Chapter 5

Answers to Problems

5.1 P(E)F2

O E C3 C3 C3 C3 C32 C3 2 C3 2 C3 2 C2 C2 C2
label aa bb cc dd aa bb cc dd 12 34 56
RjF2 F2 F6 F4 F5 F3 F4 F5 F3 F6 F2 F1 F1
E 2 -1 -1 -1 -1 -1 -1 -1 -1 2 2 2
P i R Rj F 2 2F2 -F6 -F4 -F5 -F3 -F4 -F5 -F3 -F6 2F2 2F1 2F1

P(E)F2 % 4F1 + 4F2 – 2F3 – 2F4 – 2F5 – 2F6 % 2F1 + 2F2 – F3 – F4 – F5 – F6

This is the same result as P(E)F1, and therefore the companion function can be found in the
same way as shown in the text, either by another projection or a transformation by a
symmetry operation.

P(E)F4

O E C3 C3 C3 C3 C32 C3 2 C3 2 C3 2 C2 C2 C2
label aa bb cc dd aa bb cc dd 12 34 56
RjF4 F4 F6 F5 F1 F5 F2 F2 F6 F1 F3 F4 F3
E 2 -1 -1 -1 -1 -1 -1 -1 -1 2 2 2
P i R Rj F 4 2F4 -F6 -F5 -F1 -F5 -F2 -F2 -F6 -F1 2F3 2F4 2F3

P(E)F4 % –2F1 – 2F2 + 4F3 + 4F4 – 2F5 – 2F6 % –F1 – F2 + 2F3 + 2F4 – F5 – F6

This is the same result as P(E)F3, and the companion function can be obtained by the
identical addition shown in the text (p. 145).

-64-
 P(E)F5

O E C3 C3 C3 C3 C32 C3 2 C3 2 C3 2 C2 C2 C2
label aa bb cc dd aa bb cc dd 12 34 56
RjF5 F5 F3 F2 F3 F1 F1 F4 F2 F4 F6 F6 F5
E 2 -1 -1 -1 -1 -1 -1 -1 -1 2 2 2
P i R Rj F 5 2F5 -F3 -F2 -F3 -F1 -F1 -F4 -F2 -F4 2F6 2F6 2F5

P(E)F5 % –2F1 – 2F2 – 2F3 – 2F4 + 4F5 + 4F6 % –F1 – F2 – F3 – F4 + 2F5 + 2F6

To get the partner add P(E)F1 + 2P(E)F5:

P(E)F1 % 2F1 + 2F2 – F3 – F4 – F5 – F6

2P(E)F5 % –2F1 – 2F2 – 2F3 – 2F4 + 4F5 + 4F6
------------------------------------------------------
– 3F3 – 3F4 + 3F5 + 3F6 % –F3 – F4 + F5 + F6

This is the negative of the same function as previously obtained in the text.

5.2 Effect of C4(56) on G2(E) % P(E)F1

C4(56) transforms the initial functions as follows:

Before F1 F2 F3 F4 F5 F6
After F4 F3 F1 F2 F5 F6

2F1 + 2F2 – F3 – F4 – F5 – F6 6 2F4 + 2F3 – F1 – F2 – F5 – F6

= F1 – F2 + 2@3 + 2F4 – F5 – F6 % P(E)F3 = P(E)F4

The partner function can be generated from this as shown in the text.

Effect of C4(34) on G2(E) % P(E)F1

C4(34) transforms the initial functions as follows:

Before F1 F2 F3 F4 F5 F6
After F 5 F 6 F3 F 4 F 2 F1
2F1 + 2F2 – F3 – F4 – F5 – F6 6 2F5 + 2F6 – F3 – F4 – F2 – F1
= –F1 – F2 – @3 – F4 + 2F5 + 2F6 % P(E)F5 = P(E)F6

-65-
 This can be used to obtain the partner by the addition shown in the answer to 5.1 given
above.

Effect of C4(12) on G2(E) % P(E)F1

C4(12) transforms the initial functions as follows:

Before F1 F2 F3 F4 F5 F6
After F1 F2 F6 F5 F3 F4

2F1 + 2F2 – F3 – F4 – F5 – F6 6 2F1 + 2F2 – F6 – F5 – F3 – F4

2F1 + 2F2 – F3 – F4 – F5 – F6 % P(E)F1 = P(E)F2

This just gives back the initial function.

