How to Draw

Molecules ...
Just Like Lewis Dot
Diagrams, Only Easier &
Better
1 Introduction
Our primary goal for today is to be able to draw diagrams of molecules, such as we see in figure 1.

Figure 1: A Few Small Molecules

These diagrams tell us that the F2 molecule has a single bond, the CO2 molecule has two double
bonds, and the HCN molecule has one single bond plus one triple bond.
There are two main methods for constructing such diagrams.

Part I of this document describes the smart way to do it. This involves counting the bonding
and antibonding electrons, so we call it the BABE method. As a corollary of the BABE method, we
have the hole counting method, which is particularly easy to use, and applies to a wide range of
electron-rich molecules, including many organic molecules. We shall see that BABE is nicely
consistent with theory, including Molecular Orbital theory.
Part I is meant to give a clean and self-contained explanation of BABE and of holecounting. That means it is suitable for readers who have never heard of Lewis dot structures.

Part II provides background and context. In particular it contrasts the BABE method with an
older, widely known method, namely the Lewis dot method. It involves counting bonding and
non-bonding electrons, so it could be characterized as BNBE. We shall see that the usual rationale
for the Lewis method is incompatible with the available data, and incompatible with modern theory.

* Contents
1 Introduction
PART I
2 How to Draw Molecules
2.1 Electron-Rich Molecules
2.1.1 Fluorine
2.1.2 Nitrogen
2.1.3 Oxygen
2.1.4 Cyanic Acid and Isocyanic Acid; Formal Charge
2.1.5 Acetic Acid

2.1.6 Sulfur Dioxide

2.1.7 Sulfate Ion and Sulfuric Acid
2.1.8 Methanesulfonic Acid
2.1.9 Biomolecules
2.1.10 Summary
2.2 How to Explain Hole-Bonding
2.3 Spectroscopy and Energy Levels
2.4 Limitations of the Naive Approach, and How to Overcome Them
3 Some Supporting Ideas
3.1 The Aufbau Principle
4 Approximate Bond Angles
4.1 Basic Ideas
4.2 Some Examples
4.2.1 Methane
4.2.2 Ammonia
4.2.3 Water
4.2.4 CO2 and the Energy of Double Bonds
4.2.5 HCN, C2H2, and the Energy of Triple Bonds
4.2.6 CO3=
4.2.7 Nitrogen Dioxide Radical
4.2.8 Other Cases
5 Some Peculiar Examples
5.1 The Nitro Group
5.2 Sulfur Hexafluoride
5.3 Sulfur Tetrafluoride
PART II
6 Overview of the Contrast
7 Fun With Lewis Dot Diagrams ... Or Not
7.1 Hydrides and Other Successes
7.2 Oxygen
7.3 Sulfate Ion; Expanded Octets
7.4 Methanesulfonic Acid
7.5 Divergent Predictions
7.6 Additional Remarks
8 Conceptual Foundations (or lack thereof)
8.1 Contrast: Molecular Orbitals versus Lewis
8.2 Contrast : Hole Bonding versus Lewis
8.3 What Implies What
8.4 The Aufbau Principle (or not)
9 Discussion
10 Additional Background and Context
10.1 The Hole Concept
10.2 Coulombic Models
11 Summary
12 References

* PART I
2 How to Draw Molecules

In this section, we discuss a way of diagramming molecules that has many advantages. It is in all
ways preferable to Lewis dot diagrams.
The diagrams provide predictions about the number of bonds. This is important, because it implies,
among other things, predictions about the existence, non-existence, and reactivity of molecules.
That is, when we learn there are four single bonds in CH 4 and two double bonds in CO2, we also
learn that reactions that produce molecules of CH 4 are vastly more plausible than reactions that
produce molecules of CH2, and similarly molecules of CO2 are more plausible than molecules of
CO4.
2.1 Electron-Rich Molecules

Our first goal is to draw (and understand!) some simple molecules. Before we start talking about
rules, let alone theories, lets look at some data.
NN
OO
FF
Ne Ne

triple bond
two units of bond strength
single bond
no bond

5 electrons in the N=2 shell

6 electrons in the N=2 shell
7 electrons in the N=2 shell
8 electrons in the N=2 shell

The trend is clear: In this part of the periodic table, more electrons means less bonding. We shall
see that this trend holds true for a broad class of interesting and important molecules (but not for all
molecules).
We can construct a simple rule for drawing such molecules and predicting the number of bonds. We
will begin by stating the rule and applying it to some examples, and then in section 2.2 we will
explain the rationale behind the rule.
The procedure begins with figure 2, which can be considered a fragment of a simplified periodic
table. Hydrogen is shown with one dot, representing one hole, because it is one electron shy of the
helium closed-shell configuration. Similarly fluorine has one hole, because it is one electron shy of
neon.

Figure 2: Hole-Count of Some Common Atoms

This is just the venerable idea of valence, with the twist that the dots represent hole-valence, not
electron-valence. The elements in figure 2 all have the property of being electron rich which
means they have relatively few holes.
This meaning of the word hole has been in use for many decades. You may have encountered it in
connection with P-type semiconductors. It is pretty much synonymous with vacancy. It means
nothing more or less than a place where an electron could have been, but is not. In particular, holes

could perfectly well be high-lying vacancies, not just low-lying vacancies. (This stands in contrast
to ordinary back-yard holes, which always extend downward, below the general terrain.)
2.1.1 Fluorine

We can use the fluorine molecule as our first example. To diagram the F 2 molecule, we start by
drawing the atoms. Figure 3 shows two independent fluorine atoms. Each carries one hole, as we
know from figure 2.

Figure 3: Two Fluorine Atoms

We are free to slide the holes around. Figure 4 shows the same thing, with the holes rearranged to
face each other.

Figure 4: Two Fluorine Atoms, Rearranged

Finally, we connect the dots. Two holes make one chemical bond, as shown in figure 5.

Figure 5: Diagram of the Fluorine Molecule

Unlike a Lewis dot diagram, this figure does not portray electrons, but rather holes. (The notion of
holes has been part of physics since 1930, as discussed in section 10.1.) That is, each little dot in the
diagram represents the absence of an electron from an antibonding orbital.
Notice the logic here: In electron-rich molecules, what we casually call bonding might more
logically be called absence of antibonding. Its a double negative.
2.1.2 Nitrogen

As our next example, we turn to nitrogen. The procedure is the same. To diagram the N 2 molecule,
we start by drawing the atoms. Figure 6 shows two independent nitrogen atoms. Each carries three
holes, as we know from figure 2.

Figure 6: Two Nitrogen Atoms

Next, we slide the holes around so that they face each other. Finally, we connect the dots. Two holes
make one chemical bond. The nitrogen molecule contains a triple bond, as shown in figure 7.

Figure 7: Diagram of the Nitrogen Molecule

Figure 7 is the normal, conventional way of diagramming the N 2 molecule.

Meanwhile, figure 8 shows the same thing with a little extra detail. It shows three antibonding

orbitals (in color) each of which bears two holes.

Figure 8: Extra-Detailed Diagram of the Nitrogen Molecule

2.1.3 Oxygen

Lets consider oxygen, which is a slightly tricky case. If you blindly follow the simplified
procedure outlined above, you will wind up with a double bond. That is, the molecule will be
diagrammed as O=O.
However, we are allowed to use additional information.

We might know from observation that oxygen molecules are paramagnetic.

We might know from observation that liquid oxygen is blue (unlike liquid nitrogen, which is
colorless).

We might know (from theory or from spectroscopy) that the two highest non-empty
molecular orbitals in the oxygen molecule have the same energy. There is a rule (Hunds rule) that
says that if two electrons can unpair, they will.
For any (or all!) of those reasons, we can conclude that oxygen contains unpaired electrons.
Therefore diagramming the molecule as OO is vastly preferable to O=O. The preferred diagram is
also shown in figure 9.

Figure 9: Diagram of Oxygen Molecule

Figure 9 shows the O2 molecule with a moderate level of detail. Meanwhile, figure 10 shows the
same thing with a little extra detail. It is useful to compare this against figure 8. In both cases we

have three antibonding orbitals. The difference is that in nitrogen, the orbitals are fully populated,
for a total of three units of bond strength, while in oxygen, there is only fully-populated orbital and
two half-populated orbitals, for a total of two units of bond strength.

Figure 10: Extra-Detailed Diagram of Oxygen Molecule

To repeat: hole-counting alone does not suffice to tell us whether OO or O=O is the preferred
representation of the oxygen molecule. We need additional evidence from observation and/or
theory to tell us that OO is the right answer. This is discussed more fully in section 2.4.

2.1.4 Cyanic Acid and Isocyanic Acid; Formal Charge

To arrive at the most appropriate structure, a good rule of thumb is to distribute the bonds so as to
minimize formal charge. This tends to minimize the electrostatic potential energy, which is an
energetically favorable thing to do, other things being equal. 1
The procedure is to break each bond apart into two holes. (That is, break each dash into two dots.)
For the bond between atom A and atom B, assign one of the dots to each atom. Then compare the
number of dots assigned this way to the number of dots the atom normally has, where normal is
defined by the table in figure 2. If the assignments are the same, the atom has zero formal charge. If
the atom has more holes than normal, it bears a positive formal charge, and if it has fewer holes
than normal, it bears a negative formal charge. Remember, these dots represent holes, not electrons,
and each hole makes a positive contribution to the formal charge.
The idea of assigning half of each bond to each of the atoms makes perfect sense for symmetrical
molecules, but is only a first approximation in non-symmetrical molecules. It is certainly not a law
of nature.
Applying this rule gives us:
HOCN
HN=C=O

It is certainly possible to imagine something like HO=C=N, but it would be disfavored due to
needlessly high electrostatic potential energy. It would have positive formal charge on the oxygen,
and negative formal charge on the nitrogen ... unlike the formulas in equation 1, which have zero
formal charge on each of the atoms.
It is amusing to note that both of these acids have the same conjugate base, namely the cyanate ion:
NCO(())

The symbol to the right of the O in equation 2 is not a dash; it is a superscript minus sign. I know a
minus sign looks a lot like a dash, but in this case it is a minus sign. You can recognize it as such
partly because it is up in the exponent, and partly because it does not join two atoms.
You could imagine something like O=C=N(()) (which could be called the isocyanate ion), but since
oxygen is more electronegative than nitrogen, it makes sense to put the negative charge on the
oxygen.

Note: Electrons are very much more mobile than protons. The isocyanate ion would convert
immediately to the cyanate ion. In contrast, unionized isocyanic acid converts slowly, if at all, to
cyanic acid.
2.1.5 Acetic Acid

Lets move on to a larger and more complicated molecule, namely acetic acid. The first step, as
always, is to figure out what molecule were talking about. In this case its systematic name is
ethanoic acid, and its formula may be written CH 3COOH.
The easiest way to proceed is to draw the molecule, placing next to each atom one dot for each hole
that it contributes counting down from the appropriate atomic closed-shell configuration as
indicated in figure 2. Therefore the acetic acid molecule looks like
H
.

O
..

.
..
H. .C. .C
.
.
.
H

.
O. .H

Then we just connect the dots. Observation tells us this molecule is non-paramagnetic, so the
simple rule two dots make a dash is all we need. The result is:
H
O
|
//
H - C - C
|
\
H
O - H

Each dash represents one bond. For electron-rich molecules such as this, each bond is due to holes
(i.e. the absence of electrons) in a high-lying antibonding orbital.
With a little experience, you discover that it is possible to save some work by treating the methyl
functional group as a black box, like this:
O
//
(CH ) - C
3
\
O - H

or equivalently
O
//
Me - C
\
O - H

It is also interesting to see what happens when the acetic acid molecule reacts with water. In the
classical approximation, the result can be drawn as:
O
//
Me - C
\ ((-))
O

(+)
H - O - H
|
H

Acetate
Ion

Hydronium
Ion

We see that on the acetate ion, there is an oxygen with a formal negative charge, and only one hole
associated with it, while on the hydronium ion, there is an oxygen with a formal positive charge,
and three holes associated with it.
In this diagram, the formal charges are indicated in double parentheses, because they dont tell you
anything you dont already know. If the indications were missing, you could easily reconstruct
them based on the valence of each atom and the number of bonds it has.
The rest of this subsection is tangential to our main discussion, but it has tremendous practical
ramifications, so wed better mention it: In the classical approximation there are two different ways
of drawing the acetate ion, namely
(-)

O
/
Me - C
\\
O

<-->

O
//
Me - C
\ (-)
O

Either one of the oxygens could get the double bond, while the other one gets the formal negative
charge. Actually neither of these alternatives is really correct, for the same reason that neither of the
Kekul structures correctly describes benzene. In fact you get a quantum-mechanical resonance.
That is, the superposition of the two alternatives has a lower energy than either of them separately.
The process of resonance involves moving an electron from one oxygen to the other. This only
works in the ion. In contrast, in the neutral acid molecule, there are two alternatives, but the rate of
quantum-mechanical tunneling from one to the other is negligibly small, since it would require
moving a proton, not just an electron. This is important, because it means the ions energy is
resonantly lowered while the neutral molecules energy is not lowered. This makes the acid much
stronger than it would otherwise be, by making it energetically favorable for the acid to get rid of
its proton.
2.1.6 Sulfur Dioxide

Now lets examine sulfur dioxide. As usual, the starting point is to draw the atoms with the right
number of dots, i.e. the right number of holes. This gives us a rough idea of whats going on, as
shown in figure 11.

Figure 11: Sulfur Dioxide : Starting Point

Next, as usual, we see if we can combine the dots into bonds, to create a ball-and-stick model of the
molecule. The two possibilities are shown in figure 12.

Figure 12: Sulfur Dioxide : Two Ball-and-Stick Models

The final answer will be some sort of resonance i.e. a quantum mechanical superposition of the two
ball-and-stick models. This is shown in figure 13.

