Stokes' Hypothesis

When we think of viscosity, we often picture the thick, slow flow of honey—a simple resistance to being smeared or sheared. This intuitive concept, characterized by a single coefficient of shear viscosity, seems to capture the "stickiness" of fluids in our everyday world. But is this the complete picture? Does a fluid also resist being uniformly compressed or expanded, and if so, how does that change our fundamental understanding of fluid motion? This question reveals a crucial gap between a surface-level perception and a more profound physical reality.

This article delves into the heart of this question by exploring Stokes' hypothesis, an elegant and powerful simplification that has shaped the field of fluid dynamics for over a century. To build a comprehensive understanding, we will proceed in two parts. First, the chapter on ​​Principles and Mechanisms​​ will unpack the theoretical and physical basis for a second type of friction—the bulk viscosity—and explain how Sir George Stokes' brilliant postulation seeks to eliminate it, revealing where this assumption succeeds and where it fails. Subsequently, the chapter on ​​Applications and Interdisciplinary Connections​​ will journey through the vast practical consequences of this hypothesis, demonstrating its role in everything from enabling modern engineering simulations to explaining the behavior of sound waves and the structure of supersonic shock fronts.

Let’s begin our journey with a simple, everyday experience: pouring honey. You know instinctively that it’s “thicker” or more “viscous” than water. If you were to spread it on a piece of toast, you’d feel a resistance to the motion of your knife. This resistance to being smeared, or more technically, to being sheared, is the most familiar face of viscosity.

Physicists in the time of Newton captured this idea with beautiful simplicity. Imagine two parallel plates with a layer of fluid between them. If you slide the top plate while keeping the bottom one still, the fluid sticks to both plates. The layer next to the moving plate moves along with it, the layer at the bottom stays put, and the fluid in between is sheared. The force you need to apply to the top plate is proportional to how fast you move it. The resulting internal friction in the fluid is called ​​shear stress​​, and the property of the fluid that relates this stress to the rate of shearing is called the ​​dynamic viscosity​​, or more commonly, the ​​shear viscosity​​. It is universally denoted by the Greek letter $\mu$ (mu). For a simple shear flow, this relationship is wonderfully linear: the stress is just $\mu$ times the velocity gradient. This single constant, $\mu$, seems to perfectly describe the "stickiness" of fluids like water, oil, and honey in our daily experience.

But is that the whole story? Is resistance to shearing the only kind of internal friction a fluid can have? This question opens a door to a much richer and more subtle understanding of fluid motion.

What happens if a fluid isn't being sheared, but is being compressed or expanded uniformly, like a sponge being squeezed or a gas cloud expanding into space? Does viscosity play a role then? This is a much less intuitive idea. We can shear honey, but how do you "thicken" the act of compression itself?

To get a handle on this, we must upgrade our tools. The simple idea of a single stress is not enough when a fluid can be twisted, sheared, and compressed all at once. We need a more powerful mathematical object: the ​​stress tensor​​, denoted by $\boldsymbol{\sigma}$. Think of it as a complete switchboard of forces within the fluid. It tells us the force in any direction acting on any surface at any point.

For a fluid at complete rest, things are simple. The stress is just the familiar, everyday pressure, $p$, which pushes equally in all directions. In the language of tensors, we write this as $\boldsymbol{\sigma} = -p\boldsymbol{I}$, where $\boldsymbol{I}$ is the identity tensor (a matrix of ones on the diagonal and zeros elsewhere). The minus sign is a convention; we think of pressure as a force pushing inward on a fluid element.

Now, let the fluid move. The total stress is the sum of this resting pressure and an additional part due to the motion—the ​​viscous stress tensor​​, $\boldsymbol{\tau}$. So, our total stress is $\boldsymbol{\sigma} = -p\boldsymbol{I} + \boldsymbol{\tau}$.

