High-temperature all-metal induction furnace for melting samples of minerals and / or rocks for extracting gases under ultra-high vacuum

Description

technical field

[0001] The present invention relates to the field of extraction of rare gases present in minerals and / or rocks.

[0002] It relates more specifically to a furnace operating by electromagnetic induction heating, at high temperature (HT) and under ultra-high vacuum (UHV). "High temperature" here and within the scope of the invention means maximum furnace temperatures of at least 1500°C, typically around 1800°C. "Ultra-high vacuum" here and within the scope of the invention means pressures inside the furnace chamber of 10⁻⁸ to 10⁻⁹ mbar, typically a pressure on the order of 5 x 10⁻⁸ mbar at 800°C.

[0003] It aims more specifically to improve the performance of existing UHV HT furnaces and other known extraction systems consisting of lasers. Previous art

[0004] The noble gases, also known as rare gases, are chemical elements belonging to group 0 of the periodic table. Under normal temperature and pressure conditions (273 K, 1 atm), they are monatomic gases with the respective symbols He (Helium), Ne (Neon), Ar (Argon), Kr (Krypton), and Xe (Xenon). Their saturated outer electron shells, with two electrons for He and eight for Ne, Ar, Kr, and Xe, give them a particular physical characteristic: chemical inertness with respect to other elements. This unique property makes them excellent geochemical tracers and leading geochronometers. Each noble gas has several isotopes: two for helium (3-4 < He); three for neon (20-21-22 < Ne) and argon (36-38-40 < Ar); six for krypton (75-80-82-53-84-86< Kr) and finally nine for xenon (124-126-128-129-130-131-132-134-136< Xe).

[0005] The elemental and isotopic composition of these gases has been constantly evolving since the accretion of the Earth 4.56 billion years ago through nuclear reactions, whether radiogenic (radioactivity), nucleogenic (nuclear reactions) or cosmogenic (production of isotopes by interactions with cosmic radiation).

[0006] Furthermore, the elemental and isotopic compositions of all terrestrial reservoirs containing rare gases (atmosphere, crust, and shallow and deep mantle) have also changed as a result of: to degassing; to mantle differentiation; to leaks (helium) into space at the level of the upper atmosphere; to human activity since the industrial revolution (supply into the atmosphere of radiogenic helium following the exploitation of fossil fuels (coal, gas, oil) and tritiogenic supply following nuclear tests in the atmosphere which generated 3< He by decay of 3< H).

[0007] Studying and understanding fluid-rock interactions at the crust and / or mantle level is a major challenge in geoscience and requires powerful tools to extract rare gases from rocks and / or minerals in order to access the geochemical information necessary to understand the processes mentioned above.

[0008] Several extraction techniques are known: one can refer to publication [1] which lists the most used techniques in research laboratories that allow access to rare gases trapped in rocks and minerals.

[0009] One technique involves heating samples in order to extract the rare gases they contain.

[0010] Extracting rare gases by heating samples has always been a major challenge in the geosciences. It requires high-performance tools to meet laboratory requirements, namely extracting gases trapped in rocks or minerals at high temperatures as quickly as possible in an ultra-high vacuum chamber where outgassing of rare gases from internal surfaces is negligible.

[0011] The heat extraction systems currently used by the scientific community for the aforementioned extraction can be classified into several categories: resistive furnaces (either double-walled or single-walled), bulb furnaces, electromagnetic induction furnaces with glass enclosures, and lasers. All these known systems are associated with a purification line and a mass spectrometer.

[0012] Double-walled resistive furnaces are based on a technology from the 1980s that was very widespread in laboratories: see for example publication [2], [3].

[0013] These are ovens that use an electric current to heat a resistance, most often made of tungsten, by Joule effect.

[0014] We represented in figure 1 , such a double-walled UHV high-temperature furnace comprising two independent chambers 10 and 20 which do not communicate with each other.

[0015] The first enclosure is the external enclosure 20 in which a tungsten resistance 21, powered from the outside by electrical through-holes 21T, is heated by Joule effect.

[0016] The radiation emitted by this device heats a second chamber, a tantalum (Ta) tube 10, to a high temperature. This tube is connected to an ultra-high vacuum purification line 11, which purifies the gases extracted from a sample heated in a metal crucible 12, such as one made of molybdenum (Mo), tantalum (Ta), magnesium oxide (MgO), alumina (Al₂O₃), or boron nitride (BN). The samples in the crucible can thus be melted at temperatures typically between 1600 and 1800°C.

[0017] The ultra-high vacuum (UHV) purification line 11 includes chemical (chemisorption) and physical (physisorption) adsorption traps to purify the extracted gas. Chemisorption traps all reactive species (H₂O, CO₂, CxHy, etc.). Physisorption, generally used after chemisorption, separates noble gases based on their low-temperature condensation point on a surface (e.g., activated carbon).

[0018] The external enclosure 20 is connected via a flange not shown to a turbomolecular pumping unit 22, dedicated to the evacuation of gases, including oxygen, inside the external enclosure 20, in order to avoid damaging the heating element 21 by oxidation.

[0019] The main advantage of this type of double-walled furnace remains its double-chamber vacuum system, which facilitates the purification of extracted noble gases. Indeed, the hydrogen (H₂) produced at high temperatures by the tungsten heating element, which could degrade the efficiency of the chemical adsorption traps used in the purification line 11, is pumped out by the pumping unit 22.

[0020] However, this type of high-temperature, high-voltage (HV) furnace under UHV, which is state-of-the-art, has many disadvantages which can be summarized as follows.

[0021] The tantalum tube 10 is massive, with a significant Ta mass. It naturally contains large quantities of atmospheric noble gases, which, through diffusion at high temperatures, contaminate the noble gas measurements from the samples. Degassing this tantalum tube is mandatory to reduce the diffusion of atmospheric noble gases into the extraction chamber 20. This operation is often lengthy and tedious.

[0022] Furthermore, the heating rate of crucible 12 is slowed by the tantalum tube, which initially absorbs the radiation emitted by the heating element to raise its temperature. The crucible is then heated by conduction to reach the target temperatures, i.e., between 1600 and 1800°C. Approximately 15 to 20 minutes are required to reach these temperatures. The thermal inertia of the significant metallic mass, consisting of the crucible and the tantalum tube (typically 700 to 800 g), also hinders rapid cooling of the furnace. Another 15 to 20 minutes are required to cool the tantalum tube to a temperature below 500°C.

[0023] Achieving a seal between the Ta tube and the purification line is tricky and often a source of leaks. This seal is created using two flat flanges 13 between which a gold gasket 14 is compressed. This connection method is expensive simply because of the gold gasket. It is difficult to implement because the positioning of the gold gasket relative to the flanges 13 is delicate. Indeed, it tends to generate leaks.

