Combined Analytic Hierarchy Process and Weighted Interval Method Models for the Geological Evaluation of CO₂ Storage in Coal Goaf

Abstract

The increasing concentration of CO₂ in the atmosphere is a major factor contributing to climate change. CO₂ storage in coal goaf is a convenient, effective, and economical solution. Methods to quickly and effectively evaluate geological conditions are urgently required. The main influencing factors are geological safety, storage potential, economics, and environmental protection; these include 4 aspects, 38 indexes, and 4 index levels that can be quantified using classification levels. We established a geological evaluation model, using analytic hierarchy process (AHP) and weighted interval methods. AHP was used to determine its elements, indicators, and inter-layer relationships, as well as to clarify its structural relationships. The weight interval method is used to evaluate unstable elements, reducing their difficulty, and constant values are used to assign weights of stable elements to increase accuracy. This model was applied to assess the suitability of the goaf in Yaojie mine for geological CO₂ storage. The results revealed that this goaf is an above average CO₂ storage space, which was consistent with previous research. This geological CO₂ storage evaluation model may also be used to assess the CO₂ storage suitability of other coal goafs.

1. Introduction

Carbon dioxide (CO₂) is one of the most abundant greenhouse gases [1,2]. Significant emissions of CO₂ trap heat in the Earth’s atmosphere, causing global temperatures to increase in a phenomenon known as the greenhouse effect. Indirect effects include polar ice-cap melting and climate alterations, which can lead to a rise in sea levels and the inundation of low-altitude land [3,4]. Several techniques have been developed to mitigate the release of CO₂ and mitigated the impact of climate change. Carbon capture and storage (CCS) is a significant solution that uses large-scale technology to mitigate climate change by removing CO₂ emissions from fossil-fueled power plants [5,6,7]. As it is currently impossible to completely abandon fossil energy sources, CCS is the primary technical method and supporting guarantee to achieve the temperature control goal of the Paris Agreement [8].

CO₂ storage, which presents one of the most difficult challenges for CCS technology, can be divided into three categories: mineral storage, oceanic storage, and geological storage. The mineral storage of CO₂ is a chemical reaction between CO₂ and metal oxides to form solid carbonates and other by-products [9,10,11]. This storage method has the advantages of a lengthy storage time and low regulatory costs, but the high cost of the energy and reactants required for the storage process presents issues. The ocean storage of CO₂ consists of injecting CO₂ to form hydrates that are disposed into the ocean at great depths, or injecting CO₂ at great depths into the ocean to liquefy or form hydrates under suitable conditions [12]. Submarine earthquakes or water pressure releases can cause CO₂ leakages. This results in ocean acidification [13], which affects marine life and causes disturbances to ecosystems. The geological storage of CO₂ uses a mechanism that mimics the storage of fossil fuels in nature. CO₂ is injected through pipelines into strata with specific geological conditions and at specific depths. Geological environments such as old oil and gas fields, difficult-to-exploit coal seams, and deep groundwater layers are often used for CO₂ storage [14]. Geological storage is considered to be a more effective method for the long-term storage of CO₂ because of site feasibility and economic aspects, among other issues. As of November 2020, 28 CO₂ storage reservoirs were active in the global within geological structures, and 37 others were in development.

CO₂ is a predominant greenhouse gas in the Earth’s atmosphere. A major source of anthropogenic CO₂ is the combustion of fossil fuels. Coal is an important fossil energy source and industrial raw material. It is the largest and most abundant fossil energy material in China [15]. The total amount of CO₂ emitted by coal combustion has more than doubled in the past 20 years because of the continuous increase in total energy consumption (Figure 1). The characteristics of China’s energy structure have been defined as “lacking oil, little gas, rich coal”. This identifies coal resources as the main energy source and this structure does not change [16]. Under the goal of carbon neutrality, coal enterprises must seek breakthroughs and participate in the carbon emission market. Finding CO₂ geological reservoirs and developing simple and inexpensive storage technologies are particularly important for the in situ storage of CO₂. Large areas of underground goaf containing fractured zones are formed, providing significant potential space for CO₂ underground storage (Figure 2). The geological storage of CO₂ in coal goaf with good cap rocks is an interesting option.

Fossil-fuel power plants in China are the main consumers of coal and, consequently, the largest industrial fixed-point source of CO₂ emissions. This extensive use of coal provides convenient CO₂ storage potential in goaf areas to achieve the sustainable development concept of “where does coal originate and where do the solid waste and CO₂ generated from coal use return to?”. Comparatively, the cap rock of goaf is intact and closer to the CO₂ production area.

The storage of CO₂ in goaf has significant advantages compared with other geological storage methods. Goaf (Figure 2) can represent a vast area; the collapse area formed by the collapse of the overlying strata in goaf has a much larger porosity and geometric size of voids than other geological reservoirs. This results in a much lower CO₂ injection pressure than other storage methods. The tunnels and chambers not only provide a significant area for CO₂ storage but can also be used as overpressure storage tanks in combination with appropriate sealing measures [18,19]. Mines also accumulate detailed geological data during long-term mining processes that other geological storage methods cannot provide.

Currently, only feasibility studies on the storage of CO₂ in coal goaf exist [20,21,22,23]. The geological conditions of CO₂ storage in goaf are rarely evaluated. It is necessary to study the geological conditions of CO₂ storage because CO₂ leakage can harm groundwater, organisms, and the environment. Effective methods to evaluate the geology are urgently required. In this study, we establish a mechanism for CO₂ storage in coal goaf and evaluate the geological characteristics to provide a reference for CO₂ storage in goaf.

