On the flash temperature in sliding contacts

This paper presents a new analytical theory for flash temperature in sliding contacts that accounts for the multiscale roughness of real surfaces, demonstrating that classical models based on simplified heat source geometries fail to accurately predict temperature increases in such scenarios.

The Gist

The Big Idea: The "Flash" in Flash Temperature

Imagine you are rubbing your hands together quickly to warm them up. You feel a sudden, intense heat in the specific spots where your skin touches. That instant, localized spike of heat is what scientists call "flash temperature."

In the real world, this happens whenever two solid objects slide against each other—like car tires on a road, ice skates on ice, or even the tectonic plates shifting during an earthquake. The friction turns kinetic energy (movement) into heat. Because the contact points are tiny and the movement is fast, the heat doesn't have time to spread out; it gets trapped, creating a microscopic "hotspot" that can reach temperatures high enough to melt rock or ice in a split second.

The Problem: The Old Maps Were Wrong

For decades, scientists used "classical theories" (developed by experts like Jaeger, Archard, and Greenwood) to predict how hot these spots get.

The Analogy: Imagine trying to map a city's traffic jams. The old theories assumed the city was made of perfect, round circles (like roundabouts) or squares. They assumed the "traffic" (heat) came from smooth, uniform sources.

The Reality: Real surfaces are nothing like smooth circles. They are like mountain ranges. If you zoom in on a rock or a piece of rubber, it looks bumpy. If you zoom in further, those bumps have smaller bumps on them, and those have even smaller bumps. It's a fractal structure with roughness at every single scale, from the size of a grain of sand down to the size of an atom.

The old theories failed because they ignored this "multiscale" nature. They tried to smooth over the tiny, jagged peaks, which led to wildly inaccurate predictions about how hot things get.

The New Theory: A Multiscale Heat Map

The authors of this paper (Müser and Persson) created a new, sophisticated mathematical theory that accounts for roughness at all levels.

The Analogy: Instead of looking at the city as a few roundabouts, their new theory looks at the entire highway system, the side streets, the alleyways, and the driveways all at once. They realized that heat doesn't just come from one big contact point; it comes from thousands of tiny "micro-contacts" that are all heating up and influencing each other.

They developed a way to calculate the average temperature of these hotspots by looking at how the "roughness" of the surface correlates with the heat. Think of it as a recipe that says: "Take the roughness of the surface, mix it with the speed of the slide, and add the material's ability to conduct heat, and you get the exact temperature spike."

Key Findings: What They Discovered

1. The "Hot Track" Effect

When an object slides fast, the heat doesn't just stay under the contact point; it leaves a trail behind, like a hot iron moving across fabric.

• The Metaphor: Imagine a hot dog rolling over a blanket. The spot under the hot dog is scorching, but the blanket behind it is still warm for a moment.
• The Discovery: At high speeds, these "hot tracks" from previous contact points overlap with new contact points. This creates a cumulative heating effect that the old theories completely missed.

2. Rubber on Concrete vs. Granite on Granite

The authors tested their theory on two very different scenarios:

• Rubber on Concrete: Like a tire on a road. Rubber is soft and squishy. The theory showed that for rubber, the heat spreads out differently depending on how fast you go.
• Granite on Granite: This is crucial for earthquakes. When tectonic plates slide past each other, they are essentially giant blocks of granite grinding together.
  • The Shock: The old theories predicted that the rocks wouldn't get hot enough to melt. The new theory showed that because of the multiscale roughness, the frictional heat is so intense that the quartz in the granite can actually melt or turn into a glass-like substance at speeds typical of earthquakes (around 1 meter per second). This melting acts like a lubricant, causing the earthquake to slip more easily and suddenly.

3. Why the Old Math Failed

The paper proves that if a surface has roughness spanning many different sizes (which almost all real surfaces do), the classical "circular heat source" math is useless.

• The Metaphor: It's like trying to measure the volume of a sponge by pretending it's a solid block of wood. You get the wrong answer because you ignore all the holes and the complex internal structure.

Why Does This Matter?

