Introduction

Helium 3 is an increasing valuable comodity world wide with an estimated market in 2029 of about USD 200 million [1]. Almost all helium-3 is being employed for security purposes across the world; in terms of application, helium 3 is used for neutron detectors having a share about 30% [2]. Cryogenics, quantum computers, neutron scattering facilities, oil and gas exploration and various medical applications are other uses. Helium-3 is widely employed in medical imaging methods, including magnetic resonance imaging (MRI), in clinical respiratory implications such as asthma, chronic obstructive pulmonary disease, cystic fibrosis, radiation-induced lung damage and transplantation [1].

In the current global context of climate change, ensuring stable and safe energy production while sustaining global economic development is of great important. Nuclear fusion is one of the most promising source of renewable energy due to its potential to reduce both a greenhouse gas emissions and dependence on traditional energy sources. In addition, nuclear fusion is considered to have no risk of nuclear accidents and to be highly resource-efficient [3]. In future power plants based on fusion energy, a mixture of deuterium and tritium is the primary choice for fuel with the inconvenience that following the fusion reaction is generated high-energy neutrons and helium:.

$$D + T \to {\text{He}}^{4} \left( {3.5{\text{MeV}}} \right) + n\left( {14.1{\text{MeV}}} \right)$$

Due to the disadvantage of neutron-induced material damage and its impact on longevity within the extreme environment of a fusion reactor, an alternative promising fuel has been proposed to reduce or eliminate neutron formation [4]. This reference suggests that fusion reactors using the Deuterium-Helium-3 fuel cycle, which employs the reaction below, are considered a superior option compared to deuterium–tritium reactors.

$$D + {\text{He}}^{3} \to {\text{He}}^{4} \left( {3.6{\text{MeV}}} \right) + p\left( {14.7{\text{MeV}}} \right)$$

The nuclear reaction that occurs when helium-3 and deuterium fuse creates normal helium and a proton, which wastes less energy and is easier to contain. Nuclear fusion reactors using helium-3 could therefore provide a highly efficient form of nuclear power with virtually no waste and no radiation. The research involved in the present work aims to obtain necessary information for the design and commissioning of an experimental demonstrative setup with a low energy consumption for purifying and enrichment of helium-3 from a nuclear power plant cover gas.

Romania as a Source for Helium-3

Helium-3 is a rare and valuable isotope with growing significance for various applications. However, much of the helium-3 on Earth is not currently accessible. It exists in low concentrations within the atmosphere, oceans, and natural gas reservoirs, making its extraction from these sources inefficient and costly. Therefore, it is essential to explore efficient and cost-effective methods for separating and enriching helium-3, particularly from by-products of industrial and nuclear processes.

One promising source of helium-3 is the decay of tritium, a by-product of nuclear reactors, such as those based on the CANDU (Canadian Deuterium Uranium) design. These reactors use heavy water as a neutron moderator. During operation, the interaction of neutrons with deuterium in heavy water leads to the production of tritium. Most of this tritium accumulates as tritiated water within the reactor systems. Additionally, a smaller fraction of tritium is found in the cover gas (helium-4) due to radiolysis effects. Over time, tritium undergoes beta decay to form helium-3, which migrates into the cover gas.

To recover tritium and subsequently helium-3, advanced isotope separation technologies are required. For tritium recovery, current approaches include gas chromatography, permeation, and cryogenic distillation. In Romania, a research project is underway to develop an advanced modular tritium separation system based on the CECE (Combined Electrolysis Catalytic Exchange) process. By integrating such systems, the recovery rate of tritium from heavy water and cover gas in nuclear reactors can be significantly increased.

Romania’s two CANDU reactors at Cernavodă, which have been operational for an extended period, have produced substantial amounts of tritium. The first reactor has generated approximately 60 Ci/kg of tritium in the heavy water moderator, while the second reactor has produced around 40 Ci/kg. Over time, the decay of this tritium has led to the accumulation of helium-3 in the cover gas and associated systems. It is estimated that a single 700 MW CANDU unit could generate a stockpile of up to 1.5 kg of helium-3 through tritium decay. Romania plans to build two additional CANDU reactors and establish a heavy water detritiation plant, which will further enhance its capacity to recover tritium and helium-3.

In addition to tritium decay, another potential source of helium-3 is natural helium extracted from natural gas. Although natural helium-3 concentrations are low, recovering it from helium already separated during natural gas processing may be cost-efficient. This approach could supplement the helium-3 stockpile generated from nuclear processes.

In conclusion, Romania’s nuclear infrastructure, particularly its CANDU reactors, provides a unique opportunity to generate and stockpile helium-3. By leveraging advanced tritium recovery and helium-3 separation technologies, the country can play a significant role in addressing the global demand for this rare and valuable isotope.

