Coolant

Summary of Physical Properties of Typical Coolants
Basic physical properties of materials that are commonly considered as suitable for acting as the coolant in a nuclear reactor are summarised in the second of the following two tables. First the desirable properties for a coolant are outlined in the table below. The figure to the right shows the neutron capture cross-sections for the discussed coolants (except helium) for a range of energies.


 * $$^\dagger $$ Extrapolated from reference [Sobolev 2007]. Strictly speaking the extrapolation is only valid down to the melting point temperature.

Fanning, T., 2007 “Sodium as a Fast Reactor Coolant”, Presentation on behalf of the Nuclear Engineering Group at Argonne National Laboratory http://www.ne.doe.gov/pdfFiles/SodiumCoolant_NRCpresentation.pdf [Accessed 5th Feb 2010]

Azad, A-.M., “Critical temperature of the Lead-Bismuth Eutectic (LBE) Alloy”, Journal of Nuclear Materials, 341 pp. 45-52

Morita, K., Maschek, W., Flad, M., Yamano, H., and Tobita, Y., “Thermophysical Properties of Lead-Bismuth Eutectic Alloy in Reactor Safety Analyses” Journal of Nuclear Science and Technology, 43 pp. 526-536

Stacey, W.M., “Nuclear Reactor Physics” 2nd Edition, Wiley

Sobolev, V., 2007, “Thermophysical Properties of Lead and Lead-Bismuth Eutectic” Journal of Nuclear Material, 362 pp 235-247.

NNDC, Updated October 2009, “&sigma;igma Evaluated Nuclear Data Files (ENDF) Retrieval and Plotting” Version 3.1, National Nuclear Data Center, USA. http://www.nndc.bnl.gov/sigma/ [Accessed 9th March 2010]

Sodium
The melting point of Na is low, reducing the possibility of coolant freezing. The coolant vessel does not need to be pressurised because the Na boiling point is higher than reactor operating temperatures (the boiling point is not as high as it is for Pb or LBE [IAEA 2002]). The large thermal conductivity and specific heat capacity allows for fast heat transfer away from the core. The Na vessel can be designed to ensure passive flow, even if all systems shutdown. No special measures are required to protect stainless steel from Na, they are chemically compatible materials. Impurities in the Na flow can however cause measurable damage and must be monitored. Many of the fission products are soluble in Na. Ultrasound is used to monitor equipment that is immersed in Na. In 2006 the price of nuclear reactor grade Na was $3,400 m-3 [Fanning 2007]. Na is chemically reactive with air and water. It burns, releasing heat when in contact with air or water; in water it may also be explosive as a result of the liberation of hydrogen. The Na aerosols and chemical reaction products stick to surfaces; some of these products will damage equipment and are toxic to human health. Induced radioactivity arises from 23N(n,&gamma;)24Na, which has a &beta;--decay half-life of $$T_{1/2} = 14.97$$ hours, usually a 1.38 or 2.76 MeV &gamma;-ray is emitted promptly following the decay. Due to the intensity of &gamma;-ray radiation, designs for Na cooled reactors have an additional coolant loop, this coolant is also Na but care is taken to ensure it does not become radioactive. In the event that Sodium and the final water coolant mix together, this reduces the probability that the hazard is exacerbated through the mixing of radioactive Sodium with air/water.

Fanning, T., 2007, “Sodium as a Fast Reactor Coolant”, Presentation on behalf of the Nuclear Engineering Group at Argonne National Laboratory http://www.ne.doe.gov/pdfFiles/SodiumCoolant_NRCpresentation.pdf [Accessed 5th Feb 2010]

IAEA, 2002, “Comparative Assessment of Thermophysical and Thermohydraulic Characteristics of Lead, Lead-Bismuth and Sodium Coolants for Fast Reactors” International Atomic Energy Association, IAEA-TECDOC-1289

Lead
The boiling point of lead is very high, it is improbable that it will boil leaving the core exposed. Its melting point is 601 K, this is high enough that it poses a risk of freezing should the reactor temperature drop. It also poses a technical challenge in melting the lead and getting it into the coolant vessel in the first instance. Similarly shutdowns of the reactor may require that the lead be kept in a molten state. This is potentially expensive. It has been suggested that the lead might be heated and initially forced into the coolant vessel through using pressurised helium [Fernández et al. 1996]. This would require 350 kW of electricity over the course of 1 month for 10,000 tonnes of lead. The cost only of the electricity would be ∼£10,000 / month.