What does this viscous stress, $\boldsymbol{\tau}$, look like? By applying some very general principles—namely, that the fluid itself has no intrinsic preferred direction (​​isotropy​​) and that the laws of physics don't depend on how you're spinning or moving as an observer (​​frame indifference​​)—we arrive at a remarkable conclusion. For a vast class of fluids (the Newtonian fluids), the most general form of the viscous stress tensor requires two independent constants, not just one!

We can write this relation in a conceptually clear way:

Don’t be intimidated by the symbols! The equation tells a simple story. The first term, $2\mu\boldsymbol{D}'$, describes the stress that arises from the fluid changing its shape without changing its volume (this is shear, in its most general form). It is governed by our old friend, the shear viscosity $\mu$. The second term, $\zeta (\nabla \cdot \vec{v}) \boldsymbol{I}$, describes the stress that arises purely from the fluid changing its volume. The quantity $\nabla \cdot \vec{v}$ is the rate of volume expansion per unit volume. And the new coefficient, $\zeta$ (zeta), is a second type of viscosity called the ​​bulk viscosity​​ or ​​dilatational viscosity​​.

So, there it is! A fluid’s internal friction is, in general, characterized by two distinct properties: its resistance to being sheared ($\mu$) and its resistance to being compressed or expanded ($\zeta$).

Having two viscosity coefficients is, frankly, a bit of a nuisance. The bulk viscosity $\zeta$ is notoriously difficult to measure, far more so than $\mu$. For nearly a century, physicists and engineers wondered if they could somehow get rid of it. It was the great 19th-century physicist Sir George Gabriel Stokes who proposed an ingenious way out.

Stokes offered a beautifully simple physical argument. Imagine a tiny blob of fluid that is expanding or contracting uniformly in all directions, so its shape remains a perfect sphere as its volume changes. In such a pure "dilatation," there is no shearing. Stokes postulated that in this special kind of motion, the average force per unit area on the blob's surface—the ​​mechanical pressure​​—should be exactly equal to the fluid's ​​thermodynamic pressure​​, the same pressure ($p$) it would have if it were just sitting in a box at equilibrium.

In other words, he guessed that the act of pure expansion or compression, if uniform, shouldn't introduce any extra viscous pressure. The friction should only come from the rubbing and sliding motions of shear.

This seemingly modest postulate has a powerful mathematical consequence. It forces the bulk viscosity $\zeta$ to be exactly zero! This famous simplification, $\zeta=0$, is known as ​​Stokes' hypothesis​​. (In some older textbooks, the viscous stress tensor is written using a different constant, $\lambda$, where the hypothesis takes the form $\lambda = -\frac{2}{3}\mu$. This is mathematically equivalent to $\zeta=0$.

If this hypothesis holds, it means that viscous forces only do work when a fluid element is changing its shape. If a fluid element undergoes a pure, uniform expansion, the viscous forces are completely dormant and perform no work at all. For decades, this elegant simplification was built into the very foundation of fluid dynamics, the Navier-Stokes equations. But was Stokes' brilliant guess correct?

To answer whether Stokes was right, we cannot stay at the level of continuum mechanics. We must dive into the world of atoms and molecules. It turns out that a fluid's bulk viscosity is a direct window into the microscopic dance of its constituent particles.

​​A Monatomic Gas: The Ideal Case​​

Consider a gas made of single atoms, like helium or argon. You can think of these atoms as tiny, perfect billiard balls. The only form of internal energy they have is their translational kinetic energy—the energy of their random, zipping motion. When you compress this gas, the work you do is transferred almost instantaneously to this random motion through collisions. The system adapts to its new volume with incredible speed. There is no internal "sluggishness," no delay in redistributing the energy. Consequently, there is no dissipative friction associated with the volume change. Rigorous kinetic theory confirms this intuition: for a monatomic gas, the bulk viscosity $\zeta$ is, for all practical purposes, zero. Here, Stokes' hypothesis is a spectacular success.

​​Complex Fluids: The Reality of Relaxation​​

Now, let's look at a more complex gas, like carbon dioxide ($\text{CO}_2$). A $\text{CO}_2$ molecule is not a simple sphere. It can rotate, and its chemical bonds can vibrate like tiny springs. These rotational and vibrational states are a way for the molecule to store internal energy, in addition to its translational motion.

