Performance rankings were influenced by composition of

Performance
indices of hot liquid sodium exposed sacrificial surface layers in fast breeder
reactors

K. Mohammed Haneefa1*, Manu
Santhanam1, F.C. Parida2

1 Department of Civil
Engineering, IIT Madras, Chennai, India

{mhkolakkadan, manusanthanam}@gmail.com

2 Radialogical Safety
& Environment Group

Indira Gandhi Centre for Atomic Research,
Kalpakkam, Tamilnadu

[email protected]

Abstract. Leakage accidents
in sodium cooled fast breeder reactors trigger various thermo-chemical
degradations of structural materials used for their construction. The
interactions of hot liquid sodium with concrete at around 550 °C and above are
investigated in this paper. Potential materials were designed and tested.
Degradation mechanisms and extend of damages were identified; and subsequently
performance indices were developed. Four different types of cements, eight
different w/c ratios and geopolymer composites were investigated in this
study.  Comprehensive mechanical, physical,
chemical and microstructural characterizations were performed before and after
exposure. Micro-analytical tools such as SEM (SE and BSE), TG/DTA, XRF, XRD and
thin section petrography were used for characterizing the degradation
behaviour. Study revealed that the performance rankings were influenced by
composition of concrete and water to cement ratios employed for conventional
cement-based systems. Performance indices for geopolymer composites were
superior to the conventional cement-based systems in hot liquid sodium hostile
environment.

Keywords: Fast breeder
reactors, hot liquid sodium interactions, sacrificial layer, limestone mortars,
geopolymers

 

1 Introduction

Liquid
sodium is used as a coolant in Fast Breeder Reactors (FBRs). Sodium leakage
accidents in FBRs results in formation of a pool of hot liquid sodium or sodium
spray on Structural Concrete (SC). These interactions (at around 550°C and
above) elicit various thermo-chemical degradations of concrete structures
employed in inert atmosphere (equipment cells or reactor cavity) or in air (containment
and steam generator building). To prevent the SC from deterioration in FBRs, a
sacrificial surface layer is employed over it. Even though various researchers
have studied the hot liquid sodium and concrete interactions, there are still
uncertainties persist in quantifying the fundamental degradation mechanisms 1-9.
The present study summarizes the investigations of Haneefa et al. 10-20 and
reports performance indices for 24 different mixes for hot liquid sodium
exposed sacrificial surface layers in FBRs.  
    

2
Experimental Designs

Four
types of cements were used in the study namely Ordinary Portland Cement (OPC),
Portland Pozzolana Cement (PPC), Portland Slag Cement (PSC) and High Alumina
Cement (HAC). Studies were conducted with a range of water to cement ratios
(w/c) 0.4 to 0.6 for OPC. At a w/c of 0.55, performances of different types of
cements were studied. River sand and limestone aggregates were used in the
study to evaluate their suitability for hot sodium hostile environment.
Different geopolymer composites with varying molarities of NaOH (8M, 12M, 16M
and 18M), fixed solid sodium silicate to NaOH ratio (s/n) of 1.5 and varying
activator fly ash ratios (0.45 and 0.50) were tested to check their suitability
for sacrificial surface layer. Class F fly ash was used for making geopolymers.
The experimental program was divided in to three phases. In the first phase,
the constituent materials used were tested for physical properties. Apart from
that, mineralogical and microstructural characterizations were performed using
micro-analytical techniques. In the second phase, all the constituent materials
and mixes were tested for thermal performance. 
The specimens were exposed to 550 °C (ideal operation conditions of FBRs)
for duration of 30 minutes. The duration was a deemed one based on a maximum
time required to flow away the accidently spilled hot liquid sodium and reach
the collection pits in FBRs 13-15. 
Thermal performance test was conducted in a maffle furnace with an
average rise in temperature of 0.60 °C/s (Fig.1). Hot liquid sodium
exposure studies were conducted at Indira Gandhi Centre for Atomic Research
(IGCAR), Kapakkam, India. Specially designed carbon steel vessels equipped with
dismountable thermal insulations of aluminium cladding (Fig. 2) were used with
1200 W electric surface heater. The temperatures were monitored and controlled
by the help of long and flexible thermocouples. After the required sodium fire
exposures, the samples hung from the top were removed and allowed to cool in
ambient temperatures (Fig. 3 and Fig. 4). Post-test specimens were stored in an
electronic desiccator after cleaning with ethyl alcohol and drying. 

