Selection Rules from Direct Products

1. The Quantum Origin of Selection Rules

From quantum mechanics, you know that the probability of a transition from an initial state ${\psi }_i$ to a final state ${\psi }_f$, induced by an interaction with light, is governed by the transition amplitude. For most common spectroscopic transitions (UV-Vis, IR), the dominant interaction is the electric dipole coupling. The transition probability is then proportional to the square of the transition dipole moment integral:

Here, $\hat{\mu }=e\mathbf{r}$ is the electric dipole operator. A transition is "allowed" if this integral, ${\mu }_{fi}$, can be non-zero. If symmetry dictates that this integral must be zero, the transition is "forbidden".

Our central task, therefore, boils down to a single question: under what conditions is the integral $\int {\psi }_f^{\ast }\hat{\mu }{\psi }_{i}d\tau$ guaranteed to be zero by symmetry?

2. The Fundamental Theorem for Integrals

Group theory provides a beautifully simple answer to this question, which is a cornerstone of its application in quantum physics.

An integral $\int Fd\tau$ over all space can be non-zero only if the integrand $F$ transforms as the totally symmetric irreducible representation of the system's symmetry group.

The totally symmetric irrep is the one-dimensional representation where all characters are +1 (often labeled A, A₁, or A₁g). A function transforming this way is unchanged by any symmetry operation. Since the integral itself is just a number and must be invariant under any symmetry operation, the function being integrated must also possess this total symmetry for the integral not to vanish.

3. The Direct Product and the General Selection Rule

The integrand in our transition dipole moment is a product of three functions: ${\psi }_f^{\ast }$, $\hat{\mu }$, and ${\psi }_i$. Each of these functions serves as a basis for some representation (generally irreducible) of the molecular point group. Let's denote their representations as ${\Gamma }_f$, ${\Gamma }_{\mu }$, and ${\Gamma }_i$, respectively.

To determine the symmetry of the product of these functions, we use the direct product of their representations. The character of the direct product representation ${\Gamma }_A\otimes {\Gamma }_B$ for a given operation $R$ is simply the product of the individual characters:

The full integrand, ${\psi }_f^{\ast }\hat{\mu }{\psi }_i$, therefore transforms as the direct product representation ${\Gamma }_f^{\ast }\otimes {\Gamma }_{\mu }\otimes {\Gamma }_i$. (For groups with real characters, which includes all point groups you'll commonly encounter, the complex conjugate representation ${\Gamma }_f^{\ast }$ is identical to ${\Gamma }_f$).

Combining this with the fundamental theorem for integrals gives us the general selection rule:

A transition from state ${\psi }_i$ to ${\psi }_f$ is symmetry-allowed if and only if the direct product representation ${\Gamma }_f\otimes {\Gamma }_{\mu }\otimes {\Gamma }_i$ contains the totally symmetric irreducible representation (A₁ or A₁g).

An equivalent and often more practical statement is:
A transition is allowed if ${\Gamma }_f$ is contained in the decomposition of the direct product ${\Gamma }_{\mu }\otimes {\Gamma }_i$.

To use this rule, we need the representation for the dipole operator, ${\Gamma }_{\mu }$. Since $\hat{\mu }=e\mathbf{r}$, its components $({\mu }_x,{\mu }_y,{\mu }_z)$ transform identically to the Cartesian coordinates $(x,y,z)$. The character table for any point group lists which irreps transform as $x$, $y$, and $z$ in its final columns.

4. Application 1: Electronic Transitions in an O_h Field

Let's apply this to the d-orbital splitting in an octahedral field from our last lesson. We found that the d-orbitals split into levels of E_g and T_2g symmetry.
Question: Is the electronic transition $T_{2g}\to E_g$ allowed by electric dipole absorption?

1. Identify the representations:

   • Initial state: ${\Gamma }_i=T_{2g}$
   • Final state: ${\Gamma }_f=E_g$
   • Dipole operator: From an O_h character table, you can see that $(x,y,z)$ transform together as the irrep ${\Gamma }_{\mu }=T_{1u}$.

2. Form the direct product: We need to compute ${\Gamma }_{\mu }\otimes {\Gamma }_i=T_{1u}\otimes T_{2g}$. We do this by multiplying their characters for each class of the O_h group.

