True Stress and Strain

How do we truly measure the strength of a material? When we pull on a metal bar, our standard engineering measurements often tell a puzzling story: after a certain point, the material appears to get weaker before it finally snaps. This apparent paradox highlights a critical gap between a component's external behavior and the actual physical reality its internal structure experiences. This discrepancy can have profound consequences for engineering design and safety analysis.

This article delves into the crucial distinction between engineering stress and the more physically accurate "true stress." We will uncover why this seemingly simple correction is fundamental to understanding material science. In the following chapters, we will first explore the principles and mechanisms behind true stress and strain, revealing how they paint a picture of continuous strengthening known as strain hardening. Following this, we will examine the far-reaching applications of this concept, from predicting catastrophic failure and building virtual engineering simulations to understanding the intricate dance of microscopic defects that gives materials their character. By the end, you will understand why the true stress-strain curve is not just a mathematical refinement but a universal language for describing material behavior.

Imagine you are pulling on a piece of licorice. As you pull, it not only gets longer, but it also gets thinner. If you were to simply record the force you are applying, you would notice something peculiar. After a certain point, it feels like it gets easier to stretch, even though the candy is getting closer to snapping. Does this mean the licorice is somehow getting weaker before it breaks? Our intuition, and a naive measurement, would suggest so. But the reality of what the material is experiencing is far more fascinating and, as we shall see, quite different. This simple act of stretching holds the key to understanding the profound difference between how we often measure strength and what strength truly is.

In science and engineering, we like to quantify things. When we pull on a sample of metal, we measure the force, $F$, and how much it elongates, say from an initial length $L_0$ to a current length $L$. To make these measurements independent of the sample's size, we normalize them. The most straightforward way is to divide the force by the original cross-sectional area, $A_0$, and the change in length, $L - L_0$, by the original length, $L_0$. These give us what are called ​​engineering stress​​ ($\sigma_{\mathrm{eng}}$) and ​​engineering strain​​ ($\epsilon_{\mathrm{eng}}$):

These are wonderfully practical quantities. They tell a designer how a component of a given initial size will behave under a specific load. It's the "before and after" picture. However, they don't tell the whole story. They don't accurately describe the physical state of the material during the deformation.

Think back to the licorice. As it stretches, it thins. The force you are applying is concentrated over a progressively smaller area. The atoms inside the material are being pulled apart with an intensity that is actually increasing dramatically. To capture this "atom's-eye view," we need a more faithful set of measures. We define ​​true stress​​ ($\sigma_{\mathrm{true}}$) as the force $F$ divided by the instantaneous cross-sectional area $A$. For strain, we consider that each little bit of stretching happens relative to the length at that moment. Summing up all these little relative stretches from the beginning to the end gives us the ​​true strain​​ ($\epsilon_{\mathrm{true}}$), which mathematically turns out to be a natural logarithm:

This logarithmic form has a beautiful property: strains become additive. Stretching by 10% and then another 10% of the new length isn't a 20% stretch in total, but true strain captures this compounding effect perfectly. These "true" measures describe the actual physical conditions within the material from moment to moment.

Now, let's plot these two descriptions of our tensile test and see what they reveal. If we plot engineering stress versus engineering strain, we see a curve that rises, reaches a peak, and then, puzzlingly, begins to fall until the specimen breaks. The peak of this curve is a famous and important quantity called the ​​Ultimate Tensile Strength (UTS)​​. The downward slope after the UTS seems to confirm our initial intuition: the material gets weaker before it fails.

But what happens when we plot the true stress versus the true strain? The picture changes completely. The true stress curve keeps rising, all the way to fracture! The material isn't getting weaker at all; it's continuously getting stronger. This remarkable phenomenon is called ​​strain hardening​​ or ​​work hardening​​.

The difference isn't trivial. It can be enormous. Consider a real tensile test on a metal specimen. Well into the test, after the engineering stress has peaked and dropped to about $320 \text{ MPa}$, the corresponding true stress within the material has soared to nearly $885 \text{ MPa}$! Near the point of fracture, the engineering stress can under-predict the actual stress experienced by the material by more than 45%. Relying on the engineering curve to understand the material's state would be like trying to guess the speed of a race car by only looking at the driver's heart rate—it's related, but it misses the direct physics of the situation.