[0024] Furthermore, the Ta tube recrystallizes under the effect of heat. Partial (or total) pumping of the rare gases extracted from the samples in chamber 10 to the second chamber 20 can occur through cracks resulting from the recrystallization of the tube via the secondary pumping unit 22. This phenomenon has been frequently observed. It gives erroneous results because, by definition, it underestimates the concentrations of the measured rare gases.

[0025] Furthermore, the tantalum tube is frequently damaged due to the melting of silicates present in the samples. The alloys formed between the tantalum and the molten silicates reduce the lifespan of the tantalum tube. Repeated high-temperature heating cycles cause the tantalum to flow towards the base of the tube, which then deforms and bulges. This deformation can be dangerous for the integrity of the furnace. Periodic replacement of the tantalum tube is costly, typically around €1500 per tube. It requires complete dismantling of the furnace, with the risk of damaging the expensive watt-based heating elements and heat shields.

[0026] As mentioned above, the first chamber 20 requires a dedicated secondary pumping unit to protect the heating elements from oxidation by atmospheric oxygen (which would destroy them). This pumping unit represents an additional cost for this extraction method, typically around €5,000. Maintenance of the pumping unit must also be factored into the cost, with an estimated price of €1,500 per service call.

[0027] Thus, the total cost of a double-walled UHV high-temperature furnace is very high, not only in terms of purchase, typically estimated by inventors at around €80,000 for a turnkey furnace, but also due to the maintenance required.

[0028] To eliminate the problems associated with the tantalum tube used in double-walled furnaces, the inventors of the present invention have created a single-walled furnace, as described in patent application FR2973105.

[0029] In essence, this single-walled furnace contains a gas extraction crucible and a heating element within the same chamber under ultra-high voltage (UHV). An electric current heats a tantalum heating element by Joule heating, into which the crucible containing the sample, made of boron nitride, is inserted. The crucible is then heated to approximately 1400-1450°C by radiation.

[0030] While this single-walled oven is generally satisfactory, it nevertheless has several drawbacks as follows.

[0031] First, the maximum temperature the furnace can reach is limited to approximately 1400-1450°C. It is not suitable for melting refractory samples such as olivine, which require a temperature of 1700-1800°C for melting. Gas extraction from refractory samples like olivine can, however, be carried out by diffusion. The sample particle size must not be too coarse, otherwise the extraction time will increase drastically. A particle size larger than 300-500 µm does not allow for a satisfactory extraction yield.

[0032] Next, the heating element is subjected to significant mechanical stress during heating, which weakens it. Its integrity can be preserved by gradually increasing the electrical power, but this comes at the expense of rapid heating. Therefore, a heating time of 20 to 30 minutes is recommended to reach the maximum temperature. Nevertheless, it is observed that this heating element is destroyed after an average of 40 to 50 cycles. This results in a significant operating cost for its replacement, which is around €350 per element.

[0033] Furthermore, the surface condition of the heating element fixings on the electrical passages must be particularly careful, otherwise the temperature reached by the oven will decrease.

[0034] Finally, boron nitride crucibles exhibit relative chemical inertness with respect to the elements present in minerals. However, they release significant amounts of nitrogen at high temperatures. Using Ta or Mo crucibles does not allow reaching 1450°C with this type of furnace.

[0035] Electromagnetic induction heating furnaces are also implemented for the extraction of rare gases from samples: see in particular publication [4] [5].

[0036] Such an induction oven is shown in figure 2A metal crucible 12, designed to receive the samples, is supported by a suitable silica support 15 and held within a double-walled glass enclosure 10. This enclosure is connected to a gas purification line 11 via a CF-type flange 13. A tube 16, which guides the sample to the metal crucible 12, is arranged on top of the crucible. The metal crucible is subjected to induced electric currents when subjected to a variable magnetic field supplied by an inductor 3. The inductor's windings 30, located outside the enclosure 10, surround the crucible 12 along its entire height. Water circulation within the double wall 17 of the enclosure 10 ensures the cooling of the furnace. The currents induced in the metal crucible 12 allow for a rapid temperature increase up to 1800°C, or even 2000°C if necessary.

[0037] The main drawbacks of this induction furnace are related to the use of glass for the chamber. Atmospheric helium diffuses through the chamber wall because glass is porous to this element. The residual quantities of helium detected in this type of chamber are therefore, by definition, significant, as highlighted in publications [4] and

[16] , and thus preclude the analysis of samples with low helium content. It is not possible to replace the glass chamber with a metal one due to the arrangement of the induction coils 30 on its exterior.

[0038] Furthermore, the metal crucible 12 rests on a crucible holder 15 made of silica. The differences in thermal expansion between the metal and the silica are the cause of the fracturing observed in the crucible holder during heating cycles. To minimize the extent of this fracturing, it is recommended to heat and cool the crucible slowly, which contradicts the need for a very rapid temperature increase in the furnace. Typically, approximately one hour is required to heat the crucible to a temperature of 1800°C.

[0039] Finally, the crucible's dimensions are very small. Its volume of 2.7 cm³ does not allow for the melting of large quantities of samples.

[0040] Bulb furnaces were also used for the extraction of rare gases: see publication [6]. A diagram was shown in figure 3Such a bulb furnace operates using a halogen lamp 4. Confining the light within a chamber 10 under UHV radiation heats a metal foil 6 containing a sample in the form of a powder or minerals. The light is transmitted into the chamber through a viewing window 5. The temperature is measured with a thermocouple 7 mounted on a dedicated feedthrough. This feedthrough is welded to a CF-type flange 13 at the end of the chamber 10. The chamber 10 is connected to a gas purification line 11 via another CF-type flange 13.

[0041] The main drawback of bulb furnaces is the maximum temperature reached, which cannot exceed 900 to 1000°C. These temperatures are insufficient for the extraction of rare gases present in rock and / or mineral samples.

[0042] In addition to all the known UHV high-temperature furnaces described above, CO2, diode, and Nd:YAG lasers (Neodymium-doped Yttrium aluminum garnet) are excellent tools for gas extraction. They allow for the rapid melting of a sample by focusing the energy of a laser beam onto its surface. This technique appears suitable only for the analysis of small samples with a mass between 0.1 g and 150 mg: see publications [7] and [8]. It would therefore be best suited to gas-rich samples.

[0043] There is therefore a need to improve the extraction of rare gases present in rock and / or mineral samples, in particular to overcome the drawbacks of state-of-the-art extraction systems as presented in the preamble, especially with a view to improving extraction performance, enabling efficient extraction and analysis from gas-poor samples, enabling rapid temperature rise up to high temperatures of at least 1800°C, enabling rapid cooling of the sample after melting, enabling easy maintenance in working order, while reducing manufacturing and maintenance costs.

