Borghi-Peters diagram

The dance between fire and turbulence is a captivating yet profoundly complex phenomenon that governs the performance of everything from car engines to power plants. Predicting how a flame will behave—whether it will be stable, how fast it will burn, and what shape it will take—when subjected to a chaotic, swirling flow presents a major challenge for scientists and engineers. A simple description of 'wind' and 'fire' is insufficient; we need a systematic way to classify this interaction.

To address this challenge, the Borghi-Peters diagram was developed as a conceptual map for turbulent combustion. This article provides a comprehensive overview of this essential tool. The first chapter, ​​Principles and Mechanisms​​, delves into the fundamental physics behind the diagram, explaining how the battle between chemical reactions and turbulent mixing is distilled into key dimensionless parameters like the Damköhler and Karlovitz numbers, which define distinct combustion regimes. The subsequent chapter, ​​Applications and Interdisciplinary Connections​​, explores the diagram's crucial role in the modern world, from guiding the choice of models in high-fidelity computer simulations to reconciling experimental observations with theoretical predictions. By the end, you will understand how this elegant map brings order to the fiery chaos of turbulent flames.

Imagine trying to light a candle in a breeze. A gentle draft might make the flame dance and flicker, but it remains lit. A stronger, gusty wind, however, might blow it out entirely. This simple experience holds the key to a deep and beautiful problem in physics and engineering: the intricate dance between fire and turbulence. What determines whether the flame survives, how fast it spreads, and what shape it takes? To answer this, we can't just think about "the wind"; we must appreciate that turbulence is a complex hierarchy of swirling eddies, from large gusts down to tiny, invisible whorls. Similarly, a "flame" is not just a blob of light; it's a delicate, self-propagating wave of chemical reaction sustained by the transport of heat and fuel. The interaction between these two complex phenomena governs everything from the efficiency of a car engine to the safety of an industrial furnace.

To bring order to this complexity, scientists like Antoine Borghi and Norbert Peters developed a powerful tool: a map. But instead of longitude and latitude, this map uses special coordinates that capture the essence of the battle between chemistry and fluid motion. This "map of combustion" is the Borghi-Peters diagram, and it allows us to see, at a glance, the entire landscape of possible flame behaviors.

Let's begin by characterizing our two dancers. A flame, left to its own devices in a still mixture of fuel and air, has two key properties. First, it has a ​​laminar flame speed​​, which we'll call $s_L$. This is the speed at which it naturally advances, like the speed of a fire front spreading across a quiet field. Second, it has a ​​laminar flame thickness​​, $\delta_L$, which is the width of the zone where the chemistry and heating actually happen. From these, we can define a fundamental ​​chemical timescale​​, $\tau_c \approx \delta_L / s_L$. This is, roughly, the time it takes for the flame to complete its chemical business and advance by one flame thickness. It's the flame's natural rhythm.

Now for the turbulence. The most obvious feature of a turbulent flow is its intensity, the root-mean-square velocity fluctuation, $u'$. The largest, most energetic swirls, or eddies, have a characteristic size, the integral length scale, $l$. These large eddies define a ​​turbulent timescale​​, $\tau_t \approx l / u'$, which represents the time it takes for a large gust to turn over or traverse its own size.

The first great question we can ask is: which is faster, the chemistry or the large-scale turbulent mixing? The ratio of these two timescales gives us our first crucial dimensionless number, the ​​Damköhler number​​, $Da$:

The Damköhler number tells us about the overall stability of the flame.

• If ​​$Da \gg 1$​​, the turbulent mixing is much slower than the chemistry ($\tau_t \gg \tau_c$). The flame has plenty of time to burn through a pocket of fuel before a large eddy can tear it apart. The flame is robust and will persist, though it will be wrinkled and stretched by the flow. This is the hallmark of the ​​flamelet regimes​​.

• If ​​$Da \ll 1$​​, the turbulent mixing is much faster than the chemistry ($\tau_t \ll \tau_c$). The turbulence shreds and disperses the fuel and hot products so quickly that the flame doesn't have time to establish a stable, propagating front. The flame structure is torn asunder, and the reaction becomes a disorganized, soupy mess distributed throughout a volume. This is the ​​broken reaction​​ or ​​distributed reaction​​ regime, and in this state, the flame is very fragile and can be easily extinguished.

The Damköhler number gives us the big picture, but it doesn't tell the whole story. Turbulence is not just made of large eddies. According to the celebrated theory of Andrey Kolmogorov, large eddies break down into a cascade of smaller and smaller eddies, until they become so small that their energy is dissipated into heat by viscosity. The smallest of these eddies have a size called the ​​Kolmogorov length scale​​, $\eta$, and a characteristic time, the ​​Kolmogorov timescale​​, $\tau_\eta$.

