kinetic energy becomes less significant as the ‘big box’ is no longer much bigger than the initial two ‘litvle boxes’). As a result, the total energy curve passes through a minimum and then climbs to a strongly positive value as the two nuclei are pressed together. VALENCE BOND THEORY 325 Fig, 16.4 The bonds in Ny oe bul by allowing the electrons Inthe N2p arias 1 pai. However, enly one orbital on each stam can form saben the ovals perpendicular to the ax form monde, A similar description is used for molecules built from atoms that contribute more than one electron to the bonding. For example, to construct the VB description of Ny, we consider the valence-electron configuration of each atom, which is 2s¢2p! 2p! 2p! Itis conventional to take the z-axis to be the inter- nuclear axis, so we can imagine each atom as having a 2p, orbital pointing towards 2 2p, orbital on the other atom, with the 2p, and 2p, orbitals perpendicular to the axis (Fig. 14.4). Each of these p orbitals is occu pied by one electron, so we ean think of bonds as being formed by the merging of matching orbitals on neighbouring atoms and the pairing of the electrons that occupy them. We get a cylindrically symmetric o bond from the merging of the two 2p, orbitals and the pairing of the electrons they contain. However, the remaining p orbitals cannot merge to give @ bonds because they donot have cylindrical symmetry around the internuclear axis. Instead, the 2p, orbitals merge and the two electrons pair to form 2 bond. A bond is s0 called because, viewed along the inter- nuclear axis, it resembles a pair of electrons in ap orbital (and wis the Greek equivalent of p). Similarly, the 2p, orbitals merge and their electrons pair to form another m bond. In general, a bond arises from the merging of two p orbitals that approach. side-by-side and the pairing ofthe electrons that they contain. It follows that the overall bonding pattern in N) is a.@ bond plus two m bonds, which is consistent with the Lewis structure :N=N? in which the atoms are linked by a triple bond. Solt-test 14.1 Desevbe the VB ground stete of «Cl, molecule Lrsnaene o(tp, C18 bons 326 CHAPTER 14: QUANTUM CHEMISTRY: THE CHEMICAL BOND lL y ig 145 Tho bonding in an HO molecule canbe pictured in terms ofthe paving ofan alecron belonging to one Kom wih an alectron in an O2p extal the otner bond is formed Thewise, out using s perpenciuler O29 orl. The precicwed boond angle is 80%, which is in poor agreement with the ‘experimental bond angle (104 14.4 Polyatomic molecules Each 6 bond in a polyatomic molecule is formed by the merging of orbitals with cylindrical symmetry about the relevant internuclear axis and the pairing ‘of the spins of the electrons they contain, Likewise, bonds are formed by pairing electrons that occupy atomic orbitals of the appropriate symmetzy (broadly speaking, of the appropriate shape}, © Abrictillustration The valence electron configuration ‘of an atoms 2s%2p%2} 2p} The two unpaired electrons inthe O2p oritels ean esen pair with an electron in ats cial, and each combination results in “he formation of ‘bona (each bond has evndtcal symmetry about the respective Oi intermucleor distance). Becouse the 20, and 2p, otitels leat 90° to each athe, the 0 0 bonds ‘they form sao i a 90° to eaen ater Ig. 14.5). We pre dict therefore, that 0 shaulé be an engulr melecul, ‘whic itis. However, te model precisa bord ange of 90°, whereas the actual bond anges 104%. @ Selttest 142 Give 6 VB description of NH, and pracict the bond engle ‘ofthe molecule on the basi o his descrston Lnsnetee iN2p 1st bonds 3 While broadly correct, VB theory seems to have :wo deficiencies. One is the poor estimate it provides for the bond angle in 11,0 and other molecules, such as NIT,. Indeed, the theory appears to make worse pre- dictions than the qualitative VSEPR model, which predicts HOH and HINET bond angles of slightly less than 109° in IT,O and NIT, respectively. The second ‘major deficiency isthe apparent inability of VB theory to account for the number of bonds that atoms can form, and in particular the tetravalence of carbon, ‘To appreciate the laiter problem, we note that the ground-state valence configuration ofa catbon atom is 2542p! apt, which suggests cha it should be eap- able of forming only two bonds, not four. 14.5 Promotion and hybridi Two modifications solve all these problems. ies, we allow a valence electron to be promoted fom a fal atomic orbital to an empty atomic orbital as a hond js formed: that results in two unpaired electrons instead of two paired electrons, and each unpaired electron can participate in bond formation. In ear- bon, for example, the promotion of 2s electron toa 2p orbital leads to the configuration 2s!2p! 2p! 2p}, with four unpaired electrons in separate orbitals These electrons may pair with four electrons in orbitals provided by four other atoms (such as four His orbitals ithe molecule is CH), and asa result the atom can form four 6 bonds. Promotion is worthwhile ifthe energy it requires ean be mote than recovered inthe greater strength or numberof bonds that ean be formed. ‘We can now sce why tetravalent carhon is so common. The promotion energy of carbon is small because the promoted electron leaves a doubly occu- pied 2s orbital and enters a vacant 2p orbital, hence significantly relieving the electron-clectron repulsion ivexperiences in the former. Furthermore, the energy requited for promotion is more than recovered by the atoms ability to form four bonds in place of the two bonds ofthe unpromoted atom. Promotion, however, appeats to imply the presence of three o bonds of one type (in Cy, from the merg- ing of [11s and C2p orbitals) anda fourth 6 bond of a dissinety different type (formed from the merging of Is and C23). Ie is well known, however, that all four bonds in methane are exactly equivalent both in ‘terms of their chemical properties and their physical properties (their length, strengths, and sifinesse}. This problems overcome in VB theory by drawing oon another technical feature of quantum mechanies that allows the same electron distribution 0 be described in different ways. In this ease, we can describe the electron distribution in the promoted atom cither as arising from four electrons in one 8 and three p orbitals, or as arising from four electrons in four different mixtures of these orbitals. Mixtures {more formally, linear combinations) of atomic orbitals on the same atom are called hybrid orbitals. ‘These wavefunctions interfere destructively or con- seruetively in different regions and give rise to four bond sp" hybrid Fig. 146 The 2s and three 2p orbitals of 2 carbon ators hybridize, andthe resulting hybrid cxbatals point towerds the comers of regul ‘evaheson. Each 0 bond is formed by ‘the warring ofan elecsronin an Hs eral witn an electron in ane of the hyd aaitals, The resulting molecule ' regu tetrahedra new shapes (we saw in Section 12.3 that interference isa characteristic of waves}. The specific linear com- binations that give rise to four equivalent hybrid orbitals are +P tP, HPL Pa=S—Pe—Py +P e PEEP, Me=S4 PL PLP e A brief comment In general, linear combination of swe functions fand g's e\f+ 9, where e, and ¢, sre numerical costficints, 3 a Inear combination is & mote general term than ‘sur Ine um, = ¢ = (44) Asa result ofthe constructive and destructive inter ference between the positive and negative regions of the component orbitals, each hybrid orbital has a large lobe pointing towards one corner of a regular tetrahedron (Fig. 14.6). Because each hybrid is built from one s orbital and three p orbitals, iti called an, sp* hybrid orbital Te is now easy to see how the valence bond descrip- tion of the methane molecule leads to a tetrahedral molecule containing four equivalent C—IT bonds. It is energetically favourable (in the end, after bonding, has been taken into account] for the carbon atom, to undergo promotion. The promoted configuration, has a distribution of electrons that is equivalent to fone electron occupying each of four tetrahedral hybrid orbitals. Each hybrid orbital of the promoted atom contains a single unpaired electron; a hydrogen, ‘Is electron can pair with each one, giving rise to a 9 bond pointing in a tetrahedral direction. Because cach sp* hybrid orbital has the same composition, all, four o bonds are identical apart from their orienta tion in space. Hybridization is also used in the VB description of alkenes. An ethene molecule is planar, with ITCH and IGG bond angles close to 120°. To reproduce o Fig. 1427 (a) Tigonal planar hybrid 2aton is obteined when ans and two p orbitals are hybridized. The three lobes iin {plane and make an angle of 120" to each athe. (3) The Femaining © orbital nthe valence shel of an sa-nybrisizes stam les perpenciculer to the plane ofthe three hybris. this o-bonding structure, we think of each C atom as being promoted to a 2s!2p!2p!2p! configuration. However, instead of using all four orbitals to form hybrids, we form sp? hybrid orbitals by allowing the s orbital and two of the p orbitals to interfere. As shown in Fig. 14.7, the three hybrid orbitals 542! p, +N Pe ~ GM pe GB, lie in plane and point towards the corners of an equilateral triangle. The third 2p oxbital (2p. is not inchuded in the hybridization, and its axis is per- pendicular to the plane in which the hybrids lie. The coefficients 2", etc in the hybrids have been chosen to give the correct directional properties ofthe hybrids “The squares ofthe coefficients give the proportion of cach atomie orbital in the hybrid. All three hybrids have sand p orbitals inthe ratio 1:2, as indicated by the label sp* ‘The sp!