Hydrogen Inventory Research Articles

Hydrogen in the first wall of nuclear fusion plasma devices

Abstract The interaction of hydrogen isotopes with the inner walls of the plasma vessel is important for the physical, radiological and economical performance of controlled nuclear fusion devices. It can be subdivided into three major areas: (1) recycling phenomena describing the balance of incident and released hydrogen, (2) the development of hydrogen inventory in, and (3) the permeation through the first wall. In a first step, the composition and structure of the wall material must be characterized. It is the surface layer of this material where most interaction processes (trapping, reflection, release and diffusion) are controlled. In a second step, the radiation environment of the first wall must be defined; it can generally not be obtained by simulation experiments. Material composition and radiation environment are responsible for erosion and deposition of surface layers generally incorporating large amounts of hydrogen. IBA is the most important method of first wall analysis because it offers a choice of different analyzing methods in one device. Specific complications arise, however, because profiling of mobile hydrogen requires measurements during exposure; further, detrapping by the analyzing beam may in some hydrogen-containing layers amount to several thousand atoms per ion.

The surface chemistry of dissolving labradorite feldspar

Elastic recoil detection (ERD) analysis was used in conjunction with Rutherford backscattering (RBS) analysis to determine depth profiles of hydrogen, silicon, aluminum and calcium in labradorite crystals reacted under various pH conditions. The inventory of hydrogen in the mineral is strongly affected by solution pH. Hydrogen extensively infiltrates the mineral during reaction for 264 hours with solutions in the pH range 1–3. Infiltration is accompanied by extensive removal of sodium, calcium and aluminum from the mineral. This incongruent reaction proceeds to several hundreds of angstroms of depth and produces a silicon-rich surface which is amorphous to electron diffraction. The amount of hydrogen in the reacted layer is much less than is predicted from knowledge of the quantity of cations leached from the feldspar. These low inventories of hydrogen suggest that hydrogen-bearing groups in the reacted layer repolymermize subsequent to ion exchange and depolymerization reactions. This repolymerization eliminates hydrogen from the layer. At higher pH conditions (pH > 5), hydrogen inventories in the crystals decrease with time relative to an unreacted reference crystal. Hydrogen does not infiltrate beyond the first few unit cells of feldspar. Thus, dissolution in slightly acid, near-neutral, and basic solutions proceeds at the immediate surface of the feldspar. Within the limit of the RBS technique, there is no evidence for incongruent dissolution at these conditions.

Hydrogen and helium recycling in tokamaks with carbon walls

This paper presents a review of hydrogen and helium recycling phenomena in tokamaks with limiters and walls largely made out of carbon (graphite, a-C:H layers). The key points of interest are the plasma fuelling efficiency, the wall pumping phenomena as observed in JET, TFTR, TEXTOR and other machines under various fuelling schemes (gas, neutral beams, pellets), the release of hydrogen/helium from material surfaces during and after plasma discharges and the long term retention (total particle inventory) of hydrogen in graphite or carbonised structures in tokamaks. The effect of a combined hydrogen/helium plasma on recycling is also discussed. It is shown that only part of the above phenomena can be understood in terms of processes between hydrogen/helium and carbon as known from simulation experiments (ion beams, gas discharge facilities) and that others (in particular the JET wall pumping phenomenon) have still to be explained. Possible mechanisms are outlined and discussed by means of global models.

Hydrogen trapping and transport in carbon

The present paper reviews phenomena associated with the implantation of hydrogen isotopes into graphite and carbonized layers, with particular emphasis on the implanted layer, i.e. the depth range where the injected atoms are deposited. Different modifications of graphite and carbonaceous layers are characterized with respect to their structure, and the influence of radiation damage is treated. Recent evidence is demonstrated for the occurrence of local hydrogen and methane molecule formation, and the diffusion of these molecules through the bulk. The ‘saturated layer’ formed during hydrogen implantation is modelled by a dynamic balance between ion deposition and trapping, thermal or ion-induced detrapping, and local recombination and molecular diffusion. Corresponding calculations yield satisfactory fits to monoenergetic implantation and thermal effusion experiments. The inventory in the saturated layer is only a small contribution to the total hydrogen inventory of carbon walls in fusion machines. Some recycling phenomena can be explained by the physics of the implanted layer, whereas others, like wall pumping, require different mechanisms for their explanation.

