Objective Time Derivative

In the study of continuum mechanics, describing how material properties like stress or strain change over time is fundamental. While the standard time derivative is sufficient for stationary systems, it fails when the material itself is flowing, deforming, and rotating. This introduces a critical problem of perspective: how can we formulate physical laws that are consistent for all observers, regardless of their own motion? The simple material derivative is not "objective"—it incorrectly registers changes due to pure rotation, violating the Principle of Material Frame-Indifference.

This article tackles this fundamental challenge by introducing the concept of the ​​objective time derivative​​. It provides a robust framework for accurately describing material evolution in complex flows. The first chapter, ​​Principles and Mechanisms​​, will delve into why the simple derivative is insufficient, explore the mathematical toolkit used to correct it, and introduce key objective rates like the Jaumann and upper-convected derivatives, explaining their distinct physical meanings. Following this, the chapter on ​​Applications and Interdisciplinary Connections​​ will demonstrate the profound impact of using the correct derivative in real-world scenarios, from modeling complex fluids in rheology to simulating large deformations in solid mechanics and ensuring consistency with the laws of thermodynamics.

To understand the world of moving, flowing, and deforming things—be it the churning of cream into butter, the flow of paint from a brush, or the slow crawl of glaciers—we need a language to describe how properties change. Our first instinct, inherited from introductory physics, is to think about the rate of change with time, the familiar derivative $d/dt$. But when the medium itself is in motion, this simple idea quickly runs into a profound and beautiful complication, a problem of perspective.

Imagine you are trying to describe the stress inside a piece of taffy as it's being pulled and twisted. You want to know how the stress at a particular speck of sugar is changing as that speck is carried along by the flow. This rate of change, following a material particle, is what we call the ​​material time derivative​​. For any tensor quantity $\boldsymbol{A}$ (like stress or strain), it's written as $\dot{\boldsymbol{A}}$ and is composed of the local rate of change at a fixed point in space plus a term that accounts for the particle moving to a new location with a different value of $\boldsymbol{A}$. Mathematically, $\dot{\boldsymbol{A}} = \partial \boldsymbol{A} / \partial t + (\boldsymbol{v} \cdot \nabla) \boldsymbol{A}$, where $\boldsymbol{v}$ is the fluid's velocity.

This seems straightforward enough. But now, let’s introduce a twist. What if we, the observers, are watching this from a spinning merry-go-round? One of the most fundamental tenets of physics, the ​​Principle of Material Frame-Indifference (MFI)​​, states that the laws of nature must be the same for all non-accelerating observers, regardless of their rigid-body rotation. A constitutive law, which is an equation describing a material's behavior (e.g., how its stress relates to its deformation), must not depend on whether the physicist writing it is sitting in the lab or on a spinning chair.

The simple material derivative, $\dot{\boldsymbol{A}}$, spectacularly fails this test.

Consider a simple thought experiment: a fluid undergoing a pure rigid-body rotation, like coffee being stirred gently in a cup. Let's imagine a tensor quantity $\boldsymbol{A}$—perhaps representing the orientation of microscopic fibers suspended in the fluid—that is simply "stuck" in the fluid and rotates with it. From the perspective of someone riding along with that fluid element, the tensor $\boldsymbol{A}$ is absolutely constant. Yet, for us in the laboratory frame, it is clearly changing as it rotates. If we calculate its material derivative, we find that $\dot{\boldsymbol{A}}$ is not zero! Instead, we get $\dot{\boldsymbol{A}} = \boldsymbol{W}\boldsymbol{A} - \boldsymbol{A}\boldsymbol{W}$, where $\boldsymbol{W}$ is the ​​spin tensor​​, which describes the local rate of rotation of the fluid.

This non-zero derivative is a "phantom" rate of change. It doesn't represent any intrinsic change in the material's state; it's purely an artifact of viewing a rotating object from a stationary frame. When we formalize this by looking at how $\dot{\boldsymbol{A}}$ transforms under a change of observer characterized by a rotation $\boldsymbol{Q}(t)$, we find that the new derivative $\dot{\boldsymbol{A}}'$ is not just the rotated version of the old one, $\boldsymbol{Q}\dot{\boldsymbol{A}}\boldsymbol{Q}^T$. Instead, it picks up extra, unwanted terms related to the observer's own spin: $\dot{\boldsymbol{A}}' = \boldsymbol{Q}\dot{\boldsymbol{A}}\boldsymbol{Q}^T + \boldsymbol{\Omega}\boldsymbol{A}' - \boldsymbol{A}'\boldsymbol{\Omega}$, where $\boldsymbol{\Omega}$ is the spin of the observer's frame. Because of these extra terms, the material derivative is said to be ​​non-objective​​. We cannot use it to formulate physical laws, because a law relating $\dot{\boldsymbol{A}}$ to something else would give different results for different observers.

