Projection operators and symmetry-adapted linear combinations (SALCs)

Intuition Beginner

When atomic orbitals on different atoms mix to form molecular orbitals, symmetry imposes a powerful constraint: only orbitals of matching symmetry type can combine. A symmetry-adapted linear combination (SALC) is a specific mixture of atomic orbitals that transforms cleanly under the molecule's symmetry operations — it belongs to exactly one row of the character table. SALCs are the "symmetry-approved" building blocks from which molecular orbital diagrams are assembled.

The projection operator is the mathematical tool that generates SALCs. The idea is straightforward: start with any single atomic orbital (a basis function), apply every symmetry operation in the point group, weight each result by a number from the character table, and add everything together. The result is either a valid SALC or zero. If you get zero, try a different starting orbital.

Why does this matter? Without SALCs, you would have to guess which combinations of atomic orbitals produce valid molecular orbitals. The projection operator removes the guesswork entirely. It guarantees that every SALC has a definite symmetry label, and only SALCs of the same label can interact. This is the foundation for building molecular orbital diagrams systematically.

Visual Beginner

Four hydrogen 1s orbitals in a square (D${}_{4h}$ arrangement) being combined by projection operators into four SALCs: one totally symmetric (all in-phase, A${}_{1g}$), one with alternating phases on opposite corners (B${}_{1g}$), one with alternating phases on adjacent corners (B${}_{2g}$), and one degenerate pair (E${}_u$) with nodal planes through the square.

Worked example Beginner

Construct the SALCs of the O–H bonds in water (H${}_2$O, C${}_{2v}$).

Label the two hydrogen 1s orbitals ${\varphi }_1$ (left H) and ${\varphi }_2$ (right H). The C${}_{2v}$ point group has four operations: E (identity), C${}_2$ (180-degree rotation), ${\sigma }_v$(xz) (reflection in the molecular plane), and ${\sigma }_v^{\prime }$(yz) (reflection perpendicular to the molecular plane). Track what each operation does to ${\varphi }_1$:

Operation	Effect on ${\varphi }_1$
E	${\varphi }_1$
C${}_2$	${\varphi }_2$
${\sigma }_v$(xz)	${\varphi }_1$
${\sigma }_v^{\prime }$(yz)	${\varphi }_2$

To project onto A${}_1$ symmetry (characters: 1, 1, 1, 1), multiply each result by the A${}_1$ character and add:

$\text{SALC}(A_1)=1\cdot {\varphi }_1+1\cdot {\varphi }_2+1\cdot {\varphi }_1+1\cdot {\varphi }_2=2({\varphi }_1+{\varphi }_2)$

After normalising: ${\psi }_{A_1}=\frac{1}{\sqrt{2}}({\varphi }_1+{\varphi }_2)$. This is the symmetric combination — both O–H bonds in phase.

To project onto B${}_2$ symmetry (characters: 1, -1, -1, 1):

$\text{SALC}(B_2)=1\cdot {\varphi }_1+(-1)\cdot {\varphi }_2+(-1)\cdot {\varphi }_1+1\cdot {\varphi }_2=0$

Wait — that gives zero. The issue is that B${}_2$ is not the right irrep for these orbitals. Projecting onto B${}_1$ (characters: 1, -1, 1, -1):

$\text{SALC}(B_1)=1\cdot {\varphi }_1+(-1)\cdot {\varphi }_2+1\cdot {\varphi }_1+(-1)\cdot {\varphi }_2=2({\varphi }_1-{\varphi }_2)$

After normalising: ${\psi }_{B_1}=\frac{1}{\sqrt{2}}({\varphi }_1-{\varphi }_2)$. This is the antisymmetric combination — the two O–H bonds out of phase. The O–H stretching SALCs of water are therefore A${}_1$ + B${}_1$.