5.3 IE4E5dJ = I(F1 – F2)(F3 – F4)dJ = 0 Clearly, the other two combinations E4E6 and E5E6
will give 0.

IE4E1dJ = I(F1 – F2)(F1 + F2 + F3 + F4 + F5 + F6)dJ = 0 Other combinations with E1 will

have the same result.

IE4E2dJ = I(F1 – F2)(2F1 + 2F2 – F3 – F4 – F5 – F6)dJ = 2 – 2 =0

IE5E2dJ = I(F3 – F4)(2F1 + 2F2 – F3 – F4 – F5 – F6)dJ = –1 + 1 =0
IE6E2dJ = I(F5 – F6)(2F1 + 2F2 – F3 – F4 – F5 – F6)dJ = –1 + 1 =0

IE4E3dJ = I(F1 – F2)(F3 + F4 – F5 – F6)dJ = 0

IE5E3dJ = I(F3 – F4)(F3 + F4 – F5 – F6)dJ = 1 – 1 =0
IE6E3dJ = I(F5 – F6)(F3 + F4 – F5 – F6)dJ = –1 + 1 = 0

-66-
5.4 The terms of the full P(T2)sA operator are shown below. The 8C3 operations have zero
characters and can be skipped.

Td E ... C2 C2 C2 S4 S4 S4 S43 S43 S43

Label x y z x y z x y z
RjsA sA sC sD sB sB sC sC sD sB sD
T2 3 -1 -1 -1 -1 -1 -1 -1 -1 -1
PiRjsA 3sA -sC -sD -sB -sB -sC -sC -sD -sB -sD

Fd Fd Fd Fd Fd Fd
bd ac bc ad ab cd
sC sA sD sA sA sB
1 1 1 1 1 1
sC sA sD sA sA sB

Gathering the terms:

P(T2)sA % 6sA – 2sB – 2sC – 2sD % 3sA – sB – sC – sD

This is the result obtained from the P(T)sA operator in the subgroup T.

5.5 (a)

D2h E C2(z) C2(y) C2(x) i F(xy) F(xz) F(yz)

'H 4 0 0 0 0 0 0 4

'H = Ag + B3g + B1u + B2u

-67-
By inspection
P(Ag) % sA + sB + sC + sD
M1(Ag) = ½(sA + sB + sC + sD)

Projection for B3g:

D2h E C2(z) C2(y) C2(x) i F(xy) F(xz) F(yz)

RjsA sA sD sB sC sC sB sD sA
B3g 1 -1 -1 1 1 -1 -1 1
P i R j sA sA -sD -sB sC sC -sB -sD sA

P(B3g)sA % sA – sB + sC – sD Y M2(B3g) = ½(sA – sB + sC – sD)

This is clearly orthogonal to M1(Ag).

Projection for B1u:

D2h E C2(z) C2(y) C2(x) i F(xy) F(xz) F(yz)

RjsA sA sD sB sC sC sB sD sA
B1u 1 1 -1 -1 -1 -1 1 1
P i R j sA sA sD -sB -sC -sC -sB sD sA

P(B1u)sA % sA – sB – sC + sD Y M3(B1u) = ½(sA – sB – sC + sD)

This is clearly orthogonal to both M1(Ag) and M2(B3g) .

Projection for B2u:

D2h E C2(z) C2(y) C2(x) i F(xy) F(xz) F(yz)

RjsA sA sD sB sC sC sB sD sA
B2u 1 -1 1 -1 -1 1 -1 1
P i R j sA sA -sD sB -sC -sC sB -sD sA

P(B2u)sA % sA + sB – sC – sD Y M3(B2u) = ½(sA + sB – sC – sD)

This is clearly orthogonal to both M1(Ag), M2(B3g) , and M3(B1u).

-68-
5.5 (b)

D2d E 2S4 C2 2C2' 2Fd

'H 4 0 0 0 2

'H = A1 + B2 + E

If we do the work in the subgroup D2, we will not only reduce the number of terms, but more
importantly we will lift the E degeneracy, allowing separate projections for the two
degenerate SALCs. In the descent from D2d to D2, A1 6 A, B2 6 B1, and E 6 B2 + B3.

