See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/361777700

Pore-scale imaging of hydrogen displacement and trapping in porous media

Preprint · July 2022

DOI: 10.31223/X54346

CITATIONS READS

2 622

9 authors, including:

Eike Marie Thaysen Ian Butler

National Spanish Research Council The University of Edinburgh
25 PUBLICATIONS   497 CITATIONS    137 PUBLICATIONS   3,853 CITATIONS   

SEE PROFILE SEE PROFILE

Aliakbar Hassanpouryouzband Damien Freitas

The University of Edinburgh The University of Edinburgh
57 PUBLICATIONS   1,724 CITATIONS    8 PUBLICATIONS   84 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

MiReCOL View project

PANACEA View project

All content following this page was uploaded by Eike Marie Thaysen on 05 July 2022.

The user has requested enhancement of the downloaded file.

 1 This manuscript is submitted for publication in Applied Energy. Please note that, this

2 manuscript has not undergone peer-review and is not yet formally accepted for publication.

3 Subsequent versions of this manuscript may have slightly different content. If accepted, the

4 final version of this manuscript will be available via the ‘Peer-reviewed Publication DOI’

5 link on the right-hand side of this webpage.

7 Pore-scale imaging of hydrogen

8 displacement and trapping in porous media
9 Eike M. Thaysen1*, Ian B. Butler1, Aliakbar Hassanpouryouzband1, Damien Freitas1,

10 Fernando Alvarez-Borges2, Samuel C. Krevor3, Niklas Heinemann1, Robert Atwood2,

11 Katriona Edlmann1

1
12 School of Geoscience, Grant Institute, The King's Buildings, The University of Edinburgh,

13 James Hutton Road, Edinburgh, EH9 3FE, United Kingdom

2
14 Diamond Light Source Ltd, Harwell Campus, Didcot OX11 0DE, United Kingdom

3
15 Department of Earth Science and Engineering, Imperial College London, London SW7
16 2AZ, United Kingdom
17

*
= corresponding author email, phone number: [email protected], +34 682079236

1
18 Abstract

19 Hydrogen can act as an energy store to balance supply and demand in the renewable energy

20 sector. Hydrogen storage in subsurface porous media could deliver high storage capacities but

21 the volume of recoverable hydrogen is unknown. We imaged the displacement and capillary

22 trapping of hydrogen by brine in a Clashach sandstone cylinder at 2-7 MPa pore fluid pressure

23 using X-ray computed microtomography. Hydrogen saturation obtained during drainage at

24 capillary numbers of <10-7 was ~50% of the pore volume and independent of the injection

25 pressure. Fluid configuration in the pore space was indicative of a water wetting system at all

26 conditions. Capillary trapping of hydrogen via snap-off during secondary imbibition at a

27 capillary number of 2.4x10-6 systematically increased with pressure, 20%, 24% and 43% of the

28 initial hydrogen trapped at 2, 5 and 7 MPa, respectively. Injection of brine at increasing

29 capillary numbers up to 9.4x10-6 reduced capillary trapping. Based on these results, we

30 recommend more shallow, lower pressure sites for future hydrogen storage operations in

31 porous media.

32

33 Keywords: geological hydrogen storage, µCT, porous media, recovery, drainage, imbibition

34

35 Highlights:

36  Hydrogen injectivity and recovery in rock is imaged with x-ray computed micro-CT

37  Hydrogen recovery decreases with increasing reservoir depth

38  Hydrogen recovery increases with increasing brine flow rate

39  Hydrogen trapping occurs via snap-off processes

40  Nitrogen is a poor proxy for hydrogen

41

2
42 1. Introduction

43 The storage of surplus electrical energy could increase the utility of renewable energy sources

44 thereby reducing the environmental impact of energy generation1, 2. Storage mediums like

45 batteries, compressed air, capacitors, or flywheels are only applicable for the short-term and

46 small-scale electricity storage3. The use of hydrogen (H2) as an energy vector however could

47 provide long-term storage to balance the intermittent demand and supply challenges affecting

48 renewable methods3. Considering the low energy density of H24, storage of H2 in large storage

49 sites is the only solution for economical, gigawatt to terawatt-scale H2 storage. Underground

50 storage of H2 in salt caverns is an established technology5, yet, only subsurface porous media

51 (saline aquifers and depleted oil and gas fields), can provide TWh storage capacities that could

52 balance seasonal demands4. Relative to cavern storage, inter-seasonal storage in porous media

53 is poorly understood and introduces a range of other complexities2, including potential pore-

54 clogging by microbial growth6, gas loss out of diffuse lateral boundaries, viscous fingering of

55 H2 into reservoir brines, residual trapping and possible reactions with the primary formation

56 fluid and reservoir minerals7-10. Each of these requires consideration of multiphase flow,

57 transport and reaction2, 8.

58 Recent years have seen several studies on H2 geochemistry11-13 and the elucidation of many

59 important flow characteristics of H2 including; Hydrogen relative permeability14, 15, interfacial

60 tension (IFT)16, 17, and H2/brine/rock contact angles14, 16-24. Yet, additional data describing the

61 H2 fluid flow in different porous formations and under varying conditions are vital to make

62 accurate predictions of the H2 plume development and to define optimum production

63 strategies8. Of particular relevance is the degree of capillary trapping, which leads to a reduced

64 volume of recoverable H2, in this way impacting the economic feasibility of the operation8. The

65 capillary forces that control capillary trapping also control the relative permeability8. As such

3
66 studies on residual trapping give insight to a range of crucial input parameters for pore-scale

67 and reservoir H2 storage models.

68 The effects of the brine pore fluid pressures and of brine flow rate on capillary trapping of H2

69 are hithertho unknown. The brine pore fluid pressure is linked to the reservoir depth via the

70 hydrostatic gradient, hence an investigation of the latter could facilitate the definition of an

71 optimal storage depth for gas recovery. The flowrate affects the capillary number, Nc, via

72 equation 1:
𝑞⋅µ
73 𝑁𝐶 = (1)
𝜎

74 where q is the flow, µ is the viscosity and σ is the interfacial tension (IFT). As Nc increases,

75 viscous forces dominate capillary forces and capillary trapping of the nonwetting phase

76 decreases25. This capillary desaturation with increasing Nc, depends on the rock type26 and on

77 the kind of nonwetting phase25 and is observed around critical capillary numbers of 10–5 to 10–
8 25
78 .

79 A non-destructive standard technique for visualization and analysis of wetting and non-wetting

80 phase displacement processes in porous media is x-ray micro-computed tomography (µCT)27.

81 Pore-scale models are directly reliant on µCT images to build up the basic models for

82 appropriate understanding of gas storage operations28. Previously published µCT studies on

83 two-phase flow in porous media involving gas have largely focused on supercritical carbon

84 dioxide (sCO2) in sandstones and limestones, e.g.27, 29-34. Reported saturations of sCO2 in non-

85 aged (i.e. not wettability altered by exposure to oil or organic acids) water-wet rocks range

86 from 30-60% during drainage and 15-30% during imbibition, at NCa between 10-8 to 10-6 and

87 10-8 to 2x10-5, respectively27, 30, 32-34. Studies using subcritical, gaseous carbon dioxide (CO2),

88 which could be more readily compared to H2, are scarce. Yet, the characteristic trapping curves

89 for sCO2 and CO2, which show the residual saturation (Snnwr) as a function of initial saturation

90 (Snwi), are not significantly different35. Saturations of nitrogen (N2) during drainage and

4
 91 imbibition in sandstones with 6-22% porosity is 43-64% and 43%, respectively, at NC between

92 1-7x10-8 36, 37
. Carbon dioxide trapping increases at lower pressures due to decreasing water

93 contact angles38. Unlike in CO2 storage, trapping is not desirable in geological H2 storage as it

94 leads to unrecoverable H2.

95 Some observational studies have suggested that H2 is less wetting on rocks than CO219, 21. At

96 the time of writing two µCT studies, one micro-model study and one nuclear magnetic

97 resonance (NMR) study on H2 exist: Al-Yaseri et al. (2022)39 used NMR to find H2 Snwi and

98 Snwr of 4% and <2%, respectively, in a Fontainebleau sandstone at 0.4 MPa and ambient

99 temperature. Higgs et al. (2021)16 µCT-imaged H2 injection into a 5 mm diameter and 7.6 mm

100 length core of Bentheimer sandstone at pressures of 6.8 to 20.8 MPa and documented

101 decreasing IFT between H2 and water with increasing pressure (72.5 mN/m at 6.9 MPa to 69.4

102 mN/m at 20.7 MPa). Jha et al. (2021)40 conducted a single cycle H2-brine displacement

103 sequence in a 5 mm diameter and 15 mm length Gosford sandstone, and used µCT to calculate

104 an Snwi of 65% during drainage and a Snnwr of 41% after brine imbibition. Rock samples in Jha

105 et al. (2021)40 and Higgs et al. (2021)16 were sufficiently short for capillary end effects to

106 dominate the flow behaviour, where the wetting phase accumulates close to the production face

107 of the core41. The experiment by Jha et al. (2021)40 was performed at ambient pressure and

108 leaves open the question of variations in H2 wetting and flow behaviour at reservoir

109 conditions19. Increasing pore fluid pressures have been reported to increase the gas saturation

110 and pore network connectivity for nitrogen (N2) in a brine wet, aged Bashijiqike tight sandstone

111 (0.6% gas saturation at 2 MPa versus 43% at 8 MPa)36. The effect of NC increases from 7.7x10-
7
112 to 3.8x10-4 on the H2 flow through a sandstone micro-model at 0.5 MPa was investigated by

113 Lysyy et al. (2022)20 who found that H2 saturation increased correspondingly from 18% to

114 79%20.

