Traitement en cours

Veuillez attendre...



Aller à Demande


Note: Texte fondé sur des processus automatiques de reconnaissance optique de caractères. Seule la version PDF a une valeur juridique

[ EN ]



This i nvεntion rel ates to thermomechani cal processing of di spersi on- strengthened precious metal alloys. The present invention can provide alloys
containing platinum, palladium, rhodium and gold which are useful in the production of glass fibers.
Cnε of the most exacting applications of platinum is in the production of glass fibers. Molten glass often at temperatures ranging from 1200 to 1600ºC passes through a series of orifices in a bushing. Advances in glass fiber production are demanding both larger bushings and higher operating temperatures.
Structural components such as these at elevated temperatures under constant loads experience continuous dimensional changes or creep during their lives. This creep behavior depends upon the interaction between the external conditions (load, temperature) and the
microstructure of the component. In recent times,
increased resistance to creep of material systems has been accomplished by using a dispersion of very small, hard particles (called dispersoids) to strengthen the
microstructure of the component. These systems have become to be known as d ispersi on- strengthened metals and alloys and the dispersoids used are usually oxides. A recent development in dispersion-strengthening is mechanical alloying which uses a high energy ball mill to achieve the intimate mechanical mixing typical of the process. An attritor mill or vibratory mill also can be used.
The present invention provides for the
thermomechani cal processing of dispersion-strengthened precious metal alloys.
The invention is comprised of a series of
mechanical deformation and annealing cycles to help develop a creep resistant microstructure. Specifically, I achieve this by rolling and annealing a powder compact of
dispersion-strengthened precious metal. The material may be cross-rolled as well as longitudinally rolled or just longitudinally rolled.

FIGURE 1 is a schematic drawing of the rolling operation.
According to the process of this invention, the proceoure used to thermomechanically process the compact was to roll the compact for a 10 percent reduction in area then anneal the rolled specimen. The reduction in area is carried out under a pressure that elongates the rolled specimen without substantially widening it. Generally, the annealing is carried out for a period of time and at a temperature sufficient to develop a specimen with a minimum creep rate. Preferably the annealing is carried out for five minutes at at 1,900°F (1,040°C) before further
rolling. The total extent of deformation ranges from 50 to 20 percent reduction in area and generally is approximately an 85 percent reduction in area. This roll/anneal cycle was selected to help develop a creep resistance
microstructure. The roll/anneal cycles are continued until the 35 percent reduction in area is accomplished.
There are several high-energy ball mills
comn.erci al ly available either using a stirrer or vibration to induce mechanical alloying. Stainless steel bearings or grinding media and the powder charge go into the
cylindrical container of the mill . The high-energy impacts are affected by the rotating impeller. In the internal arrangement of the attritor mill, impact events occur in the dynamic interstices of the media created by the
impeller during stirring.
Dispersion-strengthened precious mεtals are known in the art and are commercially available. One such material is that available from Johnson, Matthey & Co.
Limited, under their designation ZGS. The above indicated ZGS material consists essentially of platinum in which the disperoid is zirconia; the latter is present in an amount of about 0.5% by volume.
The dispersion-strengthened precious mftals of this invention generally comprise a precious metal, or precious metal alloy, preferably platinum, as the
dispersing medium, or matrix, and a dispersoid of a metal oxide, metal carbide, metal silicide, metal nitride, metal sulfide or a metal boride which dispersoid is present in effective dispersion-strengthening amounts. Usually such amounts will be between about 0.1 percent to about 5.0 percent by volume. Preferably the dispersoid will be an oxide. Exemplary of metal compounds which may be employed as the dispersoid are compounds of metals of Group IIA, IIIA, 111 E (including non-hazardous metals of the Actinidε and Lanthanide classes) , I VB , VB , VI B and VI IB. More specifically exemplary of suitable metals are the
following: Be, Kg, Ca, Ba, Y, La, Ti , Zr, Hf M o W, Ce Na , Ga , and Th as Well as Al .
Several mechanical alloying experiments were performed using the attritor mill to generate a composite powder for consolidation. Wash heats intended to coat a thin layer of platinum on the internal working surfaces of "the attritor mill were carried out. This "conditioning" treatment was intended to prevent iron contamination of subsequent milling experiments, but several washes were required before the iron contamination was reduced to what was believed to be an acceptable level.
The samples then ere consolidated by vacuum hot pressing (VHP) at elevated temperatures and pressures. In the alternative, the samples can be consolidated by first cold pressing at elevated pressures followed by sintering at elevated temperatures. VHP generally is carried out at a temperature ranging from 1300 to 1700°C under a pressure ranging from 500 to 10,000 psi for a time ranging from 10 to 30 minutes. Preferably, the temperature ranges from

