Einstein's A and B coefficients

The dance between light and matter is a fundamental process that governs everything from the glow of distant stars to the technologies that power our modern world. For a time, our understanding was incomplete, based on just two processes: atoms absorbing light to jump to higher energy states and spontaneously emitting light to fall back down. However, this simple picture could not explain how matter and radiation could coexist in perfect thermal balance. In 1917, Albert Einstein identified a missing piece of the puzzle, a third process that would not only solve this paradox but also lay the theoretical groundwork for one of the 20th century's most transformative inventions. This article delves into Einstein's profound insights. The following chapters will explore the fundamental principles of the three core processes of light-matter interaction and the far-reaching applications of this powerful framework.

Imagine an atom, a tiny solar system of electrons orbiting a nucleus. This system isn't static; it lives and breathes by interacting with the universe around it, primarily by absorbing and emitting light. But how exactly does this dance between matter and light work? At first glance, it seems simple. An atom in a low-energy "ground" state can swallow a photon of precisely the right energy to jump to a higher "excited" state—a process we call ​​absorption​​. What goes up must come down. After a while, the excited atom will, of its own accord, spit out a photon and fall back to the ground state. This is ​​spontaneous emission​​, a random act both in timing and in the direction of the emitted light. It’s the process that makes stars shine and fluorescent signs glow.

For decades, these two processes were thought to be the whole story. But in 1917, a young Albert Einstein, while pondering the nature of thermal equilibrium, realized something was missing. He performed a thought experiment of beautiful simplicity, the essence of which we can explore. Imagine a sealed box filled with atoms and light, all at a constant temperature. In this cozy equilibrium, an unceasing exchange of energy takes place. For every atom that absorbs a photon and jumps up, another atom must fall down, emitting a photon, to keep the populations of the energy levels and the intensity of the light constant.

Einstein realized that if absorption and spontaneous emission were the only players, the game was rigged. At higher temperatures, there's more light, so absorption would happen faster. But spontaneous emission is an internal property of the atom; it doesn't care how much light is around. The rates wouldn't balance correctly against Planck's well-established law for blackbody radiation. The system could never reach equilibrium. To solve this paradox, Einstein postulated a third process, a new kind of interaction: ​​stimulated emission​​.

In this process, a photon of the correct energy doesn't get absorbed by an excited atom. Instead, it tickles the atom, inducing it to fall to the ground state and release a second photon. The magic is that this new photon is a perfect clone of the first: it has the same frequency, the same direction, the same phase, and the same polarization. One photon goes in, two identical photons come out. A beam of light passing through a collection of such excited atoms would be amplified, not diminished. This was the theoretical seed of the laser, planted decades before one could be built.

To put these ideas on solid ground, Einstein introduced a set of phenomenological constants, now known as the ​​Einstein A and B coefficients​​. They are the rulebook for the three-part dance of light and matter.

• ​​Spontaneous Emission (The A Coefficient):​​ The rate at which an excited atom in state 2 decays to state 1 on its own is given by $R_{\text{spon}} = A_{21} N_2$, where $N_2$ is the population of excited atoms. The coefficient $A_{21}$ is simply a probability per unit time. If you have a collection of excited atoms, $A_{21}$ tells you what fraction of them will decay each second. Its inverse, $\tau = 1/A_{21}$, is the average ​​lifetime​​ of the excited state. A "forbidden" transition, often discussed in astrophysics, is simply one with a very small $A_{21}$ and therefore a very long lifetime. For instance, an observed lifetime of half a second for a particular nebula transition immediately tells us that $A_{21} = 2.00 \text{ s}^{-1}$, a value many orders of magnitude smaller than typical atomic transitions which have lifetimes on the order of nanoseconds.

• ​​Absorption and Stimulated Emission (The B Coefficients):​​ These two processes depend on the presence of external light. Their rates are proportional to the spectral energy density of the radiation field, $\rho(\nu)$, at the transition frequency $\nu$.

  • The rate of absorption is $R_{\text{abs}} = B_{12} \rho(\nu) N_1$.
  • The rate of stimulated emission is $R_{\text{stim}} = B_{21} \rho(\nu) N_2$.

The $B$ coefficients are a measure of how strongly the atom couples to the light field. A large $B$ coefficient means the atom is very susceptible to being stimulated, either to absorb a photon and jump up, or to emit one and fall down. From a dimensional standpoint, since $\rho(\nu)$ has units of energy-per-volume-per-frequency (in SI, $\text{kg} \cdot \text{m}^{-1} \cdot \text{s}^{-1}$), the B coefficient must have units that convert this into a rate ($\text{s}^{-1}$), which turns out to be $\text{m} \cdot \text{kg}^{-1}$.