[0044] Industrial vacuum induction furnaces are known for example from US 3300564A or EP3029165A.

[0045] The aim of the invention is to meet at least part of this need. Description of the invention

[0046] To this end, the invention relates, in one of its aspects, to a high-temperature, ultra-high-vacuum furnace for the extraction of rare gases present in mineral and / or rock samples, comprising: a metallic and airtight enclosure, particularly airtight against atmospheric gases, comprising in its upper part an opening for connection to a gas purification line for gases released into the enclosure and / or to a device for gravity-feeding a sample, and an opening for connection to a pump adapted to create ultra-high vacuum inside the enclosure; a crucible made of electrically conductive material, adapted to contain at least one sample of minerals and / or rocks; a support made of electrically insulating material, resting on the lower part of the metallic enclosure and supporting the crucible at a certain height within the enclosure; at least one induction coil, electrically powered from outside the enclosure by at least one insulated feedthrough through a wall of the enclosure, and arranged in the enclosure around the crucible; a tube made of electrically insulating material, resting on the lower part of the metallic enclosure.arranged between the induction coil and the crucible supported by the stand.

[0047] Thus, the invention essentially consists of providing a high-temperature UV furnace with a single metallic enclosure in which an electrically conductive crucible, preferably made of tantalum, is placed on an insulating support, preferably ceramic, and heated by induction by a coil wound around it. The insulating tube, preferably made of quartz, arranged between the induction coil and the crucible, advantageously serves as a surface on which condensable species can condense. The quartz insulating tube primarily protects the induction coil. Indeed, significant quantities of material from the samples are vaporized in the furnace. If these were to condense on the coil, they would fill the gaps between the turns of the induction coil, which would eventually touch. The consequence of this phenomenon would likely be a short circuit, which would probably induce a drop in the crucible temperature.

[0048] The oven according to the invention makes it possible to solve all the extraction problems encountered in research laboratories to quickly extract, under satisfactory analytical conditions, the rare gases trapped in geological samples (whole rocks, separated minerals).

[0049] The oven according to the invention also has many characteristics and advantages compared to all the extraction techniques used to date in research laboratories, which can be summarized as follows.

[0050] First, the furnace's single chamber is metallic. This allows it to be heated to a minimum of 200°C to degas the internal walls. By definition, it is impermeable to atmospheric noble gases. A metal-chambered furnace eliminates the problems encountered with state-of-the-art glass-chambered induction furnaces, namely the fragility of the structure in relation to the crucible when it is heated, but especially the porosity of the walls to helium. This porosity introduces significant quantities of atmospheric helium into the chamber of glass-chambered induction furnaces, adding further uncertainty to the measurements. This contribution can represent more than 99% of the total helium measured.

[0051] Furthermore, the heating of the furnace according to the invention is of the inductive type. This allows temperatures exceeding 1800°C to be reached. As a result, the furnace according to the invention thus surpasses, in terms of heating power, single-wall resistive heating furnaces according to the prior art and lamp furnaces according to the prior art, as presented in the prior art, which can only reach 1400°C and 1000°C respectively.

[0052] In the furnace according to the invention, the crucible, preferably made of tantalum, rests on an insulating support, preferably ceramic. This avoids problems associated with the tantalum tube used in prior art double-walled furnaces, namely recrystallization of the tube walls, which leads to partial or total pumping of the extracted gases by the secondary pumping unit used in this type of furnace, and to an underestimation of the calculated rare gas concentrations. It also eliminates mechanical and thermal stresses on the crucible, as it simply rests on the insulating support, preferably ceramic.

[0053] The effect of inductive heating, combined with the absence of mechanical and thermal constraints on the crucible, allows for a very high heating rate, reaching 1800°C in less than 5 minutes. Cooling is equally rapid. Currently, no laboratory furnaces with these characteristics exist internationally, whether single-walled resistive furnaces, double-walled resistive furnaces, bulb furnaces, or induction-heated furnaces with glass chambers. The time saved during heating, estimated by the inventors at approximately 15 to 20 minutes per extraction, undeniably improves the quality of rare gas extraction from samples, notably by reducing rare gas blanks.

[0054] The characteristics of the oven according to the invention (maximum temperature, heating and cooling speed, analytical blank) are reliable and reproducible over time.

[0055] Furthermore, the crucible degassing rate is rapid; a few hours at 1800°C is sufficient. The same applies to the walls of the metal chamber, typically at 200°C. This allows for the rapid attainment of highly satisfactory analytical conditions, namely very low residual quantities of atmospheric rare gases in the chamber, negligible compared to those extracted from a sample. Initial results concerning helium and neon place the furnace according to the invention among the highest-performing furnaces in terms of analytical blanks. This is explained by the very high temperature achieved, at least 1800°C, and by the small mass of the crucible to be degassed, typically around 120g. This same degassing process is much longer with a state-of-the-art glass induction furnace due to a significantly slower temperature rise time, necessary to minimize thermal effects on the crucible holder.This is also true with a state-of-the-art double-walled resistive furnace, due to the significant mass of the tantalum tube, typically around 700g, and the longer heating time. Finally, it is also true with a state-of-the-art single-walled furnace, due to the lower crucible degassing temperature (1400°C).

[0056] Furthermore, some geological samples exhibit very low gas concentrations. This characteristic has thus far necessitated melting large quantities of samples to extract sufficient rare gases for detection. The crucible according to the invention can have a significant volume, typically on the order of 15 cm³. This potential volume is up to five times greater than that offered by crucibles used in state-of-the-art glass induction furnaces. Therefore, this large potential volume for the crucible according to the invention makes it possible to consider melting substantial quantities of sample, typically 1 to 2 g per sample, in order to analyze gas-poor samples. This capability is, for example, not available to techniques using a laser as a heating element, which are limited by the diameter of the laser beam to melting only a few tens of milligrams of sample per analysis.And, for reasons of exorbitant cost, it is not advisable to use high-power lasers with a surface area at the focal point large enough to melt 1 to 2 g of sample.

[0057] Furthermore, changing the crucible is very easy, as it rests solely on an insulating support, preferably ceramic. Changing the crucible in double-walled resistance furnaces, according to current best practices, is much more complicated, as it requires dismantling the tantalum tube, with the risk of accidentally breaking the heating elements, the replacement cost of which is very high, estimated at over €12,000.