When you rapidly compress this gas, you first pump energy directly into the translational motion of the molecules—they start flying around faster. However, it takes a small but finite amount of time, a ​​relaxation time​​, for this extra energy to be shared and equilibrated with the rotational and vibrational modes. During this tiny lag, the gas is not in a state of local thermodynamic equilibrium. This process of delayed energy transfer is irreversible and dissipates energy, much like friction. This very dissipation is what we macroscopically measure as a non-zero bulk viscosity!

Therefore, for polyatomic gases, and indeed for most liquids, ​​Stokes' hypothesis fails​​. For water at room temperature, the bulk viscosity $\zeta$ is about three times the shear viscosity $\mu$. In more complex systems, the effect can be dramatic. For polymeric liquids or for fluids near their critical point (like $\text{CO}_2$ under high pressure), the internal relaxation mechanisms become so pronounced that the bulk viscosity can be thousands of times larger than the shear viscosity.

This "second viscosity" becomes profoundly important in phenomena involving very rapid compressions, where the fluid has no time to relax. The propagation of high-frequency sound waves, for instance, is heavily damped by bulk viscosity. The extreme compression within a shock wave traveling faster than sound is another domain where Stokes' hypothesis is completely inadequate, and the full theory, including $\zeta$, is essential.

There is a final, important piece to this puzzle. For a vast number of engineering and physics problems—like the flow of water in a pipe or air over a car at low speeds—the fluid's density hardly changes. We can model the flow as ​​incompressible​​, which is a kinematic constraint stating that the volume of any fluid parcel is constant. Mathematically, this means $\nabla \cdot \vec{v} = 0$.

Look back at our expression for the viscous stress. If $\nabla \cdot \vec{v} = 0$, then the entire term involving the bulk viscosity, $\zeta (\nabla \cdot \vec{v}) \boldsymbol{I}$, vanishes, regardless of the value of $\zeta$! This means that for incompressible flows, the bulk viscosity is completely irrelevant. The equations of motion simplify and depend only on the shear viscosity $\mu$.

This explains why Stokes' hypothesis was so successful for so long. For a huge class of everyday fluid flows, it simply doesn't matter whether the hypothesis is true or not. But it is a crucial mistake to think that because water is often modeled as incompressible, its bulk viscosity must be zero. The truth is merely that an experiment on incompressible flow is incapable of measuring it.

We end where we began, with pressure, but now with a far deeper appreciation. In the world of non-equilibrium fluid dynamics, we must distinguish between two kinds of pressure:

1. The ​​thermodynamic pressure​​, $p_{th}$, is the pressure you know from the ideal gas law, a state variable that relates to density and temperature.

2. The ​​mechanical pressure​​, $\bar{p}$, is the physical, average normal stress exerted by the fluid. It's what you would actually measure with a pressure gauge that tumbles along with the flow.

In perfect equilibrium, these two are identical. But as we've seen, when a fluid is being compressed or expanded, they can differ. Their relationship elegantly summarizes our entire discussion:

The difference between the pressure measured by a tiny instrument and the pressure predicted by thermodynamics is precisely the bulk viscous stress.

Stokes' hypothesis, then, can be seen in its most profound light as the assumption that the mechanical and thermodynamic pressures are always one and the same. It is an assumption of perpetual local equilibrium. We have seen that this is a beautiful and powerful approximation, true for the simplest of gases, but it breaks down in the rich and complex world of real molecules and rapid changes. The failure of Stokes' hypothesis is not a flaw in our theories; rather, it is a window into the fascinating, intricate dance of molecules working to find their balance in a world in constant motion.

In our previous discussion, we met Stokes' hypothesis not as a dry mathematical formula, but as a stroke of physical intuition—a brilliant guess about the hidden nature of friction within a fluid. We saw how it proposes a simple, elegant relationship for how a fluid resists being squeezed or stretched. Now, we embark on a journey to see where this idea takes us. It is one thing to have a neat theory, but the real test, and the real fun, is to see what it can do. Where does this seemingly small assumption about viscosity appear in the grand tapestry of the physical world? You may be surprised to find its signature written in the silence of a fading echo, the fiery glow of a spacecraft’s reentry, and even in the turbulent heart of a distant star.