Surface chemistry of labradorite feldspar reacted with aqueous solutions at pH = 2, 3, and 12

The reaction of feldspar with an aqueous solution is examined by complementing dissolution rate measurements with analysis of mineral surface chemistry. Rates of feldspar dissolution were measured in H 2O—HCl and D 2O—DCl solutions. These measurements were combined with elastic recoil detection (ERD) analysis of hydrogen isotope concentration, and Rutherford backscattering analysis (RBS) of silicon, aluminum and calcium concentrations near the surface of the mineral. Dissolution rates of labradorite feldspar in H 2O—HCl solutions ( pH = 1.7) are 33% more rapid than in D 2O—DCl mixtures ( pD = 1.7). The depth of penetration and inventory of hydrogen in the feldspar is a strong function of solution pH, temperature, and reaction time. Hydrogen infiltrates the feldspar more extensively from an acidic solution than from a basic solution, and complete isotopic exchange between the hydration layer and water proceeds in time on the order of hours. The hydrolysis of bridging Si—O—Al bonds by reaction with a strongly acidic solution for several hundreds of hours progresses to depths of several hundreds of Angstroms into the mineral. Calcium is also removed from the mineral to this depth during reaction with an acidic solution. The composition of the reacted surface, however, cannot be explained solely on the basis of ion exchange or depolymerization reactions. The data suggest that the silicon-rich surface of feldspar continually repolymerizes during reaction, and that this repolymerization eliminates hydrogen from the hydration layer.

Surface and bulk changes in iron nitride catalysts in [formula omitted] mixtures

Preparation of the ζ-, ε-, and γ′-iron nitride phases was confirmed by Mössbauer spectroscopy, X-ray diffraction, and quantitative mass spectrometry of NH 3 evolved during decomposition. Computer fitting of the ε-nitride Mössbauer spectra with a distribution of hyperfine fields shows the conversion of iron with two nitrogen nearest neighbors (Fe 2nn) to Fe 3nn as the nitrogen content increases. The dynamic response of the nitrides to H 2 CO mixtures at reaction temperatures was followed by constant-velocity Mössbauer spectroscopy and transient mass spectrometry. The rapid decomposition of the iron nitrides in H 2 at 523 K occurs with surface reaction as the rate-limiting step, initially. At lower temperatures or after significant nitride decomposition, the data are best fit with a shrinking core model. For reaction at 473 K, the Mössbauer effect identified an α-Fe shell, a ζ-Fe 2N core, and an ε-Fe xN transition region. Surprisingly, loss of the pure nitride phase is barely retarded for H 2 CO mixtures compared to H 2 alone at 523 K. Mass spectrometric studies show that the freshly prepared nitride has a substantial hydrogen inventory, equivalent to a monolayer of NH 3 for ζ-Fe 2N. On exposure to synthesis gas, the nitride catalysts produce no methane until one to two monolayers of N have been removed, but carbon is deposited on the catalyst by the Boudouard reaction. Mass spectral measurements show no evidence for active nitrogen on the surface after the synthesis reaction has been established. Both Mössbauer spectroscopy and mass spectral measurements confirm, however, that following the initial loss of nitrogen, bulk carbonitrides form which lose their nitrogen very slowly as the reaction proceeds. These data suggest that differences in the performance of iron and iron nitride catalysts may be strongly influenced by the way surface carbon is deposited during reactor startup.