5
115 In this work, we used µCT to investigate the displacement and capillary trapping of H2 by brine

116 in 4.7 mm diameter and 54-57 mm length Clashach sandstone cores as a function of gas and

117 brine pore fluid pressure (2-7 MPa) and flow rate (20 to 80 µl min-1, corresponding to H2 and

118 brine bulk NC of 1.7-6.8x10-8 and 1.2-9.4x10-6, respectively). We show that the H2 Snwi is

119 independent of pore fluid pressure but that higher brine pore fluid pressures during secondary

120 imbibition increase capillary trapping of H2, indicating increased trapping at elevated reservoir

121 depths. Higher flowrates during secondary imbibition on the other hand decrease capillary

122 trapping. Our work advances the current understanding of H2 flow in porous media by

123 delivering the first realistic estimates on how much H2 can be recovered from the subsurface

124 after injection, disregarding any H2 loss by microbial or geochemical reactions, as a function

125 of reservoir depth and flow rate.

126

127 2. Materials and Methods

128 2.1 Core-flooding experiments

129 We performed non-steady and steady state, two-phase core flooding experiments with H2

130 (purity 99.9995 vol.%, BOC Ltd.) and brine (0.5 M CsCl or 2 M KI, Sigma-Aldrich) in a

131 Clashash sandstone (Permian, Aeolian sandstone from near Elgin in Scotland. Clashach

132 composition: ~96 wt.% quartz, 2% K-feldspar, 1% calcite, 1% ankerite42, porosity 11.1-

133 14.4%42-44) at ambient temperature (~293 K). The Clashach sandstone, with its low

134 mineralogical variability, yet still comparably big pore space, was chosen in order to facilitate

135 the visualization of the injected H2 and the interpretation of the results.

136 Two types of experiments were carried out: The first set of experiments (from now on referred

137 to as UoE (Univeristy of Edinburgh) experiments) was carried out using the µCT instrument

138 at the University of Edinburgh and were aimed at imaging the displacement and capillary

6
139 trapping of H2 by brine as a function of saturation after drainage and imbibition under different

140 experimental conditions. Three UoE experiments were carried out that investigated the effect

141 of injection pressure (2-7 MPa), whereof each was repeated once (UoE exp. 1-3, Table 1). The

142 standard error on the H2 saturation in the repeated experiments was calculated as the standard

143 deviation devided by the square root of the number of repeated experiments. One UoE

144 experiment looked at the effect of the H2/brine injection ratio on the H2 saturation (UoE exp.

145 4, Table 1). This experiment was carried out to evaluate the effect of H2 injection into aquifers

146 with different magnitudes of brine flow (hydrodynamic aquifers), and to simulate the far field

147 situation, where H2 and brine move together. Another UoE experiment investigated the effect

148 of secondary drainage and imbibition (UoE exp. 5, Table 1). UoE exp. 6 used N2 instead of H2.

149 This experiment was undertaken in order to compare the flow behaviour of the two gases,

150 which is of relevance because N2 is sometimes used as an analogue for H2, e.g. in permeability

151 measurements11, 15. Because rearrangement processes in the pore volume were noted previously

152 for N236, UoE exp. 7 (Table 1) was undertaken to examine the stability of H2 in the rock volume

153 over an experimental duration (10 hours): Ten PV of H2 were injected into a brine-saturated

154 rock and keeping the pressure constant inside the pressure vessel, with imaging undertaken at

155 the start and after 10 hours. The effect of bulk capillary number on drainage and imbibition

156 was deduced by a comparison of the results from UoE exp. 2 and 5 which used flowrates of 20

157 and 80 µl min-1, respectively, and 5 MPa injection pressure (Table 1).

158 The second type of experiment (from now on referred to as the ‘dry rock experiment’; Table

159 1) used synchrotron radiation (Diamond Light Source, I12-JEEP tomography beamline) to

160 capture the time-resolved displacement of H2 by brine in a dry, H2-saturated rock. This

161 experiment served as a base of comparison to the displacements of H2 in wet, brine-saturated

162 rock (UoE experiments).

163 All experiments used a bespoke x-ray transparent core holder for a 5 mm diameter rock core,

7
164 Table 1: Overview over experiments

Experiment Type of x- Description Sample H2/brine Flowrate Injection Confining Pore volumes Repetitions of
ray source injection Pressure Pressure injected experiment
ratio(s) (µl min-1) (MPa) (MPa)

UoE exp. 1 laboratory Primary drainage and 1 - 20 7 9 10 1 entire

imbibition in non-steady state repetition, 1
displacements of H2 and brine repetition on
drainage only
UoE exp. 2 laboratory Primary drainage and 1 - 20 5 9 10 1
imbibition in non-steady state
displacements of H2 and brine
UoE exp. 3 laboratory Primary drainage and 1 - 20 2 9 10 1
imbibition in non-steady state
displacements of H2 and brine
UoE exp. 4 laboratory Simultaneous, steady state 1 1:4, 1:1, 20 5 9 10 -
injections of H2 and brine 4:1
UoE exp. 5 laboratory Primary and secondary 1 - 80 5 9 10 -
drainage and imbibition in
non-steady state displacements
of H2 and brine
UoE exp.6 laboratory Non-steady state 1 - 20 5 9 10 -
displacements of N2 and brine
UoE exp.7 laboratory H2 stability 1 - 80 5 9 10 -
Dry rock synchrotron Non-steady state, brine 2 - 5 5 10 10 -
injection into dry, H2 saturated
rock

165

8
166 which was a scaled-up version of the cell described by Fusseis et al. (2014)45. In UoE

167 experiments a carbon-fibre reinforced PEEK pressure vessel was used to permit good x-ray

168 transparency. The dry rock experiment used an aluminium pressure vessel to comply with the

169 health and safety requirements at Diamond Light Source. The specifications of the pressure

170 vessels of PEEK and aluminium were engineered with safety factors of 2 times or more over

171 and above the maximum applied confining pressure (10 MPa; Table 1).

172 Rock cores for the experiments were obtained by diamond drill coring with a water-flushed

173 chuck, followed by preparation of the core ends by grinding on a lathe. Experiments used a set

174 of four high-pressure pumps (Cetoni NemesysTM, flowrate range 0.072 nl s-1 to 13.76 ml s-1):

175 One for the injection of H2, one for the injection of brine, one to hold the backpressure and one

176 for the confining pressure (Fig. 1). A bespoke manifold system composed of high-pressure 1/8”

177 and 1/16” 316 stainless steel and 1/16” PEEK tubing (Swagelok, Top Industrie and Cole

178 Parmer, respectively) connected the pumps to the core-flood cell (Fig. 1). Additional pressure

179 transducers (ESI Technology; accuracy 0.1% full-scale) were coupled to the flow system at the

180 inlet and outlet to allow for higher precision pressure monitoring than was possible using the

181 integral pressure gauges in the syringe pumps. Cyclic H2 and brine injections used a Clashach

182 outcrop sample without further cleaning of 4.7 mm diameter and a relatively long length of 54-

183 57 mm to avoid the influences of capillary end effects46, 47. To prevent leakage of H2 into the

184 confining fluid, the rocks were jacketed in aluminium foil and poluolefin heatshrink tubing and

185 sealed with silicone adhesive between the conical-ended pistons within the pressure vessel. In

186 UoE experiments, a water-wet rock was first saturated with brine (0.5 M CsCl) at a flow rate

187 of 70 µl min-1. Afterwards, H2 was injected (drainage) into the brine-saturated rock at flow

188 rates of 20-80 µl min-1, based on desired capillary-regime NC of 1.7-6.8x10-8 (The viscosity of

189 H2 is 9.01 μPa s at 298K and 4.7 MPa48 and the IFT between H2 and water is 72.6 mN m-1 at

190 298K and 5 MPa49, 50). Subsequently, the brine was reinjected (imbibition) at flow rates of 20-

9
 Figure 1: Experimental setup showing the manifold system that connected the X-ray transparent pressure vessel

to a set of four high-pressure Cetoni NemesysTM pumps: one to inject H2, one inject brine, one to maintain

backpressure and one to maintain confining pressure. The materials for the connections were 316 stainless steel

(black), HPLC (green) and PEEK or carbon fibre reinforced PEEK (blue). The pressure vessel consisted of

carbon fibre reinforced PEEK in UoE experiments and of aluminium in the dry rock experiment (see text).

Pressure and flow rate control was achieved with the Q-mix software.

191

192 80 µl min-1, resulting in NC of 2.35-9.45x10-6 (using the same IFT between H2 and water of

193 72.6 mN m-1 at 5 MPa and 298 K49, 50 and a viscosity of 1.247 Pa s at 5 MPa that was estimated

194 from the reported 1.2503 Pa s and 1.233x10-3 Pa s at 0.1 MPa and 25 MPa, respectively, and

195 298 K51). The NC the N2 experiment was 3.5x10-8 (using an IFT of 73 mN m-1 between N2 and

196 water 1t 298K and 10 MPa35 and a viscosity of 1.89x10-5Pa s at 5 MPa and 295K48). Each

197 injection used ten pore volumes to ensure completely flushing of the sample cores with the

198 injected fluid.

