Pulsatile flow

While the study of fluid motion often begins with the simple, unchanging world of steady flow, reality is far more dynamic. From the gusts of wind to the beat of our own hearts, the world is fundamentally unsteady. This article delves into pulsatile flow, a ubiquitous and vital form of periodic, unsteady motion. It addresses the inadequacy of steady-flow models for describing many critical phenomena in both nature and technology. By exploring this topic, you will gain a deeper understanding of the forces that shape rhythmic fluid systems. The journey begins in the first chapter, "Principles and Mechanisms," which lays the groundwork by defining key concepts like streamlines, pathlines, and the dual nature of fluid acceleration. It culminates in the introduction of the Womersley number, the critical parameter governing the battle between inertia and viscosity. The second chapter, "Applications and Interdisciplinary Connections," then demonstrates how these principles manifest in the real world, from creating measurement challenges in engineering to orchestrating the very rhythms of life in the cardiovascular, cerebrospinal, and lymphatic systems.

In our journey to understand the world, we often begin with the simplest cases. In the study of fluid motion, this starting point is ​​steady flow​​. Imagine a perfectly calm river, where at any single point you choose—near the bank, in the middle, deep down—the water's velocity never changes. The speed is constant, the direction is constant. This is the essence of steady flow. But nature is rarely so placid. The wind gusts, the tides ebb and flow, and our own hearts beat in a relentless rhythm. The world is fundamentally unsteady, and one of the most important and beautiful types of unsteadiness is ​​pulsatile flow​​.

Let's get our language straight, because in physics, precise words are the tools of clear thought. Consider a simple closed-loop pipe with a pump. When the pump is off, the fluid is at rest. Now, we switch it on. For a brief moment, the pump ramps up, pushing the fluid from rest into motion. During this ramp-up, the velocity at every point in the pipe is changing from one moment to the next. This is, by definition, an ​​unsteady flow​​. The local velocity is a function of time; mathematically, its partial derivative with respect to time, $\frac{\partial \mathbf{v}}{\partial t}$, is not zero.

Once the pump reaches its final operating speed and maintains a constant volume flow rate, say $Q$, does the flow become steady? The answer, perhaps surprisingly, is "it depends where you look". If our pipe has a constant diameter everywhere, then the average velocity is the same all along its length. Since the flow rate $Q$ is now constant, the velocity at any fixed point is also constant. The flow has become steady.

But what if our pipe includes a conical reducer, a section where it smoothly narrows? Mass must be conserved, so as the cross-sectional area $A$ decreases, the fluid must speed up ($V = Q/A$). A particle moving through this reducer is constantly accelerating, even though the overall flow rate $Q$ is no longer changing in time. Is this flow steady or unsteady? At any fixed point within the reducer, the velocity is constant because $Q$ and the area at that point are constant. So, the flow is still ​​steady​​! What we are observing is the difference between a flow that is ​​uniform​​ (velocity is the same at different points along a path at one instant) and ​​non-uniform​​ (velocity varies with position). Our flow in the straight sections is both steady and uniform, while in the reducer, it is steady but non-uniform.

Pulsatile flow, then, is a special kind of unsteady flow, one that is typically periodic. Think of a pump that drives the fluid not with a constant push, but with a rhythmic pulse. The velocity at any point might be described by a function like $U(t) = U_{avg}(1 + \epsilon \sin(\omega t))$, where it oscillates around an average value $U_{avg}$. This is the heartbeat of our subject.

A common intuition is to imagine that if you could see the "lines" of a flow at a given instant, a small particle would simply travel along one of these lines. These instantaneous lines of flow, which are tangent to the velocity vector at every point, are called ​​streamlines​​. They give us a snapshot of the flow's structure right now. In a steady flow, this intuition is correct. The streamlines are fixed, and they are identical to the actual trajectories of fluid particles, which we call ​​pathlines​​.

But in an unsteady flow, the world is far more interesting. The streamlines themselves are changing from moment to moment. A particle starts moving in the direction of the streamline at its current location, but by the time it has moved a little, the streamline has already changed. The particle is perpetually chasing a target that is itself moving.