Another view of the control of the isotopic composition of initial DT plasmas in tokamaks

The present paper reviews phenomena associated with the implantation of hydrogen isotopes into graphite and carbonized layers, with particular emphasis on the implanted layer, i.e. the depth range where the injected atoms are deposited. Different modifications of graphite and carbonaceous layers are characterized with respect to their structure, and the influence of radiation damage is treated. Recent evidence is demonstrated for the occurrence of local hydrogen and methane molecule formation, and the diffusion of these molecules through the bulk.The ‘saturated layer’ formed during hydrogen implantation is modelled by a dynamic balance between ion deposition and trapping, thermal or ion-induced detrapping, and local recombination and molecular diffusion. Corresponding calculations yield satisfactory fits to monoenergetic implantation and thermal effusion experiments.The inventory in the saturated layer is only a small contribution to the total hydrogen inventory of carbon walls in fusion machines. Some recycling phenomena can be explained by the physics of the implanted layer, whereas others, like wall pumping, require different mechanisms for their explanation.

Sticking coefficient of molecular and atomic hydrogen on palladium

Hydrogen recycle and inventory in first-wall materials of fusion reactors are strongly affected by surface and bulk properties—such as coefficients for hydrogen adsorption on the surface, and diffusion in the bulk. Here we present results for the sticking coefficients of molecular and atomic hydrogen on palladium, obtained from measurements of retained hydrogen subsequent to exposure to H2 and sub-eV H0 fluxes. Characteristic curves of the retained hydrogen versus incident fluence were used to determine the sticking coefficients αm and αa for H2 molecules and H0 atoms, respectively.

Simultaneous gas- and plasma-driven hydrogen transport in solids

The transport parameter formalism for evaluating the steady-state hydrogen (tritium) inventory, permeation rate, and recycle time for surfaces exposed to the plasma of an operating magnetic confinement fusion reactor has been extended to include synergistic effects of gas- and plasma-driven permeation. This steady-state formalism includes hydrogen trapping, recombination barriers to release at inner and outer surfaces, diffusion, and effects of thermal gradients (e.g., Ludwig–Soret effect) and is applicable to simultaneous plasma-driven and gas-driven hydrogen injection. In addition to providing a simple way of calculating the magnitude of these important tritium-related concerns, the transport parameter together with the ratio of the recombination coefficients at the two surfaces can also be used to classify the nature of these processes as to whether they are rate limited by bulk diffusion or surface recombination. New simple equations are derived which define these various regimes, and a new regime is identified which could influence permeation-probe measurements of plasma-edge fluxes.

Calculated particle inventories in the TFTR moveable limiter and first wall

A model for the time evolution of the particle inventories within the device components of TFTR has been developed. The DIFFUSE code is used to calculate the one-dimensional diffusion and trapping of particles in the wall and limiters. The DEGAS neutral transport code computes the charge-exchange fluxes onto these components. This combined model will be useful in determining the concentrations of isotopes in the device components, overall pumping by the wall and limiter during discharges, degassing rates between discharges, and the rate of isotope exchange in the machine. Example results for a 1.4 MA, 80 cm, ohmically heated discharge shaped by the TFTR moveable limiter are given. One preliminary finding was that the hydrogen inventory in the wall is comparable to that in the limiter. A bound on the diffusivity of hydrogen in graphite was also suggested by these calculations. Pumping of this discharge by the Zr/Al getter system was also simulated, and was found to be negligible.

The effect of bulk hydrogen inventory on the chemical erosion of graphite

Methane formation due to sub-eV H0 impact on pyrolytic graphite is characterized by an initial transient effect, followed by a steady state behavior. The nature of the transient is strongly affected by the preexposure history of the sample, viz., sample bakeout temperature during preparation, preheating of sample before H0 exposure, and previous exposure of sample to H0 fluxes. Preheating of the sample at ≥2200 K for several minutes results in the elimination of the transient. By contrast, the steady state CH4 production does not appear to be sensitive to previous sample history for samples initially baked at ≥1700 K.