140C to 1500°C under a pressure of 3,000 to 6,000 psi for a time of 15 to 25 minutes. Generally, the cold pressing is carried out at a pressure ranging from 2,000 to 10,000 psi for up to 5 minutes followed by sintering at a temperature ranging from 120G to 1700°C for 2 to 6 hours.
Approximately one kgm of -325 mesh (-44 micron) platinum sponge from Englehard was blended with an amount of yttria (Y203) to give nominally 0.65 volume percent (0.15 weight percent) oxide loading in the final compact. The yttria was nominally 200-600 angstrom in size. The platinum matrix starting powder for the experiment
consisted of v ery fine, near spherical particles or chained aggregates. Most of the particles below 2 microns appeared to be single crystals. The starting powder had a fairly high specific surface area.
The pov-der mixture was charged into the
container of the attritor mill while it was running. The grinding media had been previously loaded to give a volume ratio of media to powder of 20:1. The grinding media used was a hardened 400 series stainless steel bearing nominally 3/8 inch (0.953 cm) diameter. The impeller rotational speec was selected at 130 rpm .
Samples of powder were removed at various times to obtain information on the changes in particle morphology and specific surface area with milling time. The first sample was taken after one hour of milling and indicated that flake generation was in progress.
After milling for three hours, another powder sample was taken for metal 1 ographi c characterization.
While more flakes were generated, the extent of plastic deformation seemed to have increased. Flake cold welding appeared to have taken place as well . The composite flake appeared to have three or four component flakes cold welded together. No edge cracking appeared in the composite flake suggesting that work hardening saturation had not been reached at this point.
After milling for 23 hours, the composite flakes appeared to thicken. This clearly demonstrates the cold welding aspect of the milling action. Along with cold v.elding, the flake diameter appeared to increase.
The experiment was continued for 71 hours then terminated, and the powder was removed for further
processing .
There appeared to be a fairly high initial surface area generation rate. The iron contamination in the milled powder was greatly reduced compared to the previous experiments and reflects the coating action that appeared to minimize wear debris generation during milling.

The maximum iron contamination level in the powder was approximately 300 wppm. The milled powder was consolidated by vacuum hot pressing and thermomechani cal ly processing into sheet for creep testing, the details are to follow.
Example I produced a powder of relatively low iron contamination. Since this experiment resulted in small powder lots (nominally 60 gms) taken at various times during the milling experiment, each sample was individually consolidated by vacuum hot pressing (VHP) εt 1,450°C under 5, 000 psi (34.5 MN/m2) for twenty minutes. The resultant compacts were nominally 1 inch (2.54 cm) in diameter.
Relative density of specimens a re listed.

The thermomechanical processing (TMP) used on the compact consisted of several roll/anneal cycles. The basic operation involved rolling a sheet specimen and cropping pieces after various rolling passes for microstructural characterization. The procedure used was to roll the compact for a 10 percent reduction in area then anneal the rolled specimen for five minutes at nominally 1,040°C before further rolling.
Specimen D was the most responsive to the TMP cycles. After the 10th rolling pass, the grain structure was fairly elongated. The lack of oxide clusters during optical metal lographic examination suggested that the milling action had worked the yttria into the platinum matrix. A metal 1 ographic analysis of the same region showed the development of a moderate grain aspect ratio

(grain length to thickness ratio in the viewing plane). A the number of roll/anneal cycles increased, the grain aspect ratio (GAR) increased. At this stage a moderate GA also had been developed in a transverse direction. The significance of this observation is that the grains took o the shape of a pancake structure thin in a direction perpendicular to the sheet yet extended in the other two directions. Since a GAR seems to extend in two directions in the rolled sheet and the state of stress in a bushing tiP Plόte is biaxial, this transverse GAR development may be vtry beneficial for good creep resistance in bushing applications.
After the 16th rolling pass, the elongation of the grains had increased significantly. A higher
magnification view of the same region revealed the degree of grain elongation and fineness of the grain size. The transverse GAR had also significantly increased. These elongated grain morphologies a r e desirable microstr uctures for gooc creep resistance.
Creep Testing
All the creep testing was dona in air using constant load machines, the elongation was measured by an LVDT connected to a multi-point recorder and a precision cigital voltmeter. Specimen temperature was monitored with a calibrated Pt/Pt-Rh thermocouple attached so that the bead was adjacent to the gage section of the creep
specimen. The creep specimen was a flat plate type with a gage length of approximately 2.25 inch (5.72 cm). The tensile stress was applied parallel to the rolling
direction (longitudinal direction). The general procedure was to hang the specimen in the furnace to reach thermal equilibrium then start the rig timer upon application of the load. Periodic temperature and extension measurements were made either until the specimen failed or the test was terminated (specimen removal or furnace burn-out) .
Creep results were obtained from specimens that were processed according to Example II except that these specimens w e r e milled 10 hours and received the above thermomechanical processing treatment of 10% reduction in area per pass with an intermediate anneal at nominally 1040ºC for 5 minutes. The extent of deformation was nominally an 85% reduction in area. The first specimen had a varied creep history that started by applying a tensile stress of 1.000 psi (6.89 Mn/m2) at 2,400ºF (1,316°C). The resultant secondary creep rate was too low to adequately measure; therefore, the temperature was increased to
2,600ºF (1,427 ºC) and a secondary creep rate of 4.5x10-6 hr-1 was observed. After approximately 118 hours the stress was increased to 1,400 psi (9.65 Mn/m2) and a new secondary creep rate of nominally 3x10-5 hr-1 was recorded.

These creep rates are two o rd e rs of magnitude less than that for the previously indicated ZGS under the same testing conditions. The ZGS material will have a stress rupture life of at least 48 hours when tested at 1400 ºC and 1000 psi in the rolling direction of the sheet.
The general microstructure of the crept specimen indicated that the grains were highly elongated in the rolling direction (creep stress direction also) and the grain boundries were ragged. There appeared to be evidence of subgrains in the structure as well. The microstructure observed in this specimen was typical of that of a good creep resistant material as evidenced by the exceptionally good creep properties.