Figure 13: Sulfur Dioxide : Symmetric

If you want to attribute half a unit of negative formal charge to each oxygen atom, and a full unit of
positive charge to the sulfur atom, that is not necessary but is harmless provided you dont take it
too literally. There is no well-defined dividing line that determines which part of the electron cloud
belongs to one atom or another.
In any case we know from experiment that the SO 2 molecule is bent and has a nonzero electric
dipole moment. (Contrast this with CO2 which is linear and has no dipole moment.)
Also, spectroscopy and molecular orbital calculations tell us that the 3d orbitals make no
observable contribution to the bonding in SO 2.
As always, diagrams of the sort we are constructing here describe covalent bonding. A bond
between dissimilar atoms will exhibit some percentage of ionic character. A single bond between
sulfur and oxygen should have roughly 22% ionic character.
2.1.7 Sulfate Ion and Sulfuric Acid

Another example that illustrates the simplicity and power of the hole-bonding approach is the
sulfate ion, SO4=. As usual, we start by drawing the atoms, with the correct number of dots, where
each dot represents a hole in an antibonding orbital. We started with ten dots, because each of the
five atoms is divalent according to figure 2, but we erased two of the dots because of the double
negative charge carried by the ion. Remember the dots represent holes, and the two negative
charges fill those holes.

Figure 14: Sulfate Ion

We then replace pairs of dots with dashes representing a two-hole bond. The diagram practically
draws itself. The molecule is tetrahedral and completely symmetric:

Figure 15: Sulfate Ion : Hole-Pair Bonds

If you want to think of this as having a +2 charge on the central sulfur atom, and a 1 charge on
each of the four oxygens, thats optional but harmless. That is consistent with basic
electronegativity ideas, and semi-quantitatively consistent with the electron-density data from
detailed quantum chemistry calculations.
There is no trouble drawing a sensible diagram for the sulfuric acid molecule, as shown in figure 16.

Figure 16: Sulfuric Acid

This stands in contrast to the usual Lewis dot approach, which has some problems, as discussed
in section 7.3.
2.1.8 Methanesulfonic Acid

A diagram of the methanesulfonate ion is shown in figure 17. In the ion, all three oxygens are
equivalent. In contrast, in the methanesulfonic acid molecule, as shown in figure 18, the acid
hydrogen breaks this symmetry.

Figure 17: Methanesulfonate Ion

Figure 18: Methanesulfonic Acid

When we check the formal charge, we find a formal charge of +2 on the sulfur atom, and a formal
charge of 1 on each of the three oxygen atoms in the methanesulfonate ion. This is consistent with
the overall charge (1) on the ion. This is also consistent with what we know about
electronegativity; oxygen is considerably more electronegative than sulfur.

Similarly in the acid molecule, we find a formal charge of +2 on the sulfur atom, zero formal
charge on the acid oxygen, and a formal charge of 1 on the other two oxygens. Again this is
consistent with the overall charge balance, and consistent with what we know about
electronegativity.
For a discussion of how our diagrams differ from the prior art, see section 7.4.
It must be emphasized that the minus sign in double parentheses in figure 17 is strictly parenthetical;
you could erase it and the diagram would have the same meaning. You can verify and/or reconstruct
the parenthetical minus sign by adding up the formal charge on each atom in the drawing. It would
be quite wrong to calculate the formal charge according to the drawing and then add the
parenthetical charge; that would be counting the same charge twice.
Such a parenthetical charge indication stands in contrast to a formula such as H , which is an ion,
unlike H which is a neutral atom. The minus sign in the H formula is not parenthetical. Similarly
OCN is an empirical formula, and the minus sign is necessary to indicate its charge, in contrast to
NCO which is a structural diagram, and the charge can be inferred from the structure, namely
NCO(()).
In mathematics, parentheses are used for
grouping, with no implication that the contents
are merely parenthetical.

In written English, parentheses are used for

parenthetical remarks.

I use double parentheses when I want to make it clear that something is parenthetical, not a
grouping.
2.1.9 Biomolecules
Figure 19 shows a sketch of a molecule with a fair bit of biological significance, namely glucose.

Figure 19: Glucose Molecule

To sketch this, you need not only the empirical formula, but also some clue about which isomer is
desired. The isomer shown here is quite common in nature. This figure is meant to show the
number of bonds and the topology (i.e. what is connected to what); it is not meant to show every
detail of the three-dimensional shape.
Most biomolecules are sufficiently electron-rich that the simplest hole-counting notions suffice.

Figure 19 is what you get by using the hole-counting method. It is consistent with the way

professional biologists and biochemists sketch this molecule. If you doubt this, go
to http://www.google.com/search?q=glucoseand look at the results.
We are quite aware that hole-counting is not the only way of predicting the topology and the
number of bonds. There is another method that is almost universally taught in high-school
chemistry textbooks ... but it is not consistent with the known facts about molecules, and is not
consistent with the way professionals draw things. This is discussed in Part II.

Hole-counting agrees with real-world practice.

2.1.10 Summary

To summarize this section:

We confine our attention to electron-rich molecules. We assume we have been given the
basic topology of the molecule, and our first task is to determine which bonds are single, which are
double, et cetera.

Find the number of holes contributed by each atom. You can do this by reference to figure 2,
or simply by counting down from the next-highest noble gas element.

Each hole can be represented by a dot.

Draw bonds connecting the atoms. A bond is represented by a dash. A pair of dots is replaced
by a dash.

If necessary, re-arrange the bonds so that the formal charge (if any) on each atom follows the
rules of electrostatics:
o
Opposite formal charges attract, and should not be separated without good reason,
hence HOCN not HO=C=N not as discussed in section 2.1.4.
o

Like formal charges repel each other. Example: (())N=C=N(()) not NCN((=)).

o
Negative formal charge, if any, is more likely to be found on the more electronegative
atoms. By the same token, positive formal charge, if any, is more likely to be found on the more
electropositive atoms. Hence NCO(()) not O=C=N(()) as discussed in section 2.1.4.

It is very unusual for atoms in the top rows of the table (N3, up to and including chlorine)
to participate in more than four bonds. There are only so many antibonding orbitals available, so
there are only so many bonds you can make, no matter how many holes are available.
If you are ever tempted to violate this rule, it may mean that other rules are being violated as
well. Double-check that the compound really is electron-rich, and double-check the rules
about formal charge.


In some cases, such as O2, there will be half-strength bonds. Each such bond is represented
by a lone dot (not by a dash, and not by a pair of dots). (You cant detect such cases just by
counting holes; you need to make use of theory and/or data, such as paramagnetism data or
spectroscopic data.)

Sometimes, due to a symmetry, there are two equally-good ways to draw the molecule, as
in section 2.1.5. In such a case it is likely that the right answer is a quantum mechanical
superposition of the two possibilities. There is no very good way to draw a ball-and-stick model of
such a superposition. Usually the best you can do is draw both of the possibilities, and leave it to
the readers to form the superposition in their minds eye. (This is a lamentable limitation to the
whole ball-and-stick modeling endeavor; the laws of physics affect electrons differently from balls
and sticks.)
Hole-bonding diagrams correctly predict the number of bonds in every case where Lewis dot
diagrams had any hope of making correct predictions ... and they have the advantage that they can
be extended to additional cases, plus the advantage of being consistent with the paramagnetism
data, the spectroscopic data, et cetera.
2.2 How to Explain Hole-Bonding

At the simplest level, we can explain the hole-bonding method by saying there is abundant
observational and theoretical evidence that antibonding orbitals exist.
If somebody asks what, exactly, is the evidence, we can either tell them, or (if necessary) say that
the full details are beyond the scope of the course.
If you want a mental picture that folks can use to make the antibonding idea more concrete, you can
model each antibonding orbital as being a spring in compression, trying to destroy the molecule by
pushing the atoms apart. In particular, it behaves like an air-spring, i.e. a parcel of gas under
pressure, pushing the atoms apart. Each bond that we draw represents a net attraction, because it
portrays the absence of a repulsive force.
For a truly introductory treatment, stop here!
To a more-sophisticated audience, I would point out that the air-spring analogy is really rather
good. The pressure of the gas comes from the kinetic energy of the particles ... as does, ultimately,
the repulsive force of the antibonding orbitals.
If people can accept that the suns size is determined by a tradeoff between an attractive interaction
(gravitational PE) and an outward pressure (due to the KE of the particles), they ought to be able to
tolerate the idea that atoms also make a tradeoff between an attractive interaction (electrostatic PE)
and an outward pressure (KE again).
At the next level of detail, I remark that because of degeneracy, electrons produce a greater amount
of pressure than you would have expected based on a nave classical model. That is, an atom is
more like a neutron star than like the sun, as discussed in reference 1.

Also at the not-quite-introductory level, I would point out that compared to F 2,

O2 has fewer electrons but more bonding. The same remark applies at every step in the sequence
Ne2 F2 O2 N2, in the sense that progressively fewer electrons means
progressively more bonding. The number of bonds is observable in terms of the nonexistence of
Ne2 and the progressively increasing binding energy in the rest of the sequence. This is direct
evidence in support of the antibonding idea: we see progressively fewer electrons and progressively
more bond-strength. So ... this basic reactivity data provides evidence in support of the holebonding idea. What could be more appropriate than that? Its not necessary to make it more
complicated than that.
2.3 Spectroscopy and Energy Levels

At the next level of detail, I would point to the spectroscopic data, which has been around for over
100 years, as strikingly direct evidence for molecular orbitals.
A good way to visualize the orbitals, and the filling of orbitals, is to draw energy level diagrams
such as figure 20 through figure 23. If you want more information about what these diagrams mean,
see e.g. reference 2.
In all such diagrams, energy increases vertically. The energy is not shown to scale, but the ordering
of the levels is shown correctly. The electrons in the core levels are shown in gray, because they
almost never participate in chemical reactions.
The bonding versus antibonding character of each level is encode three ways in the diagram: The
tips of the level turn up for bonding levels, and turn down for antibonding orbitals. The name of the
level has a * in it for antibonding orbitals, and not for bonding orbitals. Finally, off to the right,
the levels are explicitly labeled antibonding or bonding.
The name of each level tells us something about the symmetry of the molecular orbital
wavefunction, and about how it can be derived from the atomic orbitals of the atoms that make up
the molecule.
In all of these four cases, there are three high-lying antibonding orbitals.

Figure 20: F2 Energy Levels

Figure 21: O2 Energy Levels

Figure 22: N2 Energy Levels

Figure 23: C2 Energy Levels

Figure 24: B2 Energy Levels

2.4 Limitations of the Naive Approach, and How to Overcome Them

The nave hole-counting procedure as described above is more-or-less mechanical and mindless. It
correctly predicts the number of bonds in all situations where the Lewis fairy-tale had any hope of
giving the right answer, namely in all situations where the molecule is sufficiently electron-rich that
all its bonding orbitals are filled. (That means, among other things, that adding an electron to the
molecule for instance by replacing C with N, or replacing N with O will reduce the net number
of bonds by half a unit, by occupying an antibonding orbital.)
For molecules that are not so electron-rich, nave counting doesnt suffice. The Lewis approach has
no hope of getting the right answer for such molecules, and cannot be repaired or extended. In
pleasant contrast, the molecular-orbital description does of course get the right answer. The
canonical illustration of non-nave counting is dicarbon. The octet fairy-tale predicts that it should
have a quadruple bond, but the real molecule has only a double bond. As we move along the
series FF, OO, NN, every time we remove two electrons we increase the number of bonds, by
depopulating antibonding orbitals, as shown in figure 20 through figure 22. But when we extend the
series, stepping from dinitrogen to dicarbon, we remove two electrons from a bonding orbital, as
shown in figure 23. Therefore the number of bonds goes down to 2, not up to 4. If all we cared about
was bond-strength, we would blissfully write C=C ... but that would give a misleading picture of
the hole-count. In some cases, we need a picture that assigns the correct formal charge in dicarbon.
In such cases, the preferred picture is C/=\C, where

The two horizontal lines represents two unfilled antibonding orbitals, as always. This
indicates two units of bond strength.

The diagonal lines indicate a pair of orbitals, one bonding and one antibonding, both
unfilled. These cancel each other out, making no net effect on the bond strength. If all we cared
about was bond strength, we wouldnt mention these at all, and we would just write C=C, but
sometimes we prefer to write C/=\C because it helps us account for what all the holes are doing.
The symbol C/=\C tells us we have a total of eight holes, i.e. four holes per carbon, i.e. a net neutral
charge ... and a net bond strength of two.
To repeat: you cant get the right answer for dicarbon just by nave counting; you have to actually
know something about the molecule. That is, you have to look at the energy levels of the molecular
orbitals, as shown in figure 23. For more information, see e.g. reference 2.

Similarly, for very electron-poor molecules such as Li 2, you are better off counting electrons than
counting holes. We depict this differently, namely LiLi in which the curly symbol represents
the presence of bonding electrons (counting up from helium) not the absence of antibonding
electrons (counting down from neon).
Amusingly, dicarbon can be represented in two ways, either as C/\C (counting up from helium) or
as C/=\C (counting down from neon):

Method a: Start from all valence orbitals filled, and count down, counting holes. This gives
us a bond strength of 2, due to holes in three high-lying antibonding orbitals ... plus holes in one
bonding orbital.

