Silvano Tosti

Experimental and simulation of both Pd and Pd/Ag for a water gas shift membrane reactor

Abstract The study of the water gas shift reaction performance in terms of complete conversions is presented. The behaviour of a membrane reactor (MR) consisting of a tubular microporous ceramic within a thin palladium membrane was compared with a membrane reactor using a palladium/silver membrane. Membranes were developed in order to obtain a metallic layer thick enough to avoid any defects of the metallic layer and ensure infinite hydrogen selectivity with respect to other gases. The lumen of both membrane reactors was filled with the catalyst. The experiments were carried out by using nitrogen as inert gas in the streep having a flow rate ranging between 1×10−4 and 4×10−4 mol s−1 in co-current and counter-current mode in the temperature range 331–350°C and in the feed molar flow range 3.05×10−5–7.1×10−5 mol s−1. Hydrogen was the only one gas passing through both membranes. A complete separation of hydrogen from the other gases of the reaction system was obtained. The water gas shift reaction conversion was close to 100% by using the Pd/Ag membrane. A mathematical model was developed to interpret the experimental data. It described the system under isothermal conditions and considered an axial differential mass balance in terms of partial pressure for each chemical species. The simulation study and the experimental results show a satisfactory agreement and both highlight the possibility to shift towards 100% the conversion of the considered reaction.

Pd–Ag membrane reactors for water gas shift reaction

Abstract Pd–Ag thin wall permeators have been obtained by coating ceramic porous tubes with thin Pd–Ag metal foils (50xa0μm). A procedure of cold rolling and annealing has been used for producing thin metal foils. These membranes and membrane reactors have been proposed to recover hydrogen (and its isotopes) from tritiated water by using the water gas shift reaction, and by the reverse reaction (CO 2 conversion) for applications in the fusion reactor fuel cycle. The rolled membranes have been tested at 135–360xa0°C with a hydrogen transmembrane pressure in the range 130–180xa0kPa and hydrogen flow rates up to 1.02×10 −4 xa0molxa0s −1 . Both a complete hydrogen selectivity and a good chemical and physical stability have been observed through long-term tests. The tests on the membrane reactors have been carried out at the temperature of 325–330xa0°C with a feed pressure of 100xa0kPa; in particular, reaction conversion values close to 100% (well above the equilibrium value, about 80%) have been attained with the water gas shift reaction. These tests have demonstrated their applicability to the fusion fuel cycle as well as to the hydrogenation or dehydrogenation processes involving the use or the production of highly pure hydrogen.

Sputtered, electroless, and rolled palladium-ceramic membranes

Abstract Three techniques were used to produce palladium–ceramic (Pd–ceramic) composite membranes for hydrogen separation and production. They are sputtering, electroless deposition and rolling of thin Pd alloy films over ceramic porous tubes. After studying and developing the three coating techniques, an extensive testing and characterizing work was carried out on these thin film composite membranes. The results show that in the sputtered (0.5–5xa0μm) and electroless (2.5–20xa0μm) composite membranes, the thermal cycling of the hydrogenated metallic layer produces membrane failures. Such failures are characterized by crack formation and metal film peeling. This fact has been explained by an evaluation of the shear stresses at the metal–ceramic interface due to the differential elongation between the palladium (Pd) coating and the ceramic support under thermal cycling and hydrogen loading. The rolled membranes (50–70xa0μm), however, because of the particular coating solution, have shown a complete hydrogen selectivity and good chemical and physical stability in long-term tests.

Characterization of thin wall Pd–Ag rolled membranes

Abstract A thin wall Pd–Ag tube (thickness 50 μm ) obtained by a procedure of cold-rolling and annealing of thin metal foils has been characterized in long-term tests for determining the hydrogen permeability under operating condition of a temperature range from 300°C to 400°C and transmembrane differential pressure from 50 to 100 kPa . During testing, the physical and mechanical stability of the rolled membranes have been observed and high hydrogen fluxes have been measured. The long-term tests have also shown the modification of the surface structure of the membrane due to the hydrogen–metal interaction and the thermal cycling: as a consequence, an increase of the mass transfer properties of the hydrogen through the material (diffusivity and solubility) has been observed and the membrane tube has attained very high performances in about 3 months of operation in terms of hydrogen permeability. Furthermore, the tests have demonstrated that these Pd–Ag membranes have the capability to separate hydrogen from gas mixtures with a complete hydrogen selectivity and can be used to produce ultra-pure hydrogen for applications in energetic fields in membrane reactors by molecular reforming in membrane reactors (i.e. isotopic hydrogen separation, fuel cell, etc.).

