Redlich-Teller product rule

When an atom in a molecule is swapped for one of its heavier isotopes, the molecule's vibrational frequencies predictably shift. But how can we determine the exact nature of this change without re-solving the entire quantum mechanical problem? This question lies at the heart of vibrational spectroscopy, and the answer is found in the elegant Redlich-Teller product rule. This fundamental principle of molecular physics provides a precise, quantitative link between the "music" of molecular vibrations and the masses of the constituent atoms, allowing chemists and physicists to cross-verify experimental data with remarkable accuracy.

This article delves into the theoretical underpinnings and practical power of this rule. The first chapter, ​​Principles and Mechanisms​​, will deconstruct the rule, starting from a simple diatomic model and building up to the sophisticated framework of Wilson's FG matrix method and the general Teller-Redlich theorem. We will explore how molecular symmetry miraculously simplifies these complex relationships. Following this, the chapter on ​​Applications and Interdisciplinary Connections​​ will reveal the rule's profound impact beyond pure spectroscopy, demonstrating its role in understanding thermodynamic equilibria and the rates of chemical reactions.

Imagine you have a guitar string. You pluck it, and it vibrates at a certain pitch—a certain frequency. This frequency depends on three things: the length of the string, its tension, and its mass per unit length. Now, what if you could swap the string for another of the exact same length and tension, but made of a slightly heavier material? You would intuitively expect the pitch to drop. The heavier string is more sluggish, more resistant to being whipped back and forth, and so it vibrates more slowly.

In a surprisingly deep sense, the bonds between atoms in a molecule are like that guitar string. They vibrate at characteristic frequencies, which we can "hear" using techniques like infrared and Raman spectroscopy. And just like with the guitar string, we can change the "mass" of the atoms participating in the vibration without altering the "tension" of the chemical bond that holds them together. This is done through ​​isotopic substitution​​—swapping an atom for a heavier or lighter version of itself (e.g., replacing a hydrogen atom, $H$, with its heavier isotope, deuterium, $D$).

The deep principle that allows us to do this is the ​​Born-Oppenheimer approximation​​. This cornerstone of quantum chemistry tells us that the light, zippy electrons move so much faster than the heavy, sluggish nuclei that we can consider the nuclei to be frozen in place when we calculate the electronic structure and the corresponding potential energy. This potential energy surface, which dictates the forces between the atoms—the "tension" in our guitar string analogy—depends only on the positions of the nuclei, not their masses. By changing an atom to an isotope, we change its mass, but we don't change the electrical forces holding the molecule together. The music sheet remains the same; only one of the instruments in the orchestra has been made heavier. The question is, how does this affect the symphony of vibrations the molecule can play? The ​​Redlich-Teller product rule​​ provides the beautifully elegant answer.

Let's start with the simplest possible molecule: a diatomic, like carbon monoxide ($CO$) or hydrogen chloride ($HCl$). We can model its vibration as two masses, $m_X$ and $m_Y$, connected by a spring with a force constant $k$. Classical mechanics tells us the frequency of vibration, $\omega$, depends not on the individual masses, but on the ​​reduced mass​​, $\mu = \frac{m_X m_Y}{m_X + m_Y}$. The frequency is given by the simple formula:

As we discussed, the force constant $k$ is a property of the electronic bonding and is unchanged by isotopic substitution. So, if we measure the frequency $\omega_{XY}$ of a molecule $XY$, and then the frequency $\omega_{X'Y}$ of its isotopologue where $X$ is replaced by the isotope $X'$, we know that the ratio of their squared frequencies is simply a ratio of their reduced masses:

This is the product rule in its most basic form. It's a "product" rule because for polyatomic molecules, we will be dealing with products of many frequencies. But the core idea is already here: a relationship between vibrational frequencies that depends only on the masses of the atoms.

