Carbon capture technologies fall into two broad camps: those that trap CO2 at its source (power plants, factories, refineries) and those that pull it directly from the atmosphere. Within each camp, several competing approaches are at different stages of development, from commercially deployed systems to lab-scale prototypes. Here’s what’s actually being built, how each technology works, and where the biggest obstacles remain.
Point-Source Capture: Three Core Approaches
Most carbon capture today happens at the smokestack, where CO2 concentrations are high enough to make separation practical. Three main methods dominate.
Post-combustion capture strips CO2 from exhaust gases after fuel has already been burned. The most common version pumps flue gas through a chemical solvent that reacts with CO2 and binds to it. The solvent is then heated to release a concentrated stream of CO2 for storage. Recovery rates range from 80 to 98%, depending on whether the system uses solid adsorbents or liquid solvents. This is the most mature approach and the easiest to retrofit onto existing plants, but the energy cost of heating and regenerating the solvent is significant.
Pre-combustion capture converts fossil fuels into a mix of carbon monoxide and hydrogen (called synthesis gas) before burning. A secondary reaction shifts the carbon monoxide into CO2 and more hydrogen, and the CO2 is separated before the hydrogen is used as fuel. Removal efficiency reaches about 90%. The main limitation is that it requires a gasification or reforming step upfront, so it’s better suited to new facilities designed around the process than to retrofitting old ones.
Oxy-fuel combustion takes a different path entirely: instead of burning fuel in regular air (which is mostly nitrogen), it burns fuel in nearly pure oxygen. The result is an exhaust stream that’s roughly 95% CO2, making separation almost trivial. Recovery rates hit 90 to 98%. The catch is that producing high-purity oxygen in the first place demands a large air separation unit, which adds cost and energy.
Direct Air Capture
Direct air capture (DAC) pulls CO2 straight from the ambient atmosphere, where concentrations are roughly 0.04%. That’s orders of magnitude more dilute than flue gas, which makes the engineering challenge far harder and the energy requirements far higher. Two main approaches are being scaled.
Liquid solvent systems, pioneered by Carbon Engineering (now part of Occidental Petroleum), pass air through a solution of potassium hydroxide. The CO2 dissolves into the liquid and forms a carbonate. A calcium loop then strips the carbonate out, regenerates the potassium hydroxide for reuse, and liberates a pure stream of CO2. The process requires substantial thermal energy to complete the calcium loop.
Solid sorbent systems, developed most prominently by Climeworks, use filters coated with materials that chemically bind CO2 at low temperatures. Once saturated, the filters are heated to around 80 to 120°C, releasing the CO2 for collection. Climeworks’ Mammoth plant in Iceland, which opened in 2024, is the largest operational DAC facility, though its capacity of roughly 36,000 tons per year is tiny relative to global emissions.
Cost remains the central barrier. As of 2023, DAC runs between $600 and $1,000 per ton of CO2 removed. Economies of scale could push that toward $500, but to compete with other climate strategies, the price needs to fall below $100 per ton. At current rates, the math simply doesn’t work for large-scale deployment without heavy subsidies or premium carbon credit markets.
Bioenergy With Carbon Capture (BECCS)
BECCS pairs biomass energy with carbon capture to create a system that can, in theory, go net-negative. Plants absorb CO2 as they grow. When that biomass is burned for energy or converted into biofuels, the resulting CO2 is captured and stored underground instead of released. The net effect is removing carbon from the atmosphere while still generating usable energy.
Right now, only about 2 megatons of biogenic CO2 are captured per year, mostly from bioethanol production, and less than 1 megaton of that actually goes into permanent storage. The International Energy Agency’s project pipeline suggests capture from biogenic sources could reach around 60 megatons per year by 2030. That falls far short of the roughly 185 megatons per year the IEA says would be needed by 2030 in a net-zero scenario.
BECCS is already being integrated into several industrial sectors. Biomass co-firing is commercial in pulp and paper mills, cement plants, and steel blast furnaces. But most cement plants planning to combine biomass with carbon capture aim for carbon neutral at best, not carbon negative, because they’re only partially substituting biomass for fossil fuels or only capturing a fraction of emissions. Scaling BECCS also raises questions about land use, biodiversity, and competition between fuel crops and food production.