2. Methods and Materials

2.1. Methods of CO₂ Geological Storage

CO₂ geological storage is the most direct and effective method of reducing carbon emissions. Common underground storage methods include using deep saline aquifers in sedimentary basins [24,25], oil and gas fields [26,27], unminable deep coal seams [28,29], and post-mining salt karst cavities [30]. Researchers have proposed the use of CO₂ to replace CH₄ or oil for the extraction and storage of CO₂ [31]. The exploration of CO₂ geological storage methods has primarily focused on deep storage (over 1000 m), with disadvantages such as slow injection rates, high displacement pressure, high comprehensive investment costs, obvious geological constraints, long distances from CO₂ production concentration areas, and the inability to store CO₂ on a large scale. It is necessary to identify widely distributed, convenient, and low-cost storage areas for CO₂ potentially at the expense of a certain storage capacity under the premise of safety according to the Chinese situation.

The critical pressure of CO₂ is 7.38 MPa, and its critical temperature is 31.1 °C (Figure 3a). CO₂ remains in a liquid state despite temperature changes when the pressure exceeds 7.38 MPa. The density of CO₂ in the process of CO₂ geological storage gradually increases with an increase in the injection depth, but the change is minor. The density of CO₂ reaches a supercritical state when the depth reaches or exceeds 800 m (Figure 3b).

The burial depth of coal in China is generally 500–1000 m. The feasibility of CO₂ storage in the goaf of medium and shallow coal mining is a topic of concern for Chinese geologists [14,20,21,22,23,33]. A pilot experiment for CO₂ storage was conducted in the Southern Shanxi Formation of the Qinshui Basin in Shanxi Province (buried at depths of 472.34~478.70 m), with an injection of 192.8 tons of liquid CO₂. The storage effect was satisfactory [14,33]. Wang et al. [21] studied the occurrence conditions of natural CO₂ in the Yaojie mining area of Gansu Province (<400 m) and concluded that it is feasible to store CO₂ in coal goaf. The geological evaluation of CO₂ storage in coal goaf provides a reference for the future geological storage of CO₂ in coal goaf.

2.2. Mechanism of CO₂ Storage in Coal Goaf

The mechanism of CO₂ storage in goaf is divided into physical trapping and chemical trapping.

2.2.1. CO₂ Physical Trapping

The physical storage mechanisms include geological trapping and bound residual CO₂ trapping.

(1)

Geological trapping

Geological trapping uses the upper structure of a reservoir space (the associated after-effects of tectonic or orogenic processes) to prevent the upward migration of CO₂ under buoyancy, thereby of storing CO₂. A stratigraphic trap, a structural trap, or a combination of both can support CO₂ storage. Such structures are suitable for normal-pressure storage. Traps include anticlines and closed faults, illustrated in Figure 4. Liu et al. [19] proposed the rebuilding of storage facilities in goaf for supercritical CO₂ storage. The mining depth (500–1000 m) of coal mines in China represents middle and shallow layers, permitting storage under normal pressure. Stable roadway spaces can be stored in artificially reconstructed storage tanks with a supercritical storage capacity.

(2)

Residual CO₂ trapping

An amount of CO₂ exists in the pores of a medium because of the effects of capillary force, surface tension, and organic carbon adsorption. The carbon content and porosity in the cover layer, the internal surface area (20–300 m²/g), and the coal pillar of coal seams are higher; thus, CO₂ is more easily absorbed [34]. The properties of the cover layer are similar to those of the coal seam, potentially permitting storage of up to 60 m³ (standard temperature and pressure, or STP) CO₂ per ton of coal at relatively low gas pressures of approximately 5 to 6 MPa [35]. Coal seams and cover layers can store at least five times (even ten times, for the most captive and porous layers) the quantity of gas that is traditionally contained within a classical reservoir rock.

2.2.2. CO₂ Chemical Trapping

The chemical storage mechanisms include solubility trapping and mineral trapping.

(1)

Solubility trapping

Solubility trapping is the dissolution of CO₂ in underground fluids. The solubility of CO₂ in water is influenced by the temperature, pressure, and salinity of the water. These factors affect the storage time.

(2)

Mineral trapping

Mineral trapping is the fixation of CO₂ by carbonate minerals through a reaction with certain components of the surrounding rock (phyllosilicate minerals) and water (sulfate). It is affected by the storage depth, salinity, and pH value of the fluid and the mineral composition of the surrounding rock. Mineral carbonation is a low-temperature/low-pressure process, making it feasible for use in relatively shallow geological settings.

2.3. Geological Evaluation Methods

2.3.1. Introduction to AHP and WIM Models

Studying the geological conditions of CO₂ storage in goaf can not only prevent CO₂ leakage to the surface or near-surface environment but also ensure that health, safety, and environmental issues are addressed. Thus, a comprehensive analysis of the geological conditions of CO₂ in the goaf of coal is necessary. Geological CO₂ storage involves still-immature technology, especially in the goaf of coal. There is also a lack of an evaluation system to assess the suitability of coal goaf for geological CO₂ storage. In this study, we used the results of previous studies from CO₂ storage sites such as deep salt water, oil and gas reservoirs, unminable coal seams, and karst cavities. They were all geologically evaluated using AHP [34].