Check your understanding Beginner

Formal definition Intermediate+

The projection operator

Let $G$ be a finite group of order $h=|G|$ with irreducible representations ${\Gamma }_1,\dots ,{\Gamma }_k$. The projection operator for irrep ${\Gamma }_i$ (of dimension $l_i$) is:

$${\hat{P}}^{{\Gamma }_i}=\frac{l_i}{h}\sum _{R\in G}\overline{{\chi }_i(R)}\hat{R}$$

where ${\chi }_i(R)$ is the character of operation $R$ in irrep ${\Gamma }_i$, and $\hat{R}$ is the operator that applies the symmetry operation $R$ to the basis function. Applying ${\hat{P}}^{{\Gamma }_i}$ to any function $f$ yields a function that transforms as ${\Gamma }_i$ (or zero, if $f$ has no component of that symmetry).

SALC construction procedure

Given a set of equivalent basis functions $\{{\varphi }_1,\dots ,{\varphi }_n\}$ on atomic positions in a molecule of point group $G$, the step-by-step procedure for constructing SALCs is:

1. Determine the point group $G$ and identify the character table.
2. Construct the reducible representation ${\Gamma }_{\text{basis}}$ by recording, for each class of symmetry operation, how many basis functions are left in their original position (the number of unmoved basis functions). This gives the characters of ${\Gamma }_{\text{basis}}$.
3. Decompose ${\Gamma }_{\text{basis}}$ into irreps using the reduction formula from 16.02.02 pending. This tells you how many SALCs of each symmetry type to expect.
4. For each irrep ${\Gamma }_i$ in the decomposition, apply the projection operator ${\hat{P}}^{{\Gamma }_i}$ to a starting basis function ${\varphi }_j$. If the result is nonzero, normalise it to obtain a SALC. If the result is zero, try a different starting basis function.
5. For degenerate irreps (dimension > 1), the projection operator applied to a single basis function produces only one partner function. Generate the remaining partner(s) by applying a nontrivial symmetry operation to the first SALC, or by starting from a different basis function.

Normalisation

After projection, the raw SALC has the form $\psi ={\sum }_j{c}_j{\varphi }_j$. Normalise by dividing by $\sqrt{{\sum }_j|c_j{|}^2}$ (assuming the ${\varphi }_j$ are orthonormal):

$${\psi }_{\text{norm}}=\frac{1}{\sqrt{{\sum }_j|c_j{|}^2}}\sum _j{c}_j{\varphi }_j$$

For real basis functions with unit overlap, this simplifies to dividing by the square root of the sum of squared coefficients.

Transfer operators for degenerate representations

For a degenerate irrep ${\Gamma }_i$ of dimension $l_i$, the full projection operator can be decomposed into component operators:

$${\hat{P}}_{\mu \nu }^{{\Gamma }_i}=\frac{l_i}{h}\sum _{R\in G}\overline{[D^{{\Gamma }_i}(R){]}_{\mu \nu }}\hat{R}$$

where $D^{{\Gamma }_i}(R)$ is the full matrix representation. The diagonal operator ${\hat{P}}_{\mu \mu }^{{\Gamma }_i}$ projects onto the $\mu$-th partner function. In practice, the character-based projection operator (using only traces) produces a sum of all partner functions. To separate them, apply a symmetry operation that mixes the degenerate components to the first projected SALC.

Counterexamples to common slips

• The projection operator applied to the wrong basis function gives zero, not an error. Zero is a valid (and informative) result. It means the starting function has no component of the target symmetry. Move to another basis function.

• SALCs are not molecular orbitals. A SALC is a symmetry-adapted combination of basis functions on one set of equivalent atoms. A molecular orbital results from combining SALCs with central-atom orbitals (or other SALCs) of matching symmetry. The SALC is the input; the MO is the output.

• Degenerate SALCs are not unique. Any unitary transformation of the partner functions within a degenerate set is equally valid. The SALCs you obtain depend on your choice of starting function and coordinate system. Only the symmetry labels and the spanned subspace are invariant.

Core model Intermediate+

SALCs for BF3 (D3h)

BF${}_3$ has three equivalent fluorine atoms arranged in an equilateral triangle. Consider the three fluorine 2p${}_z$ orbitals (perpendicular to the molecular plane), labelled ${\varphi }_1$, ${\varphi }_2$, ${\varphi }_3$.