So, what causes this strain hardening? When we plastically deform a crystalline metal, we are forcing tiny line defects called ​​dislocations​​ to move through the crystal lattice. As deformation proceeds, these dislocations multiply and get tangled up with each other, forming complex, gridlocked structures. This microscopic traffic jam makes it progressively harder for other dislocations to move, so a greater stress is required to produce further strain. In a sense, the material's internal structure adapts to resist the deformation. It learns from its experience and becomes stronger. Because the true stress-strain curve captures this intrinsic material response, it is the basis for creating accurate mathematical models of material behavior, for example by converting a known engineering stress-strain relationship into its true form.

This leaves us with a beautiful paradox. If the material is continuously getting stronger (as the true stress curve shows), why does the force we need to apply eventually decrease (as the engineering stress curve shows)?

The answer lies in a dramatic competition. As we pull on the specimen, two opposing effects are in play:

1. ​​Strengthening:​​ The material strain-hardens, increasing its resistance to flow. This effect, on its own, would require an ever-increasing force to continue the deformation.
2. ​​Weakening:​​ The specimen's cross-sectional area decreases as it elongates. This geometric effect means a smaller area is available to support the load, which, on its own, would make it easier to continue pulling.

At the beginning of the plastic deformation, hardening wins. The material strengthens faster than it thins. The load required, $F = \sigma_{\mathrm{true}} A$, goes up. However, the rate of hardening doesn't stay constant. As the dislocation tangles become denser, a competing mechanism called ​​dynamic recovery​​ kicks in, where some dislocations manage to rearrange and annihilate each other. This causes the hardening rate—the slope of the true stress-strain curve—to decrease with increasing strain.

Eventually, a critical point is reached. The hardening rate drops to a level where it can no longer compensate for the rapid reduction in area. At this precise moment, the geometric weakening effect takes over. The total force required to continue deformation reaches its maximum and begins to fall. This is the UTS on the engineering curve.

This is the onset of ​​plastic instability​​, or ​​necking​​. From this point onwards, the deformation is no longer uniform along the specimen. It concentrates in the weakest spot, which begins to thin down rapidly like the waist of an hourglass, leading swiftly to fracture. This elegant balance is captured by the celebrated ​​Considère criterion​​, which states that instability begins when the rate of hardening exactly equals the value of the true stress:

For many materials whose behavior can be described by a simple power law called the ​​Hollomon equation​​ ($\sigma_T = K \epsilon_T^n$), this criterion leads to a wonderfully simple result: the true strain at which necking begins is numerically equal to the material's strain-hardening exponent, $n$. This demonstrates the incredible predictive power that arises from understanding these competing effects.

We have built a powerful picture, but a true scientist always asks: "Is this the whole story?" Let's refine our understanding one last time. Our ability to convert between engineering and true measures, and to derive the necking criterion, hinges on a key assumption: ​​plastic incompressibility​​. When metals are deformed plastically, their volume remains almost perfectly constant. Stretching in one direction is compensated by contraction in the other directions, so we can say $A_0 L_0 \approx A L$. This allows us to calculate the shrinking area from the measured elongation. This assumption is excellent for large plastic strains, though for the initial, small elastic strains, a more precise calculation would use the material's ​​Poisson's ratio​​ to find the area change.

But what happens after necking starts? The simple picture begins to break down. The smooth cylindrical geometry is gone. Inside the curved neck, the stress is no longer a simple, uniform tension. The curvature induces a complex ​​triaxial stress state​​—a combination of pulling along the axis and pulling inward from the sides.

This means that the "true stress" we calculate simply as Force/Area at the neck is no longer the true flow stress governing the material's plastic deformation. It's an approximation. To find the genuine material response at these extreme strains, scientists must use analytical corrections, like the ​​Bridgman correction​​, or sophisticated modern techniques like ​​Digital Image Correlation (DIC)​​ paired with computer simulations.

This final complexity does not diminish the beauty of our model; it enhances it. It shows us that true stress and strain are not just mathematical tricks. They are the first and most crucial step toward revealing the intrinsic behavior of a material. They take us from the engineer's external view to the physicist's internal reality, uncovering the hidden story of a material's struggle and ultimate failure, written in the language of force, geometry, and the beautiful, complex dance of dislocations.

