Patent Application
20220118515
The present invention relates to powders based on niobium-tin compounds, in particular of the composition NbxSny where 1≤x≤6 and 1≤y≤5 for the production of superconducting components, wherein the powders have a low oxygen content, a process for the production thereof and also the use of such powders for the production of superconducting components.
Superconductors are materials whose electrical resistance drops to zero when the temperature goes below a particular temperature, known as the critical temperature. In the superconducting state, the interior of the material remains free of electric and magnetic fields and the electric current is transported without any losses. Superconductors are used, inter alia, for producing strong, constant magnetic fields or for producing low-loss transformers which for the same power have smaller dimensions and mass than conventional transformers and thus have advantages, especially in mobile operation.
Superconductors can be classified into various categories such as metallic superconductors, ceramic superconductors and high-temperature superconductors. Since, at the latest, the discovery of the critical temperature of niobium-tin (Nb3Sn) of 18.05 K, niobium and its alloys have moved into focus as materials for the production of superconductors. Thus, superconducting cavity resonators made of niobium are used, for example, in particle accelerators (including XFEL and FLASH at the DESY in Hamburg or CERN in Geneva).
Superconducting wires are of particular interest as superconducting components, and these are used, inter alia, for producing superconducting coils. Kilometer-long wires having conducting fibers/filaments having a thickness of only a few microns are generally necessary for strong superconducting coils, and these require complicated production processes.
For the production of such wires, in particular on the basis of niobium-tin alloys, recourse is made essentially to the bronze process in which a Cu—Sn alloy is used as starting material.
Thus, EP 0 048 313 describes superconducting wires based on bronze-Nb3Sn which can be employed at high magnetic fields and are characterized by a cubic phase in the bronze-Nb3Sn wire and comprise stabilizing alloy constituents from the group Li Be Mg Sc Y U Ti Zr Hf V Ta Mo Re Fe Ru Ni Pd Zn Al Ga In Tl Si Ge Sb in the percent by weight range from 0.01 to 7, based on the proportion of Nb, and/or from 0.05 to 10, based on the proportion of bronze in the wire, which largely prevent formation of a tetragonal phase and/or reduce tetragonal deformation (1−c/a).
As an alternative, superconducting wires based on niobium-tin alloys can be produced by the PIT (powder-in-tube) process in which a pulverulent tin-containing starting compound is introduced into a niobium tube and is then drawn to give a wire. In a last step, a superconducting Nb3Sn boundary layer is formed between the niobium-containing sheathing tube and the tin-containing powder introduced by means of a heat treatment. As regards the tin-containing starting compound, the phase composition, chemical purity and particle size, which must be no greater than the diameter of the finished filament, are critical.
T. Wong et al. describe, for example, the PIT process and the production of the tin-containing starting compound for the example of NbSn2 (T. Wong et al., “Ti and Ta Additions to Nb3Sn by the Powder in Tube Process”, IEEE Transactions on Applied superconductivity, Vol. 11, No. 1 (2001), 3584-3587). A disadvantage of the process is that a multistage process made up of milling steps and thermal treatments of up to 48 hours is necessary for a satisfactory reaction of niobium with tin to form NbSn2. Furthermore, the general teaching is that the oxygen content should be very low.
U.S. Pat. No. 7,459,030 describes a production process for a superconducting Nb3Sn wire by the PIT process, in which a tantalum-tin alloy powder is used as starting compound. To produce this, use is made of K2NbF7 and K2TaF7, which are reduced to the respective niobium metal and tantalum metal before the reaction with tin. However, the process described has the disadvantages of some restrictions to the use of these niobium and tantalum metals. Thus, only metals having a maximum content of oxygen of less than 3000 ppm and hydrogen of less than 100 ppm can be used. Exceeding the oxygen content leads to a lower quality of the finished wire. At hydrogen contents above 100 ppm, safety problems occur in the process, since the hydrogen escapes during the thermal treatment. Furthermore, the process described has the disadvantages that the target compounds contain a high content of unreacted tin and the finished wire core also contains tantalum-containing compounds, which can have an adverse effect on the superconducting properties of the wires. Furthermore, sparingly soluble metal fluorides such as MgF2 or CaF2 are formed in the reduction of the starting compounds K2NbF7 and K2TaF7, and these cannot be separated off completely. In addition, all fluorine-containing compounds in the process chain are very toxic.
A. Godeke et al. give an overview of the conventional PIT processes for the production of niobium-tin superconductors (A. Godeke et al., “State of the art powder-in-tube niobium-tin superconductors”, Cyrogenics 48 (2008), 308-3016).