-hybridized C atoms each form three bonds with either the by hybrid ofthe other C atom or with the ITs orbitals. The o framework therefore consists of bonds at 120° to each other. Moreover, provided the two CIT, groups lie in the same plane, the two electrons in the unhybridized C2p, orbitals «an pair and form a x bond (Fig. 14.8). The forma- tion of this © bond locks the framework into the planar arrangement, for any roration of one CH, ‘group relative to the other leads to a weakening of the x bond (and consequently an increase in energy of the molecule}. A similar description applies to a linear ethyne {acetylene} molecule, I—C=C-H Now the carbon atoms are sp hybridized, and the @ bonds are built from hybrid atomic orbitals ofthe form hy be (a4) 4s) pe Note that the s and p orbitals contribute in equal proportions. The two hybrids lie along the z-axis. 4328 CHAPTER 14; QUANTUM CHEMISTRY: THE CHEMICAL BOND Fig. 14.8 The vslence-bond description ofthe structure of ‘atbor-cerbon double band, asin ethene, The electrons in ‘the two $9 nya that point towards each other pair and form a 9 band. Eletvons nthe two p oritals that are pe. penaieulor te the plane of the hybrids par, and form ar bora Te elocirons in the remaining hybrid araials are usec to ‘orm bonds to other atoms fin ethene ite, 0 H stom Fig. 189 The elecvonie structure of athyne acetylene, The ‘slocons in the two sp hyords on each atom pair to form 0 bonds ther with he other C stom ar with an ator. There maining two unhybrdized 2p oroitals on each atom are per enduro the 9 the electrons in eovresponcling orbitals ‘on each atom gairto form Iwo Lords. The overall electron distribution 's evinces, ‘The electrons in them pair either with an electron in the coreesponding hybrid orbital on the other C atom ‘or with an electeon in the His orbitals. Electeons in the two remaining p ozbitals on each atom, which are perpendicular to the molecular axis, pair to formtwo perpendicular x bonds (as in Fig. 14.9). ‘Other hybridization schemes, particularly those involving d orbitals, are often invoked to account for (or at least be consistent with) other molecular geo- rmetties (Table 14.1). An important point to note is that the hybridization of N atomic orbitals always results in the formation of N hybrid orbitals. For example, sp'd? hybridization results in six equivalent hybrid orbitals pointing towards the corners of a regular octahedron, This octahedral hybridization scheme is sometimes invoked to account for the structuze of octahedral molecules, such as SF,. Solttest 143 Describe the bonding ina PC, molecule in VB arms. Anse: Sve o bends fred rom spy on Table 14.1 Hybrid orbitals Number Shape Hybridization® 2 Linear 9 3 Triganel planer sot a Tetrehodral so 5 Trigonal bipyramidal p's 6 Octahectl sod * otter combnatons ae posse °30- 08 Tie Tao Bond angle 9" "ao 103 Tor 108 708 Tio ‘utotplane angle, 6 Fig. 14.10 The vataton ot hyoriiztion with bond angle in {a} angular, tiganal pyramidal molecules. The vaical axis 19¥e8 the rato at p 10'S cheracter, so high velues indeate mostly p charactor ‘The ‘pure’ schemes in Table 14.1 are not the only possibilities: it is possible to form hybrid orbitals ‘with intermediate proportions of atomic orbitals. For example, as more p-orbital character is included in an sp-hybridization scheme, the hybridization changes towards sp" and the angle between the hybrids changes continuously from 180° for pure sp hybridization to 120° for pure sp* hybridization. Ifthe proportion of p character continues to be increased (by reducing the proportion of s orbital), then the hybrids eventu- ally become pure p orbitals at an angle of 90° to each, other (Fig. 14.10). Figure 14.11 shows contour plots of hybrid orbitals as the ratio of 2p character to 2s, character increases. Now we can account for the structure of 11,0, with its bond angle of 104°. Each, O-II 9 bond is formed from an O atom hybrid orbital with a composition that lies between pure p Fig, 14.11 Contour plots showing the amplitudes of so" hybrid orbitals. To construct these pls, we heve used nyorogene 2s and 2p orbs (which would lead to a bond angle of 90°) and pure sp? (which would lead to a bond angle of 120"). The actual bond angle and hybridization adopted are found by calculating the energy of the molecule as the bond angle is varied, and looking for the angle at which the energy is a minimum, 14.6 Resonance ‘Another term introduced into chemistry by VB theory is resonance, the superposition of the wave- functions representing different electron distribu. tions in the same nucleas framework, To understand what this means, consider the VB description of a purely covalently bonded HCl molecule, which could be written Yee 2) = val) ¥e2) + Vul2) Vet) We have supposed that the bond is formed by the spin paiting of electrons in the Hs orbital, iy and the Clap, orbital, eq. However, there is something ‘wrong with this descrsption:it allows electzon 1 to be oon the Hatom when electoa is on the Clatom, and vice versa, but it does not allow for unequal sharing of electeon density between the atoms. On physical grounds, we should expect the purely covalent char- acter of HCL to be an incomplete description of the molecule: because the Cl atom has higher ionization energy and electron affinity than the H atom, we ean expect the ionic form HCI to play azole. The wave- fanetion for this ionic structure, in which both ele- eons ate in the Cl2p, orbital, s Yevert 2) = Ve HWel2) Towever, this wavefunction alone is unrealistic, because TIClis not an ionic species. A better descrip tion of the wavefunction for the molecule is as a superposition of the covalent and ionic descriptions, and we write (with a slighty simplified nozation) Vise.= Vier * Wren ‘with A (lambda) some numerical coefficient. In gen- eral, we write Y= Veovslene * Moric (14.7) WHEE Vigyainr i the wavefunction for the purely covalent form of the bond and Vga. is the wavefunc- tion for the ionic form of the bond. According to the ‘general rules of quantum mechanics, in which prob- abilities are related to squares of wavefunctions, we interpret the square of Aas the relative proportion of the ionic contribution. If 2? is very small, the cova- lent description is dominant. If 2 is very large, the ionic description is dominant. We find the numerical value of 2 by using the variation theorem. First, we write down a plausible ‘wavefunction, atrial wavefunetion, forthe molecule, such as the wavefunction in eqn 14.7 where 2 is a variable parameter. The variation theorem then states that ‘The energy of a trial wavefunction is never less than the trac energy. ‘The theorem implies that if we vary 2 until we achieve the lowest energy, then the wavefunction with that value of 2s the best available of that par- ticular kind ‘The approach summarized by eqn 14.7, in which swe express a wavefunction as the superposition of ‘wavefunctions coreesponding to a variety of struc- tures with the nucle in the same locations, i called resonance. In this ease, where one structure is pure sovalent and the other pure ionic, it is called ionie-covalent resonance. The interpretation of the wavefunction, which is calle a resonance hybrid, is that if we were to inspect the molecule, then the probability that it would be found with an ionic stracture is proportional to 22. For instance, we might find that the lowest energy is reached when 2.= 0.1, s0 the best description of the bond in the molecule in terms of a wavefunction like that in eqn 147 isa resonance structure deseribed by the wave- function Y= Yoouier + O-1 Vine This wavefunction implies thatthe probabilities of finding the molecule im its covalent and ionic forms are in the ratio 100:1 (because 0.1? = 0.01). 30. CHAPTER 14; QUANTUM CHEMISTRY: THE CHEMICAL BOND ‘One of the most famous examples of resonance is in the VB description of benzene, where the wave- funetion of the molecule is written asa superposition of the wavefunctions of the two covalent Kekulé structures (1) and (2): o-9 Y= Wis Wie ‘The two contributing structures have identical ener sis, so they contribute equally to the superposition The effect of resonance (which is represented by a double-headed artow) in this case isto distribute ddouble-bond character around the ring andro make the lengcs and strengths ofall the carbon-carbon bonds identical. The wavefunction is improved by allowing resonance because it allows for a more Accurate description ofthe location ofthe electrons, fndin particolar the distribution can adjstintoa sare of lower energy. This lowering is called the resonance stabilization ofthe molecule and in the context of VB theory, is largely responsible for the unusual stability of aromaticrngs. Resonance always lowers the energy, and the lowering is greatest when the contributing structures have similar energies. The wavefunction of benzene is improved sill further, and the calculated energy of the molecule is lowered further sill, if we allow jonic-covalent resonance oO - ‘00, by allowing a small admixture of struc- ‘tures such as that shown in (3). 