Here is where Einstein's genius shines brightest. These coefficients are not independent. The atom's intrinsic desire to decay ($A_{21}$) and its susceptibility to being tickled by light ($B_{21}$) are deeply connected. By insisting that his three processes must perfectly reproduce Planck's law for blackbody radiation in that sealed box, Einstein found a stunningly simple and profound relationship:

This equation is a cornerstone of quantum optics. It reveals a fundamental unity. The properties of an atom ($A_{21}$, $B_{21}$) are inextricably linked to the properties of the vacuum and the light that propagates through it ($h$, $c$), and to the very energy of the transition ($\nu$). The powerful $\nu^3$ dependence tells us something remarkable: at very high frequencies (like UV or X-rays), the ratio $A/B$ becomes enormous. This means spontaneous emission overwhelmingly dominates stimulated emission, which is a major reason why building an X-ray laser is profoundly more difficult than an infrared one.

This relationship allows us to ask fascinating questions. For instance, at what temperature does the rate of stimulated emission equal the rate of spontaneous emission? For this to happen, $B_{21}\rho(\nu, T)$ must equal $A_{21}$. Using the connection we just found, this implies that $\rho(\nu, T) = \frac{8\pi h \nu^3}{c^3}$. Plugging this into Planck's law, we can solve for the temperature and find $T = \frac{h\nu}{k_B \ln(2)}$. For a typical visible light transition, this temperature is thousands of Kelvin. This tells us that under most "normal" conditions on Earth, spontaneous emission is by far the dominant decay path. To make stimulated emission important, you need an enormous density of photons, far beyond what thermal equilibrium can provide.

So, when does a material absorb light, and when does it amplify it? The answer lies in the competition between absorption ($N_1 B_{12} \rho(\nu)$) and stimulated emission ($N_2 B_{21} \rho(\nu)$). Light is amplified if the rate of stimulated emission exceeds the rate of absorption.

A subtle point arises from the degeneracies of the energy levels, which are the number of distinct quantum states that share the same energy. Let the ground state have degeneracy $g_1$ and the excited state have $g_2$. The B coefficients are related by $g_1 B_{12} = g_2 B_{21}$. This reflects a principle of detailed balance: the intrinsic probability of the light-matter interaction is the same in both directions.

For a beam of light passing through the material to be perfectly transparent—neither amplified nor attenuated—the rates must be equal:

Using the relation between the B coefficients, this simplifies to a condition on the populations alone:

This ratio is the tipping point. If $\frac{N_2}{N_1} \frac{g_2}{g_1}$, as is almost always the case in nature (atoms prefer lower energy states), absorption wins and the material attenuates light. To get amplification, we need to achieve the opposite, an unnatural condition known as ​​population inversion​​:

Creating a population inversion is the central challenge in building a laser. You must force more atoms into the excited state than the ground state (or at least more than this critical ratio).

This framework beautifully explains the difference between an LED and a laser. We can define a "coherence metric" $\eta$ as the fraction of total emissions that are stimulated: $\eta = \frac{R_{\text{stim}}}{R_{\text{spon}} + R_{\text{stim}}}$. Using the A/B relation, this can be written as $\eta = \frac{\rho_0}{\rho_0 + A/B}$. In an LED, we create excited states that decay mostly on their own. The energy density $\rho_0$ is small, so $\eta \approx 0$ and the light is incoherent. In a laser, we use an optical cavity to trap photons, building up a colossal energy density $\rho_0$. As $\rho_0$ becomes much larger than $A/B$, $\eta \to 1$. Stimulated emission dominates, and the output is a perfectly coherent beam.

Can we create a population inversion just by shining really bright light on a simple two-level system? Let's try. As we increase the intensity of our pumping light, $\rho(\nu)$, the absorption rate ($B_{12} \rho(\nu) N_1$) goes up, pushing atoms into the excited state. But as $N_2$ increases, the rate of stimulated emission ($B_{21} \rho(\nu) N_2$) also goes up, pushing them back down! The system reaches a steady state where the upward and downward rates balance. In the limit of infinite pumping power, the spontaneous emission term $A_{21}$ becomes negligible, and the balance is purely between absorption and stimulated emission, leading to the transparency condition we found earlier: $\frac{N_2}{N_1} \to \frac{g_2}{g_1}$. You can never push past it. The maximum fraction of atoms you can get into the excited state is $\frac{g_2}{g_1+g_2}$, which for equal degeneracies is only 0.5. You can saturate the transition, but you can't invert it. This is why practical lasers require more clever schemes, using three or four energy levels to separate the pumping process from the laser transition.

Finally, what determines the values of A and B for a given atom? They are not arbitrary. Quantum mechanics tells us they are determined by the ​​transition dipole moment​​, $|\vec{\wp}_{21}|$, a quantity that measures how much the atom's electron cloud "sloshes" as it transitions between the two states. It turns out that both $B_{12}$ and $B_{21}$ are directly proportional to $|\vec{\wp}_{21}|^2$. And since $A_{21}$ is proportional to $B_{21}$, it must also be proportional to $|\vec{\wp}_{21}|^2$.