He-3 Separation and Enrichment Techniques

The separation of helium-3 (3He) from helium-4 (4He) is one of the most challenging aspects of helium-3 recovery. The most effective techniques for achieving high-purity helium-3 (up to 99.9%) include superfluid filtration, cryogenic distillation, and adsorption [6, 7]. Bentoumi et al. [8] investigated the direct recovery of helium-3 from the moderator cover gas system in Pressurized Heavy Water Reactors (PHWRs). The amount of recoverable helium-3 depends largely on the activity of tritium in the moderator. For a standard 700 MWe PHWR unit, this amount typically ranges from approximately 0.1 to 0.7 m3 (STP) per year. A critical factor influencing the cost and efficiency of helium-3 recovery is the loss and replenishment rate of the moderator cover gas, which can reach up to 30% of the total cover gas inventory daily. For the production of helium-3 from natural helium, a combination of rectification and adsorption methods has been proposed [9, 10]. Experimental studies using activated charcoal adsorbers immersed in liquid helium at atmospheric pressure have shown promising results. When processing mixtures containing 16.5% to 95.1% helium-3, the final product achieved purity levels between 99.9981% and 99.9999%. Reference [5] highlights the potential for extracting helium-3 from the cover gas at the Cernavodă Nuclear Power Plant. Simulations performed using CHEMCAD software analysed helium isotope distillation. However, the study lacked experimental validation and did not calculate the physical dimensions of the distillation column, limiting the practical applicability of the findings.

To advance helium-3 recovery and separation, further development of cryogenic distillation processes is essential. This method offers high efficiency and the ability to achieve high-purity helium-3 while reducing energy consumption. Moreover, optimizing helium-3 recovery from cover gas mixtures can help mitigate environmental impacts. Despite its importance, research on the cryogenic distillation of helium isotopes has been limited in recent years. This process is crucial for isotope separation at extremely low temperatures but remains underexplored. The following chapter focuses on helium-3 separation and enrichment through cryogenic distillation. It assesses an existing cryogenic distillation (CD) system using in-house design and simulation software based on the Lewis-Matheson method and the tridiagonal matrix Wang-Henke method [11], enhanced with an improved computational algorithm. This simulation model provides performance data for new operating conditions and suggests optimization strategies for helium-3 recovery and separation.

Design and Simulation of the Cryogenic Experimental Column for Helium Separation

For the design of the separation cryogenic column we used a mathematical model based pn on the Underwood method and tridiagonal matrix Wang-Henke method with an improved computational algorithm. These methods can be applied for both design and simulation of a distillation process for helium isotopes. The column is assumed to be composed of N theoretical stages, where the 1st stage is the reboiler and N + 1th stage is the condenser. The basic equations are derived from the requirements for conservation of material, isotopes and for phase equilibrium on each stage. We have acquired extensive knowledge in the design and simulation of cryogenic distillation columns specifically for hydrogen isotopes, and we actively apply this expertise in the development and optimization of our systems based on helium isotopes having 4He a boiling point of 4.22 K and 3He a boiling point of 3.2 K at standard pressure.

Material balance:

$$\sum_{k=2}^{j}\left({F}_{k}-{W}_{k}-{U}_{k}\right)={L}_{i}+{U}_{1}+{V}_{1}-{V}_{j+1}$$
(1)

Phase equilibrium relationships:

$${y}_{i,j}={K}_{i,j}*{x}_{i,j} ;\,\,\, {K}_{i,j}={p}_{i,j}^{0}/{P}_{j}$$
(2)

Summations of mole fractions:

$$\sum_{i=1}^{m}{x}_{i,j}=1$$
(3)
$$\sum_{i=1}^{m}{y}_{i,j}=1$$
(4)

Heat balance:

$${Q}_{j}={q}_{j}+{L}_{j-1}*{h}_{j-1}+{V}_{j+1}*{H}_{j+1}+{F}_{j}*{H}_{Fj}-\left({L}_{j}+{U}_{j}\right)*{h}_{j}-\left({V}_{j}+{W}_{j}\right)*{H}_{j}$$
(5)

Component material balance:

$$\left({L}_{1}+{U}_{1}\right)*{x}_{i,1}={V}_{2}*{y}_{i,2}-{V}_{1}{*y}_{i,1}$$
(6)
$${L}_{j-1}*{x}_{i,j-1}+{V}_{j+1}*{y}_{i,j+1}+{F}_{j}*{z}_{i,j}-\left({V}_{j}+{W}_{j}\right)*{y}_{i,j}- \left({L}_{j}+{U}_{j}\right)*{x}_{i,j}=0$$
(7)
$${L}_{N-1}*{x}_{i,N-1}+{F}_{N}*{z}_{i,N}-{V}_{N}*{y}_{i,N}-{L}_{N}*{x}_{i,N}=0$$
(8)

where: i = 1,…,m; j = 2,…,N-1; m = number of components; N = number of theoretical plates; Ki,j = equilibrium constants; p0i,j = vapor pressure of pur components (bar); Pj = pressure on the each plate. Li and Vi are the liquid flow rate and vapor flow rate that goes out from plate i, xi and yi are the compositions of the each component on plate i at liquid and vapor phases respectively, Fj feed molar flow rate entering plate j, hj, Hj – liquid and vapor enthalpies on each plate, qj, Qj – heat gain and heat lost by each plate, Wj and Uj are the liquid flow rate and vapor flow rate that goes out from plate j, xj and yj are the compositions of the each component on plate j.