3 Resonance is not a flickering between the con- tributing states; it is a blending of their character- istics, much as a mule is a blend of a horse and a donkey. It is only a mathematical device for achiev- ing a closer approximation o the true wavefunction of the molecule than that represented by any single contributing structure alone (14.8) Molecular orbitals In molecular orbital theory, electrons are treated as spreading throughout the entire molecule: every elec- tron contributes to the strength of every bond. This theory has been more fully developed than valence bond theory and provides the language that is widely used in modern discussions of bonding in small inorganic molecules, d-metal complexes, and solids. To introduce it, we follow the same strategy as in Chapter 13, where the one-electron hydrogen atom, was taken as the fundamental species for discussing atomic structure, and then developed into a descrip- tion of many-clectron atoms. In this section we use the simplest molecule of all, the one-electron hydro- gen molecule-ion, IT}, to introduce the essential features of bonding, and then use it as a guide to the structures of more complex systems. 14.7 Linear combinations of atomic orbitals A molecular orbital is a one-electron wavefunction that spreads throughout the molecule. The mathem- atical forms of such orbitals are highly complicated, even for such a simple species as II, and they are sunknown in general. All modern work builds appro ximations to the true molecular orbital by formulat- ing models based on linear combinations of the atomic orbitals on the atoms in the molecule. First, we recall the general principle of quantum mechanies—which we used earlier to construct VB wavefunctions—that if there are several possible outcomes, then we superimpose—add together—the ‘wavefunctions that represent those outcomes. In IT3, there are two possible outcomes: because an electron, spreads throughout the molecule, it may be found cither in an atomic orbital centred on A, yg, or in an. orbital centred on B, vq. Therefore, we write VE CAYK OR (14.92) where cy and cy are numerical coefficients. A wave- fanction constructed in this way is called a linear combination of atomic orbitals (LCAO) and the corresponding molecalar orbital is called an LCAO- MO. The squares of the coefficients tellus the rela- tive proportions of the atomie orbitals contributing to the molecular orbital. In a homonuclear diatomic molecule an electron can be found with equal prob- ability in orbital A or orbital B, so the squares of the coefficients must be equal, which implies that eq = ey. The two possible (annormalized) wave- functions are therefore vevtve (14.96) First, we consider the LCAO with the plus sign, Y= Wy + Ypsas this molecular orbital will urn out to have the lower energy of the two. The form of this, orbital is shown in Fig. 14.12. Ii called a 6 orbital because it resembles an s orbital when viewed along, the axis. More precisely, it is so called because an. electron that occupies a 6 orbital has zero orbital angular momentam around the internuclear axis, just C C C wo C & Region of constructive interference Fig, 14,12 The formation of a bonding molecule’ ertitel (eo crit fo) Two Hs oratals carne together. (a) The atorsc ‘ibtels overlap, interfere constructively, ond give ise to on fenhaneed amplsude in ne nterctear region The resulting ‘ibtel nas eylndheel symmetry about the interaucles axis. ‘When itis eceupied by two pated electrons, 0 give the con- figurtion & we hove © 0 nd Voce Yan Wavefunetion, Fig. 14.13 The jnormalzed boning malecular arial wave- function along the intornucloar axis. Note that there is an enhancement of empltude between the nucle, so there son increased probabilty of finding the bonding electrons in that region as an electron has zer0 orbital angular momentum around an axis passing through the nucleus. Because, as we shall see, itis the @ orbital of lowest energy, it islabelled 16. An electron that occupies a 6 orbital is called a 6 electron. In the ground state ofthe 13 ion, there is a single Io electron, s0 we report the ground- state configuration of TTf as 10! ‘We can sce the origin of the lowering of energy that is responsible forthe formation of the bond by examining the LCAO-MO. The two atomic orbitals ate like waves centred on adjacent nuclei. In the imernuclear region, the amplitudes interfere con- structvely and the wavefunction has an enhanced amplitude there (Fig. 14.13). The three contributions that we listed for bonding in VB theory (Section 143) apply here too: there isan accumulation of electron density between the two nuclei, removal of electron density from close to the nuclei, and a lowering of MOLECULAR ORBITALS Fig, 1414 J schematic representation of the contibutions te the overlap integral. a) $= 0 because the orbitals are far ‘part and their producti always smal) Sis lege fout less than 1) because the product iis large over a substantial region. Ic) S = 0 beceuse the positive region of overlap is ‘exactly cancelled by the negatve region. kinetic energy as a result of the electron spreading over both nuclei The accumulation of probability density in che internuclear region is measured by the overlap integ- ral, 5. As we show in Derivation 14.2, when S= 1 there is perfect overlap between two atomic orbitals, when $= 0, there is no overlap at all (Fig. 14.14). Broadly speaking, the greater the overlap integral, the stronger is the bonding effect of electrons in the molecular orbital they form. Derivation 14.2 Overlap integrals An overhpinograis ckeulatedby dvising space up into ‘large numberof small regions, mutialying together the values of yy andy im each regio, then adding together fntograting| the resuling predicts for all the regions Formally, we express this rule by wating s-[var (4.10) ‘whore a (dee tau) i an infintesimal volume element Hor instance, dr =dxsyazin three-d mensional Carcsian ‘oottinetes. ti small wherever ys large, ena vice versa (such as when two hyerogen nucle are far apart, the aroducts reall small anc the integrals alse smal this comtesponds to small value of 5. Atypical bonding lstances, ys nd yg are bath lroein he internuceer re ‘g0n,s0theirprocuts thera are ltge, and the integrals alse large: this corresponds toa value of Sapproaching 1 typically, about 0.4.1 the to nucle’ ae concent the ‘wo atomie orbitals have identical values everywhere, nc the integral ofthe" proauets gives Its possible, but rot easy, to evaluate the overlap integral fr nyeogone oral, nd for two tsoritls on hycrogen nuclei separated by dstance F, the resus sais i can aa 332. CHAPTER 14; QUANTUM CHEMISTRY: THE CHEMICAL BOND Tnternucear seperation, Fig. 14.15 The variation of the overlap integral wth internuclear astance for wo His oroitas, This function is plotted in Fig {ctor guarantees thatthe overlap integral goes to 260 atlarge separations 4.15, The exponential 14.8 Bonding and antibonding orbitals A 1o orbital is an example of a bonding orbital, 2 molecular orbital tha, if occupied, contributes to the strength of a bond berween two atoms. Asin VB theory, we can substitute the wavefunction Y= V4 + Vj into the Schrodinger equation for the molecule-ion with the nuclei ata fixed separation R and solve the equation for the energy. The molecular potential energy curve obtained by plotting the energy against Riis very similar to the one drawn in Fig. 14.1. The energy of the molecule falls as R is decreased from large values hecause the electron is increasingly likely to be found in the internuclear region as the two atomic orbitals interfere more effectively. However, at small separations, there is too litle space between the nuclei for significant accumulation of electron density there. In addition, the nucleus-nucleus repol- sion Vescoue (given in eqn 14.3} becomes large and the kinetic energy of the electron is not lowered by very much. As a result, after an inital decrease, at small internuclear separations the potential energy curve passes through a minimum and then rises sharply to high values. Calculations on IL} give the equilibrium bond lengths 130 pm and the bond dis- sosiation energy as 171 kJ moll; the experimental values are 106 pm and 250 kJ mol", so this simple LCAO-MO description of the molecule, while inac- curate, is not absurdly wrong. Now consider the alternative LGAO, the one with ‘a minus sign: Y= Yq ~ yp. Because this wavefunction e @ co) structive interference Fig, 14,16 The formation ofan antibonding molecular orbital (2 0* orbital. fol Two His orbitals come together. tb) The atomic ortals overlap with opposite signs las depicted by Gtteron shades of grey, intertare destructively, and gv rise to decreased empit in the internuclear region. There is «2 nodal plane exactly hallway between ine nucle on which any electrons that occuny the orbital wil nt be found, is also cylindrically symmetrical around the inter- nuclear axis itis also a ¢ orbital, which we denote 1o* (Fig. 14.16}. When substituted into the Schrédinger equation, we find that it as a higher energy than the bonding Io orbital and, indeed, it has a higher energy than either of the two atomic orbitals Selltest 14.4 Show that the molecule otal wtten above is zero on fa plane curing tough te internuclear axis et ts mid point. Take eacn atomic orbital tbe ofthe form a, ‘with qmeesured ‘tom nucleus A any measured fom nucleus 8. Unser Thestome opal ancl or ales ‘sistant om the ru ‘We can trace the origin of the high energy of the 16 orbital to the existence of a nodal plane, a plane ‘on which the wavefunction passes through zero. This, plane lies half-way between the nuclei and cuts through the internuclear axis. The two atomic orbitals cancel on this plane as a result of their destructive interference, because they have opposite signs. In drawings like those in Figs 14.13 and 14.16, we represent overlap of orbitals with the same sign (asin the formation of 10) by shading of the same tit; the overlap of orbitals of opposite sign (asin the forma- tion of 16") is represented by one orbital of a light, tint and another orbital of a dark tint. ‘The 16° orbital is an example of an