The totally symmetric SALC is clearly M1(A) = ½(sA + sB + sC + sD). In D2, the P(B1)sA
projection is found as follows:

D2 E C2(z) C2(y) C2(x)

RjsA sA sD sB sC
B1 1 1 -1 -1
P i Rj s A sA sD -sB -sC

M2(B1) = ½(sA – sB – sC + sD) Y M2(B2) = ½(sA – sB – sC + sD) in D2d

The first of the degenerate functions is found as the projection P(B2)sA in D2:

D2 E C2(z) C2(y) C2(x)

RjsA sA sD sB sC
B2 1 -1 1 -1
P i Rj s A sA -sD sB -sC

M3(B2) = ½(sA + sB – sC – sD) Y M3(Ea) = ½(sA + sB – sC – sD) in D2d

-69-
 The second of the degenerate functions is found as the projection P(B3)sA in D2:

D2 E C2(z) C2(y) C2(x)

RjsA sA sD sB sC
B3 1 -1 -1 1
P i Rj s A sA -sD -sB sC

M4(B3) = ½(sA – sB + sC – sD) Y M4(Eb) = ½(sA – sB + sC – sD) in D2d

The orthogonality of all four functions is easily shown.

If the doubly degenerate projection is carried out in D2d on sA, we obtain an initial projection
P(E) % sA – sD, because only the operations E and C2 have nonzero characters. The
normalized function M3'(Ea) = 1//2 (sA – sD) is orthogonal to our previously obtained M1 and
M2 expressions, which are the same functions generated by P(A1)sA and P(B2)sA in D2d. The
companion can be generated by applying the operation S43 (for example) to M3'(Ea) to obtain
M4'(Eb) = 1//2 (sB – sC). Both M3'(Ea) and M4'(Eb) are orthogonal to each other and to M 1(A1)
and M2(B2). They are related to the previously obtained functions as M3(Ea) % M3'(Ea) +
M4'(Eb) and M4(Eb) % M3'(Ea) – M4'(Eb). The distinction is a matter of definition, with neither
pair of degenerate SALCs being "right" and the other "wrong".

5.6 For all of these planar cyclic systems, the projections can be read off from the characters for
the appropriate representations in the Cn subgroup of the molecule's true Dnh point group (n
= 3, 4, 5).

(a) C3H3

' = A' + E' in D3h Y ' = A + E in C3

The A projection and resulting normalized wave function are

P(A)p1 % p1 + p2 + p3 Y A1 1//3 (p1 + p2 + p3)

From the complex conjugate pair of E representations we obtain the following two
projections:

P(Ea)p1 % p1 + ,p2 + ,*p3

-70-
 P(Ea)p1 % p1 + ,*p2 + ,p3

We can add and subtract these two to obtain real functions, realizing that

, + ,* = 2 cos 2B/3 = 2(–½) = –1

, – ,* = 2i sin 2B/3 and ,* – , = –2i sin 2B/3

The imaginary constant 2i sin 2B/3 can be factored out of the subtractive combination prior
to normalization. Thus, we obtain the following real SALCs:

P(Ea)p1 + P(Eb)p1 % 2p1 – p2 – p3

P(Ea)p1 – P(Eb)p1 % p2 – p3

These give the normalized functions

A2 = 1//6 (2p1 – p2 – p3)

A3 = 1/2 (p2 – p3)

(b) C4H4

' = Eg + A2u + B2u in D4h Y 'z = A + B + E in C4

By inspection of the C4 character table, the A and B functions are

A1(A) = 1/2 (p1 + p2 + p3 + p4)

A4(B) = 1/2 (p1 – p2 + p3 – p4)

The two E projections are

P(Ea)p1 % (p1 + ip2 – p3 – ip4)

P(Eb)p1 % (p1 – ip2 – p3 + ip4)

By addition and subtraction, these give the two normalized functions

A2(E") = 1//2 (p1 – p3)

A3(E$) = 1//2 (p2 – p4)

-71-
(c) C5H5

' = A2" + E1" + E2" in D5h Y ' = A + E1 + E2 in C5

By inspection, A1(A) = 1//5 (p1 + p2 + p3 + p4 + p5). From the C5 character table we can
read off the two E1 projections and the two E2 projections. The following relationships will
be useful in transforming these into real functions:

Let T = 2B/5 = 72o and 2T = 4B/5 = 144o

, + ,* = 2 cos T = 2 (0.3090)
, – ,* = –2i sin T = –2i (0.9511)

,2 + ,2* = 2 cos 2T = 2 (–0.8090)

,2 – ,2* = –2i sin 2T = –2i (0.5878)

The two E1 projections are

Pp1(E1a) % (p1 + ,p2 + ,2p3 + ,2*p4 + ,*p5)

Pp1(E1b) % (p1 + ,*p2 + ,2*p3 + ,2p4 + ,p5)

Adding:

Pp1(E1a) + Pp1(E1b) % 2p1 + 2p2 cos T + 2p3 cos 2T + 2p4 cos 2T + 2p5 cos T

% p1 + p2 cos T + p3 cos 2T + p4 cos 2T + p5 cos T

% p1 + 0.3090p2 – 0.8090p3 – 0.8090p4 + 0.3090p5 % A2(E1)

Normalizing:

N2IA22dJ = 1 + 0.0955 + 0.6545 + 0.6545 + 0.0955 = 2.5

N = (1/2.5)½ = (2/5)½

Normalized function:

A2(E1) = (2/5)½{p1 + p2 cos T + p3 cos 2T + p4 cos 2T + p5 cos T}

Subtracting:

-72-
 Pp1(E1a) – Pp1(E1b) % –2ip2 sin T – 2ip3 sin 2T + 2ip4 sin 2T + 2ip5 sin T

% p2 sin T + p3 sin 2T – p4 sin 2T – p5 sin T

% 0.9511p2 + 0.5878p3 – 0.5878p4 – 0.9511p5 % A3(E1)

Normalizing:

N2IA32dJ = 0.9045 + 0.3455 + 0.3455 + 0.9045 = 2.5

N = (1/2.5)½ = (2/5)½

Normalized function:

A3(E1) = (2/5)½{p2 sin T + p3 sin 2T – p4 sin 2T – p5 sin T}

The two E2 projections are

Pp1(E2a) % (p1 + ,p2 + ,2p3 + ,2*p4 + ,*p5)

Pp1(E2b) % (p1 + ,*p2 + ,2*p3 + ,2p4 + ,p5)

Adding:

Pp1(E2a) + Pp2(E2b)% 2p1 + 2p2 cos 2T + 2p3 cos T + 2p4 cos T + 2p5 cos 2T

% p1 + p2 cos 2T + p3 cos T + p4 cos T + p5 cos 2T

% p1 – 0.8090p2 + 0.3090p3 + 0.3090p4 – 0.8090p5 % A4(E2)

Normalizing:

N2IA42dJ = 1 + 0.6545 + 0.0955 + 0.0955 + 0.6545 = 2.5

N = (1/2.5)½ = (2/5)½

Normalized function:

A4(E2) = (2/5)½{p1 + p2 cos 2T + p3 cos T + p4 cos T + p5 cos 2T}

Subtracting:

Pp1(E2a) – Pp1(E2b) % –2ip2 sin 2T – 2ip3 sin T + 2ip4 sin T + 2ip5 sin 2T

-73-
 % p2 sin 2T + p3 sin T – p4 sin T – p5 sin 2T

% 0.5878p2 + 0.9511p3 – 0.9511p4 – 0.5878p5 % A5(E2)

Normalizing:

N2IA52dJ = 0.34555 + 0.9045 + 0.9045 + 0.3455 = 2.5

N = (1/2.5)½ = (2/5)½

Normalized function:

A5(E2) = (2/5)½{p2 sin 2T + p3 sin T – p4 sin T – p5 sin 2T}

-74-
5.7 The three basis sets, their reducible representations in D3h, and the species of which they are
composed are shown below.

