The Palaeocene Lista Shale: A Planned Carbon Capture and Storage Top Seal for the East Mey CO₂ Storage Site

Abstract

Top seals and overburden above reservoirs at geological carbon capture and storage (CCS) sites can be major concerns when they are at risk of being mineralogically and texturally unstable in the presence of high-pressure CO₂. Here we report on the pore systems, mineralogy, and surface area attributes of the Palaeocene Lista Shale, the caprock to the Mey Sandstone at the UK’s planned East Mey CCS site. The core was logged, and then mineral quantification was undertaken with X-ray powder diffraction mineralogy, light optics and electron microscopy analyses. Laser particle size analysis was used for grain size determination. Porosity, pore throat diameter, surface area and pore body size were measured via mercury intrusion porosimetry and nitrogen adsorption analyses. The mudstone facies from the Lista Shale are dominated by smectite-rich matrix and silt-grade quartz, with small quantities of chlorite and sodic-plagioclase. Chlorite, sodic-plagioclase, and even smectite are known to be capable of reacting with, and potentially leading to mineral sequestration of CO₂. The mean pore throat and pore body diameters are 17 and nearly 18 nm, respectively, showing that the Lista is mesoporous; the similarity of pore body and pore throat dimensions reveals a predominance of plate and slit pores. Gas adsorption analyses revealed that the overall pore structure is complex, with a high tortuosity of fluid movement through a complex clay-rich matrix (this equates to a mean fractal dimension D2 value of 2.67). Gas adsorption analyses have also shown that grain surfaces are moderately complex (rough) due to the dominance of clay aggregates (this equates to a mean fractal dimension D1 value of 2.56). D2 being higher than D1 suggests that there is a relatively low potential to physically store CO₂ gas on grain surfaces. Conversely, the ability of the CO₂ to react with minor quantities of chlorite and sodic plagioclase, or even with smectite, could lead to increasing surface area of the remaining shale minerals with newly exposed reactive silicates leading to further enhanced mineral trapping of the injected CO₂. The restricted pore throat size linked to small grain size and poor sorting, and reflected by the high fractal D2 value, plus limited grain surface complexity, reflected by the low fractal D1 value, collectively suggest that mineral trapping of the injected CO₂ would be relatively slow (on the order of 1000s of years) if CO₂ penetrated the top seal.

1. Introduction

Mudstones represent a major control of fluid flow in sedimentary basins and in near surface environments. Mudstones routinely function as aquicludes or aquitards in sedimentary basins, restricting water movement patterns [1]. In petroleum systems, mudstones act as top seals to many hydrocarbon reservoirs [1,2]. Mudstones are now recognised as important controls on the vertical flow and trapping of CO₂ in carbon capture and storage (CCS) sites [3].

The success of long-term carbon dioxide (CO₂) storage in geological formations is typically expected to require a strong containment system for the injected CO₂ [4]. An important element of the containment system is the top seal stratigraphically above the storage formation [4]. Unlike during the appraisal of oil or gas discoveries, where a successful top seal can be assumed, it is essential to characterise and understand the top seal at any potential CO₂ storage site to minimise the risk of CO₂ leakage. Top seal characterisation involves the use of a range of methods to assess the petrophysical, geochemical and geomechanical properties.

Petrophysical properties, including porosity, pore structure and permeability, are potentially affected when, or if, mudstone interacts with CO₂ [5,6]. It has been suggested that the storage capacity of CO₂ by mudstone can be reduced due to reaction with CO₂ [7]. Significant mineralogical alteration in mudstone has been reported in several studies of CO₂ injection [6,7,8,9]. The reported alterations in mudstone structure were partly caused by the dissolution of CO₂ in formation waters, that subsequently generates carbonic acid, with the latter leading to induced mineral dissolution and precipitation processes [6,10].

One of the essential parameters that control interaction processes at an aqueous fluid-solid interface is a specific surface area [11]. Clay minerals tend to have the highest specific surface areas of any mineral group in sediments, sedimentary rocks and soils due to their small grain size and complex crystal morphology [11]. Here we seek to characterise the specific surface area, mineral composition, and both the primary (depositional) and secondary (diagenetic) factors that influence essential properties such as pore structure, pore throat diameter, and grain size distribution, in the top seal of a proposed CCS project at the Palaeocene East Mey formation in the UK North Sea, as part of the larger Acorn project.

The Acorn project, one of several CCS initiatives currently progressing in the UK [12,13] aims to utilise the Palaeocene Mey Sandstone Member as a storage site within an open saline aquifer located in the Moray Firth Basin in the North Sea, offshore UK (Figure 1A) [14]. The Palaeocene Lista Shale serves as the top seal for the Mey Sandstone Member. Risk assessment of the East Mey storage site was conducted by the Acorn research and technical team to define leakage potential and threats of leakage from, for example, an undetected open fault or a leaky abandoned well [13]. Notably, in the study area, leakage via geological factors was perceived to be less likely to happen than leakage via old abandoned wells [13]. Nonetheless, to understand and minimise the risk of possible leakage potential of CO₂ from the underlying reservoir via the top seal, we have undertaken a full characterisation of the Lista Shale top seal.

A general question of relevance to all future CCS projects is what rock characteristics and specific aspects of a seal’s geological history result in a good seal at a CCS site. This question will be addressed here by reference to the Lista Shale. A synthesis is presented at the end of the paper that draws on the Lista Shale results and interpretation as well as the outcomes of several related studies. Characterising the top seal involves a combination of core analysis, petrographic assessment, and mineral quantification with pore system analysis. In this study, fractal dimension analysis has been applied to examine the pore surface and pore structure of the Lista Shale. The specific research questions that we have answered for the Lista Shale as a potential CCS top seal are as follows:

• What are the characteristic pore sizes?
• What are the genetic origins of the pore types present?
• What is the best way to determine a specific surface area?
• What controls reactive surface area?
• Are there compositional controls on pore structure and surface area?
• Is the Lista Shale likely to be a good seal for the injected CO₂?

2. Geological Background

The turbidite-derived stratigraphic intervals of the Cenozoic Palaeogene cycles of the Central North Sea became economically important in northwest Europe after the first discovery of the hydrocarbon reservoirs at the Arbroath Field in 1969 and the Forties Field in 1970 [15,16]. The collection of production and exploration datasets, and variable access to the core, from the mature North Sea hydrocarbon province has afforded an opportunity to investigate regional scale deep water sedimentation and how it is related to tectonic framework and sea level changes during the Palaeogene in the central North Sea [15]. In this study, we will investigate the relationship between deep water sedimentation of the turbidite system with the mineralogical composition in the Lista Shale CCS caprock.

The Chalk, underlying the Cenozoic clastic succession of interest here, was deposited in the Upper Cretaceous and lowermost Palaeocene (Figure 1). The Ekofisk Formation of the Chalk Group immediately underlies the mudstones of the Palaeocene Montrose Group, which is subdivided between the Maureen and Lista Formations [15,17].

Figure 1. (A) Location map of the East Mey storage site in the Outer Moray Firth, Central North Sea, UK, adopted from the Millenium Atlas [18]. TheStudy area is within the Blair field, specifically at well 16/21a-13. (B) Lithostratigraphy of the East Mey CO₂ storage site, with the storage interval highlighted in red. The reservoir for CO₂ storage is the Mey Sandstone of the Palaeocene Lista Formation within the Montrose Group. This study focuses on the Lista Shale as the top seal for the Mey Sandstone Member.

Figure 1. (A) Location map of the East Mey storage site in the Outer Moray Firth, Central North Sea, UK, adopted from the Millenium Atlas [18]. TheStudy area is within the Blair field, specifically at well 16/21a-13. (B) Lithostratigraphy of the East Mey CO₂ storage site, with the storage interval highlighted in red. The reservoir for CO₂ storage is the Mey Sandstone of the Palaeocene Lista Formation within the Montrose Group. This study focuses on the Lista Shale as the top seal for the Mey Sandstone Member.

The Montrose Group is composed of pelagic shales and turbidite sands [15,17] (Figure 1). The Maureen Formation contains discrete turbidite sandstone horizons (members) separated by pelagic shales. The Lista Formation sits above the Maureen Formation and contains pelagic shales and turbiditic sands. The Mey Sandstone Member, a planned CCS reservoir, sits within the Lista Formation [15,17]. The pelagic Lista Shale represents the top seal of the Mey Sandstone Member reservoir [14,19]. To be clear, the Mey Sandstone Member has also been locally named the Andrew Sandstone Member and the Balmoral Sandstone Member [14,15].

The Moray Group lies above the Montrose Group (Figure 1), consisting of the Sele Formation, associated with the Forties Member and the Balder Tuffaceous Member [15,17]. The thick Quaternary Nordland Group, at the top of the succession, is undifferentiated and dominated by mudstone, claystone and marls [14,19].

