Nitrous oxide (N2O) is the third largest contributor to climate change after carbon dioxide and methane. Unlike most other super pollutants, it is also long-lived, persisting in the atmosphere for more than a century. While aggressively cutting nitrous oxide emissions is essential and reductions are possible in the energy and industrial sectors, agricultural sources remain far harder to mitigate and there is currently no viable pathway for reducing emissions to zero. This raises the question: could nitrous oxide be removed from the atmosphere after it has been emitted, through enhanced natural sinks or engineered processes?
Spark asked me to explore this question as part of their broader commitment to addressing unsolved climate challenges. Spark’s approach is to identify unmanaged climate risks — like large sources of unabated emissions — and accelerate the development of the fields needed to address them. Their work spans from active field building programs to deep analysis into high-potential fields and early-stage frontier research into possible opportunities where the science is still very nascent. Nitrous oxide removal fits squarely in the third category: new frontiers.
This guest essay presents a summary of my research into the nascent science of nitrous oxide removal, including an examination of how potential removal pathways may work. The goal is to help evaluate whether nitrous oxide removal is a solution worth pursuing through more focused efforts.
Nitrous oxide is a major contributor to climate change and the third largest source of greenhouse gas emissions after carbon dioxide and methane. These emissions have increased atmospheric nitrous oxide concentrations by nearly 25% since the pre-industrial era, reaching 336 parts per billion (ppb) in 2022. Despite this seemingly small concentration, nitrous oxide is responsible for approximately 0.1°C of warming on account of its high potency as a greenhouse gas — each ton has the global warming potential (GWP) of 273 tons of carbon dioxide. This warming effect persists beyond the next century, similar to that of carbon dioxide, due to nitrous oxide’s 109-year lifetime in the atmosphere (thus the GWP equivalency is the same whether measured over a 20-year or 100-year time horizon).
This high potency can also be seen in estimates of the economic damages from emissions: The EPA’s 2023 Report on the Social Cost of Greenhouse Gases estimates nitrous oxide’s social cost at $54,000 per ton, compared to $190 for carbon dioxide and $1,600 for methane. Beyond this climatic role, nitrous oxide is also an ozone-depleting substance, like chlorofluorocarbons (CFCs), which are responsible for causing the Antarctic ozone hole. Ozone loss is now recovering with the regulation of these gases by the Montreal Protocol. Nitrous oxide emissions are now the dominant ozone-depleting emissions, with an impact equivalent to 15% of peak CFC emissions in the 1980s. This effect is estimated to have an additional social cost of $1,700 per ton in human health and agricultural impacts. This impact may be even greater in a future scenario with steep carbon dioxide and methane reductions: because these gases have an indirect role in increasing ozone, unabated nitrous oxide emissions in such a scenario may reverse the ozone hole’s recovery.
Nearly two-thirds of direct anthropogenic nitrous oxide emissions stem from the application of nitrogen-based fertilizers and manure in agriculture. A portion of this nitrogen is converted to nitrous oxide through microbial processes both on the field and in downstream and downwind ecosystems that receive nitrogen-laden runoff emanating from agricultural fields. Nitrous oxide emissions are often linked to other nitrogen losses, such as ammonia, nitrogen oxides, and nitrate leaching, which can degrade local air, soil, and water quality. Increasing global population and per capita consumption are projected to increase agricultural nitrous oxide emissions by 25% by 2050. While existing efforts focus on reducing the need for nitrogen fertilizer and optimizing its management through existing best practices and new innovations (See Spark’s exploratory work on Nitrogen 2.0 and the 2024 Global Nitrous Oxide Assessment), there is no line of sight to reducing these emissions to zero.
Other anthropogenic sources do have significant mitigation potential and, in some cases, existing cost-effective solutions. 30% of nitrous oxide emissions come from the combustion of fossil fuels and biomass, caused by high-temperature conversion of nitrogen compounds in the fuel. These emissions are expected to decline if a transition to cleaner energy sources is successful. The remaining industrial emissions originate from the production of adipic and nitric acid, two important chemical feedstocks. A range of cost-effective treatments for these processes is already available and, in theory, these emissions can be almost entirely mitigated, although only account for 5% of anthropogenic nitrous oxide emissions.
Ammonia, the precursor to nitrogen fertilizers, has also been proposed as a carbon-free fuel for maritime shipping.Yet, incomplete combustion can form nitrous oxide, analogous to carbon monoxide’s formation in fossil fuel combustion. While this production can likely be mitigated by appropriate engineering and operating conditions, it presents a significant risk to ammonia’s role as a sustainable fuel: a conversion of just 0.4% to nitrous oxide would have the same climate impact as the avoided fossil fuel emissions. A four-fold increase in ammonia production would be needed to power the entire maritime shipping sector, with the potential to significantly perturb global nitrogen cycles if even a small fraction were released into the environment.