Method b: Start from all valence orbitals empty, and count up, counting electrons. This gives
us a bond strength of 2, due to electrons in three low-lying bonding orbitals ... plus electrons in one
antibonding orbital.
As a separate issue, nave counting does not suffice to predict that OO is a better description of the
oxygen molecule than O=O. We start by considering both possibilities, and then use the available
experimental and/or theoretical information to decide which version is better. One option is to
observe that liquid oxygen is attracted by a magnet. (In contrast, liquid nitrogen shows no
comparable attraction.) Alternatively, you dont even need a magnet, and you dont need a fancy
spectrometer: If you put liquid oxygen and liquid nitrogen in transparent flasks, you can see from
across the room that they are different. The oxygen has a pale blue color, due to the unpaired
electrons. A third option is to look at the molecular orbital energy-level diagram, where you
discover that there are two degenerate orbitals into which the highest-lying electrons can be placed.
In accordance with Hunds rule, the electrons will unpaired whenever there are enough
energetically-accessible orbitals to make unpairing possible. All in all, we have multiple lines of
evidence telling us that OO is vastly preferable to O=O.
Similar remarks apply to B2. A glance at figure 24 allows us to predict that it has one unit of bond
strength, that it is paramagnetic (which it is), and that its spectrum shows a triplet Zeeman splitting
of the ground state (which it does). To figure out what is going on in this molecule, it is easier to
count electrons rather than holes, counting up from helium rather than counting down from neon
(although counting down would give the right answer also). To depict this molecule in a way that
shows the bonding as well as the antibonding, you can write B\;/B. Here we use a semicolon
instead of a colon to indicate that we are counting electrons, not holes. Similarly we write the
slanted lines the other way (i.e. \ / instead of / \) to indicate electrons not holes.
In general, if you observe that a molecule is paramagnetic, it tells something about the degeneracy
of the highest molecular energy level, and conversely if you know the energy levels you can predict
the paramagnetism. This is in stark contrast to the Lewis approach, which makes irreparably wrong
predictions about O2, and cannot describe B2 at all.
Consider the contrast:
Heres the central unifying idea: In general,
you need to think about the molecular orbitals.

For electron-rich molecules, all the bonding

orbitals are filled, so a simple mechanical

You need to figure out which bonding orbitals

are occupied and which antibonding orbitals
are occupied.

hole-counting procedure suffices. Holecounting is not fully general, but its domain of
applicability covers a remarkably large number
of molecules, including most of the molecules
commonly encountered in organic chemistry.

Some peculiar additional examples are discussed in section 5.

3 Some Supporting Ideas

3.1 The Aufbau Principle

Aubau is the German word for building up.

The Aufbau principle has been known since the earliest days of quantum mechanics. It is a rule of
thumb, not a rigorous theorem. Among other things, it tells us that when we go from O 2 to F2, we
expect the two molecules to have qualitatively similar energy-level diagrams, which is just what we
see in figure 21 and figure 20. The main difference is that in F2, the levels are more fully occupied.
Same levels, more occupation.
There are exceptions to the Aufbau principle. You can see that in going from N 2 to O2, there is one
pair of levels that re-order themselves. But the rest of the energy-level diagram is pretty much
undisturbed.
It is worth noting that as we go from OO to F-F, the number of bonds decreases. F-F has more
electrons but less bonding. That is consistent with the Aufbau principle, because we are not
depopulating pre-existing bonding orbitals; we are populating a new antibonding orbital.
For more on this, see section 8.4.

4 Approximate Bond Angles

4.1 Basic Ideas

It is important to know the shape of molecules. For instance, the well-known properties of water
depend crucially on the fact that the molecule is bent, not linear in shape.
Lewis dot structures are often praised as being a stepping-stone toward VSEPR (Valence Shell
Electron Pair Repulsion) which is a convenient method for predicting bond angles. Like Lewis dot
diagrams, VSEPR is based on the unjustifiable idea of molecular octets, and an unjustifiable focus
on bonding orbitals (to the neglect of antibonding orbitals).
The purpose of this section is to point out that approximate bond angles can be
predicted without basing the prediction on molecular octets. This is the final nail in the coffin of
molecular octet notions.
This will be presented as an empirical rule of thumb, without any deep derivations.

The ideas here in section 4 are much less well-founded than in other sections. When we say to use a
wooden molecule of a carbon atom, pre-drilled with four holes in a tetrahedral pattern, you may
reasonably ask why we should start there. The answer is that empirically and retrospectively,
starting there leads to qualitatively-correct predictions in a goodly number of simple cases.
Disclaimer: Lets be clear: The models presented here in section 4 are partly bogus and partly
fortuitous.

As models of where the electrons are located and what they are doing in the molecule, such
models are completely bogus. You should not imagine for a moment that a macroscopic classical
model could properly represent anything as ultramicroscopic and utterly quantum-mechanical as
the electrons in a molecule.

As a model of how the nuclei in the molecule are arranged, these models are justified a
posteriori by the fact that they make qualitatively correct predictions in a number of interesting
cases. Validity is restricted atoms on the right-hand side of the upper rows of the periodic table, and
even then there may be exceptions.
You shouldnt stake your life on any predictions made by such models but lets keep things in
perspective: These models are in no ways worse, and are in some ways better, than models based
on Lewis octets in molecules. The a posteriori justification is about the same, and the a
priori wrong assumptions are fewer.
So here it is: The rule-of-thumb is that for elements in the N=2 and N=3 rows of the periodic table,
the bonding pattern will be tetrahedral, or a subset of tetrahedral, unless there is compelling reason
for it to be otherwise.
4.2 Some Examples
4.2.1 Methane

As a first example, methane is tetrahedral. Thats not very tricky. Why shouldnt it be tetrahedral? It
is easy to build a hands-on model of this, using wooden balls for atoms and tiny coil springs for the
bonds. The carbon atom is pre-drilled with four holes in a tetrahedrally-symmetric pattern. Pictures
can be found in reference 3.
4.2.2 Ammonia

An important, interesting example is ammonia. Let the vertices of the basic tetrahedron be
called A, B, C, and D. In the classical approximation,2 sites B, C, and D are occupied by ligands,
while the A site is unoccupied. This is an example of what I call a subset of the tetrahedral
configuration. This particular subset is called trigonal pyramidal ... but thats just another way of
saying tetrahedral with one vacancy. Again it is easy to build a hands-on model of this. Again the
carbon atom is pre-drilled with four holes in a tetrahedral pattern. Three of the holes receive bonds,
while the fourth one goes vacant. Pictures can be found in reference 3.
4.2.3 Water

The next example is water. The hydrogen ligands occupy sites C and D, while sites A and B are
unoccupied. I am not saying that A and B are occupied by lone pairs(which is what VSEPR
would say); Im just saying they are unoccupied. This configuration is properly called bent ... but
thats just another way of saying tetrahedral with two vacancies.
I am explicitly ducking the question of detailed bond angles. The theory, at its present state of
development, predicts symmetry, not details. In the case of water, it says that the molecule is bent,
not straight. It says that the bond angle is approximately the tetrahedral angle, but does not predict
whether the angle is slightly greater or slightly less than the mathematically-perfect tetrahedral
angle.
4.2.4 CO2 and the Energy of Double Bonds

Similarly carbon dioxide is linear. It is conventional to diagram the molecule as O=C=O ... but that
is somewhat misleading, because it suggests that all four bonds are straight and parallel to each
other. A somewhat more informative diagram is shown in figure 25.

Figure 25: Carbon Dioxide : Linear Molecule, Bent Bonds

At this level of approximation, we can say that the four bonds depart the central carbon atom in the
four tetrahedral directions, and then bend as necessary to terminate on the oxygen atoms. Two of
the bonds lie are in the xy plane, while the other two lie in the xz plane.
Again it is easy to build a ball-and-spring model of this. The model depends on the fact that the
bonds are bendable and springy. Pictures of such models can be found in reference 3.
Heres why this is important: We expect a double bond to be stronger than a single bond, but
nowhere near twice as strong (other things being equal). We can explain this by saying in a double
bond, the bonds are bent and under stress, which costs energy. We can understand this energy as
follows: The holes bear a positive charge, so there is some electrostatic energy involved in bending
them closer together. There is also an increase in kinetic energy, since momentum depends directly
on the curvature of the wavefunction, and kinetic energy depends on momentum squared. A
tetrahedral arrangement of straight bonds has minimal potential energy (since the bonds are as far
apart as possible) and minimal kinetic energy (since the bonds are straight).
4.2.5 HCN, C2H2, and the Energy of Triple Bonds

The HCN molecule is linear, not bent, and its easy to see why. There is a triple bond to the
nitrogen, specifically HCN. Here (as always) we use a dash to represent two holes, i.e. the
absence of antibonding electrons, not the presence of bonding electrons. We think about the shape
as follows: start with the default tetrahedral configuration. The hydrogen is bonded to the A vertex.
The nitrogen is bonded to the B, C, and D vertices. The nitrogen ion core sits at a location that is
the average or the sum of the B, C, and D vectors. If you look at the geometry of the tetrahedron,

you see that this puts the nitrogen directly opposite the hydrogen ... so we have an easily-visualized
model of why HCN is linear.
Similarly, acetylene (HCCH) is linear.
A triple bond is even more highly stressed, more energetic, and more reactive than a double bond.
Among other things, this explains why acetylene has remarkably high fuel value.
4.2.6 CO3=

Consider the carbonate ion, CO3=. In the classical approximation, you can draw it with one doublebonded oxygen and two single-bonded oxygens. From this you can easily (and correctly) predict
that it will be planar. At the next level of sophistication, moving beyond the classical
approximation, there will be resonance, i.e. the three oxygens will take turns having the double
bond, so the molecule is actually threefold symmetric. This configuration is called trigonal planar.
In this case you can build three different not-quite-correct classical models; the correct symmetry is
obtained by averaging these three.
4.2.7 Nitrogen Dioxide Radical

Lets look at NO2. The constituent atoms are shown in figure 26.

Figure 26: Nitrogen Dioxide Constituent Atoms

There are seven holes, which is enough to make three and a half bonds. Note that any molecule
where not all the electrons are paired is called a radical. There are many ways of drawing the bonds
in such a molecule. At this level of detail, it is not at all obvious whether we should draw the
molecule as linear or bent. Here are some facts that we might use as the basis for an analogy:

A triatomic molecule with two single bonds is bent. Example: H 2O (water).

A triatomic molecule with one single bond and one double bond is bent. Example: NOCl
(nitrosyl chloride).

A triatomic molecule with two double bonds is linear. This makes sense because of the
symmetry of the situation. Example: CO2 (carbon dioxide).
Hole-counting theory does not provide enough information to determine whether NO 2 is linear or
bent. The situation is somewhere between NOCl and CO 2, so it could go either way. However,
experiment tells us that it has a nonzero permanent dipole moment, so it must be bent. Therefore we
choose to draw basis states as shown in figure 27.

Figure 27: Nitrogen Dioxide Basis States

The actual molecule is shown in figure 28. It is a resonant superposition of the two basis states
shown in figure 27. Each oxygen is bonded to the nitrogen by one ordinary bond plus a partial bond
with 3/4ths of a unit of bond strength.

Figure 28: Nitrogen Dioxide Molecule

We can contrast this with figure 29, which shows a hypothetical linear molecule. It has four partial
bonds, each of which has 7/8ths of a unit of bond strength (for a total of 3.5 units). Hole-counting
does not rule this out, but experiment tells us that this shape is disfavored relative to the bent shape.

Figure 29: Nitrogen Dioxide Wrong

The physics here can be partially rationalized as follows: There are many contributions to the
overall energy. Resonance makes a contribution that tends to favor the linear structure. Meanwhile,
bond-bending (as discussed in section 4.2.4) makes a contribution that tends to favor the bent
structure. Hole-counting cannot tell you which of these contributions will win out.
4.2.8 Other Cases

Things get trickier if we consider bonding that involves elements in later rows of the periodic table,
beyond N=3 ... but lets not worry about that right now.
Some harder-to-explain examples (including SF 6 and SF4) are discussed in section 5.

5 Some Peculiar Examples

Here are some examples that dont fit the simple pattern. It should be emphasized that there are
millions of molecules for which you can draw a satisfactory bonding diagram without thinking very
hard, but there are a few where you cant.
5.1 The Nitro Group

One example that requires special handling is the nitro group, as found for instance in nitromethane
(CH3)NO2. Mindlessly playing the connect-the-dots game might produce the following diagram:
O - O
\ /
N
|
HCH
H

which is straightforward, elegant, symmetric, and wrong. The main problem is that its inconsistent
with the IR spectroscopy data, which shows floppy, low-frequency bending of the ONO angle,
inconsistent with closure of the three-member ring via an OO bond. Instead it supports the
following ringless structure:

(-)
O
O
\\ /
N(+)
|
HCH
H

(-)

<-->

O
\ //
N(+)
|
HCH
H

I imagine theres also NMR, NQR, and reactivity data confirming theres a positive charge on the
nitrogen.
To understand this result, you need to consider bond angles, as discussed in section 4. The natural
angle for the ONO and OON bonds should be close to the tetrahedral angle (109 degrees).
Distorting these angles to form a 606060 degree triangle is not energetically feasible.
To say the same thing more simply: In the real molecule, the oxygen atoms are significantly farther
apart than the drawings above might suggest.
5.2 Sulfur Hexafluoride

As a second peculiar example, consider SF 6. Naive hole-counting tells us this molecule has 8 holes,
which is enough for four ordinary bonds ... but theres a problem, because there are six ligands. It is
totally unacceptable to have 6 bonds with 2/3rds of a unit of bond-strength apiece, because QM
tells us that using s and p orbitals gives us only four bonds, and only four places to put bonds. The
only possible explanations for SF6 must involve contributions from some supplementary orbitals
in addition to the conventional valence orbitals 3s and 3p. Hypotheses to be considered include 2p,
3d, 4p, and perhaps others. The energy levels of these supplementary orbitals is high, but not so
wildly high as to make them completely inaccessible. This completely changes the hole-counting
process, because there is a whole new set of bonding and antibonding molecular orbitals to be
considered.
5.3 Sulfur Tetrafluoride

As a third peculiar example, consider SF 4. Naive hole-counting tells us this molecule has six holes.
That means it cannot have four ordinary chemical bonds; it can have at most three ... even though
there are four ligands. Thats actually the right answer. Apparently the four fluorines take turns
being bonded, with each one being bonded three quarters of the time. This is not different in
principle from the nitro group discussed in section 2.4, or the carbonate ion discussed in section 4,
except that in those cases the ligands take turns having a second bond, while in SF 4 the ligands take
turns having their first bond. You might imagine this makes SF 4 somewhat unstable against
disintegration, and youd be right.
I dont know of any useful convention for drawing a 3/4-strength bond.
We should also consider the hypothesis that SF 4 has two ordinary bonds (each using a pair of holes)
and two half-bonds (each using one unpaired hole). However, the lack of paramagnetism indicates
this is not what happens. I dont have any deep theoretical basis for predicting this, so for now lets
just consider it an observed fact.

