JP7637172B2 - Plasma equipment for the production of high quality spherical powders at high capacity - Google Patents

Description

CROSS-REFERENCE TO RELATED APPLICATIONS
This application claims priority to U.S. Provisional Application No. 62/171,618, filed Jun. 5, 2015, currently pending, which is incorporated herein by reference.

The subject matter of this disclosure relates primarily to the production of high purity spherical powders for use in additive manufacturing (3D printing) and hot isostatic pressing (HIP).

There is an increasing demand for fine, high-quality spherical metal powders. For example, one of the main applications of such powders is the 3D printing industry, which requires fine size cuts of high-quality, spherical, satellite-free Ti-6Al-4V (titanium alloy), typically 45-45 microns or 45-106 um. Cuts of 106-250 um cut size are used for Hot Isostatic Pressing (HIP).

This application is raising the bar in quality as it consumes the highest quality powders currently available on the market. Many criteria are used to evaluate powder quality, i.e. powder sphericity, powder particle size distribution, absence of satellites (significantly smaller particles attached to the main particle). One of the current problems is that the production capacity of such fine powders is very limited. Another problem is that the industry is looking for very fine and specific cuts while typical atomization systems produce a wide range of particle sizes.

Many methods have been developed over the years to produce powders by micronization.

For example, U.S. Patent No. 5,707,419 discloses a method for melting and atomizing titanium wire using a plasma torch. In this disclosure, the feed rate of titanium, for example, is limited to 0.75 kg/h with a wire diameter limited to 3.2 mm (1/8 inch), and the plasma torch is fixed in a position at an angle of 30° to the feed axis. This 30° angle has been determined to be the optimum angle under certain circumstances. The torch is then locked at this particular angle to ensure alignment with the wire. This method has the advantage of repeatability between runs and also minimizes the chance of missing the wire, but applicant's experience has demonstrated that this configuration is not optimal.

In PCT Patent Publication No. WO2011/054113, a method is proposed to improve the productivity of plasma atomization using electrodes for preheating. Using electrodes for preheating is a very complicated process. In this arrangement, there are typically (3) electrodes as well as three (3) plasma torches to ensure uniform heating. The wire is heated by arcing the wire from each electrode. Thus, three currents are passed through the wire, heating it by resistive heating. Note that this means that six power supplies are required for operation, but the more power sources present, the more difficult it becomes to manage the heat directed to the wire, plus the capital and operating costs also increase significantly.

This arrangement also presents some mechanical inconveniences. For example, to perform plasma atomization, alignment of the torch with the wire is very important. All electrodes and torches, as well as the wire, must converge at the same exact spot. Since the space around the apex convergence point is very limited, the design of the assembly is dictated by mechanical constraints, not by the process itself.

It would therefore be desirable to have a simplified device that would allow for increased productivity of plasma atomization. It would also be beneficial to have a system that allows for control of particle size distribution.

U.S. Pat. No. 5,707,419 International Publication No. 2011/054113

It would therefore be highly desirable to provide a novel apparatus for producing fine powders.

It would also be highly desirable to provide new methods for producing fine powders.

Embodiments described herein provide, in one aspect, an apparatus for producing powder from a wire by plasma atomization, the apparatus comprising:
- a device for preheating the wire,
at least one plasma torch for heating and atomizing the wire;
a chamber for cooling the atomized particles to a solid state;
Equipped with.

In another aspect, embodiments described herein provide an apparatus for producing powder from a wire by plasma atomization, the apparatus comprising:
a plasma torch for atomizing the wire, the plasma torch being adapted to pivot and rotate in the reactor;
a chamber for cooling the atomized particles to a solid state;
Equipped with.

Further, in another aspect, embodiments described herein provide an apparatus for producing powder from a wire by plasma atomization, the apparatus comprising:
a plasma torch for atomizing the wire, the angle of the plasma torch with the wire being adapted to be adjusted in order to adjust the particle size distribution of the powder;
a chamber for cooling the atomized particles to a solid state;
Equipped with.

Further, embodiments described herein provide, in another aspect, a method for producing powder from a wire by plasma atomization, the method comprising:
- preheating the wire;
- heating and atomizing the wire;
- cooling the atomized particles to a solid state;
Includes.

Further, embodiments described herein provide, in another aspect, a method for producing powder from a wire by plasma atomization, the method comprising:
Providing a plasma torch, the plasma torch being adapted to pivot and rotate in the reactor;
- adjusting the position of the plasma torch in the reactor;
- using a plasma torch to atomize the wire;
- cooling the atomized particles to a solid state;
Includes.