199 In the dry rock experiment, H2 was directly injected into a dry rock at a flow rate of 70 µl min-
1
200 . Subsequently, the brine (2 M KI) was injected at a flow rate of 5 µl min-1, resulting in NC of

201 5x10-7 (using the same IFT between H2 and water of 72.6 mN m-1 at 5 MPa and 298 K49, 50 and

202 a viscosity of 1.07x10-3Pa s for 0.6 M KI and 293 K42).

10
203 The combined application of an x-ray transparent core holder and µCT allowed the

204 visualization of the fluid saturation distributions at pore scale at each injection step. The

205 difference in the x-ray attenuation coefficient of the fluids (H2 and 0.5 M CsCl/ 2 M KI)

206 provided an excellent contrast between the two fluid phases and the rock on the acquired µCT

207 images, combined with the respective radiation energy in the two different laboratories.

208 3D volumes were acquired from the lower central portion of the sample to avoid the impact of

209 capillary end effects on fluid saturation46, 47. For the UoE experiments, image acquisition used

210 a µCT instrument built in-house at the University of Edinburgh, comprising a Feinfocus 10-

211 160 kV reflection source, a Micos UPR-160-air rotary table and a Perkin-Elmer XRD 0822 1

212 MP amorphous silicon flat panel detector with a terbium doped gadolinium oxysulfide

213 scintillator. Data acquisition software was developed in-house. The following settings were

214 used for UoE experiments: 120 keV, 16 W, 2 seconds exposure time, 1200 projections and 2

215 frames per stop. The voxel size was 5.4 μm3. In the dry rock experiment, time-resolved imaging

216 of the H2 and brine displacement processes was achieved by means of a 65 keV monochromatic

217 beam detected by a high-resolution imaging camera with optical module 2 (PCO.edge 5.5, 7.91

218 x 7.9 μm/pixel with FoV 20 mm x 12 mm) using 17-25 milliseconds exposure time and 900

219 projections. The voxel size was 7.9 μm3.

220 2.2 Image analysis

221 Tomographic reconstructions were undertaken by filtered back projection using Octopus 8.952

222 on a GPU accelerated workstation. All subsequent image processing and analysis of

223 tomographic data was performed using Avizo Version 9.1.1 (FEI, Oregon, USA). Data from

224 UoE experiments were processed using a non-local means filter53. Processing of the dry rock

225 experiment used a combination of median filter and unsharp mask to reduce image

226 noise.Segmentation of UoE experiment data used a global threshold on the 2D greyscale image

227 histogram, and encompassed two phases. In the water-saturated scans, water and rock were

11
228 treated as two discrete phases. In scans after brine and H2 injections, the H2 was treated as one

229 phase and the brine and rock as a single separate phase, following protocols of Andrew et al.

230 (2014)32. Holes and spots which were at the resolution limit of the data were removed from all

231 datasets (applied thresholds corresponded to 33 and 53 voxels, respectively). Based on the

232 segmented image of the water-saturated scan in UoE experiments, a pore size distribution was

233 calculated. The 3D image was separated into individual pores and throats using Avizo’s

234 ‘separate objects’ module, which calculates a chamfer distance map of the pore-space and then

235 applies a marker based watershed algorithm to the distance map to define discrete pore bodies

236 as catchment basins separated by the watershed which marks the location of pore throats (SI

237 Figure S1d). Supporting information Figure S1 shows the work flow for the water-saturated

238 scan.

239 In scans following brine and H2 injections in UoE experiments, the segmented image was

240 analysed in 3D using the ‘labeling’ and ‘label analysis’ modules to identify, label and measure

241 the volume of each H2 cluster. Hydrogen cluster size distributions were compared to the pore

242 size distribution to evaluate the H2 connectivity and identify trapping mechanisms during brine

243 imbibition.

244 2.3 Capillary pressure

245 Recovered Clashach cores were submerged in 25% w/v NaOH solution (Fisher Scientific) for

246 2 hours to remove the aluminium foil from the core surface, and rinsed in successive milli-Q

247 water, acetone and ethanol ultrasonic baths. Subsequently, the cores were cut and squared to

248 the dimensions of 25 mm, overlapping the µCT visualized rock volumes, and cleaned

249 ultrasonically with milli-Q water to remove grinding products. Mercury injection capillary

250 pressure (MICP) was performed on the cleaned cores, using a micromeritics automated

251 mercury injection equipment (Autopore IV 9500) to estimate the capillary pressure-saturation

252 relationship and pore size distribution. The pressure range tested was from vacuum to 379 MPa.

12
253 3. Results

254 3.1 Characterization of the pore space

255 The µCT-evaluated porosity of the

256 Clashach sandstone from the segmented

257 volume of the water-saturated rock was

258 12.5% to 13.5%, depending on the imaged

259 region of the rock core. The MICP-

Fig. 2. µCT-derived pore size distribution (grey) and
260 evaluated pore throat size distribution
MICP-derived pore throat size distribution (transparent
261 showed a large number of very small pore
blue). Note that MICP derived counts were multiplied
262 throats with <5 µm radius and a small by a factor of 20 to facilitate visualization of the results.

263 number of small to intermediate size pore All plots used a bin size of 30.

264 throats (>5 to 90 µm) (Fig. 2). The largest pore throat had a radius of 195 µm (Fig. 2). The

265 µCT-evaluated size distribution of the pores evidenced a narrow pore size distribution with

266 comparably small pores with radii <50 µm (Fig. 2, 4a).

267 3.2 Hydrogen wetting behaviour and stability in UoE experiments

268 Hydrogen sat in the centre of the pore bodies. Residual brine sat in corners, pore throats (Fig.

269 3b and c) and, as a subtraction of the water saturated scan from the H2-and brine filled rock

270 revealed, in thin films around the grains (Fig. 3d). The injected H2 remained stable within the

271 pore volume under no-flow conditions and at constant pore fluid pressure over a time period

272 of 10 hours which was the maximum experimental duration (SI Fig. S2).

273 3.2 Effect of pore fluid pressure on hydrogen connectivity, saturation and recovery

274 Hydrogen saturation during drainage was independent of the pore fluid pressure with 49.8%,

275 51.7% and 39.7%-52.6% saturation at pore fluid pressures of 2, 5 and 7 MPa, respectively (Fig.

276 4a-c, f, SI Fig. S3). Hydrogen connectivity during drainage generally showed one large,

13
277
278 Fig. 3: (a) Water-saturated Clashach sandstone with the water shown in black and the rock in different shades of grey. (b) and (c) Brine-saturated Clashash sandstone after
279 injection of H2. H2 (black) fills the centre of the pores while the brine (dark grey) remains in corners and small pore throats around grains. (d) and (e) Subtraction of the water-
280 wet scan from the brine-saturated scan after H2 injection, following registration of the brine-saturated scan after H2 injection to the water-saturated scan, revealing discontinuous
281 brine thin films around grains. The rim around the Al foil in (d) is caused by continued shrinkage of the Al foil onto the rock during the experiments.

14
282
283 Fig. 4: 3D rendering of H2 and N2 clusters with saturation percentages in UoE experiments. Discrete clusters were
284 rendered in colours, where mainly one color marks one large, connected cluster and different colors indicate
285 several, not connected clusters. (a-c) Effect of pore fluid pressure on H2 clusters and saturation after drainage and
286 after primary imbibition. (a) 2 MPa, (b) 5 MPa and (c, f) 7 MPa, all at a constant flow rate of 20 µl min-1
287 corresponding to capillary numbers of 1.7x10-8 and 2.4x10-6, respectively. Large, connected clusters that existed
288 after drainage were broken down to numerous smaller clusters after imbibition, with apparently no clear
289 relationship between H2 saturation and pore fluid pressure. Experiments were repeated once at 2 and MPa, and at
290 7 MPa twice for drainage runs and once for imbition. For experiments at 2 MPa and 5 MPa averages and standard
291 errors for the H2 saturation are reported. For experiments at 7 MPa, due to the discrepancy in the results, both of
292 the full primary drainage and imbibition experiments are visualized in (c) and (f). (d) Effect of cyclic injections
293 on H2 clusters and saturation: Averages and standard errors of the H2 saturation after primary and secondary
294 drainage, and after primary and secondary imbibition, all at 5 MPa pore fluid pressure and a flowrate of 80 µl
295 min-1 corresponding to a capillary number of 9.4x10 -6. (e) Nitrogen clusters and saturations during drainage and
296 imbibition at 5 MPa pore fluid pressure and a flowrate of 20 µl min-1. For the full display of the results see SI Fig.
297 S3 and SI Fig S4.

15
298 connected cluster at all pore fluid pressures except for one out of three images at 7 MPa which

299 showed three large disconnected clusters (Fig. 5, SI Fig. S3). During drainage, the largest H2

300 cluster had a volume of 1x108 μm3 at all pore fluid pressures except for the one run at 7 MPa

301 with the disconnected clusters were the largest volume was 7x107 μm3 (Fig. 5e).