The reducible representation for three equivalent basis functions:

D${}_{3h}$	E	2C${}_3$	3C${}_2$	${\sigma }_h$	2S${}_3$	3${\sigma }_v$
Unmoved	3	0	1	3	0	1
${\Gamma }_{\text{basis}}$	3	0	1	3	0	1

Using the reduction formula with $|G|=12$:

$n_{A_1^{\prime }}=\frac{1}{12}[1(3)(1)+2(0)(1)+3(1)(1)+1(3)(1)+2(0)(1)+3(1)(1)]=\frac{3+0+3+3+0+3}{12}=1$

$n_{A_2^{\prime }}=\frac{1}{12}[1(3)(1)+2(0)(1)+3(1)(-1)+1(3)(1)+2(0)(1)+3(1)(-1)]=\frac{3+0-3+3+0-3}{12}=0$

$n_{E^{\prime }}=\frac{1}{12}[1(3)(2)+2(0)(-1)+3(1)(0)+1(3)(2)+2(0)(-1)+3(1)(0)]=\frac{6+0+0+6+0+0}{12}=1$

$n_{A_1^{\prime \prime }}=\frac{1}{12}[1(3)(1)+2(0)(1)+3(1)(1)+1(3)(-1)+2(0)(-1)+3(1)(-1)]=\frac{3+0+3-3+0-3}{12}=0$

$n_{A_2^{\prime \prime }}=\frac{1}{12}[1(3)(1)+2(0)(1)+3(1)(-1)+1(3)(-1)+2(0)(-1)+3(1)(1)]=\frac{3+0-3-3+0+3}{12}=0$

$n_{E^{\prime \prime }}=\frac{1}{12}[1(3)(2)+2(0)(-1)+3(1)(0)+1(3)(-2)+2(0)(1)+3(1)(0)]=\frac{6+0+0-6+0+0}{12}=0$

So ${\Gamma }_{\text{basis}}=A_1^{\prime }+E^{\prime }$.

A${}_1^{\prime }$ SALC. Apply ${\hat{P}}^{A_1^{\prime }}$ to ${\varphi }_1$ (characters: 1, 1, 1, 1, 1, 1):

${\hat{P}}^{A_1^{\prime }}{\varphi }_1=\frac{1}{12}[1\cdot {\varphi }_1+1\cdot {\varphi }_2+1\cdot {\varphi }_3+1\cdot {\varphi }_1+1\cdot {\varphi }_2+1\cdot {\varphi }_3+1\cdot {\varphi }_1+1\cdot {\varphi }_3+1\cdot {\varphi }_2+1\cdot {\varphi }_1+1\cdot {\varphi }_2+1\cdot {\varphi }_3]$

Collecting carefully using all 12 operations (E, C${}_3$, C${}_3^2$, C${}_2^{(1)}$, C${}_2^{(2)}$, C${}_2^{(3)}$, ${\sigma }_h$, S${}_3$, S${}_3^5$, ${\sigma }_v^{(1)}$, ${\sigma }_v^{(2)}$, ${\sigma }_v^{(3)}$):

$=\frac{1}{12}[4{\varphi }_1+4{\varphi }_2+4{\varphi }_3]=\frac{1}{3}({\varphi }_1+{\varphi }_2+{\varphi }_3)$

Normalised: ${\psi }_{A_1^{\prime }}=\frac{1}{\sqrt{3}}({\varphi }_1+{\varphi }_2+{\varphi }_3)$

E' SALCs. Apply ${\hat{P}}^{E^{\prime }}$ to ${\varphi }_1$ (characters: 2, -1, 0, 2, -1, 0):

${\hat{P}}^{E^{\prime }}{\varphi }_1=\frac{2}{12}[2\cdot {\varphi }_1+(-1)\cdot {\varphi }_2+(-1)\cdot {\varphi }_3+0+0+0+2\cdot {\varphi }_1+(-1)\cdot {\varphi }_3+(-1)\cdot {\varphi }_2+0+0+0]$

