Elsevier

Applied Geochemistry

Volume 106, July 2019, Pages 120-133
Applied Geochemistry

Gaseous carbonation of cementitious backfill for geological disposal of radioactive waste: Nirex Reference Vault Backfill

https://doi.org/10.1016/j.apgeochem.2019.04.020Get rights and content

Highlights

  • Carbonation of a potential backfill resulted in fully, partially and uncarbonated zones.

  • Potassium (and to a lesser extent sodium) were concentrated in the carbonated region.

  • Sulphur and aluminium were found to be enriched just ahead of the carbonation front.

  • Carbonation impacted the porosity and mineral assemblage of the backfill material.

Abstract

The ability of Nirex Reference Vault Backfill (NRVB), a cement backfill material, to capture carbon dioxide from Intermediate Level Radioactive waste packages after repository backfilling, has been assessed. Large-scale trials assessed the physical and chemical reaction of carbon dioxide with the hardened backfill grout. A carbonation front, radial in nature, was observed extending into the grout and three distinct regions were identified in the hardened grouts. A carbonated region, a carbonation front, and a partially carbonated zone were discerned. Potassium, and to a lesser extent sodium, were concentrated in the carbonated region just behind of the main reaction front. The area just ahead of the carbonation front was enriched in both sulphur and aluminium, while sulphur was found to be depleted from the carbonated material behind the main reaction front. Within the main carbonated region, virtually all of the hydrated cement phases were found to be carbonated, and carbonation extended throughout the grout, even within material indicated by phenolphthalein solution to be uncarbonated. Importantly, carbonation was observed to impact both the mineral assemblage and porosity of the cement backfill; it is therefore important to understand these characteristics in terms of the long term evolution of NRVB and its groundwater buffering safety function within the geological disposal facility near-field.

Introduction

Geological disposal in an engineered facility underground is the UK Government's policy for disposal of higher activity radioactive waste. (Department of Energy and Climate Change, 2014). Such geological disposal facilities (GDFs) or repositories are based on the use of a multi barrier containment approach, which involves the application of engineered barriers, working in combination with natural geological features, to reduce the rate of radionuclide release to the biosphere. In the UK, an illustrative concept for disposal of Intermediate Level Waste (ILW) in a fractured crystalline rock (e.g. granitic), is that packages of grouted waste will be emplaced in sub-surface vaults and surrounded with a Portland cement-based backfill called Nirex Reference Vault Backfill (NRVB) (Pusch et al., 2017).

The bulk of gaseous emissions from ILW packages are expected to be hydrogen (H2), mainly produced from the corrosion of metallic waste products and methane (CH4) and carbon dioxide (CO2), produced via the microbial degradation of organic waste materials under anaerobic or aerobic conditions (Radioactive Waste Management Limited, 2016a; Amec Foster Wheeler, 2017). A small proportion of the gas produced would include tritium (3H), 14C species (including 14CH4 and 14CO2) and radon (Rn-222). It is desired that, after backfilling, either the cementitious material in the waste packages, or in the backfill, would capture any waste CO2 (including 14CO2), thus retarding its egress to the geosphere (Hoch et al., 2016).

In ordinary Portland cements used for construction purposes, CO2 from the atmosphere diffuses through gas-filled pores and dissolves into the pore solution forming aqueous HCO3. The uptake of acidic CO2 into the alkaline pore solution reduces the internal pH of the binder, and the dissolved carbonate also reacts with calcium-rich hydration products present in the matrix, mainly with portlandite (Ca(OH)2), calcium silicate hydrate (C–S–H1) and the various calcium aluminate hydrates present, to form solid calcium carbonates, silica gel and hydrated aluminium and iron oxides (Johannesson and Utgenannt, 2001; Živica and Bajza, 2001; Fernández-Bertos et al., 2004).