M. Lopez et al. describe the synthesis of nano-intermetallic Nb3Sn by mechanical alloying and heat treatment at low temperatures (M. Lopez et al., “Synthesis of nano intermetallic Nb3Sn by mechanical alloying and annealing at low temperature”, Journal of Alloys and Compounds 612 (2014), 215-220). The Nb3Sn produced in this way has a proportion of 87% by weight of Nb3Sn and 8% by weight of NbO.
However, all processes known in the prior art for producing superconducting wires composed of Nb3Sn have the disadvantage that a significant proportion of oxygen is carried over to the target compounds by introduction of oxygen with the elements niobium and tin and also while carrying out the process, for example by means of air. For this reason, the process described in U.S. Pat. No. 7,459,030, for example, is restricted to the use of niobium and tantalum metal powders having an oxygen content of not more than 3000 ppm and tin having an oxygen content of not more than 2000 ppm. A high proportion of oxygen in the target compound can lead, inter alia, to occupation of the interstitial lattice sites by oxygen atoms and also to formation of a separate NbO phase, which can be detected by X-ray diffraction analyses. The niobium bound in this way is thus no longer available for further reactions such as the formation of the Nb3Sn boundary layer. In addition, the solid-state diffusion of tin and niobium which is necessary for formation of the boundary layer is hindered. This not only has an adverse effect on the yield and efficiency of the production process, but the presence of oxygen can also lead to significant impairment of the superconducting properties, for example the critical current density or the residual resistance ratio (RRR), of the target compound and of the wire.
It is therefore an object of the present invention to provide suitable starting compounds for the production of superconducting components, in particular superconducting wires, which starting compounds allow an efficient reaction without impairment of the superconducting properties of the target compounds.
It has surprisingly been found that this object is achieved by a powder which does not have any separate NbO or SnO phases.
The present invention therefore firstly provides a powder for producing superconducting components, comprising NbxSny where 1≤x≤6 and 1≤y≤5, wherein the powder does not have any separate NbO and/or SnO phases. This can be seen from, in particular, the powders not having any NbO and/or SnO reflections in the X-ray diffraction pattern, for example determined on pulverulent samples using an instrument from Malvern PANalytical (X'Pert-MPD with semiconductor detector, X-ray tubes Cu LFF with 40 KV/40 mA, Ni filter).
In a preferred embodiment, the NbxSnY compound is a compound selected from the group consisting of Nb3Sn, Nb6Sn5, NbSn2 and mixtures thereof.
Analyses of conventional powders as are provided by the prior art show that these have a separate NbO phase which shows up as reflections in the X-ray diffraction pattern, as can be seen from
In a preferred embodiment, the powders of the invention are characterized by the oxygen content in the powder being less than 1.5% by weight, preferably less than 1.1% by weight and particularly preferably from 0.2 to 0.75% by weight, based on the total weight of the powder. The oxygen content of the powder can, for example, be determined by means of carrier gas hot extraction (Leco TCH600).
Apart from a low oxygen content, the powder of the invention also displays excellent phase purity, which is revealed by, inter alia, it having only a small proportion of crystalline phases of compounds other than the respective niobium-tin target compound. In a preferred embodiment, the powder of the invention is therefore characterized in that the compounds Nb3Sn and/or Nb6Sn5 and/or NbSn2 make up a proportion of in each case more than 92%, preferably more than 95%, particularly preferably more than 98%, based on all crystallographic phases detected and determined by Rietveld analysis of an X-ray diffraction pattern of the powder of the invention.
In a preferred embodiment, the powders of the invention are characterized in that the powder comprises three-dimensional agglomerates having a size having a D90 of less than 400 μm, preferably from 220 to 400 μm, determined by means of laser light scattering, the agglomerates are made up of primary particles which have an average particle diameter of less than 15 μm, preferably less than 8 μm, determined by means of scanning electron microscopy, and the agglomerates have pores of which 90% or more have a diameter of from 0.2 to 15 μm, determined by means of mercury porosimetry.
The D90 is the value which indicates the percentage of agglomerates in the powder which have a particle size of less than or equal to the size indicated.
In the production of superconducting wires, it has also been found to be advantageous to use powders having a small particle size. For this reason, preference is given to an embodiment of the powder of the invention in which the powder has a particle size D99 of less than 15 μm, preferably less than 8 μm, particularly preferably from 1 μm to 6 μm, determined by means of laser light scattering. The D99 here is the value which indicates the proportion of particles in the powder which have a particle size of less than 15 μm. The particle size can be realized, for example, by milling of the powders.