In the previous chapter, we embarked on a conceptual journey, distinguishing the "apparent" story of a material's deformation, told by engineering stress and strain, from the "real" story, narrated by true stress and strain. We saw that by accounting for the changing shape of a material as it deforms, we get a truer picture of the forces its internal structure is actually withstanding. This might seem like a mere academic correction, a bit of mathematical housekeeping. But it is not. This shift in perspective is a key that unlocks a vastly deeper understanding of why materials behave the way they do. It transforms our ability to predict their fate, from the slow sag of a beam over decades to the violent shattering of a plate in a microsecond.

In this chapter, we will explore the marvelous applications of this "true" perspective. We will see how it allows us to foresee failure, to unify seemingly disparate properties, to build virtual worlds for engineering design, and to peer into the microscopic chaos that gives materials their strength. The true stress-strain curve is not just a graph; it is a fundamental language for conversing with the physical world.

Watch someone stretch a piece of taffy or a rubber band until it snaps. You will notice a fascinating event just before it breaks: a small region will begin to narrow dramatically, a phenomenon engineers call "necking." This localized thinning is a form of instability, a point of no return. Once the neck forms, all subsequent stretching concentrates there, leading swiftly to rupture. For an engineer designing a bridge cable or an artificial hip joint, knowing when this instability will occur is not a matter of curiosity—it is a matter of life and death.

The engineering stress-strain curve registers this event as its highest point, the Ultimate Tensile Strength (UTS). But it doesn't explain it. It tells us what happens, but not why. The true stress-strain curve gives us the 'why'. The instability of necking is a dramatic competition: as we pull on the material, its cross-section shrinks, which tends to increase the stress. At the same time, the material "work hardens"—it becomes stronger and more resistant to further stretching. As long as the rate of hardening is greater than the rate of stress increase due to area reduction, the deformation remains stable and uniform. The tipping point—the onset of necking—occurs precisely when the material's capacity to harden can no longer compensate for its shrinking cross-section.

Remarkably, this complex physical argument boils down to a beautifully simple mathematical condition first articulated by Armand Considère. The instability begins when the slope of the true stress-strain curve becomes equal to the value of the true stress itself: $\frac{\mathrm{d}\sigma_T}{\mathrm{d}\epsilon_T} = \sigma_T$.

For many metals, the plastic portion of the true stress-strain curve is wonderfully described by a simple power-law relationship known as the Hollomon equation: $\sigma_T = K \epsilon_T^n$, where $K$ is the strength coefficient and $n$ is the all-important strain-hardening exponent. If we apply Considère's criterion to a material that follows this law, we stumble upon a result of profound elegance: necking begins when the true strain is exactly equal to the strain-hardening exponent.

$\epsilon_T = n$

Think about what this means! A single number, $n$, which describes the shape of the material's 'true' response curve, directly predicts the macroscopic point of catastrophic failure. A material with a high $n$ value hardens rapidly and can endure a large amount of uniform stretching before it begins to neck. A material with a low $n$ value will surrender to instability much sooner. By carefully measuring the true stress-strain curve from a tensile test, we can extract these parameters and build predictive models of material behavior.

Imagine you want to assess the strength of a large, critical steel component, perhaps a support column in a power plant. You can't just cut a piece out of it to perform a tensile test. Is there a less destructive way? What if you could learn about its strength simply by pressing a tiny, sharp diamond into its surface? This is the essence of a hardness test. Intuitively, it feels like a harder material should also be a stronger one. The concept of true stress provides the rigorous link that confirms this intuition.

The eminent physicist David Tabor showed that the measured hardness of a metal is, to a good approximation, directly proportional to its true flow stress, $\sigma_f$. The flow stress is simply the true stress required to cause the material to flow plastically. This relationship is often expressed as $H_V \approx C \sigma_f$, where $C$ is a constraint factor (typically around 3) that accounts for the complex, three-dimensional stress state beneath the indenter.