1. Earthquakes: Understanding flash heating helps geophysicists understand why earthquakes happen and how they stop. If the rocks melt, the friction drops, and the ground slips violently.
2. Tires and Brakes: Engineers can design better tires and brake pads that don't overheat and fail, by understanding exactly how heat builds up on rough surfaces.
3. Ice Skating: It explains why ice is slippery. The flash temperature melts a thin layer of ice, creating a lubricating film of water under the skate.

The Bottom Line

This paper is a "fix" for a broken thermometer. For years, we tried to measure the heat of sliding objects using a ruler that assumed everything was smooth and round. The authors built a new, high-tech thermometer that understands that the world is rough, jagged, and complex. They showed that when things slide fast, the heat is much more intense and widespread than we ever thought, with massive implications for everything from your car's brakes to the safety of our planet during earthquakes.

Technical Summary

1. Problem Statement

Friction between sliding solids generates localized heat at asperity (microscopic contact) points, leading to rapid, transient temperature spikes known as flash temperatures. These spikes can reach several hundred to over 1000°C, significantly influencing friction coefficients, wear rates, and material properties (e.g., melting, phase transitions, or amorphization).

The Core Limitation: Classical theories for estimating flash temperature (developed by Jaeger, Archard, and Greenwood) rely on simplifying assumptions that fail for real-world surfaces. Specifically, they:

• Approximate frictional heat sources as isolated circular or square shapes.
• Ignore the multiscale nature of real surface roughness, which exists across many decades of length scales (from macro-asperities down to atomic scales).
• Neglect the thermal coupling and collective heating effects between closely spaced micro-asperities within a macro-contact region.

The authors argue that for surfaces with multiscale roughness, these classical approximations lead to severe errors in predicting flash temperatures, particularly in applications like rubber friction and earthquake dynamics (granite-on-granite sliding).

2. Methodology

The authors present a new analytical theory that accounts for surface roughness across arbitrarily many length scales.

• Theoretical Framework:

  • The problem is modeled as heat diffusion in a semi-infinite solid ($z > 0$) with a heat source $\dot{q}(x,t)$ on the surface $z=0$.
  • The heat source is assumed proportional to the local normal stress: $\dot{q} = \mu v \sigma(x - vt)$, where $\mu$ is the friction coefficient, $v$ is sliding speed, and $\sigma$ is the normal stress.
  • The solution utilizes Fourier transforms to solve the heat diffusion equation. The authors derive temperature-stress and temperature-temperature correlation functions.
  • Key inputs include the surface roughness power spectrum $C(q)$ and the stress correlation function, which incorporates the effective modulus $E^*$ and the relative contact area.

• Key Definitions:

  • Flash Temperature ($T_{flash}$): Defined not as a peak point temperature, but as a weighted average of the temperature over the contact area, weighted by the local contact stress:
    $$T_{flash} = \frac{\langle T(x)\sigma(x) \rangle}{\langle \sigma(x) \rangle}$$
  • Correlation Functions: The theory derives the spatial correlation of temperature fluctuations, $g_T(r)$, and compares them to stress correlations, $g_\sigma(r)$.
  • Derivatives: The theory calculates effective temperature gradients ($\nabla T_{flash}$) and Laplacians ($\nabla^2 T_{flash}$) to characterize the shape and width of the temperature profile within the contact.

• Numerical Implementation:

  • The theory is applied to specific systems: Rubber on Concrete and Granite (Quartz/Silica) on Granite.
  • Numerical results are generated using the measured power spectra of concrete and run-in granite surfaces.
  • The authors also perform numerical simulations of heat diffusion on periodic arrays of heat sources to validate the conceptual framework and illustrate thermal overlap effects.

