Lessons learnt from the impact of volcaniclastic rocks on hydrocarbon systems and their application to CO₂ storage

Volcaniclastic rocks are generally considered to make poor hydrocarbon reservoirs. This is because they contain an abundance of mineralogically and mechanically unstable components that during burial generally lead to a quicker reduction in reservoir quality compared to “clean” siliciclastic rocks. Under favourable conditions however, such as fracturing and/or dissolution of unstable particles, volcaniclastic rocks may act as reservoirs. Indeed, volcaniclastic reservoirs are known, for example, from Japan, Indonesia and Georgia. Alternatively, secondary alteration can retard permeability resulting in volcaniclastic rocks acting as seals in conventional or unconventional plays. It is because of their highly reactive nature and suitable compositions that attention is also turning to the potential of volcaniclastic rocks to store CO₂ through carbonation and thereby, permanently fixing carbon with increased storage security.

Despite the roles volcaniclastic rocks can play within hydrocarbon systems or CO₂ storage, there is still a paucity of quantitative studies characterising them. This is commonly due to their complex secondary mineralisation pathways which can occur during diagenesis and burial. Where data does exist, it is difficult to compare like-for-like due to the application of various, often non descriptive, naming schemes. As a result, volcaniclastic rocks are generally considered problematic and are often overlooked or avoided during the development of exploration plays. It is, therefore, imperative to quantify the nature, thickness and architecture of volcaniclastic rocks to better understand their role in hydrocarbon prospectivity and CO₂ sequestration in volcanic-bearing sedimentary basins.

Thicknesses and architectures of interlava volcaniclastic sedimentary rocks have been extensively quantified from three well-exposed onshore volcanic analogues, namely East Greenland, the Faroe Islands, and the Blue Nile and Mekele basins, Ethiopia. The latter, for example, consists of well-developed fluvial–lacustrine interlava sequences locally >200 m in thickness. Field mapping combined with multiple analyses (geochronology, geochemistry, biostratigraphy, petrographic point-counting and conventional heavy mineral analysis) of these Ethiopian interlava sequences is shedding light on the sedimentary systems that operated during volcanism and basin subsidence. This research is helping to predict lithofacies distribution and to contextualise diagenetic pathways.

More than 300 interlava (volcaniclastic and siliciclastic) samples have been collected from the three onshore field analogues. The samples cover a range of maximum burial depths (0–5 km) and the volcanic components vary in composition (mafic–felsic, subalkaline–alkaline) and texture (crystalline volcanic fragments to volcanic glass). Sixty-four volcaniclastic samples, with air permeabilities from 0 to 173 mD, exhibit a significant disparity in their helium (mean: 25.3%) and total optical (mean: 1.8%) porosities. The high proportion of porosity <30 µm (minimum optical resolution) is due to the abundance of secondary clays and zeolites. This is supported by subsequent Mercury Injection Capillary Pressure (MICP) and Quantitative X ray Diffraction (QXRD) analysis. Principal component analysis of these data show that those samples with a greater abundance of zeolites compared to secondary clays (e.g. smectite) have better sealing capacities. Entry pressures for a subset of 33 samples range from 4 to 613 psia that, under surface conditions, could support hydrocarbon column heights of up to 1181 m (3875 ft) and 671 m (2201 ft) for oil and gas, respectively. The corresponding CO₂ columns heights can also be calculated under various subsurface conditions that suggest the volcaniclastic rocks could also act as baffles to CO₂ migration.

The combination of field observations (burial depths, distributions, depositional environments) and analytical techniques (Poro–Perm, MICP, QXRD) is allowing for a holistic assessment of the factors which control mineralisation, and as a consequence, whether the volcaniclastic rocks will act as hydrocarbon reservoirs or seals. Further research will build on these initial results to fully understand the reaction pathways within volcaniclastic rocks and ultimately their interaction with dissolved CO₂ to assess their storage capabilities.