Carbon capture and storage (CCS) or carbon capture, utilization, and storage (CCUS) is recognized internationally as an indispensable key technology for mitigating climate change and protecting the human living environment (Fig. 1)[1–3]. Both the International Energy Agency (IEA) [4] and the Carbon Sequestration Leadership Forum (CSLF) [5] have stated that, in order for the energy sector to achieve net zero emissions by 2050, the global scale of CCS in 2030 and 2050 must respectively be 10–15 times and 100 times greater than the current 40 Mt·a^–1 as of 2020. The Intergovernmental Panel on Climate Change (IPCC) [6] has projected that the cost of mitigation will rise by 138% in 2100 if CCS technologies are not adopted. CCS is listed as one of three mandatory emission reduction technologies in the four key emission reduction technology pathways [7]in the 1.5 °C special report released by the IPCC in 2018.

Fig. 1. The CCS industry chain and its driving factors. UAV: unmanned aerial vehicle.

The IEA [8] has summarized four strategic areas in which CCUS should be used to address emissions: existing infrastructure, lowcarbon hydrogen production, the most challenging emission in sectors such as heavy industry and aviation, and removing carbon from the air. Both CCS technology and renewable energy technology are key technologies for mitigating climate change; however, from the perspective of land use, CCS is an underground-space utilization technology, and coal-fired power plants plus CCS require a much smaller land area than solar and wind power plants. Coal power/natural gas power generation plus CCS is still the cornerstone of power production safety and stability on the premise that renewable energy cannot provide large-scale industrial power and presents intermittency and energy-storage problems that are difficult to solve. In addition, CCS technology is the only effective option for direct and rapid emission reductions from fossil-energy-based large-scale sources such as the steelmaking, cement, and chemical industries [8]. CCS will become a large industry on the same scale as the oil and gas industry (Fig. 2).

Fig. 2. The CCS industry chain and its driving factors. MMV: measurement, monitoring, and verification; BECCS: bioenergy with carbon capture and storage; DAC: direct air capture; EOR: enhanced oil recovery.

Scholars’ predictions for the role CCS will play in China’s carbon neutrality pathway differ[9–11], with different prediction models corresponding to differences in the magnitude of CCS’s contribution to China’s emission reduction and in the amount of government investment in the future. Some Chinese scholars regard CCS technology as a ‘‘minimum” technology rather than a necessary technology, and thus do not rank it high enough in importance to predict its contribution to China’s emission reduction. However, China is heavily dependent on coal, so we consider a higher emission reduction contribution value from CCS to be preferable. For example, according to the latest Asian Development Bank (ADB) forecast, the contribution of CCS to emission reduction in China must reach (0.3–1.2) × 10⁸ t·a^–1 in 2030, (8.5–25) × 10⁸ t·a^–1 in 2050, and (13–26) × 10⁸ t·a^–1 in 2060 in order for China’s carbon neutrality goal to be achieved [12]. Given the anticipated emission reduction needs of the Belt and Road countries in the future, widespread use of CCS will give Chinese enterprises more room to expand. Negative emission technologies, such as direct air capture (DAC) and bioenergy with carbon capture and storage (BECCS), are futuristic and require scaling up to determine their economic viability.

CO₂-enhanced oil recovery (EOR) has been carried out in the United States and Canada since the 1960s. The world’s first large-scale CO₂-EOR project, Scurry Area Canyon Reef Operating Committee (SACROC), has been implemented by Chevron in the oilfield in Scurry County, Texas since January 26, 1972 [13]. The CO₂ for this project comes from the natural CO₂ fields in Colorado and is pipelined to the oilfield for flooding. More than 175 million tonnes of natural CO₂ in total were injected in the SACROC project during 1972–2009 [14].

However, the concept of CO₂ capture, transport, and storage in the modern sense, as a means of reducing anthropogenic CO₂ emissions, was first proposed by Marchetti [15]. The Sleipner CCS project, which began in 1996, and the IEA Greenhouse Gas (IEAGHG) Research and Development Programme Weyburn–Midale CO₂ Monitoring and Storage Project (Weyburn Project for short), which began in 2000, were the first international demonstrations of the large-scale capture, utilization, and storage of anthropogenic CO₂ emissions.

The Sleipner CCS project, which is a scientific research and large-scale commercial demonstration project implemented by Equinor under the influence of the Norwegian carbon tax policy, captures the CO₂ separated in the natural gas purification process and injects it into deep saline aquifers for storage and emission reduction. The project has stored more than 20 million tonnes of CO₂ in total since 1996. The project has carried out advanced monitoring for 20 consecutive years and has achieved remarkable scientific research results [16].

The Weyburn Project is the world’s most complete multidisciplinary scientific research project on the geological storage of CO₂. It has been ongoing for 12 years with a scale of 1.8 Mt·a^–1 of CO₂ injection and has safely stored over 35 million tonnes of CO₂ since October 2000. The Weyburn Project is conducted at the largest geoscience test site in the world, which was established in the Weyburn field in Southern Saskatchewan, Canada, with the joint support of Natural Resources Canada, the US Department of Energy, the Saskatchewan government, and other government agencies and enterprises[17–19]. After completion of the planned scientific research, the project has been converted into a commercial project.