 

Compressive strength, flexural strength as per ASTM C
348 21, mass loss behaviour and abrasion resistance conforming to IS
1237-1980 22 were assessed understand the thermal effects.  Scanning Electron Microscopy (SEM) with back
scattered imaging on polished specimens and secondary electron images on
fracture surface, thin section petrography on 30µm slides using plane and
crossed polarized light, X-Ray Fluorescence (XRF), X-Ray Diffraction and
thermogravimetric differential thermal analysis (TG/DTA) were used to perform a
comprehensive forensic analysis of different mixes after exposure to 550 °C
with and without sodium.

 

3
Results and Discussions

3.1
Thermal performance Indices

Table
1 provides the mixes and their compositions used in the study. Performance
indices were developed based on performances in compressive strength, flexural
strength, mass loss and abrasion resistance after exposure to 550°C for 30
minutes 10-20.  Mixes are then ranked
on a scale 1 to 8, for 1 being the best and 8 being the worst for limestone and
river sand mortars. For the geopolymer mixes the ranking was from 1 to 4 for
pastes and mortars separately. For both the types of aggregate, the cement
mortars with lower w/c ratios performed well in thermal in ranking. Similarly,
among the different types of cement, the performance of Portland pozzolana
cement was better in most of the cases. However, the relative performance of
river sand mortar was inferior to limestone mortars. The reductions in
compressive strengths after thermal exposure were 8.4, 9.1, 11.5, 16.1, 17.4,
12.9, 15.9 and 13.2 % for limestone mortars LS1 to LS8. The corresponding
values for river sand mortars were 12.3, 15.3, 18.1, 23.4, 22.4, 19.0, 19.1 and
17.6% respectively for the mixes RS1 to RS8. Similar trends were observed for
flexural strength and mass loss. However, abrasion resistance of river sand
mortars was better compared to limestone. This effect was resulted from the
mineralogy of river sand and limestone (calcite has a Mohs hardness scale
number of 3, whereas quartz is 7).  Geopolymer
pastes exhibited increase in compressive strengths upon heating at 550°C for 30
minutes.  Hence, for the geopolymer
pastes strength indices were based on absolute values. Among the different
pastes, the mix GM3 with 16 M ranked first. Increment in strength of
geopolymers may be due to the attainment of high level polymerization upon heat
treatment. Similar trends were observed for other indices, except for abrasion.
Further, geopolymer mortars with limestone aggregate were tested. The mixes
with lower molarities ranked higher indicating mixes becoming more brittle with
high molarities of NaOH. Moreover, the mix 18 M was severely cracked and broken
upon thermal exposure. 

3.2 Indices
based on sodium fire performance

Based
on performance indices from thermal study, limestone mortars with different
cement types and w/c, geopolymer pastes and geopolymer mortars were considered
for sodium fire tests. For cement based mortars calcium depletion and sodium
enrichment indices were developed based on XRF data. Apart from these indices,
aluminium depletion index was developed for geopolymer composites. These
indices were calculated as (Reduction/increment in elemental composition after
sodium fire) i / (Average elemental composition before exposure).
Calcium depletion index and sodium enrichment index for limestone aggregate
(LS) were found to be 0.436 and 802.5 respectively. Figure 5 depicts calcium
depletion indices for cement based limestone and limestone mortars after sodium
fire. The indices after sodium fire corroborate the observations from the
thermal performance indices. As the w/c ratio increases, the calcium depletion
indices showed an increasing trend. Similar trends were observed in sodium
enrichment indices. The mixes with higher w/c might have resulted in more
release of free water and subsequent formation of NaOH and hydrogen gas. These
consequences intensify the reaction kinetics during the sodium fire associated
with more calcium depletion and sodium enrichment by possible cationic
exchanges.  Since the present study
simulates the worst case of sodium accidents, the maximum w/c ratio for FBRs
sacrificial surface layer should be equal or less than 0.4 to extend the
reinstatement period of sacrificial surface layers.