The ability of the natural nitrogen cycle to accommodate these new and increasing anthropogenic sources is limited, as evidenced by the increasing atmospheric concentration. Nitrous oxide’s long atmospheric lifetime is indicative of its stability in the atmosphere, with decomposition via UV light and oxygen radicals in the stratosphere as the only major sinks.
A very small number of researchers are starting to evaluate potential nitrous oxide removal pathways to understand whether nitrous oxide can be removed from the atmosphere or broken down to nitrogen and oxygen. These pathways involve either enhancing natural sinks or deploying novel engineered processes and may take the form of relatively self-contained industrial activities (closed system approaches) or outdoor interventions in the atmosphere or on land (open system approaches).
Such removal would augment crucial emissions reduction efforts, which are unfortunately both being implemented too slowly and ultimately limited in their extent: The widespread deployment of current mitigation technologies is only sufficient to flatten, but not decrease, annual emissions, counteracting increased demand for fertilizer and ammonia. Yet, pathways that limit warming to 2°C require a 20% reduction in anthropogenic nitrous oxide emissions by 2050 (from 2019 levels). Atmospheric removal may be able to bridge this gap and is the only pathway for addressing historical emissions.
Beyond this, nitrous oxide’s long atmospheric lifetime means its climate impact may be considered fungible with carbon dioxide over the next century (approximately, and not considering the significant and distinct non-climatic effects of both gases). If net nitrous oxide emissions are not sufficiently reduced, even deeper cuts in carbon dioxide will be required to balance the radiative budget and meet climate goals. Conversely, removing nitrous oxide could help compensate for shortfalls in carbon dioxide mitigation.
At current concentrations, nitrous oxide accounts for 15% of the warming potential of all greenhouse gases in the atmosphere, equivalent to the warming effect of 90 ppm carbon dioxide. Thus, the technical challenge of removal at its very low atmospheric concentration (0.336 ppm) is partially ameliorated by its high potential impact.
The potential scale of nitrous oxide removal, and the time it takes to develop and scale up, is not well constrained for any approach. Scaling closed system approaches will likely face similar challenges as direct air capture of carbon dioxide (DAC): a relatively high cost, a dependence on the availability of low-emission energy, and the material, labor, and engineering constraints of large industrial projects. It may be possible to integrate nitrous oxide removal with DAC, increasing the total climate mitigation potential per air volume by 22%. While open-system approaches may avoid some of these challenges, their scale is potentially limited by land use, fundamental constraints imposed by the earth system, and social license to operate. The timeline to validate such approaches may also be longer than that for closed systems due to the longer lifecycle of field experiments and the need to develop measurement, reporting, and verification (MRV) systems.
As one benchmark, a total scale of 2 Mt nitrous oxide per year by 2050 is needed to close the gap between the deployment of existing mitigation strategies and a 2°C scenario, increasing to 3.6 Mt per year by 2100 (equivalent to 0.55 and 0.98 Gt carbon dioxide per year, respectively). A portfolio of approaches will likely be necessary to achieve the total desired scale. At this early stage of development, having a diverse research portfolio can also help to reduce the risks that any specific approach ends up proving nonviable. As the future feasibility and scale of nitrous oxide removal is unknown, it’s also crucial that we don’t depend on potential future removal solutions at the expense of greater ambition today with current climate solutions.
Nitrous oxide removal could be accomplished by a variety of chemical and biological mechanisms, analogous to the wide range of carbon dioxide and methane removal approaches under development. These mechanisms take advantage of nitrous oxide’s unique chemical properties to perform selective chemical reactions, either separating nitrous oxide from air or converting it into less harmful products.
Nitrous oxide’s 109-year atmospheric lifetime is a result of its unstable state in the atmosphere, as energy is released in an exothermic reaction when it decomposes to nitrogen and oxygen. However, this reaction does not proceed spontaneously at normal atmospheric conditions, requiring some form of activation energy to kick-start the process. As a result, nitrous oxide has a much longer lifetime than more reactive gases like methane. The natural nitrous oxide sinks are over 99% atmospheric processes, with natural mechanisms in soils and wetlands accounting for less than 0.1% of the total natural sink.
Photolysis is a decomposition process in which a nitrous oxide molecule is broken apart by the energy delivered by a photon of light. About 90% of the natural sink of nitrous oxide is due to photolysis in the stratosphere by ultraviolet(UV) light from the sun. For this to occur, the photon must have sufficient energy to break the N-O bond (shorter wavelengths of light, like UV, have higher energy per photon) and have a wavelength that is absorbed by nitrous oxide.
Nitrous oxide is colorless, meaning it does not absorb light in the visible wavelengths, but it does absorb a small fraction of light in the UV range, below 240 nm (and in the infrared, which causes its radiative forcing effect). Experiments using 140-230 nm light show that every time a nitrous oxide molecule absorbs a photon, it breaks apart into N2 and a lone O atom (which later combines with a second O to form O2). In other words, each absorbed photon causes one molecule to decompose, which means the efficiency of this process — the quantum yield — is one. Yet, sunlight in these wavelengths is blocked as it passes through the atmosphere by ozone (O3), which strongly absorbs UV light and naturally accumulates in the stratosphere. Thus, nitrous oxide emitted at the surface must first mix vertically in the atmosphere and be transported to the stratosphere in order to be subject to photolysis. Once there, O2 and O3 still absorb the majority of incoming photons due to their higher concentrations, thus the apparent quantum yield (AQY), incorporating the fraction of photons which reach a nitrous oxide molecule, is much lower.