The success of the Weyburn Project has established the irreplaceable role of CCS technology in reducing greenhouse gas emissions (Fig. 3). Firstly, it is the world’s first project to carry out the large-scale capture and long-distance transportation of low-cost and high-concentration CO₂ from the use of coal (i.e., the Dakota Gasification Company) for EOR and storage. It has demonstrated that it is possible to use CCS technology to reduce the CO₂ generated by coal—which contributes the highest proportion of carbon emissions in fossil energy use—on a large-scale, rapid, and lowcost basis, so as to make clean use of coal. Secondly, the project benefits from relying on CO₂-EOR; in the absence of government subsidies, the project has been in good operation for more than 20 years and has established the most successful CCS commercialization model. Thirdly, after solving the problem of how to effectively and efficiently capture and store CO₂ emission sources from the high-concentration coal chemical industry, the organizers of the Weyburn Project targeted the capture and storage of lowconcentration CO₂ emitted from coal-fired power plants, and built the world’s first 1 Mt·a^–1 post-combustion CO₂ capture facility at Unit 3 of the SaskPower Boundary Dam Power Station (abbreviated to BD3), so as to transport the CO₂ to the Weyburn oilfield for EOR and storage. Moreover, since the excess CO₂ captured at the power plant can be stored in nearby saline aquifers when the demand of the oil company for CO₂ is low, the SaskPower cooperated with the Petroleum Technology Research Centre (PTRC) to build the Aquistore scientific research facility for the geological storage of CO₂ in deep saline aquifers.

Fig. 3. Scientific research facilities and solid scientific research laid the foundation for the success of the Weyburn Project. CSA: Canadian Standards Association; BD3: Unit 3 of the SaskPower Boundary Dam Power Station.

The Weyburn Project has also made the utilization of CO₂ flooding into the best business model for CCS development without government incentives; therefore, ‘‘U” has been added to the concept of CCS for ‘‘utilization” and is widely praised internationally. However, among the commitments and efforts countries are currently making to achieve carbon neutrality goals, some forms of CO₂ utilization, such as CO₂-EOR, have not been fully effective in reducing emissions on a large scale [20]. Thus, this paper does not discuss CO₂ utilization such as CO₂ flooding and so forth.

There are many classification methods for CCS projects, with that of the Global CCS Institute (GCCSI) [21] being the most wellknown. The GCCSI takes projects that capture, transport, and store 400 kt·a^–1 CO₂ from industrial emission sources, or projects that capture and store 800 kt·a^–1 CO₂ from coal-fired power plants, as the threshold for large-scale pilot testing, demonstration, or commercialization projects. According to the GCCSI 2020 report, there are currently 65 commercial projects in the world, 26 of which are in operation, while others are at different stages of development.

CCS projects can also be classified according to the source of CO₂—that is, CO₂ from natural gas separation (e.g., Sleipner), carbon capture in the coal chemical plant (e.g., Great Plains Coal Gasification Plant), hydrogen production in oil refining (e.g., Shell Quest [22] and Tomakomai), coal power plants (e.g., SaskPower Boundary Dam and Petra Nova), carbon capture in steelworks (e.g., Al Reyadah), biomass energy capture (e.g., Decatur), and DAC (e.g., Carbfix and Climeworks, Heidelberg Cement’s ongoing Norwegian Norcem’s Brevik cement plant, and the Canada Lehigh Cement Company). Projects that directly use CO₂ from the natural CO₂ fields for flooding and storage are not considered to be carbon emission reduction projects, and thus are not considered as CCS projects.

The storage used in CCS projects can be divided into types such as storage in saline aquifers, storage with CO₂-EOR, storage in abandoned oil and gas fields, and storage in basalt [23]. CO₂ storage projects in basalt include Carbfix [24] in Iceland and Tomakomai [25] in Japan. However, although the Tomakomai project successfully injected CO₂ in sandstone, attempts at CO₂ storage in basalt were not ideal. As a result, the Tomakomai reservoir has low injectivity and does not exhibit high-temperature conditions like those of the basalt of the Carbfix project; thus, no mineralized storage has been achieved. Given these results, the potential for future CO₂ storage in basalt does not appear to be high at present.

The development history of CCS projects, the scientific problems encountered, and the technological progress achieved can be better understood by classifying CCS projects according to scientific research and commercial demonstration or operation. As mentioned above, although the United States has carried out large-scale CO₂ flooding since 1972, CO₂-EOR technology focuses on EOR that part of injected CO₂ remains in the reservoir and part of CO₂ comes out with oil from producing wells, rather than on the geological storage of CO₂ and CO₂ emission reduction. When CO₂ flooding technology turns to CO₂ geological storage in either saline aquifers or oil reservoirs, the first questions to be answered are: Is the CO₂ safely stored? How can the location of the stored CO₂ be verified? How can the CO₂ storage capacity of different scales be verified? What technology and equipment are used to monitor the safety of CO₂ storage over the long term? These questions must be solved by CCS research projects and infrastructures; they are also the basis for the commercial promotion of CCS. The Weyburn Project was a scientific research project with the goal of achieving CO₂ monitoring and storage, while carrying out monitoring and testing as completely as possible[18,26,27]. The standard Canadian Standards Association (CSA) Z741 geological storage of carbon dioxide was based on research from this project, and was in turn the basis for the international standard International Standards Organization (ISO)/Technical Committees (TC) 265 carbon dioxide capture, transportation, and geological storage.

