For many years, there has been interest in utilising thorium-232 as a nuclear fuel since it is three to five times as abundant in the Earth’s crust as uranium. The cost of nuclear power generation in the conventional way is on the rise. There is a great pressure from international community to reduce carbon emission. After disasters in Chernobyl, Fukushima, Windscale, Kysthym, etc., nuclear energy carried a dreaded stigma. So many countries are looking for a path to generate energy that is distinctly different, but pays off in the long run. More abundantly available in India than uranium, thorium is cheaper, safer, and greener. Thorium is fertile, means it cannot sustain a chain reaction of its own like uranium or plutonium, yet it can produce fissile material if neutrons are provided from an external source.
A thorium reactor would work by having Th-232 capture a neutron to become Th-233, which decays to uranium-233, which is able to undergo fission as shown in FIG – 1. The process creates energy and once the process starts it sustains of its own turning more thorium-232 nearby into the same nuclear fuel. This route is very attractive to India, as India has meager reserves of uranium, but the known reserve of Indian thorium in the form of monazite in Kerala placer sand and in other deposits of India is about 12.47 Mt (against the world reserve of 16 Mt) – which contains about 1 Mt of thorium dioxide.
If this mineral is used properly, it can meet India’s energy demand for the next century and beyond. Advanced heavy water reactor of 300 MW capacity designed by BARC, would use thorium based fuel and would likely to make a significant push in power generation by thorium power cycle. An alternative route to use thorium very effectively, economically, and with very high safety parameters is provided by Accelerator Driven Subcritical Reactor (ADSR) system. The concept of using ADSR based on thorium-U-233 fuel cycle was first proposed by Carlo Rubbia in 2013.
Working principle of ADSR
The concept of an ‘accelerator driven subcritical reactor’ is based on the idea of coupling a fission reactor to a particle accelerator. Conventional nuclear power plants require the use of critical reactors, in which the number of neutrons released by fission is just enough to sustain a controlled chain reaction. An ADSR relies instead on a subcrtical reactor, in which more neutrons are absorbed than generated and hence a chain reaction is not possible. A typical arrangement of ADSR sytem is shown in FIG – 4.
The core of such an ADS is mainly thorium, located near the bottom of a tall tank. A typical research ADSR reactor vessel of 100MWtis of size 10 meter in diameter and 20 meter in height. It is filled with some 8000 to 10,000 tonnes of primary coolant like lead, sodium or lead-bismuth at high temperature, which circulates by convection around the core. Thus the hot core due to fission reaction is cooled and kept at designed temperature level without overheating. The hot coolant is extracted from the reactor by a pump to recover the heat for generation of steam.
To run such a subcritical reactor at constant fission level, additional neutrons are needed, which are generated by bombarding a heavy metal high atomic number target like tungsten, thorium, depleted uranium, tantalum, lead, lead-bismuth etc., placed inside the core of the reactor with a beam of high energy protons, usually >500MeV supplied by a linear accelerator or cyclotron and the process is known as spallation. Up to one neutron can be produced per 25MeV. The spallation neutrons could not cause additional fission events in the system. However, if the spallation target is surrounded by a blanket assembly of nuclear fuel rods, containing nuclear fuels such as uranium, plutonium, or thorium-232 which can breed fissionable U-233, then fission reaction can be sustained.
In such an ADS system, the neutrons produced by spallation would cause fission in the fuel, assisted by further neutrons arising from that fission. Natural and abundant Th-232 when bombarded with neutrons transmutes into Th-233, which is an excellent fissile material. Th-233 beta decays to Pa-233. The Pa-233 that is produced emits another electron and anti-nutrino by a second beta decay to become U-233, the fuel. A U-233 nucleus yields more neutrons for every thermal neutrons it absorbs – when it splits in a reactor and these neutrons are released in the surrounding fuel. This gives better neutron economy in the reactor system. The decay product is either recycled into the new fuel or may be reused in the same fuel to sustain the chain reaction and produce heat energy to generate steam for the steam turbine. FIG – 2 shows spallation event using molten lead as the target and multiplication of neutrons through the fission in the core surrounding the spallation area. The reaction that takes place in the ADS reactor is given below:
neutron (n) ß- ß-
n + Th-232—–> Th-233—–> Pa-233—–> U-233 (fuel)
Uranium-233 is a very good fissile isotope and absorbs neutron. About 94% of all absorption reaction results in fission, when the nucleus splits into two smaller nuclei along with release of a few neutrons, typically 2.48 neutrons per fission and release of energy in the form of heat and gamma rays. Therefore, 6% of all absorption reaction results in radiative capture of neutrons.
94%
n + U-233—–> U-234 ———> fission ( Xe-137 + Sr-94 +3n) + energy
6%
n + U-233 ——> U-234——–> U-234 + g
Fission of one atom of U-233 generates 197.9 MeV = 81.95 TJ/kg energy. No control rods are necessary to regulate output power of the ADS reactor, instead the same can be controlled by variation of the proton beam intensity.