These tiny, fast-moving eddies pose a new kind of threat. Are they small enough to get inside the flame structure? To answer this, we compare the flame's chemical timescale, $\tau_c$, with the Kolmogorov timescale, $\tau_\eta$. This ratio defines the ​​Karlovitz number​​, $Ka$:

The Karlovitz number tells us if the flame's internal structure is safe from the meddling of the smallest eddies.

• If ​​$Ka \ll 1$​​, the chemical processes are much faster than even the smallest eddies ($\tau_c \ll \tau_\eta$). Equivalently, the flame thickness is much smaller than the smallest eddies ($\delta_L \ll \eta$). The turbulence, at all its scales, remains outside the flame front. The flamelet picture holds perfectly.

• If ​​$Ka \ge 1$​​, the smallest eddies are now faster than the chemical processes ($\tau_c \ge \tau_\eta$), and smaller than the flame thickness ($\eta \le \delta_L$). This is a critical transition! These tiny eddies can now invade the flame's inner sanctum. A premixed flame has a layered structure: a relatively broad preheat zone, where the incoming cold gas is heated by diffusion, and a much thinner inner reaction layer, where the chemistry ignites. When $Ka$ first exceeds unity, the eddies ($\eta$) are smaller than the preheat zone ($\delta_L$) but may still be larger than the reaction layer. They penetrate the preheat zone, stirring it up and dramatically enhancing the transport of heat and species. This turbulent stirring steepens local temperature gradients, increasing the ​​scalar dissipation rate​​ and marking the breakdown of the simple laminar flamelet assumption. The flamelet is no longer a simple 1D structure.

We now have all the pieces to construct our map. The Borghi-Peters diagram is typically plotted on a log-log scale with the velocity ratio $u'/s_L$ on the vertical axis and the length scale ratio $l/\delta_L$ on the horizontal axis. These two ratios beautifully summarize the state of the system.

The boundaries separating the different combustion behaviors are simply lines of constant $Da$ and $Ka$.

The three most important boundaries are:

1. ​​$Da = 1$​​: This line, given by the equation $u'/s_L = l/\delta_L$, separates the stable flamelet regimes from the unstable broken reaction regime.
2. ​​$Ka = 1$​​: This curve, which on this plot follows the equation $u'/s_L = (l/\delta_L)^{1/3}$, marks the line where the smallest eddies begin to penetrate the flame's internal structure. It is the border between the classical flamelet regimes and the thin reaction zones regime.
3. ​​$u'/s_L = 1$​​: This horizontal line provides a useful distinction within the flamelet regime. When $u' \lt s_L$, the turbulence is weak and only gently wrinkles the flame. When $u' \gt s_L$, the turbulence is strong enough to cause large-scale corrugations.

With our map in hand, we can take a tour of the different territories of turbulent combustion. Imagine we are in a laboratory, and we can control the turbulence ($u', l$) interacting with a specific flame ($s_L, \delta_L$).

• ​​Wrinkled and Corrugated Flamelets ($Ka \ll 1$)​​: Here, turbulence is too slow and its smallest eddies are too large to affect the flame's internal structure. The flame remains a thin, continuous sheet. If turbulence is weak ($u' s_L$), the sheet is gently ​​wrinkled​​. If turbulence is strong ($u' > s_L$), the sheet becomes highly folded and ​​corrugated​​, which massively increases its surface area and thus the overall burning rate.

• ​​Thin Reaction Zones ($Ka > 1, Da > 1$)​​: As we increase turbulence intensity, we cross the $Ka=1$ boundary. We are now in a new and fascinating territory. The smallest eddies ($\eta$) are now smaller than the flame's preheat thickness ($\delta_L$). They can invade the preheat zone, but they are not yet small enough to disrupt the much thinner core reaction layer. The result is a flame whose preheat zone is broadened and "puffed up" by turbulent transport, while its chemical heart remains an intact, albeit heavily strained and contorted, thin sheet. Consider a practical example with a methane-air flame where calculations give a Damköhler number $Da \approx 2$ and a Karlovitz number $Ka \approx 13$. This combination of $Da > 1$ and $Ka > 1$ places it squarely in the ​​thin reaction zones​​ regime. This is a common state in many practical devices like gas turbines.

• ​​Broken Reactions ($Da \lesssim 1, Ka \gg 1$)​​: If we crank up the turbulence even further, we eventually cross the $Da=1$ boundary into the most chaotic regime. Here, the turbulence is so fast at all scales that it completely overwhelms the chemistry. The large eddies are fast enough to prevent a stable front from forming ($Da \ll 1$), and the small eddies are so tiny they can shred even the inner reaction layer ($Ka \gg 1$). The very concept of a "flame front" dissolves. Instead, we have a volume of intensely mixed reactants, hot products, and intermediate species, where reactions occur in a distributed, disorganized fashion. This is the ​​distributed reaction regime​​, a state where the flame is on the verge of being blown out completely.