The Sleipner project involves offshore monitoring research on the large-scale geological storage of CO₂. Four-dimensional (4D) seismic monitoring and time-lapse gravity monitoring from this project revealed the principle behind CO₂ migration and accumulation underground during the process of CO₂ injection. Other internationally influential CCS science and technology infrastructures (Table 1[28,29]) include the Aquistore, Cooperative Research Center for Greenhouse Gas Technologies (CO2CRC) Otway, Tomakomai, Ketzin, and the world’s first biomass CCS research project, Decatur Project, among others. These science and technology infrastructures have a high level of construction technology and advanced monitoring technologies and equipment, and cover pre-injection site research on the geological storage of CO₂, monitoring and research in the injection stage, and continuous monitoring and research after injection (post-injection and post-closure). Early research infrastructures for the geological storage of CO₂ include a geological CO₂ storage test site in Nagaoka, Japan, and the K12-B Offshore CO₂ Injection Project in the Netherlands, among others.

Table 1 Typical global CCS science and technology infrastructures and their characteristics.

VSP: vertical seismic profiling; DAS: distributed acoustic sensing; DTS: distributed temperature sensing; GPS: Global Positioning System; HiPACT: high pressure acid-gas capture technology; JGC: Japan Gasoline Company; BASF: Badische Anilin-und-Soda-Fabrik; InSAR: interferometric synthetic aperture radar.

However, on the whole, these CCS science and technology infrastructures focus on geological storage and are designed for conducting research on the measurement, monitoring, and verification (MMV) of geological storage and testing the feasibility of monitoring technology for storage capacities from 10 kt·a^–1 to 1 Mt·a^–1 . Except for the Weyburn Project, whose research focus is storage with EOR, the research focus of all the other projects is CO₂ storage in saline aquifers. This is because saline aquifers have much larger storage space than oil reservoirs in terms of storage potential. Projects involving CO₂ storage in saline aquifers capture and store CO₂ in the vicinity of the emission sources, thereby avoiding the cost of building long-distance CO₂ transportation pipelines. Moreover, storage of CO₂ in saline aquifers does not need to be taken into consideration the reduction or termination of oil companies’ demand for CO₂ due to changes in international oil prices. Such projects generally have a lower net-carbon emission reduction cost than the storage of EOR projects. When storing CO₂ in saline aquifers, it is not necessary to capture a very high concentration of CO₂. Therefore, CO₂ storage in saline aquifers is the current direction of CCS development.

However, CO₂-EOR should not be discounted, as this process creates a revenue stream to pay for a portion of the currently expensive capture process. Technically speaking, CO₂ geological storage in saline aquifers is currently carried out by means of direct injection; that is, no fluid is pumped out of the saline aquifers to reduce the in situ pore pressure. This method differs from the CO₂ flooding method, which involves flooding oil and gas reservoirs with CO₂ to replenish the storage pressure. Injecting CO₂ into saline aquifers is more likely to result in a pore pressure that is higher than the in situ pore pressure, which in turn leads to the rupture of caprocks, opening of reservoir cracks, opening of faults, and other leakage risks. However, the ongoing international CO₂ storage project in saline aquifers (e.g., Sleipner, Shell Quest, Tomakomai, Aquistore, CO2CRC Otway) proves that such risks are controllable.

The difficulty of CO₂ capture lies in its large scale and the need to achieve low energy consumption or low cost when it is used for low-concentration emission sources (e.g., coal-fired and gas-fired power plants, steelworks, the cement and chemical industries, and waste incineration). According to a breakdown of the capital expenditure investment of 790.3 million CAD in the Shell Quest CCS project [30], the costs of the capture, transportation, and storage of CO₂ with a concentration of 17% in the 1 Mt·a^–1 heavy oil hydrogen-production process are about 55%, 16%, and 5%, respectively. The cost of CO₂ capture for low-concentration emission sources always accounts for the great majority of CCS investment, while the cost of geological storage is the lowest. In addition to the high cost of investment in carbon capture facilities, there is still room for the optimization and improvement of the compression process and steam energy consumption; moreover, the very high solvent degradation rate requires the replenishment of solvents in large quantities during the operation of capture facilities, increasing the operating cost and the cost of capture.