Increasing the natural photolysis sink does not appear promising as the AQY, which limits the reaction rate, is essentially controlled by the atmospheric composition. Increasing the AQY would require decreasing the amount of O2 and O3 in the stratosphere to allow more UV light from the sun to hit nitrous oxide. An artificial UV light source operating at the surface, where the ozone concentration is lower, may achieve a slightly higher AQY, yet will also generate O3, a harmful air pollutant when at ground level. The amount of energy required for an artificial light source is likely to be significant given the low AQY.
The remaining 10% of natural nitrous oxide sink is due to reaction with energetic radicals, also in the stratosphere. When ozone absorbs UV light it can also undergo photolysis, releasing a lone oxygen atom in an energetically excited state called O(1D). Most commonly, this O(1D) collides with N2 or O2 and is deactivated, ultimately combining with a second radical and returning to O2 or O3. Upon collision with nitrous oxide, O(1D) has sufficient energy to initiate a reaction, forming either N2 and O2 or 2 NO. (These NO can later react with O3, the mechanism by which nitrous oxide depletes the ozone layer). While the quantum yield of O(1D) production from O3 and the O(1D) reaction with nitrous oxide are each approximately one, the majority of O(1D) is deactivated by N2 and O2 (due to their higher concentrations increasing the likelihood of a collision), thus the AQY of the entire pathway is very low.
Beyond O(1D), nitrous oxide is relatively unreactive with other natural atmospheric radicals. The reactions with OH and Cl radicals, the primary atmospheric sink for methane and other organic compounds, are six orders of magnitude slower than that with O(1D).
While the stratospheric O(1D) pathway is well understood, to our knowledge there have been no efforts to enhance this pathway or replicate it in an engineered system. Generating O(1D) through artificial ozone photolysis is straightforward, yet the majority of these radicals will simply deactivate in collisions with N2 and O2 — resulting in a low AQY and significant energy requirement.
Photocatalysts are substances which are able to catalyze a chemical reaction by absorbing a photon of ultraviolet or visible light. That absorption excites one of the photocatalyst’s electrons which can react with nearby molecules, including nitrous oxide.
Photocatalysts are typically semiconductors with an energy difference between valence and conduction bands called a bandgap. Semiconductors normally behave as insulators, with all electrons stuck in the low energy valence bands. A photon can excite an electron into the conduction band, leaving behind a hole, if it has energy greater than the bandgap. The electron and hole can each react with other molecules adsorbed to the surface of the photocatalyst. The efficiency of this process is also described by an apparent quantum yield (AQY), the number of desired reactions which occur for every photon delivered to the system. This metric is influenced by both intrinsic material properties and extrinsic conditions:
The AQY is important for understanding how much energy is needed to run a system with artificial light, or the area of photocatalyst and sunlight needed for a sunlight-powered passive system. A recent analysis by Randall et al. indicates an AQY of >1% and >0.1%, respectively, are needed for cost-effective removal in these systems.
Various nitrous oxide removal approaches have been described in literature using TiO2-based photocatalysts. Two research groups testing photocatalytic air treatment to reduce other pollutants in agricultural barns have shown incidental removal of nitrous oxide at near-ambient concentration using TiO2-based commercial coatings and artificial light. Yet, the percent reductions are small (5-12%) and a follow-up study failed to replicate these results under similar conditions. A company, Crop Intellect Ltd., has also demonstrated atmospheric removal by dispersing a proprietary TiO2-based catalyst on agricultural fields and observing a ~1 ppb drawdown on ambient nitrous oxide concentration (extrapolated to ~2 tons CO2e per hectare per season). While promising, these studies do not report sufficient detail for efficacy or feasibility analysis, such as AQY or long-duration performance, and reveal significant knowledge gaps in transferring laboratory studies of photocatalysis into the real world.
The activation energy for decomposition can be supplied by increasing the temperature. While there is no minimum temperature threshold, the decomposition rate is negligible below approximately 700°C, and increases exponentially above this. Research on thermal decomposition has historically focused on the use of pure nitrous oxide as a spacecraft propellant (where the reaction itself generates significant heat) or emissions destruction from industrial processes when the concentration is high, offsetting the significant heating energy. Reducing this energy requirement is key for making thermal destruction viable at lower concentrations.
One approach is to incorporate a heat recovery system into the thermal process, removing waste heat from the treated exhaust and using it to preheat the incoming air. Such regenerative thermal reactors, designed and used for a range of emissions control applications, can recycle 90-95% of the exhaust’s heat, reducing the amount of external heat needed by a factor of 10-20, respectively.