We can see the elegance of this principle in a clever thought experiment. Suppose we have four isotopologues of a diatomic molecule: the original $XY$, one with just $X$ substituted ($X'Y$), one with just $Y$ substituted ($XY'$), and the doubly substituted one ($X'Y'$). We can measure the three frequencies $\omega_{XY}$, $\omega_{X'Y}$, and $\omega_{XY'}$. Can we predict the fourth frequency, $\omega_{X'Y'}$, without ever making that molecule or knowing any of the masses or the force constant? The product rule tells us yes! The underlying relationship between frequency and reduced mass leads to a startlingly simple connection:

This result is a little piece of magic. It's a direct consequence of the fact that the reciprocal of the reduced mass, $\frac{1}{\mu} = \frac{1}{m_X} + \frac{1}{m_Y}$, is additive. All the messy details about bond strengths and atomic masses cancel out, leaving a pure, clean relationship between the observable frequencies themselves.

Moving from a diatomic to a polyatomic molecule is like going from a single violin to a full orchestra. A molecule like water ($H_2O$) or ammonia ($NH_3$) doesn’t just have one way to vibrate. It has multiple, complex vibrational patterns called ​​normal modes​​, where all atoms move in-phase with the same frequency. How do we even begin to describe this?

The definitive mathematical framework for this is ​​Wilson's FG matrix method​​. In this picture, the vibrational problem is broken into two parts:

1. The ​​F matrix​​, or force constant matrix. This is the ​​potential energy​​ part of the problem. It contains all the information about the stiffness of the bonds and the energy it costs to bend the angles between them. It is our "music sheet," defining the harmony of the molecule. Because it's a feature of the electronic potential energy surface, the F matrix is ​​unchanged​​ by isotopic substitution.
2. The ​​G matrix​​, or inverse kinetic energy matrix. This is the ​​kinetic energy​​ part of the problem. It depends on the geometry of the molecule (bond lengths and angles) and, crucially, the masses of the atoms. This is our "orchestra," with each atom being a musician whose mass influences how it can move. The G matrix ​​changes​​ with isotopic substitution.

The vibrational frequencies ($\omega_k$) are found by solving the secular equation, $\det(\mathbf{GF} - \lambda \mathbf{I}) = 0$, where $\lambda_k = \omega_k^2$. This equation essentially says that the final "music" (the frequencies) arises from the interplay between the "music sheet" ($\mathbf{F}$) and the "orchestra" ($\mathbf{G}$).

One of the fundamental properties of determinants is that $\det(\mathbf{AB}) = \det(\mathbf{A})\det(\mathbf{B})$. Therefore, the product of all the eigenvalues of $\mathbf{GF}$ is simply $\det(\mathbf{GF}) = \det(\mathbf{G})\det(\mathbf{F})$. This gives us a profound result:

Now, consider what happens when we compare a molecule to its isotopologue (denoted by a prime). We can write the same equation for the isotopologue: $\prod_{k} (\omega'_k)^2 = \det(\mathbf{G'})\det(\mathbf{F'})$. Since the F matrix doesn't change ($\mathbf{F} = \mathbf{F'}$), if we take the ratio of these two product equations, the determinant of the force constant matrix—that complicated, difficult-to-determine "music sheet"—miraculously cancels out!

This is the heart of the Redlich-Teller product rule. It states that the ratio of the products of the squared frequencies is equal to the ratio of the determinants of their G-matrices. We have traded a problem involving unknown forces for one that depends only on mass and geometry.

The story gets even better. Most molecules have some form of symmetry. For a bent $XY_2$ molecule like water ($C_{2v}$ point group), the vibrations can be classified into different ​​symmetry species​​. For instance, there are vibrations where the molecule stretches and bends symmetrically ($A_1$ symmetry) and a vibration where one bond stretches while the other compresses ($B_2$ symmetry). This symmetry allows us to break the large $\mathbf{G}$ and $\mathbf{F}$ matrices down into smaller, independent blocks for each symmetry species. The product rule can then be applied to each block separately!

