1. Introduction
Engineered Safety Features (ESFs), such as the Second Shutdown System (SSS), are additional to the main safety system and are employed to address accidents in the event of a failure of the main safety system. The function of SSS is limited to the shutdown of the reactor under emergency conditions (i.e., against accidents with failure of First Shutdown System (FSS)). On the other hand, FSS functions include start-up, shutdown, reactor power control, SCRAM (Safety Control Rod Axe Man) the reactor under emergency conditions, and burn-up compensation. In the course of developing safety programs for research reactors, SSS has been globally adopted as one of the most promising countermeasures. Some efforts to study, design, and implement SSS have been performed, such as the design of a complementary scram system for liquid metal cooled nuclear reactors[1], design of an additional safety rod for the Ghana research reactor-1[2], development of a diverse secondary shutdown system for a low power research reactor[3], and implementation of SSS in ETRR2[4] and OPAL[5].
Some designs have been investigated in the Tehran Research Reactor (TRR) by considering heavy water based concepts for SSS[6, 7]. In these designs, heavy water was used as a moderator in normal operation and as a SSS in emergency conditions by dumping. The long dumping time of heavy water, major variations needed in the reactor structure, and difficulties in the production and operation of heavy water are some of the drawbacks of the aforementioned designs for TRR[6, 7].
Another design in TRR is based on liquid neutron absorber injection as a SSS. However, the decrease of excess reactivity in normal operation and relatively large acting time are drawbacks of this design[8].
Considering the drawbacks of earlier designs for TRR, this paper presents the most important adopted principles in the design of an independent, fast acting, diverse, safe, simple, and practical Shutdown System (SS), that complies with the design and safety criteria. Further, in order to avoid a failure mode and to achieve diversity, the design of the SSS is such that all of the components, parts, and shutdown mechanism are entirely different.
The rest of this paper is organized as follows. Section 2 describes the general features and characteristics of the TRR; section 3 reviews the necessary requirements and characteristics regarding the design of one SSS; section 4 contains a discussion on the neutronic analysis investigating parameters dependent on the design specifications and associated results, and section 5 summarizes the conclusions of this work.
2. Tehran Research Reactor (TRR)
TRR is a medium power moderated and reflected light water research reactor with a maximum thermal power capacity of 5 MW[9]. This reactor has been operated continuously and safely for 50 years since first criticality was established on 11th November 1967. The reactor core is immersed in demineralized water and is mounted on a grid plate with an array of 9×6 holes with a grid pitch of 8.1×7.71 cm2. It is comprised of Standard Fuel Elements (SFEs), Control Fuel Elements (CFEs), a Regulating Rod (RR), Graphites (GRs), and Irradiation boxes (IRs), as shown in Fig. 1 for the equilibrium core No. 70[10]. The total and active lengths of the Fuel Elements (FEs) are 65.5 and 61.5 cm, respectively.
As can be seen from the left and right of Fig. 1, the three peripheral spaces of this core are occupied with beam tubes and a thermal column, respectively. Due to the lack of outer space in the core, the proposed locations for the SSS are seven achievable IRs, as shown in the right of Fig. 1.
Each SFE and CFE contains 19 and 14 fuel plates, respectively; each fuel plate is composed of low enriched U3O8, dispersed in aluminum sandwiched between two aluminum plates. The FSS is composed of four shim safety rods for reactor shutdown in less than 1 second and one RR for fine power regulation. This system also consists of a set of control plates arranged in pairs and moving inside the modified fuel elements. The control plates of CFE are composed of Ag, In, and Cd, with mass fractions of 80, 15, and 5%, respectively, and the RR is built using stainless steel.
Different parts of the Reactor Protection System (RPS), such as the area monitoring detectors, in-core neutron flux detectors, and water level detectors provide the scram signals. These scram signals are used for the actuation of each SS if they exceed pre-set values.
There is one interlock between the two SSs that only the FSS can be actuated by the scram signal command sent via the FSS detectors, but the FSS and SSS would be actuated if the scram signal command is sent from the SSS detectors.
3. Design of SSS
3.1. Design principles
The most crucial issue in designing a SSS is that it requires clear and accurate specifications without affecting the reactor core design. Mobile and stationary components are possible in any design. Nevertheless, it is recommended to use stationary parts whose functioning does not depend on an external input.