Old-fashioned VSEPR predicts a remarkable see-saw shape for SF 4. This would be considered a
triumph for VSEPR, except that molecular dynamics studies indicate that that its not really a good
description of whats going on. I dont know how to describe the shape of this molecule, and I
certainly dont have a simple way of predicting the shape.
We must also consider the hypothesis that d-orbitals are involved, which would throw off the hole
count.

* PART II
6 Overview of the Contrast
The relationship between the various available methods is shown in figure 30, which is a plot of
what you get out (results) as a function of what you put in (effort).

Figure 30: Results versus Work for Various Methods

The figure makes several important points, including:

For advanced problems, the full molecular orbital approach is called for. However, for
simple problems, this approach is not worth the trouble.

For the very simplest problems, hole-counting makes the same predictions as the Lewis
approach, with a similarly small amount of work. For slightly more advanced problems, the more
work you put into hole-counting, the more correct results you get out. The farther you push it, the
more it begins to look like the molecular orbital approach.

The farther you push the Lewis approach, the worse it looks. It does not connect to the
sophisticated molecular orbital approach. If/when you want to learn the sophisticated approach, you
will have to unlearn any notion of octets in molecules. Unlearning is always difficult.

Choosing between hole-counting and molecular orbitals involves judgment and requires
tradeoffs, trading off effort versus accuracy of the results. In contrast, choosing between holecounting and Lewis octets is an easy decision, because hole-counting is Pareto superior. That is, it
is sometimes better and never worse.

Compared to Lewis dot structures, it appears that BABE and hole counting are better in every way,
for the following reasons:
1.
In every case where the Lewis dot method had any hope of getting the right answer, hole
counting gives the right answer.
2.
In every case where the Lewis dot method had any hope of getting the right answer, hole
counting is no harder. It is usually easier.
3.
The whole idea of filled Lewis octets in molecules is an impediment to deeper
understanding. It must be completely unlearned before you do any sort of spectroscopy or
molecular physics. In contrast, hole counting is a stepping stone on the way towards deeper
understanding.
4.
There are simple cases (including the familiar O 2 molecule) where the Lewis dot approach
stubbornly makes wrong predictions. In contrast, hole counting makes it possible to draw diagrams
that correspond to reality.
We shall see that the Lewis dot method is limited both as to domain and to range. By domain, we
mean the class of molecules that it applies to. By range, we mean what it says about a given
molecule. The tradeoff between domain and range is shown in table 1. In the table, domain runs
horizontally, while range runs vertically.
Molecule
Number of Bonds
All Electrons Paired
Filled Octet

CH4
CO
Yes
Yes
Yes
Yes
Yes
no
Table 1: Lewis Scorecard

O2
Yes
no
no

C2
no
Yes
no

B2
no
no
no

Remember that this section is just an overview; supporting evidence will be presented
in section 8 and later sections.
In table 1, you can that if we severely restrict the domain, keeping only the first column (namely
simple hydrides such as CH4, NH3 etc.), then over this domain the Lewis method makes quite a
range of successful predictions. If you want to sell the Lewis method as a theory of simple
hydrides, that would be OK. Alas, people often try to sell it as something much more than that.
The second row of the table is important, because it represents not just the CO molecule but a wide
class of electron-rich molecules, including most biomolecules. The Lewis method can predict the
number of bonds for such molecules. However, the idea that atoms in such molecules have filled
octets just like neon is hogwash, as we know from the spectroscopy data, from modern theory,
and from many other lines of evidence.
Note that C2 is just barely outside the domain of electron-rich molecules. CO is sufficiently
electron-rich, while C2 is not.
We now switch switch from a column-by-column discussion to a row-by-row discussion of the
table.

As you can see from the top row of table 1, if we restrict the range to predictions as to number of
bonds and nothing else, the domain gets bigger. If you want to sell the Lewis method as an
unexplained, unprincipled mnemonic for predicting only bond order in electron-rich molecules
made from certain elements in the upper-right corner of the periodic table, I might even buy it.
As a minor point of terminology, the term bond order is sometimes used to denote the distinction
between single, double, and triple bonds. In this context the order is just a number, as in order-ofmagnitude, or the order of a polynomial; it does not refer to the sequencing (or ordering) of the
bonds left-to-right or anything like that.
The situation soon becomes ugly because most textbook authors cannot resist the temptation to
explain the Lewis method in terms of atomic physics principles. They throw around physics
ideas such as the energy of filled shells in neon, and then claim that sharing allows each atom in
(say) the O2 molecule or the F2 molecule to have a filled shell just like neon. Scare quotes are
necessary because usually nearly every word of such an explanation is false. 3
There are of course other limitations. The basic Lewis method obviously has no hope of explaining
a molecule such as SF6. Some authors try to patch this up in terms of an expanded
octet.4 However, given that there arent really any octets in typical molecules, it seems pointless to
talk about expanding something that doesnt exist.

7 Fun With Lewis Dot Diagrams ... Or Not

This section serves to put hole-bonding in context, relative to the widely-taught Lewis dot method.
Lewis dot diagrams have lots of problems, and it is possible to do much, much better, with zero
additional work, using hole-counting and related methods as discussed in section 2.
7.1 Hydrides and Other Successes

Of course the Lewis dot method is not entirely without merit. Sometimes it makes correct
predictions. For starters, as far as I can tell, it gives a satisfactory description of molecules in the
sequence CH4, NH3, H2O, HF, Ne, which are all well-behaved molecules ... and conversely it
correctly predicts that related entities such as CH 3 are not well-behaved molecules under ordinary
conditions. Another reasonably satisfactory example is NaCl, as discussed in section 10.2.
7.2 Oxygen

There are, alas, other cases where Lewis dot diagrams stubbornly and irreparably make wrong
predictions.
Reference 4, from UC Berkeley, shows a conventional Lewis dot structure for the oxygen molecule,

O2, which can also be called dioxygen. This is the way chemistry is taught in a lot of places, not
just Berkeley. Figure 31 portrays the same idea, using a style I prefer because it is slightly more
explicit. The black dots represent electrons. The reddish shaded areas represent orbitals. The
proponents of such diagrams like to call attention to the following good points:

There are a total of 12 valence electrons involved, as there should be.

Each oxygen atom is connected to four doubly-occupied orbitals, in accordance with the socalled octet rule. [See reference 5 for a statement of the octet rule(s).] In particular, this upholds
the DCBO rule (double counting of bonding orbitals), which lies at the heart of the Lewis method.
That is, the four electrons that make up the two bonds are counted as part of the first oxygens
filled octet and then counted again as part of the second oxygens filled octet.

The diagram predicts that O2 contains a double bond.

Figure 31: Conventional Unsatisfactory Lewis Dot Diagram for O2

However, this diagram is nonsense. It cannot possibly be correct. The most obvious problem is that
O2 is well known to be paramagnetic. If you pour liquid oxygen into the gap of a magnet, it will
stick to the magnet. There is an animated GIF image of this in reference 6 and a nice still image
in reference 7. Paramagnetism arises from unpaired electrons. Figure 31 categorically predicts that all
electrons are paired. Therefore:

The diagram falsely predicts that O2 is not paramagnetic.

There have been various attempts to solve this problem. One attempt, from the University of
Illinois, is shown in reference 8. Figure 32 portrays the same idea, using my preferred style.

Figure 32: Single-Bond : Unsatisfactory Lewis Dot Diagram for O2

This solves the paramagnetism problem, but creates several others.

It violates every imaginable version of the octet rule. There are only seven electrons in the
vicinity of each oxygen atom. Since octets is one of the guiding principles for the construction of
Lewis dot diagrams, if we sacrifice the octet rule in this case it calls into question practically
everything that anyone has ever done using diagrams of this sort.

The unpaired electron sticking out would make this a viciously reactive radical, which is
inconsistent with our common knowledge about the behavior of the oxygen molecule.

It predicts a single bond in the O2 molecule. Since the fact that figure 31 predicts a double
bond is usually cited as one of the great successes of the Lewis dot method, a diagram that predicts
a single bond must be considered a failure.
So lets try again. Hope springs eternal.

Reference 9 shows an oxygen molecule with one regular bond and two half-bonds. Figure 33 portrays

the same idea, using my preferred style.

Figure 33: Two Half-Bonds : Unsatisfactory Lewis Dot Diagram for O2

Alas this, too, has its problems.

We have five orbitals attached to each oxygen atom. That seriously calls into question the
conceptual foundation of Lewis dot diagrams, since it is mathematically impossible for an atom to
have more than four orbitals in the N=2 shell. If you dont want to give up the notion of octets
entirely, you at least have to give up the DCBO rule.

If there are five orbitals available for O 2, why are there not five orbitals available for F 2? By
the aufbau principle, we should be allowed to add two electrons to figure 33, and the result should
be isoelectronic with F2. This predicts a triple bond in F2, in gross disagreement with conventional
wisdom.
We would prefer not to have a theory of chemistry based on rote dogma ("four legs good, two legs
bad, eight dots good, ..."). Instead we would like to have a rule based on physics, mathematics, and
logic.
For isolated atoms in the N=2 and N=3 rows, there is a valid octet rule, to wit: there are four
orbitals near each atom, and each orbital can be occupied by at most two electrons (because of the
exclusion principle). So far so good. The problem arises only when we try to extend this rule to the
atoms in molecules. There arises a question of how to interpret the word near in this rule. The
Lewis method chooses an interpretation that leads to DCBO (double counting of bonding orbitals)
and this is irreparably wrong.
The five orbitals in figure 33 means either that the figure is wrong, or we need to entirely abandon
DCBO in general and Lewis dot diagrams in particular.
7.3 Sulfate Ion; Expanded Octets
Figure 34 shows the Lewis dot diagram for the sulfate ion. Remember, in this section, the dots in the

diagrams represent electrons, not holes.

Figure 34: Sulfate Ion Lewis Dot Diagram

This isnt particularly terrible. However, it is unnecessarily uglier and more complicated than the
conventional diagram we saw in figure 15.

Within the Lewis formalism, just as in other formalisms, it is conventional to represent a bond by a
dash, so when a pair of dots represents a bond we can replace it with a dash. In the case of the
sulfate ion, the result is shown in figure 35.

Figure 35: Sulfate Ion Lewis Structure

This can be compared and contrasted with figure 15.

If you are interested in formal charge, figure 35 can be interpreted as assigning a +2 formal charge to
the central sulfur atom, and 1 formal charge to each of the oxygen atoms. This is the same
interpretation as we saw in section 2.1.7. It is consistent with the known electronegativity, and
consistent with the known symmetry.
For some reason, figure 36 is commonly (albeit not universally) presented as the Lewis structure
for the sulfate ion. It can be found in the Encyclopedia Britannica, on the wikipedia site, and in
many textbooks.

Figure 36: Sulfate Ion Another Lewis Structure?

This has the small advantage of having zero formal charge on the sulfur atom. In return it incurs the
large disadvantage of violating the Lewis octet rule. There are six bonds to the central sulfur
atom. Another disadvantage is that it violates the known symmetry; the molecule is observed to be
tetrahedral, all four oxygen atoms are observed to be equivalent. It seems odd that even articles that
say in the text that the oxygens are equivalent and tetrahedral persist in using a non-symmetric
drawing.
You could try to salvage the symmetry by saying the final structure is a quantum mechanical
superposition of all the various ways of arranging the single and double bonds, leading to an
average bond strength of 1.5 everywhere.
No matter what you do with the symmetry, we still have a violation of the Lewis octet rule since
there are still six bonds to the central sulfur atom. Sometimes people try to wave away this
violation by talking about an expanded octet, presumably involving the sulfur atoms d orbitals.
This is troublesome because it doesnt explain why an octet might be expanded in some cases but
not others.

In particular, the symmetry tells us there cant be any d orbitals involved. The tetrahedral shape has
the symmetry of an sp3 hybrid; any nontrivial admixture of d orbitals would change the symmetry.
Since there was never a rational physical basis for Lewis octets to begin with, dont hold your
breath waiting for a physical basis for the expanded octets.
One possible Lewis structure for sulfuric acid follows the same pattern, as shown in figure 37.

Figure 37: Sulfuric Acid Lewis Structure

Just as figure 36 is a modification of figure 35, there is a widely-seen alternative to figure 37,
namely figure 38.

Figure 38: Sulfuric Acid Another Lewis Structure?

Again this violates the Lewis octet rule.

The expanded octet problem is not confined to sulfates; the various sulfonic acids are almost
universally drawn with six bonds to the central sulfur atom. See section 7.4.
7.4 Methanesulfonic Acid

It is worth noting that the drawing of methanesulfonic acid in figure 18 (in section 2.1.8) shows single
bonds everywhere, and only four bonds to the central sulfur atom.
Thats remarkable, because almost all references show this molecule with double bonds to two of
the oxygen atoms, for a total of six bonds to the sulfur atom, as shown in figure 39.

Figure 39: Methanesulfonic Acid Lewis Structure?

