- Enhanced Rock Weathering (ERW) is a CDR method that accelerates CO2 sequestration but faces efficiency and scalability challenges due to energy demands, rock availability, and soil conditions.
- Biochar offers long-term carbon storage with diverse applications, though its effectiveness depends on production methods, decay rates, and energy requirements.
Enhanced rock weathering (ERW) or Enhanced Weathering (EW) is a carbon dioxide removal (CDR) method that accelerates natural mineral carbonation. The process sequesters CO2 into rocks over timescales ranging from months to decades, depending on factors such as mineral composition, particle size, soil temperature, and pH.
ERW application as a CDR method involves a range of supply chain processes, including rock excavation, materials transportation, rock grinding, and application on land (see Figure 1). For instance, from crushing and grinding the rocks alone, the CO2 removal efficiency of ERW can decrease by up to 24%. Additionally, the method has limited sequestration potential per unit of rock, e.g., around 0.3 tCO2/t for basalt and 0.9 tCO2/t for dunite [1], [2]. As a result, these supply chain emissions are amplified by the large quantities of rock required, which limits the CO2 removal potential from EW due to rock availability and extraction constraints.
The rock type and particle size strongly influence the CO2 removal efficiency of the EW method. Although considered permanent, CO2 removal from EW is not instantaneous and requires time. This rate can be accelerated by reducing the particle sizes and applying the particles on land with suitable soil pH and temperature. However, the process to reduce particle size requires more energy, which may increase costs and CO2 emissions associated with rock grinding, depending on the energy source. Similarly, land with suitable pH and temperature for ERW application is also limited.
Rock excavation for rock mining is represented in MESSAGE as ERW_MINING technology. The technology takes either lightoil or diesel as an input and erw_(rock) at the primary level as the output commodity.
| Parameter | Value | Unit | Note |
|---|---|---|---|
| var_cost | 6.2 | $/t | Uniformly applied in all regions |
| input | 4.2 | kWh/t | Either diesel or lightoil |
Rock transportation for ERW involves transport from the quarry to the grinding facility and from the grinding facility to the cropland for application. To simplify the representation in the model, these two transportation segments are merged as ERW_TRANS technology. Variation of the total distance for this is taken into account, using data from [3], which quantifies the relation between distance and fraction of cropland area that can be covered. In the model, this data is aggregated into five bins for simplification
| Parameter | Value | Unit | Note |
|---|---|---|---|
| var_cost (per km) | 0.03 | $/t.km | Uniformly applied in all regions, vary by bin |
| input | 0.3 | kWh/t.km | Either diesel or lightoil |
| input | 1 | t | erw_(rock) (e.g., rock = basalt), primary |
| output | 1 | t | erw_(rock), secondary |
Distance:
|
| km | Each bin covers 20% of the total land area in each modelled region |
The initial particle size diameter is assumed to be 5,000 µm, with the target diameter after grinding varying between 10 μm and 100 μm. The energy required for crushing and grinding to 500 µm is assumed 25 kWh/t [2]. Secondary grinding energy requirements to reduce the particle sizes further are estimated using empirical data from [3].
| Parameter | Value | Unit | Note |
|---|---|---|---|
| inv_cost | 6.5 | $/tpa | - |
| capacity_factor | 0.95 | - | - |
| technical_lifetime | 10 | yr | - |
| fix_cost | 7.1 | $/tpa.yr | - |
| input | 25 | kWh/t | Primary grinding from 5,000 µm to 500 µm, electricity at the secondary level |
|
| kWh/t | Secondary grinding from 500 µm to the size targets, electricity at the secondary level |
| input | 1 | t | erw_(rock), secondary |
| output | 1 | t | erw_(rock)_(size), final |
It is important to note that the output particle sizes from these grinding processes are not uniform but follow a distribution. These values are expressed as P80 diameter, which refers to the screen size through which 80% of the sample's mass passes. Here, we use the Rosin-Rammler model to estimate the size distribution:
This distribution is important to be reflected in the model as it can significantly affect the rate at which CO2 is removed from the atmosphere using the ERW method, which is illustrated in Figure 4.