The CO₂ capture process also presents various additional challenges, including limited land space, a limited water resource utilization rate, tolerance to other flue gas components (i.e., the need to adapt to the emission sources of different flue gas components including iron and steel, cement, smelting, etc.) and other local power plant standards (e.g., the main operation and maintenance intervals and the reduction of output loads). The scale of capture is also an important indicator of technological advancement, as large-scale post-combustion capture has many benefits. Many of the abovementioned performance issues are being solved. For example, companies such as China Huaneng Group are exploring how to adjust the solvent makeup for each flue gas in order to reduce degradation issues.

The scientific community has carried out a great deal of research in the field of low-concentration CO₂ capture technologies[31–34], but there are few technologies that can be commercialized on a large scale. As the world’s first post-combustion CO₂ capture facility for a megaton-scale coal-fired power plant, the completion and operation of the SaskPower BD3 capture facility in Canada is a milestone. The capture facility has been in stable operation since it was first put into operation on October 2, 2014. As of the end of May 2021, it has captured a total of 4.143519 million tonnes of CO₂. Its annual actual capture capacity is greater than 75% of the designed scale. The capture process uses Shell’s Cansolv technology, which is an integrated SO₂–CO₂ thermal capture process that uses a proprietary amine solvent. These solvents are regenerated using low-pressure steam from the power plant. At the SaskPower BD3 capture facility, the high energy requirements for the regeneration of amine-based solvents, solvent degradation, and so forth require optimization and improvement [35]. Among the second-generation capture technologies supported by the US Department of Energy, Linde/Badische Anilinund-Soda-Fabrik (BASF)’s advanced aqueous amine solvent process [36] achieves a capture energy consumption of less than 2.7 GJ·t^–1 of CO₂ (at an unknown operation scale), and the smallest amine-based solvent degrades during 5500 h of operation. Baker Hughes’ compact carbon capture technology is more suitable for most emission sources for which the space for capture land is not reserved, and has a good application prospect.

The coal-powered Petra Nova plant contains the largest coalfired power-plant CO₂ capture facility in the world, with an annual CO₂ capture capacity of 1.4 Mt·a^–1 . CO₂ capture is carried out using Kansai Mitsubishi’s CO₂ recovery process. The process uses a proprietary KS-1 solvent (a sterically hindered amine), which is regenerated using the steam from an independent gas-fired power plant. The CO₂ generated from the Petra Nova post-combustion plant is used for EOR, so low oil prices will affect the operation of the capture facility until regulations are in place to reduce industrial carbon footprints.

Projects with landmark CO₂ capture facilities include the capture project of the United Arab Emirates (UAE) Al Reyadah steel plant, which has a capture capacity of 800 kt·a^–1 CO₂ and uses the gas-based direct reduced iron (DRI) process; the UK Drax biomass power-generation capture facility, which is under construction, with a projected capture capacity of 4 Mt·a^–1 ; the 400 kt·a^–1 Norcem Brevik CO₂ capture project by Norway Heidelberg Cement; the 400 kt·a^–1 capture facility for the Fortum Oslo Varme waste incineration process; the 6 Mt·a^–1 CO₂ capture facility of the Mustang Station Natural Gas Power Plant; the full-scale 800 kt·a^–1 capture facility that is about to be built for the Canadian Heidelberg Caspian Lehigh Cement Plant; and the 750 kt·a^–1 capture facility for chemical processes such as hydrogen production by Shell (Canada) Polaris. The 3 Mt·a^–1 CO₂ capture facility of the Kemper County integrated coal gasification combined cycle (IGCC) was unsuccessful; however, the ambition shown by this project and the accumulation of experience from it are still useful as an exploration of large-scale pre-combustion CO₂ capture technology.

The characteristics of industrialized technological progress: The experience and lessons learned in the early operation of technologies are applied to improve the design and operation of the nextgeneration equipment, thereby improving efficiency and reducing capital and operating costs. It was the success of the SaskPower BD3 capture facility that led to the subsequent establishment of the second-generation post-combustion capture technology for the Shand Power Station, which is based on the operation practice of the first-generation CO₂ capture facility but reduces the project cost by 67%[8,37].

At present, China does not possess full-process CCS science and technology infrastructures, and current CCS research is carried out through small-scale full-process CCS pilot test projects and CO₂- EOR demonstrations[38–41]. The main goal of these projects is to study CO₂-EOR in order to improve oil recovery. Field-level research on the safety of CO₂ geological storage in China is insufficient. Although many scholars have carried out related monitoring studies[42,43], there is not enough evidence to confirm CO₂ storage capacity, CO₂ distribution, and storage safety; moreover, Chinese scholars are not significantly involved in researching the geological storage of CO₂. China is inherently deficient in its CCSrelated scientific research level, technical level, and investment funds, and a very big gap exists between China’s CCS research infrastructures and those of advanced countries.

China’s early CO₂-EOR projects in China National Petroleum Corporation (CNPC) and Sinopec used CO₂ from the natural CO₂ fields for CO₂ flooding and storage. However, as mentioned earlier, such projects cannot be recognized as CCS and emission reduction projects. The subsequent Shenhua Ordos Saline Aquifer CCS Demonstration, Jingbian CCS, Sinopec Zhongyuan Oilfield CO₂- EOR, and CNPC Changqing Oilfield CO₂-EOR are demonstrations based on low-cost and high-concentration CO₂ captured from coal chemical plant[38–41].