The heat transfer properties of lead are not as favourable as for sodium, but they are good. Due to the large size of the nucleus, lead is a particularly poor moderator of neutrons, this is favourable for increasing the transmutation of minor actinides. The high boiling point means its vessel will not require pressurisation. The Energy Amplifier design proposal, which uses lead as its coolant, would correspond to a pressure of approximately 28 bars at the bottom of the coolant vessel, this is low compared to the &#126;160 bars that pressurised water reactors operate at [Fernández et al. 1996]. Four isotopes of lead are stable and naturally occurring. 208Pb makes up 52.4% of natural lead and it has the lowest neutron capture cross-section of the four stable isotopes. It has been proposed that there may be a benefit to operating a lead cooled reactor with 208Pb enriched lead [Khorasanov et al. 2009]. This would improve the neutron economy of the reactor, enabling it to be run with less fuel. Macroscopic analysis of the lead capture cross-section data has shown that pure 208Pb would capture &#126;4 times less neutrons than natural lead, see teh figure to the right for the neutron capture cross sections of natural Pb and pure 208Pb. The 2005 price of lead was $15,000 m3 [Fanning 2007].

Lead is chemically inert. The intensity of induced radioactivity is low Neutron activation of 204Pb leads to the creation of 205Pb, which decays into 205Tl with a half-life of $$T_{1/2} = 17$$ million years. The decay energy is low and no &gamma;-ray emission follows it. The creation of 210Po is a potential hazard (the section on lead-bismuth eutectic for details on hazards and managing 210Po). In lead the dominant mechanisms for its population are:


 * 208Pb(n,&gamma;)209Pb $$\rightarrow$$ &beta;-($$T_{1/2} = 3.25$$ hours) $$\rightarrow$$ 209Bi(n,&gamma;)210Bi $$\rightarrow$$ &beta;-($$T_{1/2} = 5$$ days) $$\rightarrow$$ 210Po


 * 208Pb(n,&gamma;)209Pb(n,&gamma;)210Pb $$\rightarrow$$ &beta;-($$T_{1/2} = 22.2$$ years) $$\rightarrow$$ 210Bi $$\rightarrow$$ &beta;-($$T_{1/2} = 5$$ days) $$\rightarrow$$ 210Po

Using 208Pb enriched lead will increase the amount of 210Po produced in the coolant. A key issue facing the use of flowing lead as a coolant is that it is corrosive to and erodes the stainless steel structure of the reactor. The rate at which corrosion and erosion take place is increased with higher temperatures of lead and increased velocity of its flow [IAEA 2002]. Increased corrosion at high temperature is in opposition to the desire to operate the reactor at high temperatures to improve its heat transfer efficiency and therefore revenue. Lower coolant velocities require a larger flow area than higher velocities in order to achieve the same heat extraction rate. Large quantities of lead will therefore be required, which will increase costs. Pitting is a particularly challenging feature of steel corrosion due to lead (or LBE) [Roussanov et al. 2003]. Pitting is the creation of multiple local but separate corrosion-erosion centres. Pitting can be difficult to detect as it causes little surface damage or material loss, however it can extend deep beneath the surface of the steel and therefore potentially cause significant damage.

Preventative action can be taken to reduce the corrosion and erosion affects of a liquid lead (or LBE) flow against stainless steel [Zelenski et al. 2007]. The inclusion of oxygen in the coolant will lead to the creation of a ferrite oxide layer on the surface of the steel, which protects against the dissolution of steel. The thickness of the oxide layer that is formed is dependent on the composition of steel that is chosen. The lead coolant will have to be dissolved from fuel assemblies when they are removed from the reactor using acid.