Catalytic membrane reactors for tritium recovery from tritiated water in the ITER fuel cycle

Abstract Palladium and palladium–silver permeators have been obtained by coating porous ceramic tubes with a thin metal layer. Three coating techniques have been studied and characterized: chemical electroless deposition (PdAg film thickness of 10 μm), ion sputtering (about 1 μm) and rolling of thin metal sheets (50 μm). The Pd-ceramic membranes have been used for manufacturing catalytic membrane reactors (CMR) for hydrogen and its isotopes recovering and purifying. These composite membranes and the CMR have been studied and developed for a closed-loop process with reference to the design requirements of the international thermonuclear experimental reactor (ITER) blanket tritium recovery system in the enhanced performance phase of operation. The membranes and CMR have been tested in a pilot plant equipped with temperature, pressure and flow-rate on-line measuring and controlling devices. The conversion value for the water gas shift reaction in the CMR has been measured close to 100% (always above the equilibrium one, 80% at 350°C): the effect of the membrane is very clear since the reaction is moved towards the products because of the continuous hydrogen separation. The rolled thin film membranes have separated the hydrogen from other gases with a complete selectivity and exhibited a slightly larger mass transfer resistance with respect to the electroless membranes. Preliminary tests on the sputtered membranes have also been carried out with a promising performance. Considerations on the use of different palladium alloy in order to improve the performances of the membranes in terms of permeation flux and mechanical strength, such as palladium/yttrium, are also reported.

Numerical approach for a study of the hydrogen isotopes separation by palladium alloy membranes

In this work, a simple and stable method is proposed to design a hydrogen permeator operating with Pd/Ag membranes in countercurrent or cocurrent mode. The method is based on an iterative calculation that requires as input data a tentative value of the outlet permeate composition. The isotopic competition effect in the permeation process has been included in the flux formula. The model supported by a computer code has been used to study the hydrogen isotopes extraction from the gas stream arising from a tritiated water gas shift reactor under conditions relevant to a solid breeding blanket of ITER (International Thermonuclear Experimental Reactor) size. The model and the code have also been tested with experimental data reported in the open literature.

Analysis of tritium permeation in the steam generators of the SEAFP/SEAL fusion power reactor

Abstract The key uncertainty in estimating tritium permeation from fusion reactor cooling system components is the permeability coefficient, which is strongly dependent on the component material, pressure, temperature, chemistry, etc. Published tritium permeability values, for typical cooling system materials, range at least six orders of magnitude, resulting in estimates of tritium permeation carrying a large degree of uncertainty. In this work, both theoretical and empirical approaches are used for estimating tritium permeation and environmental releases for the water cooled SEAFP reactor. The steam generators are the major source of tritium permeation from the primary coolant. The theoretical model is based on the thermodynamic equilibria and the mass transfer phenomena at steady state conditions. A calculation code, from the theoretical model, allowed us to describe the behavior of the system under the main operating conditions. A parametric analysis has been carried out by varying the hydrogen concentration. The beneficial effect of oxide layers in reducing the permeation of hydrogen isotopes through steam generator tube materials has been observed experimentally by others, and is modeled here by a permeation reduction factor (PRF). The oxide layer effect has also been observed in operating steam generators, as demonstrated in the analysis based on CANDU operating experience. Such experience, which covers the range of conditions of the water cooled SEAFP reactor, has been used to estimate tritium permeability for the SEAFP steam generators. Hence, the use of measured data from CANDU plants has reduced the range of uncertainty in tritium permeability to approximately one order of magnitude. It has also permitted a meaningful estimate of tritium permeation to be performed. The best estimate is ≈16 Ci year−1 per steam generator.