The beauty of this framework is that it can be enriched with even more physical detail.

• ​​The Gibson Scale​​: We can ask: what is the size of an eddy whose characteristic velocity is exactly equal to the flame's own speed, $s_L$? This defines a special length scale called the ​​Gibson scale​​, $l_G$. Eddies larger than $l_G$ have velocities greater than $s_L$ and are thus strong enough to wrinkle the flame front. Eddies smaller than $l_G$ are too feeble; the flame propagates faster than they can stir, effectively smoothing out their influence. The Gibson scale provides a physical threshold for which eddies are relevant for wrinkling the flame.

• ​​The Lewis Number​​: So far, we have implicitly assumed that heat and fuel diffuse at the same rate. But what if they don't? The ratio of thermal diffusivity to mass diffusivity is called the ​​Lewis number​​, $Le$. If $Le 1$, the fuel diffuses into the reaction zone faster than heat can escape. This enriches the reaction, making the flame hotter and faster ($s_L$ increases). If $Le > 1$, the opposite occurs, and the flame becomes weaker and slower. Changing the Lewis number of the fuel mixture fundamentally changes the flame's properties ($s_L, \delta_L$) and therefore shifts all the regime boundaries on the Borghi-Peters diagram! This reveals a deeper unity: the macroscopic behavior of a turbulent flame is tied not only to the flow but also to the most fundamental molecular transport properties of the gas itself.

By starting with simple questions about time and length, we have constructed a complete and predictive map. The Borghi-Peters diagram is a testament to the power of dimensional analysis in physics, transforming a problem of bewildering complexity into an elegant and intuitive landscape, allowing us to understand and ultimately design the engines and furnaces that power our world.

Having journeyed through the principles that give the Borghi-Peters diagram its structure, we might now ask the most important question of all: "So what?" What good is this beautiful map of fire? It turns out this is not merely an academic exercise in classification. The diagram is a powerful and indispensable tool, a bridge connecting the abstract world of theory with the practical realms of engineering design, experimental physics, and the frontiers of scientific simulation. It is our guide for navigating the complex, fiery dance of turbulence and chemistry.

Imagine you are an engineer tasked with designing a new, more efficient gas turbine engine for an airplane, or a cleaner internal combustion engine for a car. You cannot afford to build hundreds of physical prototypes; it would be too slow and expensive. Instead, you turn to the immense power of computer simulations—what we call Computational Fluid Dynamics, or CFD—to test your designs. But here you face a dilemma. The dance between the turbulent flow and the achemical reactions of the flame is happening at scales impossibly small and fast to capture fully on any computer, even a supercomputer. The intricate flame front might be thinner than a human hair, while the engine combustor is the size of a coffee can. Resolving everything would take millennia.

So, you must make a choice. You must use a simplified model for the flame. But which one? Do you model the flame as a thin, wrinkled sheet? Or as a region where fuel and air are rapidly mixed before reacting? Making the wrong choice means your simulation will be nonsense, leading to a faulty design.

This is where the Borghi-Peters diagram becomes your navigator's chart. By calculating a few key parameters of your engine's environment—the intensity of the turbulence ($u'$), the size of the largest swirls ($L$), and the basic properties of your fuel flame ($S_L$ and $\delta_L$)—you can compute your position on the map. This position tells you the regime your flame lives in.

If your calculations land you in the "corrugated flamelets" or "thin reaction zones" regimes, the chart tells you that flamelet-based models are the way to go. These models treat the turbulent flame as a collection of thin, laminar-like flame structures that are wrinkled and stretched by the flow. Advanced techniques like the Artificially Thickened Flame (ATF) model or Flamelet-Generated Manifolds (FGM) are built on this very assumption. The diagram gives you the confidence to use them, telling you that their underlying physical picture is sound for your specific problem. For instance, knowing you are in the thin reaction zones regime tells you that you must choose a model that not only accounts for the wrinkling of the flame surface but also for the way small turbulent eddies enhance heat transport within the flame's preheat zone.

Conversely, the diagram also erects crucial warning signs. Suppose your calculations place you in the "broken" or "distributed reaction" regime, where the Karlovitz number ($Ka$) is enormous. Here, the diagram warns you that the very concept of a "flame surface" has broken down. The smallest turbulent eddies are so viciously energetic that they have torn the flame apart from the inside, smearing the reaction out into a volumetric mess. In this case, a model based on Flame Surface Density (FSD), which presumes a well-defined surface exists, would be physically meaningless and is guaranteed to fail. The diagram doesn't just tell you which tools to use; it tells you which ones to leave in the toolbox.

The diagram's utility extends far beyond the world of simulation. It is a powerful predictive tool that helps us understand how a flame's character will change when its environment changes. Consider the immense difference between a flame burning in the open air of a laboratory and one burning inside the high-pressure chamber of a modern jet engine. The pressure might increase by a factor of 10, 20, or even more. How does the fire change its nature?