Imagine you are tasked with predicting the weather, designing a quieter submarine, or modeling the flow of blood through an artery. Your first step is to write down the laws of nature that govern the fluid's motion. The fundamental principles are clear: conservation of mass (what goes in must come out) and Newton's second law, which for a fluid becomes the conservation of momentum.

This gives us a set of beautiful equations, but we quickly run into a frustrating problem. We have more unknown quantities—density $\rho$, pressure $p$, temperature $T$, internal energy $e$, and the three components of velocity $\mathbf{v}$—than we have equations to solve for them. It’s like trying to solve a Sudoku puzzle with too few starting numbers; there isn't a unique solution. This is the famous "closure problem" of fluid dynamics. The system is mathematically open, waiting for more physical information.

To close it, we need to specify how the fluid behaves. We need rules that connect stress to motion, and temperature to energy. These are the "constitutive relations." For a viscous fluid, the most critical of these is the law for the stress tensor, $\boldsymbol{\sigma}$. This tensor tells us about all the internal forces a fluid element exerts on its neighbors. A piece of it is the familiar pressure, $p$, which acts equally in all directions. The other part, the viscous stress $\boldsymbol{\tau}$, is more complex. It describes the shearing, stretching, and squeezing forces. For a vast class of common fluids—so-called Newtonian fluids—the viscous stress is proportional to the rate at which the fluid is being deformed. The most general form for this is:

Here, $\mu$ is the familiar shear viscosity that resists sliding motion, like honey spreading on toast. The term $\nabla \cdot \mathbf{v}$ represents the rate of volume expansion or compression. And the new character, $\zeta$, is the bulk viscosity, which describes an additional resistance to this change in volume. Measuring $\zeta$ is notoriously difficult. This is where Stokes' hypothesis enters, with its bold and simplifying proposition: that for many fluids, the bulk viscosity is effectively zero ($\zeta=0$). With this assumption, the viscous stress tensor takes on the clean and practical form we often see in textbooks:

By providing this missing piece of the puzzle, Stokes' hypothesis, combined with the energy equation and equations of state, "closes" the system. It gives us a complete and solvable set of rules: the compressible Navier-Stokes equations. This set of equations is the very foundation of Computational Fluid Dynamics (CFD), the powerful simulation tool that has revolutionized everything from aircraft design to cardiovascular medicine. Stokes' simple guess becomes a cornerstone of modern engineering.

Have you ever wondered why a sound fades away? You might think it's just because the energy spreads out as the sound wave expands, and that’s part of it. But there is a more subtle and interesting process at work. The organized energy of the sound wave is being actively consumed and turned into disorganized heat by the fluid it travels through. This is the work of dissipation, and Stokes' hypothesis gives us a precise way to calculate it.

Whenever a fluid deforms, the viscous forces do work. Just like rubbing your hands together creates warmth, this internal work on the fluid generates thermal energy. The rate of this irreversible conversion of mechanical energy into heat is described by the viscous dissipation function, $\Phi$. Using Stokes' hypothesis, this function can be shown to be:

where $\mathbf{D}$ is the rate-of-deformation tensor (the symmetric part of the velocity gradient). This equation tells us exactly how much heat is generated per unit volume, both from shearing motions ($\mathbf{D}:\mathbf{D}$) and from compressive motions ($(\nabla\cdot\mathbf{v})^2$). In most everyday, low-speed flows, this heating effect is tiny and can be safely ignored. The ratio of viscous heating to heat conduction is captured by a dimensionless number called the Brinkman number, and when it is small, dissipation plays a minor role in the overall energy balance.

