Structure and Accelerated Carbonation of Mature Cement Matrices for immobilisation Paulo Borges P'bo

1
Structure and Accelerated Carbonation of Mature
Cement Matrices for immobilisationPaulo Borges
(P.borges_at_sheffield.ac.uk) Dr. Neil B.
Milestone Dr. Cyril LynsdaleUniversity of
Sheffield- Department of Engineering Materials
Sir Robert Hadfield Building Mappin Street
Sheffield S1 3JD United Kingdom

• INTRODUCTION
• Blended cement pastes containing pulverized fuel
  ash (PFA) and blast furnace slag (BFS) are used
  in the immobilisation of low level waste (LLW)
  and intermediate level radioactive waste (ILW).
• Up to 90 BFS and 75 PFA are currently used in
  blended pastes to limit the evolution of heat
  during hydration, to avoid cracking due to
  thermal gradients during cooling and to maintain
  integrity of the matrix.
• It is expected that those pastes will have long
  term durability (hundred of years) and
  accelerated tests must be carried out to assess
  changes in their structure and properties.

• EXPERIMENTAL / RESULTS

• EXPERIMENTAL / RESULTS
• Table 1 shows the formulations used. OPC is a
  reference sample. 31 PFAOPC and 91 BFSOPC are
  formulations currently used in immobilisation.
  9S1CA is the same 91 BFSOPC formulation now
  activated with a sodium silicate solution. The
  last formulation is a 100 activated PFA. Sodium
  silicate and sodium hydroxide have been used in a
  water / solids ratio of 0.30. The choice of one
  formulation will depend on some fluidity results
  and an analysis of their microstructure (SEM).
• Blended cement pastes were cured at two
  different curing regimes (T1 up to 40 degrees,
  and and T2, up to 80 degrees) with relative
  humidity gt 90. Activated PFA (formulations FA1
  and FA2), were cured at 60ºC for 24 hours and
  then prepared for SEM.

• Figure 4(a) shows a BEI of 91 BFSOPC system
  activated with sodium silicate after 10 months of
  hydration. It is a mature paste and most of the
  reactions have already taken place. However, it
  is possible to see that a large number of BFS
  particles did not react (lighter grey grains).

Table I Proposed formulation of pastes for
assessment of durability
NEW APROACH
4 (a) - 9S1CA-T2 (10 months)

• Figure 4(b) is a SEI of the 31 PFAOPC system
  after 10 months of hydration. The microstructure
  seems very compact. It shows that most of PFA
  particles have reacted and much C-S-H was formed
  although 75 of solids were PFA and only 25 was
  cement.

• Other systems can be used for immobilisation
  since they show good chemical and physical
  stability.

• Alkaline solutions in contact with
  aluminosilicates lead to the formation of two
  different structures, depending on the amount of
  calcium available (Figure 1).
• When mild or high calcium is available (BFS,
  blended cements) a modified calcium silicate
  hydrate (C-S-H) structure is formed.
• If low calcium content is available (metakaolin
  and PFA) the high alkaline environment
  disintegrates the aluminosilicates network to
  form an amorphous aluminosilicate binder commonly
  known as geopolymer (Figure 2).

4 (b) - 3F1C-T2 (10 months)
Figure 4 SEM image of mature blended pastes

• Accelerated carbonation was carried out in 10
  months old samples using 15 CO2, 60 R.H. and
  25C. Samples were round cylinders of diameter
  14 mm (Figure 5).

Figure 5 9S1C-T2 after 13 days of accelerated
carbonation showing carbonated (A) and non
carbonated (B) zones
Figure 1 Activation of aluminosilicates

• Figure 6 shows that, as expected, neat OPC has
  the lowest carbonation front due to its low
  permeability.
• Interestingly, activated 91 BFSOPC has higher
  carbonation than its non activated analogue. This
  indicates that the activation of 91 BFSOPC
  might not be a good solution for improving
  durability.
• 31 PFAOPC paste carbonated completely after
  seven days of test. This indicates that,
  comparatively, this paste is the most vulnerable
  to carbonation.

• A simple activation of the 91 BFSOPC system
  currently used in immobilisation could be an
  example of the former. In this case, the
  activation could improve the reactivity of BFS
  particles.
• Geopolymers show good chemical stability and
  have been studied as alternative materials for
  immobilisation.
• Although both systems could be used,
  immobilisation researches seem to neglect the
  importance of carrying out accelerated tests to
  predict long term durability of such systems.

Figure 2 Structure of geopolymers
Figure 6 Carbonation over time in mature pastes

• AIMS
• This PhD project aims to investigate the long
  term durability of pastes used for immobilisation
  by using a set of accelerated tests. So far, the
  main objectives are
• - Comparison of two different types of activators
  of PFA and their influence in the microstructure.
  
• Assessment of the structure of mature pastes (10
  months old) currently used for immobilisation and
  verify the degree of hydration of pozzolanic
  materials (PFA and BFS).
• Assessment of the performance of such pastes to
  Accelerated Carbonation.

(a) PFA Na2SiO3
(a) PFA NaOH

• FUTURE WORK
• Quantify the extent of the hydration of BFS (or
  remaining unreacted particles) by chemical
  dissolution and by SEM image analysis.
• Based on the individual rates, develop a model
  of prediction of carbonation over time.
• Run other accelerated tests, such as wetting and
  drying and accelerated leaching.

Figure 3 PFA activated with alkaline solution
after 24 hours

• ACKNOWLEDGMENTS
• The authors want to acknowledge Nexia Solutions,
  (formerly BNFL/NSTS) for funding this PhD
  project and the Department of Civil and
  Structural Engineering (University of Sheffield)
  for their support.