[0058] The maintenance cost of the furnace according to the invention is very low, as it consists solely of replacing the crucible. Typically, the cost of a tantalum crucible is €285. By comparison, the maintenance cost of single-walled resistive furnaces is higher, as it includes both the replacement of the crucible, currently around €50, and the periodic replacement of the heating element, estimated at €350. The maintenance cost of double-walled resistive furnaces is even higher, as it includes the replacement of the tantalum tube, estimated at €1,500, and the maintenance of the pumping unit, which costs approximately €1,500 per overhaul.

[0059] Finally, the overall cost for developing the furnace according to the invention, equipped with an HF signal generator, is quite acceptable for commercial purposes. Compared to double-walled resistive furnaces and techniques using a laser (CO2, diode, etc.) as a heating medium, the inventors estimate, as a first approximation, that the cost of a furnace according to the invention can be up to two to four times lower.

[0060] According to an advantageous embodiment, the oven includes a cooling circuit with a heat transfer fluid integrated into the side, bottom, and top walls of the chamber. Preferably, the heat transfer fluid is water.

[0061] According to an advantageous variant, the metal enclosure consists of cylinders joined together by welding and fixing flanges. Preferably, the cylinders and fixing flanges are made of stainless steel, preferably type 304L and 316LN respectively.

[0062] According to another advantageous variant, the distance between the induction coil and each of the lower, upper and lateral internal walls is at least equal to 10 mm.

[0063] The conducting crucible is preferably made of a material chosen from tantalum (Ta), molybdenum (Mo), platinum (Pt), or iron (Fe). In fact, all metal crucibles can be used within the scope of the invention.

[0064] The electrically insulating support is preferably made of a material chosen from among ceramics and quartz. In practice, any electrically non-conductive material can be used.

[0065] Preferably, the electrical insulating tube should be made of a material chosen from quartz, glass-ceramic, or vitreous carbon. Other electrical insulating materials may also be suitable.

[0066] According to an advantageous embodiment, the connection opening is linked to a bypass piece to connect the upper part of the enclosure to both the gas purification line and a storage carousel as a gravity-feeding device for a sample.

[0067] The oven according to the invention may also have one or both of the following advantageous characteristics: the lower end of the electrical insulating support is advantageously housed in a centering guide made in the lower inner wall of the enclosure; the lower end of the electrical insulating tube is housed in a centering guide made in the lower inner wall of the enclosure; the upper end of the electrical insulating tube is arranged as close as possible to the upper inner wall of the enclosure. Detailed description

[0068] Other advantages and features of the invention will become clearer upon reading the detailed description of illustrative and non-limiting examples of implementation of the invention, with reference to the following figures, among which: there figure 1 is a schematic longitudinal cross-sectional view of a double-walled, resistive high-temperature furnace under ultra-high voltage (UHV) according to the state of the art; the figure 2 is a schematic longitudinal cross-sectional view of a high-temperature, ultra-high-voltage (UHV) furnace of the electromagnetic induction heating type within a glass enclosure, according to the state of the art; the figure 3 is a schematic longitudinal cross-sectional view of a high-temperature, ultra-high-voltage (UHV) bulb-type furnace, according to the state of the art; the figure 4 is a schematic longitudinal cross-sectional view of a high-temperature, ultra-high-voltage (UHV) furnace of the electromagnetic induction heating type according to the invention; the Figures 5 and 5Aare respectively longitudinal cross-sectional and top-down views of the lower part of the oven according to the figure 4 ; THE Figures 6 and 6A are respectively longitudinal cross-sectional and top views of the upper part of the oven according to the figure 4 ; there figure 7 illustrates, in the form of a curve, the evolution of the temperature of the crucible of a furnace according to the invention as a function of the settings of the induction generator; the figure 8 illustrates, in the form of a curve, the heating and cooling rate of the crucible of a furnace according to the invention.

[0069] For the sake of clarity, the same references designating the same elements of a high-temperature ultra-high vacuum furnace according to the prior art and according to the invention are used for all the Figures 1 to 6A .

[0070] Throughout this application, the terms "vertical", "lower", "upper", "bottom", "top", "below", "above", "height" are to be understood by reference to an oven according to the invention with its enclosure in a vertical operating configuration.

[0071] It is specified that the different elements according to the invention are represented only for the sake of clarity and that they are not to scale.

[0072] It is also specified that the terms used for CF flanges are those commonly used for vacuum applications. Thus, CF flanges (also called ConFlat flanges) are knife-edge flanges with holes, made of 304 or 316 stainless steel, and used for ultra-high vacuum applications. The assembly, consisting of a copper gasket sandwiched between two CF flanges, is secured with nuts and bolts. CF flange sizes are indicated by the nominal inside diameter (DN) in millimeters. CF flanges are available in standard diameters from 16 to 300 mm (DN16 CF, DN40 CF, DN63 CF, DN100 CF, DN160 CF, DN200 CF, DN250 CF, DN300 CF).

[0073] THE figures 1 to 3 The aspects relating to high-temperature furnaces under ultra-high voltage (UHV) according to the prior art have already been discussed in the preamble. They are therefore not detailed below.

[0074] We now describe a high-temperature (HT) furnace under UHV conditions according to the invention as illustrated in Figures 4 to 6A .

[0075] The oven according to the invention 1 comprises first of all a gas-tight metallic enclosure 10, delimited by an internal diameter Di, a height H and defining an internal volume V. In the illustrated example, Di = 135 mm, H = 101 mm, and V ≈ 1450 cm 3< .

[0076] The metal enclosure 10 consists of an upper part 1S and a lower part 1I assembled together by two standard CF flanges 13S, 13I between which a copper gasket is interposed to ensure sealing against the atmosphere.

[0077] An upper flange 13C connects the interior of the furnace chamber to a gas purification line, not shown, via a tube 19. In the illustrated example, flanges 13S and 13I are DN 160 CF flanges, the upper flange 13C is DN CF40, and the gasket is of the OFHC type (Oxygen Free High Conductivity). More specifically, the upper flange 13C has a connection opening 4 in its center for a gas purification line for the gases released into the chamber. Preferably, the connection opening 4 is connected to a bypass fitting to connect the upper part 1S of the chamber to both the gas purification line and a storage carousel as a gravity-feed sample delivery device.

[0078] As illustrated in Figures 5 and 5AThe lower part 1I of the enclosure 10 comprises two hollow cylinders 100, 101 arranged concentrically and welded at their upper ends to the mounting flange 13I. In the example shown, the cylinders 100 and 101 have diameters of 159 mm and 135 mm, respectively. The lower part of the cylinders 100, 101 is welded to a third cylinder 102 via a circular plate 103, which forms the lower inner wall of the furnace enclosure.

[0079] The spacing between cylinders 100 and 101 allows water circulation to prevent overheating of the furnace's side walls. Water enters this water-cooling circuit RL via an inlet tube 105, while optimized filling is achieved through the discharge tube 106 welded between the two cylinders 100 and 101.