Lets consider the pros and cons of figure 39, according to the usual Lewis notions. We find (a) there
is formal charge neutrality everywhere, which is good; (b) there are filled Lewis octets on the

oxygen atoms on the outside of the molecule; although (c) the so-called octet rule has been
seriously violated on the central sulfur atom.
Again as mentioned in section 7.3, sometimes people try to wave away the octet violation by talking
about an expanded octet, presumably involving d orbitals.
In any case, in the BABE scheme, there is no problem, as discussed in section 2.1.8. Since we know
that filled Lewis octets never existed to begin with, since we are not double-counting bonding
orbitals, we have no fear of unfilled octets (on the oxygens or anywhere else).
7.5 Divergent Predictions

Recall that one place where the BABE method diverges from Lewis octets is that the Lewis method
stubbornly predicts that the O2 molecule has the O=O structure. It cannot accommodate the OO
structure.
Here we have a second place where the methods diverge. The BABE method stubbornly predicts
single bonds (and nothing but single bonds) in the sulfate ion, sulfuric acid, and methanesulfonic
acid. In contrast, the Lewis method (as practiced by experienced professionals) usually makes a
different prediction than the BABE method does concerning sulfate, and always (as far as I can tell)
makes a different prediction regarding methanesulfonic acid.
This should have observable consequences in the electron density. The double bond has zero
electron density along its midline. Is this observed in real life, or not? I dont know for sure, but I
predict not.
Heres one reason why the BABE method cannot predict a double bond in these molecules. In
going from figure 35 figure 36, we re-arranged some electrons. We took some electrons from a lone
pair and used them to make another bond. This cannot happen in the BABE approach, because
there are no lone pairs. The concept of lone pairs does not exist, so theres nothing to re-arrange.
You cant just add another bond by pulling it out of the air, since that would change the overall
electrical charge of the molecule.
In case youre wondering whether there might be holes in a d orbital that could be used to form the
additional bond, thats a clever idea, but it wont work. In these electron-rich molecules, bonds are
holes in antibonding orbitals. All the low-lying d orbitals are bonding, and you cant make bonds by
putting holes in such orbitals. There are lots of games you can play with d orbitals, but none of
them (so far as I can tell) increase the bond order in these molecules.
7.6 Additional Remarks

Note that the easy method of counting formal charge by counting bonds, as suggested
in section 2.1.4, only works within the BABE formalism, i.e. when the bonds represent pairs of holes
in antibonding orbitals in electron-rich molecules. In contrast, in the Lewis scheme, counting
formal charge would be trickier, because we would need to account for lone pairs as well as bonds.
This is another of the reasons why the BABE method is just plain easier.

If you are only interested in bond order, and you are sure you will never do any spectroscopy or any
molecular structure calculations, you might conclude that Lewis dot diagrams are OK, for electronrich compounds (with a few exceptions such as O2).
Please keep in mind that except for the hydrides mentioned in section 7.1, Lewis dot diagrams make
false predictions about the spectroscopy and molecular orbitals of almost all molecules, not just
oxygen. O2 is just the canary in the coal mine, in the sense that it is where we first noticed the
problem. But it did not cause the problem, and ignoring the dead canary will not solve the problem.
As Kim Philby said: To betray, you must first belong. Applying this to the notion of filled Lewis
octets, the fact that this fundamentally wrong notion makes some correct predictions about bondorder is a big part of what makes it so pernicious.

8 Conceptual Foundations (or lack thereof)

8.1 Contrast: Molecular Orbitals versus Lewis

The bond-strength diagrams discussed in section 2 are firmly based on sound theory, as we now
discuss.
This is called Molecular Orbital theory, or MO theory for short. Like most theories, it seems simple
if you understand it, and formidably complicated if you dont understand it. Like any good theory,
watered-down, approximate, qualitative versions are available ... and thats where we should start.
Here then are some of the qualitative things we know about MO theory:
The MO explanation is consistent with the
paramagnetism data.
The MO explanation is consistent with the
optical spectroscopy data.
The MO explanation is consistent with
reactivity data, including things like B2 and
Be2.
The MO explanation is consistent with theory:
atomic physics, quantum mechanics, and all
that.

The Lewis dot diagram stubbornly predicts no

paramagnetism, contrary to observations.
The Lewis dot diagram has little to say about
spectroscopy, and what it does say is
categorically wrong.
There are many cases including B2 and Be2
where Lewis dot diagrams make predictions
that cannot be reconciled with observations.
The Lewis dot approach was published about
ninety years ago (reference 10), considerably
before there was a comprehensive quantummechanical understanding of molecular
bonding (reference 11). The Lewis approach
cannot be reconciled with present-day
theoretical understanding.

A reasonably-accessible discussion of molecular orbitals, including a sketch of the energy-level

diagram for O2, can be found in reference 2. Another useful source is reference 12, which contains
energy-level diagrams for some interesting molecules (water, hydrogen fluoride, ammonia,
methane, ethane, ethene, ethyne).
Among the things we learn is that the Lewis approach is a dead duck. We shouldnt be surprised by
this; it made the wrong predictions in figure 31, it was violated in figure 32, and it was toppled off its

foundations in figure 9. The MO explanation tells us that these are not isolated or superficial
mistakes, but instead result from a systematic and fundamental misconception. DCBO is just
wrong.
The MO explanation tells us that there should
be eight molecular orbitals (not all of which
will be filled). We know from classical physics
and mathematics (Liouvilles theorem) that
eight is the right answer.

In the Lewis dot approach, e.g. figure 31, we

see six molecular orbitals, 2 unshared on the
left, 2 shared in the middle, and 2 unshared on
the right. We know (experimentally and
theoretically) that this is the wrong number.

Ironically enough, this means that while figure 9 has too many orbitals to be a valid Lewis structure
(7 instead of 2+2+2), according to the molecular orbital explanation it doesnt have too many just
the wrong kind.
This leads us to a key point: The data and the theory tell us that antibonding orbitals exist (as well
as bonding orbitals of course). In particular, the two electrons that are responsible for the
paramagnetism of O2 sit in antibonding orbitals, as shown in figure 21.
This in turn means that there is no hope of fixing the Lewis dot formalism, except by throwing it
out and starting over, as was done in section 2. A proper theory cannot be based on DCBO. A proper
theory must represent antibonding electrons (not just bonding and non-bonding electrons). A proper
theory must have a way to take into account the energy levels of the molecular orbitals, to provide a
foundation for applying Hunds rules.
8.2 Contrast : Hole Bonding versus Lewis

Consider the contrast:

The old Lewis approach requires folks to
believe in the unjustified and unjustifiable
DCBO rule. It requires them to believe that
O2 has six molecular orbitals (two on the left,
two shared, and two on the right).

The new hole-bonding approach requires folks

to believe in the well-founded notion of
antibonding orbitals (as well as bonding
orbitals) and degrees of occupation thereof. In
fact O2 has eight molecular orbitals (four
bonding and four antibonding).

At this level of detail, the hole-bonding story is more truthful and less complicated than the Lewis
story.
At this point,

We have said no false things, unlike the Lewis approach.

We have said far more true things than the Lewis approach ever will.

Our approach remains conceptually simpler than the Lewis approach will ever be, since we
have introduced fewer new abstractions.

Our diagrams are easier to draw, for ordinary molecules, since there are fewer dots involved.

8.3 What Implies What

Let us examine what does and what doesnt constitute evidence for the Lewis octet rule. This
will help us understand its domain of validity. A high-level summary of the situation is shown
in figure 40. The rest of this section is devoted to explaining the details.

Figure 40: Inference Diagram

When applied to a molecule like O 2, the Lewis octet rule says that the number of actual valence
electrons, plus the number of electrons involved in bonds, must add up to 16, i.e. one octet per
atom. The point is that the electrons involved in bonds are intentionally double-counted. Since there
are two electrons per bond, this leads to the bond-order formula:

bond order =

Now let us contrast this with a proper molecular-orbital explanation.

We start with the would-be molecule Be2. The MO diagram nicely explains the lack of BeBe chemical bonding; both the s bonding orbital and the s* antibonding orbitals are filled,
resulting in no net bonding. So trying to form Be 2 using 2s orbitals is like trying to form He2 using
1s orbitals ... its just not gonna happen. This correctly contrasts with the observed LiLi bonding,
in which only the s bonding orbital is filled. (Lewis dot diagrams cannot begin to explain the
pattern going from He2 to Li2 to Be2.)
At the other end of the N=2 row, let us consider the would-be molecule Ne 2. The MO
diagram correctly predicts no NeNe chemical bonding, because all eight molecular orbitals
are filled. For every filled bonding orbital, there is a filled antibonding orbital, so there is
zero net benefit in forming the molecule. This correctly contrasts with the F 2 molecule, in
which one of the antibonding orbitals is empty, leading to a net bond-strength of 1.
As we move another step to the left, we come to O 2. Relative to F2, this has two fewer
electrons, depopulating one more antibonding orbital. (Actually it half-depopulates two such
orbitals.) In any case the result is a net bond-strength of 2.

Moving again to the left, in N2 we have three fully-depopulated antibonding orbitals, for a
bond-strength of 3.
For the four cases Ne2, F2, O2, and N2, we find that equation 3 summarizes the facts. Note the
word summarizes.
The point is that equation 3 only summarizes the four facts, without explaining any facts
whatsoever. Its at best a mnemonic. Its little more than numerology. To see the distinction
between summarizing and explaining, just extend the sequence one more step to the left:
When we get to C2, we depopulate another orbital ... but this time we depopulate
a bonding orbital, reducing the bond-strength to 2.
The bond-strength peaks at 3 at N2. From there, moving either to the left or to the right
causes bond-strength to decrease.
Equation 3 gets worse and worse as we continue moving to the left across the N=2 row.

So lets see how the contestants have scored:

1.
When applied to pairs of atoms from the N=2 row, equation 3 is right about half the time. This
comes from comparing the formula to observed results, and is independent of whatever derivation
or rationale (if any) you think lies behind the formula.
2.
The Lewis octet rule has less validity than the bond-order formula. That is, if we try to
interpret equation 3 in terms of octets, we are right less than half of the time, because the octet
picture predicts not just bond-order, but also falsely predicts that all the electrons in O 2 are paired.
And the Lewis octet picture is grossly incompatible with the spectroscopic data.
3.
If we interpret equation 3 terms of molecular orbitals, we realize it applies if-and-only-if all
four of the molecules bonding orbitals are full, i.e. when we are considering putting electrons into
zero or more of the three antibonding orbitals that lie at the top of the energy ladder. So in some
sense its an antibonding formula, not a bonding formula.
4.
More importantly, the MO explanation tells us when we should and when we shouldnt pay
attention to equation 3. The MO explanation gives us a way of understanding the whole sequence
He2, Li2, Be2, B2, C2, N2, O2, F2, Ne2 (including the spectroscopic data and the paramagnetism data).
To repeat: equation 3 summarizes the data over a narrow range, but is not (by itself) an explanation.
Explaining the formula in terms of octets only makes things worse. MO theory explains equation 3,
explains its limitations, and explains lots more besides.
8.4 The Aufbau Principle (or not)

This continues the discussion that began in section 3.1.

It is worth noting that the Aufbau principle is utterly incompatible with the Lewis octet idea. Lets
do an example, starting with O2. The Lewis theory tells us that each atom in the molecule will
have an octet of electrons around it. If we now construct F 2 in accordance with the Aufbau

principle, we leave the oxygenic electrons alone, and add two more electrons. There are now more
than eight electrons around each atom.
For most molecules, this is an inescapable conflict. Either the Aufbau principle is wrong, or the
Lewis octet idea is wrong. (The wrongness of the Lewis octet idea is discussed in section 7.)
The Lewis model explains bond order in terms
of bonding pairs and non-bonding pairs.
If you think in terms of bonding and nonbonding, there is no Aufbau-compatible way to
explain how F2 can have more electrons but
less bond strength.

MO theory explains bond order in terms of

bonding and antibonding.
It is perfectly compatible with the Aufbau
principle to say that when we go from OO to
F-F, we leave the bonding levels alone but add
electrons to an antibonding level.

There is a big difference between antibonding and non-bonding. There is a big difference between
BABE and BNBE (i.e. Lewis dot structures).

9 Discussion
Almost all scientific results, experimental and theoretical, are approximate and imperfect.
One criterion for judging scientific approximations is that they ought to
be controlled approximations. That is, the error ought to be small, we ought to have an upper bound
on how big the error is, and we would like to have a way of making the error smaller if and when
we need to.

As an example, consider the notion that kinetic energy is m v2 in the non-relativistic limit.
That is an excellent approximation in ordinary classroom situations, but thats not the main point.
The point is that the approximation can be made better and better by limiting the range of
velocities, and/or by including higher-order correction terms. For that matter, the fully relativistic
expression is available if needed.

As a counterexample, consider the rule about eating oysters in months with an R in them
(September through April). Oysters depend in logical ways on the weather, but there is no logical
connection from that to the way we spell the names of our months. So the oyster/R rule, to the
extent that it has any validity at all, is somewhere between a coincidence and a mnemonic. Even if
the rule works in one part of the world, you have little or no confidence that it will work in another
part of the world, because it cannot possibly be sensitive to the relevant variables.
Lewis dot diagrams do not do well under this criterion. The Lewis octet rule seems to be an all-ornothing proposition. Either the O2 molecule has just six molecular orbitals, or it does not. In fact it
does not, as we know from the paramagnetism data, the spectroscopy data, et cetera. So Lewis dot
diagrams must be put into the same category as the oyster/R mnemonic.
Another rule of good scientific practice is to be up-front about whatever approximations are being
made. This is where the most immediate improvements can be made. If you feel you must draw
Lewis dot diagrams, go ahead, but make sure the audience understands that:


Lewis dot diagrams are at best phenomenological and empirical, not based on firm
theoretical foundations.

They make many incorrect predictions in addition to whatever correct predictions they make.

Much better explanations of whats going on are available.

Another rule of good scientific practice asserts that it is bad manners to criticize a result for being
imperfect unless youve got something better to suggest. In this case the suggestion is to use the
bond-strength diagrams described in section 2, which are preferable to Lewis dot diagrams in every
way.
Similarly, the bond-order formula equation 3 is valid over a limited range, but still has more validity
than the Lewis octet picture. It can be seen either as a narrow corollary of the molecular orbital
explanation, or it can be used on its own merits as an empirical / phenomenological / mnemonic /
numerological rule.
It is perfectly possible to visualize the results of the bond-order formula. You can draw pictures of
molecules including FF, OO, NN, C/=\C, B\;/B, and LiLi ... without connecting any of this to
molecular octets. We connect this to correct theory by saying that each or represents one unit
of bond strength. A represents a filled bonding orbital, while a represents an unfilled
antibonding orbital.
Things go awry when people try to use the successes of the bond-order formula as evidence for the
Lewis octet rule, by back-tracking along arrow (2) in figure 40. It is not logical to follow arrow (2)
while ignoring arrows (1), (3), and (4). It is not fair to pay attention to the occasional fortuitous
successes of the DCBO rule, while ignoring its failures and ignoring the many successes of
competing hypotheses.
To say the same thing in other words: What many people claim as successes for the molecular octet
rule are nothing of the sort; they are really just successes of the bond-order formula for electronrich molecules equation 3 with no necessary connection to the molecular octet rule.