The second approach is to reduce the required temperature by using a thermal catalyst. These materials have active sites that adsorb nitrous oxide and enable it to decompose at a lower temperature. Such behavior has been demonstrated for a wide range of materials, from noble metal nanoparticles to transition metal oxides and metal-doped zeolites.
As with thermal destruction, thermal catalysts can operate over a range of temperatures, with higher reaction rate at higher temperatures. The T50, the temperature at which 50% of nitrous oxide is decomposed, is commonly used to compare performance. Although, like a photocatalyst’s AQY, this metric depends on extrinsic factors like the flow rate and concentration. Many materials have demonstrated T50 < 300°C, and the highest performance, cobalt spinel (Co3O4) doped with other metals, shows T50 < 200°C, resulting in a 3-4x reduction in heating energy compared to thermal destruction.
More experimental characterization of these catalysts is needed at atmospheric concentrations; typical experimental conditions are 1000 ppm to 5% nitrous oxide in an inert carrier gas, yet oxygen and water vapor are shown to inhibit decomposition, increasing T50 by ~100°C for some catalysts. Finally, more thorough characterization of reaction kinetics, inhibition mechanisms, and durability is needed to evaluate the feasibility of a thermal catalyst for atmospheric nitrous oxide removal.
All biological systems require nitrogen to synthesize a variety of critical substances. This nitrogen is ultimately sourced from, and returns to, dinitrogen (N2) in the atmosphere. The biological nitrogen cycle describes the series of steps by which N2 is fixed into more reactive nitrogen chemicals which are then used and metabolized. The final step in the denitrification portion of the cycle is to reduce nitrous oxide to N2, which is performed by the nitrous oxide reductase enzyme (N2OR), the only known enzyme to catalyze this reaction.
Nitrous oxide reducing organisms are typically found in ecosystems which are net sources of nitrous oxide and thus have locally elevated concentration, although ecosystems have been observed to act as net sinks in rare occasions. In either case, it may be possible to enhance the nitrous oxide uptake (although in net-source ecosystems such efforts may be less fruitful than those to reduce nitrous oxide production in the first place). Such natural microbial communities also play a key role in modulating other greenhouse gases (carbon dioxide, methane) and the overall ecosystem health, thus any effort to alter this natural balance should include a holistic assessment of impacts.
Nitrous oxide-reducing organisms can also be hosted in a bioreactor. Several systems have been studied to treat high concentration (100+ ppm) emissions and to remove nitrous oxide produced during denitrification of wastewater, alongside other efforts to reduce in-situ production directly. To our knowledge, bioreactors have only been studied for emissions mitigation with no efforts focusing on atmospheric removal. The challenge of operating at ambient concentration is especially acute for biological processes. While gaseous nitrous oxide can directly contact the active sites of thermal and photocatalysts, it must first dissolve in an aqueous phase to reach an enzyme’s active site. Specific investigation of this mass transfer limitation and the kinetics of nitrous oxide metabolism at low concentration is needed to evaluate the feasibility of bioreactors.
Nitrous oxide is the third largest source of greenhouse gas emissions after carbon dioxide and methane and the largest ozone-depleting emission. Current mitigation efforts focus primarily on reducing emissions from agriculture, energy, and industrial sources. Their widespread deployment is only sufficient to flatten, but not decrease, annual emissions, yet a 2°C pathway will require a 20% reduction by 2050, equivalent to 2 Mt nitrous oxide per year. Researchers are starting to evaluate whether nitrous oxide can be removed from the atmosphere, either by enhancing natural sinks or deploying novel engineered processes, and if such approaches can scale to help address this challenge.
Nitrous oxide removal could be accomplished by a variety of chemical and biological mechanisms, analogous to the wide range of carbon dioxide and methane removal approaches under development. The emerging body of research on these approaches indicates significant challenges to climate-beneficial and cost-effective deployment, yet also indicates open research questions which may yield solutions.
All of these mechanisms, except biological, have inherently high energy requirements in the form of UV light or heat. The cost and embodied emissions of that energy alone is likely to limit the feasibility of closed systems (where that energy must be supplied), requiring significant improvements in catalyst performance and other system elements. Open system approaches, including the dispersal of photocatalysts or enhancing ecosystem denitrification, avoid this energy requirement yet have poorly constrained cost and effectiveness estimates.
Given the early stage and anticipated technical challenges, it is imperative to research a diverse portfolio of removal approaches, lest any specific approach proves nonviable, in parallel with aggressive overall greenhouse gas emissions reductions. Looking ahead, field-building efforts focused on feasibility assessment, consideration of broader environmental impacts, social license to operate, and MRV are essential for proactively guiding future research and development. By improving our understanding of these mechanisms, this work can help determine what role, if any, nitrous oxide removal can play in augmenting existing mitigation efforts and can increase the likelihood of developing feasible solutions in time to significantly impact global climate goals.