To build a proper physical law, we need to craft a new kind of time derivative, an ​​objective time derivative​​, that is immune to the observer's rotation. The goal is to create an operator that measures only the intrinsic changes in a tensor, separating them from the trivial effects of rigid rotation.

Our toolkit for this task comes from a closer look at the fluid's motion. Any local motion can be described by the ​​velocity gradient tensor​​, $\boldsymbol{L} = \nabla\boldsymbol{v}$. This tensor holds all the information about how the neighborhood of a point is moving. We can decompose it into two parts with very clear physical meanings:

1. The ​​rate-of-deformation tensor​​, $\boldsymbol{D} = \frac{1}{2}(\boldsymbol{L} + \boldsymbol{L}^T)$, is the symmetric part. It describes how the material is being stretched and sheared.
2. The ​​spin tensor​​, $\boldsymbol{W} = \frac{1}{2}(\boldsymbol{L} - \boldsymbol{L}^T)$, is the skew-symmetric part. As we've seen, it describes the local angular velocity, or the rate at which the material element is spinning.

The non-objectivity of the material derivative comes from it being "contaminated" by the spin $\boldsymbol{W}$. The most intuitive way to fix this is to simply subtract this rotational contribution. This leads us to the ​​Jaumann derivative​​, also known as the co-rotational derivative:

$\overset{\circ}{\boldsymbol{A}} = \dot{\boldsymbol{A}} - \boldsymbol{W}\boldsymbol{A} + \boldsymbol{A}\boldsymbol{W}$

This definition is beautifully constructed. The correction terms $-\boldsymbol{W}\boldsymbol{A} + \boldsymbol{A}\boldsymbol{W}$ are designed to precisely cancel out the phantom rate of change that arises from rigid rotation. If a tensor is just rigidly rotating with the fluid and undergoing no intrinsic change, its Jaumann derivative is exactly zero. It measures the rate of change as seen by an observer who is spinning along with the material element. It seems we have found our perfect, objective tool. But is the story truly this simple?

A fluid element doesn't just spin; it also stretches. This stretching motion is just as important as rotation, and our derivatives must account for it properly. This realization opens the door to a richer family of objective derivatives, each with its own profound physical meaning.

The most important of these is the ​​upper-convected derivative​​ (or Oldroyd-B derivative):

$\overset{\triangledown}{\boldsymbol{A}} = \dot{\boldsymbol{A}} - \boldsymbol{L}\boldsymbol{A} - \boldsymbol{A}\boldsymbol{L}^T$

At first glance, this seems more complex than the Jaumann derivative. It uses the full velocity gradient $\boldsymbol{L}$, not just the spin part. But its physical interpretation is just as beautiful: it represents the rate of change of a tensor as measured in a coordinate system that is not only rotating, but also stretching with the fluid.

Why would we need different objective derivatives? The answer reveals a deep connection between physics and geometry. The "correct" derivative to use depends on the geometric nature of the quantity you are measuring.

Think of a polymer solution. The stress in the fluid comes from the stretching of long-chain molecules. Let's represent a single polymer chain by a vector $\boldsymbol{q}$ connecting its ends. As the fluid deforms, this vector is carried along and stretched by the flow. A quantity that transforms this way—by being "pushed forward" with the deformation—is called a ​​contravariant​​ tensor. The conformation tensor, $\boldsymbol{A} = \langle \boldsymbol{q}\boldsymbol{q} \rangle$, which represents the average stretch and orientation of all polymer chains, is a contravariant tensor. The upper-convected derivative is the one, and only one, that naturally describes the transport of such objects. In fact, if a polymer network deforms perfectly with the fluid (affine deformation), its upper-convected derivative is zero.

Conversely, there is a ​​lower-convected derivative​​, $\overset{\triangle}{\boldsymbol{A}} = \dot{\boldsymbol{A}} + \boldsymbol{L}^T\boldsymbol{A} + \boldsymbol{A}\boldsymbol{L}$. This derivative is the natural choice for ​​covariant​​ tensors, which can be thought of as measuring grids or surfaces that are deformed by the flow. Thus, the choice is not arbitrary; it is prescribed by the underlying physics and geometry of the problem.