The burial and thermal histories of the East Mey site are relatively straightforward, with no significant uplift or erosion events identified since deposition commenced in the Palaeocene (66 to 56 Ma, Figure 2) [14]. The burial depth of the East Mey CCS site, which is close to the Blair Field, is approximately 6000 to 6600 ft (1800–2000 m), with a present-day temperature of about 60 to <70 °C [14]. Note that this relatively low maximum temperature, for a relatively short period of time, implies that mineral diagenesis and fabric alteration will not be highly advanced as it is only at the earliest stages of mesodiagenesis (burial diagenesis) [20] (Figure 2).

3. Methodology

3.1. Characterisation Workflow

The characterisation workflow used in this study has been adopted from the Armitage workflow [21]. To start, the core was logged to investigate sedimentary features. Second, representative core samples were collected from the Lista Formation. Polished thin-section samples were used to carry out quantitative petrographic assessment via SEM/EDS and optical microscopy. Mineralogy was determined using XRD analysis. To relate mineralogy with porosity, mercury intrusion porosimetry analyses (MICP) and nitrogen gas adsorption experiments were undertaken to investigate the different pore types and mean pore size diameters, along with surface area determination (Figure 3). These data were then used for a risk assessment of the Lista Shale.

3.2. Samples and Data

The Blair Field, within the planned CCS site for the East Mey project, is currently operated by Harbour Energy PLC (London, UK) [13,14,19]. Well 16/21a-13 has been selected for this study because, perhaps unusually, the core was taken through sealing lithologies during the assessment of local petroleum resources. Eleven representative samples were collected, made into 30 µm polished thin sections and were prepared for backscattered electron microscopy using standard methods (

Table 1. Sampling of the Lista Shale from well 16/21a-13 including all methods undertaken in this study. The symbol (X) refers to samples taken from each technique and md refers to measured depth.

Depth (md-ft)	BET	MICP	XRD	Thin-Section	Optical Microscope	SEM	LPSA
6975	X	X	X	X	X	X	X
6981	X	X	X	X	X	X	X
6982	X		X				X
6987	X		X	X	X	X	X
6989	X		X	X	X	X	X
6990	X	X	X				X
6992	X		X	X	X	X	X
6995.5	X		X	X	X	X	X
7001	X	X	X	X	X	X	X
7005	X	X	X	X	X	X	X
7008	X	X	X	X	X	X	X
7013	X		X				X
7015	X	X	X	X	X	X	X
7024.5	X	X	X	X	X	X	X

).

3.3. Lista Formation Core Logging

Approximately 62 feet of core was originally cut from well 16/21a-13. From this core, approximately 60% was in fine-grained top seal facies of the Lista Formation. For lithology, grain size, sedimentary structures, bioturbation index (BI), trace fossil and macrofossil identification, the core was logged at high resolution (mm-scale). The bioturbation index scale ranges from 0 to 6, where 0 is no trace of bioturbation and 6 represents total bioturbation [22]. Wireline logs were recorded through the reservoir and top seal at the time the well was drilled (by North Sea Sun Oil Company in 1984) but they proved to be of minimal help for detailed analysis of the lithology and rock properties during this study as neutron and shear compressional log data were not collected [14].

3.4. Microscopy

Optical microscopy was undertaken on polished thin sections using an Olympus BX51 microscope with an Olympus SC50 camera employing both plane polarised light (PPL) and crossed polarised light (XPL) imaging [21,23]. The Olympus BX51 microscope with the SC50 camera was manufactured by Olympus Corporation, Tokyo, Japan

Prior to electron microscopy, polished thin sections were carbon coated according to the method used by AlNajdi, et al. [21] to quantify pore shapes. Quantification was achieved using backscattered electron (BSE) and secondary electron (SE) microscopy on a Zeiss Gemini 450 SEM System at the Shared Research Facility (SRF) at the University of Liverpool.

Mineral proportions were determined and mineral maps were produced using secondary X-ray analysis in a dedicated scanning electron microscopy-energy dispersive spectroscopy SEM-EDS [14]; in this case, we used the University of Liverpool‘s FEI WellSite QEMSCAN used by [14,24].

3.5. Mineralogy via X-Ray Diffraction (XRD)

X-ray diffraction (XRD) analyses were undertaken to quantify the mineralogy of the core. Sample preparation was performed according to the standard methods applied at the University of Liverpool for mineral quantification and clay separation [14,21,25,26]. Mineral quantification was achieved using “HighScore Plus ^®” software version 4.9A using the relative intensity ratio (RIR) method [27], with reference data from the International Centre for Diffraction Data’s Powder Diffraction File-4+ Release 2022.

3.6. Laser Particle Size Analysis (LPSA)

The value of particle size determination during the analysis of top seals at carbon storage sites derives from its relation to pore volume and surface area [28]. Large particle sizes in top seals lead to high pore volumes, which potentially facilitate fluid flow and diffusion [28]. Linking laser particle size with N₂ adsorption (see Section 3.8) reveals the effect of particle size distribution on the gas adsorption process and provides unique insight into the relationship between surface area and pore volume measurements. Therefore, textural data from 14 core samples were acquired using a laser particle size analyser (LPSA) at the University of Liverpool; see AlNajdi, et al. [21] for details about sample preparation; the equipment used was a Beckman-Coulter LS13-320, manufactured by Danaher Corporation, California, USA. The data were processed using GRADISTAT version 9.1 to reveal particle size distributions [29,30].

3.7. Mercury Intrusion Capillary Pressure (MICP)

Mercury intrusion capillary pressure (MICP) analysis was used to characterise the pore system and pore size distribution for the Palaeocene Lista Shale [14]. Eight selected samples were analysed by MCA services (Cambridge, UK) [14]; see Peng, et al. [31] for details of the methodology and sample preparation. The relationship between pore throat diameter and applied pressure was defined by the Washburn equation [32].

$$\mathrm{Washburn}\mathrm{equation}:\mathrm{D}\mathrm{p}\mathrm{t}=\left(4\mathsf{\gamma }\mathrm{c}\mathrm{o}\mathrm{s}\right(\mathsf{\theta }\left)\right)/\mathrm{P}\mathrm{c}$$

(1)

where Dpt is the pore throat diameter, Pc is capillary entry pressure, $\gamma$ is the surface tension of the liquid, and $\theta$ is the contact angle in Radians between the specimen surface and mercury [14,32]. The contact angle varies as a function of fluid types and the specific mineral; however, the mercury-air contact angle ($\theta$) is 141° and the surface tension of the mercury (γ) is 0.48 N/m [14,31,33,34].

3.8. N₂ Gas Adsorption via BET Analysis

Rock chips from the core were collected from the same depths as those used for LPSA and MICP analyses; see AlNajdi, et al. [21] and Luffel and Guidry [35] for details of the methods employed.

Gas adsorption measurements were performed using a Nova4200e surface area analyser, configured according to the static volumetric method to quantify gas adsorption. Prior to analysis, each sample underwent de-gassing in line with the ISO9277 [36].

The traditional pore diameter model was initially based on the Kelvin Equation (2), which was subsequently refined by Barrett, et al. [37] to account for multilayer adsorption, thus making it widely applicable for the calculation of pore diameter within the mesopore range [37,38]. The Barrett, Joyner, and Halenda (BJH) model describes the phenomena of capillary condensation within pores [37]:

$$\mathrm{l}\mathrm{n}(\mathrm{P}/({\mathrm{P}}_{\mathrm{o}})=-(2\mathsf{\gamma }\mathrm{V}\mathrm{m})/\mathrm{R}\mathrm{T}({\mathrm{r}}_p-{\mathrm{t}}_c)$$

(2)

where ln(P/${\mathrm{P}}_{\mathrm{o}}$) is the relative pressure of the nitrogen gas, $\gamma$ is the surface tension of the bulk, $V_m$ is the molar liquid nitrogen volume, R is the gas constant, T is the temperature at which the isotherm is calculated (77K), $r_p$ is the pore radius, and $t_c$ is the thickness of the adsorbed multilayer film [38,39].

3.9. Fractal Dimension Analysis for Complex Pore Structure and Surface

In recent years, fractal theory has been applied to characterise the heterogeneity of pore structures in rocks and various other materials [40,41,42,43,44,45,46,47,48,49,50]. AlNajdi, et al. [21] discussed how to interpret the results of fractal calculations in terms of top seal properties; the same approach will be employed in this study of the Lista Shale. The fractal dimension index (D), derived from the N₂ adsorption isotherm, ranges between 2 (regular/smooth surfaces) and 3 (irregular/rough surface) [6,51]. The Frenkel-Halsey-Hill (FHH) equation was used to estimate the D value from N₂ adsorption data [52]. Equation (3) presents the FHH model [51],

$$\mathrm{l}\mathrm{n}\mathrm{V}=(\mathrm{D}-3)\mathrm{l}\mathrm{n}(\mathrm{l}\mathrm{n}{\mathrm{P}}_{\mathrm{o}}/\mathrm{P})+\mathrm{C}$$

(3)

where V is the N₂ volume adsorbed at equilibrium pressure (cm³), ${\mathrm{P}}_{\mathrm{o}}$ is the saturated vapour pressure, P is the equilibrium pressure (MPa), C is a constant, and D is the fractal dimension. D is calculated from the slope of the linear regression correlation between lnV and ln(−ln(${\mathrm{P}}_{\mathrm{o}}$/P)).