Nitrous oxide (N2O) is the third largest contributor to climate change after carbon dioxide and methane. Unlike most other super pollutants, it is also long-lived, persisting in the atmosphere for more than a century. While aggressively cutting nitrous oxide emissions is essential and reductions are possible in the energy and industrial sectors, agricultural sources remain far harder to mitigate and there is currently no viable pathway for reducing emissions to zero. This raises the question: could nitrous oxide be removed from the atmosphere after it has been emitted, through enhanced natural sinks or engineered processes?
Spark asked me to explore this question as part of their broader commitment to addressing unsolved climate challenges. Spark’s approach is to identify unmanaged climate risks — like large sources of unabated emissions — and accelerate the development of the fields needed to address them. Their work spans from active field building programs to deep analysis into high-potential fields and early-stage frontier research into possible opportunities where the science is still very nascent. Nitrous oxide removal fits squarely in the third category: new frontiers.
This guest essay presents a summary of my research into the nascent science of nitrous oxide removal, including an examination of how potential removal pathways may work. The goal is to help evaluate whether nitrous oxide removal is a solution worth pursuing through more focused efforts.
Nitrous oxide is a major contributor to climate change and the third largest source of greenhouse gas emissions after carbon dioxide and methane. These emissions have increased atmospheric nitrous oxide concentrations by nearly 25% since the pre-industrial era, reaching 336 parts per billion (ppb) in 2022. Despite this seemingly small concentration, nitrous oxide is responsible for approximately 0.1°C of warming on account of its high potency as a greenhouse gas — each ton has the global warming potential (GWP) of 273 tons of carbon dioxide. This warming effect persists beyond the next century, similar to that of carbon dioxide, due to nitrous oxide’s 109-year lifetime in the atmosphere (thus the GWP equivalency is the same whether measured over a 20-year or 100-year time horizon).
This high potency can also be seen in estimates of the economic damages from emissions: The EPA’s 2023 Report on the Social Cost of Greenhouse Gases estimates nitrous oxide’s social cost at $54,000 per ton, compared to $190 for carbon dioxide and $1,600 for methane. Beyond this climatic role, nitrous oxide is also an ozone-depleting substance, like chlorofluorocarbons (CFCs), which are responsible for causing the Antarctic ozone hole. Ozone loss is now recovering with the regulation of these gases by the Montreal Protocol. Nitrous oxide emissions are now the dominant ozone-depleting emissions, with an impact equivalent to 15% of peak CFC emissions in the 1980s. This effect is estimated to have an additional social cost of $1,700 per ton in human health and agricultural impacts. This impact may be even greater in a future scenario with steep carbon dioxide and methane reductions: because these gases have an indirect role in increasing ozone, unabated nitrous oxide emissions in such a scenario may reverse the ozone hole’s recovery.
Nearly two-thirds of direct anthropogenic nitrous oxide emissions stem from the application of nitrogen-based fertilizers and manure in agriculture. A portion of this nitrogen is converted to nitrous oxide through microbial processes both on the field and in downstream and downwind ecosystems that receive nitrogen-laden runoff emanating from agricultural fields. Nitrous oxide emissions are often linked to other nitrogen losses, such as ammonia, nitrogen oxides, and nitrate leaching, which can degrade local air, soil, and water quality. Increasing global population and per capita consumption are projected to increase agricultural nitrous oxide emissions by 25% by 2050. While existing efforts focus on reducing the need for nitrogen fertilizer and optimizing its management through existing best practices and new innovations (See Spark’s exploratory work on Nitrogen 2.0 and the 2024 Global Nitrous Oxide Assessment), there is no line of sight to reducing these emissions to zero.
Other anthropogenic sources do have significant mitigation potential and, in some cases, existing cost-effective solutions. 30% of nitrous oxide emissions come from the combustion of fossil fuels and biomass, caused by high-temperature conversion of nitrogen compounds in the fuel. These emissions are expected to decline if a transition to cleaner energy sources is successful. The remaining industrial emissions originate from the production of adipic and nitric acid, two important chemical feedstocks. A range of cost-effective treatments for these processes is already available and, in theory, these emissions can be almost entirely mitigated, although only account for 5% of anthropogenic nitrous oxide emissions.
Ammonia, the precursor to nitrogen fertilizers, has also been proposed as a carbon-free fuel for maritime shipping.Yet, incomplete combustion can form nitrous oxide, analogous to carbon monoxide’s formation in fossil fuel combustion. While this production can likely be mitigated by appropriate engineering and operating conditions, it presents a significant risk to ammonia’s role as a sustainable fuel: a conversion of just 0.4% to nitrous oxide would have the same climate impact as the avoided fossil fuel emissions. A four-fold increase in ammonia production would be needed to power the entire maritime shipping sector, with the potential to significantly perturb global nitrogen cycles if even a small fraction were released into the environment.
The ability of the natural nitrogen cycle to accommodate these new and increasing anthropogenic sources is limited, as evidenced by the increasing atmospheric concentration. Nitrous oxide’s long atmospheric lifetime is indicative of its stability in the atmosphere, with decomposition via UV light and oxygen radicals in the stratosphere as the only major sinks.