For the production of superconducting components by additive manufacturing processes, for example LBM (laser beam melting), EBM (electron beam melting) and/or LC (laser cladding), it has been found to be advantageous to use powders having a particular spherical particle shape. Here, it has surprisingly been found that the powders of the invention can very readily be atomized by known methods to give powders having sphere-like particles, for example using the EIGA (electrode induction-melting gas atomization) method. In a preferred embodiment, at least 95% of all powder particles of the powder of the invention therefore have a Feret diameter of from 0.7 to 1, preferably from 0.8 to 1, after atomization, where the Feret diameter is for the purposes of the present invention defined as the smallest diameter divided by the greatest diameter of a particle, able to be determined by evaluation of SEM images.
The powder of the invention preferably has a specific surface area determined by the BET method of from 0.5 to 5 m2/g, preferably from 1 to 3 m2/g. The specific surface area determined by the BET method can be determined in accordance with ASTM D3663.
To produce superconducting components having acceptable properties, it is indispensable for the chemical purity of the powders used to be high and foreign substances to be introduced only in controlled form as dopants. Materials, in particular metallic impurities and fluoride-containing compounds, unintentionally introduced in the process should be minimized. In a preferred embodiment, the powder of the invention has a fluorine content of less than 25 ppm, preferably less than 10 ppm, where the ppm are by mass. In a further preferred embodiment, the powder of the invention has a total content of unintentional metallic impurities with the exception of tantalum of less than 0.8% by weight, preferably less than 0.5% by weight, particularly preferably less than 0.25% by weight, in each case based on the total weight of the powder.
In a preferred embodiment, the powder of the invention additionally contains dopants. The addition of suitable dopants makes it possible to adapt the properties of the powder as required, and it has surprisingly been found that the dopants do not have to meet any particular requirements but rather it is possible to use the customary dopants known to a person skilled in the art.
Some of the processes described in the prior art for producing superconducting wires based on Nb3Sn start out from a tantalum-tin alloy or from an intermetallic tin alloy based on tantalum and niobium as precursor powder. However, this has the disadvantage that residues of tantalum remain in the later Nb3Sn wire filament and the superconducting properties of the products can be impaired in this way. In the context of the present invention, it has surprisingly been found that the addition of tantalum can be dispensed with without the effectiveness of the reaction being adversely affected. In a preferred embodiment, the powder of the invention is therefore essentially free of tantalum and tantalum compounds. In a particularly preferred embodiment, the proportion of tantalum and compounds thereof in the powder of the invention is less than 1% by weight, preferably less than 0.5% by weight, particularly preferably less than 0.1% by weight, in each case based on the total weight of the powder.
The powders of the invention have a low oxygen content which is shown, inter alia, by no reflections for NbO and/or SnO being able to be detected in the X-ray diffraction pattern of the powders of the invention. The present invention therefore further provides a process for producing the powders of the invention, which process makes it possible to realize this property, where the process of the invention comprises the reaction of niobium metal powder with tin metal powder and also a reduction step in the presence of a reducing agent, where the amount of reducing agent added is based on the previously determined total content of oxygen in the two metal powders used. The reactant is one selected from the group consisting of magnesium, calcium, CaH2 and MgH2 and mixtures thereof.
In a preferred embodiment of the process of the invention, the niobium metal powder is reacted with tin metal powder in a first step and the product obtained is subsequently subjected to a reduction step in the presence of a reducing agent, where the amount of reducing agent added is based on the previously determined content of oxygen in the product obtained from the first reaction.
To make the process efficient, it has been found to be advantageous to carry out the reaction of the niobium metal powder with the tin metal powder directly in the presence of a reducing agent. For this reason, preference is given to an embodiment of the process of the invention in which the reaction of the niobium metal powder with the tin metal powder is carried out in the presence of a reducing agent.
It has surprisingly been found that the formation of separate oxygen-containing phases such as NbO and SnO can be reduced further when the reaction of the metallic starting compounds is carried out in the presence of a gaseous reducing agent. In particular, the reactant is one selected from the group consisting of magnesium, calcium and mixtures thereof. It has surprisingly been found that the use of these reducing agents, especially in the gaseous state, enables the formation of NbO and SnO phases in the powder to be reduced, while the residues of the reducing agent can be removed simply from the product powder without leaving a residue.
The removal of the oxidized reducing agent can be effected in a simple way by washing. For this reason, preference is given to an embodiment of the process of the invention in which the powder obtained is additionally subjected to a washing step. It has surprisingly been found that particularly efficient removal of any residues of the reducing agent can be achieved when mineral acids are used as washing liquid. For this reason, preference is given to an embodiment in which the washing step is washing with mineral acids. The mineral acids are preferably selected from the group consisting of sulfuric acid, hydrochloric acid and nitric acid.
