2-4 Standard Valence-Bond Theory for N2O and F2O2
23
bond for CH 3 N=O. No matter how we vary the contributions of structures (1), (2)
and (3) to the resonance, it is not possible to account for both bond-length observations simultaneously. However, often a “resonance shortening correction” is
invoked, which may increase the theoretical N-N and N-O bond-numbers to
values closer to 3 and 2 respectively. This type of correction may have a valid
basis in theory (see for example Ref. 13a), but we contend that better valencebond structures may be constructed, which can also rationalize the observations.
We shall postpone consideration of this matter until Section 2-5(b).
2-4(c) Standard Valence-Bond Theory and 2 2
F O
We shall use F 2 O 2 to provide a second example that illustrates some unsatisfactory
features of the standard Lewis descriptions for many electron-excess systems.
In Section 2-3(b) we have indicated that the O-O bond-length
9 of 1.217 Å for
F 2 O 2 is similar to the double-bond length of 1.207 Å for the O 2 ground state. The
O-F bond-lengths of 1.575 Å are appreciably longer than the 1.42 Å for the O-F
single-bonds of F 2 O. A set of nine valence-bond structures that conform to the
Lewis-Langmuir octet rule is displayed in Figure 2-10.
If rule (a) of Section 2-4(a) is invoked, we would select the standard Lewis
structures (1)-(4) to be the important valence-bond structures for the F 2 O 2 ground
state. If we then invoke rule (b) (the electroneutrality principle), we would deduce
the order of importance for these Lewis structures to be (1) > (2) = (3) > (4). If we
assume that structure (1) alone represents the electronic structure of F 2 O 2 , then we
cannot account for the observed bond-lengths of this molecule. (By contrast it may
be noted that for hydrogen peroxide H 2 O 2 , the valence-bond structure which is the
same as structure (1), with Η replacing F, is in accord with the observations that
O-O and O-H bond-lengths of 1.464 Å and 0.965 Å are essentially those of O-O
and O-H single bonds
16
.) For F 2 O 2 , it is necessary to assume that structures (2) and
(3) at least make substantial contributions to the ground-state resonance scheme. A
justification for this assumption is also provided by electronegativity considerations, which allow fluorine atoms to acquire formal negative charges (as they do
in structures (2) and (3)). If as is sometimes done, we assume that resonance between structures (2) and (3) alone can be used to describe the electronic structure
of F 2 O 2 , then we can account for the similarity of the O-O bond-length to that of
an O-O double-bond, and also for the lengthening of the O-F bonds relative to
those of O-F single-bonds. However, such a description ignores the contribution
of Lewis structure (1) to the resonance. The absence of formal charges for this
structure would suggest that it is also important. If we include structure (1),
together with structures (2) and (3), then the O-O bond-length would be predicted
to be rather longer than a double bond. To restore agreement between theory and
experiment, it is then necessary to assume that valence-bond structure (4) (with an
O-O triple bond) also contributes to the resonance, and that its contribution to the
resonance scheme is equal to that of structure (1). The very different sets of formal
charges (and bond-arrangements) would not permit structures (1) and (4) to have
23
bond for CH 3 N=O. No matter how we vary the contributions of structures (1), (2)
and (3) to the resonance, it is not possible to account for both bond-length observations simultaneously. However, often a “resonance shortening correction” is
invoked, which may increase the theoretical N-N and N-O bond-numbers to
values closer to 3 and 2 respectively. This type of correction may have a valid
basis in theory (see for example Ref. 13a), but we contend that better valencebond structures may be constructed, which can also rationalize the observations.
We shall postpone consideration of this matter until Section 2-5(b).
2-4(c) Standard Valence-Bond Theory and 2 2
F O
We shall use F 2 O 2 to provide a second example that illustrates some unsatisfactory
features of the standard Lewis descriptions for many electron-excess systems.
In Section 2-3(b) we have indicated that the O-O bond-length
9 of 1.217 Å for
F 2 O 2 is similar to the double-bond length of 1.207 Å for the O 2 ground state. The
O-F bond-lengths of 1.575 Å are appreciably longer than the 1.42 Å for the O-F
single-bonds of F 2 O. A set of nine valence-bond structures that conform to the
Lewis-Langmuir octet rule is displayed in Figure 2-10.
If rule (a) of Section 2-4(a) is invoked, we would select the standard Lewis
structures (1)-(4) to be the important valence-bond structures for the F 2 O 2 ground
state. If we then invoke rule (b) (the electroneutrality principle), we would deduce
the order of importance for these Lewis structures to be (1) > (2) = (3) > (4). If we
assume that structure (1) alone represents the electronic structure of F 2 O 2 , then we
cannot account for the observed bond-lengths of this molecule. (By contrast it may
be noted that for hydrogen peroxide H 2 O 2 , the valence-bond structure which is the
same as structure (1), with Η replacing F, is in accord with the observations that
O-O and O-H bond-lengths of 1.464 Å and 0.965 Å are essentially those of O-O
and O-H single bonds
16
.) For F 2 O 2 , it is necessary to assume that structures (2) and
(3) at least make substantial contributions to the ground-state resonance scheme. A
justification for this assumption is also provided by electronegativity considerations, which allow fluorine atoms to acquire formal negative charges (as they do
in structures (2) and (3)). If as is sometimes done, we assume that resonance between structures (2) and (3) alone can be used to describe the electronic structure
of F 2 O 2 , then we can account for the similarity of the O-O bond-length to that of
an O-O double-bond, and also for the lengthening of the O-F bonds relative to
those of O-F single-bonds. However, such a description ignores the contribution
of Lewis structure (1) to the resonance. The absence of formal charges for this
structure would suggest that it is also important. If we include structure (1),
together with structures (2) and (3), then the O-O bond-length would be predicted
to be rather longer than a double bond. To restore agreement between theory and
experiment, it is then necessary to assume that valence-bond structure (4) (with an
O-O triple bond) also contributes to the resonance, and that its contribution to the
resonance scheme is equal to that of structure (1). The very different sets of formal
charges (and bond-arrangements) would not permit structures (1) and (4) to have