This distinction is not just a mathematical subtlety; it has dramatic and observable consequences. These derivatives are fundamentally different in any flow that involves stretching (where $\boldsymbol{D} \neq \boldsymbol{0}$). For instance, the difference between the upper-convected and Jaumann derivatives of a tensor $\boldsymbol{A}$ is given by $-(\boldsymbol{D}\boldsymbol{A} + \boldsymbol{A}\boldsymbol{D})$. This difference, which depends on the rate of stretching $\boldsymbol{D}$, can lead to completely different physical predictions.

Let's return to our polymer solution, described by a classic Maxwell model. The choice of objective derivative defines the material's behavior.

• ​​A Cautionary Tale​​: If we use the intuitively simple Jaumann derivative and simulate the fluid in a simple shear flow (like spreading honey on toast), the model makes a bizarre prediction: the normal stresses, which are responsible for the famous rod-climbing (Weissenberg) effect, begin to oscillate wildly. This is completely unphysical; real polymer solutions do not do this.

• ​​The Physically Correct Choice​​: If we instead use the upper-convected derivative—the one we know is physically appropriate for describing polymer stretch—the model correctly predicts a positive, steady normal stress that grows with the shear rate. It captures the essential physics.

The profound importance of this choice is at the heart of a major challenge in computational science: the ​​High Weissenberg Number Problem (HWNP)​​. In strong flows (high Weissenberg number), the upper-convected model correctly predicts that polymer molecules undergo a dramatic transition from a coiled to a highly stretched state. This physical reality leads to an explosive growth in stresses. While physically correct, these enormous stresses and their sharp gradients are a nightmare for numerical simulations, often causing them to fail. The problem is not that the mathematics is wrong, but that the physics it describes is wonderfully extreme.

The journey from a simple time derivative to a family of objective rates is, therefore, far more than a mathematical exercise. It is a deep exploration of the nature of observation and motion. It reveals a hidden unity between the abstract geometry of tensors, the physical behavior of materials like polymer solutions, and the practical challenges of modern scientific computing. The beauty lies in discovering that by carefully crafting our mathematical tools to respect fundamental physical principles, we can unlock a true and predictive description of our complex, flowing world.

Having grappled with the mathematical machinery of objective time derivatives, one might be tempted to ask, "Is all this formalism truly necessary?" Is this not just a subtle mathematical game played by theoreticians, a fine point of little consequence in the real world of flowing rivers, bending steel, and brewing coffee? The answer, it turns out, is a resounding "no." The principle of objectivity is not a mere mathematical nicety; it is a profound physical requirement that stands as a gatekeeper, ensuring that our models of the physical world are sensible, consistent, and in harmony with the most fundamental laws of nature, from the mechanics of everyday materials to the grand laws of thermodynamics.

The journey to appreciate this begins with a simple observation. Imagine stirring a cup of thick honey with a spoon. The honey resists; it feels thick and viscous because you are deforming it. Now, imagine putting the same cup of honey on a spinning turntable. The honey rotates as a solid block, but it does not deform internally. You would not expect the honey to spontaneously develop internal stresses just from being spun around. Our physical laws must be smart enough to distinguish between a true deformation (stirring) and a simple rigid rotation (spinning).

For a simple fluid like water—a "Newtonian" fluid—this distinction comes for free. The stress in a Newtonian fluid is directly and instantaneously proportional to the rate of deformation. If there is no deformation, there is no stress. But the world is filled with materials far more complex and interesting than water. Think of polymer solutions like paint or shampoo, biological fluids, molten plastics, or even the solid earth deforming over geological time. These materials have memory. Their current state of stress depends not just on what is happening to them right now, but on their entire history of stretching and shearing.

To model such "viscoelastic" materials, we need equations that describe how the stress evolves over time. And this is precisely where the ordinary material time derivative, $\dot{\boldsymbol{\sigma}}$, reveals its fatal flaw. If we were to use it in our constitutive law, it would incorrectly predict that a material develops stresses simply by being rotated, even with zero deformation. It fails the turntable test! An objective time derivative, such as the Jaumann rate, is the mathematical cure for this ailment. It is ingeniously constructed to measure the rate of change of stress relative to a frame that rotates with the material element itself. It effectively "subtracts" the spurious changes due to pure spin, ensuring that a vortex flow, for instance, is correctly interpreted by the constitutive law.

Nowhere is the application of objective derivatives more vital than in the realm of complex fluids and rheology—the science of flow. Consider a dilute polymer solution, a liquid containing long-chain molecules. These molecules are like microscopic strands of spaghetti tumbling and stretching in the solvent.

A classic model for such a fluid is the Oldroyd-B model, which beautifully illustrates the power of our concept. It pictures the total stress as a sum of contributions from the simple Newtonian solvent and the complex polymeric part. The polymer stress is governed by an equation that balances its tendency to relax back to a coiled-up state against the stretching imposed by the flow. This evolution equation must be objective, and the model employs the ​​upper-convected derivative​​ to achieve this.