Table1
Mixes used in the study and performance indices based on 30 minutes exposure to
550 °C 10-20

 

Figure
7 represents calcium depletion, sodium enrichment and aluminium depletion
indices for geopolymer composites. Unlike the cement based systems, geopolymer
system contains more sodium content. As illustrated in the Figure 7, geopolymer
pastes were less affected by sodium fire. However, the mortars exhibited
significant changes upon sodium fire compared to geopolymer pastes. The
developed indices pronounced that the performance of higher molarity
geopolymers were deficient to the ones with lower concentrations. The elemental
depletion indices after sodium fire of geopolymers revealed that the reaction
kinetics of hot liquid sodium with geopolymer composites were less and
eventually the degradation was minimal compared to conventional cement based systems.
Additionally, From the XRD study; Na4SiO4, Na3SiO3,
Na2CaSiO4, NaOH, Na2CO3 NaAlO2,
Ca(OH)2, Na (Hexagonal) and Na (Cubic) were found as reaction
products.

 

Fig.
7 Major elemental depletion indices for geopolymer composites

Table 2
provides glimpses of TG/DTA analysis (a typical pattern is presented in Fig. 8)
for all sodium fired specimens along with shape retaining indices ({mass of
unaffected inner core after sodium fire i / mass of the specimen
before any exposure} ×100)) and residual strength indices (absolute values of
residual strength after sodium fire in MPa). Among the cement-based systems,
only the mixes with 0.40 and 0.45 w/c ratios exhibited enthalpy changes
corresponding to the decomposition of calcite (CaCO3); which
indirectly portrays that most of the calcite was decomposed in the mortars with
higher w/c upon sodium fire.  These
results resembled that the degradation of mortars was worst at higher water
cement ratios due to transport of hot liquid sodium into the inner core.
Similarly, the formation of Ca (OH)2 was high at higher w/c.   Corroborating
trends were observed in shape retaining indices calculated based on absolute
values of mass retained after sodium fire. There were no inner cores present in
the mixes with 0.50, 0.55 and 0.60 w/c. Due to high level of disintegration residual
strength calculation was not possible for cement based system. The mix with fly
ash blended PPC showed a sign of minor DTA peak 751.7°C related to
calcite decomposition which implies presents of unaffected limestone in the
mortar after sodium fire.