Ocean-Based Carbon Removal
The ocean already absorbs roughly a quarter of human CO2 emissions. Ocean alkalinity enhancement (OAE) aims to accelerate that natural process by increasing the water’s capacity to absorb and store carbon. One approach adds crushed alkaline minerals (like olivine or limestone) to seawater. Another generates sodium hydroxide electrochemically and introduces it to the ocean, where it drives the conversion of dissolved CO2 into stable bicarbonate that remains locked in seawater for thousands of years.
These methods are still largely in the modeling and early laboratory stage. The vast majority of results on OAE’s chemical and biological impacts come from simulations, and researchers have flagged an urgent need for empirical data from mesocosm studies and small field trials before scaling up. The potential is enormous given the ocean’s sheer volume, but so is the ecological uncertainty.
Advanced Materials: MOFs and Membranes
The workhorse solvent in conventional carbon capture, an amine-based liquid, degrades over time and requires a lot of energy to regenerate. Two categories of advanced materials aim to do the job more efficiently.
Metal-organic frameworks (MOFs) are crystalline materials with an extraordinarily porous structure, like a molecular sponge. They can be tuned at the atomic level to selectively grab CO2 while ignoring other gases. The key advantage over amine solvents is lower regeneration energy: less heat is needed to release the captured CO2 and reset the material for another cycle. Researchers are now screening thousands of experimentally synthesized MOFs to identify candidates that can work under real-world conditions, including the humidity present in actual flue gas.
Membrane separation takes a different approach. Instead of chemically binding CO2, thin polymer or inorganic films act as selective barriers, allowing CO2 to pass through while blocking nitrogen and other gases. Pure polymer membranes face a fundamental tradeoff: making them more permeable (faster throughput) tends to reduce their selectivity (how well they distinguish CO2 from other molecules). Inorganic membranes, such as those based on zeolites, have pushed past this tradeoff in lab settings. Mixed-matrix membranes, which embed inorganic particles in a polymer base, are an active area of development trying to combine the best properties of both.
What Happens to Captured CO2
Capturing carbon is only half the equation. The CO2 has to go somewhere permanent, or be turned into something useful.
Geological Storage
The most established option is injecting compressed CO2 deep underground into saline aquifers, depleted oil and gas reservoirs, or unmineable coal seams. Over time, multiple trapping mechanisms kick in: the CO2 gets physically trapped beneath cap rock, dissolves into brine, or reacts with surrounding minerals to form stable carbonates. Basalt formations are particularly promising because the reactive minerals in basalt can convert CO2 into solid rock within years rather than centuries, essentially locking it away permanently through mineral carbonation.
Carbon Utilization
Rather than simply storing CO2, carbon utilization converts it into commercial products. Mineral carbonation transforms CO2 into stable minerals like magnesite or calcite, which can be used in cement production. Captured CO2 can also serve as a feedstock for synthetic fuels, methanol, or other biochemicals. The appeal is obvious: turning a waste product into revenue. The limitation is that only permanent storage actually removes CO2 from the carbon cycle. Synthetic fuels, for instance, release their carbon right back into the atmosphere when burned. Mineralization into building materials, by contrast, locks carbon away for the life of the structure.
The Scale Problem
The technologies described here range from commercially deployed to barely past the concept stage. Point-source capture on industrial facilities is real and operating today, but covers only a fraction of global emissions. DAC works but costs 5 to 10 times more than it needs to for widespread adoption. BECCS could theoretically deliver net-negative emissions, but the current project pipeline would need to triple to meet climate targets by 2030. Ocean-based methods haven’t left the lab in any meaningful way.
The gap between what exists and what’s needed is not primarily a technology problem. Most of these systems work at small scale. The bottleneck is cost reduction, energy supply (many capture processes are energy-intensive, and that energy needs to be clean), and the infrastructure to transport and store billions of tons of CO2 per year. Every major climate scenario that limits warming to 1.5 or 2°C assumes carbon capture plays a significant role. Whether the technology scales fast enough to fill that role is the open question.