Let's look at the symmetric ($A_1$) vibrations of a bent $XY_2$ molecule, like in problem. The G-matrix for this symmetry block contains terms that depend on the reciprocal atomic masses (i.e., $1/m_X$ and $1/m_Y$), the bond length ($R$), and the bond angle ($2\alpha$). The expressions for the matrix elements look rather complicated. But when we calculate the determinant of this $2 \times 2$ matrix, something wonderful happens. All the terms containing the bond angle $\alpha$ cancel out, leaving an astonishingly simple result for its mass dependence: $\det(\mathbf{G}^{(A_1)}) \propto \frac{1}{m_Y}(\frac{1}{m_Y} + \frac{2}{m_X})$. The resulting ratio of frequency products is therefore independent of the molecular geometry for this symmetry block! (See also. The same holds true for the $A_1$ vibrations of a pyramidal $XY_3$ molecule like ammonia. Applying the product rule for an isotopologue $X'Y_2$:

Again, we have a precise prediction about observable frequencies that depends only on the atomic masses. This power to isolate the effect of mass from the intricacies of bonding and geometry makes the product rule an invaluable tool for physicists and chemists to verify their assignments of vibrational spectra. If the experimental frequency shifts upon isotopic substitution don't match the prediction of the product rule, the assignment is likely wrong.

The G-matrix determinant method is powerful, but it requires some heavy mathematical lifting. Is there a more direct way? Edgar Teller and Otto Redlich provided an even more general and magnificent formulation of the rule. Their theorem directly relates the frequency product ratio to fundamental properties of the molecule: its total mass, its moments of inertia, and the masses of its atoms.

The full ​​Teller-Redlich product rule​​ for a single symmetry species $\Gamma$ is a masterpiece of physical insight. It states that the ratio of the frequency products is connected to the ratio of mass products, but also to how the overall translation and rotation of the molecule are affected by the mass change.

For any given symmetry species, the vibrations belonging to it are, by definition, "uncontaminated" by overall translations and rotations that belong to other symmetry species. The theorem accounts for this by including terms for the total mass $M$ and the principal moments of inertia $I_x, I_y, I_z$. The full rule looks formidable, but its logic is sound. It essentially "corrects" for the fact that some of the degrees of freedom for a set of atoms contribute to the molecule's overall translation and rotation rather than to its internal vibration.

Let's see this in action for the asymmetric stretch ($B_2$ symmetry) of our bent $XY_2$ molecule. For this symmetry, the overall translation of the molecule along one axis ($T_y$) and rotation about another axis ($R_x$) also transform as $B_2$. The theorem cleverly uses the total mass $M$ and the moment of inertia $I_x$ to account for these motions. After a beautiful derivation where the ratios of total mass and moment of inertia are calculated, they combine with the atomic mass ratios in such a way that many terms cancel, yielding a final, clean expression for the frequency ratio:

Unlike the symmetric modes, this result depends on the bond angle $\alpha$. This makes physical sense: the asymmetric stretch involves the atoms moving in a way that is intimately tied to the molecular shape and its rotational properties. The theorem beautifully captures this connection. It demonstrates a deep unity in molecular motion, linking internal vibrations with the gross translation and rotation of the entire molecule through the elegant language of symmetry. When we apply the rule to all the vibrations of a molecule, as seen for $BF_3$, these individual pieces combine into a "complete" product rule that provides a comprehensive check on the entire vibrational spectrum.

From a simple spring to a full orchestral symphony governed by the laws of symmetry, the Redlich-Teller product rule stands as a testament to the elegant and often surprisingly simple relationships that underpin the complex world of molecular dynamics. It allows us, with nothing more than a knowledge of mass, to predict and understand the intimate dance of atoms.

After our journey through the principles and mechanisms of the Redlich-Teller product rule, you might be tempted to file it away as a specialized tool for the vibrational spectroscopist—a clever but narrow formula for cross-checking frequency assignments. And it is certainly that. But to leave it there would be like admiring the gearwork of a single clock without realizing it's part of a grand astronomical observatory. The true beauty of a deep physical principle lies not in its specificity, but in its reach. The product rule is a spectacular example, a single thread that weaves through the vast and colorful tapestry of chemistry and physics, connecting phenomena that, at first glance, seem to have nothing to do with one another.