D3h E 2C3 3C2 Fh 2S3 3Fv

'F 3 0 1 3 0 1

'F = A1' + E'

D3h E 2C3 3C2 Fh 2S3 3Fv

'z 3 0 -1 -3 0 1

'z = A2" + E'

D3h E 2C3 3C2 Fh 2S3 3Fv

'2 3 0 -1 -3 0 1

'2 = A2' + E'

If we carry out the projections in the subgroup C3, all three reducible representations will be
' = A + E. Therefore, once we find the forms of the SALCs for one case, the other two
cases will have the same forms. In addition, working in C3 lifts the double degeneracy,
allowing us to read off the pair of projections from the characters for each species. Taking
the sigma case as the model, the A projection and resulting normalized wave function are

-75-
 P(A)N1 % N1 + N2 + N3 Y M1(F) = 1//3 (N1 + N2 + N3)

From the complex conjugate pair of E representations we obtain the following two
projections:

P(Ea)N1 % N1 + ,N2 + ,*N3

P(Ea)N1 % N1 + ,*N2 + ,N3

We can add and subtract these two to obtain real functions, realizing that

, + ,* = 2 cos 2B/3 = 2(–½) = –1

, – ,* = 2i sin 2B/3 and ,* – , = –2i sin 2B/3

The imaginary constant 2i sin 2B/3 can be factored out of the subtractive combination prior
to normalization. Thus, we obtain the following real SALCs:

P(Ea)N1 + P(Eb)N1 % 2N1 – N2 – N3

P(Ea)N1 – P(Eb)N1 % N2 – N3

These give the normalized functions

M2(F) = 1//6 (2N1 – N2 – N3)

M3(F) = 1/2 (N2 – N3)

It follows that the other two sets have SALCs with the same forms:

M4(z) = 1//3 (N4 + N5 + N6)

M5(z) = 1//6 (2N4 – N5 – N6)
M6(z) = 1/2 (N5 – N6)

M7(2) = 1//3 (N7 + N8 + N9)

M8(2) = 1//6 (2N7 – N8 – N9)
M7(2) = 1/2 (N8 – N9)

-76-
5.8 The four basis sets, their reducible representations in D4h, and the species of which they are
composed are shown below.

D4h E 2C4 C2 2C2' 2C2" i 2S4 Fh 2Fv 2Fd

'F 4 0 0 2 0 0 0 4 2 0

'F = A1g + B1g + Eu

D4h E 2C4 C2 2C2' 2C2" i 2S4 Fh 2Fv 2Fd

'z 4 0 0 -2 0 0 0 -4 2 0

'z = A2u + B2u + Eg

D4h E 2C4 C2 2C2' 2C2" i 2S4 Fh 2Fv 2Fd

'2 4 0 0 -2 0 0 0 4 -2 0

'2 = A2g + B2g + Eu

If we carry out the projections in the subgroup C4, all three reducible representations will be
' = A + B + E. Therefore, once we find the forms of the SALCs for one case, the other two
cases will have the same forms. In addition, working in C4 lifts the double degeneracy,
allowing us to read off the pair of projections from the characters for each species. Taking
the sigma case as a model, the A projection and resulting normalized wave function are

-77-
 P(A)N1 % N1 + N2 + N3 + N4 Y M1(A) = 1/2 (N1 + N2 + N3 + N4)

Likewise, for the B projection and resulting normalized wave function we have

P(B)N1 % N1 – N2 + N3 – N4 Y M2(B) = 1/2 (N1 – N2 + N3 – N4)

The two E projections are

P(Ea)N1 % N1 + iN2 – N3 + iN4

P(Eb)N1 % N1 – iN2 – N3 – iN4

By adding and subtracting we obtain the following two normalized functions

M3(E) = 1//2 (N1 – N3) and M4(E) = 1//2 (N2 – N4)

5.9 (a) For trigonal planar sp2 hybrids, ' = A1' + E in D3h. Use the rotational subgroup C3 to
read off the projections for the s = A and (px, py) = E AOs as SALCs of the hybrids. From
the A projection we obtain the normalized function

s = 1//3 (Q1 + Q2 Q3)

For the two complex conjugate E projections we have

P(Ea) % Q1 + ,Q2 + ,*Q3

P(Eb) % Q1 + ,*Q2 + ,Q3

Add and subtract these two imaginary functions to obtain real functions, realizing that

, + ,* = 2 cos 2B/3 = 2(–½) = –1

, – ,* = 2i cos 2B/3 and ,* – , = –2i cos 2B/3

The real functions, then, are

P(Ea) + P(Eb) % 2Q1 – Q2 – Q3

P(Ea) – P(Eb) % Q2 – Q3

The normalized expressions for the AOs are

px = 1//6 (2Q1 – Q2 – Q3)

py = 1/2 (Q2 – Q3)

