Lecture Notes from CHM 1341
17 June 1996

Gilbert Newton Lewis

Lewis was the driving force behind the excellent Chemistry Department at The University of California at Berkeley. But far more significant than his contributions to California were those he made to Chemistry. It was he who first suggested a rationale for chemical bonding.

That regularity of valence and chemical properties in the Periodic Table suggested the filling of blocks or shells of electrons where early electron occupants of a shell yielded one kind of chemistry and late occupants another. Moreover, the periodicity meant that "buried" shells weren't having much effect, and so the notion of a valence shell being the outer, "currently filling" one controlling the destiny of bonding arose.

Lewis was struck by the obstinacy of the last electron to fill a shell...all the way to the right of any period. That last electron "closed" the shell at the rare gas configuration, and rare gases were so self-satisfied that they would not engage in bonding.

Since Lewis's day, some rare gases have been forced to bond, but they do not produce robust compounds. They are a laboratory curiosity.

So he reasoned that Nature, while abhorring a vacuum, adored that closed shell octet electronic structure. His suggestion that the same impulse was at work in bonding led to the notion of the sharing of bonding electrons which each atomic participant can either claim to be part of its now closed valence shell octet or can release, falling back to an inner shell octet. In this sense, chemical bonding is synergistic in that the atomic participants complete all of their octets by electron sharing or electron transfer. Clearly, atoms early in the period would gain an octet most readily by losing their few valence electrons while atoms near the end of a period would seek to capture electrons for their few remaining vacancies in the valence shell. That would rationalize ionic (charge separation) bonding. Atoms midway in the period would need to involve themselves in many bonds to share enough electrons to simulate a closed shell. That would explain the high valencies of the mid-period atoms.

To facilitate the explanation of these effects, Lewis developed Lewis Symbols which surrounded atoms on 4 sides with dots representing only the valence electrons. Given that electrons repel one another, it wasn't hard to imagine that each succeeding one would seek a physical domain as far away as possible from other electrons of the shell. So the filling of Lewis symbols with the first 4 electrons is done with one to a side; it is only the last 4 electrons which have no choice but to pair up with one another as in

             .     .     .     .     ..     ..
.Na  .Mg.   .Al.  .Si.  :P.   :S:   :Cl.   :Ar:
                   '     '     '     ''     ''

The unpaired electrons, equivalent in number to the atom's valence, were considered available for bonding, while the "lone pair" electrons were considered non-bonding. This rationalizes the valencies (1, 2, 3, 4, 3, 2, 1, 0) across the periods. If like atoms bond, such as the chlorine atoms of the chlorine molecule,

     .. ..
    :Cl:Cl:
     '' ''

they share their single unpaired electron to complete what appears to be an octet around each. This perfect sharing is called covalent bonding. In the other extreme, if magnesium gives up its 2 valence electrons one each to a chlorine,

    -..   2+ ..-
    :Cl: Mg :Cl:
     ''      ''

Mg becomes a dipositive ion, has the closed shell (never shown) beneath the valence shell to content itself, the chlorines each capture an electron to fill their one valence shell vacancy, becoming (mono) negative ions in the process, and the molecule is held together not by shared electrons but rather by simple electrostatic attraction of opposite charges in an ionic bond.

When Lewis dots are used in molecules, the resulting diagram is a Lewis structure; Lewis symbol is reserved for the atomic diagrams.

The octet notion works well for all Main Group except period 1 and early period 2 elements. For example, lithium is univalent, and can form the lithium diatomic molecule, Li:Li, or lithium hydride, Li:H. For that matter, the hydrogen diatom, H:H, suffers the same conceptual problem: no octet. Instead, these atoms seek the stable structure of the first rare gas, He:, with its duet of electrons completing the n=1 shell.

Thus the ionic LiCL and the polar covalent HCl might have Lewis structures of

    +  .. -          ..
  Li  :Cl:   and   H:CL:
       ''            ''

respectively, where one partner is filling a duet while the other is filling an octet. Still the same principle at work.