Current CO₂ capture experiments and demonstrations based on low-concentration CO₂ sources include the CR-Power (Haifeng) carbon capture test facility built by China Resources Holdings in Guangdong in May 2019, the 100 kt·a^–1 CO₂ capture unit of the Shanghai Shidongkou Second Power Plant built by the China Huaneng Group in 2009, the 60 kt·a^–1 pre-combustion capture facility of the Tianjin IGCC Power Plant completed by the China Huaneng Group in 2017, the 50 kt·a^–1 post-combustion capture facility of the Anhui Baimashan Cement Factory, and the newly built 150 kt·a^–1 CO₂ capture facility of the Guohua Jinjie Power Plant of the National Energy Group in Shaanxi. However, there is no downstream large-scale storage or utilization project for these CO₂ capture facilities, so they cannot complete long-term capture and full-load operation tests. There is a large gap between China and the United States and Canada in terms of the testing of capture technologies, including the testing of different types of adsorbents, operational efficiency, operational stability, energy consumption, and—especially—the scale of capture.

CCS projects are based on actual geological storage capacity, the safety of geological storage, and the monitoring, reporting and verification (MRV) of CO₂ emission reductions [44]. This is the basis for enterprises to enter the carbon trading market and obtain government incentives, such as the 45Q tax credit in the United States and carbon tax in Norway and Canada. Early in 2008, the National Academy of Engineering regarded carbon storage as one of the 14 major unsolved engineering challenges in the 21st century. In 2019, the MIT Technology Review listed carbon storage as the first of the top 10 technical challenges in the world. The most critical scientific and technical issue is how to study and develop a series of technical methods for the MMV of field-level geological storage of CO₂, and then to assess the underground storage risk and storage capacity, and ensure the safety of long-term underground storage. The US Department of Energy began funding CCS research in 1997, and its initial investment focused on geological storage[45,46]. It is only once the geological storage problem is solved that the destination of captured CO₂ can be ascertained, which in turn drives the development of CCS projects.

Concerns about the safety and risks of the geological storage of CO₂ have always been one of the primary CCS-related topics in the scientific community and among the public[47,48]. The risks of the geological storage of CO₂ mainly come from three aspects: ① Direct leakage channels may be present, such as injection or production wells connecting geological storage bodies, underground water layers, and the atmosphere[49–51]; ② the CO₂ injected underground might induce earthquakes, open faults, or breaks through caprocks, resulting in leakage; and ③ external forces such as natural earthquakes might damage geological CO₂ storage bodies (including borehole walls) so as to cause leakage. During the operation of injection or production wells, the wellbore may narrow, expand, rupture, or collapse, and corrosion of old well casings and cement sheaths may occur, creating leakage paths with the highest risk in CO₂ geological storage. Wellbore integrity ensures the safety of geological storage. Therefore, the US National Environmental Protection Agency (EPA) has established the Class VI Rule for CO₂ geological storage wells, which aims to ensure the safety of wellbores and the long-term safety of geological storage during CO₂ injection.

Two CCS scientific research facilities in Nagouka and Tomakomai, Japan, focus on investigating whether natural earthquakes will cause damage and leakage risks to geological CO₂ storage facilities. Geological CO₂ storage bodies in these two areas of Japan have experienced multiple earthquakes, including earthquakes with magnitudes of 6.8 and 6.7 and epicenters just 20 and 30 km away, which did not cause any leakage in the projects[52–54]. Scholars from the Research Institute of Innovative Technology for the Earth (RITE) checked the borehole wall integrity of the Nagouka Project before and after the earthquakes, assessed the seismic bottom hole pressure and injection facilities, conducted air tightness/pressure tests [55], and demonstrated the safety of the wellbore by means of acoustic cement bond log (CBL) and hole imaging.

However, CO₂ that has been injected underground might induce earthquakes, resulting in danger due to the opening of faults or the breakthrough of caprocks. This issue presents difficulties in safety monitoring and early warning, as the relevant pressure-change law and CO₂ flow direction are unknown. A paper published by Zoback and Gorelick [56] in 2012 triggered a debate among scientists by pointing out that CO₂ injection and geological storage may induce earthquakes, and that even a small earthquake will cause CO₂ to escape to the surface. The main problem with this perspective lies in the fact that Zoback’s experimental research was based on granite; however, it is extremely unlikely for granite to be used as the sealing layer or caprock in actual storage. Secondly, it is unreasonable to deem that the permeability of both joints and fractures in caprocks will increase in all cases so as to form a path allowing the migration of CO₂ to the surface. The degree of increase in the permeability of joints and fractures depends on multiple factors such as rock type, stress state, and fillers. In fact, in many cases, large faults exhibiting a large degree of slippage play a sealing role and have no impact on permeability. This is the case with reservoirs in California (USA), Iran, and even China’s Bohai Bay Basin; despite frequent large earthquakes, oil and gas reservoirs can still be preserved very well without leakage. In these regions, the faults themselves play a sealing role rather than being fluid-migration channels; in fact, the reservoirs of oil and gas were originally formed in previous geological periods because the faults blocked the migration of oil and gas. The fault discussed by Zoback is a huge fault extending from a deep injection layer to the surface. The possibility of such faults in a basin is very low, and they can be avoided in the selection of geological storage sites. MIT scholars [57] and Carnegie Mellon University scholars [58] have all deemed the possibility of generating huge faults by earthquakes induced by CO₂ injection and storage to be extremely low.

