US3089227A - Beryllium sheet having third dimensional ductility - Google Patents

Description

i May 14, 1963 Y MW com F M YANS 3,089,227 BERYLLIUM SHEET HAVING THIRD'DIMEINSIONAL DUCTILI'I'Y Original Filed July 28. 1959 Pu m MEDOE M39 M24 .r N k ...v

on v

. kowmzo Tomm- Tw PwmAD mmwmwm INVENTOR.

FRANCIS MYANS F. M. YANS 3,089,227

BERYLLIUM SHEET HAVING THIRD DIMENSIONAL DUCTILITY May 14, 1963 4 Sheets-Sheet 2 Onginal Filed July 28, 1959 R N o m u %vn m W M S Ewxw Ema: mo. Gwxw Qmjom $96 92 50:55 M & $521 58 U231 29% .20: mo. MERE H 301 wziu 22% 5 6: F

O In

BY W muzwm mm 023 May 14, 1963 F. M. YANS 3,089,227

BERYLLIUM SHEET HAVING THIRD DIMENSIONAL nucmm Onginal Filed July 28, 1959 4 Sheets-Sheet 3 24 I l I l I 0 As UPSET A UPSET AND ANNEALED A AT 750C FORIHOUR.

3.2 g I z a 5 o C F' 2.8 g 1 Q 0 6 3 (Z) 3 2.4 2 Lu D w J 5 52 a0 7 i g 52 g BI-AXIAL L m z w l f, 1.5 a D 2 (D [:J E 2 2 S. c: (5 5 '5 L2 q O o E! E 0.. g

x o 7 UPSET AND ANNEALED 4 v AT 75oc FOR IHOUR l I I l v l o 2:| 4:: 62! 8:| I03 12:! 14:1

' FENAL SHEET AREA REDUCTION IN P ETTING U s INITIAL BILLET AREA (93%;. J

INVENTOR.

FRANCIS M. YAN

ties.

Unite States Patent C) 3,089,227 BERYLLIUM SHEET HAVING THIRD DIMEN- SIONAL DUCTILITY Francis M. Yans, Brooklinc, Mass, assignor, by mesne assignments, to the United States of America as represented by the United States Atomic Energy CommISSIOII Continuation of application Ser. No. 830,173, July 28,

. 1959. This application Dec. 27, 1962, Scr. No. 247,774

12 Claims. (Cl. 29-182) This invention relates to a beryllium sheet of improved ductility and a method of producing same. More particularly, this invention relates to beryllium sheet having third dimensional ductility. Beryllium sheet having third dimensional ductility refers to the ability. of the sheet to undergo relatively large changes in deformation along the plane of the sheet as well as in a direction perpendicular to the plane thereof.

This application is a continuation of my copending application Serial No. 830,173, filed July 28, 1959, for Beryllium Sheet Having Third Dimensional Duetility, now abandoned.

Beryllium sheet having two dimensional ductility refers to the ability of the sheet to undergo relatively large changes in deformation along a direction parallel to the plane of the sheet.

Beryllium metal possesses some extremely attractive properties for use as a structural material. For example, its strength-to-weight ratio is approximately twice that of aluminum and three times that of stainless steel. Equally significant is its combination of high moduli, high strength and low density (0.0658 lb. per cubic inch). This combination of properties would, but for its extreme brittleness, make beryllium sheet highly desirable for use in high speed aircraft, missiles and numerous other applications. For such applications, the main attractions of beryllium are even more favorable when compared to other materials now in use for the same purpose. For example, the density of beryllium is approximately that of magnesium. Its modulus of elasticity is approximately 8 times that of magnesium, 3 times that of titanium, 4 /2 times that of aluminum and 1 /2 times that of steel. Furthermore, its relatively high strength and melting temperature permit design for service at temperatures up to about 1100" F.

The nuclear properties of the metal beryllium are also unique. It is the only metallic moderating material in existence, and it has a low neutron absorption cross section. These characteristics, coupled with the metals excellent corrosion properties in air up to temperatures of about 1400 F. make it a very attractive cladding material for gas-cooled nuclear reactors.

