US3851027A

US3851027A - Shock-bonding of hard particles

Info

Publication number: US3851027A
Application number: US00320060A
Authority: US
Inventors: A Balchan; G Cowan
Original assignee: EI Du Pont de Nemours and Co
Current assignee: ETI EXPLOSIVES TECHNOLOGIES INTE
Priority date: 1973-01-02
Filing date: 1973-01-02
Publication date: 1974-11-26
Anticipated expiration: 1991-11-26

Abstract

1. IN A PROCESS FOR SHOCKNG HARD PARTICLES SO AS TO BOND THEM TOGETHER BY EXPLOSIVELY PROPELLING A PROJECTILE TOWARD A BODY OF SOLID MATERIAL CONTAINING THE PARTICLES IN A MANNER SUCH THAT THE PROJECTILE COLLIDES WITH SAID BODY OF SOLID MATERIAL IN AN AXIALLY PROGRESSIVE MANNER AT CONFORMING COLLISION SURFACES, THEREBY INTRODUCING A SHOCK WAVE INTO THE BODY OF SOLID MATERIAL, THE IMPROVEMENT WHICH COMPRISES (1) FORMING SAID BODY OF SOLID MATERIAL FROM (1) A CARRIER MATRI HAVING DISPERSED THERETHROUGH INTERSTICES IN WHICH THE HARD PARTICLES ARE AMASSED INTO AGGREGATES AND (2) A MATRIX-CONTAINMENT MEANS, SAID MATRIX HAVING A POST-SHOCK DEFORMABILITY WHICH IS GREATER THAN THAT OF SAID PARTICLES, AND HAVING A LOWER DEGREE OF POROSITY THAN SAID AGGREGATES, WHEREBY THE MOTION OF SAID AGGREGATES IS RETARDED IN THE DIRECTION OF SHOCK TRAVEL, THE PEAK PRESSURE OF THE SHOCK WAVE INDUCED IN THE HARD-PARTICLE AGGREGATES BEING AT LEAST ABOUT 100 KILOBARS; AND (B) AFTER THE COLLISION, SEPARATING THE BONDED BODIES PRESENT IN THE INTERSTICES FROM THE MATRIX.

D R A W I N G

Description

NOV. 26, A. 5. BALCHAN ET AL SHOCK-BONDING OF HARD PARTICLES Filed Jan. 2, 1973 United States Patent O 3,851,027 SHOCK-BONDING OF HARD PARTECLES Anthony S. Balchan, Hagerstown, Md., and George R.

Cowan, Woodbury, NJL, assignors to E. I. do llont de Netnours and Company, Wilmington, Del.

Filed Jan. 2, 1973, Ser. No. 320,060 Int. Cl. C(llb 31/06 US. Cl. 26484 8 Claims ABSTRACT OF THE DISCLOSURE Hard particles, e.g., diamond particles, are bonded together to form dense polycrystalline bodies by amassing the particles into aggregates within interstices dispersed throughout a carrier matrix, preferably formed from adjacent apertured segments, e.g., metal disks, the aggregateladen matrix, together with containment means therefor, forming a body of solid material which is caused to collide progressively in an axial direction with an explosively propelled projectile along conforming surfaces so as to introduce into the body a substantially uniform shock wave spanning the bodys lateral boundaries; and, after the collision, separating the resulting bonded bodies formed in the interstices from the matrix. Bonded bodies 0.5 mm. and larger can be formed, e.g., stone-size polycrystalline diamonds. A circularly cylindrical body of solid material and coaxial surrounding hollow circular projectile cylinder are preferred to achieve optimum recovery of bonded bodies.

BACKGROUND OF THE INVENTION This invention relates to an improved method of bonding hard particles, e.g., diamond particles, together by means of shock waves.

The compaction or bonding of a mass of powder to form a dense body by introducing a shock wave into the powder mass has been known for a number of years, the major effort in this technique having been expended on the compaction of relatively soft, i.e., readily deformable, materials by the detonation of a layer of explosive adjacent to a container for the powder. The bonding together of hard particles, e.g., diamond, cubic boron nitride, silicon carbide, etc., by shock waves, however, has posed a special problem owing to the fact that it has not been possible to induce in such particles, by the above shocking system, the degree of plastic flow needed to achieve strong interparticle bonding.

US. Pat. 3,399,254 describes the bonding of diamond particles directly to each other by subjecting a mass of diamond powder to a shock wave at a pressure of at least about 300 kilobars. In a technique shown in this patent, diamond powder is confined in a container having the form of a flat slab or disk, and the container in turn is confined in a massive shock-resistant metal block. A shock pressure pulse is applied to the flat container surface and the powder by head-on impact of the surface with an explosively driven metal projectile plate. This shock technique is capable of developing the required shock pressure in the powder mass but is subject to certain limitations deriving from the geometry of the shockproducing assembly and the mass of powder. For example, the necessity of confining the powder in a flat container gives rise to recovery problems owing to the fact that, when the shock pressure is high, a fiat container is difiicult to maintain intact and consequently is snceptible to rupture, with loss of powder. Also, the flat container requirement generally precludes the use of the head-on impact technique with powders that may become molten as a result of the passage of the shock wave, inasmuch as the probability of loss of such materials via the relatively large-area closures needed in flat containers is high.