10 Additional Background and Context

10.1 The Hole Concept

The Hall effect (reference 13) was discovered in 1879. It was found that some metals (e.g. aluminum
and magnesium) have positive Hall coefficients, indicating positively-charged majority carriers
what we would nowadays call holes. Note that this was many years before the date conventionally
assigned to the discovery of the electron (reference 14).
The notion of holes is central to any discussion of solid-state electronics. This goes back to Peierlss
1929 explanation of the Hall effect (reference 15).
Holes have been part of quantum field theory since Diracs epochal work in 1930 ( reference 16, page
293):

The annihilation of a negative-energy electron is to be understood as the creation of a hole in the

sea of negative-energy electrons, or the creation of a positron.
The idea of holes in the context of bonds between atoms is not new, either. In 1969, Baym
(reference 17, page 454) remarked that a halogen could be regarded
... as having a weakly bound hole which also easily forms chemical bonds.
Im not quite sure why he put the word hole in scare quotes.
10.2 Coulombic Models

Lets consider a crystal of NaCl. It exhibits ionic bonding, in contrast to the covalent bonding in
(say) a silicon crystal. It can be described reasonably well in classical terms: the Na + ion and the
Cl ion are modeled as classical hard spheres (of appropriate sizes), and they are held together by
Coulomb interactions.
This is consistent with the idea of Lewis octets, in the sense that the Na + ion and the Cl ion can be
explained in terms of filled atomic octets. (Remember, atomic octets actually have some basis in
fact, whereas molecular octets almost never do.)
It must be emphasized that the Coulomb interaction between ions is quite different from the
covalent bond. In the real world, some bonds exhibit a mixture of ionic and covalent character.
NaCl crystals are near the ionic extreme, while silicon crystals are at the covalent extreme.
Coulombs law can be expressed in a variety of equivalent forms. These include:

V(r) = [

where V(r) is the voltage (i.e. electrical potential) at location r due to a charge Q at the origin. The
quantity in square brackets is a universal constant, called Coulombs constant. It is occasionally
written kC but more commonly as 1/(4 0).
An equivalent expression is:
F(r) = [

where F(r) is the force on a charge q at location r due to a charge Q at the origin. The r/|r| factor is
a unit vector in the r direction. The force is directed radially outward if Q and q have the same sign.
The Coulomb interaction depends on the charge (but not the mass) of the particles. It is a longrange force, in the sense that the interaction energy falls off like 1/r. In contrast, the size and
strength of the covalent bond depend on the mass (as well as charge) of the electron. The covalent
bond is necessarily a short-range interaction: at large distances the wavefunction falls off
exponentially, at a rate determined by the mass of the electron.
Coulombs equation tells us the electrostatic potential energy, but we must also include the kinetic
energy. We must include the kinetic energy if we are to have any hope of explaining many of the
most well-known and important chemical facts, including:

The fact that atoms have any nonzero size.

The fact that some bonds are directional. The nontrivial shape of everything from water
molecules to DNA molecules depends on the existence of directional bonds. Coulombs equation
(plus some hand-waving) give an acceptable description for ionic compounds such as NaCl, and for
structureless molecules such as CH4, but not for more complicated structures. The Coulomb
interaction is purely radial, so it has virtually no ability to control bond angles.

The non-rotation in ethylene (H2C=CH2) as opposed to the rotation in ethane (H3CCH3).

The umbrella inversion in the ammonia molecule.

The nontrivial variation of bond order as a function of atomic number, such as we see in the
sequence C=C, NN, OO, FF.

et cetera.

Charles-Augustin de Coulomb died in 1806, more than 100 years before the advent of quantum
mechanics. Therefore speaking of a Coulombic model rather strongly implies a classical model.
We know that any correct description of covalent bonding requires non-classical ideas far, far
beyond what you can get from equation 4.
Some people who are fed up with Lewis dot diagrams try to solve their problems by suggesting we
can describe everything (or almost everything) in terms of Coulombic models. This is not a good
idea. Lets be clear: Talking about a Coulombic model of the water molecule or the ethylene
molecule is at best misleading. It is at best an abuse of the terminology.
Coulombs law is part of the description of any chemical bond (including covalent as well as
ionic), but is equally true that bolts are part of a car. We dont talk about the bolt model of how
cars work, and we shouldnt talk about the Coulombic model of bonding, except perhaps for
ionic compounds such as NaCl. Coulombs law is a law of physics, but it is only one law among
many; it is not a complete description.

(In contrast, it is OK to talk about a quantum mechanical model. Thats because QM explicitly
incorporates all the other laws of physics. On the RHS of the Schrdinger equation we find the
Hamiltonian, which includes all the contributions to the energy. In contrast, the Coulomb energy
calculated using equation 4 is just one contribution to the energy, one contribution among many.)

11 Summary
There is nothing you can do via Lewis dot structures that you cant do at least as easily and at least
as accurately via hole-counting. Hole counting, within its domain of applicability, is based on
sound theory, whereas Lewis dot diagrams are based on a fairy tale.
There are many cases where Lewis dot diagrams predict bond-order in agreement with the facts.
However, this agreement must be considered fortuitous, because there are also many disagreements
that ought not be overlooked. The method recommended here accounting for antibonding orbitals
as well as bonding orbitals retains all the good features of Lewis dot diagrams while getting rid of
the disagreements.
There is an octet rule for atoms that is
reasonably well founded in atomic physics for
a single row-2 atom. Truly there is something
special about the complete octet (i.e. closed
shell) in neon.
In reality, there are eight valence-shell
molecular orbitals in molecules such as O2.
This is the starting point for the theory of
chemistry described here. This is the right
answer.
Naive hole-counting wont tell you whether
OO or O=O is preferable. Making no
prediction about this question is better than
making a wrong prediction. We leave the door
open for other theoretical and/or experimental
information to decide the question.
Hole-counting gives a good accounting of its
own limitations. It is easy to understand
that nave hole-counting works in electron-rich
molecules, including a very wide range of
important biomolecules. Meanwhile, it is easy
to see that slightly less-nave accounting is
needed to explain C=C and other systems that
are not electron-rich ... without in any way
contradicting what was said about electronrich systems.

Chemistry is primarily about molecules, not

isolated atoms, especially not noble-gas atoms.
For molecules, the Lewis approach couples the
octet idea to DCBO (double counting of
bonding orbitals), which is a disaster.
The Lewis approach says O2 has only six
molecular orbitals: two on the left, two shared,
and two on the right. This is the wrong answer.

Lewis octets stubbornly predict O=O, in

defiance of the well-known facts. There is no
way to draw a Lewis octet that fits the facts.

Lewis theory does not explain its own

limitations. There is no way to explain C=C in
terms of Lewis octets, nor any good way to
explain why not. Octets are either a
fundamental principle, or theyre not. If the
octet principle explains N2, why doesnt it
explain O2 or C2?

Anybody who says my approach is identical to the Lewis approach isnt paying attention.
In every case where Lewis dot diagrams make
correct predictions, hole-counting makes the
same predictions (i.e. the correct predictions)
with the same amount of effort, or less.
Hole-counting is based on molecular orbital

Lewis dot diagrams are, of course, better than

nothing. The successes of Lewis dot diagrams
area a subset of the successes of hole-counting.
The Lewis dot diagram scheme is based on

theory. It is a shorthand, summarizing (not

supplanting) some well-established facts.

Within MO theory we can visualize bond order

just as easily as we could using Lewis dot
diagrams. For instance we can draw NN,
where each dash represents an unfilled
antibonding orbital. In any case where you
could have drawn a Lewis dot diagram, you
can draw a bonding diagram consistent with
MO theory and consistent with the facts
with the same amount of effort. In other cases,
the correct diagram is slightly more work, but
its a small price to pay for correctness.

DCBO, and on the neglect of antibonding, and

on other assumptions that are inconsistent with
whats really going on in the molecule. It is a
miracle that Lewis dot diagrams make as few
wrong predictions as they do.
Lewis dot diagrams are easy to draw, easy to
visualize, compact, and elegant ... but rife with
error.

12 References
1.

John Denker, Pressure, Degeneracy, Exchange Interaction, Neutron Stars, Atoms,

Etc. ./degeneracy.htm
2.

Mark R. Leach Diatomics : Molecular Orbital Theory http://www.metasynthesis.com/webbook/39_diatomics/diatomics.html

3.

John L. Park, Model

Building http://dbhs.wvusd.k12.ca.us/webdocs/Bonding/Lab-ModelBuilding/
especially http://dbhs.wvusd.k12.ca.us/webdocs/Bonding/LabModelBuilding/H2O-NH3.jpg
and http://dbhs.wvusd.k12.ca.us/webdocs/Bonding/Lab-ModelBuilding/CO2HCN.jpg
4.

The lower-left corner of the image shows the conventional Lewis dot diagram for O 2;
used in a UC Berkeley chemistry course circa 1995:
./img48/berkeley-chem1-oxygen.gif
5.

John Denker, Models and Pictures of Atomic

Wavefunctions www.av8n.com/physics/wavefunctions.htm
6.

Liquid O2 pouring into the gap of a

magnet. http://www.chem.wisc.edu/deptfiles/genchem/demonstrations/Movie
s/o2parasm.gif

7.

Liquid O2 poured into the gap of a

magnet. http://demoroom.physics.ncsu.edu/multimedia/images/demos/5G3020
.jpg
8.

Single bond in O2 molecule. http://www.chem.uiuc.edu/clcwebsite/jpeg/O2.jpg

9.

Part B of the figure shows one bond plus two half-bonds in O 2 molecule;
used in a chemistry course at Penn State:
./img48/psu-o2-lewis.gif
10.

Gilbert N. Lewis "The Atom and the Molecule" JACS 38 762786

(1916) http://dbhs.wvusd.k12.ca.us/webdocs/Chem-History/Lewis1916/Lewis-1916.html
11.

Linus Pauling, The Nature of the Chemical Bond (1939).

12.

W. Locke, Introduction to Molecular Orbital

Theory http://www.ch.ic.ac.uk/vchemlib/course/mo_theory/main.html
13.

Edwin Hall, On a New Action of the Magnet on Electric Currents American Journal of
Mathematics 2 (1879).
14.

J.J. Thompson, Philosophical Magazine 44 293 (1897).

15.

Rudolf Peierls, On the theory of the Hall effect, Physikalisches Zeitschrift 30 273274
(1929).
16.

Paul A. M. Dirac, Principles of Quantum Mechanics (1930).

17.

Gordon Baym, Lectures on Quantum Mechanics (1969).

Of course other things are not equal, and the kinetic energy of the wavefunction is at least as
important as the potential energy, so the charge-distribution rule will remain a rule of thumb,
not a law of nature.
2

The classical approximation works OK for liquid ammonia, and for ammonia in aqueous
solution. It fails spectacularly for isolated ammonia molecules, because the protons tunnel
from one side to the other. For almost all molecules heavier than ammonia, the classical
approximation works OK for describing the position of the nuclei, and quantum mechanics is
only needed for describing the electrons. But for the isolated ammonia molecule, even the
nuclei must be described quantum mechanically. There is a wonderful discussion of this in
Feynman volume III chapter 9.
3

It could be argued that the Lewis theory works for simple hydrides such as CH 4, NH3 etc.,
but a theory that only works for a small number of obviously-exceptional cases cant be
considered a respectable theory, especially when it claims to work for a wide class of cases
where it does not.
4

As a minor point, we note in passing that expanded octet is a contradiction in terms. Octet
means 8, and 8 is not expandable. I assume they meant to say expanded shell or
something like that.
[Contents]
http://www.av8n.com/physics/draw-molecules.htm
http://archives.library.illinois.edu/erec/University%20Archives/1505050/Rogers/Text7/Tx74/

http://chem.libretexts.org/LibreTexts/Lansing_Community_College/LCC%3A_Organic_Chemistry_I
%3A_251/Chapters/Chapter_1/1.1

https://www.youtube.com/watch?v=7RR098ulxVI

Lesson Transcript
Instructor: Nissa Garcia

Nissa has a masters degree in chemistry and has taught high school science and college level chemistry.
Sulfurous acid has a chemical formula of H2SO3. It is known for its bleaching properties so it is used to bleach various products. In this lesson, we will
learn about sulfurous acid, some of its uses, and how to draw its Lewis structure.

What Is Sulfurous Acid?

There are some chemical compounds that are well known because of how useful they are and others we recognize because of its odor. In the case of
sulfurous acid, we can recognize it because of its suffocating sulfurous odor.
Sulfurous acid has a chemical formula of H sub 2 SO sub 3. This compound is formed when sulfur dioxide (SO sub 2) is dissolved in water. It is a weak
acid, so it is not as corrosive as strong acids, like hydrochloric acid or sulfuric acid, and it is an oxoacid since it is an acid with oxygen atoms in its
chemical formula.

Chemical Formula of Sulfurous Acid

Sulfurous acid acts as a good reducing agent because it has the ability to take up oxygen from the air and other substances that are rich in oxygen. It is
also exhibits good bleaching properties.
Did you know that sulfurous acid is a major component of acid rain? Also, because of its bleaching properties, it is used to bleach various products that
we are all familiar with and use - paper products and straw products, like bags and hats.