4. Results

In the following sections, the results from core logging, mineralogical assessment, petrographic assessment, gas adsorption, mercury intrusion, and particle size analyses are presented alongside the derived fractal dimensions. The results from each technique will be interpreted and compared in the discussion section.

4.1. Core Logging

Core samples from the Lista Shale Formation are predominantly mudstone, exhibiting fine lamination at scales of 1 to 5 mm, with localised turbidite sandstone beds that typically range from millimetres to centimetres scale (Figure 4). The cores are grey to dark grey and are locally bioturbated. The bioturbation index is approximately 3, dominated by trace fossils such as Planolites and Zoophycos (Figure 4 and Figure 5). Traces of slickensides are present in the core and are associated with localised fractures (Figure 5). Centremetre-sized siderite nodules are locally present in parts of the core (Figure 4 and Figure 5).

4.2. Facies Associations

Six facies were identified in core: unbioturbated laminated mudstone; bioturbated mudstone; unbioturbated laminated silty mudstone; bioturbated silty mudstone; thin-bedded, very fine to fine, unbioturbated sandstones; and thicker bedded (10 cm), structureless unbioturbated sandstones (Figure 5, Figure 6 and Figure 7).

Based on the facies classification for the Mey Sandstone Member of the Lista Formation published by Kilhams, et al. [15], three facies associations, linked to different environments in the turbidite fan system, are present in the Lista core (Figure 4 and Figure 5).

Facies association A (hemipelagic mudstone) is from the basin plain beyond the turbidite fan system, dominated by bioturbated mudstones (Figure 7A). Facies association B (mud-prone heterolithic) is from the outer fan and is dominated by mudstone and silty mudstone with local thin-bedded sandstones. Facies association B is characterised by bioturbation, with traces of Planolites and Zoophycos (Figure 6A–C). Facies association C (sand-prone heterolithic) is from the mid-fan and is dominated by bioturbated mudstone with localised 10 cm-thick sandstone beds.

4.3. Mineralogy and Petrography

Whole rock XRD analysis reveals that smectite is the dominant clay, comprising 20 to 47%, with minor amounts of mica and kaolinite, and trace amounts of chlorite (Figure 8 and

Table 2. Mineral analysis through X-ray diffraction (XRD) for well 16/21a-13.

Depth (ft)	Quartz	K-Feldspar	Plagioclase	Illite/Muscovite	Smectite	Chlorite	Kaolinite	Pyrite	Calcite
6975	18	8	6	30	31	3	4	0.5	0
6981	39	6	10	14	24	2	3	0	0
6982	28	9	5	24	27	3	4	0	0
6987	34	6	6	23	22	3	5	0	0.5
6989	33	7	7	21	22	4	5	1	0
6990	36	7	11	14	22	3	6	0	0
6992	22	7	6	23	36	3	3	0	0.5
6995.5	53	10	7	9	9	4	8	0.5	0
7001	27	6	9	16	35	2	5	0	0
7005.9	25	6	5	21	38	2	3	0.5	0
7008	52	5	9	9	20	2	2	0	0
7013	20	7	5	23	35	3	7	0	0
7015	30	7	6	11	36	3	6	0.5	0
7024.5	26	6	5	10	47	2	4	0.5	0

). The Lista samples are silt-rich, containing between 18% and 53% quartz. Feldspar silt grains make up 8 to 10% of the samples, with plagioclase (specifically albite) slightly more prevalent than K-feldspar. Trace amounts of pyrite and calcite were also identified (Figure 8; Table 2).

Conventional light microscopy revealed the presence of detrital silt-grade quartz and trace fossils (Figure 9). Light microscopy also showed that radiolaria microfossils are localised around fibrous aragonite cement (Figure 6(A1,A2)).

Electron microscopy proved to be valuable for characterising the Lista Shale samples. SEM-EDS images confirmed that the shale is primarily composed of smectite, illite and quartz, aligning with the XRD results (Table 2). High-resolution backscattered electron microscopy (BSEM) images identified detrital quartz, K-feldspar and micas (Figure 10A–D). Additionally, BSEM revealed the presence of pyrite framboids, which are common in the Lista Shale (Figure 10A,E), further validating the XRD findings (Table 2).

Illite is present (Figure 10F), confirming the XRD results (Table 2). The milky white sub-rounded clast in the BSEM image in Figure 6A is a fragment of chalk (Figure 6(A3)). BSEM images in Figure 6(A4–A6) show the composition of the chalk and the Fe-Mn-rich carbonate around the chalk. In Figure 7, core image A shows closed but sutured fractures filled with Fe-Mg-rich carbonate solution (Figure 7(A1–A3)). Core image B in Figure 7 is from mud-prone heterolithic facies association B and is laminated with coarse silt. However, the laminae are not only different in grain size but also in composition (Figure 7B). The rare coarse silt laminae are carbonate cemented, with cement localised around detrital quartz and K-feldspar (Figure 7(B1,B3)).

4.4. Pore Types

Based on the classification for CCS top seals proposed by AlNajdi and Worden [23], two distinct pore types were observed (Figure 11). Interparticle pores are common between clay minerals (smectite) and silt grains composed of mica, quartz and feldspar (Figure 10A). Other types of interparticle pores are slit pores between the stack of clay aggregates (Figure 10B,D), or plate pores between clay platelets (Figure 11C,E). The intraparticle pores are primarily located within pyrite framboids (Figure 11F).

A schematic illustration is shown in Figure 11, linking the pore types to the adsorption process of liquid nitrogen and their ability to adsorb and store gases; this will be further addressed in the discussion section.

4.5. Particle Size Analysis (LPSA)

Particle size analysis reveals a range of unimodal, bimodal and polymodal distributions, with unimodal being dominant. Samples tend to be poorly sorted (Figure 12A;

Table 3. Summary of results from nitrogen adsorption-desorption isotherms, MICP and LPSA, along with sorting and mean grain size (MGS) data. The surface area, pore volume and pore diameter measurements are obtained using the BET and BJH models for the Lista Shale from well 16/21a-13. Sorting and grain size data were analysed using GRADISTAT version 9.1, based on the Folk and Ward method [29]. ND: not determined for this sample.

Depth (ft)	BET Measurements				MICP Measurements					LPSA Measurements
	Multi BET Surface Area (m2/g)	BJH Surface Area (m2/g)	BJH Pore Volume (cc/g)	BJH Pore Diameter (nm)	Average Pore Diameter (nm)	MICP Porosity	Median Pore Diameter (MICP)	Average Pore Throat Diameter (nm)	Total Pore Area (m2/g)	Sorting Φ	MGN (µm)
6975	13.32	13.35	0.066	3.16	17.55	15.00	0.014	15.1	21.24	1.22	4.64
6981	11.41	17.21	0.071	3.16	18.68	15.71	0.013	14.07	23.99	1.19	4.22
6982	11.56	16.26	0.062	3.16	17.3	ND	ND	ND	ND	1.2	3.65
6987	7.06	10.93	0.048	3.14	18.49	ND	ND	ND	ND	1.33	4.12
6989	7.35	11.00	0.045	3.12	17.47	ND	ND	ND	ND	2.13	8.87
6990	4.11	7.34	0.031	3.40	17.84	14.15	0.018	23.53	12.56	2.51	19.46
6992	12.10	15.71	0.067	3.16	16.89	ND	ND	ND	ND	1.19	4.86
6995.5	4.56	7.66	0.029	3.16	17.68	ND	ND	ND	ND	2.17	23.63
7001	10.87	16.28	0.061	3.12	18.13	15.74	0.013	14.64	23.07	1.21	5.05
7005.9	11.05	16.32	0.062	3.16	18.23	16.33	0.014	14.74	24.12	1.37	4.47
7008	11.61	17.07	0.061	3.14	18.3	17.02	0.016	17.46	21.2	1.2	3.84
7013	11.22	16.17	0.058	3.12	18.19	ND	ND	ND	ND	1.31	5.06
7015	11.81	16.90	0.065	3.16	17.58	16.25	0.013	13.68	25.52	1.18	3.54
7024.5	9.87	15.00	0.064	3.14	17.53	15.45	0.012	13.78	23.96	1.16	3.76

). The particle size varies from very fine medium silt to medium silt. The distribution of grain sizes ranges from sandy silt to silt. The mean grain size, also determined using the Folk and Ward method in GRADISTAT, is 7.08 µm (Figure 12B), representing very fine silt-grade material given that the boundary between clay and silt is 4 µm. The mean sorting, calculated using the Folk and Ward method, is 1.45 φ, indicating very poorly sorted primary sediment [24].