This insight explains many things. If we could engineer an atom to double its transition dipole moment, the B coefficient would increase by a factor of 4, and the A coefficient would also increase by a factor of 4. It also gives a physical meaning to "forbidden" transitions. A transition is forbidden not because it's impossible, but because the symmetry of the electron wavefunctions makes the transition dipole moment zero. Higher-order, much weaker interactions, like the electric quadrupole moment (which scales differently with atomic size), can still cause the transition, but the resulting A and B coefficients will be minuscule. This fundamental connection also means that if you can measure a quantity related to absorption (like the total absorption cross-section, $\Sigma$), you can directly calculate the atom's spontaneous emission lifetime, $\tau$, further cementing the inescapable unity between these seemingly disparate processes.

From a simple puzzle about thermal equilibrium, Einstein uncovered a deep and beautiful set of rules governing all of light-matter interaction, paving the way for one of the most transformative technologies of the 20th century. The A and B coefficients are not just abstract parameters; they are the language in which the universe describes the ceaseless, elegant dance between atoms and light.

Now that we have acquainted ourselves with the intricate machinery of Einstein’s $A$ and $B$ coefficients, you might be tempted to think of them as a quaint piece of theoretical physics—a clever solution to an old puzzle about a box of atoms and light. But to leave it there would be like discovering the alphabet and never reading a book. These coefficients are not a destination; they are a passport. They grant us access to an astonishingly diverse range of phenomena, from the heart of our most advanced technologies to the far-flung cosmos, and even to the very nature of spacetime itself. The simple rules governing spontaneous emission, absorption, and stimulated emission form a universal language describing the dance between matter and energy. Let us now see where this passport takes us.

Perhaps the most celebrated child of Einstein's coefficients is the laser. The name itself—Light Amplification by Stimulated Emission of Radiation—is a direct homage to the $B_{21}$ coefficient. The magic of a laser lies in its ability to create a cascade of perfectly cloned photons, all marching in step with the same frequency, phase, and direction. This is the work of stimulated emission. An incoming photon encounters an excited atom and, instead of being absorbed, tempts the atom to release a second, identical photon.

But here we encounter a problem. In any system left to its own devices at some temperature, there will always be more atoms in the lower energy state than the upper one. As we saw when first deriving the coefficients, nature prefers the path of least energy. This means that for any incoming photon, it's far more likely to be absorbed by a ground-state atom than to stimulate emission from an excited one. The result is net absorption, not amplification. A system in thermal equilibrium is a light-eater, not a light-creator.

To build a laser, we must cheat. We have to force the system out of equilibrium by "pumping" it with energy so furiously that, for a moment, we achieve the unnatural state known as a ​​population inversion​​. This doesn't just mean having more atoms in the excited state than the ground state ($N_2 > N_1$). The real condition is more subtle: the population per available quantum substate must be higher in the upper level. That is, we need $\frac{N_{2}}{g_{2}} \gt \frac{N_{1}}{g_{1}}$, where $g$ is the degeneracy of the level. Only when this condition is met can the rate of stimulated emission, $N_2 B_{21}$, finally overwhelm the rate of absorption, $N_1 B_{12}$. Once this threshold is crossed, the chain reaction begins. A single photon begets two, two beget four, and an avalanche of coherent light is born. Every laser pointer, every fiber-optic cable, every surgical scalpel made of light is a testament to this engineered victory of stimulated emission over absorption.

While engineers were busy building lasers, chemists and biologists realized that the same coefficients provide a powerful toolkit for eavesdropping on the private lives of molecules. The rate of spontaneous emission, the $A$ coefficient, determines the characteristic lifetime of an excited state—how long, on average, a molecule will "glow" after being excited. This is not just an abstract number; it's a measurable property that tells us about the molecule's structure and environment.

Consider modern marvels like quantum dots, which are semiconductor nanocrystals used as fluorescent markers in biological imaging. By carefully tuning their size, scientists can make them glow in any color of the rainbow. When a quantum dot has multiple pathways to decay back to a lower energy state, its total lifetime is determined by the sum of all the individual spontaneous emission rates. A shorter lifetime implies a "brighter" or more efficient fluorophore, a tiny lantern for peering into the machinery of a living cell.