antibonding orbital, an orbital that, if occupied, decreases the strength of a bond between two atoms. The anti- bonding character of the 10° orbital is partly a result of the exclusion of the electron from the internuclear Weavetunetion, Fig. 14.17 The inormalzed) antbonding molecular orbs wavefunetion slong the intemucleat ax’, Note tht there is 8 decrease in armpiude between the nucle, so there is @ ‘decreased probstility 0" finding the banding electrons in thet rego. region and its relocation outside the bonding region, where it helps to pull the nuclei apart rather than, pulling them together (Fig. 14.17). An antibonding orbital is often slightly more strongly antibonding than the corresponding bonding orbital is bonding although the ‘gluing’ effect of a bonding electron and the ‘antigluing’ effect of an antibonding eleetron are similar, the nuclei repel each other in both cases, and this repulsion pushes both levels up in energy ‘We need to be aware of a few points regarding notation, For homonuclear diatomic molecules, it is helpful to identify the inversion symmetry of a molecular orbital, especially when discussing elec- tronic transitions (Chapter 18). By “inversion sym- metry’ ir meant the behaviour of a wavefunction, when it is inverted through the centre (more form= ally, the centre of inversion) of the molecule. Thus, if wwe consider any point of the ¢ bonding orbital, and then project it through the centre of the molecule and out an equal distance on the other side, then we arrive at an identical value of the wavefunction (Fig. 14.18). This so-called gerade symmetry {from the German word for ‘even’) is denoted by a subscript g, as in G,, On the other hand, the same procedure applied to the antibonding 6° orbital results in the same size but opposite sign of the wavefunction. This ungerade symmetry (‘odd symmetry’) is denoted by a subscript u, as in G,, This inversion symmetry Fig, 14.18 The goradelungerade character of @ bonding and sniboncing orbitals, classification (or ‘parity’) is not applicable to hetero- nuclear diatomie molecules (like CO) as they do not have a centre of inversion. 14.9 The structures of diatomic molecules In Chapter 13 we used the hydrogenic atomic orbitals and the building-up principle to deduce the ground electronic configurations of many-electron atoms. Here we use the same procedure for many-electron diatomic molecules (such as H, with two electronsand. even Br, with 70), but using the Hj molecular orbitals as a basis. The general procedure is as follows: 1. Construct molecular osbitals by forming linear combinations of all suitable valence atomic orbitals supplied by the atoms (the meaning of suitable’ wall be explained shorty}; N’ atomic orbitals result in N moleculae orbitals 2. Accommodate the valence electrons supplied by the atoms s0 as to achieve the lowest overall ‘energy subject tothe constraint of the Pauli exclu- sion prinipl, that no more than two electrons may occupy a single orbital (and then must be paired} 3. If more than one molecular orbital of the same energy is available, add the electrons to each individual orbital before doubly occupying any one orbital (because that minimizes electron electron repulsions} 4. Take note of Hund’s rule (Section 13.11), that if electrons occupy different degenerate orbitals, then they do so with parallel spins ‘The following sections show how these rules are used in practice, Selt-test 145 How many molecular orbitals can be bulk from the valence shel exital in 0,7 (answer: 81 14.10 Hydrogen and h “The first step in the discussion of II, the simplest many-electron diatomic molecule, is to build che molecular orbitals. Because each II atom of Il, con- tributes 2 1s orbital (as in IT), we can form the 16 (more precisely, 16,) and 16° (that is, 10,) bonding and antibonding oxbitals from them, as we have seen already. At the equilibrium internuclear separation these orbitals will have the energies represented by the horizontal lines in Fig. 14.19. m molecules 334 CHAPTER 14; QUANTUM CHEMISTRY: THE CHEMICAL BOND Energy > Fig. 14.19 A molecular erital energy level diagram for cofbitals constructed fom {Is,1stoverlap, the seperation of ‘the levels corespanding othe egulitrism bond length \@ His, Hay 14.20 The ground elecwonie configuration of Hy is ined by accommodating the two electtons in the lwvest avalable orbital ne bonding orb Eneray > ‘There are two electrons to accommodate (one from each atom). Both can enter the 16, orbital by pairing their spins (Fig. 14.20). The ground-state configuration is therefore 162, and the atoms are joined by a bond consisting of an electron pair in a bonding o orbital. These two electrons bind the two naclei together more strongly and closely than the single electron in T13, and the bond length is reduced from 106 pm to 74 pm. A pair of electrons in a 6 orbital is called a obond, and is very similar to the @ bond of VB theory. The two differ in certain details of the electron distribution between the two atoms joined by the bond, but both have an accumulation of density between the nuclei We can conclude that the importance of an elec- tron pair in bonding stems from the fact that two is the maximum number of electrons that can enter cach bonding molecular orbital. Electrons do not ‘want’ to pair: they pair because in that way they are able to occupy a low-energy orbital AA similar argument shows why helium is a mona- tomic gas. Consider a hypothetical Ie, molecule Each He atom contributes a 1s orbital to the linear combination used to form the molecular orbitals, and s0 we can construct 16, and 16, molecelar orbitals. ‘They differ in detail from those in TT, because the Hels orbitals are more compact, but the general Eneray > Fig. 1421 The ground electron configuration of the fout- faction molecule Hi, has ‘wo banding alectrons and so antibonding leerrons, Ishes a higher energy then the sepat= ‘Sted atoms, and so He, 's unstable relative to vo He atoms, shape is the same, and for qualitative discussions wwe can use the same molecular orbital energy-level dliagcam as for H,. Because each atom provides two clectrons, there are four electrons to accommodate Two can enter the 16, orbital, but then itis fll (by the Pauli exclusion principle). The next two electrons ust enter the antibonding lot, orbital (Fig. 14.21} ‘The ground clectronie configuration of Hey is therefore 1o2102. Because an antibonding orbital is slightly more antibonding than a bonding orbital is bonding, the He, molecule has a higher energy than the separated atoms and is unstable. Hence, two ground-state He atoms do not form bonds to each cther, and helium isa monatomie gas. Example 14.1 Judging the stability of diatomic molecules Decide whether Li is thely to exist on the assumption ‘hat only the valence s orbitals contnbute tolts molecular orbital. Strategy Deve wh: molecu orbitals can be formed ‘rom the available valence orstls,renk them in order ot fenetgy, then fesa in the slectrans supplied by the val fence exbtals ofthe atoms, Judge whetne” the bonding ornet antibonding effect between the atoms. Solution Each molecular orbital is bulk rom 2s atomic cial, which give one tending and one anibandingcom= bination 0, and 1a, respectively). Each Li atom supplies cone valence elecvon the two electrons ithe 19, ori 1a give she configuration 108, whichis bonding. Selt-test 146 le LM tely to ana f the Li ator uses only ts 25 orbital for bonding (newer Yes, lia it6P 14.11 Period 2 tomic molecules ‘We shall now see how the concepts we have intro- duced apply to other homonuclear diatomic mole- calles, such as Ny and Clyy and diatomic ions such as OF Inline with the building-up procedure, we first consider the molecular orbitals that may be formed from the valence orbitals and do not (at this stage) trouble about how many electrons are available. In Period 2, the valence orbitals are 2s and 2p. Suppose first that we consider these two types of orbital separately. Then the 2s orbitals on each atom. overlap to form bonding and antibonding com- binations that we denote 16, and 16,, respectively Likewise, the two 2p, orbitals (by convention, the internuclear axis is the z-axis} have cylindrical sym- metry around the internuclear axis. They may there~ fore participate in o-orbital formation to give the bonding and antibonding combinations 26, and 20, respectively (Fig. 14.22). The resulting energy levels of the 6 orbitals are shown in the MO energy level diagram in Fig, 14.23, Note that we number the o, orbitals in sequence (16,, 26, ..