As a result of the additional treatment with an amount of reducing agent based on the total content of oxygen in the process of the invention, the restrictions in respect of the oxygen content of the starting materials used, as in the prior art, for example as described in U.S. Pat. No. 7,459,030, no longer apply. Significantly higher oxygen contents are tolerable while achieving improved phase purity of the target compounds. Nevertheless, the oxygen content should not be too high. For this reason, preference is given to an embodiment of the process of the invention in which a niobium metal powder containing less than 3% by weight of oxygen, preferably from 0.4 to 2.5% by weight, particularly preferably from 0.5 to 1.5% by weight, and/or a tin metal powder containing less than 1.5% by weight, particularly preferably from 0.4 to 1.4% by weight, of oxygen is used, where the figures are in each case based on the total weight of the powder.
It has surprisingly been found that the morphology of the niobium metal powders used is not subject to any limitations. It is possible to use powders comprising porous agglomerates which consist of three-dimensionally connected primary particles or else powders consisting of irregular or spherical particles without porosity.
To prevent formation of sparingly soluble MgF2 and CaF2 in the powder of the invention, a niobium metal powder having a very low fluoride content is preferred. For this reason, niobium metal powders produced by reduction of niobium oxides are preferred over niobium metal powders produced by reduction of fluorine-containing compounds, for example K2NbF7. In a preferred embodiment, the niobium metal powder used contains less than 10 ppm of fluorine, preferably less than 5 ppm, particularly preferably less than 2 ppm.
The powder of the invention is particularly suitable for producing superconducting components. The present invention therefore further provides for the use of the powder of the invention for producing superconducting components, in particular for producing superconducting wires. The superconducting component is preferably produced by powder-metallurgical processes or additive manufacturing processes. In a preferred embodiment, the superconducting wires are produced by the PIT process.
The present invention further provides for the use of the powder of the invention in additive manufacturing processes. The additive manufacturing processes can be, for example, LBM (laser beam melting), EBM (electron beam melting) and/or LC (laser cladding).
The present invention will be illustrated with the aid of the following examples, but these should not be construed as constituting any restriction of the inventive concept.
Niobium metal powder was reacted with tin metal powder in the presence of magnesium as reducing agent under various conditions and the products obtained were washed with sulfuric acid and analyzed. Powders for which the reaction of the starting compounds was carried out conventionally without reducing agent and subsequent washing were employed as comparative experiments. The tin metal powder used had a particle size of less than 150 μm and an oxygen content of 6800 ppm in all experiments.
The results are summarized in table 1, with the information on the oxygen contents being determined by means of carrier gas hot extraction (Leco TCH600) and the specific surface area being determined by the BET method (ASTM D3663, Tristar 3000, Micromeritics). The particle size was in each case determined by means of laser light scattering (MasterSizer S, dispersion in water and Daxad11, 5 min ultrasonic treatment). The trace analysis of the metallic impurities such as Mg was carried out by means of ICP-OES using the following analytical instruments PQ 9000 (Analytik Jena) or Ultima 2 (Horiba). X-ray diffraction was carried out on pulverulent samples using an instrument from Malvern-PANalytical (X'Pert-MPD with semiconductor detector, X-ray tubes Cu LFF with 40 KV/40 mA, Ni filter).
The niobium metal powder used for producing the powders of examples 2 and 4 was obtained by a method analogous to the production process described in WO 00/67936 by reaction of NbO2 with magnesium vapor. The niobium metal powder obtained had an oxygen content of 8500 ppm, a hydrogen content of 230 ppm, a fluoride content of 2 ppm and an agglomerate size D50 of 205 μm and D90 of 290 μm. The average size of the primary particles was 0.6 μm and the pore size distribution of the agglomerates was bimodal with maxima at 0.5 and 3 μm. Such niobium metal powders display a high porosity which, contrary to expectations, does not lead to a higher oxygen content and formation of an NbO and SnO phase in the NbSn powder. Accordingly, niobium metal powders having a high porosity can also be used in the process of the invention.
In the case of the powders of example 1 and 3 and of the two comparative experiments, niobium metal powders according to the prior art without internal porosity of the particles were used, with these having an oxygen content of 2900 ppm, a hydrogen content of 10 ppm and a particle size having a D90 of 95 μm. Examples 1 and 3 show that a low oxygen content and the avoidance of the NbO and SnO phases can also be achieved using these starting materials.
The powder of example 2 was subsequently milled in an oxygen-free atmosphere, leading to a D90 of 3.1 μm and a D99 of 4.9 μm. It was surprisingly observed that milling of the powder did not lead, contrary to expectations, to an increase in the oxygen content, which was 0.78% by weight in the milled powder, nor to formation of an NbO and SnO phase.
It has also surprisingly been found that reaction of metals in the presence of magnesium does not lead to residues of the reducing agent remaining in the product. Rather, it was found that the content of Mg in the powder according to the invention is in the normal range.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2019 000 905.3 | Feb 2019 | DE | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2020/052826 | 2/5/2020 | WO | 00 |