Interest in beryllium has been further accentuated by its attractive thermal properties. Its high thermal conductivity (comparable to aluminum) tends to reduc; transient thermal gradients. Its coetficient of thermal expansion (approximately one half that of aluminum) tends to minimize excessive dimensional changes and related thermal stresses.

Until now, the principal factors which have prevented the use of this unusual metal in structural applications which require it to be in sheet form is its lack of third dimensional ductility and its anisotropic cracking proper- The addition of numerous alloying elements to beryllium has failed to increase its ductility or ease of fabrication. Third dimensional ductility is an absolute necessity for any application where beryllium undergoes strain in two directions simultaneously. If the sheet is strained in two directions simultaneously in the plane of the sheet it must, in order to avoid failure, accommodate some strain in a direction perpendicular to the plane of the sheet. In almost every conceivable practical application of beryllium sheet, the material is subjected to some form of biaxial or complex strain. For example, in fastening beryllium sheet, riveting is employed. Since the riveted sheet is subjected to biaxial strains in the plane of the sheet, it must, in order to avoid failure,

accommodate some strain perpendicular to the plane of the sheet. Hence, some degree of third dimensional ductility is necessary whenever the sheet is subjected to biaxial strains.

Beryllium sheet having ductilities of the order of 30% elongation in a direction parallel to the plane of the sheet (two dimensional ductility) has been produced by hot extruding flats from cold-compacted beryllium powder, followed by rolling perpendicular to the extrusion direction at elevated temperatures. However, this sheet is characterized by an almost complete lack of ductility in the third dimension (perpendicular to the plane of the sheet). Moreover. such sheet exhibits severe crack propagation in an extremely anisotropic fashion.

Because of its high sensitivity to crack propagation and inability to withstand complex biaxial strains, it has heretofore been extremely difficult to fabricate a ductile beryllium sheet with a practical material yield. Furthermore, beryllium sheet could not be worked at or near room temperature into forms having any appreciably small radius of curvature without experiencing a high degree of anisotropic cracking.

It is, accordingly, a major object of this invention to provide a beryllium sheet having a useful measure of ductility in a direction perpendicular to the plane of the sheet as well as in a direction parallel to the plane of the sheet.

Another object of this invention is to provide a process for producing a beryllium sheet having a useful measure of third dimensional ductility.

Still another object of this invention is to provide ductile beryllium sheet having a reduced tendency for crack propagation in an anisotropic fashion.

A further object of this invention is to provide a beryllium sheet which can accommodate plastic deformation at or near room temperature.

Other objects and advantages within the scope of this invention will be appreciated from the following description and drawings. In the drawings:

FIGURE 1 is a stereographic projection of the (0002) basal plane pole population for beryllium sheet fabricated by extruding a powdered compact of beryllium at 1850- l950 F. to effect a reduction ratio of 12:1 and then rolling (at about 1850" F.) the extruded flat perpendicular to the extrusion direction to effect a further reduction in area or 321.

FIGURE 2 is a stereographic projection of the (0002) basal plane pole population for beryllium sheet fabricated by upsetting a powdered beryllium compact at about 1850 F. to effect a reduction in area of 6: 1.

FIGURE 3 is a stereo graphic projection of the (IOIO) prism plane pole population of the extruded and crossrolled sheet.

Patented May 14, 1963 cation methods such as casting.

e,ose,227

FIGURE 4 is a stereographic projection of the (1010) prism plane pole population of the upset sheet.

FIGURE 5 is a graph showing the variation in two dimensional and three dimensional ductility with increasing reduction in area by upsetting, at a temperature in the range 1850 F. to 1950 F., a powdered beryllium compact clad in a thin deformable container.

FIGURE 6 is a scmilogarithmie graph showing the variation in the logarithm of maximum bend deflection of beryllium sheets having various ratios of width to thickness.

The reason for the lack in third dimensional ductility of beryllium sheet has not been entirely understood. Some workers have shown that certain of the transition elements, especially iron, in solid solution have a detrimental efi'ect on ductility. Others have pointed out that, because of the extreme anisotrophy of beryllium single crystals, the texture (crystallographic orientation) of polycrystalline material is an important factor in determining its mechanical properties.