4.6. Mercury Intrusion Porosimetry (MICP)

Mercury intrusion porosimetry (MICP) analysis was performed on eight samples (Figure 13B). The results, presented in terms of pore throat diameter and listed in Table 3, also include the derived mercury intrusion-derived shale porosity values.

The average pore throat diameter of the Lista Shale is 17 nm. The porosity values determined by MICP range from 14% to 17% (Table 3), with the average porosity being approximately 16% (Table 3).

4.7. Pore Structure

Adsorption involves holding the gas molecules on a thin layer on the solid surfaces, while desorption is the process whereby adsorbed gas molecules are released from the surface [36]. According to the International Union of Pure and Applied Chemistry (IUPAC), the nitrogen gas adsorption isotherms observed in Figure 14 are classified as type IV, typical of mesopore adsorbents [39,53]. As the relative nitrogen pressure increases, multilayer adsorption occurs, resulting in a characteristic “knee-bend” shape on the adsorption isotherm. In Figure 14, this knee-bend shape indicates the transition from monolayer to multilayer adsorption. Gas in the mesopores and macropores begins to condense at higher relative pressures. As the relative pressure decreases, desorption occurs, and the amount of adsorbed gas reduces. Following desorption, the desorption curve coincides with the adsorption curve on the declining pressure path, forming a hysteresis loop. The shape of the hysteresis loop is type H3, which suggests the presence of open-type pores (e.g., between plate-like particles) or wedge-shaped slit pores [54,55] as shown in Figure 11B,C.

Surface area and pore volume analyses, presented in Table 3, were calculated using the BET and BJH models. The bulk surface area ranges from 4.1 to 13.3 m²/g, while the pore volume derived from nitrogen adsorption ranges from 0.029 to 0.071 cm³/g (Table 3).

The pore body size diameters reported here are derived from the BJH model and based on the pore diameter classifications defined by IUPAC [53]. The partial volume of each pore diameter (dV/d(logD)) is plotted, and the area under the curve between two pore diameters can be used to assess the partial porosity for each pore size range [34]. The classification includes macropores, mesopores and micropores [35,56]. The results show that the dominant pore size falls within the mesopore range (Figure 13A), with an average overall pore body diameter of 17.84 nm (Figure 13A), which is similar to the pore throat diameter obtained from MICP analysis (Table 3).

4.8. Fractal Dimension Analysis Using Gas Adsorption Data

The adsorption isotherms were used to perform fractal dimension analysis using the Frenkel-Halsey-Hill (FHH) fractal model [6,21,51]. The standard adsorption and desorption isotherms typically merge in the relative pressure range of 0.40 to 0.45 (P/P_o) [6]. Consequently, the fractal dimension is derived from two distinct regions of the adsorption relative pressure (Figure 15) [6,21,51]. The first fractal region, referred to as D1 (

Table 4. A summary of the FHH fractal model-fitted lines derived from the adsorption isotherms of representative samples, chosen based on their relative surface areas. The fractal dimension D1 characterises the pore surfaces for relative pressures (P/P_o) less than 0.45, while D2 characterises the pore structures for relative pressures greater than 0.45. All depths are measured as opposed to true vertical depths.

Sample	Fractal Dimension of Pore Surfaces (Po/P < 0.45)			Fractal Dimension of Pore Structure (Po/P > 0.45)
	D1	Fitted Equation	R2 (1)	D2	Fitted Equation	R2 (2)
6975 ft	2.60	y = −0.2723x + 2.7876	0.999	2.72	y = −0.3924x + 2.7611	0.999
6981 ft	2.55	y = −0.3372x + 2.4631	0.998	2.66	y = −0.4469x + 2.4355	0.999
6982 ft	2.59	y = −0.3199x + 2.4856	0.999	2.68	y = −0.4052x + 2.4646	0.999
6987ft	2.48	y = −0.3777x + 1.8578	0.999	2.62	y = −0.5120x + 1.823	0.999
6989 ft	2.52	y = −0.3389x + 1.9949	0.998	2.66	y = −0.4703x + 1.9612	0.999
6990 ft	2.59	y = −0.3199x + 2.4856	0.999	2.68	y = −0.4052x + 2.4646	0.999
6992 ft	2.62	y = −0.3038x + 2.5429	0.996	2.69	y = −0.3701x + 2.5377	1.000
6995 ft	2.45	y = −0.3531x + 1.495	0.998	2.64	y = −0.5482x + 1.4054	0.999
7001 ft	2.57	y = −0.3095x + 2.4844	0.999	2.69	y = −0.4290x + 2.4449	0.999
7005 ft	2.56	y = −0.3221x + 2.4587	0.999	2.67	y = −0.4386x + 2.4218	0.999
7008 ft	2.57	y = −0.3120x + 2.5444	0.998	2.68	y = −0.4294x + 2.5148	0.999
7013 ft	2.60	y = −0.2955x + 2.5525	0.998	2.70	y = −0.3945x + 2.5204	0.999
7015 ft	2.57	y = −0.3285x + 2.4903	0.999	2.67	y = −0.4281x + 2.4619	0.999
7024 ft	2.43	y = −0.4500x + 1.9379	0.999	2.55	y = −0.5688x + 1.8917	0.999

), represents the pore surface and occurs at relative pressures below 0.45 [6,21,51]. The second fractal region, D2 (Table 4), represents the pore structure and is observed at relative pressures above 0.45 [6,21,51]. The fractal dimensions (D1 and D2) were calculated from the BET data, with R² (correlation coefficient) values for the fitted lines exceeding 0.9, confirming that the pores and surfaces possess fractal characteristics. The fractal dimension of the pore surface (D1) ranges from 2.4 to 2.6, indicating that some pore surfaces are smooth while others are irregular [21,51,56] (Table 4). The fractal dimension of the pore structure (D2) is around 2.7, suggesting rougher pore structures [21,51,56] (Table 4). Notably, the fractal dimension D2 is higher than D1 (Table 4).

5. Discussion

5.1. Mineralogy and Depositional Environment of the Lista Shale CCS Top Seal

The Lista Shale is finely laminated at the millimetre scale, suggesting fluctuating water flow velocities during its deposition (Figure 4). The presence of microfossils such as rare foraminifera (Figure 9), and radiolaria (Figure 6(A1)) and trace fossils such as Planolites (Figure 6A) and Zoophycos (Figure 6B,C) show that the Lista sediments were deposited in deep marine environmental settings [57,58]. Zoophycos is a trace fossil, feeding burrow with a worm-like shape commonly found in turbidite deep marine rock strata [59].

The core samples are dominated by smectite and quartz (Figure 8; Table 2). The quartz percentage varies throughout the core which relates to the shale interbedded with sandstone deposited as stacked lobes of middle and outer fan sediments (Figure 4). The variation in the thickness of sand input throughout is a characteristic of the Palaeocene turbidite system in the North Sea (Figure 4) [15,17,60]. Mudge and Bujak [60] indicated that the majority of sands associated with the Lista Formation are more widespread than the overlying Sele Formation, even though they are present in a similar area, suggesting a link to the original Mesozoic graben structure rather than the Cenozoic tectonic regime [15,61].

Three facies’ associations were identified: (A) hemipelagic mudstone; (B) mud-prone heterolithic; and (C) sand-prone heterolithic (Figure 4 and Figure 5). The term heterolithic particularly means alternating lithotype of layers, usually sand and mud [62]. The hemipelagic mudstone facies association is interpreted as basin floor deposits (Figure 4) without significant sand influx and the average grain size is clay (Figure 12) [15]. The term hemipelagic refers to a combination of gradual gravity sinking of very fine sediment, in this case with a mean grain size of 7 µm, and fine sediment delivered at the tail end of a flow event of material from a shelf towards an abyssal plain. Chalk debris flow occurs regularly as sub-rounded clasts within the mudstones (Figure 6A). The mud-prone heterolithic facies association is dominated by pelagic mudstone; the sandstone beds are usually thin (cm scale), and the average grain size is silt (Figure 12) [15]. The sand-prone heterolithic facies association is a mixture of sandstone and pelagic mud.

Mudstone and the grain size varies from clay to very fine sand (Figure 12) [15]. The sand-prone heterolithic facies are here interpreted as being marginal to the main sand channel in the turbidite fan system (Figure 4) [15].