Geoscience Australia [59] is taking the lead in studying the safety of fault sealing in a field-scale CO₂ injection project being carried out at the CO2CRC Otway scientific test site. So far, no leakage of CO₂ along the fault has been found. Of course, similar to the induction of earthquakes by oilfield water flood development and geothermal exploitation [60], large-scale fluid injection may induce earthquakes [61]. In order to prevent the induction of earthquakes during CO₂ injection, pressure control is the most important regulatory measure [62], as the activation of faults and caprock breakthroughs are mainly caused by pressure changes.

An assessment of the safety and risks of geological storage of CO₂ requires an investigation on whether the injected CO₂ will break through a complete structure or lithologic trap, such as wellbores, multiple sets of faults, or caprock combinations. The following considerations need to be ascertained and understood: the CO₂ storage capacity of geological traps; the migration and accumulation principles of CO₂ injected into reservoirs in strata; the possibility of leakage or seepage caused by the opening of faults or caprocks by the injection pressure or external forces; possible paths of CO₂ leakage; whether CO₂ will escape to the shallow surface to pollute underground drinking water and cause damage to the surface ecological environment; if CO₂ leaks through breaking caprock or opening the fault firstly but does not escape to the surface, whether multiple sets of caprocks in the overlying strata of reservoirs will continue to seal and isolate the CO₂ for a second time; and how to ensure wellbore integrity and ascertain whether reservoir engineering, technology, and so forth will cause wellbore damage and CO₂ leakage.

Research on the safety and risks of the geological storage of CO₂ has already been carried out internationally in a series of largescale geological storage and scientific research projects that focus on researching field-level safety and risk monitoring and verification. Some scholars consider that a small part of the injected CO₂ might have broken through the caprocks and faults and entered the overlying strata in the Weyburn Project (personal communication). Because it was impossible to conduct field-level monitoring in the overlying saline aquifer, the possible breakthrough position was not known. It could not be detected using geochemical methods, and no abnormal noise was found through passive seismic monitoring. During the injection process—and especially after the storage was completed—the use of geophysical (4D seismic) methods were the only feasible means to monitor and verify the caprocks and overlying strata[26,63].

China has also carried out a number of small-scale CO₂-EOR and storage projects and CO₂ storage demonstration projects in saline aquifers in Ordos[38,39,64]; however, the focus of these projects is CO₂-EOR [65]. While there are many numerical simulations for the research and verification of geological storage safety, there are few studies on the field-level and large-scale risk monitoring and verification of caprocks and faults; thus, the distribution scope, safety, and storage capacity of CO₂ underground cannot be definitively known. Surface 4D seismic monitoring of CO₂ injections has only been carried out in the Gao-89 block of the Sinopec Shengli Oilfield [66]. Although the findings of domestic CO₂-EOR projects have not been publicly reported, the leakage of CO₂ along borehole walls into the overlying strata is currently the predominant risk. No CO₂ leakage caused by fault opening has been observed in any of the CO₂-EOR projects at home and abroad.

The petroleum industry has very mature and successful experience in studying and predicting the integrity of caprocks. Of course, concerns about the storage process still remain: For example, when CO₂ is in contact with the caprock, will the geochemical reaction of CO₂ with the rocks affect the pores and permeability [67]? What is the mineralization situation of CO₂ in reservoirs? Will CO₂ diffuse to caprocks and then corrode leakage channels? What are the requirements for caprock thickness in order to seal off CO₂? Nevertheless, the preservation and sealing of existing natural CO₂ in gas fields have demonstrated that these concerns are unnecessary[68–70].

At present, more than 90% of geological CO₂ storage projects are being carried out in oil reservoirs; however, deep saline aquifers are still the largest geological CO₂ storage spaces and those with the greatest potential. White et al.[71,72] used InSAR data to observe the surface deformation near a CO₂ injection well with saline aquifers that was carried out by BP and others in In Salah, Algeria; the researchers then supported the hypothesis that the reservoir was fractured and the bottom caprock was broken through after nearly four million tonnes of CO₂ had been injected into the 950 m deep saline aquifer, although no CO₂ leakage was observed on the surface. No surface deformation has been observed in other large-scale CO₂ storage projects around the world, such as the Weyburn Project in Canada.

Verdon et al. [29] compared the stress deformation caused by the three megaton-scale geological storage projects Sleipner, In Salah, and Weyburn. The comparison showed that the impact of injection pressure on the safety of geological storage is the greatest for low-permeability saline aquifers. Pressure control is a key scientific problem to be solved for the safety and future leakage risks of geological storage of CO₂ in either reservoirs or saline aquifers [62].

