Pyrolysis isn't a single, neat chemical reaction. It's a chaotic, competitive process involving hundreds of simultaneous reactions. To make sense of it, we group them into two main stages: primary and secondary reactions.

Primary reactions are the initial thermal breakdown of the large biopolymers in biomass: cellulose, hemicellulose, and lignin. This is where the solid material first cracks apart, or depolymerizes, into smaller, volatile molecules. Think of it as the main event where the solid log turns into gases and vapors.

Secondary reactions happen immediately after. As these newly formed volatile molecules travel out of the solid particle, they can collide with each other or with hot surfaces. These collisions can cause them to react further. They might crack into even smaller gas molecules (like CO, CO₂, H₂), or they might react with each other to form larger molecules, eventually condensing into tar or re-solidifying as char.

The final product distribution—how much gas, liquid (bio-oil), and solid (char) you get—is a direct result of the competition between these pathways. If you can get the primary volatiles out of the hot zone quickly, you can minimize secondary reactions and maximize liquid yield. If they linger, they'll break down into gases or form more char.

To control the outcome of pyrolysis, we need to understand the speed, or kinetics, of these reactions. The most important tool for this is the Arrhenius equation, which connects the reaction rate constant ($k$) to temperature.

The key term here is the activation energy ($E_a$). It represents the energy barrier that must be overcome for a reaction to proceed. Each component of biomass has a different chemical structure and therefore a different activation energy for its decomposition.

As the table shows, hemicellulose has the lowest activation energy, so it's the first to decompose. Crystalline cellulose requires more energy, and lignin, with its complex aromatic structure, is the toughest to break down and reacts over a very wide temperature range. Understanding these different energy barriers is crucial for designing a pyrolysis process that targets specific products.

The heating rate is one of the most powerful levers we can pull to influence the final products. It directly controls the temperature of the biomass particles and, by extension, which reactions are favored.

Slow Pyrolysis (<10 K/min): At slow heating rates, the material heats up gradually. There is plenty of time for secondary reactions to occur. As primary volatiles are formed, they linger in the hot particle and reactor, breaking down further into thermally stable char and non-condensable gases. This process is ideal for maximizing char production, as used in making biochar for agriculture.

Fast Pyrolysis (>1000 K/min): Here, the goal is to transfer a massive amount of heat to the biomass particles almost instantly. The temperature inside the particle shoots up, rapidly exceeding the activation energy for primary decomposition. The resulting volatiles are generated so quickly that they are rapidly expelled from the particle before secondary reactions can take hold. This rapid quenching and removal is key to maximizing the yield of liquid bio-oil, often reaching 60-75% by weight.

So how do we model this complex web of reactions? There are two main approaches.

Global Kinetic Models: These are the simplest approach. They lump the thousands of reactions into a few overarching steps. A common example is a three-reaction model: Biomass → Gas, Biomass → Oil, and Biomass → Char. Each of these three “reactions” gets its own set of Arrhenius parameters ($A$ and $E_a$). While easy to compute, these models are a major simplification and don't provide insight into the specific chemical changes occurring.

Network Models: These are much more detailed. They attempt to track the formation and destruction of specific chemical species or compound groups. A network model might include dozens or even hundreds of interconnected reactions, each with its own kinetic parameters. For example, it could model the breakdown of cellulose into an intermediate called levoglucosan and then the subsequent reactions of that intermediate into other products. These models are computationally intensive but provide a much deeper, more mechanistic understanding of the pyrolysis process.

The choice of model depends on the goal. For a quick estimate of overall product yields, a global model might suffice. For understanding how to produce a specific chemical from biomass, a detailed network model is necessary.

Understanding pyrolysis kinetics allows us to move from simply burning biomass to precisely engineering its thermal deconstruction into valuable fuels and chemicals.