Up to 10% of the neutrons would come from the spallation, with the rest of the neutrons arising from fission reaction in the fuel blanket assembly. An ADS can therefore be turned off simply by switching off the current supply that produces proton beam.
A fraction of the reactor’s energy output is in turn used to power the accelerator. The main advantage of an ADSR is increased safety due to elimination of criticality related accidents. With time, the concentration of fission fragments and heavy elements in the fuel will increase to the point – where it is no longer practical to use it.
There is another innovative design that envisages subcritical operation of a molten salt ADS driven by high energy proton accelator is shown in FIG – 3. This can be a far simpler alternative to conventional ADSR with fuel rod assembly. The fuel in molten salt ADSR is a mass of cotinuously circulating molten salt of LiF-CaF2-ThF4-UF4. A spread out beam directly heats the molten salt with 1 GeV protons resulting in 15 numbers of neutrons per incident proton. The operating temperature is 6500C. Hot molten salt is pumped through heat exchanger to utilize its heat content to generate steam for the power plant. Such systems should be simple and relatively cheap to build. It provides multiple containments to protect against molten salt leaks. But unfortunately, this design has some critical drawbacks such as:
- The value of criticality coefficient, k depends on the actual location of liquid salt inside the reactor. But it is extremely difficult to maintain k at designed value within allowable tolerance, and hence the optimum critical state of the reactor.
- The radiotoxicity of the molten fuel is huge. Hence, the classic ADSR with many separate fuel rods as shown in FIG – 4 is a better arrangement.
Steam generation circuit
FIG – 5 shows a simplified diagram of loop type steam generation circuit in an ADSR using sodium as coolant. Hot molten sodium coolant at a temperature of about 510-5500C is circulated in the primary heat exchanger with the help of the primary pump, where it heats the secondary molten sodium. The primary sodium after heat transfer returns back at about 3800C to the reactor vessel to repeat the core cooling cycle. The hot secondary sodium in liquid form is pumped to the secondary heat exchanger, which is the steam generator, where feed water is heated by secondary sodium medium to form high pressure superheated steam at a pressure of 125 kg/cm2, and at a temperature of 4800C to run the steam turbine for power generation.
The most important function of the heat exchangers is to provide a barrier between the reactor coolant, which may be contaminated with radioactive fission products, and the environment. The thin tubes with large surface area act as both heat transfer element and fission product barriers.
Coolant
Production of energy in the core of the ADSR is intense and therefore the coolant used must have very good heat transfer properties. The coolants also act as neutron reflectors. Being much heavier, if neutrons hit the coolant, many of them reflect back in the core. There are mainly three types of coolants used in ADSR, those are:
- Sodium – It can remove heat effectively from the compact reactor core and remains in the liquid state over a fairly broad range of temperature. It exhibits other desirable properties like low pumping power requirement, low system pressure requirement, which means it can be used at atmospheric pressure, the ability to absorb cosiderable energy under emergency condition as its operating temperature is much below its boiling point, good neutronic properties, and in case of fuel element failure it can absorb many leaked fission product. But its main disadvantage is its chemical reactivity with air and water. It is slightly neutron decelerating.
- Lead – Lead has a very high boiling point of about about 1743 degC and has low vapour pressure. Due to its fundamental thermodynamic and neutronic characteristics, lead offers itself as a good coolant. Due to its low vapour pressure, it permits the reactor to operate at atmospheric pressure, helps prevent large losses of coolant in the event of leaks. It has low neutron absorption. Molten lead does not significantly moderte neutrons. Neutrons are not slowed down by lead, which ensures that the neutrons keep their high energy. No radioactive elements are created by absorption of neutrons by lead. Lead is very effective in absorbing gamma rays, which ensures that radiation fields outside the reactor are extremely low.
Unlike sodium it has no issue like reactivity with air or water. If there is any leak, lead would solidify. It remains liquid at 11260C, and so it can absorb any thermal excursion without any increase in operating pressure. Because of high thermal inertia, passive cooling is possible in case of emergency. Lead has high thermal conductivity of 35 W/m.K enabling effective heat transfer from the fuel elements to the coolant. But the most challenging problem with lead is that it corrodes the reactor internals. Another problem is lead being very heavy, increases the weight of the system, therefore requiring strengthening of supporting system and additional seismic protection.
- Lead-Bismuth Eutectic or LBE – LBE is a eutectic alloy of lead (44.5%) and bismuth (55.5%). Its melting point is about 1230C compared to lead, which has higher melting point of 3270C. The main problem with LBE is that it produces a considerable amount of polonium from neutron activation of bismuth. This radioactive element would dissolve in LBE, and would emit alpha rays that is hazardous. This may pose plant contamination problem.