AHP is a decision-making method that decomposes elements always related to decision making into different levels such as goals, criteria, and schemes. Qualitative and quantitative analyses can then be conducted [36]. AHP divides the various factors within these complex problems into interconnected or orderly layers, thereby rationalizing complex problems and assigning quantitative relative importance values (i.e., weighting) to each layer using fuzzy logic approaches. The geological storage of CO₂ in the goaf of coal is influenced by numerous complex factors such as geology, geography, and technology, as well as environmental, political, and economic variables.

AHP, which has a fixed factor weight, cannot process certain situations where the order is chaotic because of the importance of elements in complex systems as well as possible uncertainties between the criteria and index. An AHP based on the weighted interval method (WIM) can effectively process the fixed weight of a traditional AHP. A weight determined by the experts is amalgamated with the cognition of factors. This ensures that the interval weight is closer to the real result. The core of interval weight theory is to enable experts to assign factor weights, enhance the sensitivity to risk, and reduce the risk of the whole process to maintain each expert’s perception of risk factors.

2.3.2. Methods of AHP and WIM Models

The methods of AHP and the WIM models prevent the influence of potential risk factors and reveal the impact of potential risks on the evaluation system. The calculation procedure of the AHP and WIM models is as follows.

First, the geological evaluation grade of the research area is determined. The evaluation of the potential and suitability for CO₂ geological storage in China is classified from low to high accuracy into national grade, basin grade, target area grade, site grade, and injection grade [37]. Guo et al. [38] classified CO₂ storage sites into regional grade, basin grade, target area/target area grade, and site grade by adjusting the in scale and evaluation methods. Previous classification standards [37,38] divide the CO₂ storage sites in coal goaf into regional grade, mine field grade, mine grade, and perfusion grade based on the distribution characteristics and discontinuity of the goaf in coal. The object, or the scale size of data collection, and field investigation of CO₂ storage sites are determined according to the degree of work in the research area (

Table 1. Classification of CO₂ geological storage stages in goaf of coal.

Stage	Object	Evaluation
		Objective	Method	Scale	Potential
Regional grade	Secondary structural units in sedimentary basins	Calculate the predicted storage potential and evaluate the geological storage prospect area	Data collection, remote sensing interpretation, field research, and comprehensive research	1:50,000–1:200,000	Predicting
Mine field grade	Third structural units in sedimentary basins	Select the target area for geological storage within the structural trap	Data collection, remote sensing interpretation, field research, and comprehensive research	1:10,000–1:50,000	Control
Mine grade	Coal goaf	Provide a basis for the engineering design of favorable storage goaf	Remote sensing interpretation, field research, comprehensive research, sample collection and testing, numerical simulation	1:1000–1:10,000	Basic
Perfusion grade	Unit in the goaf	Construction design and environmental monitoring	Comprehensive research, sample collection and testing, numerical simulation	1:200–1:1000	Sealing

).

Second, a hierarchy consisting of three different elements (goal, criteria, and scheme) is built. AHP is used to determine the elements, indicators, and inter-layer relationships to clarify the structural relationships. The evaluation index system of the hierarchical analysis structure is established by selecting the geological safety, storage scale, social environmental risk, and economic suitability based on the principles of economic, moderate scale, and environmentally friendly CO₂ geological storage locations (Figure 5). The comprehensive evaluation includes 4 aspects, 38 indicators, and 4 index levels. There may be significant differences in the actual evaluation process of micro-indicators in certain regions, such as permeability and land cover ratio. These are added to the indicator layer.

Third, the evaluation units are divided and refined using a GIS grid. The evaluation units for CO₂ storage are divided based on structures (faults, folds) and tunnels, etc., as boundaries. Their corresponding reserves can then be measured.

Fourth, the raw data are non-dimensionalized. The raw data are first standardized because of the significant dimensional differences between different indicators. The non-dimensional data all between 0 and 1, compared with the standard, can be combined with the geological conditions of CO₂ storage facilities at home and abroad [37,38,39]. The main influencing factors for the storage conditions of oil, natural gas, and coalbed methane are the geological safety factors, storage potential factors, socio-economic factors, and environmental protection factors. CO₂ storage is then appropriately classified. Each indicator is then allocated into one of the following four levels (see Section 3.2 for the classification): suitable ([0.75, 1.00]), relatively suitable ([0.50, 0.75]), generally suitable ([0.25, 0.50]), and unsuitable ([0, 0.25]). Each block in a study area can then be assigned and recorded as matrix R.

$$\mathrm{R}=\left[\begin{array}{cccc}{\mathrm{r}}_{11}& {\mathrm{r}}_{22}& {\mathrm{r}}_{\mathrm{ii}}& {\mathrm{r}}_{\mathrm{nn}}\end{array}\right]$$

Fifth, a judgment matrix is established. The weights of various factors and indicators can be compared with each other by experts. The importance level is calculated based on the collected data and field research results, e.g., g_ij is a comparison result of the importance of element i and element j. The judgment matrix is formed by comparing the results of pairwise factors. The quantitative scale used to compare the importance of elements is presented in

Table 2. Comparison of element importance and quantitative scaling relationship.