$=\frac{1}{6}[4{\varphi }_1-2{\varphi }_2-2{\varphi }_3]=\frac{1}{3}(2{\varphi }_1-{\varphi }_2-{\varphi }_3)$

Normalised: ${\psi }_{E^{\prime },1}=\frac{1}{\sqrt{6}}(2{\varphi }_1-{\varphi }_2-{\varphi }_3)$

The second E' partner is obtained by applying C${}_3$ to ${\psi }_{E^{\prime },1}$:

${\hat{C}}_3{\psi }_{E^{\prime },1}=\frac{1}{\sqrt{6}}(2{\varphi }_2-{\varphi }_3-{\varphi }_1)$

An orthogonal linear combination gives a cleaner form: ${\psi }_{E^{\prime },2}=\frac{1}{\sqrt{2}}({\varphi }_2-{\varphi }_3)$

SALCs for H2O (C2v)

Water has two hydrogen 1s orbitals ${\varphi }_1$ and ${\varphi }_2$. The reducible representation:

C${}_{2v}$	E	C${}_2$	${\sigma }_v$(xz)	${\sigma }_v^{\prime }$(yz)
Unmoved	2	0	2	0
${\Gamma }_{\text{basis}}$	2	0	2	0

Decomposition: $n_{A_1}=\frac{1}{4}(2+0+2+0)=1$, $n_{B_1}=\frac{1}{4}(2+0+2+0)\cdot {\chi }_{B_1}$... More carefully:

$n_{A_1}=\frac{1}{4}[1(2)(1)+1(0)(1)+1(2)(1)+1(0)(1)]=1$

$n_{A_2}=\frac{1}{4}[1(2)(1)+1(0)(1)+1(2)(-1)+1(0)(-1)]=0$

$n_{B_1}=\frac{1}{4}[1(2)(1)+1(0)(-1)+1(2)(1)+1(0)(-1)]=1$

$n_{B_2}=\frac{1}{4}[1(2)(1)+1(0)(-1)+1(2)(-1)+1(0)(1)]=0$

${\Gamma }_{\text{basis}}=A_1+B_1$.

A${}_1$ SALC: ${\psi }_{A_1}=\frac{1}{\sqrt{2}}({\varphi }_1+{\varphi }_2)$ (symmetric, both H orbitals in phase).

B${}_1$ SALC: ${\psi }_{B_1}=\frac{1}{\sqrt{2}}({\varphi }_1-{\varphi }_2)$ (antisymmetric, H orbitals out of phase).

The A${}_1$ SALC interacts with the O 2s and O 2p${}_z$ orbitals (both A${}_1$). The B${}_1$ SALC interacts with the O 2p${}_x$ orbital (B${}_1$). The O 2p${}_y$ (B${}_2$) has no matching SALC and remains a nonbonding lone pair.

Bridge. The SALC construction for H${}_2$O and BF${}_3$ demonstrates the systematic generation of symmetry-matched basis functions from the character table. This procedure generalises directly to the ligand group orbitals of coordination complexes, where the SALCs of ligand sigma-donor orbitals combine with metal d, s, and p orbitals to produce the molecular orbital diagrams treated in 16.03.01 and 16.04.03 pending.

Exercises Intermediate+

SALCs for octahedral and tetrahedral complexes, symmetry descent, and solid-state applications Master

Ligand group orbitals for octahedral complexes (Oh)

For an octahedral complex ML${}_6$ with six sigma-donor ligands, the six ligand sigma orbitals form a reducible representation that decomposes as:

$${\Gamma }_{\sigma }=A_{1g}+E_g+T_{1u}$$

The six SALCs are:

$A_{1g}$: $\frac{1}{\sqrt{6}}({\varphi }_1+{\varphi }_2+{\varphi }_3+{\varphi }_4+{\varphi }_5+{\varphi }_6)$ — all six ligand orbitals in phase, interacting with the metal s orbital and d${}_{z^2}$ (part of E${}_g$).