[0080] The third cylinder 102 is hollowed out in its center to allow a second water circulation to cool the lower inner wall 104 of the furnace. The water supply to this water-cooling circuit Ri is also provided by means of an inlet tube 105 and a discharge tube 106. These two tubes can be reversed because the direction of water flow in cylinder 102 is irrelevant.

[0081] A first 13L flange with a 19-inch tube is welded to the outer lateral cylinder 101 of the enclosure. The 19-inch tube is welded to both cylinders 100 and 101 to ensure a perfect seal of the enclosure against the atmosphere and water circulation. A second 13L flange is attached to the first 13L flange with a copper gasket to further seal the enclosure against the atmosphere. In the example shown, the 13L flanges are DN 63 CF flanges and the gasket is of the OFHC type.

[0082] The second flange 13L is equipped with an electrical feedthrough 3T allowing the passage of high-frequency (HF) electrical signals intended to power the induction coil 3, preferably made of copper, arranged within the enclosure. The feedthrough 3T preferably consists of two copper tubes soldered onto ceramics, which are themselves hermetically sealed onto the flange 13L.

[0083] The induction coil 3 is soldered to the two copper tubes of the electrical feedthrough 3T for HF signals. Preferably, the coil is arranged in the center of the enclosure 10.

[0084] A metallic crucible 12, which is the site of the thermal phenomenon by electromagnetic induction, is placed on an electrically insulating support 15, preferably in the center of the enclosure 10.

[0085] Crucible 12 is preferably made of tantalum because, in addition to being a refractory metal with a high melting point (3020°C), it has the advantage of exhibiting a low outgassing rate of atmospheric rare gases. However, other metals can be considered for crucible 12, such as molybdenum (Mo), platinum (Pt), iron (Fe), etc. For example, platinum is preferred for extracting nitrogen from geological samples because platinum remains chemically inert with respect to nitrogen (N₂), which is not the case for tantalum or molybdenum.

[0086] The support 15 is preferably made of ceramic because it is a pure material, releases little gas, and is resistant to temperature, including brief temperature variations, typically on the order of 400°C per minute. Furthermore, ceramic is easy to machine. Any other electrically non-conductive material can also be suitable, particularly quartz.

[0087] The number of turns 30 in the induction coil is calculated based on the geometry and mass of the crucible 12. The distance between the induction coil 3 and the internal lateral walls 101, lower walls 103, and upper walls 104 is advantageously at least 30 mm, in order to minimize the effects of induction on the metal enclosure 10.

[0088] Preferably, a centering guide 150 is machined into the lower wall 103 to best position the support 15 in the center of the enclosure. The height of the support 15 is calculated so as to position the crucible 12 in the center of the turns 30 of the induction coil 3.

[0089] An electrically insulating tube 18 is positioned between the induction coils 30 and the metal crucible 12. The tube is preferably made of quartz, as it is a temperature-resistant material. Any other tube made of a non-conductive material could be suitable, particularly a glass-ceramic or vitreous carbon tube. This tube 18 is designed to facilitate cleaning of the chamber 10, since most condensable species will condense on the surface of the tube 18 and not on the walls of the chamber 10. The tube 18 also effectively protects the induction coil 3 from the condensation of volatile species.

[0090] Advantageously, the upper end of tube 18 is positioned as close as possible to the upper inner wall 104 of the enclosure 10. In the illustrated example, the upper part of tube 18 is approximately 2 to 3 mm from the upper wall 104 of the enclosure. This small distance optimizes the condensation of condensables on tube 18 rather than on the walls of the enclosure. Tube 18, preferably made of quartz, is inexpensive and can therefore be replaced as needed.

[0091] Preferably, a centering guide 180 is machined into the lower wall 103 to best position the tube 18 in the center of the enclosure.

[0092] As illustrated in Figures 6 and 6AThe upper part 1S of the chamber includes a flange 13S onto which the upper flange 13C with tubing 19 has been welded. A cone 130 has been machined into the flange 13S to best guide the samples towards the crucible 12 during their descent by gravity.

[0093] A fourth cylinder 104, forming the upper inner wall, was welded to the flange 13S. Cylinder 104 is hollowed out to allow water circulation. The water supply to this water cooling circuit Rs is also provided by means of a connection between an inlet tube 105 and a discharge tube 106.

[0094] These two tubes can be reversed because the direction of water flow in cylinder 104 is irrelevant. This water circuit Rs is designed to prevent overheating of the upper part 1S of the furnace chamber. The flange 13C with tubing 19 has an internal connection opening 4 for connecting the furnace chamber to either a sample carousel or a gas purification line. The entire assembly is connected to a mass spectrometer for measuring rare gases.

[0095] All the mounting flanges 13S, 13I, 13L, and 13C, as well as the tubes 19 and the cylinders 100, 101, 102, 103, and 104, which make up the metal enclosure, are made of 304L and 316LN stainless steel. These steels were chosen for their excellent weldability and very low helium permeability. Their mechanical strength allows the enclosure to be baked at temperatures from 10 to 300°C to desorb chemical species adsorbed onto its internal walls. The material used for the side flange with tube 13L is preferably non-magnetic 316LN stainless steel to minimize the parasitic effects of induction on the walls of tube 19.

[0096] The crucible temperature was measured through a glass window using an optical pyrometer marketed under the name "infratherm IS 8 plus" by Impac. A temperature calibration was obtained based on the heating parameters of an HF signal generator supplying the induction coil 3 via the feedthrough 3T.

[0097] As previously described, all surfaces of the furnace's metal enclosure 10—namely, the lateral surface 101, the lower surface 103, and the upper surface 104—are cooled by a water circuit, designated Ri, RL, and Rs, respectively. This cooling system is particularly effective, as these surfaces remain cool despite the prolonged heating of the crucible 12 to 1850°C. It should be noted that in the illustrated example, the tube 19 of the lateral flange 13L is not equipped with a water cooling system. External fans can be used to limit the temperature at this point to 70°C.

[0098] After assembly, and to meet the cleanliness standards required for ultra-high vacuum, the enclosure was cleaned ultrasonically in three successive baths of detergent, the commercial brand "Decon 90," and finally in 99% ultra-pure acetone. Between each bath, the enclosure was rinsed with demineralized water. This cleaning procedure removes 99.95% of the hydrocarbons present in the enclosure.