We can use our knowledge of physics and chemistry to create scaling laws that predict how a flame's properties, like its speed $S_L$ and thickness $\delta_L$, change with pressure. As pressure rises, flames typically get slower and much, much thinner. When we plug these new, high-pressure properties into our calculations for the Damköhler and Karlovitz numbers, we can plot the flame's journey across the diagram.

What we often find is that as pressure increases, a flame that was a gentle, wrinkled flamelet at atmospheric pressure moves dramatically into the thin reaction zones or even the broken reaction zones regime. The Karlovitz number can skyrocket. This tells the engineer a critical story: the physics governing the flame inside the real engine is qualitatively different from what is easily studied in the lab. The interaction with turbulence is far more intense and intimate. This insight is fundamental for designing stable and efficient high-pressure combustors, guiding engineers to account for these drastic shifts in flame structure.

Perhaps the most profound role of the Borghi-Peters diagram is as a common language—a Rosetta Stone—that allows us to compare and reconcile the three great pillars of modern science: theory, experiment, and simulation. Sometimes, these three pillars tell what appear to be conflicting stories, and the diagram is our key to unraveling the mystery.

Imagine a scenario where a laboratory experiment, using advanced laser imaging techniques like Planar Laser-Induced Fluorescence (PLIF), captures images of a flame that looks thick, fragmented, and "distributed." At the same time, a powerful Direct Numerical Simulation (DNS), which solves the fundamental equations of fluid motion and chemistry, shows the flame to be a highly wrinkled but largely continuous thin sheet. And to top it off, a theoretical analysis using the Borghi-Peters diagram agrees with the simulation, placing the flame squarely in the "thin reaction zones" regime. Who is right?

The answer, beautifully, is that everyone is right, from their own point of view. The diagram forces us to think critically about what we are really seeing. An experimentalist isn't seeing the flame directly. They are seeing light emitted or fluoresced from specific molecules. For example, if they image formaldehyde ($\text{CH}_2\text{O}$), a species that exists in the relatively thick preheat zone, they will see a broad, distorted layer, because in the thin reaction zones regime, it is precisely this preheat zone that the small eddies are penetrating and disrupting. Furthermore, the camera has a finite resolution and integrates all the light along its line-of-sight. A very thin, intricately folded flame sheet can easily appear as a thick, blurry brush when viewed this way.

On the other hand, the simulation might have its own simplifications. Perhaps it used a single-step chemical reaction scheme or assumed that all chemical species diffuse at the same rate as heat (the unity Lewis number assumption). These idealizations can make the simulated flame more "robust" and less susceptible to being torn apart than the real methane-air flame.

The Borghi-Peters diagram acts as the objective arbiter. It provides the physical context (e.g., "thin reaction zones") that allows us to understand why the experiment looks the way it does (diagnostic artifacts and species choice) and why the simulation behaves as it does (physical model simplifications). It guides us to a deeper truth by forcing us to reconcile these different viewpoints.

Science never stands still, and neither does our map of fire. The classic Borghi-Peters diagram was developed for the idealized case of a perfectly uniform mixture of fuel and air. But in many real-world devices, from diesel engines to modern gas turbines, the mixture is "stratified" or "partially premixed"—some parts are fuel-rich, some are fuel-lean, and some are perfectly stoichiometric.

Here, the diagram evolves. We can use it to track the life of a flame as turbulence intensity increases. A flame might start as a well-behaved wrinkled flamelet, but as the turbulence gets stronger, it crosses the $Ka=1$ line and becomes a thin reaction zone. As the turbulence gets stronger still, it might cross the $Da=1$ line. This transition is fascinating: it suggests the flame's character is changing from one controlled by the speed of chemical reactions to one limited by the speed of turbulent mixing, behaving more like a non-premixed flame.

For the most complex cases, we may need to invent new dimensions for our map. In a truly partially premixed flame, two processes happen at once: non-premixed mixing creates flammable pockets, and then a premixed-like flame propagates through them. Which is more important? To answer this, scientists have proposed using two different Damköhler numbers: one that compares turbulence to the propagation timescale ($Da_{\text{prop}}$), and another that compares it to the mixing timescale ($Da_{\text{mix}}$). The flame now has a location in a multi-dimensional regime space. By calculating our position in this expanded map, we can understand whether the overall burning is controlled by propagation, by mixing, or, most interestingly, by a hybrid of both.

From a practical tool for choosing a computer model to a profound framework for uniting experiment and theory, the Borghi-Peters diagram is far more than a static chart. It is a living, evolving concept that embodies our ever-deepening understanding of the beautiful and complex physics of fire. It reminds us that in science, the most elegant pictures are often the most useful.