In one form, the present involves a method wherein the crystallographic orientation of a polycrystalline beryllium article is controlled to produce a sheet of beryllium which can be strained in a direction in the plane of the sheet and, simultaneously, in a direction perpendicular to the plane thereof. In another form, my invention involves a method wherein the crystallographic orientation of a polycrystalline beryllium material is controlled to form a sheet of beryllium having improved ductility in a plane perpendicular to the plane of the sheet. In a more specific form, my invention involves the delineation of the necessary fabrication parameters which lead to a beryllium sheet of third dimensional ductility and wherein any cracks developed in said sheet propagate in an isotropic manner.

In accordance with my invention, and in order to obtain the proper crystalline orientation for beryllium sheet to have a useful measure of third dimensional ductility and to reduce the extent of anisotropic cracking thereof, a powdered compact of beryllium is subjected to a force of pure compression. One way in which a force of pure compression or equal radial flow in every direction could be applied is by rolling a beryllium billet from an' infinite number of directions. However, this is not easily done from a practical point of view. Ihave found that a practical equivalent of pure compression is to perform an upsetting operation on a beryllium billet by applying a uniform force along the billet axis, i.e., by starting with a cylinder and press-forging it into a pancaked shaped disc. This operation can be performed in a vertical press or in a horizontal extrusion press if the billet is suitably supported in the extrusion container. The billet is essentially a cylindrical compact of powdered beryllium clad in a thin deformable material such as a mild alloy-type steel. Commercially pure beryllium powder designated as QMV powder, prepared by the Brush Beryllium Company, can be used to prepare the billet. This powder has an average particle size of approximately -25 microns and has the following typical composition (percent by weight):

Assay, 99.3; BeO, .88; Fe, .078; A1, .060; Mn, .011; Mi, .013; Mg, .020; Cr, .0013; B, .0006.

From the standpoint of fabrication, as well as final mechanical properties, beryllium prepared by powder metallurgy has a number of advantages over other fabri' Powder metallurgy beryllium has a much smaller grain size in comparison with the grain size produced by other techniques. This line grain size results in improved strength and ductilities. an oxide film which acts a lubricant during deformation.

To prepare a beryllium compact for the upsetting opcration, a thin-walled cylinder of mild steel, which is closed at one end, is filled with the beryllium powder.

Also, beryllium powder particles are covered with i The powder is then cold compacted to a density ranging from 50% to of theoretical density. A mild steel cover is then sealed to the open end of the cylinder. Prior to upsetting, the entire billet is heated to a temperature below the melting point of beryllium.

As previously noted, a force of pure compression is necessary in order to achieve the desired grain orientation. To achieve the effect of pure compression, or equal radial flow in every direction, the billet should start and finish with a substantially circular cross section. This can be done by eliminating, as completely as possible, any friction between the billet and the platens of the press.

Another factor that affects the manner in which the billet deforms is its initial height-to-diameter ratio. This ratio should be less than 5 in order to effect symmetrical radial extension of the billet during upsetting. Any surface irregularities on the surface of the billet should be avoided as they cause creases to develop on the sheet during the upsetting process. I have found that the smoothness of the formed sheet is improved if the top and bottom steel cover plates are sealed to the end of the cylindrical steel clad at a point along an intermediate circumference thereof.

The upsetting may be performed on the heated billet at any temperature from 500-2l00 F. In general, the higher the upsetting temperature the lower the rupture strength and the higher the ductility of the resulting material. Upsetting at the lower temperatures requires high pressures to effect a desired reduction in area. In order to achieve easy flow during fabrication and to impart the best combination of ductility and strength to the formed sheet, an upsetting temperature in the range l5002100 F. is preferred.

A lubricant should be used between the billet and the platens of the press providing the upsetting force. Glass fiber or powder can be used for this purpose. On billets greater than about 8 inches in diameter, a thin asbestos mat impregnated with graphite has been found satisfactory. The billet is then upset at a temperature in the range 1500-2100" F. to effect a reduction in area (final area of sheet) (initial area of billet) ranging from 2:1 to 25:1. The steel cladding can then be removed by chemical and/or mechanical means.