Accelerator
The concept of particle accelerator has been in vogue since 1911, when Rutherford theorised it. He shot alpha particles, a He atom with a charge 2 +, at gold foil, when some passed though, and others bounced off. This led to his theory that an atom is mostly an empty space. As the name suggests, a ‘particle accelerator’ accelerates subatomic particles. This machine goads a beam of particles, usually electrons, or protons, or sometimes a single heavy atom like lead at rest inside a vacum chamber, where these are accelerated to an amazing speed of about 1,079,252,849 km/hr, which is about the speed of light. Once it nears the speed of light, the rate of gain of velocity decreases towards zero, though it continues to gain kinetic energy. How is it possible? Since kinetic energy = mass x velocity, so when velocity is approaching zero, then the event raises the particle’ mass instead. There are two kinds of particle accelerators: a linear one and a circular one. The linear one looks like a tunnel firing the beam in a straight line as shown in FIG – 6.
Linear accelerators can accelerate heavy ions to speeds not possible with ring type cyclotrons, because cyclotrons are limited by the strength of magnetic fields needed to keep the particles on their curved path. So linacs are most suitable for ADSR power plant, although they are costly. A simplified scheme of linac driven ADSR power plant is shown in FIG – 7.
An identified problem of a commercially viable ADSR is that the accelerator system should not suffer more than approximately 5-10 unscheduled interruptions per year, each of duration of about 1s, or else the associated costs of shutdown would make the plant financially unviable. Therefore, if three reactor cores are to be constructed, then it would be most prudent to construct them at the same site along with four accelerators (one being standby) in a configuration as shown in FIG – 8, enabling all the working 3 accelerators to provide their beam to any of the reactors at any moment when any linac fails. But this may increase the capital cost by about 17%. But the concept of dual accelerators eliminates all financial uncertainities, and ensures profitable running of the system.
The conventional linacs use normal conducting accelerating cavities cooled by water with limited duty cycles. However recently 100% CW linacs have been designed and constructed that rely on liquid helium cooled niobium superconducting structures. This eliminates the RF power losses, increasing the cavity-to-beam power efficiency to close to 100%. Another significant advantage of superconducting structures is their large aperture, which allows for very high transmission, low power losses, which is a key issue for accelerators of extreme beam power and to permit easy maintenance.
Accelerator requirement
Accelerator power:- Accelerator power, Pbeam= PADSx (1-ks)/ksGo, where
PADS= desired ADS fission power, ks= neutron multiplication factor, Go=a constant. Ks = The ratio of neutrons in generation (n+1) to the number of neutrons in generation (n). If k is less than 1, then reactor is subcritical and the chain reaction decreases in time.
Therefore accelerator power is a trade-off between acclerator power, and criticality margin, ks, (also known as keff) as shown in FIG – 9. ADS power is linerely proportional to the beam power, and hence can be adjusted by changing beam power. This is a very useful feature to smooth out fluctuating load.
- Beam energy: Optimum neutron production is obtained at beam energy at about 900 MeV. At lower energy, protons tend to lose more energy by ionization, which does not produce neutrons. At higher energy, pion production increases.
- Beam power: The range of beam power depends on core design. For industrial application the beam power should typically be between 1 – 10 MW, depending on ks and desired power output. A large beam intensities may be required to follow the load. Stability of beam power is a technical requirement.
- Beam spot size: The proton beam is directed to the spallation target through a window. The beam spot size should be large at the point of impact on the window, and may be calculated on the beam current density norm of 0.1-0.2 mA/cm2.
- Beam losses: Beam losses are to be minimized ensuring that beams are not strayed, otherwise it would irradiate accelerator elements and lower its life. For linear accelerators beam losses are limited to 1W/m.
Spallation target
The spallation is a very key component to distinguish the ADS system from other reactor system. It couples the accelerator with the reactor system. Generally, there are three main types of spallation targets:
Liquid target with windows generally use heavy metals like Hg, LBE, Pb in molten form. The design is simple, but the windows get damaged due to radiation and thermal/mechanical stress.
Liquid target without widows has simple design, and can withstand higher energy proton beam bombardment, can increase neutron yield. But it is difficult to construct, and its operation is not stable.
Solid metal target is most suitable. It can withstand high radioactivity, exposure to highly toxic decay product polonium-210, has operational stability and can withstand corrosion. Metal tungsten and to a lesser extent zirconium alloys are mostly used as the material of solid targets, because tungsten has excellent properties like – High n/p yield of about 20, high atomic number- 74, high density-19.3 g/cc, high melting point- 34000C, high thermal conductivity- 173W/m/K, chemically inert, has good life of about 1 year under severe radiation, and are not costly. It can be operated in high power region up to 100 MW. It provides safe and low-hazard operation. It generates small waste.
To be continued…
Rathindra Nath Biswas is a 1964 batch Chemical Engineering graduate from Jadavpur University, Calcutta. He further specialized in design & engineering of Benzol Rectification Plant in Giprokoks, USSR. He retired as Head, MECON, Durgapur Site Office.