Sulfurous Acid - Lewis Structure

We know the chemical formula of sulfurous acid, but how are the atoms bonded together? We will figure this out by constructing its Lewis structure.
Some rules that we need to go over before constructing its Lewis Structure are:

Hydrogen cannot be the central atom

The hydrogen atom must bond with the oxygen atom in oxoacids

The atom with the lowest electronegativity is usually the central atom

Now, let's construct the Lewis structure.

Step 1: Count the valence electrons from the Hydrogen, Sulfur and Oxygen atoms

H has 1 valence electron, S has 6 valence electrons and O has 6 valence electrons. Adding all the electrons, the total is 26 electrons.

Count the Total Valence Electrons

Step 2: Determine the central atom

In general, the central atom is the least electronegative atom (we exclude hydrogen from being the central atom). The question is, will it be Oxygen or
Sulfur?
If we look at the electronegativity values in the periodic table below, the electronegativity increases from left to right and from bottom to top.

Electronegativity in the Periodic Table

Based on their positions, sulfur is less electronegative, so sulfur will be our central atom. Here is the proper arrangement of Hydrogen, Sulfur and
Oxygen :

Spatial arrangement of the atoms

Step 3: Place electron pairs between the atoms

One electron pair between two atoms is equivalent to one line.

Draw Electron Pairs Between Atoms

Step 4: Place remaining electrons around the other atoms

Out of 26 valence electrons, we've used 10. We need to distribute the 16 valence electrons left among the atoms.

Account for the number of valence electrons

Hydrogen can only contain a maximum of two electrons, so we don't add any more electrons to Hydrogen. Sulfur and Oxygen can contain 8 electrons,
so we add more electrons to them to complete the octet (8 electrons).

Distribute the remaining valence electrons

Now, we have used all the 26 valence electrons. It looks like a good Lewis structure, but we need to do one more step.

Step 5: Calculate the formal charge

There may be a better way to draw the Lewis structure and we need to do calculate the formal charge to verify that. Here's the formula for calculating the
formal charge:

Formal Charge Formula

Let's calculate the formal charges for the Hydrogen, Sulfur and Oxygen atoms. The red dots are nonbonding electrons and the green lines are bonding
electrons. There are more than one Oxygen and Hydrogen atoms, so let's label them individually to distinguish them. Here are the formal charges for
each atom:

Formal Charges for H, S and O Atoms

To unlock this lesson you must be a Study.com Member. Create your account

http://study.com/academy/lesson/h2so3-definition-lewis-structure.html

Lab 5 - Molecular Geometry

Purpose

To explore some simple molecular structures.

To explore the relationship between bond order and bond length.

To explore resonance structures.

Goals

To compare Lewis structures to three-dimensional models.

To visualize the three-dimensional structures of some common molecules.

To obtain bond angle, bond length, and hybridization data for molecules.

To rationalize differences in predicted and measured values.

To learn how to use computer modeling software.

Introduction
The chemical and physical properties of covalently bonded materials are related to the spatial arrangement of the atoms and other electrons not involved in the actual bond
formation. There are many ways to depict the spatial arrangement in both two and three-dimensions. A Lewis structure is a two-dimensional representation of the
arrangement of the atoms, bonding electrons and non-bonding (lone pair) electrons in a covalent material.
In a Lewis structure, the nucleus is represented by the atomic symbol with a line between the atoms in a bond depicting each pair of shared bonding electrons in the structure.
Non-bonding electrons around the atoms are depicted as dots The steps to building a Lewis structure representation of a molecule are shown below. The Lewis structure of
the formate ion, CHO2-, will be used as an example.

Calculate the electrons required (ER) = the minimum number of electrons necessary to satisfy the octet rule for the non-hydrogen atoms and the duet rule for
hydrogen. For CHO2-, this would be (2 electrons x 1 hydrogen atom) + (8 electrons x 3 non-hydrogen atoms) = 2 + 24 = 26 electrons required.

Calculate the number of available valence electrons (VE) = the total number of electrons available for the molecule. For example, in CHO 2-, this would be (1 C atom
x 4 electrons) + (1 H atom x 1 electron) + (2 O atoms x 6 electrons) + (1 electron as the ion has a charge of -1) = 4 + 1 + 12 + 1 = 18 valence electrons. NOTE: For
ions, the charge must be included in this by adding the charge on an anion or subtracting the charge on a cation.

Calculate the Shared Pairs (SP) = the number of electrons to be shared in bonds. The SP is 1/2 (ER-VE); for CHO 2-, this would be 1/2 (26 - 18) = 4 shared pairs or
four bonds.

Calculate the Lone Pairs (LP) = the number of electron pairs belonging to only one atom. The LP is 1/2 (VE-(2xSP)); for CHO 2-, this would be 1/2(18 - (2x4)) = 5 lone
pairs. Notice that VE = 2 x (SP + LP).

Place the first atom in the molecular formula as the central atom, surrounded by the other atoms in the compound.

Draw bonds (shared pairs) from the central atom to each surrounding atom. The bonds are represented as lines; each line represents two electrons. The number
of lines should be equal to the number of shared pairs calculated in step 3, which in this case is four. Since hydrogen follows the duet rule, it only prefers one
bond. The fourth bond can be drawn to either one of the oxygen atoms.

Draw lone pairs on each of the non-hydrogen atoms. A lone pair is represented as two dots; each dot represents an electron. Each non-hydrogen atom prefers
eight electrons in the vicinity of the atom. If an atom has 1 bond, it requires 3 lone pairs. If an atom has 2 bonds, it requires 2 lone pairs. If an atom has 3 bonds, it
requires 1 lone pair. If an atom has 4 bonds, do not add any additional lone pairs. In our example, C requires no lone pairs, one oxygen requires 3 lone pairs and
one oxygen requires 2 lone pairs.

8
Calculate the formal charge (FC) on each atom. For an atom:

(1)
FC = VE - SP - (2 x LP)
where SP is the number of shared pairs on the atom and LP is the number of lone pairs on the atom. For molecules the sum of the formal charges of all the atoms must be
zero; for an ion, the sum will be the ionic charge. To choose the more favorable of two Lewis structures, the one with the lowest FC on each individual atom would be the most
likely candidate. Remember that the more electronegative atom will generally prefer the negative formal charge. The formal charge for each atom on the formate ion is shown
below. Note how the sum of the formal charges (0+0+0+-1) adds up to the -1 charge on the formate ion.

The structure shown above is the Lewis structure representation of the formate ion. Note that the structure could also have been draw with two bonds directed towards the
oxygen on the right. This is known as a resonance structure. Resonance structures have the same number of electrons, but may have a different number of shared pairs or
the atoms in the structure may have different formal charges.

Once a Lewis structure has been drawn, one can evaluate the bond order of shared pairs. A shared pair is a single bond with a bond order of 1; two shared pairs in a bond is
a double bond (bond order = 2) and three shared pairs in a bond is a triple bond (bond order = 3). In general, the bond strength increases and bond length decreases as the
bond order increases. (The bond length is the distance between the centers of the two atoms bonded together). Thus, a single bond is longer than a triple bond, but the triple
bond is stronger. NOTE: This general statement applies when comparing the bond order of similarly bonded atoms.
However, as noted, the Lewis structure is a two-dimensional representation of the molecule or ion. The two-dimensional diagram does not show us the geometry or shape of
the molecule. Over the years, many theories have attempted to explain the shape of covalently bonded substances; one of the simplest is
the Valence Shell Electron Pair Repulsion (VSEPR) theory. In most molecules, electrons occur in pairs, either as part of a bond (bonding pair) or as a lone pair; these pairs

occupy space around an atom. The theory proposes that these areas or electron regions repel one another. VSEPR theory's main postulate is that the regions around a given
atom will arrange themselves to minimize this repulsion by positioning themselves as far apart as possible. This leads to a predictable shape based on simple geometry.
For the purposes of the VSEPR theory, it is more helpful to think about electron groups (rather than just electron pairs) and the regions the groups occupy in space. A lone pair
or bonding pair are each considered to be an electron group. However, the four electrons in a double bond or the six electrons in a triple bond are also considered to be one
electron group. In each case they occupy one region where they are bonding two atoms together.
The position of the electron groups determines the angles at which centers of those electron regions are located. As a result, the angle between atoms that may be at the ends
of those electron groups is also determined. These angles for covalent compounds that obey the octet rule are shown in the table below. The values in the table give the
angles between the centers of the electron regions. If these are bonding groups (single, double or triple bonds), this number also is the bond angle.

Also shown in the above table are the hybridized orbitals used to form the s-bonds (sigma bonds) and to hold the lone pairs. Chemists invoke hybrid orbitals based on the
assumption that one atom will point its electron region directly at an atom to which it bonds. s orbitals do not point in any direction, and simple p atomic orbitals lie 90 o apart
along the x, y and z-axes. Regions in space that point 109, 120 or 180 apart result by taking mathematical combinations (hybrids) of the simple atomic orbitals. Hybrid
orbitals have thus been incorporated into the language that describes covalent bonding.
When we use the term molecular geometry or molecular shape, we are not describing the shape of the electron regions, but rather, the location of the atoms. The words used
to describe the shapes are therefore describing the location of the atoms. Where four atoms surround a central atom, the shape would be tetrahedral. Where three atoms
surround a central atom, the shape would either be trigonal planar or pyramidal. Where two atoms surround a central atom, the shape would either be linear or bent.
It is difficult to represent three-dimensional structures, like molecules, on a flat piece of paper. It is equally as difficult to look at a molecular structure drawn on a piece of paper
and imagine what it would look like in three-dimensions. With the help of a molecular model kit and a computer modeling program, you will be able to visualize a molecule in
three-dimensions. In this lab, you will use a computer program within WebAssign that allows molecules to be rotated, just like you could manually rotate a model built with a
model kit. You will also be able to use the computer program to obtain experimentally determined measurements, such as bond angles and bond lengths, for the molecule
under investigation.
In introductory chemistry courses, we often predict bond angles and bond lengths in simple molecules based on VSEPR or hybridization. However, our predictions are simply
approximations because other factors influence the structure of real molecules. In this lab, you will compare the bond angles and bond lengths predicted from theory to the
experimentally determined values. You will then be asked to explain differences that occur between the experimental values and the theoretical values.

Equipment

Chem-Tutor model kit

computer with internet browser

Molecular Geometry In-Lab assignment in WebAssign

About the model kit

The Chem-Tutor model kit was designed by Professor Samuel G. Levine for use by his organic chemistry students at North Carolina State University. Atoms: Atoms are
represented by jacks that have one, two, three or four pegs coming out. Each peg represents an electron region (or electron group). The electron regions (groups) can be
single bonds, double bonds, triple bonds or lone pairs. In general, black jacks are used for carbon, blue for nitrogen, red for oxygen, white for hydrogen and yellow for sulfur.
More important than color is the number of pegs. The type of atom/jack to use is determined by the Lewis structure of the molecule. Bonds: Green bonds can be used for
single, double or triple bonds. Do not bend the green bonds to force a molecule into a particular shape. Although you are strong enough to bend these bonds, it takes energy
for atoms and molecules to bend away from the predicted angles and shapes. The molecules prefer to exist in the lowest energy configurations ("straight" green bonds). Black
bonds between two carbon atoms can be used for carbon-carbon double bonds. You will notice that the pegs on each carbon define a plane, and that the two carbons are
coplanar. The atoms at the ends of the bond are fixed in place. DO NOT attempt to twist this bond. White bonds are made out of rubberized plastic, and are the only ones that
are meant to bend. They are only used for special "strained" structures that you should not encounter in this lab.

About the computer modeling software

The molecules in your WebAssign In-Lab are viewed with jmol, a javascript plug-in. The computer files viewed with jmol contain real data from measurements on these
molecules. Often we find that molecules follow what is predicted from textbooks. Occasionally, they do not. Do not be surprised if what you observe is sometimes different

than what you might predict! Rotating a molecule: A molecule in the viewing area can be rotated in all directions. This can be achieved by holding down the left mouse button
while moving the mouse. After a few tries at this, you should be able to manipulate the molecule any way that you want. To determine a bond length: Double click on one of
atoms in the bond. The cursor will change to a "+". Then double click on the other atom in the bond. The measurement will be displayed in nanometers (nm). To make the
measurement disappear, re-measure the bond in the same way. To determine a bond angle: Double click on one of the atoms of the angle. The cursor will change to a "+".
Then single click on the vertex of the angle. And last, double click on the on the third atom in the angle. The measurement will be displayed in degrees. To make the
measurement disappear, re-measure the angle in the same way.

Reagents
No chemicals will be used in this experiment.

Safety
None of the materials being used in this experiment present a safety hazard. However, the work is being done in a laboratory and the usual rules about eye protection and
proper clothing apply.

Waste Disposal
Since no chemicals are being used in this experiment, there will not be any waste for disposal.

Prior to Class
Please complete WebAssign prelab assignment. Check your WebAssign Account for due dates. Students who do not complete the WebAssign prelab are required to bring
and hand in the prelab worksheet.

Lab Procedure
You will notice the format of this lab experiment is different than other experiments. You will build a series of models and investigate them on the computer. Questions to help
you with your observations are intermingled with the procedure. Please answer the questions in your lab manual along with any other observations you make while you are
building the structures.

Launch Internet Explorer.

Open one partner's Molecular Geometry In-Lab in WebAssign.

Please print the worksheet for this lab. You will need this sheet to record your data.

Part A: Exploring Simple Structures

For each of the following molecules,

Review the correct, complete Lewis structure(s), including any resonance structures and any formal charges that you drew in your Prelab assignment.

Build the molecule with the model set.

Look at the molecule in your In-Lab assignment on WebAssign.

Fill in the table and answer the questions below.

1. Two Electron Regions Dinitrogen Oxide (N2O) Hint: Nitrogen is the central atom. 2. Three Electron Regions a. Sulfur Dioxide (SO2) b. Formaldehyde (CH2O) 3. Four
Electron Regions a. Water (H2O) b. Ammonia (NH3) c. Methane (CH4)
Table A: Exploring Simple Structures
Question A1: For each of the six molecules, how did your Lewis structures compare to the molecular models and the models on the computer? Were they the same or
different? Explain.
Question A2: For each of the six molecules, was your Lewis structure a good and accurate representation of the molecule's actual shape? Explain why or why not.