$E_g$ pair:

• ${\psi }_{E_g,1}=\frac{1}{\sqrt{12}}(2{\varphi }_1+2{\varphi }_2-{\varphi }_3-{\varphi }_4-{\varphi }_5-{\varphi }_6)$ (matching d${}_{z^2}$, with the two axial ligands enhanced and the four equatorial diminished)
• ${\psi }_{E_g,2}=\frac{1}{2}({\varphi }_3-{\varphi }_4+{\varphi }_5-{\varphi }_6)$ (matching d${}_{x^2-y^2}$, with alternating signs on the equatorial ligands)

$T_{1u}$ trio:

• ${\psi }_{T_{1u},1}=\frac{1}{\sqrt{2}}({\varphi }_1-{\varphi }_2)$ (matching p${}_z$, nodal plane through the equator)
• ${\psi }_{T_{1u},2}=\frac{1}{\sqrt{2}}({\varphi }_3-{\varphi }_4)$ (matching p${}_x$)
• ${\psi }_{T_{1u},3}=\frac{1}{\sqrt{2}}({\varphi }_5-{\varphi }_6)$ (matching p${}_y$)

where ${\varphi }_{1,2}$ are axial (z-axis) and ${\varphi }_{3,4,5,6}$ are equatorial.

The critical observation: the metal d${}_{xy}$, d${}_{xz}$, and d${}_{yz}$ orbitals (T${}_{2g}$) have no matching sigma SALC. The T${}_{2g}$ set is nonbonding with respect to sigma donation, forming the basis of crystal field theory's t${}_{2g}$ level. This is the group-theoretic origin of the eg/t${}_{2g}$ splitting in octahedral complexes.

For pi-donor or pi-acceptor ligands, the twelve pi-type ligand orbitals decompose as:

$${\Gamma }_{\pi }=T_{1g}+T_{2g}+T_{1u}+T_{2u}$$

The T${}_{2g}$ SALCs of the pi orbitals now interact with the metal T${}_{2g}$ d orbitals, introducing pi-bonding contributions that shift the t${}_{2g}$ level relative to the sigma-only picture. Pi-donor ligands raise the t${}_{2g}$ energy (anti-bonding interaction); pi-acceptor ligands lower it (back-bonding stabilization). This explains the spectrochemical series: pi-acceptors like CO and CN${}^{-}$ produce large Delta${}_o$ while pi-donors like F${}^{-}$ and H${}_2$O produce small Delta${}_o$.

Ligand group orbitals for tetrahedral complexes (Td)

For a tetrahedral complex ML${}_4$ with four sigma-donor ligands at the vertices of a tetrahedron, the four ligand sigma orbitals decompose as:

$${\Gamma }_{\sigma }=A_1+T_2$$

The SALCs are the same as for methane (A${}_1$ + T${}_2$). The metal orbital matching is:

• $A_1$ SALC interacts with metal s (A${}_1$)
• $T_2$ SALCs interact with metal p${}_x$, p${}_y$, p${}_z$ (T${}_2$) and metal d${}_{xy}$, d${}_{xz}$, d${}_{yz}$ (T${}_2$)

The metal d${}_{z^2}$ and d${}_{x^2-y^2}$ (E) have no matching sigma SALC and are nonbonding — this is the group-theoretic origin of the e/t${}_2$ splitting in tetrahedral crystal field theory (inverted relative to O${}_h$: e is lower, t${}_2$ is higher).

The tetrahedral pi orbitals (eight total) decompose as:

$${\Gamma }_{\pi }=E+T_1+T_2$$

The E SALCs of the pi orbitals interact with the metal e orbitals (d${}_{z^2}$, d${}_{x^2-y^2}$), partially offsetting their nonbonding character. The T${}_2$ pi SALCs interact with the same metal T${}_2$ orbitals that already engage in sigma bonding, adding a pi component. The T${}_1$ SALCs are nonbonding with respect to the metal d orbitals.