[0099] The inventors conducted several tests with the oven according to the invention just described, in order to validate its use in the specialized laboratory where they work. The tests are summarized below. Reproducibility of heating

[0100] The crucible was heated to high temperature three times. These three heating cycles were carried out over three consecutive days. All measurements appear consistent and indicate very good reproducibility of the heating process. figure 7illustrates the three heating cycles obtained. From these curves, it can be seen that it is possible, using a logarithmic equation, to predict the temperature of the crucible as a function of the settings of the HF signal generator. Maximum temperature reached

[0101] During these three heating cycles, a measurement at 1850°C was able to be carried out with the optical pyrometer, used for calibration, which was set for the emissivity of tantalum.

[0102] It is possible to increase this extreme temperature further because the HF signal generator was only at 40% of its maximum power.

[0103] The furnace was opened after each heating cycle to check its integrity. The inventors were able to confirm that no part inside the furnace (crucible 12, ceramic support, induction coil 3, quartz tube) had been damaged by the temperature. Verification of measurements

[0104] Further temperature measurements were taken using the optical pyrometer.

[0105] The melting of copper and nickel chips in the crucible at temperatures of 1060°C and 1435°C respectively was observed.

[0106] These measurements are in agreement with the melting temperatures of these two metals found in the tables: see publication

[31] . Heating and cooling speed

[0107] As already described, the tantalum crucible 12 is heated by induction. It rests on a ceramic support to limit heat loss by conduction.

[0108] There figure 8 illustrates the temperature measurement points as a function of time.

[0109] The temperature rise curve is extremely rapid as only 200 seconds are needed to stabilize the temperature of the crucible from 12 to 1500°C.

[0110] The inventors believe that there is currently no high-temperature UHV furnace that exhibits this characteristic.

[0111] Crucible 12 also cools down very quickly when the HF signal generator is switched off because its mass, and therefore its thermal inertia, is low. Typically, the mass of crucible 12 does not exceed 120g.

[0112] As is apparent from the figure 8 , a duration of only 160 seconds is required to reduce the temperature of the crucible from 1500°C to a temperature below 500°C. Temperature stability

[0113] The parameters of the HF signal generator were chosen to obtain a temperature of 1500°C. The temperature stabilized after 200 seconds and measurements were taken for a period of approximately 20 minutes.

[0114] The average temperature during this test was 1499°C + / - 14°C.

[0115] This result seems to indicate that the oven according to the invention has satisfactory stability during a heating cycle. Furnace degassing protocol: Neon analysis.

[0116] A furnace degassing protocol was developed to obtain very low residual quantities of neon (white). An analytical white represents the amount of gas degassed from all the internal walls of a chamber. Low chamber degassing is essential for analyzing gas-poor samples.

[0117] Each blank analysis was performed after 25 minutes of heating crucible 12 under static vacuum to simulate gas extraction. Temperature (°C) (degassing time) 20Ne (mole) Degassing 1800 (1 hour) Crucible white 1500 9.84x10 -15< Degassing 1800 (2 hours) Crucible white (n=1) 1500 1.6x10 -16< Crucible white (n=7) 1500 5.82x10 -17<

[0118] This protocol was developed after drying the chamber 10 of the oven according to the invention at 160°C for 24 hours.

[0119] The degassing of crucible 12 is very effective in chamber 10 of the furnace.

[0120] Two degassing cycles at 1800°C for 3 hours yielded very satisfactory neon blanks, namely 5.8 x 10⁻¹⁷ moles of 20Ne degassed by the crucible at 1500°C for 25 minutes. This value is perfectly suitable for analyzing samples with low neon content.

[0121] As shown in Table 1 below, the comparison of this analytical result of the oven according to the invention with other extraction systems, from world-renowned laboratories, ranks the oven according to the invention among the most efficient in the world.

[0122] It should be noted that in Tables 1 and 2, the following codes are used to specify the type of extraction system for comparative examples according to the state of the art: (A): Double-walled resistive furnace (B): Single-walled resistive furnace (C): Laser (D): Glass-enclosed induction furnace TABLE 1 Examples Extraction type Temperature in °C Nature of the crucible White 20< Ne (x10 -16< mole) Comparative examples according to publication [9] D 1500 - 6.7 according to publication

[15] D 1500 - 6.69-11.15 according to publication

[16] D 1600 Mo 11.15 according to publication [3] A 1500 Your 4.5 1500 Al 2 O 3 22 according to publication

[10] A 1400 Your 1.34-4.02 according to publication

[11] A 600-1500 Your 1.75 600-1500 Al 2 O 3 26.29 according to publication

[12] A 1600 MgO 3.3 (average) 1600 BN 3.5 (average) 1600 Mo 4.5 (average) according to publication

[17] A 1650 Mo 26.77 according to publication

[18] A 1750-1800 - 0.04 according to patent application FR2973105 B 1150 BN 2.0 1250 BN 31 According to publication

[13] B 2000 W 21.86 According to publication

[14] B 2200 Your 0.02-7.14 According to publication

[30] B 1750 2.01 According to publication

[27] C 3.5-5 According to publication

[29] C 0.11 According to publication

[30] C 8.3 Induction oven according to the invention 1500 Your 0.58

[0123] It is entirely possible to further reduce these neon whites by increasing the degassing temperature of the crucible to 1900°C for example. Furnace degassing protocol: Helium analysis

[0124] A furnace degassing protocol has been developed to obtain very low residual quantities of helium (white).

[0125] Heating the furnace to 1830°C for 30 minutes under turbomolecular pumping reduces furnace outgassing to 1.4 x 10⁻¹⁵ moles of 4He. These quantities are suitable for helium extraction from gas-poor samples.

[0126] From this table 2, we can see that, compared with other furnaces from world-renowned laboratories, the analytical result ranks the furnace according to the invention among the most efficient in the world for helium extraction. TABLE 2 Examples Extraction type Temperature °C Extraction time (min) White 4< He (x10 -15< mole) Comparative examples according to publication [4] D 600 30 46 1800 90 according to publication [9] D 1500 847.7 according to publication

[15] D 1500 446-669 according to publication

[16] D 1600 669 according to publication [8] A 1650 30 66.92 according to publication

[17] A 1650 30 17.85 according to publication

[10] A 600-1700 20.08 according to publication

[12] A 800 30 64.25 1600 30 66.03 600-800 20-30 - 1600 20-30 48.18 1800 30 154.81 800 20 - 1600 20-30 75.40 1800 20 130.72 according to publication

[19] A 1600 8.80 according to publication

[20] A 300-1500 2.23-5.80 according to publication

[21] A - 1.33-1.78 according to publication

[22] A 1700 0.33 according to patent application FR2973105 B 1450 1.12 according to publication

[14] B 2200 0.268-3.569 according to publication

[13] B 2000 10 133.845 according to publication

[23] B 1700 0.997-3.156 according to publication

[24] C 0.067-0.152 according to publication

[25] C 1300 3.322 according to publication

[26] C >1200 0.216 according to publication

[27] C 2.3-26 according to publication

[28] C 6 according to publication

[29] C 11.154 Examples of induction ovens according to the invention Example 1 1750-1830 10 1.4 Furnace degassing protocol: Xenon analysis

[0127] A furnace degassing protocol has been developed to obtain very low residual quantities of xenon (white).