The mechanical and crystallographic properties of a beryllium sheet, formed as described above, which had been upset to effect a reduction in area of 6:1 were compared with (1) a beryllium sheet formed by hot pressing to theoretical density and (2) a beryllium sheet formed by extruding a beryllium billet and then rolling the extruded billet in a direction perpendicular to the extrusion direction. The extruded and rolled sheet is known to be highly ductile in a direction parallel to the plane of the sheet.

The comparison specimens were prepared from QMV beryllium powder. Two thin-walled steel cans were filled with the powder and were hot pressed at a temperature in the range 1850-1900 F. until the powder reached theoretical density. The steel cladding was removed from one of the resultant billets. The second billet was extruded into a flat and then hot-rolled into a sheet. To obtain extruded material, the second billet was heated to a temperature in the range l200-2100 F. and inserted into an extrusion container heated to 900 F. The heated 'billet was then extruded through a rectangular die at about l900 F. to effect a reduction ratio of about 12:1. The extrusion was performed at a ram speed of about 2 in./sec. A colloidal suspension of graphite in oil was used to lubricate the extrusion. The extruded piece was then rolled perpendicular to the extrusion direction t temperature of about 1950 F. to effect a reduction ratio of 3:1.

Each of the three sheet specimens were annealed under vacuum for one hour at 750 C. cooled to room temperature and then tested for their mechanical properties. The same tensile specimen size, shape and testing technique was employed in each case. The specimens were approximately 3 inches long, approximately 0.060 inch thick with a gage length of 1 inch and a gage width of 0.300 inch. The uniaxial mechanical properties were tested by clamping one end of the test specimen to a jig and pulling the other end with a Tinius-Olsen tensile machine at a strain rate of approximately 0.004 inch/ minute. Baldwin Lima and Hamilton SR-4 type A-7 strain gauges were used to measure strain.

A summary of some typical results is given in Table I below.

Table I Hot Extruded Pressed 1.950" F.- Upset (1.850- 12:1,Cr05s- (1.850- 1,900 F.) Rolled 1,900" F.) 60,000p.s.i. 1,9-30" F. 7:1

Modulus of elas1]ticitg0(r3%s.i.).&..1.:. 41?.)(10 42x10 42x10 Yeld stren t 5 0 se (p.s.i.)--. 30,000 34,000 34,000

.e in tensile stren th W532i): 45,000 65,000 62,000 Tensile elongation (percent) 1.? 20 10 Reduction in area (percent) 1. a 11 It will be seen that the uni-axial tensile elongation (and hence the two dimensional ductility) of the hot pressed sheet was but a fraction of the extruded and rolled sheet; and the elongation of the upset specimen was comparable to the extruded and cross-rolled specimen.

The change in percent tensile elongation (uni-axial or two dimensional ductility), and ductility in a direction perpendicular to the plans of the upset sheet (third dimensional ductility) as a function of the reduction in area, is shown in FIGURE 5.

A bend test was used to evaluate the degree of third dimensional ductility in the beryllium sheet specimens. An indication of degree of third dimensional ductility is the bend deflection at fracture in specimens having a high width-toqhickness ratio.

The beryllium sheets were evaluated in a bend test apparatus. In the bend test, the deflection was recorded as a function of load. All test specimens were 0.088 inch-0002 inch thick 1: 3.0010010 inches long, and their widths were varied to obtain different width/thickness ratios.

Prior to testing, all bend specimens were surface ground to the same size, etched in 10% aqueous sulfuric acid to remove 0.007 inch on each side of the sheet. After etching, the specimens were annealed for one hour at 750 C. in a vacuum of 0.01 micron of mercury and then cooled. The grain size of all the specimens were 20-40 microns, and the shape of the grains was essentially equiaxed.

The bend test apparatus consisted of two rollers mounted for rotation about their axis on a stationary jig. Teflon hearings were used between the rollers and the jig. The rollers were 2.4 inches in diameter. The distance between the centers of,thc ,r ol lers was 3.00 inches, leaving a separation of 0.6"in'ch between the rollers. The test specimen was placed on the rollers. A round-ended ram 3 inches wide having a 0.200 radi-gs was placed over the test specimen and directly over the separation between the rollers. During testing, pressure was applied to the ram to thereby force the specimen between the rollers. The rollers rotated as the specimen moved between it, thus minimizing friction between the specimen and the rollers in contact therewith.