The effect of carbonation on the mineralogy and porosity, are the two most important characteristics that influence the ability of a cementitious backfill grout to buffer groundwater to high pH (as desired to retard release of radionuclides to the geosphere), (cf. Nuclear Decommissioning Authority. 2010; Wilson et al., 2017), and the effect has not been fully elucidated. The reaction of CO2 in the gas phase with typical waste encapsulation grouts and NRVB, in unsaturated conditions, has previously been studied in laboratory experiments e.g. Harris et al., (2003a; 2003b) and; Sun (2010) Carbonation was considered to be associated principally with the portlandite and C–S–H components, facilitated by the dissolution of CO2 into films of water condensing on the cement phases and consequently, the reaction rate is strongly influenced by relative humidity (Bamforth et al., (2012), More recently, experimental studies simulating saturated repository environments have been undertaken to examine the effect of carbonation on gas-transport properties of NRVB (Rochelle et al., 2013; Purser et al., 2015). There has also been significant interest in studying cement carbonation in deep saline brine groundwater environments to evaluate the performance of cementitious well seals in relation to the geological sequestration of CO2 (e.g. Kutchko et al., 2007; Rochelle et al., 2009; Wilson et al., 2011; Rochelle and Milodowski, 2013). Natural analogue studies provide further insights into the long-term effects of carbonation on cementitious materials (Milodowski et al., 2011; Pitty and Alexander, 2011), and indicate that both the processes and degree of carbonation depend upon the geological environment and the partial pressure of CO2 in the groundwater (Bamforth et al., 2012). However further information is required to understand this degradation mechanism in the context of the post-closure performance of a GDF. We report here on detailed experimental studies investigating the reaction of gaseous CO2 with hardened NRVB to support understanding of the backfill material following closure of the GDF.

It is important to revisit the chemistry and engineering performance of NRVB at this time, as the understanding of its role and function within a GDF have evolved since its initial formulation as the UK opens its siting process for such a facility.

Section snippets

Materials

Nirex Reference Vault Backfill (NRVB) was formulated according to (Francis et al., 1997) using Ribblesdale Sellafield specification Portland Cement (Hanson Cement) (Cann and Orr, 2010), limestone flour (Tendley Quarries (BSI, 2005)) and hydrated lime (Limbux hydrated lime supplied by Tarmac Buxton Lime and Cement (BSI and BSI, 2015)). Characterisation of each of these materials (e.g. particle size, surface area, composition etc. are provided in Vasconcelos et al. (2018)). The formulation of the

Carbonation front and visual characteristics

Staining (later confirmed by micro-focus techniques described in the characterisation sections below) was used to identify three distinct regions within each core of hardened NRVB: i) uncarbonated; ii) partially carbonated; and iii) carbonated regions. The partially carbonated region was observed as a distinct interface between uncarbonated and carbonated material, and was likely to be where a carbonation front was located; these three regions are each analysed in detail below. For each core,

Discussion

Using LA-ICP-MS, micro-tomography and EDXA, three distinct regions were identified in each sample, 1) carbonated, 2) partially carbonated and 3) uncarbonated. A carbonation front and a partially carbonated zone were identified in the partially carbonated region. Analysis by LA-ICP-MS and EDXA showed that K and Na were concentrated within the carbonated regions behind the main reaction front, and the concentration was greatest in a narrow zone up to 1–2 mm wide immediately behind where the Ca

Conclusions

The main conclusions resulting from this work are:

  • For the high lime and limestone with Portland cement NRVB grouts, while carbonation leads to the formation of a distinct carbonation front, there was clear evidence of partial carbonation occurring well beyond the main reaction front.

  • Three distinct regions were identified in the hardened NRVB grouts; carbonated, partially carbonated and uncarbonated. Within the partially carbonated region, a carbonation front and a partially carbonated zone were

Acknowledgements

The authors acknowledge the contribution of S Williams at Radioactive Waste Management in funding the experimental work that led to the preparation of this paper. The X-ray microtomography work was conducted at the Manchester X-ray Imaging Facility, and the authors thank the technical staff of that facility for their assistance in scanning the samples and calculating reconstructions. We thank Neil Bramall for assistance with LA-ICP-MS measurements. AEM, LPF, SJK, IM and AB publish with the

References (51)

  • A. Morandeau et al.

    Investigation of the carbonation mechanism of CH and CSH in terms of kinetics, microstructure changes and moisture properties

    Cement Concr. Res.

    (2014)
  • T. Nishikawa et al.

    Decomposition of synthesized ettringite by carbonation

    Cement Concr. Res.

    (1992)
  • J.L. Provis et al.

    X-ray microtomography shows pore structure and tortuosity in alkali-activated binders

    Cement Concr. Res.

    (2012)
  • C.A. Rochelle et al.

    Carbonation of borehole seals: comparing evidence from short-term laboratory experiments and long-term natural analogues

    Appl. Geochem.