In well 16/21a-13, core depths from 6974 ft to 6988 ft-md (measured depth) are part of a hemipelagic mudstone facies association and the core is dark grey, finely laminated, and contains trace fossils (Figure 4 and Figure 5) along with traces of siderite nodules identified in thin sections and SEM-EDS (Figure 8). At core depth 6987 ft-md, a complex sutured fracture was observed (Figure 7A) suggesting that this part of the core is a damage zone related to an ancient and natural fault [63]. In general, a fracture damage zone is defined as a zone surrounded by fractures, showing grain breakage, concomitant reduction of grain size, change in clay mineral orientation, and increase in crack density [63]. The main fractures in the core are branching and localised in the clay-rich layers and are developed around more competent detrital quartz grains and pyrite (Figure 7(A1)) [63]. The fractures were initially open, possibly due to early tectonic activity linked to rifting and extension [15]; the fractures were then diagenetically filled with Fe-Mg-rich carbonate pieces of cement (Table 2) (Figure 7(A1–A3)). There are traces of slickensides associated with the damage zone suggesting some degree of shear as well as extension. Fractures cannot be ignored when it comes to assessing the integrity of the caprock for CCS projects, as fractures can play a major role in being conduits to flow movements and permeability pathways. Although the fractures evident in Figure 5 and Figure 6A seem to be tightly cemented, they will have different geomechanical properties to the more clay-rich sections of the core and might be more susceptible to brittle failure under the new stress regime imposed when CO₂ is injected.

Core depths from 6991 to 6996 ft-md are part of the mud-prone heterolithic facies association, which is here attributed to outer fan deposits (Figure 5). At core depth 6995.5 ft-md, XRD data (Table 2) and BSE images show localised carbonate cement in the coarse silt beds, whereas the finer silt beds do not have calcite cement (Figure 7(B1–B3)). Calcite cementation of coarse silt is a common part of the diagenetic process in the Lista Formation; the origin of the calcite is unclear but it may be related to dissolved and reprecipitated calcite bioclasts or local redistribution favoured by the coarser grain of the cemented intervals [15]. The calcite in the coarse siltstone makes up 0.6 to 7.2% of the interval [15], which makes it potentially concerning when it comes to the integrity of the caprocks as part of CO₂-mineral interaction could lead to calcite dissolution [21].

5.2. The Genesis and Evolution of Pore Types in the Lista Shale CCS Caprock

A combination of primary sedimentary and secondary diagenetic processes leads to various pore types in mudstones [64]. Diagenetic processes during compaction and cementation influence the presence of both primary and secondary pores in these rocks. Clay-rich sediments like mudstones and shales are known to exhibit high porosity (e.g., up to 60 to 80%) at the time of deposition [2,23,65,66]. Currently, the porosity of the Lista Shale is approximately 16%, as indicated by MICP data (Table 3) and porosity data derived from density logs [14]. When compared to typical mudstone compaction curves, a porosity of 16% is common for rocks buried at around 6600 ft-true vertical depth.

Pores in the Lista Shale, after burial and diagenesis, include interparticle pores (Figure 11A–E), which are primary pores located between grains [23,64,67]. Interparticle pores between smectite platelets and between smectite and silt-grade materials are visible in SE images (Figure 11A,B,E). Intraparticle pores are observable in high-resolution images within pyrite framboids (Figure 11F). Primary interparticle porosity is the dominant pore type (Figure 11).

5.3. Pore Size Distribution in the Lista Shale CCS Caprock

Different aspects of pore size and structure are recorded by different analytical techniques. Different uses of the terms pore radius and pore diameter following mercury intrusion (MICP) and gas adsorption techniques in the literature tend to cause confusion [68,69]. For clarity, we will designate measurements from mercury intrusion as “pore throat diameter” and pore size measurements from N₂ adsorption as “pore body diameter” [21].

N₂ adsorption and MICP have different detection limits; MICP measurements can detect pore throats as small as 3.56 nm [68,70] (Figure 13B). At high mercury pressure, the rock fabric may be altered, potentially causing particle breakage and the opening of previously sealed pores [68,69]. The smallest pore body diameter detected by N₂ adsorption is 3.1 nm, while the largest detectable pore size using N₂ adsorption is 106.5 nm (Figure 13A). Despite the limitations of MICP and N₂ adsorption measurements, it has been suggested to combine both fluid invasion methods to characterise the full range of pore size distributions (Figure 16) [21,68,69,71].

From the overlapping porosity curves (Figure 16), there is a shared pore size range covered by the data from both MICP and BJH N₂ adsorption. This overlap enables the identification of the point of connection (POC) between the mercury intrusion and BJH N₂ adsorption curves, providing a complete representation of the pore size scale [21,38,69,70,72]. The overlap physically represents the filling mesopores (body and throats) leading to the formation of the N₂ monolayer.

The close correlation between the nitrogen adsorption-derived pore body diameter and mercury intrusion-derived pore throat diameter (Table 3) suggests that the pores predominantly resemble slits between grains, rather than narrow pores opening into larger pore bodies (Figure 11B). This is not surprising, given the H3 shape of the hysteresis loop which is characteristic of plate-like particles (Figure 14) [54] and considering that the Lista Shales are composed of plate-like sheet silicates such as smectite, kaolinite and mica (Figure 11D,E).

The average pore body diameter derived from gas adsorption is 17.84 nm (Figure 13A; Table 3). According to the IUPAC classification adopted by Rouquerol, et al. [71], the Lista Shale is predominantly composed of mesopores (2 to 50 nm) compared to micropores (<2 nm), with few pore bodies falling within the macropore classification (Figure 13A). Ding, et al. [55] concluded that micropores and mesopores provide the majority of surface area, facilitating gas adsorption on pore surfaces, while mesopores and macropores account for most of the pore volume, enabling gas storage as a discrete phase in shale [55]. Therefore, the dominance of mesopore bodies reflects the balance between free gas in macropore bodies and adsorbed gas in micropore bodies [21,55,73]. Schematic illustrations in Figure 11G,H show the occupation of free and adsorbed gas in slit pores and plate-like pores. Adsorption on pore surfaces is crucial for shale gas reservoir quality [74], although it may have a similarly significant impact on CO₂ adsorption on the pore surfaces of the Lista Shale.

The average pore throat diameter measured from MICP is about 17 nm (Figure 13B; Table 3). According to the classification scheme of pores proposed by [75] and adopted by Zhang, et al. [69] and AlNajdi and Worden [23], the Lista Shale pore throats (Figure 13B; Table 3) are classified as having a “transitional pore throat diameter” that ranges from 10 to 100 nm. Knowledge of the pore throat diameter, derived from MICP, is crucial for CCS projects, as it affects capillary entry pressure and the potential escape of CO₂ [23,76].

5.4. Compositional Influences on the Pore Structure, Surface Area and Fractal Dimensions of the Lista Shale

Correlation analysis with all the quantitative measurements has been undertaken, distinguishing negative and positive correlations (Figure 17). We have also derived p-value significance to identify meaningful correlations. Established statistical methods show that if the p-value is >0.05, the hypothesis (correlation) is insignificant, and if the p-value is ≤0.05, the hypothesis is significant [77]; increasingly low p-values are increasingly significant. In this research, adopting conventional statistical techniques, we have set the maximum p-value significance to equal 0.05.

There is a positive correlation between surface area (Table 3 and BJSA and MBSA in Figure 17) and pore volume (BJPV), which can be explained by the convoluted outlines of clusters of clay minerals leaving substantial aggregate-marginal, microporosity (that may be primary or secondary in origin) as well as elevated surface area (Figure 11). Xiong, et al. [78] reported a similar relationship.

The relationships between fractal dimensions and pore structure parameters are presented in Figure 17. The correlation matrix shows that fractal dimension D1 (pore surface) has a positive correlation with fractal dimension D2 (pore structure), which indicates that the Lista Shale has fractal attributes and that the fractal dimensions, D1 and D2, are not totally independent; a similar conclusion was reached by Chen, et al. [73]. As expected, there is a strong positive correlation between D1 and surface area (BJSA), reflecting the surface roughness (Figure 7B and Figure 17) [51,73].

Clay mineral proportions have been reported to influence mudstone pore volume, pore structure and surface area [78,79,80,81]. There is a positive relationship between the dominant clay mineral, smectite and surface area (Figure 17), possibly due to a greater contribution of interparticle pores in smectite-rich, silt-poor, samples (Figure 10). According to Saidian, et al. [82], smectite and illite clays usually contribute the highest surface area in clay-rich rocks due to their sheet-like morphology (Table 2).

Complex clay morphology leads to challenging diffusion and gas flow in the pore network [83]. The correlation between the pore structure and fractal dimension indicates that the Lista Shale has a low potential for CO₂ flow and diffusion due to the complex and heterogeneous pore structure.