In matrix form the three AO expressions become

-78-
Taking the transpose of the 3 x 3 matrix A to obtain the B matrix, we obtain an expression
for the hybrids as functions of the AOs:

The three individual sp2 functions are

Q1 = 1//3 s + 2//6 px
Q2 = 1//3 s – 1//6 px + 1/2 py
Q3 = 1//3 s – 1//6 px – 1/2 py

(b) For square planar dsp2 hybrids, ' = A1g + B1g + Eu. Use the rotational subgroup C4 to
read off the projections for the s = A, (px, py) = E, and dx2-y2 = B AOs as SALCs of the
hybrids. From the A and B projections we obtain the normalized functions

s = 1/2 (Q1 +Q2 + Q3 + Q4)

dx2-y2 = 1/2 (Q1 – Q2 + Q3 – Q4)

For the E projections we obtain

P(Ea) % Q1 + iQ2 – Q3 – iQ4

P(Ea) % Q1 – iQ2 – Q3 + iQ4

Adding and subtracting these give the following two normalized functions:

px = 1//2 (Q1 – Q3)

py = 1//2 (Q2 – Q4)

Gathering all four equations into matrix form gives

-79-
Using the transpose of the 4 x 4 A matrix to obtain the B matrix, we obtain

The four dsp2 function are

Q1 = 1/2 s + 1//2 px + 1/2 dx2-y2

Q2 = 1/2 s + 1//2 py – 1/2 dx2-y2
Q3 = 1/2 s – 1//2 px + 1/2 dx2-y2
Q4 = 1/2 s – 1//2 py – 1/2 dx2-y2

(c) For octahedral d2sp3 hybrids, ' = A1g + E + T1u. We can use the previously obtained
expressions for F-SALCs developed in section 5.1 to write equations for the AOs as
functions of the hybrids.

s(A) = 1//6 (Q1 + Q2 + Q3 + Q4 + Q5 + Q6)

px(T) = 1//2 (Q5 – Q6)
py(T) = 1//2 (Q3 – Q4)
pz(T) = 1//2 (Q1 – Q2)
dz2(E) = 1//12 (2Q1 + 2Q2 – Q3 – Q4 – Q5 – Q6)
dx2-y2(E) = 1/2 (Q3 + Q4 – Q5 – Q6)

In matrix form these are

-80-
Using the transpose of the 6 x 6 A matrix to obtain the B matrix, we obtain

The six hybrid functions are

Q1 = 1//6 s + 1//2 pz + 1//3 dz2

Q2 = 1//6 s – 1//2 pz + 1//3 dz2
Q3 = 1//6 s + 1//2 py – 1//12 dz2 + 1/2 dx2-y2
Q4 = 1//6 s – 1//2 py – 1//12 dz2 + 1/2 dx2-y2
Q5 = 1//6 s + 1//2 px – 1//12 dz2 – 1/2 dx2-y2
Q6 = 1//6 s – 1//2 px – 1//12 dz2 – 1/2 dx2-y2

-81-
5.10 The matrix equation for the AOs as functions of the hybrids is

Taking the transpose of the A matrix to obtain the B matrix gives the following matrix
equation for the hybrids as functions of the AOs:

The resulting individual functions are

Q1 = 1//5 s + 2//6 px – 2//30 dz2

Q2 = 1//5 s – 1//6 px + 1//2 py – 2//30 dz2
Q3 = 1//5 s – 1//6 px – 1//2 py – 2//30 dz2
Q4 = 1//5 s + 1//2 pz + 3//30 dz2
Q5 = 1//5 s – 1//2 pz + 3//30 dz2

These hybrids might be used to describe bonding in a transition metal tbp complex in which
the axial and equatorial positions made equal contributions to the bonding. The
involvement of the dz2 orbital in the hybrid set makes this mode of hybridization, in general,
inappropriate for describing the bonding in tbp structures in which the central atom is a p-
block element.