A formula based on experimental data [12] is used to calculate vapor pressure of 3He/4He.The liquid flow rates are calculated as functions of the vapor rates by an overall material balance of all plates, from the top through the j.th plate (step 1). The Eqs. (1), are combined with the Eqs. (2), to form a subset of equations. The above equations are linearized by holding the values of the flow rates and separation factors invariant and then are solved by plate to plate (step 2). Using the calculated composition, the Eqs. 4 and 5 are then solved separately for the new values of plate temperatures (step 3) and vapor flow rates (step 4). The procedure is repeated until all equations are satisfied [13,14,15].

This model enables precise simulation of isotope behaviour under cryogenic conditions, considering variables such as temperature, pressure, and feed gas composition. Initial results indicate that helium-3 purity levels of up to 99.999% can be achieved with optimized parameters, demonstrating the model's potential for guiding system design and operation (Fig. 1). Figures 2, 3 and 4 show the distribution of temperature and isotope distribution along the column, respectively, all obtained from the simulation, providing insight of the operation of the distillation column.

Fig. 1
Fig. 1
Full size image

Cryogenic distillation column design

Fig. 2
Fig. 2
Full size image

The temperatures profile along the cryogenic distillation column

Fig. 3
Fig. 3
Full size image

Liquid composition profile along the cryogenic distillation column

Fig. 4
Fig. 4
Full size image

Vapour composition profile along the cryogenic distillation column

In Figs. 3 and 4 is presenting the 3He and 4He vapor–liquid profile in the simulation distillation column. Regarding this figure it can observed that the separation process is more efficient in the rectification section. As the vapor rises along the column, its composition changes significantly, leading to a higher content of the 3He at the top.

On the other hand, the composition changes in the stripping section are less pronounced because this section primarily removes the less volatile components. These concentration changes are influenced by the temperature gradient in the column. The rectifying section generally operates at a lower temperature (near the top of the column), where the separation of components is more efficient. The stripping section operates at a higher temperature, which results in less significant changes in concentration.

The Experimental Setup Evisaged for Column Testing and Software Validation

The He-3 laboratory is part of the Tritium Laboratory—TRI-Vâlcea, which will be used for the development of technologies for He-3 extraction from tritium storage containers (ITCs or getters) and tritium separation. The applicability is immediate, given the construction phase of the CNE Cernavoda detritiation plant, which will produce the first quantities of tritium from 2026. At the same time, the laboratory will provide the necessary technical support for the development of separation and recovery processes for 3He and 4He, used as cover gases in the moderator circuit of CANDU reactors. A cryogenic distillation column will therefore be used for the separation of helium isotopes.

Based on the results of the simulation, an experimnetal setup has been envisaged for the column testing and software validation. The flow diagram of the setup is shown in Fig. 5, where the components of the cryogenic distillation system, including a distillation column, a condenser at the top of the column and a reboiler at the bottom of the column are shown. The setup consists of a column together with a condenser at the top and a reboiler at the bottom of the column, all are accomodated inside a liquid helium radiation shield, which in turn is placed inside a vacuum insulated coldbox to improve heat insulation of the column. The role of the condenser is to liquefy the vapor mixture into liquid and discharge the liquid into the top of the column as reflux. Due to gravitation, the reflux then flows down along the column through the packing material in order to separate the initial mixture. On the other hand, the reboiler is utilized to produce vapours that ascend along the column. These vapours then interact with the liquid in a counter-flow mode, with the liquid becoming saturated with the heavier component and the gas with the lighter component. Finally, the helium-3 stream (the lighter species) is collected at the top of the distillation column and a helium-4 rich stream (the heavier species) is collected at the bottom of the column.

Fig. 5
Fig. 5
Full size image

Flow diagram of the experimnetal setup

The tentatvie operating parameters of the column, as the resulted from the simulation are shown in Table 1:

Table 1 Tentative main operating and performance parameters of the experimental setup
Full size table

The construction of the experimental setup has been finised and commissioning will start soon. Figure 6 shows a picture of the setup distingushing the cryocooler on top of the coldbox, the valve and vessel manifold, togetehr with the vacuum system for the coldbox.