Imagine a simple, albeit contrived, two-dimensional flow where the velocity is given by $\vec{V} = (at)\hat{i} + U\hat{j}$, where $U$ and $a$ are constants. At any specific moment in time, say $t=t_0$, the velocity's horizontal component is constant everywhere, $at_0$. The ratio of vertical to horizontal velocity is $U/(at_0)$, which is a constant slope. Therefore, the streamlines at that instant are all straight lines. If you take a snapshot, you see a field of straight, parallel flow lines.

But what path does a particle starting at the origin actually follow? We must integrate its velocity over time. The vertical motion is simple: $y(t) = Ut$. The horizontal velocity, however, depends on time, $dx/dt = at$. Integrating this gives $x(t) = \frac{1}{2}at^2$. If we eliminate time $t$ by substituting $t=y/U$, we find the particle's pathline: $x = \frac{a}{2U^2}y^2$. This is the equation of a ​​parabola​​! The particle follows a curved path, even though at every single instant, the "road map" of streamlines consists of only straight lines. This divergence between streamlines and pathlines is a fundamental signature of unsteady flow, a beautiful consequence of the flow field evolving in time.

To understand the forces that create these curving pathlines, we must think about acceleration. When you are in a car, you feel acceleration in two ways. You are pressed back into your seat when the driver steps on the gas (the car's velocity changes with time), and you are pushed to the side when the car goes around a curve (your direction of velocity changes with space). A fluid particle experiences the exact same two kinds of acceleration.

The total acceleration of a fluid particle, known as the ​​material derivative​​, is expressed as:

The first term, $\frac{\partial \mathbf{v}}{\partial t}$, is the ​​local acceleration​​. It’s the change in velocity at a fixed point in space. This is the term that is non-zero in a pulsatile flow, representing the rhythmic speeding up and slowing down of the entire flow field. It’s what you would measure with a probe stationary in the pipe.

The second term, $(\mathbf{v} \cdot \nabla)\mathbf{v}$, is the ​​convective acceleration​​. It exists because the particle moves or is convected to a new location in space where the velocity is different. This is the acceleration you feel going around a bend or through a nozzle, even in a perfectly steady flow.

In a general pulsatile flow, a particle is subject to both. It is being accelerated because the whole flow is pulsing (local) and because it is moving through regions of different velocity (convective). This dual nature of acceleration is central to the dynamics of pulsatile systems.

We now arrive at the heart of the matter. What physical principle governs the character of a pulsatile flow? The answer lies in a battle between two fundamental properties of the fluid: ​​inertia​​ and ​​viscosity​​.

• ​​Inertia​​ is the fluid's tendency to resist changes in motion, a consequence of its mass. To accelerate a slug of fluid in a pipe requires a force. In a pulsatile flow, you are constantly asking the fluid to speed up and slow down. Inertia is the fluid's reluctance to do so. This is captured by the local acceleration term, $\rho \frac{\partial \mathbf{v}}{\partial t}$.

• ​​Viscosity​​ is the fluid's internal friction, its "stickiness." It's the force that tries to smooth out velocity differences. When you push the fluid, the layer at the pipe wall sticks, and viscosity transmits this shearing effect inwards, trying to drag the rest of the fluid along.

The entire character of pulsatile flow is dictated by the ratio of these two forces. This ratio is captured by a single, powerful dimensionless number: the ​​Womersley number​​, $\alpha$. It is defined as:

where $R$ is the pipe radius, $\omega$ is the pulsation frequency, $\rho$ is the fluid density, and $\mu$ is its dynamic viscosity. The square of the Womersley number, $\alpha^2$, represents the ratio of unsteady inertial forces to viscous forces.

Let's examine the two extremes:

• ​​Low Womersley Number ($\alpha \ll 1$):​​ This happens when pulsations are slow ($\omega$ is small), the pipe is narrow ($R$ is small), or the fluid is very viscous (like honey). In this regime, viscous forces dominate. The "sticky" forces have plenty of time during each cycle to diffuse from the wall all the way to the center of the pipe. The flow responds almost instantaneously to the changing pressure gradient. The velocity profile remains parabolic, just like in a steady pipe flow (Poiseuille flow), but its amplitude simply waxes and wanes with the pressure. This is called a ​​quasi-steady​​ flow.