But what is this flow stress? It's a point on the material's true stress-strain curve! An indentation is essentially a tiny, localized compression test. The material beneath the indenter is forced to deform plastically, and the resistance it puts up is dictated by its true stress-strain behavior. This reveals a beautiful unity: the same intrinsic material property that governs how a bar elongates also governs how it resists a sharp point. Furthermore, if we take a piece of metal and pre-strain it by stretching it, we are moving it along its true stress-strain curve to a higher flow stress. Tabor's relation correctly predicts that this work-hardened material will now be harder. This principle is the foundation for a vast range of quality control and material inspection techniques used every day across countless industries.

So far, we have spoken of simple pulling. But the real world is rarely so simple. A bolt is twisted and pulled simultaneously. The skin of an airplane is pressurized and sheared by aerodynamic forces. A car chassis during a crash is subjected to a terrifyingly complex combination of bending, twisting, and crushing. How can our simple, one-dimensional true stress-strain curve possibly help us here?

It turns out that the uniaxial true stress-strain curve is our 'Rosetta Stone'. It contains the essential, intrinsic information about the material's plastic character. By embedding this information within a more powerful mathematical framework—the theory of continuum plasticity—we can make predictions for any arbitrary three-dimensional state of stress.

Models like the Prandtl-Reuss equations, combined with a yield criterion like that of von Mises, form the engine of modern computational mechanics. We can perform a single, careful uniaxial tensile test to extract the fundamental elastic parameters (Young's modulus $E$ and Poisson's ratio $\nu$) and the plastic parameters (the initial yield stress $\sigma_{y0}$ and the hardening modulus $H$). Once these are known, the theory allows us to construct a "yield surface" in a multi-dimensional stress space. This surface represents the boundary between elastic and plastic behavior for any combination of stresses. The theory also provides a "flow rule" that predicts the direction and magnitude of plastic deformation once the boundary is crossed.

With this framework, we can ask questions like: if we know the yield stress from pulling on a bar, what shear stress will cause a shaft made of the same material to yield? The theory gives a precise answer (for a von Mises material, it's the tensile yield stress divided by $\sqrt{3}$). Or, if we combine tension and torsion, in what direction will the material flow? The theory provides the exact ratio of strain components. Remarkably, these predictions, born from a simple 1D test, hold up with stunning accuracy when compared to complex multiaxial experiments. This ability to generalize from simple tests to complex realities is what makes modern engineering simulation, such as the Finite Element Method (FEM), possible.

Our picture grows more complex, and more interesting, when we consider speed. Pulling a steel bar apart over a minute is one thing; having a projectile slam into it in a microsecond is quite another. At high strain rates, two new characters enter the stage: strain-rate sensitivity and temperature. Most materials get stronger the faster you try to deform them. At the same time, the immense energy of this rapid plastic deformation has nowhere to go. It is converted almost entirely into heat, a process known as adiabatic heating. A piece of metal in a high-speed impact can heat up by hundreds of degrees in an instant, causing it to soften dramatically.

In this extreme environment, the distinction between engineering and true measures ceases to be a subtle correction and becomes a physical necessity. Why? Because of energy. The fundamental law of thermodynamics states that the work done on a system changes its internal energy. The plastic work done per unit volume, which is the source of adiabatic heating, is correctly calculated only by integrating the true stress over the true plastic strain increment: $W_p = \int \sigma_T \mathrm{d}\epsilon_p$. True stress and true strain are "work-conjugate." Using engineering stress and strain would lead to a completely erroneous calculation of the energy dissipated and the resulting temperature rise, rendering any prediction useless.

For this reason, advanced constitutive models like the Johnson-Cook model, which are designed to predict material behavior in car crashes, ballistic events, and high-speed machining, are formulated exclusively in terms of true stress and true strain. Calibrating these models requires sophisticated high-strain-rate experiments (like those using a Split Hopkinson Pressure Bar), and the data must be converted to true measures to capture the underlying physics correctly.

Many components in our world are not designed to withstand a single, large force, but to endure millions of small, repetitive ones. The wings of an aircraft flex with every gust of wind; the crankshaft in an engine is twisted and bent with every revolution. This cyclic loading leads to a gradual accumulation of damage known as fatigue, the silent killer of mechanical structures.

How does a material's true stress-strain behavior relate to fatigue? One might be tempted to use the curve from a standard tensile test, but this would be a mistake. A material's response to being cyclically pushed and pulled is often profoundly different from its response to a single monotonic pull. Some materials get stronger with each cycle (cyclic hardening), while others get weaker (cyclic softening).