Further, embodiments described herein provide, in another aspect, a method for producing powder from a wire by plasma atomization, the method comprising:
- providing a plasma torch for atomizing the wire;
- adjusting the angle of the plasma torch relative to the wire to adjust the particle size distribution of the powder;
- using a plasma torch to atomize the wire;
- cooling the atomized particles to a solid state;
Includes.

Additionally, embodiments described herein provide, in another aspect, a powder produced by any of the above methods.

Additionally, embodiments described herein provide, in another aspect, a powder produced by any of the above devices.

For a better understanding of the embodiments described herein and to show more clearly how they may be carried into effect, reference is made to the accompanying drawings, by way of example only, which illustrate at least one illustrative embodiment.

1 is a general schematic diagram of a plasma device according to an exemplary embodiment. FIG. 13 is an enlarged detailed cross-sectional view of an induction coil and a wire guide according to an exemplary embodiment. FIG. 13 is an enlarged detailed cross-sectional view of a torch angle adjustment mechanism using a ball flange according to an exemplary embodiment. 1 is a schematic diagram of the effect of induction frequency on the temperature profile in the wire. A comparison is shown between a conventional torch and a torch according to an exemplary embodiment in terms of torch angle and closeness of the tip to the wire; in FIG. 5a, the conventional torch provides a fixed angle of 30°, which results in a longer distance, while in FIG. 5b, the current torch shows a steeper angle, which is closer to the nozzle, thereby maximizing heat and momentum transfer between the plasma and the wire. A comparison is shown between a conventional torch and a torch according to an exemplary embodiment in terms of torch angle and closeness of the tip to the wire; in FIG. 5a, the conventional torch provides a fixed angle of 30°, which results in a longer distance, while in FIG. 5b, the current torch shows a steeper angle, which is closer to the nozzle, thereby maximizing heat and momentum transfer between the plasma and the wire.

Presented herein is an apparatus P and method that can adjust the torch angle and preheat the wire to produce high quality powder, control particle size, and maximize production rates in a plasma atomization reactor. It is demonstrated herein that varying the angle and proximity of the nozzle tip to the wire can significantly affect capacity as well as particle size distribution.

As shown in Figure 1, wire 2 provided on a metal wire spool 1 is unstrained from the metal wire spool 1 and then fed through a wire feeder and straightener 3. The straightened wire 2 is fed through a pass flange 4. The wire 2 then enters a wire guide 5 surrounded by an induction coil 6 before being atomized at its apex 8 (the apex being the confluence of the wire 2 and the three torches 7) by three plasma torches 7. The powder thus produced passes through an aperture plate 9 and cools as it falls into a reactor 10.

Once preheated, the wire 2 then reaches the apex 8, which is the area where the wire 2 and the three plasma torches 7 meet for atomization. The molten atomized particles solidify back into a solid state as they fall into the chamber of the reactor 10. The powder 11 is then pneumatically conveyed to a cyclone 12, which separates the powder from its gas phase. The powder is collected at the bottom of a canister 14, while the clean gas is then sent to a finer filtration system (not shown) via an exhaust 15. The canister 14 can be isolated from the cyclone 12 by an airtight isolation valve 13.

Turning now to the induction coil 6, the apparatus P uses an induction coil to preheat the wire 2, which uses a single power source and the heat source does not interfere with the apex region. In this configuration, the preheating of the wire occurs from a single uniform and compact source. The wire temperature can be controlled by adjusting the induction power, which is a function of the current in the induction coil 6.

The induction preheating device is shown in Figure 2. The pass flange 4 is made of a non-conductive material to ensure that the entire reactor is insulated from the coil. The pass flange 4 has two airtight holes with compression tools 21 used to pass the conductors 22 of the induction coil 6 through the reactor 10.

The wire guide 5 can be designed to be either induction-responsive or induction-transparent. For example, the wire guide 5 can be made of alumina or silicon nitride, which are induction-transparent. Alternatively, the wire guide 5 can be made of silicon carbide or graphite, which are induction-responsive. In the latter case, the induction-heated hot wire guide radiates heat back to the wire 2.

For example, when the wire used is 1/4 inch diameter Ti-6Al-4V Grade 23 ELI, the optimum induction frequency for this wire has been found to be 170-290 kHz. The optimum frequency will vary depending on the material used for the wire guide, as well as the shape and dimensions and properties of the material.