	Same	Slightly	Obvious	Notable	Very	Extraordinary
Quota	0.5	0.6	0.7	0.8	0.9	1

.

$$\mathrm{G}=\left[\begin{array}{cccc}{\mathrm{g}}_{11}& {\mathrm{g}}_{12}& \cdots & {\mathrm{g}}_{1\mathrm{n}}\\ {\mathrm{g}}_{21}& {\mathrm{g}}_{22}& \cdots & {\mathrm{g}}_{2\mathrm{n}}\\ {\mathrm{g}}_{\mathrm{i}1}& {\mathrm{g}}_{\mathrm{i}2}& \cdots & {\mathrm{g}}_{\mathrm{i}\mathrm{n}}\\ {\mathrm{g}}_{\mathrm{n}1}& {\mathrm{g}}_{\mathrm{n}2}& \cdots & {\mathrm{g}}_{\mathrm{n}\mathrm{n}}\end{array}\right]$$

where is g_ik + g_ki = 1, 0 ≤ g ≤ 1.

Certain risks decrease or disappear with an increase in the degree of the understanding of unknown things in the process of field investigations and data analyses. The influencing factors can then be divided into unstable and stable factors. Stable factors do not significantly with time or there is a clear understanding of them. Unstable factors are those that undergo significant changes over time or have controversial perceptions of them. Unstable elements can be evaluated using the WIM when assigning weights to reduce their difficulty. Stable elements use values to assign weights to increase the accuracy.

Judgment matrices with different values may be produced because of differences in the cognition of experts on various elements and indicators. The interval method is used to integrate the weight matrix. [g_ik⁻, g_ik⁺] is expressed in the matrix; g_ik⁻ is the minimum value of the expert evaluation weight, and g_ik⁺ is the maximum value.

$${\mathrm{G}}^{\prime }=\left[\begin{array}{cccc}\left[{{\mathrm{g}}_{11}}^{-},{{\mathrm{g}}_{11}}^{+}\right]& \left[{{\mathrm{g}}_{12}}^{-},{{\mathrm{g}}_{12}}^{+}\right]& \cdots & \left[{{\mathrm{g}}_{1\mathrm{n}}}^{-},{{\mathrm{g}}_{1\mathrm{n}}}^{+}\right]\\ \left[{{\mathrm{g}}_{21}}^{-},{{\mathrm{g}}_{21}}^{+}\right]& \left[{{\mathrm{g}}_{22}}^{-},{{\mathrm{g}}_{22}}^{+}\right]& \cdots & \left[{{\mathrm{g}}_{2\mathrm{n}}}^{-},{{\mathrm{g}}_{2\mathrm{n}}}^{+}\right]\\ \left[{{\mathrm{g}}_{\mathrm{i}1}}^{-},{{\mathrm{g}}_{\mathrm{i}1}}^{+}\right]& \left[{{\mathrm{g}}_{\mathrm{i}2}}^{-},{{\mathrm{g}}_{\mathrm{i}2}}^{+}\right]& \cdots & \left[{{\mathrm{g}}_{\mathrm{i}\mathrm{n}}}^{-},{{\mathrm{g}}_{\mathrm{i}\mathrm{n}}}^{+}\right]\\ \left[{{\mathrm{g}}_{\mathrm{n}1}}^{-},{{\mathrm{g}}_{\mathrm{n}1}}^{+}\right]& \left[{{\mathrm{g}}_{\mathrm{n}2}}^{-},{{\mathrm{g}}_{\mathrm{n}2}}^{+}\right]& \cdots & \left[{{\mathrm{g}}_{\mathrm{n}\mathrm{n}}}^{-},{{\mathrm{g}}_{\mathrm{n}\mathrm{n}}}^{+}\right]\end{array}\right]$$

Finally, the comprehensive evaluation values of each block are calculated to evaluate the geological storage level of the block.

$$\mathrm{P}=\mathrm{R}*\mathrm{G}\prime =\left[\begin{array}{cccc}{\mathrm{r}}_{11}& {\mathrm{r}}_{22}& {\mathrm{r}}_{\mathrm{i}\mathrm{i}}& {\mathrm{r}}_{\mathrm{n}\mathrm{n}}\end{array}\right]*\left[\begin{array}{cccc}\left[{{\mathrm{g}}_{11}}^{-},{{\mathrm{g}}_{11}}^{+}\right]& \left[{{\mathrm{g}}_{12}}^{-},{{\mathrm{g}}_{12}}^{+}\right]& \cdots & \left[{{\mathrm{g}}_{1\mathrm{n}}}^{-},{{\mathrm{g}}_{1\mathrm{n}}}^{+}\right]\\ \left[{{\mathrm{g}}_{21}}^{-},{{\mathrm{g}}_{21}}^{+}\right]& \left[{{\mathrm{g}}_{22}}^{-},{{\mathrm{g}}_{22}}^{+}\right]& \cdots & \left[{{\mathrm{g}}_{2\mathrm{n}}}^{-},{{\mathrm{g}}_{2\mathrm{n}}}^{+}\right]\\ \left[{{\mathrm{g}}_{\mathrm{i}1}}^{-},{{\mathrm{g}}_{\mathrm{i}1}}^{+}\right]& \left[{{\mathrm{g}}_{\mathrm{i}2}}^{-},{{\mathrm{g}}_{\mathrm{i}2}}^{+}\right]& \cdots & \left[{{\mathrm{g}}_{\mathrm{i}\mathrm{n}}}^{-},{{\mathrm{g}}_{\mathrm{i}\mathrm{n}}}^{+}\right]\\ \left[{{\mathrm{g}}_{\mathrm{n}1}}^{-},{{\mathrm{g}}_{\mathrm{n}1}}^{+}\right]& \left[{{\mathrm{g}}_{\mathrm{n}2}}^{-},{{\mathrm{g}}_{\mathrm{n}2}}^{+}\right]& \cdots & \left[{{\mathrm{g}}_{\mathrm{n}\mathrm{n}}}^{-},{{\mathrm{g}}_{\mathrm{n}\mathrm{n}}}^{+}\right]\end{array}\right]$$

The minimum P value is calculated by multiplying the interval weight of the element or index with the grade. Suitable areas [0.75, 1.0], above average [0.50, 0.75], average areas [0.25, 0.5], and unsuitable areas [0, 0.25] are divided in the map by comparing the standard values.