After some initial evolution, most materials settle into a "stabilized" state, where each subsequent cycle of loading and unloading traces out a consistent "hysteresis loop" on a true stress-true strain plot. The tips of these stabilized loops for tests run at different strain amplitudes form a unique curve: the ​​stabilized cyclic stress-strain curve​​. This curve has a form analogous to the monotonic one, $\epsilon_a = \frac{\sigma_a}{E} + (\frac{\sigma_a}{K'})^{1/n'}$, but with its own distinct set of cyclic parameters, $K'$ and $n'$, which are generally not the same as their monotonic counterparts, $K$ and $n$. This cyclic curve is the true descriptor of the material's mechanical state during its fatigue life. It is the cornerstone of the modern "strain-life" approach to fatigue analysis, which allows engineers to predict how many cycles a component can survive before a crack initiates.

The true stress-strain curve is a masterful description of a material's behavior, but it's a macroscopic one. It doesn't, by itself, explain the "why" at a deeper level. To do that, we must zoom in, past the scale of grains, to the world of the atomic lattice. Here we find that plastic deformation is not a smooth, continuous flow, but the collective motion of countless line-like defects called dislocations.

Work hardening—the very phenomenon described by the shape of the true stress-strain curve—is the result of a microscopic "traffic jam." As dislocations glide on atomic planes, they run into each other, get tangled, and form complex networks that impede further motion. More stress is then required to push new dislocations through this increasingly dense and frustrating tangle.

Physically-based models, such as the Kocks–Mecking or Voce models, attempt to capture this dislocation drama mathematically. A powerful way to connect these microscopic theories to our macroscopic measurements is through a "Kocks plot," where we plot the instantaneous rate of hardening, $\theta = \mathrm{d}\sigma/\mathrm{d}\epsilon_p$, against the current true stress, $\sigma$. The shape of this plot is a direct fingerprint of the underlying hardening mechanisms. For many metals, a significant portion of this plot is a straight line, which points to a specific type of dynamic equilibrium between the generation and annihilation of dislocations. This allows us to extract more physically meaningful parameters than the purely empirical $n$ and $K$ from the Hollomon law, or to use more refined empirical models like the Ludwik law which accounts for an initial yield stress. This analysis provides a stunning bridge between the line we draw on a graph in the lab and the intricate dance of defects deep within the material's crystal structure.

Our story must eventually confront the end: fracture. Why, ultimately, do things break? Even if a material work-hardens, why can't it just keep getting stronger and thinner forever? It is because real materials are not perfect. They contain microscopic voids, pores, or inclusions.

As a ductile material is stretched, these voids are pulled and begin to grow. The material's overall response becomes a tug-of-war. On one side, the metallic matrix between the voids is work-hardening, trying to make the material stronger. On the other side, the growth of the voids constitutes a form of damage, reducing the effective load-bearing area and tending to soften the material.

The Gurson-Tvergaard-Needleman (GTN) model is a beautiful theory that captures this competition. It modifies the yield criterion to account for the presence of voids. The result is a model where the macroscopic stress-strain response depends on both the hardening of the matrix (described by its own true stress-strain curve) and the growth of the void volume fraction. Initially, matrix hardening dominates, and the true stress rises. But as the deformation proceeds and the voids grow larger, the softening effect from the damage begins to accelerate. Eventually, the softening wins the tug-of-war. The material loses its ability to sustain a higher load, the true stress reaches a peak, and then begins to fall, leading rapidly to the final failure as the voids link up. This sophisticated view, built upon the foundation of true stress and strain, provides a profound explanation for the process of ductile fracture from first principles.

We began with a simple idea: to measure stress and strain relative to a material's changing, instantaneous state. From this seed has grown a rich and powerful tree of knowledge. The language of true stress and strain allows us to predict the onset of instability, to relate strength to hardness, to build predictive 3D simulations from simple tests, to master the extreme realms of high-speed impacts, to design components that last a lifetime, to connect macro-behavior to micro-physics, and to understand the very process of fracture. It is a testament to how, in science, a commitment to describing reality more truthfully often rewards us with a far deeper and more unified vision of the world.