3. Key Contributions

1. Multiscale Analytical Theory: The derivation of a closed-form analytical expression (Eq. 14) for flash temperature that is valid for surfaces with roughness spanning arbitrary decades of length scales. This generalizes previous single-scale models.
2. Demonstration of Classical Failure: The paper rigorously demonstrates that classical theories (Jaeger/Archard/Greenwood) fail severely when the surface roughness spans two or more decades of length scales. The classical models assume isolated heat sources, whereas the new theory accounts for the "hot tracks" and thermal coupling between micro-asperities.
3. Definition of Effective Flash Temperature: The introduction of a stress-weighted average temperature, which provides a more physically relevant metric for frictional heating than simple peak temperatures, especially when contact areas are fragmented.
4. Application to Earthquake Dynamics: The theory is applied to granite-on-granite sliding to explain the "dynamic weakening" observed during earthquakes. It links the drop in friction coefficient to flash heating-induced softening or melting of quartz grains.

4. Key Results

• Velocity Dependence:

  • At low sliding speeds ($v \to 0$), the mean square temperature fluctuation scales as $\langle T^2 \rangle \sim v^2$.
  • At high sliding speeds ($v \to \infty$), the scaling changes to $\langle T^2 \rangle \sim v \ln v$.
  • The transition between these regimes depends on the thermal diffusivity $D$ and the characteristic roughness length scales.

• Thermal Coupling and "Hot Tracks":

  • Unlike stress, which is confined to the contact area, the temperature distribution extends far beyond the contact regions due to heat diffusion, forming "hot tracks" behind moving asperities.
  • At high speeds, these tracks overlap with downstream asperities, significantly elevating the background temperature of the contact zone.
  • The normalized temperature correlation function decays much slower than the stress correlation function, indicating a much larger effective thermal interaction range.

• Comparison with Classical Theory:

  • For surfaces with roughness over one decade of length scales, the new theory converges with classical results.
  • For two or more decades, the classical theory underestimates the temperature width and mispredicts the scaling behavior.
  • The ratio of the effective width of the temperature profile ($D_{flash}$) to the slope length ($d_{flash}$) deviates significantly from the classical prediction (3.0) for multiscale surfaces, reaching values around 5.45 in the studied granite case.

• Earthquake Dynamics (Granite on Granite):

  • Using the measured friction coefficient drop (from $\mu \approx 0.8$ to $0.35$) at sliding speeds of $\sim 1$ m/s, the authors calculate the required flash temperature.
  • The theory predicts that flash temperatures reach the melting point of quartz ($\approx 1700$ K) at sliding speeds of $\approx 0.1$ m/s.
  • This supports the hypothesis that flash heating causes the local melting or amorphization of quartz grains, leading to a lubricating silica-like layer and a drastic reduction in friction (dynamic weakening).
  • The model explains why the friction coefficient drop is observed: the temperature rise softens the material before true melting, reducing shear stress.

• Material Dependence:

  • The study highlights that for materials like rubber (low thermal conductivity), the flash temperature is highly sensitive to the multiscale nature of the roughness.
  • For quartz/silica, the conversion of crystalline quartz to amorphous silica (due to high stress/temperature) reduces thermal conductivity by a factor of 10, further enhancing local heating.

5. Significance

This paper fundamentally advances the understanding of tribological heating by bridging the gap between idealized single-scale models and the complex reality of multiscale roughness.

• Theoretical Impact: It provides the first rigorous analytical framework to calculate flash temperatures for realistic engineering and geological surfaces without resorting to purely numerical, computationally expensive simulations for every case.
• Practical Applications:
  • Tribology: Improves predictions for wear and friction in rubber tires, brakes, and seals where multiscale roughness is critical.
  • Geophysics: Offers a robust physical mechanism for "dynamic weakening" in earthquakes. It validates the flash heating hypothesis, suggesting that the rapid drop in friction during seismic slip is driven by multiscale thermal effects leading to localized melting/amorphization of fault gouge.
• Methodological Shift: It shifts the paradigm from treating asperities as isolated heat sources to treating them as a thermally coupled system, where the collective behavior of micro-asperities dictates the macroscopic thermal response.

In conclusion, Müser and Persson demonstrate that ignoring the multiscale nature of surface roughness leads to severe inaccuracies in flash temperature estimation, with profound implications for both industrial friction control and the understanding of earthquake mechanics.