3. Results and Discussion

3.1. An Example of CO₂ Geological Storage in Goaf

3.1.1. Geological Setting of the Study Area

We selected the Yaojie mining area in Gansu Province as the research area. The Yaojie mining area is located northwest of the Minhe Basin. The faults in the mining area, both compressive and torsional, are extremely developed, and there is an NNW basement fault structure at the eastern end of the area. The overlying strata comprise the Yaojie and Hengtang groups, which consist of multiple sets of mudstone, sandstone, and shale (Figure 6). The overlying strata are the Yaojie and Hengtang groups, consisting of multiple sets of mudstone, sandstone, and shale. The burial depth of the coal seam is 320–380 m, and the coal thickness is 30 m. The mining area was divided into six units based on the structural distribution. The old goaf of the mining area was selected as an example for evaluation.

3.1.2. Classification of Influencing Factors

The main influencing factors were geological safety, storage potential, socio-economic, and environmental protection. These were based on the storage conditions of oil, natural gas, and coalbed methane combined with the geological conditions of CO₂ storage facilities at home and abroad [37,38,39].

Geological Safety Factors

Geological safety factors can be subdivided into regional crustal stability, sedimentary basin properties, cap rock sealing, and hydrogeological conditions.

(1)

Regional crustal stability

The risk of CO₂ escape remains because of the changes in temperature and pressure in the earth, as well as sudden geological events such as volcanoes and earthquakes caused by tectonic movements. Regional crustal stability includes the tectonic setting as well as volcanic, earthquake, and regional fault activity.

1) Tectonic setting

The tectonic background of the research area was a concrete manifestation of the long-term stability of the crust within the area, potentially affecting the stability of the long-term storage of CO₂ in the goaf. We divide the tectonic setting into cratonic basins, inland rift basins, passive continental margin basins, and strike slip fault basins based on the stability of the coal-forming basins. The corresponding levels are presented in Table 2.

2) Volcano

Volcanic activity can penetrate the upper and lower strata, simultaneously producing faults and fractures can destroy the sealing of the CO₂ cap rock. The volcanic development zone and distance from the volcanic zone were used to evaluate volcanoes. The specific classification is listed in Table 2.

3) Earthquake

The geological storage of CO₂ in goaf is classified as grade I of the ‘Seismic Safety Evaluation of Engineering Sites (GB 17741-2005)’ [40], which refers to the storage standards of nuclear waste. The peak acceleration of ground motion, historical earthquake, the distance from the earthquake zone, and other indicators were used to evaluate earthquake (

Table 3. Grading evaluation table of the regional geological stability index.

	Suitable	Above Average	Average	Below Average
Tectonic setting	Craton Basin	Inland Rift Valley	Passive margin basin	Strike slip fault basin
Volcano	No	Low incidence area	Volcano occurrence area	Volcanic prone areas
Distance from the volcanic zone (km)	>250	100–250	25–100	<25
Earthquake	No	Low incidence area	Earthquake occurrence area	Earthquake prone areas
Historical earthquakes	Enclosed area	<4	4–6	>4
Peak ground acceleration (g)	<0.05	0.05–0.1	0.1–0.2	>0.2
Fault	No	Cracks, no faults	Mudstone filling fault	Deep faults, large cracks
Active faults	No active faults within 25 km	No active faults within 25 km, with unclear faults	No active faults within 10 km	No active faults within 10 km, with unclear faults

). The peak ground acceleration was ascertained using the ‘China Seismic Parameter Zoning Map (GB18306-2001)’ [41] from the ‘China Seismic Peak Ground Acceleration Zoning Map’.

4) Regional fault activity

We referred to the relevant nuclear power plant specifications in China for the regional fault activity requirements for CO₂ geological storage in goaf (Table 3). There were no active faults within a 10 km radius of the site. An active fault was defined as a fault movement that occurred between 10,000 and 35,000 years ago, with small-scale seismic activity and multiple historical seismic events recorded by instruments. The possibility of surface displacement might remain.

(2)

Sedimentary basin properties

The study of sedimentary basin properties during CO₂ geological storage mainly considers the stress environment, geothermal flow value, and geothermal temperature. In this study, we divided the sedimentary basins were divided into compression, compression-torsion, torsion, tension–torsion, and tension based on the dynamic environment of current sedimentary basin deformation data combined with the escape channels of a CO₂ geological storage structure.

From the data in

Table 4. Grading evaluation table of sedimentary basin properties.

	Suitable	Above Average	Average	Below Average
Mechanical environment	Compressive	Compression torsion	Torsion	Tension–torsion, tension
Geothermal flow value (mW/m2)	<30	30–50	50–90	>90
Ground temperature (°C)	<1	1–2	2–4	>4

, we concluded that the lower the values of geothermal heat flow and surface temperature, the more conducive the site to CO₂ storage.