Steady state hydrogen transport in solids

Abstract The analytical formalism for evaluating the steady state hydrogen (tritium) inventory, recycle and permeation rate and recycle time for surfaces exposed to the plasma of an operating magnetic confinement fusion reactor is reviewed and new material relevant to the application of this theory is presented. The formalism includes hydrogen trapping, diffusion, and effects of thermal gradients (e.g., Ludwig-Soret effect), and is applicable for all orders of release kinetics at the inner and outer surfaces. The problem is formulated in terms of a unitless transport parameter, W=(Rφ/D)(k1/φ)l/r, where r is the order of the release kinetics, R is the range of the implant, φ is the penetrating part of the incident flux, kl is the recombination coefficient and D is the diffusion coefficient. The steady state analytical theory is applied to several materials of interest to controlled fusion.

Steady state hydrogen transport in solids exposed to fusion reactor plasmas: Part I: Theory

A simple analytical transport parameter formalism is presented for evaluating the steady state hydrogen (tritium) inventory, recycle rate, and permeation rate for surfaces exposed to the plasma of an operating magnetic confinement fusion reactor. The formalism includes hydrogen trapping, recombination barriers to release at inner and outer surfaces, diffusion, and effects of thermal gradients (e.g., Ludwig-Soret effect). The transport parameter concept results in a natural classification of the transport behavior of hydrogen in all materials at temperatures for which the solutionized hydrogen is mobile. The analytical results provide physical insight into the processes governing hydrogen permeation and inventory and also allow simple checks on the accuracy of numerical calculations which must be used at early times in the buildup to steady state operation.

Thermally activated processes in hydrogen recycling

The thermally activated processes of hydrogen diffusion in wall materials such as stainless steels and inconels and of its surface release (including desorption) play an essential role in the hydrogen recycling in many fusion devices. Non thermally activated processes such as deposition, implantation and direct reflection (backscattering) of atomic particles, particle- and photon-induced desorption and trapping effects must also be considered. The thermal release of dissolved and adsorbed hydrogen occurs in the form of molecules at the wall temperatures of present devices and of plasma systems of the next generations. This leads one to expect that 1. (a) a large hydrogen concentration exists in the subsurface layers during the discharge and even a considerable time after its termination, 2. (b) pronounced memory effects should exist as evidenced during isotope changeover experiments; the behaviour should differ markedly when the second isotope is fed into the system at the start or when it is injected during the discharge. 3. (c) the hydrogen inventory in and its permeation through the wall should be larger than the value evaluated from its solubility at the filling pressure. These effects — and in particular their variation with the wall temperature and with the parameters of the impinging particles — are discussed and compared with available data obtained in tokamaks and in simulation devices. The domains of applicability of relevant computer codes are briefly addressed. The need for additional information on specific material data and existing methods to acquire them are indicated.

A simple theory for maximum h inventory and release: A new transport parameter

A simplified theory of hydrogen transport in materials is presented. The problem is formulated in terms of a dimensionless transport parameter, W = R(φK 0) 1 2 /D where R is the projected range, φ is the incident penetrating flux, K 0 is a phenomenological recombination coefficient and D is the diffusion coefficient for hydrogen in the material. Using this parameter, equations which describe recycle times, maximum hydrogen inventory and maximum hydrogen permeation are derived. This treatment yields simple analytic expressions for these quantities which are shown to be excellent approximations to the behavior obtained from exact numerical solutions. This transport parameter further results in a natural classification of the different types of behavior of hydrogen in materials.

Mechanically Refrigerated Liquid Hydrogen Target

A liquid hydrogen target which incorporates a mechanical refrigerator for cooling rather than using a liquid hydrogen reservoir or cold helium gas has been successfully operated at Argonne National Laboratory. Use of the mechanical refrigerator eliminates the hazard of a large liquid hydrogen inventory in the vicinity of the target. In addition, remote operation is greatly simplified over methods previously used. A commercially available 20.4 K refrigerator, standard gas handling and purifying components, and a specially constructed nonmetallic target flask comprise the mechanical portions of the target system. The target flask requires approximately 20 cc of liquid hydrogen. The system has operated satisfactorily in the external accelerated proton beam of the Zero Gradient Synchrotron.