Figure 3 shows an adjustable torch angle mechanism including a swivel ball flange 30. Three plasma torches 7 are used to atomize the wire 2. These three torches 7 are now attached to the body of the reactor head using a swivel ball flange. The ball flanges 30 each include two flanges that fit together, a bottom flange 31 and a top flange 32, and can swivel with each other. The bottom flange 31, which is connected to the reactor head, is fixed, but the top flange 32 can rotate up to an angle of 4° in all axes. Assuming the reactor head is designed to have a nominal angle of 30°, this means that the torches 7 can range anywhere from 26° to 34°.

By varying the torch angle relative to the wire 2, the location of the apex 8 (the junction of the wire 2 and the three torches 7) also moves, which affects efficiency since the length of the torches 7 is constant. To avoid such problems, a longer torch coupled to a spacer can be used. Applicant has found that the closer the nozzle tip is to the wire, the higher the capacity and the smaller the average grain size. By having a longer torch and multiple sizes of spacers, it is possible to achieve any angle while keeping the location of the apex 8 in the same place.

It seems that pivoting the torch 7 has a significant effect on the plasma atomization process. In previous systems it was stated that the optimum angle would be fixed at 30°. This statement has been questioned, but being able to pivot the torch was not a clear alternative. Hence, choosing a fixed angle was justified in the case of the old system. The current arrangement proposes to add flexibility to the system by adding pivoting of the ball flange 30 to the design.

By varying the angle of attack between the wire 2 and the plasma jet, atomization can be affected in several ways. The main difference between plasma atomization with respect to conventional gas atomization is that heat is supplied from the jet. Thus, there are two main considerations: heat transfer from the torch 7 to the wire 2, and atomization alone.

An important aspect of plasma atomization is the quality of heat transfer between the torch 7 and the wire 2. Indeed, proper matching is required. The angle of attack also affects heat transfer in two different ways, a steeper angle increases the surface area (A) to exchange heat, whereas a shallower angle promotes a higher exchange coefficient (h). A closer nozzle also increases the temperature of the plasma plume, as does the wire and its velocity, which results in a higher temperature difference (ΔT) and a higher exchange coefficient (h), respectively.
where Q is the heat transferred,
h is the heat transfer coefficient,
A is the surface area used for exchange,
ΔT is the temperature difference between the wire and the torch.

The above equation is the classical heat transfer equation. The objective is to maximize the value of Q. The angle affects both h and A. From a heat transfer perspective, the optimal angle is the angle that maximizes Q for a particular wire feed speed, size, and material.

For atomization to occur, some microdroplets must form on the surface of the wire 2 (heating phase). A gas flow is then used to separate the droplets from the wire 2 and put them in suspension in the gas phase (atomization phase). It is known that high velocities are required to break the bond between the wire and the droplets. The following equation, taken from the literature on water atomization of aluminum [5], relates the average particle size to the angle between the melt stream and the argon jet:
where V is the plasma jet velocity,
C is a global constant that regroups the different parameters and conditions;
is the angle between the melt stream and the plasma jet.

Although the application is slightly different, the concept is similar, the plasma jet replaces the water jet in the formula, and the jet is used for heating instead of cooling in the formula. In fact, this formula makes sense since faster gas velocities can separate finer particles, which requires more force to separate finer droplets from the molten wire. According to this equation, the angle has a similar effect, which makes sense considering that mechanical energy transfer is similar to heat transfer and that the angle can have the same effect on the mechanical energy transfer coefficient as it does on the heat transfer coefficient (h) mentioned above.

To optimize the plasma atomization process, it becomes more clear that the angle must be variable to accommodate different conditions. The two old equations show how the parameters are linked and that the ability to vary the angle is a key feature of the current device.

Thus, the apparatus P includes, among other things, (1) induction preheating of the wire to improve capacity, and (2) a torch mounted to the reactor head using a swivel ball flange 30 that allows flexibility regarding the angle attainable by the torch alignment to allow control of particle size distribution (powder quality).

It should be noted that conventional systems allowed production rates of about 0.75 kg/h with an average particle size of about 80 um. In fact, the optimal scenario would be to significantly increase capacity while simultaneously decreasing the average particle size distribution. However, using known techniques, it was generally accepted that increasing capacity would unfortunately result in a coarser particle size distribution.

In contrast to older methods, the current method allows for the simultaneous optimization of both parameters by increasing the transferred mechanical and thermal energy of the process.