While this scheme can work well for some polyatomic molecules, such as hydrogen peroxide,

     H                      H
     ..   ..                 \   ..
    : O : O :   alternately  :O - O:
      ¨   ¨                   ¨    \
           H                        H

it will unravel for many bonding situations if we do not augment it with the notion of bonds composed of more than one pair. A divalent atom like oxygen will bond to itself with double bonds as

  ..                     .   ..
 : O : : O :   not as  : O : O :
         ¨               ¨   '

So the preferred Lewis diagram for molecular oxygen would be

    ..
   : O = O :
         ¨ 

where each bonding pair is replaced by a bond line "-" yielding "=" for double bonds. This would make the trivalent nitrogen's diatomic molecule :N:::N: but you'll have to imagine triple bond lines since there appears to be no triple line character in HTML. (And you can't create on by underscoring the = sign because underscore marks a hotlink!)

With this in mind, we can create polyatomics with many centers as Lewis structures, for example, acetic (ethanoic acid):

          ..                    ..
     H   : O : H           H   : O - H
     ¨   ..                |    /
 H : C : C :: O :  or  H - C - C = O :
     ¨        ¨            |       ¨
      H                    H

Lewis's structures were augmented with Pauling's (Caltech) notion of electronegativity to describe the differences between covalent and ionic bonding. Electronegativity is a measure of the attraction of the electrons of a bond to a given atom. It is measured different ways by difference researchers in an effort to find a scale which is most predictive. As an example, some researchers take it to be the average of the IP (ionization potential or the energy required to rip an electron free from the neutral atom) and EA (electron affinity or the energy released when a neutral atom captures an electron), but others prefer to find some self-consistent set of electronegativities which give the best description of molecular electric dipole moments or some other measure of charge separation.

However it is measured, it applies to atoms, and the difference beween the electronegativities of two atoms is suggestive of the ionic character of a bond between them. So if they have the same electronegativity, neither will hog their bonding electrons and the bond between them will be covalent. If that difference is instead large, their bond will be quite ionic, with the atom whose electronegativity is largest receiving the lion's share of the bonding electrons.

For the electronegativity table (Fig. 3.11) in Gillepie et al., the lowest shown is 0.9 (Rb) and the highest is 4.1 (F). So the very ionic RbF bond would be characterized by an electronegativity difference of 3.2; in fact, anything over 2.0 is likely to be very ionic while anything under 1.0 is fairly characterized as covalent if polar. Any non-zero electronegativity difference implies unequal bond electron sharing; so the less electronegative partner atom will lose some its contributed bonding electron(s) some fraction of the time, rendering it partially positive (cationic) and its comrade partially negative (anionic). That charge separation is what is meant by a polar bond.

Polarity in molecules is important because the attraction of polar molecules for one another influences their intermolecular interactions and thus freezing and boiling points, for example.

For electronegativity differences between 1.0 and 2.0, the numbers are an imperfect guide to the bonding character.

                        H:2.1
   Li:1.0  Be:1.5   B:2.0   C:2.5   N:3.1   O:3.5   F:4.1
   Na:1.0  Mg:1.3  Al:1.5  Si:1.8   P:2.1   S:2.4  Cl:2.9
    K:0.9  Ca:1.1  Ga:1.8  Ge:2.0  As:2.2  Se:2.5  Br:2.8
   Rb:0.9  Sr:1.0  In:1.5  Sn:1.7  Sb:1.8  Te:2.0   I:2.2
   Cs:0.9  Ba:0.9  Tl:1.5  Pb:1.6  Bi:1.7  Po:1.8  At:2.0

It is not hard to imagine, however, that there should be electronegativity trends in the Periodic Table. The inner shell electrons are almost entirely inside the valence shell. Thus valence shell electrons "see" a core charge which is the difference between the number of protons and the number of electrons in all shells beneath them. The electronic repulsion from colleagues in the valence shell is much less significant; after all, they're trying their best to stay out of one another's way, and so they all are influenced by that same "core charge."

Thus, as we proceed across a period, the core charge increases (being the same value as the Group number), and the valence electrons respond by being ever more tightly bound. So electronegativity should increase as we move to the right in the Periodic Table.

However, as we go down a Group, the valence shell moves further and further from the nucleus, changing the attraction of the core charge inversely with the square that radial distance. So electronegativity should fall down columns; in the main, it does.

Chris Parr
University of Texas at Dallas
Programs in Chemistry, Room BE3.506
P.O. Box 830688  M/S BE2.6 (for snailmail)
Richardson, TX  75083-0688
   Voice: (972) 883-2485
     Fax: (972) 883-2925
     BBS: (972) 883-2168 or -2932 (V.32bis)
   Email: parr@utdallas.edu