A very small number of researchers are starting to evaluate potential nitrous oxide removal pathways to understand whether nitrous oxide can be removed from the atmosphere or broken down to nitrogen and oxygen. These pathways involve either enhancing natural sinks or deploying novel engineered processes and may take the form of relatively self-contained industrial activities (closed system approaches) or outdoor interventions in the atmosphere or on land (open system approaches).
Such removal would augment crucial emissions reduction efforts, which are unfortunately both being implemented too slowly and ultimately limited in their extent: The widespread deployment of current mitigation technologies is only sufficient to flatten, but not decrease, annual emissions, counteracting increased demand for fertilizer and ammonia. Yet, pathways that limit warming to 2°C require a 20% reduction in anthropogenic nitrous oxide emissions by 2050 (from 2019 levels). Atmospheric removal may be able to bridge this gap and is the only pathway for addressing historical emissions.
Beyond this, nitrous oxide’s long atmospheric lifetime means its climate impact may be considered fungible with carbon dioxide over the next century (approximately, and not considering the significant and distinct non-climatic effects of both gases). If net nitrous oxide emissions are not sufficiently reduced, even deeper cuts in carbon dioxide will be required to balance the radiative budget and meet climate goals. Conversely, removing nitrous oxide could help compensate for shortfalls in carbon dioxide mitigation.
At current concentrations, nitrous oxide accounts for 15% of the warming potential of all greenhouse gases in the atmosphere, equivalent to the warming effect of 90 ppm carbon dioxide. Thus, the technical challenge of removal at its very low atmospheric concentration (0.336 ppm) is partially ameliorated by its high potential impact.
The potential scale of nitrous oxide removal, and the time it takes to develop and scale up, is not well constrained for any approach. Scaling closed system approaches will likely face similar challenges as direct air capture of carbon dioxide (DAC): a relatively high cost, a dependence on the availability of low-emission energy, and the material, labor, and engineering constraints of large industrial projects. It may be possible to integrate nitrous oxide removal with DAC, increasing the total climate mitigation potential per air volume by 22%. While open-system approaches may avoid some of these challenges, their scale is potentially limited by land use, fundamental constraints imposed by the earth system, and social license to operate. The timeline to validate such approaches may also be longer than that for closed systems due to the longer lifecycle of field experiments and the need to develop measurement, reporting, and verification (MRV) systems.
As one benchmark, a total scale of 2 Mt nitrous oxide per year by 2050 is needed to close the gap between the deployment of existing mitigation strategies and a 2°C scenario, increasing to 3.6 Mt per year by 2100 (equivalent to 0.55 and 0.98 Gt carbon dioxide per year, respectively). A portfolio of approaches will likely be necessary to achieve the total desired scale. At this early stage of development, having a diverse research portfolio can also help to reduce the risks that any specific approach ends up proving nonviable. As the future feasibility and scale of nitrous oxide removal is unknown, it’s also crucial that we don’t depend on potential future removal solutions at the expense of greater ambition today with current climate solutions.
Nitrous oxide removal could be accomplished by a variety of chemical and biological mechanisms, analogous to the wide range of carbon dioxide and methane removal approaches under development. These mechanisms take advantage of nitrous oxide’s unique chemical properties to perform selective chemical reactions, either separating nitrous oxide from air or converting it into less harmful products.
Nitrous oxide’s 109-year atmospheric lifetime is a result of its unstable state in the atmosphere, as energy is released in an exothermic reaction when it decomposes to nitrogen and oxygen. However, this reaction does not proceed spontaneously at normal atmospheric conditions, requiring some form of activation energy to kick-start the process. As a result, nitrous oxide has a much longer lifetime than more reactive gases like methane. The natural nitrous oxide sinks are over 99% atmospheric processes, with natural mechanisms in soils and wetlands accounting for less than 0.1% of the total natural sink.
Photolysis is a decomposition process in which a nitrous oxide molecule is broken apart by the energy delivered by a photon of light. About 90% of the natural sink of nitrous oxide is due to photolysis in the stratosphere by ultraviolet(UV) light from the sun. For this to occur, the photon must have sufficient energy to break the N-O bond (shorter wavelengths of light, like UV, have higher energy per photon) and have a wavelength that is absorbed by nitrous oxide.
Nitrous oxide is colorless, meaning it does not absorb light in the visible wavelengths, but it does absorb a small fraction of light in the UV range, below 240 nm (and in the infrared, which causes its radiative forcing effect). Experiments using 140-230 nm light show that every time a nitrous oxide molecule absorbs a photon, it breaks apart into N2 and a lone O atom (which later combines with a second O to form O2). In other words, each absorbed photon causes one molecule to decompose, which means the efficiency of this process — the quantum yield — is one. Yet, sunlight in these wavelengths is blocked as it passes through the atmosphere by ozone (O3), which strongly absorbs UV light and naturally accumulates in the stratosphere. Thus, nitrous oxide emitted at the surface must first mix vertically in the atmosphere and be transported to the stratosphere in order to be subject to photolysis. Once there, O2 and O3 still absorb the majority of incoming photons due to their higher concentrations, thus the apparent quantum yield (AQY), incorporating the fraction of photons which reach a nitrous oxide molecule, is much lower.