(3)

Sealing of the cap rock

The sealing of the cap rock is a prerequisite for CO₂ storage in the goaf of coal; it is also the factor most studied by scholars. Downey [42] and Ingram et al. [43] proposed a process for comprehensive evaluation from both micro and macro perspectives. Li et al. [44] established an analytical model to ascertain the stability of cap rocks under injection production disturbance as well as a numerical model to evaluate the sealing performance of the combined cap rock. Geological macro-factors, including the lithology, thickness, continuity, spatial combination, and mining technology, were used to evaluate the sealing performance of the cap rock (

Table 5. Grading evaluation table of the sealing of cap rock.

	Suitable	Above Average	Average	Below Average
Lithology	Mudstone, marl	Shale, sandy mudstone	Argillaceous siltstone	Sandstone
Single layer thickness	>20 m	10–20 m	2.5–10 m	<2.5 m
Accumulated thickness	>300 m	100–300 m	20–100 m	<20 m
Continuity	Regional distribution	Continuity of coalfield	Almost continuously	Uncontinual
Spatial combination	Many	2	1	No
Mining technology	Backfill	Mainly for backfill	Mainly for stope caving	Stope caving

).

1) Lithology

The lithology of common cap rocks in coal includes marl, mudstone, shale, argillaceous siltstone, and sandstone. There are significant differences in the mineral composition, pore throat structure, and physical properties of the rock types. The higher the clay content in the cap rock, the lower the porosity and permeability of the rock layer, the higher the displacement pressure, and the stronger the sealing ability of the cap rock. Indicators such as the mud ratio or mud–sand ratio are used to evaluate the sealing ability of cap rocks [45,46]. The toughness of rocks also affects the sealing of the cap rock. In general, brittle rock strata are more likely to produce fractures and cracks than ductile rock strata. The higher the clay mineral content, the stronger the rock toughness. Overall, the sealing capacities of marl, mudstone, shale, argillaceous siltstone, and sandstone decreases in sequence.

2) Cap rock thickness

The cap rock thickness is the main evaluation index of sealing. Hubbert [47] observed that mudstone a few inches thick could seal an oil column hundreds of meters high. When the cap rock is thin, it is often unstable; this is unfavorable for large-scale CO₂ geological storage. A thick layer is not easily damaged by small faults, making it difficult to form connected micro-cracks. Wang et al. [48] obtained empirical data through experiments and believed that a 20 or 30 m thick coal seam roof could be used as an effective sealing thickness. In this study, we determined the thickness of the cap rock for CO₂ storage by referring to the thickness data of the cap rock for oil and gas storage (Table 5).

3) Continuity of cap rock

The continuity of cap rock is an important factor in the large-scale enrichment of CO₂. The thicker the cover layer, the larger the area, the better the continuity of lateral distribution and the safer the storage of CO₂. CO₂ storage in the goaf of coal must prevent unsealed boreholes in the process of coal exploration as well as prevent the development of faults and fractures in the cap rock. A fracture can result in the cap rock completely losing its sealing ability, especially when the fault throw is greater than the thickness of the cap rock and the fracture zone is open.

4) Cap rock space combination

CO₂ storage in goaf must possess a good regional cap layer to prevent the direct leakage of CO₂. In general, a single geological cap rock that meets the sealing capacity requirements can effectively store CO₂. CO₂ leakage may occur if the geological cap rock breaks because of factors such as earthquakes. The continuity of faults is destroyed during the long-term storage period. An overlying multi-level cap rock structure can achieve secondary interception or sealing after the direct cap rock fails if there is a multi-layer sealing combination on the direct cap rock to form a multi-level cap structure. This prevents the further upward leakage of CO₂ after breaking through the direct cover layer. Thus, the cap rock space combination is measured by the sealing ability of the cap rock. The better the quality and the greater the number of secondary cap rocks, the stronger the secondary interception and sealing ability, the higher the long-term safety, and the lower the risk of CO₂ geological storage.

5) Coal mining technology

Coal mining technology affects the continuity of CO₂ storage roofs. The filling and fallen method are used to process the roofs in goaf. The filling method results in little damage to the roof. The cap rock remains relatively complete, which is conducive to effective CO₂ storage. The fallen method causes significant damage to the roof, which then requires the consideration of factors such as the burial depth, coal mining thickness, and cover layer thickness.

(4)

Hydrogeological conditions

The influence of hydrogeological conditions on the occurrence and migration of coalbed methane and natural gas includes hydraulic migration and dissipation as well as hydraulic sealing [49,50]. The storage of CO₂ in goaf uses space in the coal goaf. The storage space is affected if filled with water. If the amount of groundwater is not significant and contains chemicals; it reacts with CO₂ and plays a role in mineral solidification. The hydrogeological conditions for CO₂ storage in goaf are evaluated before CO₂ filling. The indexes considered include the groundwater volume, the groundwater temperature, and salinity (

Table 6. Grading evaluation table of hydrogeological conditions.

	Suitable	Above Average	Average	Below Average
Groundwater volume	No	A little water	Water remains unchanged	Large
Salinity (g/L)	30–50	10–30	0–10	>50
Temperature (°C)	<2	2–3	3–4	>4

).

Storage Potential Factors

The storage potential is an important factor when evaluating CO₂ storage in goaf. Only goaf areas that can store a large amount of CO₂ have the necessary conditions for development and use. The evaluation indicators include the storage area, storage capacity, exploration degree, surface temperature, and geothermal gradient. The specific classification parameters are presented in

Table 7. Grading evaluation table of storage potential factors.