-) and the, orbitals likewise, Strictly, we should not consider the 2s and 2p, orbitals separately, because both of them can con tribute to the formation of 6 orbitals. Therefore, ina more advanced treatment, we should combine all four orbitals together to form four 6 molecular orbitals, ach one of the form Y= Cras * Vans Yhap, + Ca¥, We find the four coefficients, which represent the different contributions that ‘each atomic orbital makes to the overall molecular orbital, by using the variation theorem. However, in practice, the two lowest-energy combinations of this kind are very similar to the combination 16, and 19, of 2s orbitals. that we have described, and the two highest energy combinations are very similar to the 26, and 26, ‘a C@e = CeO 1 Fi 142207 rete xn 8 omen 2.0 bonding orbital and b) the corresponding anbonding frit when wo parbtels overlap slong on internuclear axis. Fig. 14.23 A typical molecular orsital energy level diagram {or Period 2 homanuclar atomic molecules. the valence tome orbitals are dravn in tne columns on the left an the right; the maleculsr ofials are shown in the mice Note thatthe 1 orbitals form doubly degenerate pars tthe Closely spaced ines denote orbtls Hing at exactly the same tenergy!. The sloping Ines ining the molecuib taste the ‘stom orb'tals show the principal compositan ofthe malec Use orbitals, Tis clagram i sutable for O, and Fi the con figuration of O, 18 shown, combinations of 2p, orbitals, In each case there will be small differences: the Lo, orbital, for instance, will be contaminated by some p, character and the 2G, orbital will be contaminated by some 2s charac- ter, and their energies will be slightly shifted from where they would be if we considered only the ‘pure’ ‘combinations. Nevertheless, the changee are not ‘great, and we can continue to think of 16 and 1a” as being one bonding and antibonding pair, and of 26, and 26, as being another pair, The four orbitals are shown in the centre column of Fig. 14.24. There ie no guarantee that 16, and 2c, will be in the exact, Fig. 14.24 Atypical molecular orsital energy level dagrars for Period 2 hormonuclear diatomic molecules up to and inelucing Ny 236 CHAPTER 14; QUANTUM CHEMISTRY: THE CHEMICAL BOND Fig, 14.25 (a) The interference leading tothe formation of 2 soning arial and (0) the cotrasponaing antioonaing obit location shown in the illustration and the locations shown in Fig. 14.23 are found in some molecules {see below. ‘There is one further point in this connection. As soon as we allow all four atomic orbitals to con- tribute to an LCAO it is no longer clear—except by appealing to the form of the simple pairwise LCAOs that each one resembles—whether a particular com- bination is bonding or antibonding: all we ean say is that the four linear combinations have successively increasing energies. However, the parity clasification is unaffected, and the orbitals can still be classified as gor usin homonuclear diatomic molecules, inversion, symmetry isa more fundamental classification scheme than bonding and antibonding. Now consider the 2p, and 2p, orbitals of each atom, which are perpendicular tO the internuclear axis and may overlap side-by-side. This overlap may bbe constructive or destructive and results in a bond- ing and an antibonding x orbital, which initially we label 1x and 1", respectively. The notation m is the analogue of p in atoms, for when viewed along the axis of the molecule, a orbital looks like ap orbital (Fig. 14.25). More precisely, an electron in a orbital has one unit of orbital angular momentum about the internuclear axis. The two 2p, orbitals overlap to give a bonding and an antibonding orbital, as do the two 2p, orbitals too. The two bond ing combinations have the same energy; likewise, the ‘wo antibonding combinations have the same energy. Hence, each m energy level is doubly degenerate and consists of two distinct orbitals. Typically (but not universally) the bonding effect of electrons in am orbital is less than for a 6 orbital in the same mole- cule because the electron density it represents does not lie berween the nuclei so completely. Likewise, Fig. 14.28 The geradeungerade character of bonding ard antibonding tbls. the antibonding effect of electrons in a n° orbital is typically less than when they oceupy a 6* orbital in, the same molecule. Two electrons in a x orbital con. stitute a m bond: such a bond resembles a x bond of valence bond theory, but the details of the electron, distribution are slightly different. ‘The inversion symmetry classification also applies to morbitals. As we see from Fig. 14.26, a bonding x orbital changes sign on inversion, and is therefore classified as u, On the other hand, the antibonding x* orbital does not change sign, and is therefore g. The bonding and antibonding combinations will hence forth be denoted In, and In,. The relative order of the cand orbitals in a molecule cannot be predicted without detailed calculation and varies with the energy separation between the 2s and 2p orbitals of the atoms; in some molecules the order shown in Fig. 14.23 applies, whereas others have the order shown in Fig. 14.24. The change in order can be seen, in Fig. 14.27, which shows the calculated energy levels for the Period 2 homonuclear diatomic mole- cules. A useful rule is that, for neutral molecules, the Fig. 14.27 The variation of the orbital energies of Period 2 hhomenuclear satomie molecules. Orly the valence-shell orbitals are shown, order shown in Fig. 14.23 is valid for O, and F, whereas the order shown in Fig. 14.24 is valid for the preceding elements of the period. 14.12 Symmetry and overlap ‘One centeal feature of molecular orbital theory ean, now be addressed, We have seen thats and p, orbitals may contribute to the formation of ¢ orbitals, and that p, and p, orbitals may contribute to x orbitals However, we never have to consider orbitals formed by the overlap of s and p, orbitals (or p, orbital). When building molecular orbitals, we need consider linear combinations only of atomic orbitals of the same symmetry with respect ta the internuclear axis. Because an s orbital has eylindsical symmetry around the internuclear axis, but a p, orbital docs not, the wo atomic orbitals cannot contribute to the same molecular orbital. ‘The reason for this distinction, based on symmetry can be understood by considering the interference between an s orbital and a p, orbital (Fig. 14.28}: although there is constructive interfer ence between the two orbitals on one side of the axis, there is an exactly compensating amount of destruc tive interference on the other side of the axis, and the net bonding or antibonding effect is zero. Consistent with this interpretation, the overlap of Is orbital on one atom and a 2p, orbital on another atom (with the internuclear axis) is zero. In terms of the discussion in Derivation 14.2, we seein Fig. 14.28 that at some point the product yay may be Large. However, there is a matching point in the lower half of the figure point where yj Vj has exactly the same magnitude but an opposite sign. When the integral i evaluated, there two contributions are added together and cancel. For every point in the upper half of the diagram, there isa point in the lower half that cancels it, 0 S= 0, Therefore, for symmetry reasons, there is no net overlap between the s and p orbital in this arrangement, i) or Fig. 14.28 Overlnping s and p orbitals. fa} Erckon overlap leads to nonzoro over and te the formation of an axially symmetric 0 obita. (0) Broadside overlap leads to no net ac ‘cumulation or reduction f electron density anc davs nt con ‘bute 0 bonding, ‘We now have the criteria for selecting atomic orbitals from which molecular orbitals are to be built: 1, Useall available valence orbitals from both atoms. 2. Classify the atomic orbitals as having o and = symmetry with respect to the internuclear axis, and build 6 and x orbitals from all atomic orbitals of a given symmetry. 3. From N, atomic orbitals of o symmetry, Ny 0 orbitals can be built with progressively higher energy from strongly bonding to strongly antibonding. 4, From N, atomic orbitals of x symmetry, orbitals can be built with progressively ‘energy from strongly bonding to strongly anti- bonding. The x orbitals occur in doubly degener- ate pairs As a general rule, the energy of each type of orbital (6 or a} increases with the number of internuclear nodes. The lowest energy orbital of a given species has no internuclear nodes and the highest energy orbital has a nodal plane between each pair of adja cent atoms (Fig. 14.29}, © A brief illustration A d, orital has cylindrical sym met sround 7and s0.cencontraute to orbicals. The dy, ane d,orbtala have 1 syrmmetry with respect ta the axie Fig. 14.30), so they ean contribute ton orsials. @ Energy —> Ch CO CxO cle Fig. 1429 A schematic ropresentation of the four molecular abitels thet can be formed from four s obitals in a chain ‘of four atoms. The lowest energy combination {the bottom iagram is formed trom atomic or tals wth the some sion, and thore are no internuclear nodes. The next higher orbital has ane rode [at he cenve of tne molecule. The next higne’ fofbitel has tuo internuclear nodes, and the Unpermost, highest energy orbtal has three interuclest nodes, one. btweon each neighbouring pai of atoms, and is fully ant bonding. The size ofthe spheres reflect he conburions of ‘each atom to the molecular arbatal; the shading represents sitferen signs. 