Tritium Migration in HCLL and WCLL Blankets: Impact of Tritium Solubility in Liquid Pb-17Li

The next generation of fusion power plants (DEMO) should rely on a breeding blanket (BB) able to efficiently convert the neutrons kinetic energy into heat, to ensure the tritium self-sufficiency and to adequately shield the toroidal field coils from neutrons and gamma rays. The eutectic lithium-lead alloy is a consolidate liquid blanket material, which simultaneously includes the breeder (Li) and the neutron multiplier (Pb). The assessment of the tritium inventory inside the blanket and its environmental release requires knowledge of the hydrogen isotopes interactions with blanket materials, in particular the hydrogen solubility in lithium-lead, which is defined by means of the hydrogen Sieverts constant. Several experiments, aiming to determine the hydrogen isotopes solubility in lithium-lead, have been performed in the past giving values of the temperature-dependent Sieverts constant, K , distributed in a wide range (covering about two orders of magnitude on the Arrhenius plot). Starting from a literature review of the K values, this paper provides a parametric analysis for the most influencing parameters related to tritium migration in helium-cooled and water-cooled lead-lithium DEMO BBs. This analysis has been performed using the computational code FUS-TPC for several operative scenarios and considering the different K values. This paper demonstrates that the tritium Sieverts constant in Pb-17Li has a great impact on the assessment of tritium losses, whose value can spread in more than one order of magnitude. Furthermore, the analysis suggests suitable permeation reduction factors to be adopted in the different scenarios as well as the need of addressing new accurate experiments on the solubility constant.

Tritium inventory and permeation in the ITER breeding blanket

Abstract A model has allowed us to perform the analysis of the tritium inventory and permeation in the international thermonuclear experimental reactor (ITER) breeding blanket under the hypothesis of steady state conditions. Li 2 ZrO 3 (reference) and Li 2 TiO 3 (alternative) have been studied as breeding materials. The total breeder inventory assessed is 7.64 g for the Li 2 ZrO 3 at reference temperature. The model has also been used for a parametric analysis of the tritium permeation. At reference temperature and purge helium velocity of 0.01 m/s, the HT partial pressure is ranging from 10 to 30 Pa in the breeder and 1.5×10 −3 Pa in the beryllium. At 0.1 m/s of purge helium velocity, the HT partial pressure is reduced of one order by magnitude in the breeder and becomes 5×10 −5 Pa in the beryllium. The tritium permeation into the coolant for the whole blanket is ranging from 100 to 250 mCi per day for purge helium velocity of 0.01 m/s. The analysis of the tritium inventory and permeation for the alternative Li 2 TiO 3 breeding material has been carried out too. The tritium inventory in the breeder is in the range from 6 to 375 g larger than in Li 2 ZrO 3 by about a factor 5; the tritium permeation into coolant is comparable to the Li 2 ZrO 3 one. This analysis provides indications on the influence of the operating parameters on the tritium control in the ITER breeding blanket; particularly the control of the tritium inventory by the temperature and the tritium permeation by the purge gas velocity.

Impact of tritium solubility in liquid Pb-17Li on tritium migration in HCLL and WCLL blankets

The next generation of fusion power plants (DEMO) should rely on a breeding blanket able to efficiently convert into heat the neutrons kinetics energy, to ensure the tritium self-sufficiency and to adequately shield the Toroidal Field Coils from neutrons and gamma rays. The eutectic lithium-lead alloy is a consolidate liquid blanket material which simultaneously includes the breeder (Li) and the neutron multiplier (Pb). The assessment of the tritium inventory inside the blanket and its environmental release requires the knowledge of the hydrogen isotopes interactions with blanket materials, in particular the hydrogen solubility in lithium-lead which is defined by means of the hydrogen Sieverts constant. Several experiments, aiming to determine the hydrogen isotopes solubility in lithium-lead, have been performed in the past giving values of the temperature-dependent Sieverts constant, KS, distributed in a wide range (covering about two orders of magnitude on the Arrhenius plot). Starting from a literature review of KS values, this work provides a parametric analysis for the most influencing parameters related to tritium migration in Helium-Cooled and Water-Cooled Lead-Lithium DEMO breeding blankets. This analysis has been performed by using the computational code FUS-TPC for several operative scenarios and considering the different KS values. This study demonstrates that the tritium Sieverts constant in Pb-17Li has a great impact on the assessment of tritium losses, whose value can spread in more than one order of magnitude. Furthermore, the analysis suggests suitable permeation reduction factors to be adopted in the different scenarios as well as the need of addressing new accurate experiments on the solubility constant.