Fernández, R., Mandrillon, P., Rubbia, C. and Rubio, J.A., 1996, “A preliminary Estimate of the Economic Impact of the Energy Amplifier” European Organisation for Nuclear Research CERN/LHC/96-01 (EET) Zelenskii, G.K., Ioltukhovskii, A.G., Leont’eva Smirnova, M.N., Naumenko, I.A. and Tolkachenko, S.A., 2007, “Corrosion Resistance of Fuel Element Steel Cladding in a Lead Coolant” Metal Science and Heat Treatment, 49, pp 533 538

Roussanov, A., Toryanov, V., Jachmenev, G. and Demishonkov, A., 2003, “Corrosion Resistance of Structure Materials in Lead Coolant with Reference to Reactor Installation BREST OD 300” In: “Power reactors and sub-critical blanket systems with lead and lead–bismuth as coolant and/or target material”, International Atomic Energy Agency, IAEA-TECDOC-1348, pp. 113 115

Fanning, T., 2007, “Sodium as a Fast Reactor Coolant”, Presentation on behalf of the Nuclear Engineering Group at Argonne National Laboratory http://www.ne.doe.gov/pdfFiles/SodiumCoolant_NRCpresentation.pdf [Accessed 5th Feb 2010]

IAEA, 2002, “Comparative assessment of thermophysical and thermohydraulic characteristics of lead, lead-bismuth and sodium coolants for fast reactors”, International Atomic Energy Agency, IAEA TECDOC-1289

Khorasanov, G.L., Korobeynikov, V.V., Ivanov, A.P. and Blokhon, A.I., 2009, “Minimisation of an initial fast reactor uranium-plutonium load by using enriched lead-208 as a coolant”, Nuclear Engineering and Design 239, pp. 1703 1707 (2009)

Lead-Bismuth Eutectic (LBE)
The term eutectic refers to the lowest possible temperature of solidification for any mixture of specified constituents. It is an alloy whose melting point is lower than that of any other alloy composed of the same constituents. For lead-bismuth, the eutectic alloy is reached with a mix of 44.5% lead and 55.5% bismuth [Sobolev 2007].

LBE shares many similar properties as a coolant as Lead. It has a high boiling point, similar heat transfer properties, is a very poor moderator and causes similar corrosion affects (which have similar solutions). There are some key differences. A benefit is that the melting point of LBE is 202 K lower than that of Lead, which greatly reduces the risk of the coolant freezing. LBE suffers a greater degree of induced radioactivity than Lead. This is because it has a greater propensity to produce 210Po:

209Bi(n,&gamma;)210Bi $$\rightarrow$$ &beta;-($$T_{1/2} = 5$$ days) $$\rightarrow$$ 210Po

The induced activity of LBE over the long-term is dominated by 209Bi(n,&gamma;)210mBi. This meta-stable state of 210Bi decays by &alpha;-particle emission with a half-life of $$T_{1/2} = 3.6 \times 10^6$$ years. To a lesser extent, there is also a long-lived &beta;--decay contribution ($$T_{1/2} = 3.68 \times 10 ^5$$ years) from 208Bi, which is created in the reaction 209Bi(n,2n)208Bi.