Table
2 Indices based on hot liquid sodium fire

Mix

Enthalpy Change of Calcite (CaCO3)
decomposition

Enthalpy Change of Ca(OH)2 decomposition

Shape Retaining Index

Residual Strength Index

LS1

51.75 J/g

23.90 J/g

91.5

Disintegrated

LS2

12.75 J/g

39.44 J/g   

43.2

Disintegrated

LS3

No complex peak

34.33J/g

No inner core

Disintegrated

LS4

No complex peak

42.36J/g

No inner core

Disintegrated

LS5

No complex peak

55.93J/g

No inner core

Disintegrated

LS6

Minor peak at 751.7°C

35.55J/g

83.7

Disintegrated

LS7

No complex peak

45.91J/g

49.7

Disintegrated

LS8

No complex peak

5.88J/g

41.4

Disintegrated

GP1

Not performed

Not performed

92.3

10.7

GP2

Not performed

Not performed

91.8

17.7

GP3

Not performed

Not performed

90.7

18.8

GP4

Not performed

Not performed

89.1

23.5

GM1

81.53J/g

No complex peak

57.4

22.3

GM2

98.76 J/g

Minor peak at 451.3°C

61.0

23.8

GM3

Not performed

Not performed

46.6

Disintegrated

GM4

Not performed

Not performed

Cracked

Disintegrated

Geopolymers
displayed superior indices based on CaCO3 and Ca(OH)2
enthalpy changes, shape retaining indices and residual strength indices. Among
the TG/DTA performed (the geopolymer mortars with 8M and 12 M NaOH); both the
mixes exhibited distinct complex peaks of 81.53 J/g and 98.76 J/g respectively.
The result proved that the hot liquid sodium had not intruded into the mortars
and decomposes the limestone aggregates. Moreover, there were no any sign of
enthalpy changes in TG/DTA corresponding to Ca(OH)2 decomposition.
Corroborating inferences were drawn from the shape retaining indices and
residual strength indices of geopolymer composites. The minimum residual mass
was 89.1 % for paste phase and 46.6% for mortars. The mix 18M was completely
disintegrated. Meanwhile the maximum changes in strengths were 23.5 % and 22.3%
for paste mortars respectively. Due to the high level of damage, residual
strengths were not able assess for the mortars mixes with 16M and 18M of NaOH.

3.3
General discussions on influence of microstructure changes on performance
indices

This
section describes how the microstructural alterations influence the performance
indices. Fig. 9 represents a thin section image of fire damaged river sand
mortar. Indian river sands are generally composed of weathered granite. The
hypidiomorphic and interlocked texture of granite microstructure was cracked
due to differential thermal expansion and contraction. Ferric oxidation and subsequent
cleavage staining disrupts the mineral assemblage in river sand as seen the in
the Fig. 9. These mechanisms result in more reduction in compressive strength
of river sand mortars compared to the limestone aggregate mortars. Limestone is
a mono-mineral rock with very less accessory impurity minerals in it. Its
thermal stability at 550°C is intact. Figure 10 to 13 provide SEM- back scatter
electron images of polished sodium fired specimens. The geopolymer
microstructure (Figure 10) was less affected upon sodium fire compared to the
conventional cement-based systems (Fig. 11). Moreover, the formation of cracks
in paste phases was more intensive in cement-based systems (Fig. 12) compared
to the geopolymers (Fig. 13). These observations were directly reflected on
performance indices of sodium fired specimens.

 

 

 

 

4.  Conclusions

 

The current study considered 24 different
mixes for sacrificial surface layer to protect structural concrete from hot
liquid sodium fire in FBRs and performance indices were developed based
degradation behaviour. The study recommends the following;

a)      
Use of limestone over river sand or
granite for the sacrificial surface layer

b)      
Use of low w/c ratio concretes for
sacrificial surface layer preferably less than or equal to 0.4

c)      
Effective deployment of geopolymer
technology for sacrificial layers in FBRs

 

Acknowledgement

Partial financial support from Indira Gandhi Centre
for Atomic Research, Kalpakkam, India, for the project is gratefully
acknowledged.

Reference

1
Chasanov, M.G., Staahl, G.E.: High temperature sodium–concrete interactions. J.
Nucl. Mater. 66, 217–220 (1977).

2
Barker, M.G., Gadd, P.G.: A chemical study of the sodium–concrete reactions.
In: Proceedings of the LMFBR Safety Topical Meeting, Lyon-Ecully, France,
PartIII, p. 91 (1982).

3
Fritzke, H.W., Schultheiss, G.F: An experimental study on sodium concrete
interaction and mitigating protective layers. In: 7th International Conference
on Structural Mechanics in Reactor Technology, 135–142 (1983).

4
Muhlestien, L.D., Postma, A.K.: Application of sodium–concrete reaction data on
breeder reactor safety analysis. Nucl. Safety. 25 (2), 212–22 (1984).

5 Chawla, T.C., Pederesen, D.R.: A review of modeling
concepts for sodium–concrete reactions and a model for liquid sodium transport
to the un reacted concrete surface. Nucl. Eng. Des. 88, 85–91 (1985).

6
Schultheiss, G.F., Minden, C.V., Fritzke, H.W.: Method for avoiding or reducing
the interactions and their consequences from contact of hot liquid metallic
sodium with concrete, United States Patent No. 4642300 (1987).

7 Bae,
J.H., Shin, M.S., Min, B.H., Kim, S.M.: Experimental study on sodium–concrete
reactions. J. Korean Nucl. Soc. 30, 568–580 (1998).