Increasing the natural photolysis sink does not appear promising as the AQY, which limits the reaction rate, is essentially controlled by the atmospheric composition. Increasing the AQY would require decreasing the amount of O2 and O3 in the stratosphere to allow more UV light from the sun to hit nitrous oxide. An artificial UV light source operating at the surface, where the ozone concentration is lower, may achieve a slightly higher AQY, yet will also generate O3, a harmful air pollutant when at ground level. The amount of energy required for an artificial light source is likely to be significant given the low AQY.
The remaining 10% of natural nitrous oxide sink is due to reaction with energetic radicals, also in the stratosphere. When ozone absorbs UV light it can also undergo photolysis, releasing a lone oxygen atom in an energetically excited state called O(1D). Most commonly, this O(1D) collides with N2 or O2 and is deactivated, ultimately combining with a second radical and returning to O2 or O3. Upon collision with nitrous oxide, O(1D) has sufficient energy to initiate a reaction, forming either N2 and O2 or 2 NO. (These NO can later react with O3, the mechanism by which nitrous oxide depletes the ozone layer). While the quantum yield of O(1D) production from O3 and the O(1D) reaction with nitrous oxide are each approximately one, the majority of O(1D) is deactivated by N2 and O2 (due to their higher concentrations increasing the likelihood of a collision), thus the AQY of the entire pathway is very low.
Beyond O(1D), nitrous oxide is relatively unreactive with other natural atmospheric radicals. The reactions with OH and Cl radicals, the primary atmospheric sink for methane and other organic compounds, are six orders of magnitude slower than that with O(1D).
While the stratospheric O(1D) pathway is well understood, to our knowledge there have been no efforts to enhance this pathway or replicate it in an engineered system. Generating O(1D) through artificial ozone photolysis is straightforward, yet the majority of these radicals will simply deactivate in collisions with N2 and O2 — resulting in a low AQY and significant energy requirement.
Photocatalysts are substances which are able to catalyze a chemical reaction by absorbing a photon of ultraviolet or visible light. That absorption excites one of the photocatalyst’s electrons which can react with nearby molecules, including nitrous oxide.
Photocatalysts are typically semiconductors with an energy difference between valence and conduction bands called a bandgap. Semiconductors normally behave as insulators, with all electrons stuck in the low energy valence bands. A photon can excite an electron into the conduction band, leaving behind a hole, if it has energy greater than the bandgap. The electron and hole can each react with other molecules adsorbed to the surface of the photocatalyst. The efficiency of this process is also described by an apparent quantum yield (AQY), the number of desired reactions which occur for every photon delivered to the system. This metric is influenced by both intrinsic material properties and extrinsic conditions:
The AQY is important for understanding how much energy is needed to run a system with artificial light, or the area of photocatalyst and sunlight needed for a sunlight-powered passive system. A recent analysis by Randall et al. indicates an AQY of >1% and >0.1%, respectively, are needed for cost-effective removal in these systems.
Various nitrous oxide removal approaches have been described in literature using TiO2-based photocatalysts. Two research groups testing photocatalytic air treatment to reduce other pollutants in agricultural barns have shown incidental removal of nitrous oxide at near-ambient concentration using TiO2-based commercial coatings and artificial light. Yet, the percent reductions are small (5-12%) and a follow-up study failed to replicate these results under similar conditions. A company, Crop Intellect Ltd., has also demonstrated atmospheric removal by dispersing a proprietary TiO2-based catalyst on agricultural fields and observing a ~1 ppb drawdown on ambient nitrous oxide concentration (extrapolated to ~2 tons CO2e per hectare per season). While promising, these studies do not report sufficient detail for efficacy or feasibility analysis, such as AQY or long-duration performance, and reveal significant knowledge gaps in transferring laboratory studies of photocatalysis into the real world.
The activation energy for decomposition can be supplied by increasing the temperature. While there is no minimum temperature threshold, the decomposition rate is negligible below approximately 700°C, and increases exponentially above this. Research on thermal decomposition has historically focused on the use of pure nitrous oxide as a spacecraft propellant (where the reaction itself generates significant heat) or emissions destruction from industrial processes when the concentration is high, offsetting the significant heating energy. Reducing this energy requirement is key for making thermal destruction viable at lower concentrations.
One approach is to incorporate a heat recovery system into the thermal process, removing waste heat from the treated exhaust and using it to preheat the incoming air. Such regenerative thermal reactors, designed and used for a range of emissions control applications, can recycle 90-95% of the exhaust’s heat, reducing the amount of external heat needed by a factor of 10-20, respectively.