210Po decays by the emission of up to 5.3 MeV &alpha;-particles, with an associated half-life of $$T_{1/2} = 138$$ days. One in 100,000 decays emit an 803 keV &gamma;-ray promptly following the &alpha; particle emission (the &alpha;-particle energy of such decays is correspondingly reduced to 4.5 MeV). Polonium has a melting point of 527 K and a boiling point of 1235 K, however it evaporates at temperatures beneath its boiling point. As an &alpha; particle emitter, 210Po becomes most hazardous when ingested. When the coolant is sealed it presents no danger. During maintenance that requires opening seals or through accidental coolant spillages the risks of 210 Po must be considered [Pankratov et al. 2004]. Polonium adheres to surfaces and will also enter the air as an aerosol. R&D is taking place in the development of metallic wire mesh filters [Obara et al. 2008] that take advantage of the adhesive properties of polonium and capture it as it evaporates from the coolant. Related research has identified that baking quartz contaminated with 210Po at 300 °C will remove nearly 80% of the 210Po in 5 minutes, without removing other non-radioactive adhesives [Obara et al. 2005, Obara et al. 2004]. For stainless steel, at 500 °C 40% is removed in 30 minutes and 100% for the same duration at 600 °C, again without removing other non-radioactive adhesives. It is therefore being considered that, before accessing areas of a power station where 210Po is present, the equipment should be baked to release polonium into the air, where it is captured by metallic meshes.

Bismuth is a rare material in the Earth’s crust. Including currently profitable and currently unprofitable reserves, deposits of 680 thousand tons of Bismuth have been identified around the world. In 2005 world production of Bismuth was approximately 5000 tons [Naumov 2007]. In December 2005 the Chinese price for bismuth (China provide half the world’s annual production) was $10,000 per ton, which corresponds to $9,780 m-3. However the price for reactor grade bismuth is considerably higher than this, at $86,000 m-3 (in 2005) [Fanning 2007]. Depending on its design specifications, a single LBE cooled reactor will require 5-15 thousand tons of bismuth, constructing multiple LBE cooled reactors will require a significant fraction of the world bismuth supply, not considering the demands of other industries that require the material.

Sobolev, V., 2007, “Thermophysical Properties of Lead and Lead-Bismuth Eutectic” Journal of Nuclear Material, 362 pp 235-247.

Pankratov, D.V., Efimov, E.I., Toshinskii, G. I. and Ryabara, L.D., 2004, “Analysis of the Polonium Hazard in Nuclear Power Systems with Lead-Bismuth Coolant” Atomic Energy, 97 pp. 559-563

Obara, T., Koga, T., Miura, T. and Sekimoto, H., 2008, “Polonium Evaporation and Adhesion Experiments for the Development of Polonium Filter in Lead-Bismuth Cooled Reactors” Progress in Nuclear Energy, 50, pp 556-559

Obara, T., Mura, T and Sekimoto, H., 2005, “Fundamental Study of Polonium Contamination by Neutron Irradiated Lead-Bismuth Eutectic” Journal of Nuclear Materials, 343, pp. 297-301

Obara, T., Miura, T. and Sekimoto, H., 2004, “Development of Polonium Surface Contamination Measure in Lead-Bismuth Eutectic Coolant” Presentation at the International Symposium on Innovative Nuclear Energy Systems, Tokyo, Japan, October 31st – November 4th 2004

Naumov, A.V., “World Market of Bismuth: A Review” Russian Journal of Non-Ferrous Metals, 48, pp 10-16

Fanning, T., 2007, “Sodium as a Fast Reactor Coolant”, Presentation on behalf of the Nuclear Engineering Group at Argonne National Laboratory http://www.ne.doe.gov/pdfFiles/SodiumCoolant_NRCpresentation.pdf [Accessed 5th Feb 2010]

Carbon Dioxide (CO2)
A drawback of gas coolants, such as carbon dioxide (or helium), is that they rely on the coolant being continually pressurised and pumped. The pressure at which CO2 must be pumped is appreciably lower than for helium, in this respect CO2 is more favourable. In the absence of both pressurisation and pumping there is low thermal inertia driving the flow and thus the core temperature increase rapidly [Handwerk 2007].

Supercritical CO2 has been considered as a coolant for gas-cooled reactors [Hajzlar et al. 2005]. The critical temperature and pressure of carbon dioxide are 304 K and 74 bars. Supercritical CO2 will provide improved thermal efficiency and enable direct cycle heat transfer, both of which are of economic benefit. To reach comparable thermo-physical properties to helium, supercritical CO2 operates at ~ 930 K (c.f. ~ 1130 K) and at 200 bars (c.f. 80 bars). High pressure and medium temperature are less challenging than medium pressure high temperature, due to the stress constraints at high temperature. The greater molecular mass of CO2 compared to helium leads to longer a longer depressurisation time if the pressure vessel fails.