8
Premila, M., Sivasubramanian, K., Amarendra, G., Sundar, C.S: Thermo chemical
degradation of limestone aggregate concrete on exposure to sodium fire. J.
Nucl. Mater. 375, 263–269 (2008).

9 Das,
S.K., Sharma, A.K., Parida, F.C., Kashinathan, N., 2009. Experimental study on
thermo-chemical phenomena during interaction of limestone concrete with liquid
sodium under inert atmosphere. Constr. Build. Mater. 23, 3375–3381

10 Haneefa, K.M., Santhanam, M., and
Parida, F.C.: Performance evaluation of sodium resistant mortars as sacrificial
layer in fast breeder reactors,
9th fib International PhD Symposium in Civil
Engineering, Karlsruhe Institute of Technology (KIT)-University, Karlsruhe,
Germany, July 22-24, 2012, 715-721(2012).

11 Haneefa, K.M., Santhanam, M., and
Parida, F.C.: Performance evaluation of limestone mortars for elevated
temperature application in nuclear industry, 3rd International conference on
repair, rehabilitation and retrofitting, ICCRRR-2012, University of Cape Town,
South Africa, September 12-15, 2012, 111-116 (2012).

12
Haneefa, K.M., Santhanam, M., and Parida, F.C.:  Review of Concrete performance at Elevated
Temperature and Hot Sodium Exposure Applications in Nuclear Industry, Nuclear
Engineering and Design, 258, 76-88(2013).

13
Haneefa, K.M., Santhanam, M., and Parida, F.C.: Thermal performance of
limestone mortars for use in sodium cooled fast breeder reactors, Indian
Concrete Journal, 87(12), 25-41(2013).

14
Haneefa, K.M., Santhanam, M., Ramaswamy, R., and Parida, F.C.: Hot sodium
triggered thermo-chemical degradation of concrete aggregates in the
sodium-resistant sacrificial layers of fast breeder reactors, Nuclear
Engineering and Design, ,265, 654-667 (2013).

15
Haneefa, K.M., Santhanam, M., and Parida, F.C.: Performance characterization of
geopolymer composites for hot sodium exposed sacrificial layer in fast breeder
reactors, Nuclear Engineering and Design, 2013, 265, pp.542-553 (2013).

16 Haneefa, K.M., Santhanam, M., and Parida, F.C.: Deterioration
of limestone aggregate mortars by liquid sodium in fast breeder reactor
environment, Nuclear Engineering and Design, 2014, 275, 287-299 (2014).

17 Haneefa, K.M., Santhanam, M., and Parida, F.C.:
Studies on Hot Liquid Sodium and Concrete Interactions in Fast Breeder
Reactors, 2nd International Congress on durability of Concrete, Norwegian
Concrete Association, December 4-6, 2014b, Paper 29 (2014).

18 Haneefa, K.M., Santhanam, M., and
Parida, F.C.: Performance characterization of hot sodium exposed sacrificial
concrete layer for fast breeder reactors, Proceedings of the International
Conference on Advances in Civil Engineering and Chemistry of Innovative
Materials, ACECIM’14, Department of Civil Engineering and Department of
Chemistry, SRM University, Chennai, India, March 13-14, 2014, pp. 842-847 (2014).

19 Haneefa, K.M., Santhanam, M., and
Parida, F.C.: A study on potential materials for hot sodium exposed sacrificial
layer in Fast Breeder Reactors, Proceedings of the National Conference on
Advances in Civil Engineering, ACE2K15, 23rd March, 2015, SSN
College of Engineering, Kalavakkam, Tamilnadu, India,173-181 (2015).

20 Haneefa, K.M., Santhanam, M., and
Parida, F.C.: Development of a forensic methodology for investigation of
concrete structures affected by sodium fires in fast breeder reactors, National Conference on Forensic
Structural Engineering, Vellore Institute of Technology, Chennai, Volume:
Session 1, Paper No.1, 1-18 (2016).

21
ASTM C 348: Standard test method for flexural strength of hydraulic-cement
mortars, ASTM International, U.S.A. (2014)

22 IS 1237:
Cement Concrete Flooring Tiles – Specification, Bureau of Indian Standards, New
Delhi, (1980)