• ​​High Womersley Number ($\alpha \gg 1$):​​ This is the case for fast pulsations, wide pipes, or low-viscosity fluids (like water). Here, unsteady inertia dominates. The bulk of the fluid in the core of the pipe is too "heavy" and sluggish to respond to the rapid oscillations. Only a thin layer of fluid near the wall, the ​​oscillatory boundary layer​​, can keep up. The result is a blunt, almost plug-like velocity profile where the core of the fluid slides back and forth as a nearly solid block. Furthermore, inertia introduces a ​​phase lag​​: the peak velocity no longer occurs at the same time as the peak pressure gradient, just as it takes a moment for a heavy box to start moving after you begin pushing it.

There is no better example of this than the human cardiovascular system. In our largest artery, the ascending aorta, the diameter is about 3 cm, the heart rate is about 72 beats per minute, and for blood, the Womersley number $\alpha$ is about 23. This is firmly in the high-$\alpha$ regime. The blood flow leaving our heart is an inertial, plug-like flow with a significant phase lag between pressure and flow—a far cry from the simple, steady flow models often taught first.

The battle between inertia and viscosity leads to some fascinating and non-intuitive consequences.

First, it costs more energy to pump a fluid in pulses. Consider a pulsating turbulent flow. The instantaneous pressure drop is proportional to the velocity squared. Because the average of a square is always greater than the square of the average ($\langle U^2 \rangle > \langle U \rangle^2$), the high-velocity portions of the cycle contribute disproportionately to the overall frictional loss. The time-averaged pressure drop for a flow oscillating with amplitude $\beta$ around a mean is increased by a factor of $(1 + \beta^2/2)$ compared to a steady flow with the same mean velocity. This is a "pumping penalty" paid for unsteadiness, a direct result of the energy contained in the velocity fluctuations.

Second, the inertial nature of high-frequency pulsatile flow can be beautifully described using the language of electrical engineering. The resistance of a component to flow is its ​​hydraulic impedance​​, the ratio of pressure drop to flow rate. For a purely oscillatory flow in an ideal (inviscid) fluid, the impedance is purely imaginary. This is because the pressure is not doing work against friction, but is instead working to accelerate and decelerate the fluid's mass. The fluid behaves like an inductor in an AC circuit, storing and releasing kinetic energy. This concept of ​​added mass​​ or ​​acoustic inertance​​ is crucial in designing high-frequency hydraulic systems.

Finally, pulsatile flows are not just passive recipients of an oscillating drive; they can actively interact with and organize the flow field. Consider a cylinder in a steady stream. It naturally sheds a beautiful trail of alternating vortices called a Kármán vortex street. The shedding has a natural frequency. If we now add a small pulsation to the incoming flow, something remarkable can happen. If the pulsation frequency is near a multiple or sub-multiple of the natural shedding frequency, the vortex shedding process can ​​lock-in​​ and synchronize with the external driving pulsation. This phenomenon of lock-in, common to all non-linear oscillators, shows that pulsatile flow is a dynamic participant, capable of orchestrating complex fluid-structure interactions, from the "singing" of power lines in the wind to the design of advanced flowmeters.

From simple definitions to the grand competition between inertia and viscosity, the principles of pulsatile flow reveal a world of rich, complex, and beautiful physics, governing everything from the blood in our veins to the engineering systems that power our world.

Having explored the fundamental principles of pulsatile flow, we now turn our attention to the real world. If our journey so far has been about learning the grammar of this dynamic language, this chapter is about reading the stories it tells. You will see that pulsatile flow is not some esoteric curiosity confined to the laboratory; it is everywhere. It poses subtle challenges to the engineer, orchestrates the very rhythms of life within our bodies, and drives the frontiers of technology. We will find that the same core ideas reappear in the most unexpected places, revealing a beautiful unity in the workings of nature.