The second approach is to reduce the required temperature by using a thermal catalyst. These materials have active sites that adsorb nitrous oxide and enable it to decompose at a lower temperature. Such behavior has been demonstrated for a wide range of materials, from noble metal nanoparticles to transition metal oxides and metal-doped zeolites.
As with thermal destruction, thermal catalysts can operate over a range of temperatures, with higher reaction rate at higher temperatures. The T50, the temperature at which 50% of nitrous oxide is decomposed, is commonly used to compare performance. Although, like a photocatalyst’s AQY, this metric depends on extrinsic factors like the flow rate and concentration. Many materials have demonstrated T50 < 300°C, and the highest performance, cobalt spinel (Co3O4) doped with other metals, shows T50 < 200°C, resulting in a 3-4x reduction in heating energy compared to thermal destruction.
More experimental characterization of these catalysts is needed at atmospheric concentrations; typical experimental conditions are 1000 ppm to 5% nitrous oxide in an inert carrier gas, yet oxygen and water vapor are shown to inhibit decomposition, increasing T50 by ~100°C for some catalysts. Finally, more thorough characterization of reaction kinetics, inhibition mechanisms, and durability is needed to evaluate the feasibility of a thermal catalyst for atmospheric nitrous oxide removal.
All biological systems require nitrogen to synthesize a variety of critical substances. This nitrogen is ultimately sourced from, and returns to, dinitrogen (N2) in the atmosphere. The biological nitrogen cycle describes the series of steps by which N2 is fixed into more reactive nitrogen chemicals which are then used and metabolized. The final step in the denitrification portion of the cycle is to reduce nitrous oxide to N2, which is performed by the nitrous oxide reductase enzyme (N2OR), the only known enzyme to catalyze this reaction.
Nitrous oxide reducing organisms are typically found in ecosystems which are net sources of nitrous oxide and thus have locally elevated concentration, although ecosystems have been observed to act as net sinks in rare occasions. In either case, it may be possible to enhance the nitrous oxide uptake (although in net-source ecosystems such efforts may be less fruitful than those to reduce nitrous oxide production in the first place). Such natural microbial communities also play a key role in modulating other greenhouse gases (carbon dioxide, methane) and the overall ecosystem health, thus any effort to alter this natural balance should include a holistic assessment of impacts.
Nitrous oxide-reducing organisms can also be hosted in a bioreactor. Several systems have been studied to treat high concentration (100+ ppm) emissions and to remove nitrous oxide produced during denitrification of wastewater, alongside other efforts to reduce in-situ production directly. To our knowledge, bioreactors have only been studied for emissions mitigation with no efforts focusing on atmospheric removal. The challenge of operating at ambient concentration is especially acute for biological processes. While gaseous nitrous oxide can directly contact the active sites of thermal and photocatalysts, it must first dissolve in an aqueous phase to reach an enzyme’s active site. Specific investigation of this mass transfer limitation and the kinetics of nitrous oxide metabolism at low concentration is needed to evaluate the feasibility of bioreactors.
Nitrous oxide is the third largest source of greenhouse gas emissions after carbon dioxide and methane and the largest ozone-depleting emission. Current mitigation efforts focus primarily on reducing emissions from agriculture, energy, and industrial sources. Their widespread deployment is only sufficient to flatten, but not decrease, annual emissions, yet a 2°C pathway will require a 20% reduction by 2050, equivalent to 2 Mt nitrous oxide per year. Researchers are starting to evaluate whether nitrous oxide can be removed from the atmosphere, either by enhancing natural sinks or deploying novel engineered processes, and if such approaches can scale to help address this challenge.
Nitrous oxide removal could be accomplished by a variety of chemical and biological mechanisms, analogous to the wide range of carbon dioxide and methane removal approaches under development. The emerging body of research on these approaches indicates significant challenges to climate-beneficial and cost-effective deployment, yet also indicates open research questions which may yield solutions.
All of these mechanisms, except biological, have inherently high energy requirements in the form of UV light or heat. The cost and embodied emissions of that energy alone is likely to limit the feasibility of closed systems (where that energy must be supplied), requiring significant improvements in catalyst performance and other system elements. Open system approaches, including the dispersal of photocatalysts or enhancing ecosystem denitrification, avoid this energy requirement yet have poorly constrained cost and effectiveness estimates.
Given the early stage and anticipated technical challenges, it is imperative to research a diverse portfolio of removal approaches, lest any specific approach proves nonviable, in parallel with aggressive overall greenhouse gas emissions reductions. Looking ahead, field-building efforts focused on feasibility assessment, consideration of broader environmental impacts, social license to operate, and MRV are essential for proactively guiding future research and development. By improving our understanding of these mechanisms, this work can help determine what role, if any, nitrous oxide removal can play in augmenting existing mitigation efforts and can increase the likelihood of developing feasible solutions in time to significantly impact global climate goals.
Stay in touch
Sign up to our Spark newsletter and stay updated!
Menu
Stay in touch
Sign up to our Spark newsletter and stay updated!
made by
tonik.com