An indication of the degree of third dimensional duetility can be determined from the bend deflection at fracture in specimens having a width-to-thickness ratio of at least about 16:1. To determine the exact degree of third dimensional ductility (ductility in a direction perpendicular to the plane of the sheet), two bend specimens, one of upset material, and one of extruded and transverse-rolled material, each having a width-to-thickness ratio of at least 44:1, were bent with strain gauges attached to measure strain in the plane of the sheet. Two strain gauges were applied on the tension side of the bend specimen, one gauge measuring strain parallel to and the other gauge measuring strain perpendicular to the bend axis at the line of bending. In the extruded and cross-rolled sheet, i.e., perpendicular to the extrusion direction, .l% plastic elongation was detected perpendicular to the bend axis and 0% plastic elongation parallel to the bend axis. In the upset material, 0% plastic elongation was detected parallel to the bend axis and 1.9% plastic elongation perpendicular to the bend axis in the plane of the sheet. Hence, by conservation of volume, the extruded and cross-rolled sheet had 0.1% ductility perpendicular to the plane of the sheet. while the upset sheet had 1.9% ductility in a direction perpendicular to the plane of the sheet. It will be noted from Table I that the upset sheet still has a useful measure of plastic elongation in directions parallel to the plane of the sheet.

I have thus provided a beryllium sheet having an improvement in third dimensional ductility of 1800 percent over a beryllium sheet having high ductility in the plane of the sheet but with substantially no third dimensional ductility. Moreover, as is seen from Table I and FIG- URE 5, the increased measure of third dimensional ductility is achieved at a comparatively small sacrifice in two dimensional ductility (uni-axial elongation in the plane of the sheet). v

The average fracture strength, or modulus of rupture, in bending high width-to'thickness ratios 16:l) was determined for the hot pressed, upset, and extruded and cross-rolled sheets. The average fracture strength in bending the hot pressed sheet was 93,000 p.s.i. and for the extruded and cross-rolled sheet was 92,500 p.s.i. For the upset sheet the average fracture strength in bending was 142,000 p.s.i. Reference to FIGURE 6 indicates the improved ability of the upset sheet to undergo bending at or near room temperature. On the, ordinate is plotted the maximum bend deflection (on a logarithmic scale to the base 10) in inches, of sheets fabricated by the three methods previously described; on ttrile zbscissa is plotted the width/thickness ratio of the s cc It will be noted that the upset sheet shows a definite superiority over the hot pressed or the extruded and cross-rolled material under bending strain. The upset sheet could be bent three times as much before fractur- 1?1g,ton the average, as the extruded and cross-rolled s cc The superior properties of the upset sheet were also demonstrated by the fact that over 2,000 holes ranging from inch to /1 inch in diameter have been drilled in the upset sheet without any cracks or fractures. Up set sheets as thin as 0.040 inch could be turned on a lathe without cracking. The extruded and cross-rolled sheet could not undergo this treatment without undergoing severe anisotropic cracking, with cracking developing at about 30 and to the rolling direction. Wide berylhum sheet (w./t. 16:l) made in accordance with my invention can be warm formed, at. about 1000 F., into shaped articles having a radius of curvature equal to three times the thickness of the sheet without cracking. In cas s where the upset sheet did crack, it was noted that the cracks developed along the axis of bend only.

In order to correlate the mechanical properties and the crystallographic orientation of the fabricated sheets, an X-ray diffraction analysis was performed on the upset sheet, hot pressed sheet and extruded and cross-rolled sheet to determine the type and degree of preferred orientation. The Xray ditfraction analysis used was a modified Shulz technique. Details of this technique are described in an article by L. G. Shulz, Journal of Applied Physics, volume 20, pp. 1030, 1949, and in chapter IX, Structure of Metals, by C. S. Barrett, McGraw-Hill Book Co., Inc., second edition.

The.crystallographic orientation and population of the basal and prism planes of the sheets was plotted in stereographic projection in terms of multiples of a random in tensity, i.e., R=random intensity, nR=n times random intensity. For example, 2R means that the crystalline plane is more intensely oriented. by a factor of 2, as compared-to a random intensity. For purposes of comparison, the crystalline planes of the hot pressed sheet were assumed to be oriented randomly with respect to the plane of the sheet.