    (2013)
  • R.G.W. Vasconcelos et al.

    Characterisation of a high pH cement backfill for the geological disposal of nuclear waste: the Nirex Reference Vault Backfill

    Appl. Geochem.

    (2018)
  • G. Villain et al.

    Measurement methods of carbonation profiles in concrete: thermogravimetry, chemical analysis and gammadensimetry

    Cement Concr. Res.

    (2007)
  • J.C. Wilson et al.

    Fully coupled modeling of long term cement well seal stability in the presence of CO2

    Energy Procedia

    (2011)
  • V. Živica et al.

    Acidic attack of cement based materials — a review: Part 1. Principle of acidic attack

    Constr. Build. Mater.

    (2001)
  • Amec Foster Wheeler

    “Screening Limit” on the Rate of Gas Generation from Individual Intermediate-Level Waste Packages

    (2017)
  • C. Biagioni et al.

    The tobermorite supergroup: a new nomenclature

    Mineral. Mag.

    (2015)
  • P.B. Bamforth et al.

    Cement Materials for Use as Backfill, Sealing and Structural Materials in Geological Disposal Concepts. A Review of Current Status. Serco Report, SERCO/005125/001 Issue 3

    (2012)
  • L. Black

    Raman spectroscopy of cementitious materials

  • BSI

    BS 594-1:2005, Hot Rolled Asphalt for Roads and Other Paved Areas. Specification for Constituent Materials and Asphalt Mixtures

    (2005)
  • BSI

    BS EN 459-1:2015, Building Lime. Definitions, Specifications and Conformity Criteria

    (April 2015)
  • G.C. Cann et al.

    A Technical Specification for Portland Cement, Blast Furnace Slag and Fly Ash for Use in the Encapsulation/immobilisation of Radioactive Waste Materials (8th Revision)

    (2010)
  • Cited by (7)

    • Combined LA-ICP-MS and SEM-EDX analyses for spatially resolved major, minor and trace element detection in cement clinker phases

      2022, Cement and Concrete Research
      Citation Excerpt :

      Thereby, we demonstrate the preferred incorporation of various minor and trace elements into individual clinkers phases of a real technical clinker. Collier et al. used EDX and LA-ICP-MS next to other methods to determine changes in the chemical composition of hydrated cement due to carbonation [51], but the experiments were not performed on the same area. Additionally, the individual methods seem to be promising to determine the provenance of cement [52].

    • The role of environmental conditions on the carbonation of an alkali-activated cementitious waste form

      2022, Cement and Concrete Research
      Citation Excerpt :

      These results indicate that sodium in the carbonated region moves toward the sample surface during aging and precipitates in the form of Na2SO4. The accumulation of Na behind the carbonation front and the depletion of S from the carbonated region were also observed in a backfill material mixed by OPC, limestone, and lime through laser ablation-inductively coupled-mass spectroscopy [67]. Two reasons might contribute to the migration of sodium.

    • Cenozoic to modern-day source to sink systems of Senegal: A record of provenance, transport, recycling and climate controls

      2021, Journal of African Earth Sciences
      Citation Excerpt :

      Representative subsamples of the dried sands were ball-milled to reduce particle size to <125 μm. Subsamples of the <125 μm material were micronised under water for 10 min with 10% corundum (American Elements-PN:AL-OY-03-P) using a McCrone micronising mill for bulk-rock mineralogical quantification (Kemp et al., 2016b; Collier et al., 2019). Bulk-rock X-ray Diffraction (XRD) analysis was carried out using a PANalytical X'Pert Pro series diffractometer equipped with a cobalt target tube, X'Celerator detector, and operated at 45 kV and 40 mA.

    • Evidence of self-sealing in wellbore cement under geologic CO<inf>2</inf> storage conditions by micro-computed tomography (CT), scanning electron microscopy (SEM) and Raman observations

      2021, Applied Geochemistry
      Citation Excerpt :

      Ultimately, this intermediate reaction product can react further with CO2 to produce CaCO3, alumina gel and gibbsite (Rochelle and Milodowski, 2013). For ettringite, it decomposes upon carbonation and S and Al in ettringite migrate towards the uncarbonated material (Collier et al., 2019). The aforementioned reactions govern cement integrity evolution when exposed to high concentration CO2.

    View all citing articles on Scopus
    View full text