	Suitable	Above Average	Average	Below Average
Storage area (km2)	>50	10–50	1–10	<1
Storage capacity (104t)	>500	100–500	10–100	<10
Exploration degree	Exploration	Part of exploration	General exploration	General investigation
Temperature (°C)	<−2	−2–10	10–20	>20
Geothermal gradient (°C/100 m)	<2	2–3	3–4	>4

.

Economic Factors

Economic factors are important factors for the implementation of CO₂ storage in goaf. The main indicators include the carbon source scale, storage capacity per unit area, carbon source distance, land use status, transportation mode, landform, distance from residential areas, and exploration level (

Table 8. Grading evaluation table of the economic factors.

	Suitable	Above Average	Average	Below Average
Carbon source scale (104t/a)	>50	25–50	10–25	<10
Storage capacity per Unit area (104t/km2)	>150	50–150	10–50	<10
Carbon source distance (km)	<10	10–50	50–100	>100
Distance from residential areas (m)	>1200	1000–1200	800–1000	<800
Land use status	Deserts or unused land	Grassland and forest land	Cultivated land	Residential areas
Transportation mode	Pipeline	Short-distance road	Road	Railway
Landform	Dune	Hill	Flat land	Mountain
Terrain slope (°)	0–5	5–10	10–25	>25
Exploration degree	Exploration	Part of exploration	General exploration	General investigation
Data	Sufficient and reliable	Sufficient data, generally reliable	Data and reliability are average	Insufficient

). These indicators were ascertained based on the China Agenda 21 Management Center and previous research [38,51]. A larger carbon source scale means a more concentrated carbon source, a closer carbon source distance, a lower CO₂ transportation cost, worse land use, a higher exploration degree, more data support, and a greater distance from residential area, with more suitability for CO₂ storage.

Environmental Protection

Environmental protection is an indispensable element in engineering construction. The evaluation indexes of CO₂ storage in goaf include population density, geological-hazard-prone areas, and adverse geological processes. The lower the population density, the fewer the geological-disaster-prone areas. The fewer the adverse geological effects, the greater the suitability for CO₂ storage (

Table 9. Grading evaluation table of environmental protection.

	Suitable	Above Average	Average	Below Average
Population density (/km2)	<25	25–50	50–100	>100
Geological-hazard-prone areas	No	Few	Hidden	Exist
Adverse geological processes	No	Few	Hidden	Exist

).

3.1.3. Index Weight Determination

The judgment matrix G required by AHP was obtained from an evaluation of the experts’ comparisons of the modal particles in each element. The matrix of the first-level index was obtained as follows.

$$\mathrm{G}=\left[\begin{array}{cccc}0.50& 0.80& 0.75& 0.70\\ 0.20& 0.50& 0.45& 0.30\\ 0.25& 0.55& 0.50& 0.40\\ 0.30& 0.70& 0.60& 0.50\end{array}\right]$$

Stable factors were assigned a fixed value, and unstable factors were assigned an interval value based on the comparison matrix of the geological status and factor importance of the study area. The individual weightings for all evaluation indexes related to the suitability of an assigned coal goaf for geological CO₂ storage were calculated using AHP.

Table 10. Weighting values related to the factors that control the suitability of coal goaf for geological CO₂ storage.

First-Level Parameter (A)	Second-Level Parameter (B)	Third-Level Parameter (C)	Fourth-Level Parameter (D)
Geological safety factors (A1) [0.45, 0.55]	Regional crustal stability (B1) 0.10	Tectonic setting (C1) 0.10
		Volcano (C2) 0.10	Volcano (D1) 0.75
			Distance from the volcanic (D2) 0.25
		Earthquake (C3) 0.10	Earthquake (D3) 0.65
			Historical earthquakes (D4) 0.35
		Regional fault activity (C4) 0.60	Peak ground acceleration (D5) 0.15
			Fault (D6) 0.40
			Active faults (D7) 0.55
	Sedimentary basin properties (B2) 0.15	Mechanical environment (C5) 0.60
		Geothermal flow value (C6) 0.20
		Ground temperature (C7) 0.20
	Sealing of the cap rock (B3) [0.60, 0.65]	Lithology (C8) 0.30
		Cap rock thickness (C9) [0.20, 0.30]	Single layer thickness (D8) [0.60, 0.70]
			Accumulated thickness (D9) [0.30, 0.40]
		Continuity (C10) 0.20
		Spatial combination (C11) [0.10, 0.15]
		Mining technology (C12) [0.1, 0.2]
	Hydrogeological conditions (B4) [0.10, 0.15]	Groundwater volume (C13) 0.60
		Salinity (C14) 0.10
		Temperature (C15) 0.30
Storage potential factors (A2) 0.10	Storage area (B5) [0.10, 0.20]
	Storage capacity (B6) [0.55, 0.65]
	Exploration degree (B7) 0.10
	Temperature (B8) 0.05
	Geothermal gradient (B9) 0.10
Economic factors (A3) [0.15, 0.25]	Carbon source scale (B10) 0.10
	Storage capacity per unit (B11) 0.05
	Carbon source distance (B12) 0.35
	Distance from residential (B13) [0.05, 0.10]
	Land use status (B14) [0.05, 0.10]
	Transportation mode (B15) 0.10
	Landform (B16) 0.05
	Terrain slope (B17) 0.05
	Exploration degree (B18) 0.10
	Data (B19) 0.05
Environmental protection (A4) 0.20	Population density (B20) [0.45, 0.60]
	Geological hazard-prone (B21) [0.20, 0.45]
	Adverse geological processes (B22) 0.10

lists the resulting values.