Pole figures of the upset sheet and extruded and crossrolled sheet were produced by reflection off the sides of a cube cut from the same sheets used for mechanical testing. All specimens were rotated 360 and X-ray scans (using a North American Philips Norelco unit) were taken in steps of about from the center of the pole figure to the outer edge thereof. Copper Kc: radiation was used at a maximum of 40 kilowatts and 20 milliamperes in combination with a 1 or 4 divergence slit. The crystallographic orientation of the (0002) basal and (1010) prism planes of the upset sheet and extruded and cross-rolled sheet are plotted in the stereographic projections of FIGURES 1-4.

A comparison of the orientation of basal and prism planes between the upset sheet and the extruded and rolled sheet is given in Table II.

8 sheet has 8 times as many basal planes parallel to the plane of the sheet as compared to the randomly oriented (hot pressed) sheet; the extruded and rolled sheet has 32 times as many basal planes parallel to the plane of the sheet as compared to the-random samples. Beryllium sheet having a two dimensional ductility of at least 2% and a three dimensional ductility of at least 0.5% -may be attained by upsetting a powdered beryllium billet to the point where as little as 4 and as many as 20 times the number o basal planes are oriented parallel to the plane of the resultant sheet as compared to the randomly oriented sheet.

The number of basal planes parallel to the plane of the sheet can be controlled by varying the reduction ratio (final sheet area) (initial billet area) As the reduction ratio increases from 4:1 to 15:1, the third dimensional ductility of the sheet formed by upsetting will increase to a maximum at about 6:1 to 8:1 and then decrease slightly to about .75% at a reduction ratio in the range of 15:1-30:1. At a reduction ratio of 6:1 to 8:1 the upset sheet has an elongation of about 2% in a direction perpendicular to the plane of the sheet (third dimensional ductility) and about 10% elongation in a direction parallel to the plane of the sheet (two dimensional ductility). Thus the redutcion in the number of basal planes parallel to the plane of the sheet imparts a measure of third dimensional ductilityto the upset sheet. A series of extruded and cross-rolled sheets which L d; fiRzRgndfim (with reference to hot pressed Extruded an specimen) Upset Sheet Average Orientation of Planes.

Mainly at 20 to the plane of the sheet and lie in a plane parallel to the extrusion direction, ranging Mainly at 20 to the plane of the sheet. ranging trom 0 to (0002) Basal Planes from 050 .53 at 50 .512 at 50.

Average Pole Population... 32R at 20 SE at 20.

8R at 0 0R at 0. Average Orientation of Mainly at 85 to the plane of the Mainly at to 90 to the plane Planes. sheet and in discrete directions, of the sheet.

60 and 90, with respect to the extrusion direction. (a) For prism planes 60 to the ex- (1010) Prism Planes trusligntdggcionlt f h t .5 a opaneo s so 0 Average Pole Population 8}? at to plane ot sheet ifg a a g igfi g gfif' g g gg g figg to the 2.511 at 90 to plane of sheet.

.5R at 40 to plane of sheet 8R at 85 to plane of sheet their basal planes are parallel to the sheet.

The number of basal planes (basal plane population) of the upset sheet parallel to the plane of the sheet is not as great as in the extruded and cross-rolled sheet. This should mean that the two dimensional ductility of the upset sheet is not as great as the extruded and crossrolled sheet. Table l shows this to be so. The relatively less intense orientation of basal planes in the plane of the sheet apparently enables it to undergo a greater degree of plastic deformation perpendicular to the plane of the sheet, but it is still enough to achieve a useful measure of ductility in directions parallel to the plane of the sheet.

Referring to Table II, it will be noted that the upset had been reduced in area by a ratio ranging up to as much as 30:1 exhibited negligible elongation, i.e., less than 0.1%, perpendicular to the plane of the sheet.

Theprism plane orientation of the upset sheet and of the extruded and cross-rolled sheet is shown in FIGURES 3 and 4 and is compared in Table II. In both cases, the prism planes are oriented substantially perpendicular to the plane of the sheet. There are, however, two important differences. In the upset sheet, the prism plane population perpendicular to the plane of the sheet is not as great as in the extruded and cross-rolled sheet. Upsetting apparently results in about a 75% reduction of prism plane population perpendicular to the plane of the sheet as compared to the sheet produced by extrusion and followed by cross rolling.