302

Fig. 5: (a) Pore size distribution as derived from the micro-CT image of the water-wet rock. Hydrogen cluster size
distributions after drainage and imbibition in experiments at 20 µl min-1 flowrate and pore fluid pressures of 2
MPa (b), 5 MPa (c) and 7 MPa (d), and cumulative pore size and H2 cluster size (CS) distributions at different
pore fluid pressures (e), where squares, triangles and rhombi mark the distinct repeat experiments. (f) Hydrogen
cluster size distribution after drainage for all experiments and (g) Hydrogen cluster size distribution after
imbibition for all experiments. Note the large H2 clusters of ~108 μm3 that exist after drainage in (b)-(d). A
decrease in the biggest cluster volume after imbibition in (b)-(d) along with an increase in the number of small
clusters marks the change in H2 structure during the drainage and imbibition processes. Histogram plots in (a)-
(d), (f) and (g) used a bin size of 10.
303

16
 17

304 Hydrogen clusters during drainage were at all pore fluid pressures much larger than discrete

305 pores with a maximum volume of 1.3x106 μm3 (Fig. 5e, a). Comparing all H2 cluster size

306 distributions during drainage (Figure 5f) reveals that all drainage curves, including two of the

307 distributions at 7 MPa (squares and rhombi), have largely the same distribution, however one

308 of the three distributions at 7 MPa (triangles) is distinct. This outlier experiment corresponds

309 to the experiment showing a lower Snwi (Fig. 4c).

310 Capillary trapping of H2 during imbibition seemed independent of the pore fluid pressure with

311 10%, 12% and 4-21% of trapped H2 at 2, 5 and 7 MPa, respectively (Fig. 4a-c,f, SI Fig. S3),

312 corresponding to 20%, 22% and 11-43% of the initially injected H2. During imbibition, large

313 H2 clusters were broken down into smaller clusters (Fig. 5a-d), in line with the visual changes

314 of the H2 clusters (Fig. 4a-c and f). The largest H2 clusters after imbibition remained above the

315 maximum pore size during all experiments except for one experiment at 7 MPa (triangles in

316 Fig. 5g, Fig. 5a), showing that not only was H2 trapped in discrete pore bodies but also as larger

317 H2 ganglia. The break-down of the largest H2 clusters during imbibition caused the number of

318 clusters in the size range log 4 to log 6 μm3 to increase while the number of very small clusters

319 of log 2-4 μm3 typically decreased (Figs. 5b-d). Comparing all H2 cluster size distributions

320 during imbibition (Figure 5g) shows that the distributions at 2 and 5 MPa are largely the same

321 while the imbibition distributions at 7 MPa are distinct.

322 Injections of H2 and brine into the same rock volume and at the same flow rates and pore fluid

323 pressures of 2-5 MPa were repeatable with small standard errors between 0.01-0.66% (Fig. 4a

324 and b, Fig. 5b and c). At 7 MPa very distinct Snwi and Snwr were measured (Fig. 4c and f); During

325 drainage the standard error was 4.8% at an average H2 saturation of 47.4%. The standard error

326 during imbibition was 8.5% at an average saturation of 12.9%. The pressure differences

327 between inlet and outlet during these experiments were within the error of the pressure sensors

17
 18

328 of 0.1% full-scale.

329 3.3 Effect of hydrogen/brine injection ratio on

330 hydrogen connectivity and saturation

331 We studied the effect of the H2 /brine injection

332 ratio on the H2 saturation in order to evaluate

333 the effect of H2 injection into hydrodynamic

334 aquifers, and to evaluate what happens in the

335 far field, where fluids will be moving together.

336 The results showed that the H2 saturation and

337 H2 interconnected pore volume increased with

338 increasing H2/brine injection ratio from 32.6%

339 at 4 µl min-1 H2 plus 16 µl min-1 brine to 43.2%

340 at 16 µl min-1 H2 plus 4 µl min-1 brine (Fig. 6).

341 The H2 clusters in simultaneous injection

342 experiments occupied many of the same pore

343 spaces as the clusters after H2 in the non-

344 steady state experiments at the same pressure

345 and total flowrate (Fig. 4b, Fig. 6, SI Fig. S5),

346 and even at the lowest H2 brine injection ratio

Fig. 6: Effect of H2/brine injection ratio on H2
-1 -1
347 of 4 µl min H2 plus 16 µl min brine, the H2 saturation and H2 connectivity during simultaneous
H2 and brine injection at 5 MPa (a) 4 µl min-1 H2 plus
348 clusters were large, and spanning multiple
16 µl min-1 brine, (b) 10 µl min-1 H2 plus 10 µl min-1

349 pores. The percolation threshold, i.e. one brine, (c) 16 µl min-1 H2 plus 4 µl min-1 brine.
Discrete H2 clusters were rendered in colours, where
350 connected path from inlet to outlet, was mainly one color marks one large, connected cluster
and different colors indicate several, not connected
351 apparently only reached at 100% H2 injection
clusters. With increasing injection ratio H2 saturation
352 (Fig. 4b vs. Fig. 6, SI Fig. S5). The pressure and H2 connectivity increase.

18
 19

353 differences between inlet and outlet

354 during the simultaneous injection

355 experiments were up to 0.05 MPa.

356 The H2 cluster volume distributions

357 were similar at different H2:brine

358 injection ratios (Fig. 7a). However, with

359 increasing ratio the smallest H2 clusters

360 of volume ~log 2 μm3 decreased in

361 number while the number of

Fig. 7. (a) H2 cluster size distributions during simultaneous

362 intermediate size (log 2.5 to log 6.25 injections of H2 and brine at flowrate ratios of 16:4, 10:10

363 μm3) H2 clusters and the volume of the and 4:16 µl min-1 H2:brine and 5 MPa injection pressure. As

the flowrate ratio of H2 to brine increased, the number of

364 biggest cluster increased (Fig. 7a),
intermediate size H2 clusters increased and the radius of the
365 confirming observations of increasing
biggest cluster increased. (b) Effect of flowrate during brine
366 H2 saturation and connectivity with
imbibition. All plots used a bin size of 10.

367 increasing injection ratio (Fig. 6).

368 3.4 Effect of flowrate on hydrogen saturation and recovery

369 At constant pore fluid pressure of 5 MPa, increases in the flowrate during drainage from 20 µl

370 min-1 to 80 µl min-1, corresponding to bulk NC of 1.7 x10-8 to 6.8x10-8, respectively, decreased

371 the H2 saturation from to 51.7% to 47.7% (Figure 4b and d). Correspondingly, increases in the

372 brine flowrate during imbibition from 20 µl min-1 to 80 µl min-1, corresponding to bulk NC of

373 2.4x10-6 and 9.4x10-6, respectively, reduced the Snwr from 11.5% to 7.2% (Fig. 4b and d).

374 In line with this, the H2 cluster size distributions at the two flowrates showed that larger clusters

375 were mobilized at 80 µl min-1 (maximum cluster sizes of log 6.25 μm3 at 80 µl min-1 vs. log

376 6.75 μm3 at 20 µl min-1; Fig. 7b). At both flowrates the largest H2 clusters were still bigger than

19
 20

377 the largest pore of <log 6 μm3 (Fig. 7b vs. Fig 6a), indicating that H2 was trapped also as larger

378 ganglia.

379 3.5 Secondary drainage and imbibition

380 Secondary drainage and imbibition did not significantly change the H2 saturation (47.9% and

381 7.0%) compared to primary drainage and imbibition (47.6% and 7.3%), based on results at 5

382 MPa and 80 µl min-1 flowrate (SI Fig. S4, Fig. 4d-e).

383 3.6 Dry rock experiment

384 In the dry-rock experiment, the brine entered the H2-filled dry rock via piston-like

385 displacement, (Fig. 8b), eventually recovering the H2 entirely (Fig. 8d). Before all H2 was

386 recovered, there was an intermediate stage where previously brine-filled pores (Fig. 8b)

387 showed several very small H2 bubbles (Fig. 8d).

388 3.7 Comparison to nitrogen

389 The N2 saturation was similar to the H2 saturation during drainage (48.7% vs. 51.7 ± 0.66%,

390 respectively) but N2 saturation after brine imbibition was much higher (33.9% vs. 11.5 ±

391 0.64%, respectively; Fig. 4e, SI Fig. S6). The Nc of H2 and N2 during drainage were 1.7x10-8

392 and 3.5x10-8, respectively.

393

394 4. Discussion

395 4.1 Pore space

396 The µCT-evaluated porosity of the Clashach sandstone from the segmented volume of the

397 water-saturated rock of 12.5-13.6% was within the range of the published porosities of 11.1-

398 14.4% for Clashach sandstone42-44. The distributions of pore size and pore throat size as

399 evaluated by µCT and MICP, respectively, suggested that mostly small pores of <50 µm radii

400 were joined by very small throats of <5 µm radii with a few small to intermediate size throats

20
 21

401

402 Fig. 8. The dry-rock experiment summarized. (a) H2-saturated, dry rock, (b) scan during brine imbibition shortly
403 after appearance of the first brine in the rock, (c) scan during brine imbibition after 115 minutes, with several
404 small, isolated H2 bubbles inside pore bodies (d), and (e) brine-saturated rock after two hours of imbibition
405 showing 100% recoverability of H2.
406

407 of >5 to 90 µm radii in between (Fig. 2). The largest pore throat of 195 µm radius was probably

408 measured at the surface of the rock core where the drilling process affected the pore space. The

409 pore throat distribution for our Clashach sandstone sample was very similar to the pore throat

410 distribution for Berea sandstone54. Compared to the pore throat distributions for Bentheimer

21
 22

411 sandstone and Doddington sandstone54, our Clashach sandstone sample showed smaller pore

412 throat sizes. Limestones generally show a wider pore throat size distributions than sandstones54.