Plasma atomization uses a supersonic plasma jet, which has a core temperature of about 10,000 K, to melt and atomize metal originating from a wire source in a single-step process. Although very powerful, the process is energetically inefficient; only a small portion of the heat from the plasma is transferred to the wire to melt it (about 0.4% for conventional systems), and only a small amount of kinetic energy originating from the nozzle is used to break up the droplets into finer ones (less than 0.001% for conventional systems).

In general, higher capacity can be achieved by increasing the heat transferred, and finer particles can be achieved by increasing the mechanical energy transferred (from the kinetic energy of the plume to break up the bubbles). For example, moving the average particle size distribution from 80um to 45um can require about a 78% increase in kinetic energy (momentum). These can be achieved either by increasing the feed energy input, improving the quality of transfer, or a combination of both. On the other hand, improving the energy transfer is rarely the case, as increasing the feed energy increases the operating costs of the technology.

Below is a comparison of the two techniques.

It is important to note that the data in Table 1 is an example of what can be achieved, but the actual improvements are not limited to these values. For example, with some adjustments, the capacity can be increased to about 10 kg/h and the average particle size can be even smaller.

All of the following suggested improvements work synergistically to simultaneously improve both capacity and particle size distribution, but may also be used alone.
Increased wire diameter size - Older methods limited wire size to 1/8 inch. Current methods suggest that larger wires may increase capacity for two reasons: a) increased residence time of the wire in the plasma, and b) also increased exchange surface area. Both reasons result in increased heat exchange between the plasma plume and the wire.

Induction Wire Preheating - In older systems, the wire was either not preheated before being fed into the plasma atomization area or was preheated using complex resistive heating devices. Wire preheating improves the overall heat balance of the system, as more efficient heating techniques can be used to do the hard work. Induction heating has many advantages over other heating techniques, primarily due to the ability to remotely heat objects with greater efficiency (approximately 10-40%). Wire preheating allows the wire to be brought up to melting temperature (or even above locally) and can be operated using a liquid feed. Superheating a partially liquefied wire has a positive effect on particle size distribution, as it can significantly reduce the viscosity and surface tension of the molten metal. Thus, most of the heat coming from the plasma can be used to melt and atomize or simply atomize, without wasting the mostly latent heat. Wire preheating can have a significant effect on production capacity and particle size distribution.

It is important to note that induction preheating can be used for both metallic and non-metallic materials, but using different techniques. Conductive materials such as metals and alloys can be used with wire guides made from materials that are either transparent or non-transparent to induction. On the other hand, wire stock made from non-conductive materials such as ceramics will not couple with induction and must therefore be preheated using an insert that is conductive and sensitive to induction heating in the frequency range used. In the latter case, the wire is preheated indirectly only from the wire insert, via a combination of conduction, radiation, and convection.

If the feed material can be bonded to the wire, the induction frequency can be used to fine-tune the temperature profile in the wire, which has a large effect on the shape of the wire at the apex position and indirectly affects the grain size distribution (Figure 5). A lower frequency results in a more uniform temperature, while a higher frequency results in a temperature gradient in the wire. The induction frequency is easily varied by changing the coil length; the longer the coil, the lower the frequency.

Rotating Ball Flange - In older systems, there was an inaccurate paradigm that the ideal angle was 30°, so the flange that gripped the torch was fixed at that value. However, applicant has determined that this is not the case here. The concept of rotating ball flange offers a great deal of flexibility in terms of the quality of heat and momentum transfer. The distance also varies with the angle (Figure 5a vs. Figure 5b). In fact, the closer the tip is to the wire, the higher the temperature of the plume will be and the faster the jet will be as seen by the wire. Also, the closer the nozzle tip is to the wire and in a perpendicular plane with the wire, the higher the coefficient of heat and momentum exchange will be.

Increasing the inlet argon pressure - Increasing the argon inlet pressure will result in higher gas velocity and mass flow. Increasing the pressure has a cubic effect on the kinetic energy, as follows:
where m is the mass flow of the gas, v is the velocity of the gas, and Ek is the resulting kinetic energy.

Technical data:

It should be noted that the feedstock may be made from a wide range of materials and is not limited to metals only. Examples of materials that may be atomized herein include, but are not limited to, titanium (both commercially pure titanium and titanium-based alloys), aluminum, stainless steel, Hastelloy™, Inconel, and tungsten. Examples of non-metallic materials may include ceramics, but only if they can be made in wire or rod form.