The second ditterenee to be noted is that the prism planes of the extruded and cross-rolled sheet are oriented discretely parallel at 60 o the olling direction as well as being perpendicular to the plane of the sheet, while the prism planes of the upset sheet are parallel to the plane of the sheet and are otherwise randomly oriented.

The discrete (1010) prism plane alignment apparently facilitates grain to grain crack propagation on the (1120) planes. In the upset sheet, the prism planes are random in comparison with the extruded and transverse rolled sheet. The effect of prism plane orientation is graphically illustrated by the fact that the upset sheet does not undergo severe anisotropic craclt propagation along the (lOTO) planes, as in the case of the extruded and crossrolled material.

I have described the method of my invention for obtaining a ductile and crack-free beryllium sheet in terms of an upsetting operation. Other methods may also be used to effect pure compression and thus impart isotropic mechanical properties to beryllium sheet. For example, compression rolling may be used to obtain and/or retain the advantageous properties of the upset material. By compression rolling I mean rolling from at least three separate directions perpendicular to the major axis of the billet '(i.e., in the plane of the sheet) .at a temperature to 2100 F. to effect substantially the same reduction in area per rolling pass. Compression rolling can be conveniently used in combination with an upsetting operation in cases where a further reduction in area (1%-60%) -is desired. A reduction in area greater than about 12:1

is frequently difficult to achieve by an upsetting operation alone. Thus a hot or cold powdered compact of beryllium may first be upset to achieve a reduction in area of about 10:1 and then subjected to compression rolling to complete the desired reduction in area. duced in this manner has been found to provide beiyllium sheet having as much as 1.9% ductility in a direction perpendicular to the plane of the sheet and at least 3% ductility in a direction parallel to the plane thereof.

It will therefore be understood that any method of applying a force of substantially pure compression to achieve deformation or any method involving plastic deformation which the basal (and prism planes) of a crystalline beryllium mass may be controlled to impart a useful measure of third dimensional ductility, i.e., of greater than 0.1%, to a sheet of beryllium is within the scope of my invention.

Since many embodiments might be made of the present invention and since many changes might be made in the embodiment described, it is to be understood that the foregoing description is to be interpreted as illustrative only and not in a limiting sense.

I claim:

1. A method of producing a beryllium sheet having three dimensional ductility which comprises confining beryllium powder Within a thin deformable container, pressing said container to form a billet in which the pow der is compacted to at least about 50% theoretical density, heating said billet to a temperature below the melting point of beryllium, upsetting said billet at a temperature in the range 500-2l00 F. and at a pressure sufficient to effect a reduction in area of said billet in the range 3:1 to 30:1 and thereafter removing the thus formed sheet from said container.

2. The method according to claim 1 wherein the heightto-diameter ratio of said billet, prior to upsetting, is less than about 5:1.

3. The method according to claim 1 wherein the beryllium billet is reduced in area by a ratio ranging from 6:1 to 8:l.

4. A method of producing a beryllium sheet having three dimensional ductility which comprises confining beryllium powder within a thin deformable container, pressing said container to form a billet in which the beryllium is compacted to at least about 50% theoretical density, heating said billet to a temperature below the melting point of beryllium, mounting said heatedbillet in a die, providing a lubricant between said billet and the die walls upsetting said billet at a temperature in the range 500-2l00 F. and at a pressure sufficient to effect Sheet proa reduction in area of said billet in the range 3:1 to 30:1 and thereafter removing the thus formed sheet from said container.

5. The method according to claim 4 wherein the heated billet is upset at a temperature in the range 1500-2100 6. The method according to claim 4 wherein the heated billet is reduced in area by a factor ranging from 6:1 to 8:1 at a temperature in the range l500 to 2100 F.