Let's see just how far this thread takes us.

The most immediate home for the product rule is, of course, in deciphering the symphony of molecular vibrations. A molecule, after all, is not a static collection of balls and sticks; it is a dynamic entity, constantly ringing with a unique set of vibrational frequencies—its molecular "music." When a chemist uses infrared or Raman spectroscopy, they are essentially listening to this music to deduce the molecule's structure and bonding.

Isotopic substitution is one of the most powerful techniques in this endeavor. By swapping an atom for a heavier isotope—say, a hydrogen for a deuterium—we change the mass of one of the players in the vibrational orchestra. All the frequencies will shift, but how? A-ha, says the Redlich-Teller rule, not in some arbitrary way! The shifts are coupled. For a given symmetry of motion, the product of all the new frequencies divided by the product of the old ones is fixed by a precise relationship involving the changes in mass and moments of inertia.

Imagine trying to analyze the vibrations of a water molecule, $H_2O$. It has a symmetric stretch, an asymmetric stretch, and a bending motion. Now, you make some heavy water, $D_2O$. You can measure the new frequencies, but you want to be sure you've assigned them correctly. The symmetric stretch and the bending motion both belong to the same symmetry class. The product rule tells you that the ratio of their frequencies between $H_2O$ and $D_2O$ are intertwined. If you have a good guess for the shift in the stretching frequency, the rule nails down what the shift in the bending frequency must be. This provides a wonderfully rigid check on your analysis.

This idea scales up to molecules of far greater complexity. Consider the beautiful, symmetric benzene molecule, $C_6H_6$. What happens if we replace three alternating hydrogens with deuterium to make 1,3,5-trideuteriobenzene? The molecule's symmetry drops from the highly-ordered $D_{6h}$ to a less-symmetric $D_{3h}$. The product rule, working in concert with the powerful mathematics of group theory, can still make astonishingly precise predictions about how the products of certain vibrational frequencies will shift, boiling down a complex rearrangement of dozens of vibrations to a simple, elegant ratio of atomic masses. It can even be applied to the intricate world of coordination chemistry, helping to confirm the structure of an octahedral metal complex by predicting how its infrared spectrum will change upon full isotopic substitution of the metal and its ligands.

The rule's utility doesn't stop at vibrational spectra. It reaches into the world of rotational spectroscopy, which measures the moments of inertia of a molecule with incredible precision. For a perfectly rigid, planar molecule, the moments of inertia should obey the relation $I_c = I_a + I_b$. But real molecules are never perfectly rigid; they are always vibrating. The constant tremor of out-of-plane vibrations gives rise to a small but measurable "inertial defect," $\Delta = I_c - I_a - I_b \neq 0$. This defect is a direct consequence of the molecule's zero-point vibrational energy. So, if we substitute an isotope, say the oxygen in formaldehyde, $H_2CO$, the out-of-plane vibrational frequency changes. The Redlich-Teller product rule tells us exactly how it changes, which in turn allows us to predict the resulting change in the inertial defect. A rule for vibrations has just told us something profound about a molecule's effective rotational structure. It connects the fast-paced world of vibrations to the much slower dance of molecular rotation.

Now, let's take a bold leap away from spectroscopy and ask a question from thermodynamics. If you take a tank of pure hydrogen gas, $H_2$, and mix it with a tank of pure deuterium gas, $D_2$, and wait a very long time (perhaps with a little help from a catalyst), what will you find in the tank? Will it just be a mixture of $H_2$ and $D_2$, or will they react to form HD? And if they react, what is the final equilibrium state? The reaction is: $\text{H}_2 + \text{D}_2 \rightleftharpoons 2\text{HD}$ This is a question about the equilibrium constant, $K_P$. Statistical mechanics gives us the answer: at high temperatures, the equilibrium constant is determined almost entirely by the molecular partition functions, which are a measure of all the accessible energy states—translational, rotational, vibrational, and electronic.