5.5. The Impact of Reactive Surface Area on the Lista Shale Top Seal

The reactive surface area of minerals is a dynamic property that is expected to evolve over time as geochemical reactions develop; surface area varies from mineral to mineral [84,85]. The distribution of reactive surface area of minerals within a rock can significantly influence the progress of geochemical reactions during CO₂ injection [86], which in turn affects the integrity of the caprock.

Some minerals in shales, such as quartz or feldspars, can be compact (low surface area) silt- or even sand-grade grains. Some clay minerals, especially detrital smectite, can form irregular aggregates or pseudo-grain, which lead to high D1 fractal dimensions and commensurately high surface area (Figure 17). Other clay minerals, such as chlorite or kaolinite, can occur as detrital flakes or recrystallised or neomorphic version of detrital minerals; these tend to have relatively low surface area compared to smectite (Figure 17). Carbonate minerals [21], albeit not common in the Lista Shale, tend to occur as cement in pores and probably have fairly low surface area.

Aluminosilicate minerals, such as clay minerals, plus carbonate minerals and quartz (both detrital and biogenic), can act as pore-filling, or grain-coating agents, and in some cases, a cementing framework of grains [86,87,88,89].

Clay mineral- and feldspar-rich rocks have been reported to exhibit more complex relationship between CO₂/brine and porosity than carbonate rich rocks [86]. For example, Espinoza and Santamarina [90] described a decrease in total porosity after submerging kaolinite and montmorillonite in supercritical CO₂ (ScCO₂). In contrast, Smith, et al. [91] and Miller, et al. [92] reported an increase in caprock porosity upon exposure to CO₂/brine. Both Miller, et al. [92] and Smith, et al. [91] unexpectedly found that silicate minerals were more susceptible to reaction in the presence of CO₂ than carbonate minerals, possibly due to their relatively larger reactive surface areas. These reactions, in turn, resulted in the creation of new pores larger than 1 µm with notable increase in pore connectivity [86,91,92]. The inverse relationship between pore connectivity and porosity has been attributed to the roles of clay hydration or preferred mineral reprecipitation in pore throats following dissolution [86]. This is analogous to the phenomenon of redistributional secondary porosity in sandstones leading to reduced permeability [93]. XRD mineralogy data (Table 2) reveal chlorite and plagioclase, as wells as smectite, that may be reactive to CO₂ [94], exposing newly fresh pores which contribute to increased surface area changing the physical and chemical properties of the caprock.

6. Risk Assessment of the Lista Shale Top Seal

6.1. Seal Capacity and Its Relation to Depositional Environment

Worden, et al. [14] reported the CO₂ column height for the Rodby Shale top seal to the Captain Sandstone reservoir at the Acorn injection site, as well as the Lista Shale top seal based on MICP data. Gibson-Poole, et al. [95] also provided CO₂ column heights for localised intraformational seals within the Gippsland Basin in Australia using MICP data. According to Gibson-Poole, et al. [95], intraformational seals can span a range of depositional environments from deep marine to fluvial, with CO₂ column heights varying from 53 to 1191 m. Figure 18 combines the average CO₂ column heights published by Gibson-Poole, et al. [95] for different depositional environments, along with integrated Lista data from Worden, et al. [14]. Figure 18 demonstrates that the Lista Shale is classified as intermediate in terms of CO₂ column height that can be supported before capillary entry pressure is exceeded.

6.2. Lista Shale Seal Evaluation-Caprock Integrity and Risk of CO₂ Leaks

In this section, we examine the integrity of the Lista Shale and assess the leakage risks of CO₂ from the East Mey sandstone reservoir. The methods applied here have been adapted from those reported by Espinoza and Santamarina [96] and modified by AlNajdi, et al. [21] for evaluating the seals of target caprocks at various CCS sites, based on the petrophysical and mineralogical properties of the caprock, reservoir, and the anticipated CO₂ plume. The original approach by Espinoza and Santamarina [96] looked at bulk properties and so did not consider caprock depositional or diagenetic heterogeneity or structural discontinuities.

The escape of CO₂ from reservoir to the top seal plausibly might occur by either diffusion or advection of CO₂ [96,97]. Diffusion of CO₂ is a mechanism controlled by the difference in CO₂ concentration (or more accurately its fugacity) between the base and the top of a top seal [21]. A maximum permitted or acceptable leak rate was proposed by Espinoza and Santamarina [96] to be 3 kg/m³/yr. The rate of diffusion ($\mathrm{q}{\displaystyle \frac{diff}{{\mathrm{C}\mathrm{O}}_2}}$) can be modelled by Fick’s Law [98]:

$$\mathrm{q}{\displaystyle \frac{diff}{{\mathrm{C}\mathrm{O}}_2}}=-\mathrm{n}·\mathrm{D}({∆\mathrm{x}}_{\mathrm{C}\mathrm{O}2})/{\mathrm{t}}_{\mathrm{h}}$$

(4)

where n represents fractional porosity (in the Lista, n = 0.16), $D$ is the CO₂ diffusion coefficient (here assumed to be $D$ = 10⁻⁹ m²/s [99]), $x_{\mathrm{C}\mathrm{O}2}$ denotes difference in CO₂ concentration driving diffusion, with the maximum assumed value being ~0.044 kg/m³, and $t_h$ is the thickness of the caprock, which is measured as 15 m for single caprock unit and 100 m for the whole core section based on published wireline logs from the Lista Shale [14]. Given these assumptions and the data for the Lista Shale, the modelled CO₂ diffusion rate is about 1.5 × 10⁻⁵ kg/m²/yr, which is considerably lower than the reference leak rate. Thus, diffusive loss of CO₂ seems to be unlikely to pose a concern for this top seal.

Advection of CO₂ through top seals takes place only if the capillary entry pressure of the caprock is surpassed [21,96]. The rate of CO₂ via advection ($\mathrm{q}{\displaystyle \frac{\mathrm{a}\mathrm{d}\mathrm{v}}{\mathrm{C}{\mathrm{O}}_2}}$) is affected by factors such as the pressure difference between the reservoir and the top of the top seal, the relative and absolute permeability, the density of CO₂, top seal thickness, and the viscosity of CO₂, as determined by Darcy’s Law [100]:

$$\mathrm{q}{\displaystyle \frac{\mathrm{a}\mathrm{d}\mathrm{v}}{\mathrm{C}{\mathrm{O}}_2}}=(\mathrm{k}{\mathrm{k}}_{\mathrm{C}\mathrm{O}2}^{\mathrm{r}}·{\mathsf{\rho }}_{\mathrm{C}\mathrm{O}2})/{\mathsf{\mu }}_{\mathrm{C}\mathrm{O}2}·((∆\mathrm{P}-{\mathsf{\rho }}_{\mathrm{C}\mathrm{O}2}·\mathrm{g}·{\mathrm{t}}_{\mathrm{h}}))/{\mathrm{t}}_{\mathrm{h}}$$

(5)

where $k$ represents the absolute vertical permeability (in the Lista, this is 2.59 × 10⁻²⁰ m², derived from MICP data and an assumption that k_v/k_h = 0.01). ${\rho }_{\mathrm{C}\mathrm{O}2}$ denotes the mass density of CO₂ (assumed here to be 650 kg/m³ [101]). ${\mu }_{\mathrm{C}\mathrm{O}2}$ is the viscosity of CO₂ (assumed here to be 0.05 mPa·s). $t_h$ is the thickness of the caprock, which is assumed to be 15 m [14]. ($\Delta P-{\rho }_{\mathrm{C}\mathrm{O}2}·g·t_h)$ represent the total pressure head, assumed to be up to 10 MPa. The relative permeability of CO₂, $k_{\mathrm{C}\mathrm{O}2}^r$ is assumed to be 0.64, 0.16 and 0.01, corresponding to near wellbore conditions (CO₂-saturation [$S_{\mathrm{C}\mathrm{O}2}^r$] of 0.8), mid plume (CO₂-saturation of 0.4), and far-field (CO₂-saturation of 0.1) assuming that $\left(k_{\mathrm{C}\mathrm{O}2}^r\right)$ is equal to $(S_{\mathrm{C}\mathrm{O}2}^r{)}^2$ [96].

The modelled rate of advective flux of CO₂ is illustrated in Figure 19, in the event of capillary pressure of CO₂ exceeding the Lista Shale. The rate of advective CO₂ flux varies due to different CO₂ saturations and corresponding relative permeability scaling factors, which are influenced by distance from the injection well [21]. Near the injection well, with a 10 MPa CO₂ pressure difference, the post-breakthrough CO₂ flux could reach up to 4.2 kg/m²/yr which exceeds the reference leak rate. At several hundred meters from the injection well, with a 5 MPa pressure difference, the flux may be as high as 0.55 kg/m²/yr. Further away at many hundreds of meters from the injection well, with 1 MPa pressure difference, the flux rate could be up to 0.01 kg/m²/yr. The highest risk is clearly in the wellbore region at 10 MPa pressure and the lowest which are below the reference leak rate are at 5 and 1 MPa. The greatest is clearly associated with high-pressure CO₂ injection in near-wellbore locations, but only if capillary entry pressure is exceeded.