Fig. 6
Fig. 6
Full size image

The picture of the experimnetal setup

Pre-Requirements for the Feed Gas to the Cryogenic Distillation

In this chapter, we present a new method for helium-3 pre-enrichment technology by gas-chromatography, prior to final enrichment by cryogenic distillation process. In order to better understand the mechanisms involved in helium-3 pre-enrichment technology and to select an appropriate chromatographic column system, an analytical study is required. The analytical study concerns the separation parameters (like retention time, retention factor, full width at half maximum, selectivity, resolution, and number of theoretical plates) which will are to be determined in order to select appropriate type of chromatographic column and packings for the separation and recovery optimization.

The experimental setup regarding the helium-3 pre-enrichment technology, performed at 77 K, consists of two main parts: the sample inlet system and the detection system. The sample inlet system includes a 1 ml loop connected to a VALCO valve with 10 ports which can switch from LOAD to INJECT and vice versa positions (Fig. 7). The valve is connected also to the carrier gas stream. A vacuum pump is used for cleaning the sample inlet system between the injections of different samples. A mass flow controller able to measure the pressure downstream is used to inject the gas sample into the loop; another mass flow controller establishes the values of hydrogen carrier gas flow.

Fig. 7
Fig. 7
Full size image

Sample inlet system

The second part of the setup involves a system consisting of an adsorption chromatographic column connected to a quadrupole mass spectrometer – QMS (Fig. 8). The column, made from stainless steel, is 5 m long and has an inner diameter of 1.62 mm. It is packed with 5A molecular sieve 60–80 mesh. In order to achieve the working temperature (77 K), the column is immersed in liquid nitrogen.

Fig. 8
Fig. 8
Full size image

Schematic diagram of the experimental setup

The feed gas stream is alternately injected into the gas chromatographic column, which is operated at a temperature ranging from 77 to 200 K, depending on the composition of the cover or feed gas, primarily on the ratio of helium-3 to helium-4 in the cover/feed gas.

Gas chromatographic theory involves the chromatographic separation and column efficiency. The separation is quantified by the relative retention time of two components (in our case 3He and 4He) in the stationary phase. Another measure of separation is given by the selectivity factor (α). The two gases that represent the object of this paper (3He and 4He) are fed one at a time into the sample inlet system and measured separately. Mixtures of 3He and 4He were not studied because of the small quantity of helium 3 available.

Two sets of measurements were performed, one for helium 4 and one for helium 3, using hydrogen as carrier gas, at flow rates in the range of 20–90 sccm. For each measurement, retention time, retention factor, full width at half maximum, selectivity, resolution were determined using Gauss fit. In Fig. 9, the retention time versus carrier gas flow rates is presented. In the inset of Fig. 9, the difference between the retention time of 4He and 3He at various flow rates is shown. Although the differences between the retention times are relatively low at all the studied flow rates, it can be seen that in the range 40 to 75 sccm these show optimal values for the separation of helium isotopes. Retention factor quantifies the phenomenon of retention that reflects the fact that the different compounds are retained by the column to a varying degree and is expressed as the ratio of time spent by the solute in the stationary and mobile phases according to formula 9 [16].

$$k=\frac{{t}_{r}-{t}_{0}}{{t}_{0}}$$
(9)

where tr is the retention time and t0 is the retention time of the unretained compound. The selectivity factor α (inset Fig. 10) was determined as the ratio between the retention factors of the two components. We observed a better selectivity for higher values of the carrier gas flow rates (70–90 sccm). By using Gaussian peaks fit we determined the separation parameters: retention time, retention factor, full width at half maximum, selectivity, and resolution. Using the experimental data we established the optimum operation parameter presented in Table 2 and the best value of the resolution of the two Gaussian peaks was around 0.6 in the flow rates range 70–90 sccm.

Fig. 9
Fig. 9
Full size image

Retention time versus carrier gas flow rates

Fig. 10
Fig. 10
Full size image

Retention factor versus carrier gas flow rates

Table 2 Summary of obtained data for the optimum flow rate of the carrier gas
Full size table

This preliminary enrichment stage of the He-3 and He-4 mixture, prior to processing through cryogenic distillation, is necessary because this process cannot perform enrichment starting from very low concentrations (on the order of ppm) of the lighter isotope.

Conclusions

In our investigation and evaluation of helium-3 pre-enrichment technology based on the gas chromatographic method, we observed a difference in the retention times measured for the two helium isotopes. Using Gaussian peak fitting, we determined key separation parameters: retention time, retention factor, full width at half maximum, selectivity, and resolution.

In the second part of our research, we developed software specifically designed for the modeling and simulation of a helium isotope distillation column. This software was used to perform assessments based on a mathematical model, with multiple runs yielding parameters derived directly from its implementation of the model.