338 CHAPTER 14; QUANTUM CHEMISTRY: THE CHEMICAL BOND he Fig. 1430 The types of molecule otal to whieh d ofoitals can contribute, The andi! combinations can be formes with 5. p. and d orbitals of the anpronrate symmetry, but the & otbtals canbe fred enlyby the ¢ orials ofthe two ators. Selttest 147 ‘Sketch the 'Dorbitals orbitals that resemble fourlobed orbitals when viewed along the internuclear axis) that ray be formed by the termining ‘vo d orbitals land which contibure to banding in some d-metal cluster cor pounds). Give theirinversion-symmetry classification arswars00 9.1430: banaing ate 9 antennae homonuclear diatomic molecules Figuces 14.25 and 14.24 show the general layout of the valenceshell atomic orbitals of Period 2 atome con the left and right. The lines in the middle are an indication of she energies of the molecular orbitals that can be formed by overlap of atomic orbital. From the eight valence shel orbitals (ou fom each stom), we can form eight molecular orbital: four are orbitals and fou, in two pac, ae doubly degenerate orbitals, With the orbitals established, we derive she _round-rtate electron configurations of the molecwlee by adding the appropriate numberof electrons to the orbitals and following the building-up rules. Charged species (such a the peroxide fon, O2, and Cj) need sither more or fewer electrons (for anions and cations, respectively) than the neutral molecules. We llostrate the procedre with N,, which ha ten valence electrons for this molecule we ee Fig. 14.24 ‘The first two eleetcons pair, enter, and fl the 16, orbital. The next two electrons enter and fil the lo, orbital, Six electrons remain, There are two 1%, orbitals, 0 four electrons can be accommodated in them. The two remaining electrons enter the 26, orbital. The ground-state configuration of Ny is therefore 1oZ1021"!202. This configuration is also depicted in Fig. 14.24 The strength of a bond in a molecule is the net outcome of the bonding and antibonding effects of the electrons in the orbitals. The bond order, b, in a diatomic molecule is defined as b=4N-N*) where N is the number of electrons in bonding orbitals and N* is the number of electrons in anti bonding orbitals (as judged by their resemblance to the simple pairwise LCAOs}, Each electron par in a bonding orbital increases the bond order by 1 and cach pair in an antibonding orbital decreases it by 1 For H,, b = 1, corcesponding to a single bond beeween the two atoms: this bond order is consistent with the Lewis structure HH for the molecule. In He,, which has equal numbers of bonding and anti bonding electrons (with N=2 and N° =2}, the hond order is b =, and there is no bond. In Ny, 104,20 and In, ate bonding orbitals, and n=2-+244= 8; however, 16, (the antibonding partner of 16,) is antibonding, so N* = 2 and the bond order of Ny is b=418—2)= 3. This valueis consistent with the Lewis seructure :N=N:, in which there is a tiple bond beeween the two atoms. ‘The bond order isa useful parameter for discussing the characteristics of bonds, because it correlates with bond length, and the greater the bond order beeween atoms of a given pair of atoms, the shorter the bond. The bond order also correlates with bond serength, and the greater the bond order, the greater the strength. The high bond order of N; is consistent with its high dissociation energy (942 kJ mol”’), (4.2) Example 14.3 Writing the: molecule tron configuration ofa diatomic \Whte the ground-state electron configuration of O, and caeulate the bond order, Strategy Decide which MO energy level diagram rouse (Fig, 14.23 oFFig. 14.24). Count the valance electrons and accommodate thom by using the bulking-up principle Solution Figute 14,23 appropriate for oxygen. There sae 12 valence elcrvons to accommodate. The fist 10 electrons recreate the N, configuration (with a reversal ofthe order of the 26, a 1H, arial the remainng ‘0 electrons must occupy the in orbitals. The confi- tion and bona order are snerefare toztos202 fing ‘Tris configurations also desictedin Fig. 14.29. Because 10, 204. and In, aferegatces as bending and 10, and ‘mys antibonding, he bone orders b= 8—4}=2. This bot orGer accords with the classical vow that oxygen has & double bor. Selt-test 14.3 ‘wit th olecron conigureton af F; and deduce its band order. (Ansa oprafzepintin§ B=) We see from Example 14.3 that the electron con- figuration of O, is 1o21022021R'1x2. According tothe building-up principle, the rwo In, electrons in O, will occupy different orbitals. One enters the 1, orbital formed by overlap of 2p,- The other enters its degen- erate partner, the 1m, orbital formed from overlap of the 2p, orbitals. Beeause the two electrons occupy different orbitals, by Hund’s rule they will have parallel spins (TT). Consequently, an O, molecule is sometimes rogarded as a biradical, species with two tanpaired electrons, (A tee biradical has two electron spins with random relative orientations in O; the two spins are parallel.) Molecular orbital theory therefore suggests—correctly that O; isa reactive component of the Earth's atmosphere; its most important bio- logical role is as an oxidizing agent. By contrast, Np, the major component of the air we breathe, is s0 tanreactve that nitrogen fixation, the reduction of atmospheric N, 0 NIT, by certain microorganisms, is among the most thermodynamically demanding of biological processes in the sense that it requires a great deal of energy derived from metabolic processes “The electronic configuration of , also suggests chat it will be magnetic because the magnetic fields gener- aed by the two unpaited spins do not cancel. Spcifie ally, O; is predicted to bea paramagnetic substance, a substance that ie drawn into a magnetic field, Most substances (those with paired electron spins) are diamagnetic, and are pushed ovt of a magnetic field ‘That O is in fact a paramagnetic gas is a striking confirmation of the superiority of the molecular orbital desription of the molecule over the Lewis and VB descriptions (which requiee all the electrons to be paired). The property of paramagnetism is util ized to monitor the oxygen content of incubstors by measuring the magnetism of the gases they contain, ‘An F, molecule has two more electons than an Op molecule, so its configuration is 1o21022031x!1n! and its bond order is 1. We conclude that F, is 4 singly bonded molecule, in agreement with its Lewis structure ii. The low bond order is con- sistent with the low dissociation energy of F, (154 k] mol}, A hypothetical Ne, molecule would have two further electrons: its configuration would be Jo;l0220:1n41n!202 and its bond order 0. The bond order of zero—which implies that ewo neon atoms do not bond together—is consistent with the monatomic character of neon. Example 144 19 the relative bond strengths of molecules and ions ‘The superoxide ion, O5, plays an important role in the egeing processes that ‘ake place in oxganisms. Jusge ‘wihetner OF tkely to have a lrger or stra dasock Strotegy Because 9 species with the lager bond order 'slikely have tne larger issociationeneray, we should ‘compare their electronic configurations, and assess heir bone arcers, Solution From Fig, 14.23, 0, toztof203tnging OF toitez2021n$1n2 Because the anion has the smaller bond order, ve ‘expect xo have the smaller association energy Solttest 14.9 Which can be expected 10 have the higher disor ‘tion eneray, F orF3? arse 31 14,14 Heteronuclear diatomic molecules heteronuclear diatomic molecule isa diatomic mole ccale formed from atoms of two different elements; ‘wo examples are CO and TCI. The electron distri- bution in the covalent bond between the atoms is not ‘symmetrical between the atoms because itis energet- ically favourable for a bonding electron pair to be found closer to one atom rather than the other. This imbalance results in a polar bond, which isa covalent bond in which the electron pair is shared unequally by the two atoms. The electronegativity, z (chi), of an element isthe power of its atoms to draw electrons to itself when it is part of a compound, so we can 340. CHAPTER 14: QUANTUM CHEMISTRY: THE CHEMICAL BOND Table 14.2 Electronegativities of the main-group elements 4 24 i fe 8 c N oO F 101 18 20 25 30 36 40 Neo Mg alo si Po os a og 12 «18 «18 21 25 30 K Ca Ga Ge As Se 08 419 16 18 20 24 28 Ro Sr ly Sn SbOTes os 19 17 18 19 21 25 cs Bs Tl PhB_Po 07 08 18 18 19 20 expect the polarity of a bond to depend on the rela- tive electronegativities of the elements Linus Pauling formulated a numerical scale of clectronegativity based oa considerations of bond dissociation energies, E{AB): laa~ al=0-102 x (AEK] mol)!®—(14.13a) with AE=B(A-B)—HEIA-A) + B(B-B)) (14.138) ‘Table 14.2 lists values for the main-group elements. Robert Malliken proposed an alternative definition in terms of the ionization energy, I, and the electron, affinity, E,, of the element expressed in electronvolts: Mi+E,) (14.14) ” ‘This relation is plausible, because an atom that has a high electronegativity is likely to be one that has a high ionization energy (so that itis unlikely to lose lectrons to another atom in the molecule) and a high electron affinity (so that it is energevically favourable for an electron to move towards it). The Mulliken clectronegativities are broadly in ine with the Pauling clectronegatvities. Blectronegativities show a periodi- city, and the elements with the highest clectronega- tivities are those close to fluorine in the periodic table. ‘The location of the bonding electron pair close to ‘one atom in a heteronuclear molecule results in that atom having a net negative charge, which is called 1 partial negative charge and denoted 6 There is 2 compensating partial positive charge, 81, on the other atom. In a typical heteronuclear diatomic molecule, the more electronegative element has the partial negative charge and the more electropositive clement has the partial positive charge. Selttest 14.10 Precict the signs of the charge distribution of a CH bond. answer oH Molecular orbital theory takes polar bonds into its side, A polar bond consists of two electrons in an. orbital of the form VE eAWK + on (4.asy with ¢@ no longer equal to cf. If ef> fy the electrons have a greater probability of being found on B than on A and the molecule is polar in the sense *A—B™. A nonpolar bond, a covalent bond in which the electron pair is shared equally between the two atoms and there are zer0 partial charges on each atom, has = 2. A pure ionic bond, in which one atom has obtained virually sole possession of the electron pair (as in Cs*F, 0 fest approximation), has one coefficient zero (so that A‘B- would have ej=Oand ef 1). A general feature of molecular orbitals berween dis- similar acoms is thatthe atomic orbital with the lower energy (that belonging to the more electzonegative atom) makes the larger contribution to the lowest- energy molecular orbital, Tae