Let's begin with a seemingly simple task: measuring the flow rate of a fluid in a pipe. If the flow is steady, like a placid river, the task is straightforward. But what if the flow is pulsating, driven by a piston pump, for instance? You might think you could just put a standard flow meter in the line and average its readings over time. The surprise is that this simple approach can be spectacularly wrong.

Consider an orifice meter, a common device that works by placing a plate with a hole in the pipe and measuring the pressure drop, $\Delta P$, across it. The flow rate, $Q$, is related to this pressure drop by a simple law: $Q$ is proportional to the square root of $\Delta P$. Now, if the flow $Q(t)$ is pulsating, the pressure drop $\Delta P(t)$ will pulsate along with it. A pressure gauge with a slow response will naturally report the time-averaged pressure drop, $\overline{\Delta P}$. The meter’s electronics then naively apply the steady-flow formula to calculate an "indicated" flow rate, which is proportional to $\sqrt{\overline{\Delta P}}$.

Here lies the trap. The quantity we want is the true average flow rate, $\overline{Q}$. The quantity we measure is proportional to the square root of the average pressure drop. Because of the non-linear, square-root relationship, these are not the same! The average of the square roots is not the square root of the average. In fact, due to a fundamental mathematical property of convex functions (the function $x^2$ in this case), the meter will always overestimate the true average flow rate. The same deception occurs with a rotameter, where the height of a float is balanced by a drag force proportional to the velocity squared. Here too, the float settles at a position corresponding to the average of the square of the velocity, leading to an overestimation of the true average flow.

This isn't just a measurement quirk; it's a reflection of a deeper physical reality. A pulsating flow carries more kinetic energy, on average, than a steady flow with the same average rate. This excess energy must be accounted for. For example, the energy lost to friction and turbulence as fluid passes through a valve or bend in a pipe—the head loss—is also proportional to the velocity squared. Consequently, a system with pulsating flow suffers a greater average energy loss than its steady-flow counterpart, reducing the overall efficiency of the pump and system.

The consequences of even tiny pulsations can be profound. In the world of high-performance liquid chromatography (HPLC), chemists separate trace amounts of substances in a sample. This is often done using a "gradient," where the composition of the solvent mixture is changed over time. Imagine mixing a water-based solvent with an acetonitrile-based solvent, where the acetonitrile absorbs more ultraviolet light. If the pumps delivering these two solvents have even the slightest pulsation, the composition of the mobile phase will ripple. This composition ripple is seen by the UV detector as a fluctuating baseline absorbance, creating noise that can completely obscure the tiny signals from the biomarkers you are trying to detect. The design of modern HPLC systems involves a sophisticated dance of proportioning valves and mixing chambers precisely to damp these pulsations and achieve a flat, quiet baseline. In all these engineering contexts, the pulse is a nuisance to be tamed. But in the world of biology, we find that nature has become a master of the pulse.

Our bodies are not steady-state machines; they are symphonies of pulsation. From the beat of our heart to the rhythm of our breath, life is fundamentally oscillatory. Nature has not only adapted to this reality but has evolved to exploit it in wonderfully elegant ways.

The most prominent pulse, of course, is the flow of blood ejected by the heart. As this wave of pressure and flow travels down the aorta, it interacts with the artery walls. The pulsatile nature of the flow creates a characteristic length scale—a thin region near the wall known as the oscillatory or Stokes boundary layer. For blood in the aorta, this layer is only about a millimeter thick. It is within this thin layer that the artery wall "feels" the flow, experiencing the shear stress that is a critical signal for maintaining vascular health. The character of the pulsation here is paramount.

The heart's beat echoes far beyond the circulatory system. Consider the cerebrospinal fluid (CSF) that bathes our brain and spinal cord. The skull is a rigid box of bone, and according to the Monro-Kellie doctrine, the total volume inside it—brain, blood, and CSF—must remain nearly constant. So, with each heartbeat, as arteries inside the skull expand with a pulse of blood, something must give way. That something is the CSF. The transient increase in intracranial pressure pushes a small volume of CSF out of the rigid skull and down into the more compliant spinal canal. During diastole, as the arteries relax, the CSF flows back up. This rhythmic sloshing, driven by the cardiac cycle, is a critical mechanism for circulating CSF, distributing nutrients, and, crucially, clearing metabolic waste from the brain as we sleep.