6.3. Seal Capacity

The effectiveness of geological formations for potential CO₂ storage can be assessed by using dimensionless sealing and stability numbers [96]. The sealing number represents the ratio of the capillary breakthrough pressure, Pᴄ, to the buoyant pressure, ∆P, which is determined by the thickness of the continuous CO₂ plume, h [96],

$$\mathrm{Sealing}\mathrm{number}={\mathsf{\pi }}_1=\left({\mathrm{P}}_{\mathrm{c}}^{\mathrm{*}}\right)/∆\mathrm{P}=\left(\mathsf{\psi }{\mathrm{T}}_{\mathrm{s}}\mathrm{c}\mathrm{o}\mathrm{s}\mathsf{\theta }{\mathrm{S}}_{\mathrm{s}}{\mathsf{\rho }}_{\mathrm{m}}\right)/\left(2\mathrm{e}\mathrm{h}\mathrm{g}\right({\mathsf{\rho }}_{\mathrm{w}}-{\mathsf{\rho }}_{\mathrm{C}\mathrm{O}2}\left)\right)$$

(6)

where ψ is the pore shape factor that ranges from 2 to 4; in the Lista case, it is assumed tobe 3 for slit pores. T_s denotes the interfacial tension between water and CO₂ (assumed to be 0.03 N/m after Bachu and Brant [102]). θ is the contact angle formed by the CO₂-water interface on the mineral surface (assumed to be 0.9185 Radians). Ss is the surface area, (assumed average surface area of the Lista is 9854.5 m²/kg based on BET data, Table 3). The mineral bulk density ${\rho }_m$ is assumed to be 2300 kg/m³. e is the void ratio at breakthrough, here assumed to be 0.6. h is the thickness of the continuous CO₂ plume (assumed to be 15 m based on wireline log data). $g$ is acceleration of gravity. ${\rho }_w$ is the mass density of water assumed to be 1050 kg/m³ based on regional formation water data in Warren and Smalley [103]. ${\rho }_{\mathrm{C}\mathrm{O}2}$ is the density of CO₂ assumed to be 650 kg/m³ [101].

A sensitivity test was conducted on the range of inputs to the calculated sealing number of Lista Shale, the dominant parameters as reported by AlNajdi, et al. [21] are shape factor, surface area and the thickness of CO₂ plume. The average sealing number of the Lista Shale is 39 (Figure 20). In comparison to other CCS case studies shown in Figure 20, the Lista Shale is similar to the Rodby Shale [21] and seems to offer a more effective seal than for example the Utsira Formation at Sleipner [104] and the regional seal at Rousse CCS pilot located in France [105]. The Lista Shale is slightly less effective than the In Salah (Krechba) seal [106].

6.4. Mechanical Stability of the Top Seal

Assessment of the relationship between vertical effective stress, ${\sigma }_{ZO}^{\prime }$, at the reservoir depth, $\mathrm{z}$, and the buoyant pressure, ∆P, initiated by CO₂ plume of thickness $\mathrm{h}$ can be measured via stability number (Equation (7)) [96]. Variation in pore pressure can lead to changes in vertical effective stress, which can later trigger fault reactivation and generate new fractures [96].

$$\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y} \mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}={\mathsf{\pi }}^2={\displaystyle \frac{{\mathsf{\sigma }}_{zo}^{\prime }}{∆\mathrm{P}}}={\displaystyle \frac{\left({\mathrm{Z}}_w\mathrm{g}{\mathsf{\rho }}_{\mathrm{w}}+\mathrm{Z}\mathrm{g}{\mathsf{\rho }}_{\mathrm{b}\mathrm{u}\mathrm{l}\mathrm{k}}\right)-{\mathrm{P}}_o}{\mathrm{h}\mathrm{g}\left({\mathsf{\rho }}_w-{\mathsf{\rho }}_{\mathrm{C}\mathrm{O}2}\right)}}$$

(7)

where $Z_w$ is the height of the water column above the seafloor (equal to 100 m, based on well reports). ${\rho }_{bulk}$ is the bulk mass density (assumed to be 2300 kg/m³). The sediment column height, $Z$, is assumed to be 1900 m. $P_o$ is the initial fluid pressure at the reservoir-seal interface (assumed to be hydrostatic pressure at 20.6 MPa).

A sensitivity test was applied to the range of input parameters for the calculation of the stability number (Figure 20). The stability number is primarily controlled by three parameters: the bulk density of the sediment column, the height of the sediment column, and, most significantly, the maximum thickness of the CO₂ plume [21]. The average stability number of the Lista Shale is 361. When compared to other CCS field cases, the Lista Shale is similar to the Rodby Shale [21], In Salah (Krechba) [106] and Frio [107] in terms of stability and better than the Utsira Formation at Sleipner [104].

Drawing together the work published here (Figure 1, Figure 2, Figure 3, Figure 4, Figure 5, Figure 6, Figure 7, Figure 8, Figure 9, Figure 10, Figure 11, Figure 12, Figure 13, Figure 14, Figure 15, Figure 16, Figure 17, Figure 18 and Figure 19) and other examples referred to in Figure 20A,B, we have produced a summary table that lists attributes of clastic top seal (Table 5). We have here specifically listed helpful and detrimental characteristics of each attribute. This table could serve as a checklist that could be used during the appraisal of any potential top seal to a CO₂ storage site.

7. Conclusions

• The Palaeocene Lista Shale is smectite-rich and quartz silt-bearing. It was deposited on top of the Mey Sandstone as part of a linked turbidite fan-abyssal plain system in a deep marine environment.
• The pores have been affected by both depositional and diagenetic processes; pore types include interparticle pores such as slit-shaped and plate-like pores, as well as intraparticle pores found within pyrite framboids.
• The pore bodies are primarily mesopores, with small proportions of micropores and macropores also present. The prevalence of mesopores aligns with the observed pattern in which free gas predominantly occupies macropore bodies, while adsorbed gas is mainly found within micropore bodies.
• The mean pore throat diameter measured by mercury intrusion is 17.00 nm, with a minimum of 3.56 nm and a maximum of 10,000 nm. For pore body diameters, BJH N₂ adsorption detected a minimum of 3.10 nm and a maximum of 106.5 nm. By combining two fluid invasion techniques, mercury intrusion and gas adsorption, the entire spectrum of pore body and pore throat sizes can be accurately represented. The connection at the inflection point, where mesopores begin to fill, marks the junction at which mercury intrusion pore throat measurements align with gas adsorption measurements.
• The close similarity between the N₂ adsorption pore body diameter (average 17.84 nm) and the mercury intrusion pore throat diameter (average 17.0 nm) indicates that the pores in the Lista Shale are likely to be slit-shaped.
• The complex heterogeneity of Lista Shale pore structure and mineral composition has strongly influenced the surface area. Greater quantities of smectite lead to higher surface area and pore volume, possibly due to a greater contribution of interparticle pores in the platy clay minerals that dominate in smectite-rich, silt-poor samples.
• If smectite or other aluminosilicate minerals such as chlorite or plagioclase reacted with CO₂, newly exposed minerals would potentially lead to increased reactive surface area, possibly feeding back to increased reaction rates.
• According to assessments of the diffusive flux, advection rate if capillary entry pressure is exceeded, and its overall stability and sealing characteristics, the Lista Shale shows strong potential to serve as an effective barrier against CO₂ leakage.

Table 5. Summary list of attributes for clastic top seal with listed beneficial and detrimental characteristics of each attribute for a potential CCS top seal.