[0128] Heating the furnace to 1800°C for three 30-minute cycles under turbomolecular pumping reduces furnace outgassing to 5.5 x 10⁻¹⁸ moles of ¹³²Xe. These quantities are suitable for xenon extraction from gas-poor samples.

[0129] From this table 3, we can see that, compared with other furnaces from world-renowned laboratories, the analytical result classifies the furnace according to the invention at a similar level for the extraction of xenon. TABLE 3 Example Extraction type Temperature °C Nature of the crucible White 132Xe (x10 -16< ) according to publication

[15] D 1500 - 0.004-0.009 according to publication [9] D 1500 Mo 0.005 according to publication

[32] D 1850 Your 0.194 according to publication

[16] D 1600 Mo 0.107 according to publication

[11] A 600-1500 Your 0.022 according to publication

[17] A 1650 Mo 0.223 1600 MgO 0.072 (average) 600-800 BN 0.020 (average) 1600 BN 0.045 (average) 1800 BN 0.052 (average) 1600 Mo 0.140 (average) 1800 Mo 0.058 (average) according to publication

[33] B 2150 Your 0.178 according to publication

[13] B 2000 W 0.004 according to publication

[13] B 2050 - 0.020 Examples of induction ovens according to the invention Example 1 0.055

[0130] In view of these results, and due to the very good reproducibility of the tests and the excellent reliability of the oven according to the invention (heating power, heating and cooling speed), the laboratory in which the inventors work has decided to use the oven according to the invention routinely in the laboratory to extract neon, xenon and helium from minerals (quartz, pyroxene, olivine, etc.).

[0131] Furthermore, the results obtained in neon, xenon and helium whites are very encouraging and lead the inventors to believe that Ar and Kr whites can be very weak with a furnace according to the invention as described.

[0132] Other variations and improvements can be envisaged without departing from the scope of the invention.

[0133] The invention is not limited to the examples just described; in particular, features of the illustrated examples can be combined in unillustrated variants. References cited

[0134] [1]: Zimmermann L. and Marty B. “Methods for extracting rare gases under ultra-high vacuum”, Les techniques de l’enseignement, J6632 (2014). [2]: T Staudacher, EK Jessberger, D Dorflinger and J Kiko. “A refined ultrahigh-vacuum furnace for rare gas analysis”. Journal of Physics E: Scientific Instruments 1978, Volume 11, Number 8, 781-784. [3]: Takaoka N. A low blank, Metal system for rare gas analysis. Mass spectrometry, 24, No. 1, 73-86 (1976). [4]: Marty B., Lenoble M. and Vassard N. “Nitrogen, helium and argon in basalt: A static mass spectrometry. » Chemical Geology (Isotope Geoscience Section), 120, 183-195 (1995). [5]: Chennaoui-Aoudjehane H., Modeling the solubility of rare gases He, Ne, Ar, Kr and Xe in silicate liquids at 1500°C. Thesis, 138 p. (1992) [6]: Farley KA, Reiners PW, and Nenow V., “An apparatus for high-precision helium diffusion measurements from minerals”, Anal. Chem., 71 2059-2061 (1999). [7]: Humbert, F., Libourel, G., France-Lanord, C., Zimmermann, L., & Marty, B. « CO2-laser extraction-static mass spectrometry analysis of ultra-low concentrations of nitrogen in silicates. » Geostandards Newsletter, 24(2), 255-260(2000). [8]: Foeken J.P.T., Stuart F.M., Dobson K.J., Persano C. and Vilbert D., A diode laser system for heating minerals for (U-Th) / He chronometry, Geochemistry, Geophysics, Geosystem, 7, N°4, 1-9 (2006) [9]: Moreira M. and Allègre C.J. « Rare gas systematics on Mid AtlanticRidge (37-40°N) » Earth and Planetary Science Letters, 198, 401-416 (2002).

[10] : Niedermann S, Bach W. and Erzinger.J « Noble gas evidence for lower mantle component in MORBs fromthe Southern East Pacific Rise : Decoupling of helium and neon isotope systematics » Geochemica et Cosmochimica Acta, 61, 2697-2715 (1997).

[11] : Honda M., McDougallI., Patterson D.B., Doulgeris A. and Clague D.A. « Nobles gases in submarine pillow basalt glasses from Loihi and Kilauea, Hawaii : A solar component in the Earth ».Geochemica et Cosmochimica Acta, 57, 859-874 (1993).

[12] : Maruoka T. and Matsuda J. « New crucible for noble gas extraction » Chemical Geology, 175, 751-756 (2001).

[13] Honda M., Reynolds J.H., Roedder E. and Epstein S. Noble gases in diamonds: Occurrences of solarlike helium and neon. Journal of Geophysical Research, 92, N°B12, 12.507-12521 (1987)

[14] Sumino H., Dobrzhinetskaya L.F., Burgess R. and Kagi H. Deep-mantle-derived noble gases in metamorphic diamonds from the Kokchetav massif, Kazakhstan. Earth and Planetary Science Letters, 307, 439-449 (2011)

[15] Becker R.H. and Pepin R.O. The case for a martian origin of the shergottites: nitrogen and noble gases in EETA 79001. Earth and Planetary Science Letters, 69, 225-242 (1984)

[16] Ott U. Noble gases in SNC meteorites: Shergotty, Nakhla, Chassigny. Geochimica et Cosmochimica Acta, 52, 1937-1948 (1988)

[17] : Jambon, A., Weber, H., Braun, O.« Solubility of He, Ne, Ar, Kr and Xe in a basalt melt in the range 1250-1600°C: Geochemical implications ». Geochemica et Cosmochimica Acta 50, 401-408, (1986).

[18] Lavielle B., Marti K., Jeannot J.P., Nishiizumi K. and Caffee M. The 36Cl-36Ar-40K-41K records and cosmoc ray production rates in iron meteorites. Earth and Planetary Science Letters, 170, 93-104 (1999)

[19] : Blard, P.-H. and Pik, R. « An alternative isochron method for measuring cosmogenic 3He in lava flows. » Chemical Geology, 251 (1-4). pp. 20-32. ISSN 0009-2541. (2008).

[20] : Aciego S.M., Depaolo D.J., Kennedy B.M., Lamb M.P., Sims K.W.W. and Dietrich W.E. « Combining 3He cosmogenic dating with U-Th / He eruption ages using olivine in basalt ». Earth and Planetary Science Letters, 254, 288-302 (2007).