7. A method of producing beryllium sheet having a third dimensional ductility which comprises heating a beryllium billet to a temperature below its melting point. said billet having a height-to-diameter ratio'of no greater than about 5 :1, and comprising a powdered compact of beryllium clad in a thin deformable container upsetting said billet at a temperature in the range 500-2100 F. to effect a reduction in area in the range 3:1 to 30:1,

cooling said billet, and thereafter removing the thusformed beryllium sheet from said container.

8. The method according to claim 7 wherein the beryllium billet is upset at a temperature in the range 1500- 2100" F. to effect a reduction in area in the range 6:1 to 8:1.

9. A method of producing beryllium sheet having three dimensional ductility which comprises heating a beryllium billet to a temperature below the melting point of beryllium, said billet having a height-to-diameter ratio no greater than about 5, said billet consisting of a powdered compact of beryllium clad in a thin deformable container, compression rolling said heated billet from at least three different directions perpendicular to the major axis of said billet at a temperature in the range 500- 2100 F. to effect substantially the same reduction in area per rolling pass until a desired reduction in area is achieved. and thereafter removing said container from the thus formed sheet.

10. A method of producing a beryllium sheet having third dimensional ductility which comprises heating a beryllium billet to a temperature below the melting point of beryllium, said billet consisting of a powdered compact of beryllium clad with a thin deformable container and having a height-to-diameter ratio no greater than about 5:1, upsetting said heated billet by applying force along the major axis thereof at a temperature in the range 500-2100 F. to reduce the area of said billet by a portion of said factor, then compression rolling the upset billet from at least three different directions perpendicular to the major axis'thereof to reduce the area by a substantially equal amount per rolling pass until the total required reduction in area has been effected and thereafter removing the clad from the formed sheet.

11. A method of producing a beryllium sheet having three dimensional ductility which comprises confining beryllium powder within a thin deformable container, pressing said container to form a billet in which the powder is compacted to at least about 50% theoretical density, heating said billet to a temperature below the melting point of beryllium, applying a pure compressive force to said billet at a temperature in the range 500- 2100" F. and at a pressure sufiicient to effect a reduction in area of said billet in the range 3:1 to 30:1, and thereafter removing the thus formed sheet from said container.

12. Beryllium sheet characterized by: t

(a) third dimensional ductility in the range greater than 0.1% up to 2%; (b) two dimensional ductility of at least 3%; (c) and an average modulus of ruptureof 142,000

p.s.1., said beryllium sheet being the product produced by the steps of:

(1) confining beryllium powder within a thin deformable container;

(2) pressing said container to form a billet in which 3,039,227 11 12 the powder is compressed to approximately 50% of References Cited in the file of this patent theoretical defiwy, UNITED STATES PATENTS (3) heating the billet to a temperature below the melt- 1 5 ing point of beryllium 2,885,287 Larson May 9 9 (4) applying a purely compressive force to said billet 5 at a temperature in the range 500-2100 F. and at a OTHER REFERENCIYZS' pressure sufficient to effect a reduction in area of said First Geneva confercrlct? 9" Atomic Energy, 3, billet in the range of 3:1 to 30:1, to form a sheet, 1956, PP- PY 111 Llbfafy, TK 9006 151(6- Mechanical Properties of Beryllium Fabricated by and (5) thereafter removing the thus formed sheet from 10 powdPr Metallurgy by {Raver and Wilde May 1954 said contain; (reprint), 15 pages; copy in 75/200(A).

Claims (1)

1. A METHOD OF PRODUCING A BERYLLIUM SHEET HAVING THREE DIMENSIONAL DUCTILITY WHICH COMPRISES CONFINING BERYLLIUM POWER WITHIN A DEFORMABLE CONTAINER, PRESSING SAID CONTAINER TO FORM A BILLET IN WHICH THE POWDER IS COMPOSED TO AT LEAST ABOUT 50% THEORETICAL DENSITY, HEATING SAID BILLET TO A TEMPERATURE BELOW THE MELTING POINT OF BERYLLIUM, UPSETTING SAID BILLET AT A TEMPERATURE IN THE RANGE 500-2100*F. AND AT A PRESSURE SUFFICIENT TO EFFECT A REDUCTION IN AREA OF SAID BILLET IN THE RANGE 3:1 TO 30:1 AND THEREAFTER REMOVING THE THUS FORMED SHEET FROM SAID CONTAINER.

Images