Of course, assuming that CO₂ breaks through a set of caprock or opens a fault and then enters the previous set of strata, as long as the CO₂ is blocked by multiple sets of overlying caprocks layer by layer, it is not considered to be leakage. Research conducted by Rinaldi et al. [73] showed that induced seismic activities related to fault resurrection will not necessarily open a new flow path for leakage. A single induced event in a layer is usually insufficient to substantially change the permeability over the entire length of the fault. In that case, even if changes occur in the permeability of a certain section of the fault, it does not mean that CO₂ will migrate upward along the entire fault, break through multiple sets of caprocks, and then enter the overlying drinking water layer [74].

Many scholars have carried out core experiments and numerical simulation studies on caprock sealing properties and fault opening under the conditions of the chemical and physical actions of CO₂– saltwater–strata[75,76], deepening our understanding of the possibility of the evolution of caprocks and faults during the long-term geological storage of CO₂. However, the physical and chemical changes in these reservoirs, as well as changes in the chemical reactions of CO₂–water–rocks in caprocks and faults [67], are relatively small compared with changes in formation pressure and CO₂ saturation [77], and thus are not easily observed and monitored. The safety of geological CO₂ storage has a mutually corroborating relationship with geological CO₂ storage capacity. The monitoring and verification of geological CO₂ storage capacity can identify both emission reductions and leakage risks that may be predicted when the injection volume is not equal to the storage capacity. The most critical parameters are formation pressure and CO₂ saturation, because an increase in formation pressure after CO₂ injection may induce breakthrough through caprocks and faults, as well as borehole wall breakage. In addition, CO₂ saturation is a key parameter in determining the underground distribution scope of CO₂ and calculating the storage capacity. Of course, monitoring of CO₂ distribution has to demonstrate that CO₂ may or may not break through multiple sets of caprocks or open multiple sections of faults and then enter the shallow surface or the atmosphere before the occurrence of real leakage is ascertained. Monitoring of the surface environment and monitoring of underground phreatic aquifers are also important links in determining whether CO₂ will leak to the surface or to the shallow surface.

Borehole observation is the most direct method to monitor the safety of the geological storage of CO₂. However, in most cases, the risk of breakthrough through caprocks and faults is not around the injection wells, and the scope of well monitoring is very limited. Relying on surface geophysical monitoring, and especially 4D seismic techniques, to monitor changes in underground geological storage bodies has become the preferred technical means for geological CO₂ storage projects around the world[16,27,78]. In all CCS science and technology infrastructures (Table 1), 4D seismic imaging has become the most effective evidence to prove the safety, distribution scope, and distribution status of the geological storage of CO₂. The Ketzin project accurately predicted CO₂ saturation and CO₂ storage capacity [79], which may be related to the relatively shallow CO₂ storage. The Aquistore project drew many lessons from the Weyburn Project, deployed permanent geophones, and used Vibroseis to avoid the problem of non-repetitive shot point locations in two vintages of 3D seismic monitoring with dynamite sources. Borehole distributed acoustic sensing (DAS) technology was used to obtain pressure, temperature, and VSP monitoring data at different depths at the same time without affecting CO₂ injection in the well [80]. Meanwhile, 4D surface seismic monitoring before CO₂ injection was carried out to study the impact of changes in shallow-surface elastic parameters caused by seasonal changes on 4D seismic differential information [81]. It was demonstrated that CO₂ distribution could still be imaged by means of seismic monitoring under the conditions of a 10 kt injection scale and CO₂ injection in the deepest (3400 m) saline aquifer in the world [82]. The distribution and saturation of CO₂ at a depth of 1500 m and with an injection volume of 5000 t were successfully imaged using CO2CRC Otway advanced 4D seismic monitoring with a buried DAS array, 4D VSP with optical fibers on the tubing, and continuous seismic sources [83]. To image CO₂ distribution, the Sleipner [16] and Tomakomai [25] projects used marine 4D seismic monitoring technology, which is the most effective monitoring means for the storage of CO₂ in deep saline aquifers. Sleipner [84] carried out time-lapse gravity monitoring and was the first to demonstrate the effectiveness of low-cost gravity monitoring technology for the monitoring of large-scale geological storage of CO₂. The combination of time-lapse gravity monitoring with seismic monitoring improves the accuracy of seismic prediction of CO₂ storage capacity.

However, when using conventional P-wave information, it is difficult to identify whether the difference between two vintages of seismic monitoring comes from changes in pressure or from CO₂ saturation[78,85]. Therefore, the Weyburn Project used advanced 4D and three component (4D3C) seismic monitoring technology and combined P-waves with converted waves to distinguish pressure changes from CO₂ saturation changes. Although many difficulties still remain in the processing and interpretation of 4D converted-wave data at present, and the expected results have not been achieved, this represents the forefront of seismic monitoring technology.