The beauty here is that this specific choice of derivative is not arbitrary. It can be derived directly from the underlying microscopic physics. If you model the polymer chains as tiny elastic dumbbells and analyze how they are carried along (convected) and stretched by the fluid flow, the upper-convected derivative emerges naturally from the mathematics. It is the macroscopic echo of the microscopic dance of molecules. This deep connection between the micro and macro worlds is a hallmark of great physical theories.

Of course, the upper-convected derivative is not the only player on the field. One could choose the lower-convected derivative or the corotational (Jaumann) rate, among others. Each choice represents a different constitutive assumption and leads to a different model with different predictions. For instance, in a simple shear flow, where layers of fluid slide past one another, using an upper-convected versus a lower-convected derivative results in different stress components being affected by the shear rate. This "zoo" of derivatives is not a weakness of the theory but a reflection of the rich and varied behavior of real materials. The choice of which derivative to use is a crucial part of the physicist's or engineer's art: selecting the model that best captures the essence of the material in question.

The necessity of objectivity extends far beyond fluids. In solid mechanics, when materials undergo large, permanent deformations, the same issues of rotation and frame indifference arise.

Consider the world of high-strain-rate plasticity, which governs phenomena like the behavior of metal during a car crash or in an industrial forging process. To simulate these events, computers must solve equations that track the evolution of stress as the material deforms. This stress update must be objective. For many common metals, the onset of plastic flow is governed by the intensity of the stress, a quantity that is naturally independent of rotation (isotropic). In such cases, the specific choice of objective rate (Jaumann, Green-Naghdi, etc.) is not dictated by the underlying physics of yielding, but any valid objective rate must be used to ensure the computational algorithm respects the principle of frame indifference.

Moving from the fast-paced world of impacts to the slow, relentless motion of the Earth, we find the same principles at work in geomechanics. Modeling the flow of glaciers, the convection of the Earth's mantle, or the deformation of rock in a fault zone requires a framework for finite-strain plasticity. Here, a particularly elegant approach involves the multiplicative decomposition of deformation, $F = F_e F_p$, which separates the total change in shape ($F$) into a recoverable elastic part ($F_e$) and a permanent plastic part ($F_p$). By formulating the physical laws in a special, abstract "intermediate configuration" that is tied to the plastic flow, one can define yield criteria and flow rules that are inherently objective. This sophisticated viewpoint reveals that the choice of plastic spin—how this intermediate configuration is assumed to rotate—is itself a constitutive choice that dictates how evolving material properties, like internal crystal orientations, are tracked.

Perhaps the most profound connection is the one between mechanical objectivity and the laws of thermodynamics. When a viscoelastic material is deformed, some of the work done is stored as elastic energy (like stretching a spring), and some is dissipated as heat. This viscous dissipation is what makes a batch of dough warm up in an electric mixer. The Second Law of Thermodynamics, one of the most unshakable pillars of physics, dictates that this dissipated energy, which increases the entropy of the universe, can never be negative.

Herein lies the ultimate test for our constitutive models. If we were to use a non-objective stress derivative, we could find a spinning frame of reference from which the predicted dissipation would appear to be negative. This would be tantamount to observing the mixer spontaneously drawing heat from the dough to power itself—a physical absurdity. Using an objective derivative ensures that the predicted stress power and viscous heating are frame-invariant, thereby upholding the sanctity of the Second Law in all possible observational frames. This means that a thermodynamically consistent model must be constructed as a complete package: a specific objective rate must be paired with a compatible form of the material's free energy function [@problem_id:3992208, @problem_id:3554878].

Finally, this deep theoretical chain, starting from a simple principle of observation and linking mechanics to thermodynamics, finds its purpose in modern computational science. The models we derive, describing the evolution of microscopic structures like the polymer conformation tensor $\mathbf{A}$, are often incredibly complex and expensive to simulate directly. This is where techniques like Reduced-Order Modeling (ROM) come in. By running a single, high-fidelity simulation, we can analyze the results and extract the most dominant spatial patterns of microstructural anisotropy. These patterns form a custom-built basis, a "greatest hits" of the material's behavior. Subsequent simulations can then be projected onto this highly efficient reduced basis, allowing for rapid exploration and design.

From the spinning of a honey jar to the design of advanced materials on a supercomputer, the thread of objectivity runs deep. It is a principle of consistency that unifies the microscopic and macroscopic, connects the mechanics of fluids and solids, and ensures that our physical theories are not just mathematically elegant, but are faithful to the fundamental workings of the universe.