413 4.2 H2 flow behaviour and trapping mechanisms

414 Hydrogen behaved as a non-wetting phase, filling the centre of the pores, with residual brine

415 in the pore corners and throats (Fig. 3b and c), indicating a water wetting system. The largest

416 H2 cluster was much larger than discrete pores at any pore fluid pressure during drainage (Fig.

417 5), indicating a good connectivity of the H255. Hydrogen trapping occurred via snap-off of H2

418 ganglia (Fig. 9). Snap-off competes with piston-like, i.e. pore-filling, displacement during the

419 displacement of a non-wetting fluid by a wetting fluid in porous media21, and is known as the

420 swelling of water in the corner layers of a pore throat during water invasion in water-wet porous

421 rocks until the threshold capillary pressure is exceeded, resulting in spontaneous filling of the

422 throat with water and disconnection of the non-wetting phase which can lead to trapping21.

423 Brine films around grains were not directly visible in the tomographic images (Fig. 3b and c)

424

Fig. 9: Example of a snap-off event. (a) Labelled H2-filled volume after drainage (orange), spanning over
several pores, and total pore space (transparent blue) showing one large interconnected H 2-filled pore volume.
(b) Labelled H2-filled volume after brine imbibition (different coloured shades) and the total pore space
(transparent blue) showing several, not connected H2 ganglia and the snapped-off H2 droplet (cobalt blue) in
the centre, left hand side. (c) Pore body visualisation of the same volume.

22
 23

425 but were revealed by subtraction of the water-saturated scan from the brine-saturated scan after

426 H2 injection, following the registration of the brine-saturated scan after H2 injection to the

427 water-wet scan (Fig. 3d and e). Figures 3d and 3e suggest that brine films were discontinuous

428 and very thin. When H2 was injected into a dry rock, 100% of the injected H2 could be recovered

429 (Fig. 8e) which substantiated the theory that sub-resolution brine films around grains and snap-

430 off of H2 ganglia caused decreased H2 recovery in experiments using an initially brine saturated

431 rock (UoE experiments). The occurrence of several very small H2 bubbles in the dry-rock

432 experiment (Fig. 8d) indicated Roof snap-off56 of H2 ganglia.

433 4.3 Effect of pore fluid pressure and hydrogen/brine injection ratio

434 We observed no dependence of the H2 saturation during drainage on pore fluid pressure,

435 considering that 2 out of 3 experiments at 7 MPa showed the same the same H2 saturation of

436 ~50 % as at 2 and 5 MPa (Fig. 4a-c,f, and SI Fig. S3). The one experiment at 7 MPa which had

437 only 39.7% H2 saturation (Fig. 4c, SI Fig. S3 and blue triangles in Fig. 5f) did also not have

438 the same H2 cluster size distribution as the remaining experiments (Fig. 5f), despite using the

439 same experimental settings as for all other experiments at 7 MPa, and the log archives of the

440 pore fluid pressures and injected volumes revealed no abnormalities. A shift in the distribution

441 of cluster sizes can indicate a change in wettability, regardless of the measured H 2 saturation.

442 Yet, as two of the results at 7 MPa showed a similar distribution as at the other pressures, it

443 seems likely that this experiment is an outlier. The experiment was acquired after a filament

444 change on the µCT apparatus, which implied that a slightly different part of the same rock core

445 was imaged (13.6% vs. 12.5% porosity). Yet, in principle this should not have affected the

446 results significantly, and subsequent experiments did return to show ~50% H2 saturation, e.g.

447 the H2 stability experiment (SI Fig. S2).

448 The observed constant drainage H2 saturations with increasing pore fluid pressures from 2 to 7

449 MPa are in line with a lack of a dependence of the H2 wettability on pressure increases from 2-

23
 24

450 10 MPa in Berea and Bentheimer sandstone18, with only very small increases of ~3-6° in the

451 H2 contact angles at pressure increases from 2 to 7 MPa in Basalt21 and clay24 and quartz19, and

452 with a previous findings of no change in the characteristic trapping curves for CO2 and N2 at a

453 wide range of pressure and temperature conditions35. The general anticipation of an increase in

454 gas saturation with injection pressure57, 58 may still be valid over pressure ranges larger than

455 the one investigated here. At unchanged wettability, the Snwi is controlled by the capillary

456 pressure which in turn is controlled either by the fractional flow (during simultaneous injection)

457 or by the viscous force pressure drop (during single fluid phase injection). Thus, the

458 independence of the Snwi from the pore fluid pressure indicates that the viscous pressure drop

459 is not being significantly altered by the changing pressure, e.g., the H2 viscosity change is not

460 having a major impact on the force required to drive flow. Significant loss of H2 from the gas

461 phase at higher pressures by dissolution into the brine is precluded by the low solubility of H2

462 of ~0.02 mol kgw-1 at 2.5 MPa59. The H2 stability experiments showed that H2 saturation at 5

463 MPa did not change over a time period of ten hours (49.55% at time zero vs. 49.53% ten hours

464 after; SI Fig. S2), evidencing a stable result and no H2 loss by dissolution.

465 Looking at the Snwr data only (Fig. 4a-c,f), there was no clear dependency on pore fluid pressure

466 during imbibition. Any change may, however, have been masked by the high variation between

467 the two results at 7 MPa (4-21% Fig. 4c and f). Considering that the experiment at 7 MPa with

468 the low Snwi of 40% could be identified as an outlier (see the discussion in the beginning of

469 section 4.3, and Fig. 5f, blue triangles) and that the Snwr is a function of the Snwi where a lower

470 Snwi will tend to overestimate recovery54, 60, we may disregard the Snwr of 4%. The large increase

471 in trapped H2 in the second full primary drainage and secondary imbibition experiment at 7

472 MPa (Fig. 4f) was probably due to the poorer intial H2 connectivity60. The H2 cluster size

473 distribution for this experiment showed an increase in the number of intermediate size clusters

474 compared to the other experiments (Fig. 5g, darkyellow squares), without however shifting the

24
 25

475 cluster distribution, suggesting no wettability change. This indicates that variation in one or

476 more of the other thermophysical properties –density, viscosity, or IFT – have resulted in the

477 impact on the pore scale fluid configuration (Figure 10). Meanwhile, a poorer initial H2

478 connectivity was not confirmed by the third repetition of primary drainage at 7 MPa (SI Fig.

479 S3g). More experiments at 7 MPa are needed to confirm the result of increased trapping at

480 higher pore fluid pressures.

481

Fig. 10: Density (green line), viscosity (red line), and

interfacial tension (blue line) of hydrogen versus depth

(Hassanpouryouzband et al. 2021)61.

482

483 Given a hydrostatic gradient of ~10 MPa/km an increase of the Snwr with increasing pore fluid

484 pressure, as suggested by the second experiment at 7 MPa (Fig. 4f), would indicate that deeper

485 aquifers are less favourable for H2 storage operations. However, unlike our non-steady state

486 experiments which showed barely any pressure difference between inlet and outlet, in a real

487 H2 storage operation, the well pressure is significantly higher than the reservoir pressure and

488 the brine is not injected, but naturally flowing into previously H2-saturated rock when the H2

25
 26

489 is recovered- again due to a pressure difference. Our displacement study results are hence

490 applicable to the fringe of the H2-saturated zone, only, where pressure differences are very

491 small. A decreased H2 recovery with depth would not align well with other criteria for an

492 economical and safe H2 storage operation, such as a lower cushion gas requirement with

493 elevated depth5 and reduced risks for H2-linked microbial activity at higher depths due to higher

494 temperatures6.

495 The pressure/depth effect on our results can be further related to the variation of key pore-scale

496 displacement parameters for H2 with depth (Fig. 10): The IFT of H2 reduces with depth whereas

497 the H2 viscosity increases, both of which in theory should augment the displacement of H2 with

498 water and reduce the Snwr62, 63. However, increased Snwr with decreasing IFT and increasing

499 viscosity has also been demonstrated64 which may be due to a simultaneously increased

500 likelihood of unstable displacement/ non-uniform fronts10, 65

during both drainage and

501 imbibition processes at decreasing IFT and increasing viscosity, based on the augmenting effect

502 of both parameters on NC (Eq. 1). Such unstable displacement processes at higher pore fluid

503 pressures, addition to a variation in the thickness of the brine thin films with pressure, may

504 explain the decrease in interconnected H2-filled pore volume from one large cluster at 2-5 MPa

505 to three clusters at 7 MPa and have lead to increased snap-off and trapping during imbibition

506 (Fig. 4f).

507 The Snwr of 10-21% was significantly less than the previously reported H2 Snwr of 41% for a

508 sandstone under ambient conditions40. As mentioned previously, the short length rock sample

509 in Jha et al. (2021)40 suggests that their results were affected by capillary end effects46.

510 However, the applied the relatively high bulk brine NC of 2.4x10-6 may have mobilised more

511 residual H2 than under strict capillary regime conditions.

512 During simultaneous injection of H2 and brine, which may be representative of H2 injection

513 into hydrodynamic aquifers or simulate the far field conditions, H2 saturation and H2

26
 27

514 interconnected pore volume increased with increasing H2/brine injection ratio (Fig. 6). This

515 indicated that a lower brine flow is favoured over high flow environments in terms of

516 optimising the H2 storage operation. The structure was apparently not percolating in any of the

517 simultaneous injection experiments as opposed to during 100% H2 injection (Fig. 6 vs. Fig.

518 4b), yet considering significant pressure differences of up to 0.05 MPa between inlet and outlet

519 in simultaneous injection experiments which were not observed in experiments injecting 100%

520 H2, the connections between the H2 clusters may have been broken when the injection (and

521 thereby the pressure gradient) was stopped for the scan.