Perhaps the most beautiful example of nature's ingenuity is found in the lymphatic system, our body's drainage and immune surveillance network. Deep lymphatic vessels often run alongside arteries. These vessels are thin-walled, highly compliant, and lined with frequent, one-way micro-valves. When the adjacent artery expands with each heartbeat, it compresses the lymphatic vessel. The fluid inside is squeezed, and the valves ensure it can only move in one direction—proximally, toward the chest. As the artery relaxes, the lymphatic vessel refills from the periphery. In this way, the lymphatic system "hitches a ride" on the powerful beat of the circulatory system, using the artery as an external pump to drive its own slow, but vital, current.

But what happens when these vital rhythms go wrong? In the devastating condition of septic shock, the microcirculation can break down. Even if doctors restore normal blood pressure, tiny capillaries in the tissues may experience sluggish, intermittent, and even reversed flow. This change from a healthy, brisk, unidirectional flow to a low, oscillatory one is a disaster at the cellular level. The endothelial cells lining our blood vessels are not just passive tubes; they are sophisticated mechanosensors. They can tell the difference between "good" and "bad" flow patterns. Healthy, laminar shear stress activates protective genetic programs. The pathological, oscillatory shear stress seen in sepsis does the opposite: it triggers pro-inflammatory and pro-coagulant pathways, like NF-κB. The cells start expressing molecules that make them "sticky" to leukocytes and platelets, leading to micro-clots, inflammation, and ultimately, organ failure. This is a chilling example of how a change in the physical character of a pulsatile flow can directly trigger a deadly biological cascade.

The lessons of pulsatile flow extend into our most advanced technologies, where the pulse can be a source of immense power, catastrophic failure, or profound diagnostic insight.

Just as a guitar string has a natural frequency at which it prefers to vibrate, a hydraulic system of pipes and fluid has natural frequencies of its own. The inertia of the fluid in a long pipe acts like an electrical inductor, while the compressibility of the fluid or the flexibility of a container acts like a capacitor. A pulsating pump can drive this hydraulic "LC circuit" at its resonant frequency, leading to enormous, often destructive, pressure swings—a phenomenon related to the infamous "water hammer".

The situation becomes even more complex when a flexible structure is placed in a flow. Think of a flag flapping in the wind. The flag's movement changes the flow around it, which in turn changes the forces on the flag, which changes its movement. This is a fluid-structure interaction (FSI). In some cases, the system can "lock-in" to a state of resonant self-excitation, where the frequency of the fluid pulsation becomes synchronized with the natural frequency of the structure. This can lead to violent oscillations, as famously demonstrated by the collapse of the Tacoma Narrows Bridge. Understanding this non-linear dance between fluid and structure is critical for designing everything from airplane wings to skyscrapers and even artificial heart valves.

Finally, the very act of trying to "see" these pulsatile flows inside the body presents its own fascinating challenge. In Magnetic Resonance Imaging (MRI), an image is built up piece by piece over time. If you are trying to image a carotid artery where blood is pulsing, the velocity and acceleration are changing from one moment to the next. This creates inconsistencies in the data being acquired. One result is a loss of signal, as spins within a single voxel moving at different speeds get out of phase with each other and their signals cancel out. Another is the appearance of "ghosts"—faint, repeating images of the artery smeared across the image. These artifacts arise because the periodic motion of the blood fools the imaging process. MRI physicists have developed ingenious solutions, such as gradient moment nulling (GMN) and cardiac gating, to compensate for these effects, effectively "freezing" the motion to produce a clear image. The pulse, once again, is both the object of our study and the source of our challenge.

From the simple error of a flow meter to the intricate workings of the human brain, from the silent propulsion of lymph to the violent collapse of a bridge, the principles of pulsatile flow provide a unifying thread. It is a reminder that the world is not static; it is alive with rhythm and oscillation. Understanding this pulse is not just an academic exercise—it is fundamental to understanding ourselves and the world we build.