But consider a sound wave. It is a traveling disturbance of pressure and density, which means the fluid is being repeatedly compressed and expanded. The viscous forces associated with this compression, as described by the term involving $\nabla \cdot \mathbf{v}$, continuously drain energy from the wave. Thermal conduction also plays a role, as the compressed parts of the wave are slightly hotter than the rarefied parts, and heat flows between them. When we apply the full theory, we find that the rate at which the sound wave's energy is dissipated depends on the square of its frequency ($\omega^2$) and a combination of shear viscosity, bulk viscosity, and thermal conductivity. This explains a common experience: high-pitched sounds seem to fade more quickly than low-pitched ones. The whisper of a flute dies away faster than the rumble of a distant drum. Without an understanding of viscous dissipation, which rests on assumptions like that of Stokes, the silence that follows a sound would be a deeper mystery.

The world of fluid mechanics has a wild side, and it is here, in the realm of supersonic flight, that Stokes' hypothesis becomes not just useful, but essential. When an object travels faster than the speed of sound, it creates a shock wave—an extraordinarily thin region across which the pressure, temperature, and density of the fluid change almost instantaneously.

If we look at this phenomenon through the lens of an "ideal," frictionless fluid, the shock wave is an infinitely thin mathematical jump. This is a nonsensical result. Nature abhors true infinities. A shock wave must have some thickness, however small. To understand what happens inside the shock, we must abandon the ideal fluid model and embrace the full reality of the Navier-Stokes equations.

Inside the shock, the fluid velocity changes so dramatically over such a short distance that the velocity gradients become enormous. Viscous forces and heat conduction, negligible in the smooth flow far away, become the dominant actors. It is they who manage the violent transition from supersonic to subsonic flow. To model this internal structure, one must have a rule for the viscous stress tensor. Stokes' hypothesis provides the simplest realistic model to do so. By solving these equations, we can calculate the profile of velocity and temperature through the shock, revealing that it is not a jump but a very steep, continuous transition. Understanding this structure is critical for predicting the intense heating and stresses experienced by supersonic aircraft and spacecraft reentering the atmosphere.

This is also where the frontiers of science push us to refine our ideas. In the extreme environment of very high-speed flow near a surface, the details of dissipation matter immensely. Researchers in turbulence modeling for CFD have to ask: is Stokes' simple hypothesis always enough? In these cases, the dissipation due to volume changes, or "dilatational dissipation," can become a significant factor on its own, especially when the "friction Mach number" (a measure of compressibility effects right at the wall) is high. This doesn't mean Stokes was wrong; it means that as we explore more extreme conditions, we build upon his foundational insight with more complex models, a classic example of scientific progress.

It is a common habit to think of "fluid" as meaning air or water. But the principles of fluid mechanics are far more universal. Most of the visible matter in the universe exists as plasma—a hot, ionized gas. This plasma, which forms stars and fills the space between galaxies, can, on large scales, be treated as a conducting fluid.

Do the ideas of viscosity and internal friction apply to something as exotic as the solar wind or the gas in a fusion reactor? Absolutely. The particles in a plasma, while electrically charged, still collide and exchange momentum, giving rise to viscous effects. When astrophysicists model the flow of matter accreting onto a black hole or the dynamics of a stellar explosion, they use a framework known as magnetohydrodynamics (MHD), which is essentially the Navier-Stokes equations augmented with electromagnetic forces.

In this context, the viscous stress tensor and the energy dissipated by it are calculated using the very same principles we have discussed. Stokes' hypothesis, or a more sophisticated variant that accounts for the magnetic field's influence on viscosity, is a tool used to understand how plasmas are heated and how energy is transported within them. It is a remarkable testament to the unity of physics that a concept developed by studying water and air helps us decode the behavior of matter in the most extreme environments in the cosmos.

From the quiet damping of a sound wave to the fierce structure of a shock front and the glow of a distant nebula, the legacy of Stokes' hypothesis is a powerful reminder. It shows us how a simple, physically motivated assumption can ripple outwards, providing the key to understanding a staggering variety of phenomena and enabling us to describe our world with ever-greater fidelity. The journey of discovery often begins with such an elegant, insightful guess.