opposite is true of the highest (most antibonding} orbital, for which the prin- cipal contribution comes from the atomic orbital with higher energy (the ess electronegative atom) Bonding orbitals: for 24> zu b> Antibonding orbitals: for 74> 2 ¢h< Figure 14.31 shows a schematic representation of this point, ‘These features of polar bonds can be illustrated by considering IIF. The general form of the molecular orbitals of TF is w= cys¥4; + ¢,¥oy where Yi is an TTL orbital and yp is an F2p, orbital. Because the ioniza- ton energy of a hydrogen atom is 13.6 eV, we know that the energy of the ITls orbital is-13.6 eV. As usual, the zero of energy is the infinitely separated electron, and proton (Fig. 14.32). Similarly, from the ionization energy of fluorine, which is 18.6 eV, we know that the energy of the F2p, orbital is approximately ~18.6 eV, about 5 eV lower than the ITs orbital. It follows that the bonding o orbital in TIF is mainly F2p, and the antibonding o orbital is mainly IIIs orbital in char- acter. The two electrons in the bonding orbital are most likely to be found in the F2p, orbital, so there is, a partial negative charge on the F atom and a partial positive charge on the IT atom, More electronegative Fig, 14.31 A schomatic representation of the telaive con= tebutions of atoms of ferent electronegativitas te bonding and antinonsing molec oreitals Inthe boning ort the Tora electronegative atom rakes the greater contrition {represented by the larger sohere}, and the electrons of the bond are more tke to be found on hat ator. The apposite is tue of an antinonaing orbital. A patt of the reason why {an anttonaing oraital is of high energy is thatthe electrons ‘hot eceupy itae kel to be fauna on the more elctraposi- h7.40 Energy —> nl ¢ \ lap 02a, 087y, Fig, 14.32 The stomic otto onorgy bvels of Hand F ators and the molecular eritals they forms. The bonding orbtal has predominantly F atom character and the antibonding orbital as predominantly H ators character. Energias aro in electtonvols AA systematic way of finding the coefficients in the linear combinations isto use the variation theorem and to look for the values of the coefficients that result in the lowest energy (Section 14.2) For example, when the variation principle i applied to an TT, molecule, the calculated energy is lowest when the two Ils orbitals contribute equally to a bonding orbital, How ever, when we apply the principle to HIF, the lowest energy is obtained forthe orbital y=0.24Yp + 0.97 Yp We see that indeed the F2p, orbital does make the greater contribution to the bonding 6 orbital. An even lower energy is obtained—with a lot more calculation—ifeven more orbitals ae included in the linear combination (such as F2s and F3p, orbitals) but the principal lowering of energy is achieved from atomic orbitals of similar energies. 029 028 Fig, 1433 The molecular orbital energyievel diagram for CO. Solt-test 14.11 What isthe probability of finding 6 0 electron in HF in @ Fap,orbtal? answer 88 pe cont = (0.842 100 perce Figure 14.33 shows the bonding scheme in CO and illustrates a number of points we have made. ‘The ground configuration is 10°20? 1x*36°, (The gt designation is inapplicable because the molecule is heteronuclear and the 6 orbitals are simply numbered in sequence, 16, 26, ..., and the x orbitals likewise.) ‘The lowest-energy orbitals are predominantly of O character as that is the more electronegative element. ‘The highest occupied molecular orbital (HOMO) is 36, which is a largely nonbonding orbital centred on. G, s0 the two electrons that occupy itcan be regarded as a lone pair on the C atom. The lowest unoccupied ‘molecular orbital (LUMO) is 2x, which is largely a doubly degenerate orbital of 2p character on carbon. This combination of a lone-pair orbital on C and a pair of empty orbitals also largely on C is at the root of the importance of carbon monoxide in d-block chemistry, because it enables it to form an. extensive series of carbonyl complexes by a com= bination of electron donation from the 36 orbital and electron acceptance into the 2x orbitals. The TIOMO and the LUMO jointly form the frontier orbitals of the molecule, and are of great importance for assessing its reactions. 14.15 The structures of polyatomic molecules ‘The bonds in polyatomic molecules are built in the ‘same way asin diatomic molecules, the only difference being that we use more atomic orbitals to construct 342. CHAPTER 14; QUANTUM CHEMISTRY: THE CHEMICAL BOND the molecular orbitals, and these molecular orbitals spread over the entire molecule, not just the adjacent atoms of the bond. In general, a molecular orbital is 2 linear combination ofall the atomic orbitals of all the atoms in the molecule. In 1,0, for instance, the atomie orbitals are the two Ils orbitals, the O28 orbital, and the three O2p orbitals (if we consider only the valence shell). From these six atomic orbitals we ean construct six molecular orbitals that spread overall three atoms. The molecular orbitals differ in energy. The lowest energy, most strongly bonding orbital has the least number of nodes between adjacent atoms. The highest energy, most strongly antibonding orbital has the greatest num- bers of nodes between neighbouring atoms. According to MO theory, the bonding influence of a single electron pair is distributed over all the atoms, and each electron pair (the maximum number of electrons that can occupy any single molecular orbital} helps to bind all the atoms together. In the LCAO approximation, each molecular orbital is modelled as a sum of atomic orbitals, with atomic orbitals contributed by all the atoms in the molecule Thus, atypical molecular orbital in H,O constructed from Hs orbitals (denoted yy and y) and O2s and ‘O2p, orbitals (denoted Yo, and Yo,) will have the composition Y= Vat Vou Von, FAM (14.16) Because four atomic orbitals are being used to form the LCAO, there will be four possible molecular orbitals of this kind: the lowest energy (most bond- ng) orbital will have no internuclear nodes and the highest energy {most antibonding) orbital will have a node between each pair of neighbouring nuclei (Fig. 14.34). ‘An important example of the application of MO theory isto the orbitals that may be formed from the Pp orbitals perpendicular to the molecular plane of benzene, Cill,. Because there are six such atomic orbitals, it is possible to form six molecular orbitals ofthe form VRC EWE H CWS Heat 6+ c6M6 (147) ‘The lowest energy, most strongly bonding orbital has no internuclear nodes, and has the form VEY MEME Ver Wt Me A brief comment We are ignoring normalization fectors, for laity. In ths and she follwing case it would be 16° if we ignore overlap, ‘This orbital is illustrated at the bottom of Fig, 14.35. It is strongly bonding because the constructive inter- Eneray—> Fig, 14.36 Schematic frm of the molecular orbtals of KO, ‘he blue denotes 5 orbitals, the yellow p orbital, and the green hybrids of s and p character, Dork ane: light tones Cenote positive anc negative chases, rescectvely Energy > Fig. 1435 The r oritals ond the m molecular oritl eneray level diagram of benzene. The lowestanergy orbitals fly bonding between neignbouring stoms out the us crbtalisflly antibonding. The tw pars of doubly co molecular aritels have an intermediate number rnulear nodes. As usual, ght and dark shading represents di- ‘erent signs athe wavefunction. The ofbital have opposite signs below the plane of the rng. The symmetry designa- ‘ions ave those aperopiate to « exegonalmolecule ference between neighbouring p orbitals results in a good accumulation of electron density between the nuclei (but slightly off the internuclear axis, as in the bonds of diatomic molecules). The most antibond- ing orbital has the form YO Vi Wat Wa Vat Wi Me ‘The alternation of signs in the linear combination res- ults in destructive interference between neighbours, and the molecular orbital has a nodal plane berween, cach pair of neighbours, as shown in the illustration. ‘The remaining four molecular orbitals are more difficult to establish by qualitative arguments, but they have the form shown in Fig. 14.35, and lie in energy berween the most bonding and most antibond- ing orbitals. Note that the four intermediate orbitals, form two doubly degenerate pairs, one net bonding. and the other net antibonding, “We find the energies of the six x molecular orbitals in benzene by solving the Schrodinger equation (see the following section); they are also shown in the molecular orbital energy-level diagram. There are six electrons to be accommodated {one is supplied by each C atom), and they occupy the lowest three orbitals. The resulting electron distribution is like a double doughnut. It is an important feature of the configuration that the only molecular orbitals occu- pied have a net bonding character, as this is one con- tribution to the stability (in the sense of low energy) of the benzene molecule. It may be helpful to note the similarity between the molecular orbital energy-level diagram for benzene and that for N, (see Fig. 14.24}: the strong bonding in benzene is echoed in the strong, bonding in nitrogen. A feature of the molecular orbital description of benzene is that each molecular orbital spreads either all round or partially round the C, ring. That is, bonding is delocalized, and each electron pair helps to bind together several or all of the Catoms. The de- localization of bonding influence is a primary feature of molecular orbital theory and we shall encounter it in its extreme form when we come to consider the electronic structures of solids. ‘A final point is chat the 