Handwerk, C.S., 2007, “Optimized Core Design of a Supercritical Carbon Dioxide-Cooled Fast Reactor” PhD Thesis Massachusetts Institute of Technology, Department of Nuclear Science and Engineering

Hejzlar, P., Dostal, V., Driscoll, M.J., Dumaz, P., Poulennec, G. and Alpy, N., 2005 “Assessment of Gas Cooled Fast Reactor with Indirect Supercritical CO2 Cycle” Proceedings of the International Congress on Advanced Nuclear Power Plants, Seoul, Korea, Paper 5090

Helium (He)
Helium has a high specific heat capacity, but a very low density; it therefore needs to be kept at high pressure (85 bars) and must flow at a high velocity (~ 100 ms-1). It has a low thermal conductivity. It is highly radiation stable. Helium is chemically inert, even at very high temperatures, hence its use in the High-Temperature gas-cooled Reactor (HTR). The HTR is designed to operate at over 1100 K. The high operating temperature allows for a high thermal efficiency when converting heat into electrical energy. Although helium is not chemically reactive, impurities in the coolant flow tend to be. In a real system, small leaks into the helium circuit allow for water vapour or other materials to enter the flow [Collier and Hewitt 1987].

The low molecular weight of helium leads to losses from the pressure vessel through diffusion at seals and valves. The neutron capture cross-section of helium is extremely low, it therefore does not moderate neutrons or worsen the neutron economy. Its neutron transparency does however, lead to neutron losses from the reactor geometry. The 2005 price of helium was approximately $220 m-3 at 85 bars [Fanning 2007].

Collier, J.G. & Hewitt, G.F, 1987, “Introduction to Nuclear Power” Hemisphere Publishing Corporation

Fanning, T., 2007, “Sodium as a Fast Reactor Coolant”, Presentation on behalf of the Nuclear Engineering Group at Argonne National Laboratory http://www.ne.doe.gov/pdfFiles/SodiumCoolant_NRCpresentation.pdf [Accessed 5th Feb 2010]

Water (H2O)
Because water contains hydrogen, with an atomic weight of 1, it is a very good moderator of neutrons. Therefore even in the absence of a dedicated moderator, it is not suitable for cooling a fast reactor as it will thermalise the neutrons. If an ADSR were to be designed to operate in the thermal neutron energy spectrum, then water can be considered as the coolant. There are three forms in which light water may be used in a reactor: pressurised, boiling and supercritical. Pressurised and boiling water reactors are commonplace among nuclear reactors in operation today. Supercritical water reactors make up one of the Generation IV designs. Pressurised water reactors are maintained at pressures of &#126; 150 bars, the water does not boil (in any significant quantity) at the reactor operating temperature, which is approximately 600 K. The pressurised water transfers the heat from the core to a second water circuit, which drives the turbine. In a boiling water reactor the pressure is lower, &#126; 75 bars. When heated to ~ 550 K the water boils and is directly used to drive the turbine, it is then condensed and sent back into the core.

Supercritical water reactors are intended to be operated above the critical temperature-pressure (648 K and 218 bars) of water (above the critical temperature of a material there is no distinction between its liquid or gaseous states). This is higher than for pressurised and boiling water reactors. The high operating temperature of supercritical water will allow for high thermal efficiency. It is also possible to directly send the supercritical water coolant through the turbine. Supercritical water-cooled poses materials challenges, particularly inside the nuclear core, which must withstand high temperatures, high pressures intense energy arising from thermochemical reactions and irradiation [Baindur 2008].

Baindur, S., 2008, “Material Challenges for the Supercritical Water-Cooled Reactor (SCWR)” Bulletin of the Canadian Nuclear Society, 29, pp 32-38