Tritium Transport Behavior and Parameter Sensitivity Analysis of Tritium Extraction System Based on TRAP Program ^†

Abstract

As a core subsystem of the fuel cycle, the Tritium Extraction System (TES) is responsible for the critical functions of separating, extracting, and purifying tritium from the breeding blanket. To quantify the influence of key design parameters on tritium inventory characteristics and loss mechanisms, this study, based on the helium-cooled solid breeder blanket TES designed by the China Academy of Engineering Physics, adopts an established, refined TES model and develops a numerical solver named TRAP (Tritium Retention and Analysis Program). By solving multi-component mass balance differential equations, the dynamic inventories of various species in the TES are quantitatively characterized, and systematic sensitivity analyses are conducted on parameters such as the Permeation Reduction Factor (PRF), inlet hydrogen concentration, and key component temperatures. The results of this study can provide a reference for parameter optimization and engineering design of tritium extraction systems for fusion reactors.

1. Introduction

In controlled nuclear fusion, the deuterium–tritium (D-T) reaction is the most promising path to practical fusion energy due to its high cross-section and low ignition threshold. Continuous operation of a fusion reactor requires tritium self-sufficiency, i.e., the tritium produced in the breeding blanket must compensate for core consumption and transport losses (decay, permeation, etc.). Tritium has a low natural abundance and a half-life of only 12.3 years [1,2]; it cannot be produced or stored in large quantities. Therefore, efficient recovery and purification of tritium fuel is a core engineering challenge.

As shown in Figure 1, the tritium plant ensures tritium self-sufficiency and consists of three subsystems: the inner cycle, the outer cycle, and the safety confinement system [3,4,5]. The Tritium Extraction System (TES) in the outer cycle is responsible for separating, extracting, and purifying tritium from the breeding blanket. It uses a purge gas to carry tritium out, performs initial purification, and then sends it to the Isotope Separation System (ISS). As the first subsystem that tritium enters, the performance of the TES directly affects the processing efficiency of the entire tritium plant and the tritium economy.

Figure 1. Simplified block diagram of the tritium plant model.

Engineering conceptual designs for the TES have been proposed internationally. For the ITER Helium-Cooled Pebble Bed (HCPB) blanket, Albrecht et al. [6] proposed a stepwise tritium extraction scheme. China developed the Helium-Cooled Solid Breeder Test Blanket Module (CH-HCSB TBM) and its associated tritium system [7] and performed computational analyses to obtain tritium inventory and loss data. Lei Pan et al. [8] developed an HCPB blanket tritium cycle model and analyzed the influence of parameters such as PRF, extraction efficiency, and coolant ratio. Wei Shi et al. [9] established a more detailed TES model for the Helium-Cooled Solid Breeder (HCSB) blanket, considering permeation, leakage, and HT/HTO chemical equilibrium, and conducted a sensitivity analysis on PRF.

Based on the solid breeder blanket TES designed by the China Academy of Engineering Physics [10], this paper adopts the existing TES model [9] to further conduct sensitivity analyses on the Permeation Reduction Factor (PRF), inlet hydrogen concentration, and key component temperatures, providing a reference for TES parameter optimization and engineering design.

2. Tritium Extraction System Model and Numerical Method

As shown in Figure 2, the TES for the Helium-Cooled Solid Breeder (HCSB) blanket uses a hydrogen-doped helium purge gas circuit, divided into primary extraction and secondary separation/purification circuits. In the primary circuit, helium with 0.1% H₂ purges the blanket, carrying HT and HTO via isotope exchange [11]. The gas is filtered, cooled, and passed through an ambient-temperature molecular sieve bed (AMSB) to capture H₂O and HTO. After drying, it is cooled to 103 K and enters a cryogenic molecular sieve bed to adsorb hydrogen isotopes (Q₂). In the secondary circuit, two alternating cryogenic beds are heated to 123 K for desorption. The desorbed gas is buffered, pressurized, and sent to a palladium membrane diffuser, where high-purity Q₂ permeates to the ISS for isotope separation. Unadsorbed Q₂ and He are reheated, mixed with return gas from the ISS, and replenished with H₂ to 0.1% before returning to the blanket.

Figure 2. Helium-Cooled Solid Breeder blanket Tritium Extraction System model.

To model tritium transport, mass balance differential equations for each species in each component are established using the mean residence time method [12], a diffusion-limited permeation model, and chemical equilibrium equations.

For any gas component j, the mass balance equation in component i can be expressed as:

$${\displaystyle \frac{d{I}_i^j}{dt}}={inflow}_i^j-{\displaystyle \frac{I_i^j}{{\tau }_i}}-{losses}_i^j,$$

(1)

where $I_i^j$ is the inventory (mol) in component i, j can represent H₂, HT, H₂O, HTO, τ_i is the mean residence time (h), and ${inflow}_i^j$ is the total loss rate (mol/h) of the gas, including direct leakage, permeation leakage, and tritium decay.