522 4.4 Effect of capillary number on initial and residual saturation

523 Classical pore-scale displacement theory predicts little change in residual phase saturation in

524 response to increases in flow rate until NC exceeds 10-6 -10-5. However, for most subsurface

525 reservoirs there will be rapid decreases at NC of 10-4 or more, when viscous forces become

526 dominant54. The bulk brine NC applied in this study (2.4-9.4x10-6) was within the range of little

527 saturation change but exceeded the threshold of NC <10-6 for which the flow generally is said

528 to be capillary dominated54. This may indicate that viscous forces caused a significant effect of

529 NC on the Snwr in our experiments (Fig. 4b and d), and these forces are likely to be even greater

530 at local scale than at bulk54. The H2 cluster size distribution after imbibition at NC= 9.4x10-6

531 was shifted with respect to the distribution at NC= 2.4x10-6 (Fig. 7b), indicating a change in the

532 wetting behaviour and supporting previous findings of preferential desaturation of larger

533 clusters at higher Nc66.

534 We observed a 4% decrease in Snwi in our experiments when bulk NC was increased from

535 1.7x10-8 to 6.8x10-8 (Fig. 4b and d). Critical nonwetting phase NC of 2x10–8 and 10–5 during

536 imbibition have been reported for water–gas systems and water-oil systems, respectively67,

537 indicating that the threshold of NC <10-6 for capillary dominated flow54 is not rigid. However,

538 considering reported increases in the H2 saturation after drainage with increasing NC from

27
 28

539 7.7x10-7 to 7.7x10-5 20, and acknowledging the small observed difference in Snwi, we cannot

540 exclude that the effect of flow rate was down to experimental variability.

541 4.5 Comparison to nitrogen

542 The N2 saturation was comparable to the H2 saturation during drainage at similar NC of 1-3x10-
8
543 but the Snwr after imbibition was ~20 % higher for N2 than for H2 (Fig. 4b and e). Using N2 as

544 a proxy for H2 in experimental drainage and imbibition studies is hence not advisable.

545 Considering the high degree of N2 trapping, the use of N2 as a cushion gas for H2 storage

546 operations which could reduce operational costs8 seems favourable. Our results are lower than

547 a reported 64% intial N2 saturation during drainage and 43% residual N2 saturation during brine

548 (200 g/L NaI +15 g/L CaCl2) imbibition in a Berea sandstone (20-22% porosity) at 5.5 MPa

549 and 20°C37, and higher than 43% N2 saturation during drainage in a Bashijiqike tight sandstone

550 (5.6% porosity) at 8 MPa and ambient temperature36. The trend in the differences of the N2

551 saturation in the above studies follows the same trend as the differences in the porosities of the

552 studied sandstones, with the Clashach sandstone (14% porosity) being intermediate between

553 the two other rocks. This indicates that porosity differences between the different rock types

554 applied in the above experiments defined the observed N2 saturations, yet differences in the

555 pore throats dimensions may equally have contributed or caused this. Hence Snwi and Snwr

556 depend strongly on (the local conditions within) each rock, and these rock type/local effects

557 may mask any effect of injection conditions. However, trends in the rock-specific behaviour

558 will be controlled by pressure and flow conditions. It follows from this that any rock selected

559 for H2 storage operations has to be very well characterized, in particular with respect to

560 porosity, turtoisity and capillary pressure curves.

561 4.6 Suitability of the Clashach sandstone for hydrogen storage

28
 29

562 It has been postulated that the low viscosity of H2 will cause the gas to travel swiftly, making

563 it little suitable for displacing brine10 and causing low H2 injectivity. In this work, we showed

564 that from an injectivity and recovery perspective, untreated Clashach sandstone is suitable for

565 underground H2 storage. However, sandstones aged by exposure to humic acids may be more

566 suitable analogue rocks for experimental investigations of H2 storage in porous media19.

567 Considering that aging has previously been shown to alter the wettability of H2 brine-quartz

568 systems from highly water-wet toward intermediate-wet19 the stated H2 saturations for our

569 untreated outcrop Clashach sandstone are expected to increase during drainage and decrease

570 during imbibition, further the increasing suitability for H2 storage.

571

572 5. Conclusion

573 In this work, a prima facie examination of H2 flow and displacement processes in porous rock

574 was carried out as a function of capillary numbers of 1.2-6.8x10-8 for H2 and 2.4-9.5x10-6 for

575 brine, and of pore fluid pressures between 2-7 MPa. Results showed no clear relation between

576 pore fluid pressure and H2 saturation with ~50% of the pore space saturated with H2 during

577 drainage at all pressures, and 20%, 24% and 43% of the initial H2 trapped at 2, 5 and 7 MPa,

578 respectively, during imbibition at a capillary number of 2.4x10-6, indicating that higher

579 pressure, i.e. deeper reservoirs are less favourable for H2 storage. Injection of brine at higher

580 capillary numbers reduced capillary trapping and increased H2 recovery. Hydrogen recovery

581 was distinct from N2 recovery, suggesting that N2 is a poor proxy for H2. Based on these results,

582 we recommend more shallow, lower pressure sites for future H2 storage operations in porous

583 media. Future work should aim to measure the influence of pressures above 7 MPa, elevated

584 temperatures and rock aging on H2 and brine displacement processes at a range of different

585 capillary numbers. Dynamic experiments using synchrotron light sources are recommended to

586 examine displacement mechanisms closely.

29
 30

587

588 Author Contributions

589 The experimental kit was built by Butler and Thaysen. Experiments were carried out by Thaysen,

590 Butler, Hassanpouryouzband, Freitas. Alvarez-Borges, Atwood and Humphreys were the

591 supporting beamline scientists during the experiments at Diamond Light Source. Image

592 reconstruction and tomographic analysis was carried out by Thaysen. Interpretation of the results

593 was by Thaysen, Butler, Hassanpouryouzband, Krevor and Heinemann. The manuscript was

594 written by Thaysen with review by Butler, Krevor Heinemann, Hassanpouryouzband, Freitas and

595 Edlmann. All authors have given approval to the final version of the manuscript.

596

597 Declaration of interest

598 The authors declare no competing interest.

599

600 Funding Sources

601 Thaysen, Butler, Hassanpouryouzband, Heinemann and Edlmann gratefully acknowledge the

602 funding support from the Engineering and Physical Science Research Council (EPSRC)

603 HyStorPor project [grant number EP/S027815/1] and from the Fuel Cells and Hydrogen 2 Joint

604 Undertaking (JU) under grant agreement No 101006632. The JU receives support from the

605 European Union's Horizon 2020 research and innovation programme and Hydrogen Europe

606 and Hydrogen Europe Research. This work was enabled and carried out with the support of

607 Diamond Light Source, Beamline I12-JEEP 101006632 (proposal mg26730-1).

608

609 ACKNOWLEDGMENT

610 We would like to thank Thomas Connolley for any correspondence and help with setting up

611 the experiments at Diamond Light Source, and Alexis Cartwright-Taylor for assistance and

612 discussions around the experimental results.

30
 31

613 References

614 1. Tarkowski, R., Underground hydrogen storage: Characteristics and prospects Renew

615 Sust Energ Rev 2020, 105, 86-94.

616 2. Beckingham, L. E.; Winningham, L., Critical knowledge gaps for understanding

617 water-rock-working phase interactions for compressed energy storage in porous formations.

618 Acs Sustain Chem Eng 2020, 8 (1), 2-11.

619 3. Mouli-Castillo, J.; Heinemann, N.; Edlmann, K., Mapping geological hydrogen

620 storage capacity and regional heating demands: An applied UK case study. Appl Energy

621 2021, 283 (116348).

622 4. Hassanpouryouzband, A.; Joonaki, E.; Edlmann, K.; Heinemann, N.; Yang, J.,

623 Thermodynamic and transport properties of hydrogen containing streams. Sci Data 2020, 1

624 (1), 1-14.

625 5. Heinemann, N.; Booth, M. G.; Haszeldine, R. S.; Wilkinson, M.; Scafidi, J.;

626 Edlmann, K., Hydrogen storage in porous geological formations - onshore play opportunities

627 in the Midland Valley (Scotland, UK). Int J Hydrog Energy 2018, 43 (45), 20861-20874.

628 6. Thaysen, E. M.; McMahon, S.; Strobel, G. J.; Butler, I. B.; Ngwenya, B. T.;

629 Heinemann, N.; Wilkinson, M.; Hassanpouryouzband, A.; McDermott, C. I.; Edlmann, K.,

630 Estimating microbial growth and hydrogen consumption in hydrogen storage in porous media

631 Renew Sustain Energ Rev 2021, 151 (111481), 1-15.

632 7. Matos, C. R.; Carneiro, J. F.; Silva, P. P., Overview of large-scale underground

633 energy storage technologies for integration of renewable energies and criteria for reservoir

634 identification. J. Energy Storage 2019, 21, 241-258.

635 8. Heinemann, N.; Alcalde, J.; Miocic, J.; Hangz, S. J. T.; Kallmeyer, J.; Ostertag-

636 Henning; Hassanpouryouzband, A.; Thaysen, E. M.; Strobel, G.; Wilkinson, M.; Schmidt-

637 Hattenberger, C.; Edlmann, K.; Bentham, M.; Haszeldine, S.; Carbonell, R.; Rudloff, A.,

31
 32

638 Enabling large-scale hydrogen storage in porous media: the scinetific challenges. Energy

639 Environ Sci 2021, 14, 853-864.

640 9. Paterson, L., The implications of fingering in underground hydrogen storage. Int J

641 Hydrogen Energ 1983, 8 (1), 53-59.

642 10. Chaturvedi, K. R.; Bajpai, S.; Trivedi, J.; Sharma, T., Air foams for mobility control

643 and subsurface storage of hydrogen in porous media: An experimental study. Energy & Fuels

644 2022, 36, 5036-5046.