3.1.4. Classification of Geological Conditions for CO₂ Storage

The matrix values of the influencing factors in the study area were determined using the factors listed in Table 2, Table 3, Table 4, Table 5 and Table 6 and the values of the comparison criteria. For example, the fourth level parameters (D) were subject to membership matrix.

$${\mathrm{R}}_4=\left[\begin{array}{cccc}0.80& 0.90& 0.80& 0.80\\ 0.90& 1.00& 0.60& 0.75\\ 0& 0& 0.50& 0\end{array}\right]$$

The data in Table 2, Table 3, Table 4, Table 5 and Table 6 characterized the similarity degree matrix outlined for the other grading indexes. The following example illustrates a matrix for class C reservoir characteristics, class B internal conditions, and class A comprehensive index evaluation matrix values.

$${\mathrm{R}}_3=\left[\begin{array}{cccc}1.00& 1.00& 0.80& 0.70\\ 0.825& 0.80& \left[0.768,0785\right]& 0.60\\ 0.935& 0.75& 0.75& 0.80\\ 0.635& 0& 0.80& 0\\ 0& 0& 0.60& 0\end{array}\right]$$

$${\mathrm{R}}_2=\left[\begin{array}{cccc}0.657& 0.40& 1.00& \\ 0.91& 0.60& 0.70& 0.90\\ \left[0.684,0.866\right]& 0.90& 0.90& 0.90\\ 0.72& 1.00& 0.80& 1.00\\ 0& 0.85& 0.70& 0\\ 0& 0& 1.00& 0\\ 0& 0& 1.00& 0\\ 0& 0& 0.90& 0\\ 0& 0& 0.90& 0\end{array}\right]$$

$${\mathrm{R}}_1=\left[\begin{array}{c}\left[0.684,0.873\right]\\ \left[0.595,0.695\right]\\ \left[0.755,0.833\right]\\ \left[0.685,0.945\right]\end{array}\right]$$

3.2. Comprehensive Evaluation

The weight vectors and index degree matrix were combined with the weighted average algorithm to determine a comprehensive evaluation vector for the suitability of the Yaojie coal mine to geologically store CO₂.

$$\mathrm{P}={\mathrm{R}}_1*\mathrm{G}\prime =\left[\begin{array}{c}\left[0.684,0.873\right]\\ \left[0.595,0.695\right]\\ \left[0.755,0.833\right]\\ \left[0.685,0.945\right]\end{array}\right]*\left[\begin{array}{cccc}\left[0.45,0.55\right]& 0.10& \left[0.15,0.25\right]& 0.20\end{array}\right]=\left[0.62,0.95\right]$$

The principle of the minimum suitable value indicated that the goaf of the Yaojie coal mine represented an above average location for the geological storage of CO₂. These results were similar to the results [38] of a geological CO₂ storage assessment of goaf performed by the Shaanxi Provincial Key Laboratory of Geological Support for Coal Green Exploitation, indicating that the approach outlined in our research produced effective and reasonable results. Our evaluation results for CO₂ geo logical storage had a certain reference significance.

The geological evaluation of CO₂ storage in the goaf of coal mainly comprised 4 categories, 4 levels, and 38 indexes. The selected indicators, which were macro-factors, covered geological safety, storage potential, and economic and environmental protection factors. We used the storage conditions of oil, natural gas, and coalbed methane combined with the geological conditions of CO₂ storage facilities at home and abroad for the evaluation indicator selection process. Thus, the selected indexes involved a wide range of factors and were complete.

The WIM effectively reduces the engineering risk and evaluation difficulty. It is difficult to quantify uncertain factors such as the population density and geological hazards of environmental protection indicators. Currently, experts have greater certainty and spend less time on evaluation when using the WIM because it uses minimum values to evaluate the grade, thereby reducing the evaluation risk.

Compared with traditional CO₂ storage suitability evaluation, this method inherits the hierarchical classification and logical relationship construction in traditional evaluation methods. The weight of the traditional evaluation method is determined according to the importance between the indicators. However, it is difficult to use fixed values to measure the importance of each indicator in the actual evaluation process. The weight interval method used to reduce the difficulty of comparing the importance of indicators and better handle the differences between indicators. At the same time, the principle of using the minimum appropriate value can reduce the risk and make the environment more friendly in the comprehensive evaluation.

Certain issues remain regarding the selection of indicators for this evaluation method, including the non-consideration micro-indicators. The shale content in sandstone is relatively low in certain areas and the permeability is low; thus, this may represent good cap rock.

4. Conclusions

(1) In this study, we outlined a method to evaluate the suitability of geological CO₂ storage in the goaf of coal. We divided the influencing factors into multiple macro-indicators and quantified them using classification levels. Our method combined the principles of safety, economy, and environmental protection to reduce evaluation difficulties and improve safety whilst ensuring economic feasibility and environmental friendliness.

(2) We combined our approach with the characteristics of coal goaf to develop a hierarchical index system model that could be applied to the comprehensive evaluation of any assigned coal goaf. The WIM was used to weight each indicator. This reduced the issues of weight allocation and risk, thereby providing reference values for the future assessment of the geological CO₂ storage potential of coal goaf. Furthermore, it provides theoretical support for CO₂ storage strategies and climate change mitigation.

(3) Our geological CO₂ storage assessment model was adopted to assess the geological storage potential of the goaf of Yaojie mine. This evaluation indicated that the basin is an above average area for coalbed geological CO₂ storage.