0, 1 classification of molecular orbitals is not strictly applicable to non~ linear polyatomic molecules. Instead, a classification scheme based on the actual symmetry of the mole cule is used, and you will see symbols such as a, ¢, and ty, in place of o and m!. However, the 6, 1 classification is relevant locally, in the sense that we can speak of the 6 bond between the O and an IT atom in T1,0 and the orbitals berween a pair of C atoms in benzene. In benzene, and other planar molecules, the term ‘X orbital’ is extended to mean, the delocalized molecular orbitals constructed from the p orbitals perpendicular to the ring. 14.16 The Hiickel method One of the earliest and simplest attempts to express these concepts quantitatively was introduced by Erich, 1 The symmetry dssifcation of rbtals in polyatomic molecles te deseribed in ot Phyicalchemartry (2006) ‘Hiickel in 1931. A point to bear in mind throughout is that a linear combination of N atomic orbitals results in N molecular orbitals the Tickel method provides an approximate procedure for putting these molecular orbitals in order of increasing energy. We shall introduce the method by considering ethene, CH,=CTL, and then consider larger moleces The fis step that Hckel took wa to ignore the c-bonding framework and focus solely on them ele: trons. That i, he assumed thatthe atoms had taken up the positions they have in the actual molecule, then calculated the properties of the m orbital that matched that framework. The partial justification for this approximation is that atomic orbitals that contribute to 6 molecular orbital have a different symmetry from the atomic orbitals that contribute to ‘molecular orbitals and have zero net overlap with them, Forethene that meant that he wrote ¥=¢,¥q ‘eq where and Yate the C2p, orbitals (with «x perpendicular tothe molecslar plane). We show in Derivation 14,3 thatthe Schrédinger equation for the orbitals, Fiy = Ey, then becomes the following pair of simultaneous equations for the coefiiente: (Haya ~ Eley + (Hyg ~ ES) (Hyq ~ EScq + (Hyg Eley, (14.18) ‘with the various terms defined and interpreted in the derivation, These equations are called the secular ‘equations. Derivation 14.3, ‘The secular equations When = e+ eis sudstitted into Fy Ey, we get oF + Cay = CuE a+ CEU Now mutily through by Ws: AW ata + Cea FWy = CrEWAWy + EWU ‘common sk’ in quantum chemistyis 0 turn products fof functions suchas ga) into quantties with apart lar numerical valve by integrating them over al space wal. for instance, thon becomes the averso integtl itn 6 numerical value suchas 0.2). When we intograt al four terms inthis expression we get sF lips) i Su {v2 ofa Fo 344 CHAPTER 14; QUANTUM CHEMISTRY: THE CHEMICAL BOND Thats, aban Calas ~CE¥ ES which i the fist of the two secur equations. The sec- fond equation is gonersted similaly but by mukilying ith yp instead of and then integrating In anticipation of the next step, wo can interprat some ofthe integrals. The overlap integral Swill be fama. The Insegral we have called Haq the ‘expectation value’ of ‘the hemitonian caleulated by using the wavefunction gy, ‘ented on atom A: that means it's a nd of average of ‘the hamitonian, with the contriaution tothe total eneroy weighted by tne probability, wer, thatthe electrons at ‘each point: thus * can be interpreted as the energy of the election in that oital. The same is tue of Hg, which we interpret as the energy of an electon inthe ofbitaly centtedon atm, both atoms are the same, these integrals are the seme, and we denote them 2. ‘Tho intogral Hg depends on both yj, and yy, and we can Interpret it asthe contribution to the energy cue to the accumulation of eletvon density where the two atomic cotbitals overlap, including, for instance, she Coulombic traction besween the extra accumuliion of electron density and both nucle, This term als includes the com bution tothe lowering of kinetic energy that stems ‘om the spread of the electron over both nucle. Tis Intogral land Hg, to which tis equals denoted Hiickel then made further approximations, Fes, he neglected all overlap integrals and set S=0 wher” ever it appears. This approximation turns the secular equations into (Hyg Bley + Hygea=0 aay) Heaca > (Hyp E)eg = 0 ‘Then, a¢ explained in the preceding derivation, like him we suppose that the terms H,4 and Fy, which represent the energy of an electron when iron atom A and B, respectively, are equal (that is trae in ethene and any molecule with equivalent carbon atoms) and replaced them by an empirical constant a This para- meter is approximately equal to (the negative of) the ionization energy of the atom. At this point, the equations are {a= Bley + Hyycg=0 (14.20) Hyyey + (0 Eeg=0 Finally, as explained in the derivation, he also sup- posed that the terms Hy, and Hay, which represent the energy of the interaction of the two nuclei with the accumulation of electron density in the inter- nuclear region and the effect on the kinetic energy of the electrons of spreading over both nuclei ean be replaced by the empirical parameter f(which, like oy is also a negative quantity as it represents a lowering of energy): (a Blea + Beg=0 (42a) Bex (a-Eleg=0 ‘These are the Hiickel equations for ethene, They ate two simultaneous equations for the coefficients x and ¢s We solve the Hiickel equations like solving any pair of simultaneous equations. In this case we multiply the first by Band the second by ar E: Bla Bey + Be Blo Bley + (a B)*c,=0 and then subtract the second from the firs, to obtain (P= (a- BP q=0 We know that the coeflcient cy cannot be 2er0 (for if it were, there would be no overlap and bonding), so the expression maltiplying ¢y must be zero: PHB (14.22) ‘The parameters and flare fixed, 0 this is an equa- tion for E. We rearrange itinto (al E)?= Band then take the square root of each side (allowing for two possible signs), to obtain @— E = =f. "That is, the energies ofthe orbitals are given by E-asp ‘Tis is essentially the end of the calculation, for wwe see that there are two energy levels, one at e+ and the other at e~ B, corresponding, respectively (because Bis negative), to the bonding snd anuibond- ing orbitals (Fig. 14.36). Because two electrons occupy the bonding orbital, the telecteon binding energy, Ey, of the molecule, the energy due to the eleetzons, 18 E,-2a+2p (14.242) Ifone electron is excited into the antibonding orbital, the m-clectron energy changes to Fs at B+ (a-p)=2a (14.246) ‘The difference B; ~ E, = -2B is therefore the energy needed to excite x electron in ethene. From spectro- scopy it is known that that excitation energy is equal toabout 7.7 eV, so we now know that Bis approxim- ately ~3.8 eV {corresponding to ~370 kJ mol”). ‘The Ihickel method is relatively easy to extend to more complicated hydrocarbons, but more advanced techniques—now universally implemented on math- ematical software—are needed to solve the secular (14.23) Fig, 14.26 The results of »Huckel calculation of the n orbital energies of ethene equations for the energy levels (and the coefficients of the orbitals). In addition to the approximations already mentioned, the method also supposes that the only interactions between immediate neighbours are included. In summary, the method is as follows: ‘+ Focus solely on the a orbitals ofthe system, ‘+ Sevall overlap integrals equal to zero ‘Ser all terms of the form Hy, equal t0 «(a nega tive quantity) ‘© Sevall terms of the form Hy equal to B (a negative {quantity) ifthe atoms are neighbours and to zer0 otherwise ‘+ Solve the Hiickel equations for the enezgies. In practice, this means that for N carbon atoms there are N Thickel equations; the coefficients are multi- plied by ee~ E, f, or zero depending on whether they refer toa single atom, involve neighbours, or involve atoms that are not neighbours, respectively. (o-Fley+Bey=0 Beg \a~ Bley + Be. =0 Bey r1a~ Elec + Be, =0 Bec +1a~ Eley =0 ae 4 Butadiene “The solutions hich ae best found by using mathorat- ical software but inthis simple case can also be found anual aro Eno 1.629, a+0.628 Because rand are both negative quantities, the order of increasing ene“ay, fom most banding to most antibond ing,'sa+ 1.826, 04+0.629, a—0.629, 0~1.629. @ Selftest 14.12 Caleuleie the election energy of butadiene in ite ground state. areata 4.4881 ‘The relectron binding energy of butadiene is E,=4ar+ 4.488. Ifwe had treated the molecule as two isolated bonds berween atoms A and Band atoms Cand D, we would have calculated E, = 4a + 48 {as for two ethene molecules). The difference, 0.48. (about -180 kJ mol” is called the delocalization, energy, Egyge- The delocalization energy is the addi- tional lowering of energy of the molecule due to the spreading of the = electrons throughout the molecule instead of being localized in discrete bond- ing regions. © A brief illustration The 7 electron eneray levels of benzone (s00 Problem 14.36) according to the Huckel method are Eaa229, 048,048 (Note the double degeneracy of levels with energy 2 These levels are ilustated in Fig. 14.38. The n-slecton binding energy of benzene is 5, (9+29)+ 20+) +2la+ 61 =6a+8p The rielecton binding energy of thee localized ethene molecules is 3122 +2 = 6a +63, Therefore, the delocs: ization energy of benzene is Eg = 82+ 88 16+ 8 = 28 or shout 740 ki mo, This considerable lowering of ‘energy due 1 delocazstionis & major eontbution tothe stablity ofthe benzene ring and of aromasc compounds in general. @ Computational chemistry Computational chemistry is now a standard part ff chemical research. One major application is in pharmaceutical chemistry, where the likely pharma-

You might also like