Question A3: Did the model set and computer models help you identify the molecular shape better than the Lewis structures? Do you think models are helpful with 3D
visualization?
Question A4: Did you have any other interesting observations? Please elaborate.

Part B: Bond Order vs Bond Length

For each of the following molecules,

Review the correct, complete Lewis structure(s), including any resonance structures and any formal charges that you drew in your Prelab assignment.

Build the molecule with the model set.

Look at the molecule in your In-Lab assignment on WebAssign.

Fill in the table and answer the questions below.

a. Ethane (C2H6) b. Ethene (C2H4) c. Ethyne (C2H2)
Table B: Bond Order vs Bond Length
Question B1: What conclusions can you draw about bond order and bond length?
Question B2: Looking back at your data in Part A, are all single bonds the same length? Based on these observations, can you make a generalization about the length of all
single bonds compared to double bonds or all double bonds compared to triple bonds? What general rule can you make?
Question B3: Did you have any other interesting observations? Please elaborate.

Part C: Resonance Structures

Draw the Lewis structures for Part C on your lab worksheet.
For each of the following molecules,

Review the correct, complete Lewis structure(s), including any resonance structures and any formal charges that you drew in your Prelab assignment.

Build the molecule with the model set.

Look at the molecule in your In-Lab assignment on WebAssign.

Fill in the table and answer the questions below.

a. Benzene (C6H6) Hint: Place the carbons in a hexagon. b. Carbonate Ion (CO3 2-) c. Thiocyanate Ion (SCN-) Hint: Carbon is the central atom.
Table C: Resonance Structures
Question C1: Which of the three molecules had resonance structures that were equal? Which did not? Explain.
Question C2: How can you confirm that the resonance structures are equal for a molecule? Explain.
Question C3: If there was a molecule with unequal resonance structures, which structure is the best according to the computer modeling? Can you tell which structure the
computer is displaying? How? Do your observations agree with what you have learned about formal charge?
Question C4: Did you have any other interesting observations? Please elaborate.

When you are done, clean up any model set structures and leave the sets as you found them when you arrived in the lab.

Before leaving, go to a computer in the laboratory and enter your results in the In- Lab assignment. If all results are scored as correct, log out. If not all results are
correct, try to find the error or consult with your lab instructor. When all results are correct, note them and log out of WebAssign. The In-Lab assignment must be
completed by the end of the lab period. If additional time is required, please consult with your lab instructor.
Copyright 2010 Advanced Instructional Systems, Inc. and North Carolina State University. | Credits

http://www.webassign.net/ncsugenchem102labv1/lab_5/manual.html

http://www.chem.ucla.edu/~harding/IGOC/F/formal_charge.html

Formal Charge

F.C.= # valence e- - # non-bonded e- - 1/2 # bonded e-

Here are a few examples

Bonding Links
http://www.kentchemistry.com/links/bonding/formalcharge.htm

https://superbpaper.com/?cid=1700
https://www.google.com.eg/url?
sa=i&rct=j&q=&esrc=s&source=imgres&cd=&cad=rja&uact=8&ved=0ahUKEwimn87puYHRAhX
LtBoKHfGFASoQjB0IBg&url=http%3A%2F%2Fwww.stoeth-fuchsstadt.de%2Fwhat-is-the-generalstructure-of-a-thematic-analysisessay&psig=AFQjCNFBzg1h2WiCoP1HR4eiCkTGQpiZVA&ust=1482277437165951

http://people.uwplatt.edu/~sundin/114-s6/image/1460412j.gif

What is the hybridization of so2?

https://socratic.org/questions/what-is-the-hybridization-of-so2

First draw the Lewis Structure of so2.

Note that there are 3 things or regions around the central atom. 1 single bond, 1 double bond and a lone pair.
The hybridization for this is sp2
Where:
s has 1 orbital
p has 3 orbitals
d has 5 orbitals
f has 7 orbitals
sp2 s has 1 orbital, p has 2 orbitals = 3 regions
Let's say you have 4 things or regions around the central atom like nh3

The hybridization for this is sp3

sp3 s has 1 orbital, p has 3 orbitals = 4 regions
Let's look at PCL5. It has 5 things or regions around the central atom.

The hybridization for this is sp3d

sp3 s has 1 orbital, p has 3 orbitals, d has 1 orbital = 5 regions
We can't have more than 3 orbitals on p so that's why we jumped to d orbital.
Don't forget to check this also

How do you draw VSEPR diagrams?

https://socratic.org/questions/how-do-you-draw-vsepr-diagrams
Answer:
The Valence Shell Electron Pair Repulsion Theory (VSEPR) helps us to understand the 3D structure of molecules.

Explanation:
The general concept is that the pairs of electrons repel each other and try to locate themselves as far as possible from
each other about a given nucleus.
Hence, for two pairs of electrons on a nucleus, the two pairs would locate themselves exactly opposite each other,
forming a bond angle of exactly 180.
If three pairs exist, they will locate themselves in a plane about the nucleus at angles of 120 from each other.
Here is a table of the electron pair geometries as a function of the number of electron pairs.

4.bp.blogspot.com

To determine the shape of SO2, for example, we first determine the Lewis dot structure of SO2.

upload.wikimed
ia.org

The central atom, S, has three groups bonded to it, two oxygen atoms and a lone pair.
The electron pair geometry of SO2 is trigonal planar. It is drawn as

butane.chem.uiuc.edu

The molecular geometry of SO2 is not trigonal planar.

In determining the molecular shape, we consider only the positions of the atoms, not the lone pairs.
So, the molecular shape of SO2 is bent and is represented as:

upload.wikimedia.org

The lone pair of electrons occupies a relatively large volume, since they are held by only one atom.
They compress the bond angle between the oxygens and sulfur to less than 120. The actual O-S-O bond angle is
119.5.
Here is a table listing the molecular shapes that correspond to various combinations of bonding pairs and lone pairs.

es.boundless.com

Was this helpful? Let the contributor know!

figur

Tang 04 lewis dot diagrams

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.

1. LEWIS DOT DIAGRAMS

2. LEWIS DOT DIAGRAMSDraw: H2O H O H
3. LEWIS DOT DIAGRAMSDraw: H2O H O H 1+ 6+ 1 =8 electrons
4. LEWIS DOT DIAGRAMS Draw: H2O H O H 4 electrons remaining1+6+1=8 electrons
5. LEWIS DOT DIAGRAMSDraw: H2O H O H 4 electrons remaining The octets on hydrogen are already full.
6. LEWIS DOT DIAGRAMSDraw: H2O H O H 0 electrons 4 electrons remaining
7. LEWIS DOT DIAGRAMS Draw: H2O H O H 0 electrons remaining Oxygen has a full octet, sothere is no need to form double or triple bonds
in this case
8. LEWIS DOT DIAGRAMSDraw: SO2 O S O
9. LEWIS DOT DIAGRAMSDraw: SO2 O S O 6 + 6 + 6 = 18 electrons
10. LEWIS DOT DIAGRAMSDraw: SO2 O S O 14 electrons remaining6+6+6=18 electrons
11. LEWIS DOT DIAGRAMSDraw: SO2 O S O 2 electrons 14 electrons remaining
12. LEWIS DOT DIAGRAMSDraw: SO2 O S O 2 electrons 0 electrons
13. LEWIS DOT DIAGRAMSDraw: SO2 O S O 0 electronsNow all atoms have a full octet
14. LEWIS DOT DIAGRAMSDraw: SO2 O S O The double bond can be on either side O S O These are called resonance structures.SO2 has
2 resonance structures
15. LEWIS DOT DIAGRAMSDraw: PO43- O O P O O
16. LEWIS DOT DIAGRAMSDraw: PO43- 6 6 O 6 O P O 5 O 66+6+6+6+5=29 +3 = 32 electrons
17. LEWIS DOT DIAGRAMS Draw: PO43- O O P O O 24 electrons6+6+6+6+5+3=32 electrons
18. LEWIS DOT DIAGRAMSDraw: PO43- O O P O O 0 electrons24 electrons remaining
19. LEWIS DOT DIAGRAMSDraw: PO43- O O P O O 0 electrons remainingThere are no electrons left to place
20. LEWIS DOT DIAGRAMS Draw: PO43- O O P O ONow all atoms have a full octet
21. LEWIS DOT DIAGRAMSDraw: PO43- 3- O O P O O Just one more thing

Lewis dot structures

1.

2.

3.

4.
5.
6.
7.

8.

1. Lewis Dot Structures http://www.ipfw.edu/chem/115friedel/chap9.ppt <ul><li>Lewis suggested a way to keep track of valence (outer shell)
electrons. </li></ul><ul><li>A Lewis dot structure for an atom consists of the symbol for the element and one dot for each valence electron. It
is used for the s and p block elements, the representative elements. </li></ul>
2. LEWIS DOT SYMBOLS FOR THE ELEMENTS <ul><li>A Lewis Structure (an electron dot formula) is a representation of covalent bonding
using Lewis dot symbols in which shared electron pairs are shown as a pair of dots or a line between two atoms . The lone pairs of electrons
are shown as pairs of dots on individual atoms. </li></ul>
3. Octet Rule <ul><li>When atoms form covalent bonds there is a tendency to share enough electrons so that each atom is surrounded by 8
electrons. </li></ul><ul><li>Exceptions - fewer than 8 are required for H, He, Li, Be, and B . </li></ul><ul><li>H needs 2 e ; He needs 2 e;
</li></ul><ul><li>Li is usually ionic where it loses 1 electron, </li></ul><ul><li>Be (when covalent) shares 4, </li></ul><ul><li>and B can
sometimes share 6. </li></ul>
4. Covalent Bonding <ul><li>You try it! </li></ul><ul><li>H 2 </li></ul><ul><li>F 2 </li></ul><ul><li>HCl </li></ul><ul><li>H 2 O
</li></ul><ul><li>CH 4 </li></ul><ul><li>CCl 4 </li></ul><ul><li>NH 3 </li></ul><ul><li>CH 3 CH 2 OH </li></ul>
5. Multiple Bonds <ul><li>More than one pair of electrons can be shared between two atoms. </li></ul><ul><li>O 2 </li></ul><ul><li>N 2
</li></ul><ul><li>CO 2 </li></ul><ul><li>CO </li></ul>
6. <ul><li>Sometimes, bonds are formed in which both electrons of the shared pair come from the same atom. </li></ul><ul><li> NH 4 +NH 3
+ H + </li></ul>Coordinate Covalent Bonds
7. LEWIS STRUCTURE BUILDING - GAME RULES <ul><li>Decide which atoms are bonded to which atoms. </li></ul><ul><li>Count all
valence electrons (adding or subtracting if molecule is charged). </li></ul><ul><li>Place 2 electrons in each bond. </li></ul><ul><li>Complete
the octets of the atoms attached to the central atom by adding electrons in pairs. </li></ul><ul><li>If the central atom does not have an octet,
form double bonds or triple bonds. </li></ul>
8. <ul><li>The atom at the centre must be able to make multiple bonds, and there is often only one of that atom in the formula.
</li></ul><ul><li>A symmetrical arrangement of atoms is more often correct. </li></ul><ul><li>It is very unusual for oxygen atoms to bond to

9.

10.

11.

12.

13.

each other (except in peroxides, which are very unstable). </li></ul><ul><li>Carbon atoms make 4 bonds, </li></ul><ul><li>Oxygen atoms
make 2 bonds, </li></ul><ul><li>Hydrogen and Fluorine atoms make 1 bond. </li></ul>1. Which atoms are bonded
9. <ul><li>e.g. H and F can only form ONE bond, therefore they cannot be in the middle of a molecule, they must be on the end of a molecule.
</li></ul><ul><li>H 2 O must be </li></ul><ul><li>BF 3 must be </li></ul><ul><li>try: </li></ul><ul><li>SO 3 </li></ul><ul><li>HNO 3
</li></ul><ul><li>(hint: this is an </li></ul><ul><li>oxyacid, and the H </li></ul><ul><li>comes off easily) </li></ul><ul><li>not O-H-H
</li></ul><ul><li>not B-F-F-F </li></ul>
10. Count All Valence Electrons <ul><li>Use the Periodic Table to determine the total number of electrons available for distribution among the
atoms in a molecule. </li></ul><ul><li>Example SO 3 </li></ul><ul><li>S is in Group 6 = 6 e O is in Group 6 X 3 = 18 e total: 6e + 18e = 24
e </li></ul><ul><li>Example - SO 4 2- </li></ul><ul><li>S is in Group 6 = 6 e O is in Group 6 X 4 = 24 e </li></ul><ul><li>add 2 e for the 2charge = 2e total: 6e + 24e + 2e= 32 e </li></ul>
11. Filling in the Electrons <ul><li>Place 2 electrons in each bond. </li></ul><ul><li>Complete the octets of the atoms attached to the central
atom by adding electrons in pairs. </li></ul><ul><li>Place any remaining electrons on the central atom in pairs. </li></ul><ul><li>If the central
atom does not have an octet, form double bonds or triple bonds. </li></ul><ul><li>try: HClO 3 SO 3 PO 4 3- COCl 2 </li></ul>
12. Resonance <ul><li>Sometimes, two very similar structures are equally likely: </li></ul><ul><li>e.g. for SO 2 </li></ul><ul><li>this is a
possiblility </li></ul><ul><li>but so is this </li></ul><ul><li>The double bond is delocalized, or is distributed between the two possible
structures. </li></ul><ul><li>These two lewis structures are called resonance structures. These molecules are more stable than a plain
double bond. </li></ul>
13. <ul><li>We find that CO 3 2- has 3 resonance structures: </li></ul><ul><li>Benzene C 6 H 6 has 3 delocalized double bonds, which
makes it very stable. It is often drawn as a hexagon with a circle inside, to indicate the special nature of the double bonds. </li></ul>