Symmetry descent and SALC correlation

When a complex undergoes a structural distortion that lowers its symmetry from G to a subgroup H, the SALCs must be re-expressed in the lower symmetry. The correlation proceeds through the branching of irreps:

O${}_h$ to D${}_{4h}$ (Jahn-Teller tetragonal distortion):

O${}_h$ irrep	SALC	D${}_{4h}$ irrep	Effect
A${}_{1g}$	all six in phase	A${}_{1g}$	unchanged
E${}_g$	d${}_{z^2}$-like, d${}_{x^2-y^2}$-like	A${}_{1g}$ + B${}_{1g}$	splits into two nondegenerate SALCs
T${}_{1u}$	p${}_z$-like, p${}_x$-like, p${}_y$-like	A${}_{2u}$ + E${}_u$	the axial SALC separates from the equatorial pair

The splitting of E${}_g$ into A${}_{1g}$ + B${}_{1g}$ means the two SALCs that were degenerate in O${}_h$ now have different energies. The d${}_{z^2}$-like SALC (A${}_{1g}$) is stabilised by elongation (weaker axial interaction), while the d${}_{x^2-y^2}$-like SALC (B${}_{1g}$) is destabilised (stronger equatorial interaction). This is the orbital-level description of the Jahn-Teller effect.

O${}_h$ to C${}_{2v}$ (dissociation of one ligand, forming a square pyramid):

The descent chain O${}_h$ → C${}_{4v}$ → C${}_{2v}$ shows progressive splitting: T${}_{1u}$ → A${}_1$ + E → 2A${}_1$ + B${}_1$ + B${}_2$. Each step removes degeneracy, and the SALCs must be recombined as linear combinations of the original O${}_h$ SALCs weighted by the descent coefficients.

Computational verification of SALC energies

The SALCs constructed by projection operators are symmetry-adapted but not energy-optimised — they are linear combinations of equivalent atomic orbitals with equal weights. In a real molecule, the overlap between non-nearest-neighbour atoms, differential hybridisation, and electron-electron repulsion break the equal-weight assumption. Computational chemistry methods (Hartree-Fock, DFT) verify and refine the SALC picture:

1. The symmetry labels are exact. A SALC labelled E${}_g$ in O${}_h$ remains E${}_g$ regardless of the computational method — symmetry labels are properties of the point group, not approximations.

2. The relative energies are refined. The projection-operator SALCs give qualitative MO diagrams (bonding below antibonding, nonbonding in between). Computational methods provide quantitative orbital energies that include kinetic energy, nuclear attraction, and electron-electron repulsion effects.

3. The SALC coefficients are corrected. Equal-coefficient SALCs (e.g., $\frac{1}{\sqrt{6}}({\varphi }_1+\dots +{\varphi }_6)$ for A${}_{1g}$) are the free-particle (no-interaction) limit. Actual computed MO coefficients reflect the true Hamiltonian and differ from the projection-operator result. The deviation measures the importance of interactions beyond symmetry.

4. Extended Huckel theory provides a computationally cheap intermediate: it uses the SALC symmetry to block-diagonalise the secular determinant, then solves each symmetry block independently. This gives orbital energies that respect symmetry while incorporating overlap effects.

Applications to solid-state band structures

In a crystalline solid, the point group symmetry of the unit cell is replaced by the space group, which includes translational symmetry. The SALC concept generalises to Bloch functions:

$${\psi }_{\mathbf{k}}(\mathbf{r})=\frac{1}{\sqrt{N}}\sum _{\mathbf{R}}{e}^{i\mathbf{k}\cdot \mathbf{R}}\varphi (\mathbf{r}-\mathbf{R})$$

This is a SALC of the atomic orbital $\varphi$ adapted to the translational symmetry of the lattice. The wavevector $\mathbf{k}$ plays the role of the irrep label, and the exponential factor $e^{i\mathbf{k}\cdot \mathbf{R}}$ is the character of the translation operation $\mathbf{R}$ in the irrep labelled by $\mathbf{k}$.

The connection between molecular SALCs and solid-state band structures is:

• Finite group → space group. The point group is replaced by the space group (translations + point operations). The irreps of the translation subgroup are labelled by $\mathbf{k}$; the full space-group irreps are labelled by $(\mathbf{k},\lambda )$ where $\lambda$ is an irrep of the little group of $\mathbf{k}$.