645 11. Flesch, S.; Pudlo, D.; Albrecht, D.; Jacob, A.; Enzmann, F., Hydrogen underground

646 storage-Petrographic and petrophysical variations in reservoir sandstones from laboratory

647 experiments under simulated reservoir conditions. Int J Hydrog Energy 2018, 43 (45), 20822-

648 20835.

649 12. Yekta, A. E.; Pichavant, M.; Audigane, P., Evaluation of geochemical reactivity of

650 hydrogen in sandstone: Application to geological storage. J Appl Geochem 2018, 95, 182-

651 194.

652 13. Hassanpouryouzband, A.; Adie, K.; Cowen, T.; Thaysen, E. M.; Heinemann, N.;

653 Butler, I. B.; Wilkinson, M.; Edlmann, K., Geological Hydrogen Storage: Geochemical

654 Reactivity of Hydrogen with Sandstone Reservoirs. ACS Energy Lett 2022, 7, 2203-2210.

655 14. Yekta, A. E.; Manceau, J. C.; Gaboreau, S.; Pichavant, M.; Audigane, P.,

656 Determination of Hydrogen-Water Relative Permeability and Capillary Pressure in

657 Sandstone: Application to Underground Hydrogen Injection in Sedimentary Formations.

658 Transport Porous Med 2018, 122 (2), 333-356.

659 15. Rezaei, A.; Hassanpouryouzband, A.; Molnar, I.; Derikvand, Z.; Haszeldine, R. S.;

660 Edlmann, K., Relative permeability of hydrogen and aqueous brines in sandstones and

661 carbonates at reservoir conditions. Geophys Res Lett 2022.

32
 33

662 16. Higgs, S.; Da Wang, Y.; Sun, C.; Ennis-King, J.; Jackson, S. J.; Armstrong, R. T.;

663 Mostaghimi, P., In-situ hydrogen wettability characterisation for underground hydrogen

664 storage. Int J Hydrog Energy 2021, 47 (26), 13062-13075.

665 17. Esfandyari, H.; Sarmadivaleh, M.; Esmaeilzadeh, F.; Ali, M.; Iglauer, S.;

666 Keshavarz, A., Experimental evaluation of rock mineralogy on hydrogen-wettability:

667 Implications for hydrogen geo-storage. J Energy Storage 2022, 52 (104866), 1-8.

668 18. Hashemi, L.; Glerum, W.; Farajzadeh, R.; Hajibeygi, H., Contact angle measurement

669 for hydrogen/brine/sandstone system using captive-bubble method relevant for underground

670 hydrogen storage. Adv Water Resour 2021, 154 (103964), 1-13.

671 19. Iglauer, S.; Muhammad, A.; Keshavarz, A., Hydrogen wettability of sandstone

672 reservoirs: implications for hydrogen geo-storage. Geophys Res Lett 2020, 48 (3), 1-5.

673 20. Lysyy, M.; Ersland, G.; Fernø, M., Pore-scale dynamics for underground porous

674 media hydrogen storage. Adv Water Resour 2022, 163 (104167), 1-13.

675 21. Al-Yaseri, A.; Jha, N. K., On hydrogen wettability of basaltic rock. J Pet Sci Eng

676 2021, 200, 1-5.

677 22. Hosseini, M.; Ali, M.; Fahimpour, J.; Keshavarz, A.; Iglauer, S., Basalt-H2-brine

678 wettability at geo-storage conditions: Implication for hydrogen storage in basaltic formation.

679 J Energy Storage 2022, 52 (104745), 1-6.

680 23. van Rooijen, W.; Hashemi, L.; Boon, M.; Farajzadeh, R.; Hajibeygi, H.,

681 Microfluidics-based analysis of dynamic contact angles relevant for underground hydrogen

682 storage. Adv Water Resour 2022, 164 (104221), 1-17.

683 24. Al-Yaseri, A.; D., W.-B.; Fauziah, C. A.; Iglauer, S., Hydrogen wettability of clays:

684 Implications for underground hydrogen storage. Int J Hydrog Energy 2021, 46 (69), 34356-

685 34361.

33
 34

686 25. Azin, R.; Izadpanahi, A., Fundamentals and practical aspects of gas injection

687 Springer: 2022.

688 26. Morrow, N.; Chatzis, I.; Taber, J. J., Entrapment and mobilization of residual oil in

689 bead packs. SPE Reservoir Engineering 1988, 3 (3).

690 27. Blunt, M. J.; Bijeljic, B.; Dong, H.; Gharbi, O.; Iglauer, S.; Mostaghimi, P.;

691 Paluszny, A.; Pentland, C. H., Pore-scale imaging and modelling. Adv Water Resour 2013,

692 51, 197-216.

693 28. Saraf, S.; Bera, A., A review on pore-scale modeling and CT scan technique to

694 characterize the trapped carbon dioxide in impermeable reservoir rocks during sequestration.

695 Renew Sust Energ Rev 2021, 144 (110986), 1-20.

696 29. Xu, L.; Myers, M.; Li, Q.; White, C.; Zhang, X., Migration and storage

697 characteristics of supercritical CO2 in anisotropic sandstones with clay interlayers based on

698 micro-CT experiments. J Hydrol 2020, 580 (124239), 1-17.

699 30. Iglauer, S.; Paluszny, A.; Pentland, C. H.; Blunt, M. J., Residual CO2 imaged with x‐

700 ray micro‐tomography. Geophys Res Lett 2011, 38 (L21403), 1-6.

701 31. Andrew, M.; Menke, H.; Blunt, M. J.; Bijeljic, B., The imaging of dynamic

702 multiphase fluid flow using synchrotron-based x-ray microtomography at reservoir

703 conditions. Tranp Porous Med 2015, 110 (1), 1-24.

704 32. Andrew, M.; Bijeljic, B.; Blunt, M. J., Pore-scale imaging of trapped supercritical

705 carbon dioxide in sandstones and carbonates. Int J Greenh Gas Control 2014, 22, 1-14.

706 33. Krevor, S. C. M.; Pini, R.; Zuo, L.; Benson, S. M., Relative permeability and

707 trapping of CO2 and water in sandstone rocks at reservoir conditions. Water Resour Res 2012,

708 48, 1-16.

709 34. Rahman, T.; Lebedev, M.; Barifcani, A.; Iglauer, S., Residual trapping of

710 supercritical CO2 in oil-wet sandstone. J Colloid Interf Sci 2016, 469, 63-68.

34
 35

711 35. Niu, B.; Al-Menhali, A.; Krevor, S. C., The impact of reservoir conditions on the

712 residual trapping of carbon dioxide in Berea sandstone. Water Resour Res 2015, 51 (4), 2009-

713 2029.

714 36. Cao, Q.; Gong, Y.; Fan, T.; Wu, J., Pore-scale simulations of gas storage in tight

715 sandstone reservoirs for a sequence of increasing injection pressure based on micro-CT. J Nat

716 Gas Sci Eng 2019, 64, 15-27.

717 37. Khishvand, M.; Alizadeh, A. H.; Piri, M., In-situ characterization of wettability and

718 pore-scale displacements during two- and three-phase flow in natural porous media. Adv

719 Water Resour 2016, 97, 270-298.

720 38. Sarmadivaleh, M.; Al-Yaseri, A.; Iglauer, S., Influence of temperature and pressure

721 on quartz–water–CO2 contact angle and CO2–water interfacial tension. J Colloid Interf Sci

722 2015, 441, 59-64.

723 39. Al-Yaseri, A.; Esteban, L.; Giwelli, A.; Sarout, J.; Lebedev, M.; Sarmadivaleh, M.,

724 Initial and residual trapping of hydrogen and nitrogen in Fontainebleau sandstone using

725 nuclear magnetic resonance core flooding. Int J Hydrog Energy 2022, in press.

726 40. Jha, N. K.; Al-Yaseri, A.; Ghasemi, M.; Al-Bayati, D.; Lebedev, M.; Sarmadivaleh,

727 M., Pore scale investigation of hydrogen injection in sandstone via X-ray micro-tomography.

728 Int J Hydrog Energy 2021, 46, 34822-34829.

729 41. Pak, T.; Butler, I. B.; Geiger, S.; van Dijke, M. I. J.; Sorbie, K. S., Droplet

730 fragmentation: 3D imaging of a previously unidentified pore-scale process during multiphase

731 flow in porous media. P Natl Acad Sci USA 2015, 112 (7), 1947-1952.

732 42. Iglauer, S.; Ferno, M. A.; Shearing, P.; Blunt, M. J., Comparison of residual oil

733 cluster size distribution, morphology and saturation in oil-wet and water-wet sandstone. J

734 Colloid Interf Sci 2012, 375, 187-192.

35
 36

735 43. Iglauer, S.; Paluszny, A.; Blunt, M. J., Simultaneous oil recovery and residual gas

736 storage: A pore-level analysis using in situ X-ray micro-tomography. Fuel 2013, 103, 905-

737 914.

738 44. Pentland, C. H.; Tanino, Y.; Iglauer, S.; Blunt, M. J., Capillary trapping in water-wet

739 sandstones: coreflooding experiments and pore-network modeling. In SPE International

740 Society of Petroleum Engineers Florence, Italy 2010.

741 45. Fusseis, F.; Steeb, H.; Xiao, X.; Zhu, W.-l.; Butler, I. B.; Elphick, S.; Maeder, U.,

742 A low-cost X-ray-transparent experimental cell for synchrotron-based X-ray

743 microtomography studies under geological reservoir condition. J Synchrotron Radiation 2014

744 21 (1), 251-253.