Emergency shutdown mechanisms are planned to shut down the reactor based on the following; 1) fast insertion of neutron absorbing materials into the core, 2) fast removal of fuel from the core, or 3) a fast change of the neutron leakage from the core. If there is a forced convection system, with increasing research reactor power, for example, in ETRR2 and OPAL[11], the coolant direction through the core will be upward. Conversely, for low and medium light water research reactors, such as TRR, the direction is downward. This has considerable effects on the SSS characteristics, with improved core accessibility with neighbor structures, systems, and components in the reactor pool. A planned safe research reactor requires a design a reliable shutdown system, as an amendment to the existing shutdown system. To achieve this goal, the relevant major issues for the overall reliable design of the SSS are described in the following subsections.
3.1.1 Design criteria
The basic safety criteria, such as diversity, redundancy, single failure criterion, segregation, and separation, are needed to realize a reliable SSS. We aimed to design a SSS incorporating all of the aforementioned criteria. It is recommended to minimize technical risks as much as possible by means of proven methods and materials. For example, enriched boric acid and enriched gadolinium solutions are utilized as a liquid neutron absorber in many cases[12-14].
3.1.2 Reactivity worth
The reactivity worth of the SSS should be sufficient to bring the reactor to a subcritical state with an adequate shutdown margin, assuming that the other shutdown system has failed following anticipated incidents due to any reason. The SSS shutdown margin must be larger than 1000 pcm[15, 16]. The uncontrolled or spurious withdrawal of absorber rods during services or fuel handling and the reactivity insertion from FEs falling following their erroneous or uncontrolled withdrawal, are some of these anticipated incidents.
3.1.3 Neutronic requirements
Main neutronic parameters in a research reactor, such as shutdown margin, neutron flux distribution, effective delayed neutron fraction, and power peaking factor, are treated to investigate the neutronic requirements. During the design of the SSS in the reactor core, the considered process must have a minimum negative effect on the neutron flux distribution as well as the total amount of the neutron flux in the core irradiation boxes. The power peaking factor after the SSS chamber(s) insertion into the core must not exceed the allowable limit which conservatively must be less than 3 for the TRR[9].
Although the delayed neutrons ordinarily comprise less than one percent of the neutrons released in fission, they play an important role in the control of nuclear reactors. The delayed neutrons considerably increase the reactor period[17]. The delayed neutron role in reactor kinetics can be seen from Eq. (1)[12];
where n, k, ρ, β, l, λi, and Ci in Eq. (1) are the neutron density in the reactor core, effective multiplication factor, core reactivity, delayed neutron fraction, prompt fission neutron life time, decay constant, and number density of the precursors of the ith delayed group, respectively. The summation for 6 groups of precursor nuclei, correspond to six delayed fission neutron groups. As can be seen from Eq. (1), the insertion of SSS chamber(s) into the core should not reduce the delayed neutron.
The effective delayed neutron fraction
3.1.4 Acting time
Due to the importance of time for the prevention and control of anticipated accidents, a S should have the lowest possible acting time[9].
3.1.5 Operational requirements
a. Utilization considerations
One of the most important uses of the TRR is radioisotope production from sample irradiation in the core irradiation boxes. Designing the SSS with maximum space of the irradiation boxes is therefore desirable. The capability of bringing the reactor to a safe shutdown condition and the possibility of reactor restart-up without delay, must be considered in the SSS design properties.
b. Service considerations
The SSS must be designed to allow easy servicing, with a minimum of hazards and problems for the operating personnel.
3.1.6 Cost problems
Any changes in the reactor core have related financial repercussions. The impact of ageing problems for this reactor potentially increasing these costs, is another consideration. Ensuring minimal variations in the reactor core to mitigate the financial problems is therefore important. Using plans that increase the excess reactivity, which prolong the fueling period, will result in a reduction of the fuel costs.
3.2. Proposed chambers design
It is preferable to find a SSS chamber design that can be located in the core region, as the outer core region has low neutronic worth and is comprises beam tubes, rabbits, and the thermal column. The most appropriate space in the core from the neutronic aspect is the D6 position, as shown in Fig. 1.