We assume a rock application rate of 20 t/ha-yr, based on the range of values suggested in [XYZ]. The associated energy requirement is 1.75 kWh/t, with either diesel fuel or light oil used as the energy source [4]. Additional costs, including machinery and labour, are estimated at 6.5 $/t [X].
As mentioned earlier, the rate at which CO2 is sequestered into rocks depends on different factors such as mineral composition, particle size, soil temperature, and pH. To calculate this, the weathering rate of each mineral in the rock needs to be accounted for, using Eq. 3.
The reaction rates between mineral and water, H+ and OH– ions can be calculated using Eqs. 3-6 as outlined below.
Rock characteristics parameters for those calculations are outlined in Tables 2-6 below:
Table 2. CO2 removal potential parameters [2], [5]Basalt composition is also provided by Lewis et al. [6].
Table 5. Dunite mineral composition [2], [5]In MESSAGE, we consider the ERW application on cropland. Using the soil pH and temperature data from [XYZ], a spatial analysis was performed to estimate the land-plot-specific CO2 removal rate, as illustrated in Figure Y. These estimates are used to aggregate land suitability for ERW application into five bins for simplification, represented by ERW_APPLY(N) in the model.
In MESSAGE, we assume that CO2 captured by ERW will be stored permanently. Review this article to learn more [7]
Biochar is pyrogenic carbon produced by the incomplete combustion of plant biomass, which can be used as a method to store biogenic carbon for a long period due to significant reductions in decomposition and chemical transformation rates [8].
There are several methods for producing biochar. In MESSAGE, we use fast and slow pyrolysis as archetypes. These two methods differ primarily in the composition of their outputs, which include biochar, biogas, and bio-oil. In MESSAGE, biogas and bio-oil are converted either to high-temperature heat or electricity. The CO₂ emissions from the combustion of biogas and bio-oil can either be captured, providing additional CO₂ removal, or released back to the atmosphere. Different combinations of these configurations result in eight technology archetypes implemented in MESSAGE.
We assume that biomass is transported for 100 km from the production facility to the application sites. Here, a 0.03 $/t.km of transport cost is considered, along with 0.3 kWh/t.km of light oil consumption. These parameters are incorporated in the model as biochar_trans technology parameters.
We considered two types of biochar application: permanent removal and other applications, which are represented as two different modes in the biochar_apply technology. Permanent removal refers to the long-term storage of carbon, such as incorporating biochar into concrete aggregates. Literature suggests that using biochar as an additive at up to 5% of cement mass consistently improves concrete properties. However, at higher proportions, the effects vary; some applications demonstrate further improvement, while others experience a decline in performance. To remain conservative, we limit biochar application in concrete to 2.5% of cement demand.
In addition to concrete, non-land-use applications of biochar include the remediation of contaminated soil, water filtration, wastewater treatment, and absorption in landfills or abandoned mines. These uses are subject to biochar decay, with an initial decay rate of 5.5% over the first two years, followed by a constant rate of 0.3% of biochar input per year thereafter.
References
S. Chiquier, P. Patrizio, M. Bui, N. Sunny, and N. Mac Dowell, (2022)
Energy Environ. Sci., vol. 15, no. 10, pp. 4389–4403.
S. M. Chiquier
J. Strefler, T. Amann, N. Bauer, E. Kriegler, and J. Hartmann (2018)
Environ. Res. Lett., vol. 13, no. 3, p. 034010.
P. Renforth (2012)
Int. J. Greenh. Gas Control, vol. 10, pp. 229–243.
D. J. Beerling et al. (2020)
Nature, vol. 583, no. 7815, pp. 242–248.
A. L. Lewis et al. (2021)
Appl. Geochem., vol. 132, p. 105023.
Y. Kanzaki, N. J. Planavsky, and C. T. Reinhard (2023)
PNAS Nexus, vol. 2, no. 4, p. pgad059.
Y. Kuzyakov, I. Bogomolova, and B. Glaser (2014)
Soil Biol. Biochem., vol. 70, pp. 229–236.