745 46. Pak, T. Saturation tracking and identification of residual oil saturation. University of

746 Edinburgh 2014.

747 47. Kumar, M.; Sok, R.; Knackstedt, M. A.; Latham, S.; Senden, T. J.; Sheppard, A.

748 P.; Varslot, T.; Arns, C. In Mapping fluid distributions in 3D at the pore scale: Quantifying

749 the influence of wettability and saturation

750 history on rock resistivity, SPWLA 50th Annual Logging Symposium, Society of

751 Petrophysicists and Well-Log Analysts: 2009.

752 48. Yusibani, E.; Nagahama, Y.; Kohno, M.; Takata, Y.; Woodfield, P. L.; Shinzato,

753 K.; Fujii, M., A capillary tube viscometer designed for measurements of hydrogen gas

754 viscosity at high pressure and high temperature. Int J Thermophys 2011, 32 (6), 1111-1124.

755 49. Chow, Y. T. F.; Maitland, G. C.; Trusler, J. P. M., Interfacial tensions of (H2O + H-

756 2) and (H2O + CO2 + H-2) systems at temperatures of (298-448) K and pressures up to 45

757 MPa. Fluid Phase Equilibr 2018, 475, 37-44.

36
 37

758 50. Chow, Y. T. F.; Maitland, G. C.; Trusler, J. P. M., Erratum to "Interfacial tensions of

759 (H2O + H-2) and (H2O + CO2 + H-2) systems at temperatures of (298-448) K and pressures

760 up to 45 MPa". Fluid Phase Equilibr 2020, 503 (112315), 1.

761 51. Nakai, T.; Sawamura, S.; Taniguchi, Y.; Yamaura, Y., Effect of pressure on the

762 viscosity B coefficient of cesium chloride in water Mater Sci Res Int 1996, 2 (3), 143-147.

763 52. Vlassenbroeck, J.; Masschaele, B.; Cnuddle, V.; Dierick, M.; Pieters, K.; Van

764 Hoorebeke, L.; Jacobs, P., Octopus 8: A high performance tomographic reconstruction

765 package for X-ray tube and synchrotron micro-CT. In Advances in X-ray tomography for

766 geomaterials, Desrues, J.; Viggiani, G.; Bésuelle, P., Eds. ISTE: London, UK, 2010.

767 53. Buades, A.; Coll, B.; Morel, J. M. In A non-local algorithm for image denoising,

768 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition

769 (CVPR'05), 2005; pp 60-65.

770 54. Blunt, M. J., Multiphase flow in permeable media. A Pore-scale perspective

771 Cambridge University Press Cambridge, United Kingdom, 2017.

772 55. Singh, K.; Menke, H.; Andrew, M.; Lin, Q.; Rau, C.; Blunt, M. J.; Bijeljic, B.,

773 Dynamics of snap-off and porefilling events during two-phase fluid flow in permeable media.

774 Sci Rep 2017, 7 (5192), 1-13.

775 56. Roof, J. G., Snap-off of oil droplets in water-wet pores. Soc Petrol Eng J 1970, 10 (1),

776 85-90.

777 57. Thomas, L. K.; Katz, D. L.; Tek, M. R., Threshold pressure phenomena in porous

778 media. Soc Petrol Eng J 1968, 174-184.

779 58. Boone, M. A.; Kock, T. D.; Bultreys, T.; Schutter, G. D.; Vontobel, P.; Hoorebeke,

780 L. V.; Cnudde, V., 3D mapping of water in oolithic limestone at atmospheric and vacuum

781 saturation using X-ray micro-CT differential imaging. Mater Char 2014, 97, 150-160.

37
 38

782 59. De Lucia, M.; Pilz, P.; Liebscher, A.; Kuhn, M., Measurements of H2 solubility in

783 saline solutions under reservoir conditions: preliminary results from project H2STORE.

784 Enrgy Proced 2015, 76, 487-494.

785 60. Herring, A. L.; Harper, E.; Andersson, L.; Sheppard, A.; Bay, B. K.; Wildenschild,

786 D., Effect of fluid topology on residual nonwetting phase trapping: Implications for geologic

787 CO2 sequestration. Adv Water Resour 2013, 62, 47-58.

788 61. Hassanpouryouzband, A.; Joonaki, E.; Edlmann, K.; Haszeldine, R. S., Offshore

789 Geological Storage of Hydrogen: Is This Our Best Option to Achieve Net-Zero? ACS Energy

790 Lett 2021, 6 (6), 2181–2186.

791 62. Wagner, O. R.; Leach, R. O., Effect of interfacial tension on displacement efficiency.

792 SPE J 1966, 6 (4), 335-344.

793 63. Avraam, D. G.; Payatakes, A. C., Flow Mechanisms, Relative Permeabilities, and

794 Coupling Effects in Steady-State Two-Phase Flow through Porous Media. The Case of

795 Strong Wettability. Ind Eng Chem Res 1999, 38, 778-786.

796 64. Cho, J.; Kim, T. H.; Chang, N.; Lee, K. S., Effects of relative permeability change

797 resulting from interfacial tension reduction on vertical sweep efficiency during the CO2-LPG

798 hybrid EOR process. Energy Sources 2018, 40 (10), 1242-1249.

799 65. Sabet, N.; Hassanzadeh, H.; Abedi, J., Stability of Gravitationally Unstable Double

800 Diffusive Transient Boundary Layers with Variable Viscosity in Porous Media. Aiche J 2017,

801 63 (6), 2471-2482.

802 66. Oughanem, R.; Youssef, S.; Peysson, Y.; Bazin, B.; Maire, E.; Vizika, O., Pore-

803 scale to core-scale study of capillay desaturation curves ulti multi-scale 3D imaging In

804 International Symposium of the Society of Core Analysts, Napa Valley, California, USA,

805 2013.

38
 39

806 67. Ding, M.; Kantzas, A., Capillary number correlations for gas-liquid systems. J Can

807 Petrol Technol 2007, 46 (27-32).

808

809

39
 40

810 Supplementary Information

811 Pore-scale imaging of hydrogen displacement and trapping in porous

812 media
813

814 Eike M. Thaysen2*, Ian Butler1, Damien Freitas1, Aliakbar Hassanpouryouzband1, Fernando

815 Alvarez-Borges2, Samuel C. Krevor3, Niklas Heinemann1, Robert Atwood2, Katriona

816 Edlmann1

1
817 School of Geoscience, Grant Institute, The King's Buildings, The University of Edinburgh, James Hutton Road,

818 Edinburgh, EH9 3FE, United Kingdom

2
819 Diamond Light Source Ltd, Harwell Campus, Didcot OX11 0DE, UK

3
820 Department of Earth Science and Engineering, Imperial College London, London SW7 2AZ, United Kingdom

821
822

40
 41

823
824 SI Figure S1: Image processing workflow for UoE experiments. (a) The image after reconstruction and filtering.
825 (b) The image after segmenting using a global threshold. (c) The outer part of the rock core touching the Al foil
826 was removed through application of a mask, and a chamfer distance map of the pore space was calculated by
827 assigning each voxel the distance from it to the nearest pore wall. (d) This was then used to separate the pore
828 space into individual pores by calculating the water-shed basins of the distance map.

829

41
 42

830

831 SI Figure S2: H2 stability at 7 MPa injection pressure and a flow rate of 20 µl min-1 (capillary number 2*10-8) at
832 time zero (a) and after ten hours (b). Red circles indicate subtle differences in the hydrogen filled pore volume
833 after ten hours. The difference in the hydrogen saturation for the total scanned rock volume was within error at
834 49.55% % and 49.53% for time 0 and after ten hours, respectively.
835

836

837

838

839

840

841

842

843

844

845

846

847

848

849

850

851

42
 43

SI Figure S3: Enlarged view of 3D renderings of H2 with saturation percentages in UoE experiments. Discrete clusters were rendered in colours, where mainly one colour marks
one large, connected cluster and different colours indicate several, not connected clusters. (a-d) Effect of pore fluid pressure on H2 clusters and saturation after drainage and after
primary imbibition. (a) 2 MPa, (b) 5 MPa and (c-e) 7 MPa, all at a constant flow rate of 20 µl min-1 corresponding to capillary numbers of 1.7x10-8 and 2.4x10-6, respectively..
Experiments were repeated once. For experiments at 2 MPa and 5 MPa averages and standard errors for the H 2 saturation are reported. For experiments at 7 MPa, due to the
discrepancy in the results, both results are visualized (c-d).

43
 44

SI Figure S4: Cyclic H2 injection into Clashach sandstone at 5 MPa and 80 µl min-1 corresponding to a capillary
number of 9.4x10-6, showing no significant differences in H2 saturation and H2 connectivity after primary drainage
and imbibition as compared to after secondary drainage and imbibition.

44
 45

SI Figure S5: Comparison of the distribution of the H2 clusters in the pore space between simultaneous H2 and
brine injection experiments (a-c) and the two repetitions of the non-steady state H2 injection (d) and (e), all at 5
MPa and a total flow rate of 20 µl min-1.

45
 46

SI Figure S6: Nitrogen clusters and saturations during drainage and imbibition at 5 MPa pore fluid pressure and a flowrate of 20 µl min-1

46

View publication stats