The IRs are used to perform irradiations for radioisotope production and for scientific research, and are the two most important areas of utilization of the TRR at present. Two usual irradiation strategies for sample insertion into the reactor core are:
1. The use of the two cylindrical facilities, denoted by "S1" and "S2" in the right of Fig. 2.
2. The use of a rectangular irradiation facility, with holes for insertion of irradiation samples, denoted by "Sample for irradiation" in the left of Fig. 2.
By focusing on the aforementioned criteria, such as the TRR safety criteria and operating constraints, a suitable shutdown system was selected and investigated as the SSS. This shutdown system, which depends on the reactor condition and strategy, has two different geometric chamber(s) for the injection of the liquid neutron absorber, as shown in Fig. 2. As seen in this figure, the SSS chambers design, with a chimney (left) and a two-triangular prism geometry (right), were in accordance with the TRR core characteristics, properties, and limitations. For example, these designs had no considerable drawbacks for sample irradiation in the reactor core.
The main data for the proposed SSSs are presented in Tables 1 and 2 for the chimney and the two triangular prisms, respectively.
| Features | Quantities |
|---|---|
| Cladding material | Al-6061 |
| Cladding thickness | 2 mm |
| Chamber internal width | 8 mm |
| Chamber internal height | 73.05 cm |
| Total volume of chamber | 1663 cm3 |
| Features | Quantities |
|---|---|
| Cladding material | Al-6061 |
| Cladding thickness | 2 mm |
| Each vertex isosceles triangles length | 3.9 cm |
| Each chamber internal height | 73.05 cm |
| Total volume of two chambers | 1111 cm3 |
Four different monitoring channels receiving data from the neutronic, radiological, temperature, and pool level detectors provided the scram signals to the two shutdown systems from the start of an accident. The liquid neutron absorber injection took place at actuation of any of these four command circuits following a FSS failure in less than 1 second. In normal operation, the pressure of SSS (20 kPa(0.2 bar)) was used as a monitoring tool to check for system leakage. In the event of an accident, a pressure of 103 kPa (10 bar) was needed to ensure liquid neutron absorber injection in the required time.
4. Results and discussion
The MCNPX code was used for the neutronic calculations. This code solved the problems by following the particle tracks for many histories using the stochastic method and enabled the transport of several particle types. It is a multipurpose code with continuous energy cross sections from the ENDF/B-VI.6 neutron libraries. This code is also capable of simulating the case of exact geometry[20]. There were some sources of uncertainty in the results of the MCNPX code, namely: 1) errors due to geometric simplifications in the simulation, such as ignorance of the cladding margins, and the up and down of the GRs and FEs; 2) inherent errors in the MCNPX libraries and physical models. The sources of these errors was outside the scope of the paper.
4.1. Design criteria
In order meet the diversity and redundancy criteria, the proposed design for the SSS is based upon liquid neutron absorber injection, and is completely different from existing shutdown system based on the absorber rod mechanism. The monitoring detectors and command circuit are independent from the existing equipment, which meet the independence criterion. The fail-safe criterion was also considered in the performance of this newly designed SS to enhance the SSS reliability.
4.2. Reactivity worth
The calculated reactivity worths of the SSS in different operating conditions for these proposed chambers design are given in Table 3 and 4 for the chimney chamber and triangular prism chambers, respectively. Enriched boric acid with 99% enrichment of B-10 and gadolinium nitrate with 70% enrichment of Gd-157 were used for these calculations.
| TRR core condition | Reactivity (pcm) | Relative error (pcm) |
|---|---|---|
| Equilibrium core No. 70 | 3105 | ±12 |
| SSS filled with light water | 3195 | ±13 |
| SSS filled with heavy water | 3882 | ±13 |
| SSS filled with nitrogen | 3790 | ±12 |
| SSS filled with enriched boric acid | -199 | ±11 |
| SSS filled with enriched gadolinium nitrate | -309 | ±12 |
As shown in Tables 3 and 4, the value of 3105 pcm was for the cold and full power excess reactivity. This was comprised of 1105 pcm of the reactivity feedbacks, such as the coolant temperature and density, fuel temperature, irradiation channels, and facilities, and also 2000 pcm for the maximum accessible positive reactivity, which could be inserted into the critical operational core in the event of control rods withdrawal accident.
| TRR core condition | Reactivity (pcm) | Relative error (pcm) |
|---|---|---|
| Equilibrium core No. 70 | 3105 | ±12 |
| SSS filled with light water | 3163 | ±12 |
| SSS filled with heavy water | 3718 | ±14 |
| SSS filled with nitrogen | 3605 | ±11 |
| SSS filled with enriched boric acid | -56 | ±12 |
| SSS filled with enriched gadolinium nitrate | -256 | ±13 |
As SSS serves as a back up to the FSS, this system must be capable of coping with anticipated accidents with a necessary shutdown margin. Therefore, the SSS is required to have ~3000 pcm for the acceptable safety criterion and for the shutdown margin.