An issue with CO₂ saturation—a key parameter in predicting CO₂ storage capacity—is that it is difficult to accurately determine the CO₂ saturation in a well; in particular, after CO₂ injection, it is more difficult to accurately measure CO₂ saturation in a cased well than that in the uncased well. The Nagaoka project was the first to carry out geophysical logging more than 40 times in a well after CO₂ injection and has thus become a model for identifying CO₂ changes before and after injection in reservoirs and in different stages by logging multiple times[55,86]. At present, the monitoring results of underground CO₂ distribution mainly provide the CO₂ distribution in a set of strata. When monitoring the CO₂ injected into two sets of saline aquifers—such as the volcanic and volcaniclastic Miocene Takinoue formation at 2400 to 3000 m which is the second CO₂ injection layer in the Tomakomai project—no monitoring images for CO₂ distribution have been obtained. Further research on monitoring and imaging methods to obtain the CO₂ distribution status in multiple sets of reservoirs is necessary in order to determine whether CO₂ enters other reservoirs from the injection layer, and then whether geological storage is safe. The MMV of geological CO₂ storage requires the combination and mutual corroboration of multiple disciplines such as geology, geophysics, geochemistry, petroleum engineering, and more in order to demonstrate the safety of long-term geological CO₂ storage.

Based on the experience of advanced countries in CCS development, CCS in other countries will undergo a gradual transition from low-cost and high-CO₂-concentration capture from coal chemical plant, low-cost natural gas purification and capture combined with flooding and storage, or direct storage in saline aquifers, to lowconcentration tail gas capture in coal-fired power plants, refining and hydrogen production, steelworks, cement, and so forth. The difficulty in capture technology still lies in the large-scale capture of low-concentration CO₂ tail gas with low energy consumption, while the difficulty of geological storage lies in how to carry out field-level MMV to determine geological CO₂ storage capacity, emission reductions, and storage safety.

The science and technology CCS infrastructures that have been built in advanced countries play a crucial role in the understanding of basic scientific problems related to CCS, reduction of the cost of whole-process CCS technology, monitoring of the safety of longterm geological storage, R&D of advanced technologies, and further commercialization, promotion and demonstration, and talent training. Sharing the CCS research results and knowledge of advanced countries makes it possible to accelerate the reduction of the cost and risks of CCS and the commercial layout of CCS in other countries; however, the complexity of geological conditions makes it difficult to directly replicate many technologies and achieve the desired results using them. For example, CO₂-EOR in China has not achieved the same high recovery and stable flooding results as in North America. Insufficient studies and understanding of scientific issues such as the storage status, migration and accumulation principles, and safety of CO₂ injected underground have restricted the large-scale development of CO₂-EOR and storage. Without the research and technical support of scientific CCS research facilities, even if the construction of large-scale CCS projects was carried out, it would be difficult to run the projects consecutively, and thus the goal of large-scale rapid emission reduction would not be achieved.

CCS hubs, in which various carbon emitters are matched with appropriate local carbon storage areas and carbon transport is established excellent, can achieve the cost reduction and largescale reduction of emissions from different types of emission sources within a region; thus, such hubs are the development direction of future commercial CCS projects. However, a CCS hub has higher requirements for the geological storage capacity of storage sites, and the detailed and accurate selection of geological storage sites is particularly important. The CO₂ storage potential and injection capacity of a storage site determine the scale of the capture and transmission pipelines and the construction scale of the CCS hub.

The key direction for the future development of geological CO₂ storage is the storage of CO₂ in saline aquifers, followed by the storage of CO₂ in oil reservoirs and abandoned gas fields. The storage of CO₂ in saline aquifers allows more types of high-carbon emission sources to be located nearby, which reduces the construction cost of long-distance CO₂ transportation pipelines and the carbon footprint in the production and laying of steel pipes. Storage in saline aquifers does not require the emission of CO₂ to be captured into high-purity CO₂ in order to reduce the cost of capture. During the deployment of CCS, it is necessary to improve the energy utilization efficiency and reduce the carbon footprint in each link of the entire process, calculate from the perspective of the entire process, and achieve net emission reduction at the lowest cost.

According to the construction period of successful large-scale CCS projects in advanced countries, it takes 5–10 years from site selection to the completion and operation of a CCS project. To achieve the 2030 CCS emission reductions estimated by the IEA and CSLF, action must start now; otherwise, the long-term temperature goal of the Paris Agreement cannot be achieved. Of course, the most important factors in the successful development of CCS in advanced countries such as the United States, Norway, and Canada are the country’s incentive policies (e.g., 45Q), carbon taxes, and restraint policies on enterprises’ carbon emissions. The European Union and China’s carbon markets have a positive role to play in promoting CCS technology. The increase in carbon prices will help CCS to gradually move toward profitability and commercialization.

This work is sponsored by Shaanxi Natural Science Foundation (2021JCW-04) of Department of Science and Technology of Shaanxi for Northwest University. The authors would like to thank the editors and anonymous reviewers for their helpful and constructive comments.

Jinfeng Ma, Lin Li, Haofan Wang, Yi Du, Junjie Ma, Xiaoli Zhang, and Zhenliang Wang declare that they have no conflict of interest or financial conflicts to disclose.