While the above description provides examples of embodiments, it will be understood that certain features and/or functions of the above-described embodiments may be modified without departing from the spirit and principles of operation of the above-described embodiments. Accordingly, it will be understood by those skilled in the art that the above description is intended to be illustrative and non-limiting of the embodiments, and that other variations and modifications may be made without departing from the scope of the embodiments as defined in the claims appended hereto.
References [1] P. G. Tsantrizos, F. Allaire and M. Entezarian,''Method of production of metal and
ceramic powders by plasma atomization''
．． Patent US5707419 A, 13 January 1998.
[2]C. F. Yolton and J. H. Moll, ``Method for Atomizing Titanium''. Patent US4544404, 12 March 1985.
[3] M. Drouet, ``Methods and Apparatuses for Preparing Spheroidal Powders''. Patent application WO2011054113, 12 April 2011.
[4]C. F. Yolton,''Induction melting titanium in a vacuum or nonoxidizing atmosphere
and treatment of an electric current''. Patent US5084091A, 28 January 1992.
[5]B. Verlinden and L. Froyen, ``Aluminum Powder Metallurgy,'' Training in Aluminum Application Technologies, vol. Advanced level 1, 1994.

Claims (15)

1. An apparatus for producing powder from a wire by plasma atomization, comprising:
- a device for pre-heating said wire,
at least one plasma torch for heating and atomizing said wire;
a chamber for cooling the atomized particles to a solid state ,
the device for preheating the wire comprises an induction coil;
a wire guide is provided for coupling with the induction coil for transferring heat to the wire;
An apparatus in which the wire is electrically conductive and the wire is also heated by direct coupling at the same time that it is heated by the wire guide . The apparatus of claim 1 , wherein a wire guide is provided , the wire guide configured to allow heat from the induction coil to be transferred directly to the wire. 3. The apparatus of claim 1 or 2 , further comprising a system for adjusting at least one of the angle and distance of the torch relative to the wire. An apparatus according to any one of claims 1 to 3 , wherein a feeding device is provided which is adapted to accommodate larger wire diameters. The apparatus of claim 4 , wherein the feeding device is adapted to feed and straighten the larger wire diameter. The apparatus of claim 4 or 5 , wherein the wire diameter is between 1/8 inch and 6 inches, inclusive. 1. An apparatus for producing powder from a wire by plasma atomization , comprising:
- a device for pre-heating said wire,
a plasma torch for atomizing said wire, said plasma torch being adapted to pivot and rotate in the reactor;
a chamber for cooling the atomized particles to a solid state ,
the device for preheating the wire comprises an induction coil;
a wire guide is provided for coupling with the induction coil for transferring heat to the wire;
An apparatus in which the wire is electrically conductive and the wire is also heated by direct coupling at the same time that it is heated by the wire guide . 1. An apparatus for producing metal powder from a wire by plasma atomization , comprising:
- a device for pre-heating said wire,
a plasma torch for atomizing the wire, the angle of the plasma torch with the wire being adapted to be adjusted in order to adjust the particle size distribution of the powder;
a chamber for cooling the atomized particles to a solid state ,
the device for preheating the wire comprises an induction coil;
a wire guide is provided for coupling with the induction coil for transferring heat to the wire;
An apparatus in which the wire is electrically conductive and the wire is also heated by direct coupling at the same time that it is heated by the wire guide . 1. A method for producing metal powder from a wire by plasma atomization, comprising:
- supplying a wire ;
- preheating the wire;
- heating and atomizing said wire;
- cooling the atomized particles to a solid state ,
the device for preheating the wire comprises an induction coil;
a wire guide is provided for coupling with the induction coil for transferring heat to the wire;
A method wherein the wire is electrically conductive and is also heated by direct coupling at the same time that it is heated by the wire guide . The method of claim 9 , wherein the wire is preheated by direct or indirect induction heating. The method of claim 10 , wherein a wire guide is provided for reacting with induction to transfer heat to the wire. The method of claim 10 , wherein a wire guide is provided that is induction transparent, whereby heat from an induction coil providing induction heating is transferred directly to the wire. The method of any one of claims 9 to 12 , wherein the wire has a diameter of 1/8 inch or greater. An apparatus according to any one of claims 1 to 8, wherein a wire guide is provided for coupling with the induction coil for transferring heat to the wire by conduction, radiation and convection. A method according to any one of claims 9 to 13, wherein a wire guide is provided for coupling with the induction coil for transferring heat to the wire by conduction, radiation and convection.