As can be seen from Tables 3 and 4, the reactivity worth of the proposed chambers are larger than the necessary criterion in which SSS must be able to transfer the reactor to a subcritical state with an acceptable shutdown margin of 1000 pcm[15, 16].
As shown in these tables, in spite of having a high excess reactivity for normal operation in comparison with other options, heavy water was not used due to the difficulties and costs in production, maintenance, and purification.
The performed calculations show that the determined volume of the designed SSS liquid neutron absorber injection chambers meets the required shutdown margin and injection time for filling the SSS chambers. In other words, these studies have shown that this system has characteristics to ensure the required abilities for device performance.
The integral and differential worths of the two SSS chambers were calculated and shown in Figs. 3 to 6.
As can be seen from Figs. 3 and 5, the reactivity worth of the enriched gadolinium nitrate was larger than that of the enriched boric acid, and the reactivity worth of the chimney design was larger than that of the triangular prism design. Although the difference between the reactivity worth of the two used neutron absorber solutions was less than 5%, the liquid neutron absorber with more reactivity provided a larger shutdown margin, enabling the use of the smaller chamber(s) with a shorter injection time.
The most important regions of the SSS chamber(s) with regard to reactivity worth were the middle regions; six middle parts from parts 4 to 9 shown in Figs. 4 and 6. This phenomenon is explained in the following section. The neutron flux in the middle of the core was maximum, which led to more neutrons being absorbed in the injected liquid neutron absorbers. As a result, the reactivity decreased more in these regions.
4.3. Acting time
In contrast to the earlier mentioned designs of SSS for this reactor[6-8], this proposed design had chambers with small capacities, enabling an injection time of less than 1 second. This short time for bringing the reactor to a safe shutdown condition was of the order of the FSS elapsed time of ~ 0.78 seconds. The design can be classified as fast owing to the short injection time for the liquid neutron absorber. The acting time of the SSS was calculated using Pipe Flow Expert software, which is a fluid flow and pressure loss calculations software. This software is used for designing and analyzing complex networks. It enables the user to draw a schematic of a pipe system containing up to 1000 pipes. This software can be used to include tanks, pumps, control valves, components, valves, and fittings. The fluid velocity and mass rate for filling any container are outputs of this software, and provided the acting time of the proposed SSS[21].
4.4. Neutronic parameters
The neutron flux distribution before and after the SSS chimney and SSS triangular prism chambers insertion in the TRR core were studied in detail for the three IR positions A9, C2, and D6, as shown in Figs. 7, 8, 9 and 10.
The positions A9 and C2 were far away from the position D6, dedicated to the SSS chamber(S) insertion. The SSS chamber insertion with the chimney design or SSS chambers with a triangular prism design had no considerable effect on the flux distribution of C2 and A9, because the neutron flux effects are local.
The effect of the SSS chamber(s) insertion for different energy regions was different at the irradiation position D6. The SSS presence caused a decrease in the thermal flux of up to 17.0% compared to the normal design. This is because the low energy of the thermal neutron was easily suppressed with the aluminum walls. There was an increase in the epithermal and fast flux in comparison to the normal design as a result of the moderator reduction after the SSS chamber(s) insertion. The total neutron flux, that is the sum of the thermal, epithermal, and fast flux, was 8.0% lower than the normal.
As can be seen from Figs. 7 to 10, the neutron flux level was higher for the chimney design providing an advantage of this design over the triangular prism design.
As can be seen from Fig. 10, the variations in the total neutron flux due to the SSS chamber(s) insertion into the core were not considerable, and the variations in power density, and as a result, the power peaking factor, were insignificant; therefore, this parameter remained largely unchanged.