• SALC → Bloch function. The projection operator generalises to the Bloch construction. Instead of summing over symmetry operations of a finite group, you sum over lattice translations with phase factors.

• MO diagram → band structure. A molecular orbital diagram shows discrete energy levels; a band structure shows the continuous dispersion $E(\mathbf{k})$ as $\mathbf{k}$ varies across the Brillouin zone. The number of bands equals the number of basis functions per unit cell. Bands at high-symmetry points in the Brillouin zone carry the labels of the little-group irreps — the direct analogue of SALC symmetry labels.

For a solid with the NaCl structure (O${}_h^5$, Fm$\bar{3}$m), the sigma SALCs of the halide p orbitals at the $\Gamma$ point ($\mathbf{k}=0$) decompose into exactly the same irreps (A${}_{1g}$, E${}_g$, T${}_{1u}$, T${}_{2g}$, T${}_{2u}$) as the sigma SALCs of an octahedral complex, because the little group of $\Gamma$ is O${}_h$. Moving away from $\Gamma$, the translational phase factor breaks the point-group degeneracies, producing the valence band dispersion observed in photoemission experiments.

Nontrivial applications: surface adsorption and cluster chemistry

The SALC framework extends to adsorbates on surfaces and to large metal clusters. For CO adsorbed on a metal surface with C${}_{2v}$ site symmetry, the CO 5$\sigma$ and 2${\pi }^{\ast }$ orbitals form SALCs of the site-symmetry group. The matching between adsorbate SALCs and metal surface orbital symmetries determines the adsorption geometry and binding energy — this is the orbital symmetry analysis underlying the Blyholder model of CO adsorption.

For metal clusters such as Os${}_3$(CO)${}_{12}$ (D${}_{3h}$), the twelve CO ligands form a large set of SALCs. The projection-operator approach scales systematically: the 12 sigma-donor CO orbitals decompose into a manageable number of SALCs (in D${}_{3h}$: 4A${}_1^{\prime }$ + 2A${}_2^{\prime }$ + 2A${}_1^{\prime \prime }$ + 2A${}_2^{\prime \prime }$ + 4E' + 2E''), each of which interacts with a specific combination of metal-cluster orbitals. The SALC decomposition reduces a 12-dimensional problem to a series of low-dimensional symmetry blocks.

Connections Master

To representation theory. The projection operator is a specialisation of the abstract projection from representation theory: for any finite group $G$ and irrep $\Gamma$, the operator ${\hat{P}}^{\Gamma }=\frac{\dim \Gamma }{|G|}{\sum }_g\overline{{\chi }_{\Gamma }(g)}\hat{g}$ projects onto the isotypic component of $\Gamma$ in any $G$-module. The SALC construction is the application of this abstract tool to the specific $G$-module of functions on atomic positions.

To molecular orbital theory and ligand field theory. SALCs are the ligand-group orbitals that combine with central-atom orbitals in the molecular orbital diagrams of coordination complexes. The symmetry matching enforced by SALCs determines which metal-ligand interactions are symmetry-allowed and which are forbidden. This is the bridge from group theory (16.02) to crystal field theory (16.03) and coordination chemistry (16.04), culminating in the quantitative MO diagrams of 16.04.03 pending.

To solid-state physics. The Bloch-function generalisation of SALCs connects molecular symmetry to band theory. The Brillouin zone is the dual of the translational symmetry group; the band labels at high-symmetry points are the irreps of the little group. Understanding SALCs is therefore prerequisite to understanding band structures, Fermi surfaces, and the symmetry classification of electronic states in crystals.

To spectroscopy. SALCs determine the symmetry of vibrational normal modes (as shown in 16.02.02 pending) and of electronic transitions. The selection rules derived from SALC symmetry labels predict IR and Raman activity, electronic absorption polarisations, and the intensity patterns in photoelectron spectra. Computational spectroscopy uses SALC symmetry to assign observed spectral features to specific molecular motions or electronic transitions.