-82-
5.11 The representation for the F-SALCs is based on the following vector model:

The reducible representation and its component species are

C4v E 2C4 C2 2Fv 2Fd

' 5 1 1 3 1

' = 2A1 + B1 + E

If we apply projection operators to any reference function, we will project LCAOs that are
expressions only of the members of the same set, basal or axial. The basal functions are F1,
F2, F3, F4, and the axial function is simply F5. The axial function F5 clearly has A1
symmetry, so the symmetries of the SALCs that can be formed from the basal set are the
remaining species in '. The symmetries of the two sets, then, are

'basal = A1 + B1 + E and 'axial = A1

We can carry out the basal projections in the subgroup C4, from which it is immediately
apparent that nondegenerate functions are

M1(A1) = 1/2 (F1 + F2 + F3 + F4)

M2 (B1) = 1/2 (F1 – F2 + F3 – F4)

The two E projections take advantage of the complex conjugate representations in C4:

P(Ea)F1 % F1 + iF2 – F3 – iF4

P(Eb)F1 % F1 – iF2 – F3 + iF4

By addition and subtraction, we obtain the following two real functions:

M3(E) = 1//2 (F1 – F3)

M4(E) = 1//2 (F2 – F4)

For the axial position, the SALC is simply F5:

M5(A1) = F5

-83-
The AOs on a central atom that would match with these SALCs are

A1 = s, pz, dz2 B1 = dx2-y2 E = (px, py)

We will ignore the dz2 initially, because we are assuming dx2-y2sp3 hybrids on the central
atom. If we accept the SALCs we obtained from the projection operator approach, we
would make the following matches between SALCs and central-atom AOs:

Looking at the matches between AOs and SALCs, it appears that the B1 and E SALCs are
correctly formulated, despite the basal-axial segregation inherent in our approach.
However, the two A1 SALCs are artificially exclusive with regard to the two kinds of
positions. The M1(A1) SALC would be improved by including the F5 function, which would
in no way change its A1 symmetry. This results in the following match:

If the axial and equatorial positions made equal contributions to the bonding (not a typical
result) the redefined SALC would be

M'1(A1) = 1//5 (F1 + F2 + F3 + F4 +F5)

More generally, this would be

M'1(A1) = cbasal(F1 + F2 + F3 + F4 ) + caxialF5

-84-
 If the angle between axial and basal positions were exactly 90o, there would be no
possibility of basal participation in M5(A1), because the four basal functions would fall on a
nodal plane of the matching pz orbital. In most real examples of square pyramidal
geometry, the central atom is not coplanar with the basal positions. Such departures from
perfect 2 = 90o geometry would allow some measure of constructive overlap between the
basal functions. The modified function to include such minor basal participation would be

M'5(A1) = c'axialF5 ± c'basal(F1 + F2 + F3 + F4 )

where the negative sign would be appropriate for 2 > 90o, and the positive sign would be
appropriate for 2 < 90o. If the dz2 orbital were to replace the pz orbital as the major axial
AO on the central atom, the basal positions would make a significant contribution and the
SALC would be

M"5(A1) = c"axialF5 – c"basal(F1 + F2 + F3 + F4 )

The match between dz2 and this modified SALC would be as follows:

Overlap between the annular part of the dz2 orbital and the basal functions of the SALC
would be most effective at 2 = 90o.

5.12 The purpose of this question is to explore an alternative approach to constructing SALCs
and hybrids in general, and for the tbp case in specific. In the Dahr method, an expression for
each SALC that would match with the various central-atom AOs is written down as the sum of
the projections of each pendant atom function, Ni, on the reference axes of the central atom AO,
with proper adjustment of the mathematical signs of the orbital. Each projection is given by Ni
cos 2i, where 2i is the angle between the ABi bond of each pendant atom and the reference axis
of the central AO. The d orbital functions require combining projections. In the case of the
SALC to match with dz2, this approach yields a function with unequal contributions in the
equatorial plane, which is intuitively incorrect. Recognizing this, Dahr corrects the function
empirically to maintain C3 symmetry. Beyond addressing the C3 dissymmetry, Dahr's correction
implicitly gives equal weight to the equatorial and axial positions, although this is not explicitly
stated in the paper. The final set of functions is comparable to Eqs. (5.26a - 5.26e), shown on
page 161 of the text. As such, SALCs represent a special case in which axial and equatorial
positions are made equivalent. Dahr does not note the lack of generality in these results. The
advantages and disadvantages of each approach are primarily a matter of personal preference.

-85-