So, we write down the expression for the equilibrium constant, $K_P = \frac{q_{\text{HD}}^2}{q_{\text{H}_2} q_{\text{D}_2}}$. We diligently calculate the ratios of partition functions for translation (which depends on total mass), rotation (which depends on the moment of inertia), and vibration (which depends on the vibrational frequency). Each one gives a complicated expression involving the masses of hydrogen ($m_H$) and deuterium ($m_D$). It looks like a terrible algebraic mess.

But then, a miracle happens. When you multiply all these factors together, the mass dependencies—every last $m_H$ and $m_D$—cancel out completely! The final result is astonishingly simple: $K_P = 4$.

Why this magical cancellation? It's no magic at all. It is the Redlich-Teller product rule in disguise! The relationship between the masses, moments of inertia, and vibrational frequencies is precisely what the product rule dictates. The rule is the deep physical reason that guarantees this cancellation. It shows that, putting aside the symmetry of the molecules, nature doesn't fundamentally care whether the mass is distributed as ($H_2, D_2$) or (HD, HD). The only thing that matters in this high-temperature limit is the change in symmetry. The reactants, $H_2$ and $D_2$, are symmetric, with a symmetry number $\sigma = 2$. The product, HD, is not, with $\sigma = 1$. The equilibrium constant is simply the ratio of these symmetry numbers: $K_P = \frac{\sigma_{\text{H}_2}\sigma_{\text{D}_2}}{\sigma_{\text{HD}}^2} = \frac{2 \times 2}{1^2} = 4$. The reaction proceeds strongly to the right not because of energy, but because of entropy—there are simply more ways to be HD than to be $H_2$ or $D_2$. The product rule was the key that unlocked this profound and simple truth, by showing us that all the other mass-dependent effects were a wash.

If thermodynamics tells us where a reaction is going, kinetics tells us how fast it gets there. Here too, isotopic substitution provides one of the most powerful tools for understanding reaction mechanisms: the Kinetic Isotope Effect, or KIE. A reaction involving the breaking of a C-H bond is almost always significantly faster than the identical reaction involving a C-D bond. The KIE is the ratio of these rates, $k_H/k_D$, and its value can be a dead giveaway as to what's happening at the moment of reaction—the transition state.

According to Transition State Theory, this ratio can be broken down into several factors, including contributions from changes in vibrational energy, rotational and translational energy, molecular symmetry, and even quantum mechanical tunneling. The largest piece of the puzzle is usually the vibrational part, specifically the difference in zero-point energy (ZPE). The C-H bond vibrates at a higher frequency than the C-D bond, so it has a higher ZPE. When the bond breaks in the transition state, this vibrational energy is lost. Since the C-H bond started with more ZPE, the net energy cost to reach the transition state is lower for the hydrogen-containing molecule, so it reacts faster.

But what about the other parts of the KIE? What about the ratio of rotational and translational partition functions? Once again, we find our old friend at work. Just as in the $H_2/D_2$ equilibrium problem, the mass-dependent factors from translation and rotation largely cancel out between the reactant and the transition state. This cancellation, which is formally described by the product rule, ensures that this part of the KIE is typically very close to 1. This is incredibly useful! It means experimental chemists can measure a KIE and be confident that any large value (say, a factor of 7) is not due to some mundane classical change in rotation or translation. Instead, it must be due to the interesting quantum effects: the change in zero-point vibrational energy or the even more exotic phenomenon of tunneling, where the light hydrogen atom "cheats" and passes through the energy barrier instead of over it. The product rule, by accounting for the classical mass effects, helps us clear away the fog to see the beautiful quantum landscape underneath.

From checking spectral lines to predicting equilibrium constants and dissecting reaction rates, the Redlich-Teller product rule reveals itself to be far more than a simple formula. It is a statement about the fundamental interplay of mass, symmetry, and geometry that governs the behavior of all molecules. It is a brilliant illustration of the unity of a subject, showing how a single, elegant idea can illuminate some of the deepest corners of the physical world.