To reaction mechanisms. The Woodward-Hoffmann rules for pericyclic reactions use orbital symmetry to predict whether a reaction pathway is thermally allowed or forbidden. The SALC framework provides the symmetry labels for the frontier orbitals that determine the correlation diagram between reactant and product states. The same projection-operator machinery that generates ligand SALCs also generates the symmetry-adapted frontier orbitals used in orbital correlation diagrams.

Historical notes Master

The concept of constructing symmetry-adapted basis functions from group representations dates to Frobenius and Schur in the early 1900s, who developed the representation theory of finite groups including the projection-operator formalism. Their work was purely mathematical, with no application to molecular structure.

Bethe (1929) implicitly used SALCs in his crystal-field analysis, constructing the symmetry-adapted combinations of ligand electrostatic potentials that split the d-orbital degeneracy. Bethe did not use the term "SALC" but the mathematical content is equivalent: he projected the ligand-field potential onto the irreps of O${}_h$ to determine the d-orbital splitting pattern.

Van Vleck (1935) made the connection between Bethe's crystal-field model and molecular orbital theory explicit, showing that the ligand orbitals form SALCs that interact with the metal d orbitals according to their shared symmetry labels. Van Vleck's paper ("The Group Relation Between the Mulliken and Slater-Pauling Theories of Valence," J. Chem. Phys. 3, 803–806, 1935) reconciled the two competing models of transition-metal bonding.

Cotton (1963) provided the first systematic textbook treatment of SALC construction using projection operators (Chemical Applications of Group Theory, Ch. 6). Cotton standardised the step-by-step procedure: (1) build the reducible representation, (2) decompose it, (3) apply the projection operator to obtain explicit SALCs. This procedure remains the standard pedagogical approach.

Albright, Burdett, and Whangbo (Orbital Interactions in Chemistry, 1st ed. 1985, 2nd ed. 2013) extended the SALC framework to a wide range of chemical problems including extended solids, organometallic clusters, and surface chemistry. Their treatment emphasised the qualitative power of SALC-based orbital interaction diagrams for predicting molecular structure and reactivity without full computation.

The Bloch theorem (1929) established the solid-state analogue of SALCs independently of the molecular symmetry development. The connection between the two frameworks — finite-group SALCs as the zero-dimensional limit of Bloch functions — was recognised by Bouckaert, Smoluchowski, and Wigner (1936) in their classification of electronic band states by space-group irreps.

Bibliography Master

• Cotton, F. A. — Chemical Applications of Group Theory, 3rd ed. (Wiley, 1990), Ch. 6. The canonical treatment of projection operators and SALC construction.
• Albright, T. A., Burdett, J. K. & Whangbo, M.-H. — Orbital Interactions in Chemistry, 2nd ed. (Wiley, 2013), Ch. 2. SALC-based orbital interaction diagrams for molecules, clusters, and solids.
• Miessler, G. L., Fischer, P. J. & Tarr, D. A. — Inorganic Chemistry, 5th ed. (Pearson, 2014), Ch. 4. Introductory treatment of SALCs with worked examples for common geometries.
• Van Vleck, J. H. — The Group Relation Between the Mulliken and Slater-Pauling Theories of Valence, J. Chem. Phys. 3 (1935), 803–806. Reconciliation of crystal-field and molecular-orbital approaches using SALC symmetry matching.
• Bethe, H. — Termaufspaltung in Kristallen, Ann. Phys. 3 (1929), 133–208. Implicit SALC construction in the crystal-field splitting analysis.
• Bouckaert, L. P., Smoluchowski, R. & Wigner, E. P. — Theory of Brillouin Zones and Symmetry Properties of Wave Functions in Crystals, Phys. Rev. 50 (1936), 58–67. Space-group irreps as the solid-state generalisation of molecular SALC labels.
• Bishop, D. M. — Group Theory and Chemistry (Clarendon Press, 1973). Comprehensive mathematical treatment of projection operators, transfer operators, and SALC normalisation.
• Hoffman, R. — Extended Huckel theory, J. Chem. Phys. 39 (1963), 1397–1412. Computational method that uses SALC symmetry to block-diagonalise the secular equation.