[21] : Kurz M. In situ production of terrestrial cosmogenic helium and some application to geochronology. Geochimica et Cosmochimica Acta, 50, 2855-2862 (1986)

[22] : Williams A.J., Stuart F.M., Day S.J. and Phillips W.M.Using pyroxene microphenocrysts to determine cosmogenic 3He concentrations in old volcanic rocks; an example of landscape development in central Gran Canaria. Quaternary Science Reviews, 24, 211-222 (2005)

[23] : Blard P.H., Pik R., Lavé J., Bourlès D., Burnard P.G., Yokochi R., Marty B. and Trusdell. Cosmogenic 3He prodution rates revisited fom evidences of grain size dependent release of matrix-sited helium. Earth and Planetary Science Letters, 247, 222-234 (2006)

[24] : Ammon K., Dunai T.J., Stuart F.M., Meriaux A.-S. and Gayer E. Cosmogenic 3He exposure ages and geochemistry of basalts from Ascension Island, Atlantic Ocean. Quaternary Geochronology, 4, 525-532 (2009)

[25] : Farley K.A., Libarkin J., Mukhopadhyay S. and Amidon W. Cosmogenic and nucleogenic 3He in apatite, titanite, and zircon. Earth and Planetary Science Letters, 248, 451-461 (2006)

[26] : Foeken J.P.T., Day S. and Stuart F.M.Cosmogenic 3He exposure dating of the quaternary basalt from Fogo, Cape Verdes: Implications for tift zone and magmatic reorganisation. Quaternary Geochronology, 4, 37-49 (2009)

[27] : Füri E., Aléon-Toppani A., Marty B. and Libourel G. Effects of atmospheric entry heating on the noble gas and nitrogen content of micrometeorites. Earth and Planetary Science Letters, 377-378, 1-12 (2013)

[28] : Pi T., Solé J. and Taran Y. (U-Th) / He dating of fluorite: application to the La Azul fluorspar deposit in the Taxco mining district, Mexico. Mineralium Deposita, 39, 976-982 (2005)

[29] : Nichols R. H. Jr., Hohenberg C.M. and Olinger C.T. Implanted solar helium, neon, and argon in individual lunar ilmenite grains: Surface effects and a temporal variation in the solar wind composition. Geochimica et Cosmochimica Acta, 58, 1031-1042 (1994)

[30] : Vermeesch P., Balco G., Blard P. H., Dunai T. J., Kober F., Niedermann S., Shuster D. L., Strasky S., Stuart F. M., Wieler R. and Zimmermann L.Interlaboratory comparison of cosmogenic 21Ne in quartz. Quaternary Geochronology, 26, 20-28 (2015)

[31] : CRC Handbook of Chemistry and Physics, 97th Edition.

[32] : Pujol M. Geochemistry of volatile elements in Archean rocks: characterization of ancient environments. PhD Thesis XXXp (2009)

[33] : Burgess R., Cartigny P., Harrison D., Hobson E. and Harris J. Volatile composition of microinclusions in diamonds from the panda kimberlite, Canada: Implications for chemical and isotopic heterogeneity in the mantle. Geochimica et Cosmochimica Acta, 73, 1779-1794 (2009)

[34] : Johnson LH, Burgess R., Turner G., Milledge HJ and Harris JW Noble gas and halogen geochemistry of mantle fluids: Comparison of African and Canadian diamonds. Geochimica et Cosmochimica Acta, 64, 717-732 (2000).

Claims

1. High-temperature ultrahigh-vacuum furnace (1), intended for the extraction of noble gases present in samples of minerals and / or rocks, comprising: - a metal chamber (10) that is gastight, in particular to atmospheric gases, comprising a connection opening for connecting to a purification line for purifying the gases released in the chamber (10) and / or to a gravity feed device for supplying a sample and, a connection opening (4) for connecting to a pump suitable for creating the ultrahigh vacuum inside the chamber (10), - a crucible (12) made of electrically conductive material, suitable for containing at least one sample of minerals and / or rocks, - a support (15) made of electrically insulating material, resting on the lower portion of the metal chamber (10), and supporting the crucible (12) in the chamber (10), - at least one induction coil (3, 30), powered from outside of the chamber (10) through at least one insulated bushing (3T) passing through a wall (100, 101) of the chamber (10), and arranged in the chamber (10) around the crucible (12), - a tube (18) made of electrically insulating material, resting on the lower portion of the metal chamber (10), arranged between the induction coil (3, 30) and the crucible (12) supported by the support (15).

2. High-temperature ultrahigh-vacuum furnace according to Claim 1, comprising a heat-transfer fluid cooling circuit (Rs, Ri, RL) integrated into the lateral (100, 101), lower (102, 103) and upper (104) walls of the chamber (10).

3. High-temperature ultrahigh-vacuum furnace according to Claim 1 or 2, the metal chamber (10) consisting of cylinders assembled together by welding and fastening flanges.

4. High-temperature ultrahigh-vacuum furnace according to Claim 3, the cylinders and fastening flanges being made of stainless steel, preferably of 304L and 316LN type respectively.

5. High-temperature ultrahigh-vacuum furnace according to one of the preceding claims, the distance between the induction coil (3, 30) and each of the lower (103), upper (104) and lateral (101) inner walls being at least equal to 10 mm.

6. High-temperature ultrahigh-vacuum furnace according to one of the preceding claims, the conductive crucible (12) being made of a material selected from tantalum (Ta), molybdenum (Mo), platinum (Pt), iron (Fe).

7. High-temperature ultrahigh-vacuum furnace according to one of the preceding claims, the electrically insulating support (15) being made of a material selected from a ceramic, quartz.

8. High-temperature ultrahigh-vacuum furnace according to one of the preceding claims, the electrically insulating tube (18) being made of a material selected from quartz, glass-ceramic, vitreous carbon.

9. High-temperature ultrahigh-vacuum furnace according to one of the preceding claims, the connection opening being connected to a bypass part for connecting the upper portion of the chamber (10) both to the gas purification line and to a storage carousel as a gravity feed device for supplying a sample.

10. High-temperature ultrahigh-vacuum furnace according to one of the preceding claims, the lower end of the electrically insulating support (15) being housed in a centring guide (150) made in the lower inner wall (103) of the chamber (10).

11. High-temperature ultrahigh-vacuum furnace according to one of the preceding claims, the lower end of the electrically insulating tube (18) being housed in a centring guide (180) made in the lower inner wall (103) of the chamber (10).

12. High-temperature ultrahigh-vacuum furnace according to one of the preceding claims, the upper end of the electrically insulating tube (18) being arranged as close as possible to the upper inner wall (104) of the chamber (10).