Novel Graphite Synthesis
In our last post, we discussed the importance of graphite and its associated supply, geopolitical, and pollution challenges. What if there was a way to leverage the same fossil fuels that caused our climate problem into a solution?
Because graphite is ultimately a crystalline form of carbon, several carbon-rich feedstocks can theoretically be converted into graphite under the right reaction conditions. For example, biomass can be pyrolyzed into biochar, which contains 40–80% carbon content depending on quality. This biochar can then be graphitized at a lower graphitization temperature than conventional methods using petroleum coke. Biomass is attractive because it could potentially yield carbon-negative graphite, while addressing waste biomass issues.
Coal, with a carbon content ranging from 65–92% depending on quality, has been shown to graphitize at similar temperatures as petroleum coke. Unlocking the ability to graphitize coal at scale would enable coal companies to adapt from burning fossil fuel to selling minerals, and remove the need to use a byproduct of oil refining.
Finally, methane pyrolysis (MP) leverages the United States’ dominance in natural gas production — producing more than any other nation and comprising almost a quarter of global supply — to generate hydrogen and carbon products.
None of these processes are widely-adopted on a commercial scale, but the most mature is MP. The Norwegian company Kværner (now Aker Solutions) deployed the first and only commercial-scale MP facility utilizing hot plasma technology in 1997, but the facility was decommissioned in 2003 due to insufficient quality of carbon black product. This process was later adopted by Monolith Materials, which launched its first demonstration facility in the U.S. in 2020.
Hydrogen or carbon?
Historically, MP companies like Monolith, Hazer, BASF, and Aurora Hydrogen have focused on producing hydrogen with carbon byproducts typically for use in tires and asphalt. In the hydrogen market, MP competes primarily against steam methane reformation (SMR), which produces around 95% of today’s hydrogen supply. SMR has benefits besides being a mature technology — it produces twice as much hydrogen per methane molecule due to the usage of H2O.
As a result, SMR is 75% efficient compared to MP’s 58% efficiency, but it also generates CO2 emissions. If carbon capture and sequestration are added to SMR its efficiency decreases to ~60%, which is comparable to MP. While that may seem low, MP is far less energy intensive than water electrolysis.
We’ve mostly spoken about the hydrogen product, but what about the carbon output? Carbon black is a relatively low-value material, and some startups like Molten Industries believe they can control the formation of solid carbon such that it creates more crystalline graphite instead.
While MP inherently produces both hydrogen and carbon, the process should be optimized to favor one product over the other. For hydrogen production, priorities include achieving high methane conversion and hydrogen selectivity. Conversely, processes intending for carbon to be the main source of revenue must design their systems to enhance carbon quality. One example of how these two products conflict with one another is that achieving the production densities required for hydrogen supply (SMR achieves 3000 h-1 GHSV at 20 bar) necessitates high rates of carbon deposition, which in turn typically results in defects and amorphous structures in the carbon material. Companies attempting to optimize both outputs often struggle to achieve either goal effectively.
Regardless of the feedstock, making these novel graphite synthesis technologies economically viable requires solving three main challenges:
- Decreasing the input heat required
- Increasing the amount of graphitic carbon produced
- Removing and purifying graphite from the reactor cost-effectively
Decreasing input heat
Decreasing input heat can be accomplished by using more efficient heating methods, catalysts, or autothermal processes. In traditional heating, heat is produced in one location (i.e. filament, heated wall, combustion tube), which is then transferred to the surface of the carbon feedstock via convection/radiation and then conducted inward. This technique is simple and at a high technology readiness level (TRL), but is inefficient.
As mentioned before, the only commercially-demonstrated process for MP uses plasma, which pyrolyzes methane at temperatures between 1,000°C (cold plasma) and 2,000°C (hot plasma). One of the main challenges for plasma is that the carbon output tends to be low-quality and amorphous rather than graphitic, partially because the plasma itself tends to be difficult to sustain for long durations.
In contrast, volumetric heating methods utilize the conductivity or resistance of a material to directly generate heat. Joule heating passes electric current through a feedstock, producing heat directly within the material. Induction heating uses alternating magnetic fields to induce eddy currents within a conductive material, which then generate heat via resistance. Microwave heating uses electromagnetic fields to rotate and vibrate molecules, thus generating heat. The common theme among all these techniques is that they generate heat within the volume of the material; without wasted heat, volumetric heating is more energy efficient. However, these techniques are still low-to-medium TRL and have several limitations:
- Induction/microwave heating have low penetrations depths, so still rely on conduction to transfer heat further into a material.
- Joule heating requires contact with an electrode, which can become coated with impurities or reaction byproducts.
Some companies propose autothermal processes, using combustion of some portion of the methane or hydrogen to supply heat for the endothermic methane decomposition reaction. The point of thermoneutrality, which is the amount of methane/hydrogen consumed to provide enough heat for pyrolysis, requires at least 9% of the methane feed or 15% of the hydrogen product.
As a consequence of this operation, solid carbon production may be lower. The key challenge for this technology is minimizing how much methane or hydrogen is consumed. With thermal losses and incomplete combustion, this percentage of methane or hydrogen consumption goes up, which in turn increases emissions and product losses.
Improving graphite yield
Catalysts such as nickel, iron, and amorphous carbon are another way to simultaneously reduce the heat required while increasing graphitic carbon yield. They can reduce temperature requirements from >1100°C to 600–800°C, but are easily susceptible to deactivation due to carbon deposition. Processes must reactivate the catalyst by cleaning away the carbon and/or replenish the catalyst, both of which will incur cost. As a result, most catalytic processes employ cheap “throw-away” materials like iron or carbon. The carbon produced from an iron catalyst can range in morphology, but it tends to be highly graphitic as amorphous carbon is dissolved in iron at high temperatures and re-precipitated out as graphitic sheets.
Cleaning + purifying outputs
Finally, removing the carbon deposits from the reactor and purifying them into high-quality graphite remains an unsolved challenge. Some of the most promising approaches include:
- BASF’s moving bed reactor (proposed by Exxon in 1958), where inductively-heated carbon catalyst granules flow counter to a rising methane gas stream that pyrolyzes directly onto the granules at 1400°C. This approach avoids the use of catalysts and avoids depositing carbon on the reactor walls.
- Molten media approaches (like Molten Industries, the Karlsruhe Institute of Technology, and TNO), which pass methane through a liquid metal bubble column reactor at >1000°C. The methane bubbles serve as mini-reactors where carbon deposits, rising to the surface where the floating solid carbon can be harvested.
- Hazer Group’s fluidized bed reactor, which uses an iron ore catalyst to lower reactor temperatures to between 750–850°C. Fluidized bed reactors enable good material mixing, as well as density-based separation of catalysts with/without carbon deposits.
Even after recovering the carbon material, the product must be separated from the catalyst or by carbon quality, which can be accomplished by methods like acid leaching and froth floatation.
What’s next?
At Roadrunner Venture Studios, we are interested in innovations that address these aforementioned challenges to scaling novel graphite synthesis:
- Decreasing the input heat required
- Increasing the amount of graphitic carbon produced
- Removing and purifying graphite from the reactor cost-effectively
If you are working on a technology that can tackle these challenges and are interested in how Roadrunner can help, feel free to reach out!
Note: There are a handful of other approaches as well, such as electrochemical reduction of metal carbonates and precipitation in molten media. Maybe we’ll write an analysis of those in a future post (more likely if there’s audience interest).
Note: A big thank you to Henry Moise at Stanford University for collaborating on this blog post.