But we can go deeper. The interaction is a two-way street. The strength of absorption, governed by the $B$ coefficient, tells us how likely a molecule is to absorb light of a certain frequency. In introductory chemistry, we often model molecular bonds as perfect springs obeying Hooke's Law—a simple harmonic oscillator. For such an ideal oscillator, the strength of transitions up the "vibrational ladder" would follow a simple, predictable pattern. But real molecules are not so simple. Their potential energy wells are anharmonic. This slight deviation from the perfect spring model means that the strength of the transition from the ground state to the first excited state ($v=0 \to v=1$) is not the same as from the first to the second ($v=1 \to v=2$). By precisely measuring the ratio of the corresponding Einstein $B$ coefficients, we can quantify this anharmonicity, giving us a far more accurate picture of the true nature of chemical bonds.

Let us now turn our gaze from the microscopic to the cosmic. The universe is the ultimate laboratory for light-matter interactions. In the frigid expanse of interstellar space, an excited atom might float for years before spontaneously emitting a photon. But inside a star, the situation is vastly different. The star's interior is a seething furnace, a dense plasma filled with a brilliant thermal radiation field. Here, an excited atom is constantly bombarded by photons. Stimulated emission becomes a major pathway for decay, significantly shortening the atom's excited-state lifetime. The total decay rate becomes a function not just of the atom's intrinsic properties ($A_{21}$), but also of the ambient temperature of its environment.

Einstein's framework is not even limited to electrons bound within atoms. In the hot plasma of a star's atmosphere, free electrons and ions are constantly interacting. A free electron zipping past an ion can absorb a photon, converting the photon's energy into its own kinetic energy. This process, known as inverse bremsstrahlung or free-free absorption, is a crucial source of opacity in stars, determining how energy is transported from the core to the surface. Though there are no discrete energy levels, we can still define an effective, thermally-averaged Einstein $B$ coefficient to describe this continuum absorption process, revealing the deep versatility of the original concept.

So far, our story has been one of atoms and photons. But is that the only dance in town? What if an excited system can decay by emitting a different kind of quantum? In a magnetic material, the collective spins of atoms can form waves, and the quantum of this wave is a particle called a magnon. A spin impurity embedded in such a material can relax from an excited state to a ground state by emitting not a photon, but a single magnon.

Amazingly, the relationship between spontaneous and stimulated emission of magnons follows the same logic as it does for photons. By considering thermal equilibrium between the impurity and a gas of magnons, we can derive a ratio $A/B$. However, this ratio looks different from the one for photons. It depends not on the speed of light, but on the properties of magnons, specifically how their energy relates to their momentum (their dispersion relation). This is a profound revelation: the $A/B$ ratio is not a property of the atom alone, but a property of the quantum field with which it interacts. It's a measure of the phase space available for a new particle to be born into.

This idea—that the $A/B$ ratio depends on the structure of the surrounding space—can be explored with a delightful thought experiment. What if we lived in a two-dimensional universe, a "Flatland"? The number of available modes for a photon of a given frequency to be emitted into would be different than in our 3D world. If we repeat Einstein's original derivation using the 2D formula for blackbody radiation, we find a new $A/B$ ratio!. This proves that the familiar factor of $\nu^3$ in the 3D relation is a direct fingerprint of the three-dimensional nature of our space.

The dance of emission and absorption has one final, subtle consequence that brings us to the forefront of modern physics. It doesn't just change the energy of an atom; it can destroy its quantum nature. An atom can exist in a superposition—a delicate state of being in both the ground and excited states at once. This property, called coherence, is the resource that would power a quantum computer. However, every time a random photon is absorbed, or a photon is spontaneously or stimulatedly emitted, it's like a random "measurement" of the atom's state. These interactions cause the superposition to decay, a process called decoherence. The total rate of this decoherence can be expressed directly in terms of the Einstein A coefficient and the number of thermal photons in the environment, linking the fundamental light-matter processes to the central challenge of building quantum technologies.

As a final, mind-bending crescendo, consider the very nature of empty space. The vacuum of quantum field theory is not empty; it is a roiling sea of virtual particles. Now, what happens if an atom is not at rest, but is uniformly accelerating through this vacuum? According to the Unruh effect, the accelerating observer perceives this vacuum as a warm, thermal bath of radiation. The temperature of this bath is directly proportional to the acceleration.

This means that an atom that would be perfectly stable in its ground state in an inertial frame can suddenly find itself in a warm environment and absorb a particle from the "hot" vacuum, transitioning to its excited state! This is a form of spontaneous excitation, induced by acceleration. And how do we calculate its rate? We use the same old machinery. We take the atom's Einstein $B$ coefficient for absorption and multiply it by the energy density of the Unruh thermal bath. The result is a concrete, predictable rate for an atom to be excited by nothing more than its own acceleration.

From a box of gas to the theory of everything, Einstein's simple coefficients have provided the key. They showed us how to build a laser, how to understand the stars, how to probe the chemical bond, and finally, how the very concepts of vacuum and particle depend on our state of motion. They are one of the most beautiful examples of a simple physical intuition unlocking a profound and universal truth about the workings of our world.