3. Decompose the direct product: We now use the reduction formula to decompose this reducible representation. For any irrep ${\Gamma }_{\alpha }$, the number of times it appears, $n_{\alpha }$, is:
   $n_{\alpha }=\frac{1}{|G|}{\sum }_R{N}_R{\chi }^{\alpha }(R{)}^{\ast }{\chi }^{T_{1u}\otimes T_{2g}}(R)$
   Performing this decomposition yields:
   $T_{1u}\otimes T_{2g}=A_{2u}\oplus E_u\oplus T_{1u}\oplus T_{2u}$

4. Check the condition: Our final state is ${\Gamma }_f=E_g$. Is $E_g$ contained in the decomposition $A_{2u}\oplus E_u\oplus T_{1u}\oplus T_{2u}$? No, it is not.

Conclusion: The transition $T_{2g}\to E_g$ is symmetry-forbidden. In fact, all the resulting irreps are ungerade (u), while our final state is gerade (g). This reveals a more general rule for centrosymmetric systems known as the Laporte rule: electric dipole transitions are only allowed between states of opposite parity (g $\leftrightarrow$ u). A g $\to$ g or u $\to$ u transition is parity-forbidden. This is why transition metal complexes, whose colors arise from these "d-d transitions", typically have very pale colors—the transitions are formally forbidden and thus have low probability. They only occur weakly due to mechanisms like vibrational coupling that momentarily break the perfect inversion symmetry.

5. Application 2: Vibrational Spectroscopy (IR and Raman)

The same principles apply to transitions between vibrational energy levels.

Infrared (IR) Activity

For a fundamental vibrational transition (from the ground state to the first excited vibrational state), the initial state ${\psi }_i$ is the ground vibrational state, which is always totally symmetric (${\Gamma }_i=A_1$). The final state ${\psi }_f$ has the symmetry of the vibrational mode being excited, ${\Gamma }_{vib}$. The selection rule ${\Gamma }_f\subset {\Gamma }_{\mu }\otimes {\Gamma }_i$ becomes:

This simplifies to a very direct rule: A vibrational mode is IR-active if its irreducible representation is the same as one of the irreducible representations for which x, y, or z is a basis function.

Raman Activity

Raman spectroscopy is a two-photon scattering process. A full derivation is extensive, but the result is that the relevant operator is not the dipole moment $\hat{\mu }$, but the molecular polarizability tensor, $\hat{\alpha }$. The components of this tensor transform as the quadratic functions, $x^2,y^2,z^2,xy,xz,yz$. The selection rule is analogous: A vibrational mode is Raman-active if its irreducible representation is the same as one of the irreducible representations for which a quadratic function is a basis function.

6. A Complete Worked Example: Water (C₂ᵥ)

Let's see this entire process in action for the water molecule, which belongs to the C₂ᵥ point group. We first need to determine the symmetries of its vibrational modes, and then apply the selection rules. The following video provides a clear, step-by-step demonstration.

The video demonstrates that for water (a non-linear molecule with 3 atoms, so $3N-6=3$ vibrations), the vibrational modes transform as $2A_1\oplus B_2$. By inspecting the C₂ᵥ character table:

• The A₁ irrep transforms as $z$, and also as $x^2,y^2,z^2$.
• The B₂ irrep transforms as $y$, and also as $yz$.

Therefore, all three vibrational modes ($2A_1,B_2$) are both IR-active and Raman-active. This is a powerful, concrete prediction derived entirely from symmetry.

Conclusion

In this lesson, we have unlocked one of the most practical applications of group theory in physics and chemistry. By understanding the concept of the direct product, we can predict which transitions between quantum states are allowed or forbidden based purely on their symmetry properties.

Key Takeaways:

• Spectroscopic selection rules originate from the condition that the transition dipole moment integral, $⟨{\psi }_f|\hat{\mu }|{\psi }_i⟩$, must be non-zero.
• This integral is non-zero only if the integrand's representation, ${\Gamma }_f\otimes {\Gamma }_{\mu }\otimes {\Gamma }_i$, contains the totally symmetric irrep.
• The dipole operator $\hat{\mu }$ transforms as the coordinates $(x,y,z)$, whose irreps are listed in the character table.
• This framework leads to specific rules:
  • Electronic transitions: Allowed if ${\Gamma }_f\subset {\Gamma }_{\mu }\otimes {\Gamma }_i$. For centrosymmetric systems, this implies a change of parity (g $\leftrightarrow$ u).
  • IR transitions: A vibrational mode is active if it transforms as $x$, $y$, or $z$.
  • Raman transitions: A vibrational mode is active if it transforms as a quadratic product (e.g., $x^2,xy$).

We have now seen how to classify states by symmetry and how to determine the rules for transitions between them. In the next module, we will step back to formalize the role of symmetry in the quantum mechanical framework itself, exploring Wigner's theorem and the fundamental connection between symmetry operators and the structure of Hilbert space.