Another important parameter considered for the safety criteria is the effective delayed neutron fraction given by Eq. (1). This was calculated in Table 5 both before and after the SSS chamber(s) insertion into the equilibrium core No. 70. This quantity is mainly dependent on the core material and configuration; therefore, there were minimal variations in this parameter in Table 5 for the proposed chambers.
| Configuration | βeff | Relative error (pcm) |
|---|---|---|
| Normal configuration | 710 | ±12 |
| After chimney chamber insertion | 703 | ±12 |
| After triangular prism chambers insertion | 711 | ±12 |
4.5. Operational considerations
The elapsed time of the liquid neutron absorber, from receiving the signal to complete injection, was less than 1 second, which assured that a safe condition was achieved following an accident and FSS failure. This design is compliant with the radioisotope production strategy in the TRR and did not have any effect on the radioisotope production capabilities. Due to the compact design of this plan, it is foreseeable that there will be no significant service and maintenance problems. The large excess reactivity in normal operation is one of the important advantages of this design, which causes a prolonged operation cycle in each fuel shuffling.
4.6. Cost considerations
Due to the small variations made to the TRR core during the SSS implementation, the financial costs associated with this process are limited. The insertion of the SSS chambers into the core increased the excess reactivity, corresponding to a decrease in the required fuel and fuel provision costs.
5. Conclusion
In this paper, by considering the disadvantages of the former SSS designs in MTR type reactors, two SSS chambers design were proposed. The SSS design was based on functioning principles that were different from those of the existing FSSs. The major design characteristics, as well as the core parameters variation analysis, were performed using the MCNPX 2.6.0 code. Results of the proposed SSS calculations satisfied the design requirements, such as the shutdown margin criterion and neutronic parameters (e.g., power peaking factor and effective delayed neutron fraction). The results of this work showed that this SSS design had adequately diverse and redundant features to ensure high reliability, with the main advantages as follows:
1. High reactivity worth,
2. Negligible variation in the core structure,
3. Positive effect on core reactivity in normal operation,
4. Very short acting time,
5. Small effect on total neutron flux density and prolonged operation cycle,
6. Low difficulty in operational capability, such as free space for irradiation,
7. Easiness for service and maintenance,
8. Less financial cost compared to other earlier designs.
As a result of this design study, it was confirmed that in addition to the FSS, the proposed SSS showed promise as a reliable, inexpensive, and fast acting system, satisfying the design requirements and enhancing the reactor safety with minimal impact on the performance of the reactor core.
Design of a Complementary Scram System for Liquid Metal Cooled Nuclear Reactors
, Nuclear Engineering and Design. 243: p. 87-94 (2012). doi: 10.1016/j.nucengdes.2011.12.007.Design of an additional safety rod for Ghana Research Reactor-1 using MCNP5 code
, Nuclear Engineering and Design. 245: p. 13-18 (2012). doi: 10.1016/j.nucengdes.2011.12.030.Development of a Diverse Secondary Shutdown System for a Low Power Research Reactor
, Research Reactor Utilization, Safety, Decommissioning, Fuel and Waste Management. 10: p. 81 (2003).Effect of Core Configurations on Burn-Up Calculations For MTR Type Reactors
,The OPAL (Open Pool Australian Light-Water) Reactor in Australia
, Nuclear Engineering and Technology. 38 (5): p. 443 (2006).Design of emergency shutdown system for the Tehran Research Reactor; Part I: Neutronics investigation
, Annals of Nuclear Energy. 103: p. 306-314 (2017). doi: 10.1016/j.anucene.2017.01.029.Study on Secondary Shutdown Systems in Tehran Research Reactor
, Nuclear Engineering and Design. 291: p. 224-235 (2015). doi: 10.1016/j.nucengdes.2015.05.018.A pragmatic approach towards designing a second shutdown system for Tehran research reactor
, Nuclear Technology and Radiation Protection. 31(1): p. 28-36 (2016). doi: 10.2298/NTRP1601028B.INVAP's research reactor designs
, Science and Technology of Nuclear Installations. 2011 (2010). doi: 10.1155/2011/490391.Replacement research reactor project, chapter 5c, Safety Analysis Report
.Safety Aspects of Research Reactor Core Modification for Fission Molybdenum-99 Production
,Analysis of burn up effects on kinetic parameters in an Accelerator Driven Subcritical TRIGA reactor
, Annals of Nuclear Energy. 62: p. 280-283 (2013). doi: 10.1016/j.anucene.2013.05.047.