Permeation loss of Q₂ follows:

$${\varphi }_i^{Q_2}={\displaystyle \frac{{Pe}_i{A}_i}{{PRFt\mathrm{h}}_i}}(\sqrt{R{T}_i{I}_i^{Q_2}/V_i}-\sqrt{R{T}_{GB}{C}_{GB}^{Q_2}}),$$

(2)

where ${\varphi }_i^{Q_2}$ is the permeation rate (mol/h) of hydrogen isotope Q₂ in component i,A_i is the permeable area (m²), Pe_i is the material permeability coefficient (mol·m⁻¹·h⁻¹·Pa^−0.5), th_i is the component wall thickness (m), T_i is the component temperature (K), $I_i^{Q_2}$ is the inventory of hydrogen isotope Q₂ (mol), and V_i is the component volume (m³).

Furthermore, when the purge gas flows through components with significant temperature changes, the reversible reaction H₂ + HTO ⇌ HT + H₂O within the purge gas must be considered. Let the forward reaction rate be ∆^HT. Then, the flow rate changes in each component must satisfy the following equilibrium relationship:

$$K_{eqp}={\displaystyle \frac{(F_{in}^{H_2}-{∆}^{HT})(F_{in}^{HTO}-{∆}^{HT})}{(F_{in}^{HT}+{∆}^{HT})(F_{in}^{H_{2}O}+{∆}^{HT})}},$$

(3)

where K_eqp is the chemical equilibrium constant at the corresponding temperature, and F represents the inflow rate (mol/h).

Based on the above refined TES design model and tritium transport calculation method, a system of differential equations in the form of Equation (1) is established for all components in the TES flow sheet shown in Figure 2. A numerical solver program named Tritium Retention and Analysis Program (TRAP) was developed to solve the system of mass balance differential equations for multiple components and multiple species, thereby quantitatively characterizing the dynamic inventory of each species in the TES.

3. Sensitivity Analysis

3.1. Steady-State Inventory Distribution of Main Species

Based on solving the above differential equation system with the TRAP program, the steady-state inventory distribution of the main species in each TES component was obtained, as shown in Figure 3. It can be observed from Figure 3 that HTO and H₂O exist only in the ambient-temperature molecular sieve bed and components upstream of it, while no inventory of these two species was detected in downstream components. Furthermore, the inventory of HT in the cryogenic molecular sieve bed and the hydrogen isotope separation device is much higher than that in other components, and far exceeds the inventory of HTO in each component.

Figure 3. Steady-state inventory distribution of main species in TES components under baseline conditions.

3.2. Sensitivity Analysis of Key Parameters

Sensitivity analyses were conducted on key parameters in the TES, including the Permeation Reduction Factor (PRF), inlet hydrogen concentration, and key component temperatures. The results are shown in Figure 4. As PRF increases from 20 to 100, the steady-state inventories of HT, H₂, HTO, and H₂O remain essentially unchanged. The permeation rates of both HT and H₂ decrease monotonically with increasing PRF, and at the same PRF value, the permeation rate of H₂ is always higher than that of HT. As the inlet hydrogen concentration increases from 0.05% to 0.20%, the total tritium inventory increases monotonically; permeation loss decreases, decay loss increases slightly, and the total loss decreases from 4.05 × 10⁻⁷ mol/h to 3.87 × 10⁻⁷ mol/h. With increasing temperature of the ambient-temperature molecular sieve bed, the inventories of HT and H₂ remain unchanged, while the inventories of HTO and H₂O decrease significantly.

Figure 4. Sensitivity analysis of key parameters in the TES. (a) Variation in inventories with PRF; (b) Variation in permeation with PRF; (c) Variation in total tritium inventory with inlet hydrogen proportion; (d) Variation in tritium loss with inlet hydrogen proportion; (e) Variation in HTO inventory with the temperature of the AMSB; (f) Variation in substance inventory with the temperature of the AMSB.

4. Conclusions

In this study, a numerical solver named TRAP was developed for the Tritium Extraction System (TES) of a helium-cooled solid breeder blanket. Through evaluation of three key design parameters—the Permeation Reduction Factor, the inlet hydrogen concentration, and the temperature of the ambient-temperature molecular sieve bed—it was found that varying PRF in the range of 20–100 has a negligible effect on the steady-state inventories of HT, H₂, HTO, and H₂O, while the permeation rates of both HT and H₂ decrease monotonically with increasing PRF, and the permeation rate of H₂ is consistently higher than that of HT. Increasing the inlet hydrogen concentration from 0.05% to 0.20% reduces the total loss by only about 4.4%, but increases the total tritium inventory by approximately 18%. Raising the temperature of the ambient-temperature molecular sieve bed reduces the inventories of HTO and H₂O without affecting those of HT and H₂. The TRAP program enables quantitative characterization of dynamic tritium inventories, providing a valuable reference for parameter optimization and engineering design of the TES.