Attribute	Helpful for a Good Top Seal	Detrimental to Being a Good Top Seal	Example	References
Pore body size	Preferably in the micro-pore range	When in meso- to macro-pore range	Lista Shale, Rodby Shale, Sichuan Basin	AlNajdi et al. [21], Chen et al. [51]
Pore throat size	Preferably in the micro-pore range	When in meso- to macro-pore range	Lista Shale, Rodby Shale, Mercia Mudstone	AlNajdi et al. [21], Armitage et al. [26]
Pore type	Slit pores preferred	Round pores	Lista Shale, Rodby Shale, Sichuan Basin	AlNajdi et al. [21], Chen et al. [51]
Pore volume	Low	High	Lista Shale, Rodby Shale, Sichuan Basin	AlNajdi et al. [21], Chen et al. [51]
Specific surface area	Low (to minimise reactivity)	High	Lista Shale, Rodby Shale, Sichuan Basin	AlNajdi et al. [21], Chen et al. [51]
Particle size	Low	High	Lista Shale, Rodby Shale	AlNajdi et al. [21], Pandey et al. [28]
Sorting	Poorly sorted sediment preferred	Well sorted sediment preferred	Lista Shale, Rodby Shale	AlNajdi et al. [21]
Mineralogy—clay	Inert clay minerals (e.g., illite, kaolinite)	Reactive clay minerals (e.g., chlorite, some smectites)	Lista Shale, Rodby Shale, Mercia Mudstone	Worden et al. [14], AlNajdi et al. [21], Armitage et al. [26]
Mineralogy—carbonate	Zero carbonate preferred	Calcite-rich (most reactive carbonate)	Lista Shale, Rodby Shale	Worden et al. [14], AlNajdi et al. [21]
Mineralogy—silt minerals	Quartz preferred (least reactive)	Reactive feldspar (Ca-rich are worst of all)	Lista Shale, Rodby Shale, Mercia Mudstone	Worden et al. [14], AlNajdi et al. [21], AlNajdi and Worden [23], Armitage et al. [26]
Mineralogy—silt proportions	Low (minimum brittleness)	High (increasing brittleness)	Lista Shale, Rodby Shale, Mercia Mudstone	Worden et al. [14], AlNajdi et al. [21], AlNajdi and Worden [23], Armitage et al. [26]
Mineralogy—Ca-sulphates	Low (minimum risk of volume change)	High (elevated risk of fractures and veins)	Mercia Mudstone	Worden et al. [14], AlNajdi and Worden [23], Armitage et al. [26]
Brittleness	Low	High	Lista Shale, Rodby Shale	Worden et al. [14], Ingram and Urai [110]
Unconfined compressive strength	Low	High	Lista Shale, Rodby Shale	Worden et al. [14], Ingram and Urai [110]
Thickness of top seal	Thicker seals better to minimise transport out of the storage site and risk of fault leakage	Thin seals potentially poor as they would lead to risk of leakage	Lista Shale, Rodby Shale, Sleipner, Rousse, In Salah, Ketzin, Frio, SACROC	AlNajdi et al. [21], Espinoza and Santamarina [96]
Lateral extent of top seal	Wide lateral extent preferred to minimise spillage and upward CO2 migration	Limited lateral extent would lead to risk of upwards leakage	Otway Basin	Kaldi et al. [4]
Fluid pressure relative to hydrostatic	Low fluid pressures preferred to minimise risk of capillary entry pressure being exceeded	High fluid pressures may lead to capillary entry pressure being exceeded	Lista Shale, Rodby Shale	Kaldi et al. [4], AlNajdi and Worden [23], Espinoza and Santamarina [96], Ingram and Urai [110]
Effective stress history (maximum palaeo-depth)	Sufficient to compact the mudstone, but low enough to prevent elevated brittleness	Negligible compaction is not helpful; excess compaction may not be helpful either	Lista Shale, Rodby Shale, Rousse, Mercia Mudstone, In Salah, Nordland Shale	AlNajdi and Worden [23], Ingram and Urai [110]
Thermal history (maximum palaeo-temperature)	Maximum palaeotemperature < 80 °C preferred to minimise mineral diagenesis and increasing brittleness	Maximum palaeotemperature > 120 °C as chemical compaction and mineral diagenesis lead to elevated brittleness	Lista Shale, Rodby Shale, Rousse, Mercia Mudstone, In Salah, Nordland Shale	AlNajdi and Worden [23]
Geological age of top seal	Younger rocks (e.g., Cenozoic) preferred to minimise diagenesis, but not too young to prevent sufficient compactional porosity-loss	Old rocks (e.g., Palaeozoic) have greatest chance of becoming brittle and strong (but will have lowest porosity and permeability)	Lista Shale, Rodby Shale, Rousse, Mercia Mudstone, In Salah, Nordland Shale	AlNajdi and Worden [23]

Table 5. Summary list of attributes for clastic top seal with listed beneficial and detrimental characteristics of each attribute for a potential CCS top seal.

Attribute	Helpful for a Good Top Seal	Detrimental to Being a Good Top Seal	Example	References
Pore body size	Preferably in the micro-pore range	When in meso- to macro-pore range	Lista Shale, Rodby Shale, Sichuan Basin	AlNajdi et al. [21], Chen et al. [51]
Pore throat size	Preferably in the micro-pore range	When in meso- to macro-pore range	Lista Shale, Rodby Shale, Mercia Mudstone	AlNajdi et al. [21], Armitage et al. [26]
Pore type	Slit pores preferred	Round pores	Lista Shale, Rodby Shale, Sichuan Basin	AlNajdi et al. [21], Chen et al. [51]
Pore volume	Low	High	Lista Shale, Rodby Shale, Sichuan Basin	AlNajdi et al. [21], Chen et al. [51]
Specific surface area	Low (to minimise reactivity)	High	Lista Shale, Rodby Shale, Sichuan Basin	AlNajdi et al. [21], Chen et al. [51]
Particle size	Low	High	Lista Shale, Rodby Shale	AlNajdi et al. [21], Pandey et al. [28]
Sorting	Poorly sorted sediment preferred	Well sorted sediment preferred	Lista Shale, Rodby Shale	AlNajdi et al. [21]
Mineralogy—clay	Inert clay minerals (e.g., illite, kaolinite)	Reactive clay minerals (e.g., chlorite, some smectites)	Lista Shale, Rodby Shale, Mercia Mudstone	Worden et al. [14], AlNajdi et al. [21], Armitage et al. [26]
Mineralogy—carbonate	Zero carbonate preferred	Calcite-rich (most reactive carbonate)	Lista Shale, Rodby Shale	Worden et al. [14], AlNajdi et al. [21]
Mineralogy—silt minerals	Quartz preferred (least reactive)	Reactive feldspar (Ca-rich are worst of all)	Lista Shale, Rodby Shale, Mercia Mudstone	Worden et al. [14], AlNajdi et al. [21], AlNajdi and Worden [23], Armitage et al. [26]
Mineralogy—silt proportions	Low (minimum brittleness)	High (increasing brittleness)	Lista Shale, Rodby Shale, Mercia Mudstone	Worden et al. [14], AlNajdi et al. [21], AlNajdi and Worden [23], Armitage et al. [26]
Mineralogy—Ca-sulphates	Low (minimum risk of volume change)	High (elevated risk of fractures and veins)	Mercia Mudstone	Worden et al. [14], AlNajdi and Worden [23], Armitage et al. [26]
Brittleness	Low	High	Lista Shale, Rodby Shale	Worden et al. [14], Ingram and Urai [110]
Unconfined compressive strength	Low	High	Lista Shale, Rodby Shale	Worden et al. [14], Ingram and Urai [110]
Thickness of top seal	Thicker seals better to minimise transport out of the storage site and risk of fault leakage	Thin seals potentially poor as they would lead to risk of leakage	Lista Shale, Rodby Shale, Sleipner, Rousse, In Salah, Ketzin, Frio, SACROC	AlNajdi et al. [21], Espinoza and Santamarina [96]
Lateral extent of top seal	Wide lateral extent preferred to minimise spillage and upward CO2 migration	Limited lateral extent would lead to risk of upwards leakage	Otway Basin	Kaldi et al. [4]
Fluid pressure relative to hydrostatic	Low fluid pressures preferred to minimise risk of capillary entry pressure being exceeded	High fluid pressures may lead to capillary entry pressure being exceeded	Lista Shale, Rodby Shale	Kaldi et al. [4], AlNajdi and Worden [23], Espinoza and Santamarina [96], Ingram and Urai [110]
Effective stress history (maximum palaeo-depth)	Sufficient to compact the mudstone, but low enough to prevent elevated brittleness	Negligible compaction is not helpful; excess compaction may not be helpful either	Lista Shale, Rodby Shale, Rousse, Mercia Mudstone, In Salah, Nordland Shale	AlNajdi and Worden [23], Ingram and Urai [110]
Thermal history (maximum palaeo-temperature)	Maximum palaeotemperature < 80 °C preferred to minimise mineral diagenesis and increasing brittleness	Maximum palaeotemperature > 120 °C as chemical compaction and mineral diagenesis lead to elevated brittleness	Lista Shale, Rodby Shale, Rousse, Mercia Mudstone, In Salah, Nordland Shale	AlNajdi and Worden [23]
Geological age of top seal	Younger rocks (e.g., Cenozoic) preferred to minimise diagenesis, but not too young to prevent sufficient compactional porosity-loss	Old rocks (e.g., Palaeozoic) have greatest chance of becoming brittle and strong (but will have lowest porosity and permeability)	Lista Shale, Rodby Shale, Rousse, Mercia Mudstone, In Salah, Nordland Shale	AlNajdi and Worden [23]