3,851,027 Patented Nov. 26, 1974 ice A greatly preferred technique for subjecting powder to shock pressures which are sufficiently high to cause hard particles to bond together is the progressive-collision shock technique described in US. Pat. 3,667,911, issued June 6, 1972, to A. S. Balchan and G. R Cowan, the dis closure of which is incorporated herein by reference. In this technique, an explosively propelled projectile collides progressively with powder amassed in the form of a body of solid material, e.g., a solid or hollow cylinder, plate, slab, disk, bar, or strip, the collision progressing in the direction of the bodys axis, e.g., the axis of a cylinder or the length or width of a plate, and causing a substantially uniform shock wave to span a major portion of the solid material between the bodys lateral boundaries, the spanning or transverse shock wave traveling through the body in the axial direction at a velocity equal to the velocity of the collision. The use of a cylindrical assembly is especially advantageous from the standpoint of product recoverability. For example, powder can be amassed in the form of a coherent, end-capped circular cylinder which is positioned coaxially within a hollow circular metal projectile cylinder with a spacing between the facing conforming surfaces of the cylinders, and a shock pressure pulse applied to the powder by explosively propelling the hollow cylinder in a manner such as to cause progressive collision of the cylinders at their facing surfaces. The shocked powder is held in a cylindrical container which forms from the propelled hollow cylinder and the end caps as the shock process proceeds. A double-walled container is formed when the powder is initially supported in a cylindrical container. Thus, this procedure provides good product containment not only because a cylindrical container is less susceptible to rupture than a flat container,'but also because a doublewalled container for the shocked powder can be formed to give additional strength to the shocked assembly. Inasmuch as impact of the explosively propelled cylinder with the powder-supporting cylinder is capable of generating in a given powder, with a given explosive composition and mass, a higher total pressure than can be generated by detonation of a layer of explosive adjacent to the powder-supporting cylinder, the procedure of the aforementioned patent extends the pressure capability of previous cylindrical shocking processes at the same time that it affords the good containment described above.

The attainment of sufficiently high pressure and temperature to effect the bonding of a mass of particles shocked according to the progresive-collision technique described in the above-mentioned US. Pat. 3,667,911 is dependent on a suitable selection of a number of variables associated with the shock-producing-assembly design, the explosive, and the sample to be shocked. For bonding or compacting a mass of hard particles, the pressure required with a sample having a high initial density may be quite high. With this technique the pressure attainable in a given system is dependent on the detonation velocity of the explosive employed to propel the projectile. Obviously, attempts to achieve a higher pressure in the sample by increasing the detonation velocity are subject to the limitation that an explosive having the requisite detonation velocity for the desired pressure may not be available. Also, increasing the detonation velocity in effect increases the velocity of the transverse shock wave in the sample, and the particle velocity (velocity of the sample in the direction of the progressive collision), making sample retainment more diflicult. The pressure requirement for compaction may be reduced by selecting a lower initial sample density, since the higher temperature associated with the initial porosity may allow plastic flow to occur at a lower pressure. However, a high residual particle velocity may result owing to the volume change in the porous sample. Thus, initial porosity in the sample and high detonation velocity of the explosive both aggravate the sample retainment problem, thereby jeopardizing sample recovery.

SUMMARY This invention provides an improved process for bonding together hard particles, e.g., in a powder, by shocking, the process comprising (a) explosively propelling a projectile, e.g., a hollow circular metal projectile cylinder, toward a body of solid material, such as a circular cylinder which is positioned coaxially within a projectile cylinder, the body being formed from (1) a carrier matrix having dispersed therethrough interstices in which hard particles are amassed into aggregates and (2) a matrixcontainment means, the projectile being propelled in a manner such that it collides with the body of solid material in an axially projessive manner at conforming collision surfaces, thereby introducing a shock wave into the body, the matrix having a greater post-shock deformability than the hard particles amassed therein and having a pore content lower than that of the particle aggregates, whereby the motion of the aggregates is retarded in the direction of shock travel; and (b) after the collision, separating the bonded bodies present in the interstices from the matrix. When pressure intensification is desired, the product of the density, p and the relative shock compression, AV/V is greater for the carrier matrix than for the particle aggregates at a given shock velocity. Preferably, the matrix is formed from adjacent apertured segments, e.g., superimposed metal disks. The bonded bodies formed in the interstices in the matrix are separated from the matrix after the collision by mechanical means or by a selective physical or chemical transformation of the matrix, e.g., melting or dissolution.

BRIEF DESCRIPTION OF THE DRAWING The process of the invention will be described with reference to the attached drawing which is a schematic representation of the superimposed surfaces of three disks in a matrix/hard particles disk assembly described in Example 1 and constituting a carrier matrix having dispersed therethrough apertures or pockets laden with masses of hard particles.

DETAILED DESCRIPTION In the present process, particles to be shock-bonded together to form dense polycrystalline bodies or compacts are amassed in interstices which are dispersed throughout a matrix in a manner such as to form separated pockets of hard-particle aggregates in the matrix. The matrix, which is a holder or carrier for the hard particle aggregates, is a material which is weaker, i.e., more deformable or ductile after shock, than the particles. Amass'ing of the hard particles in dispersed, separated pockets in a weaker matrix serves to repress the formation of cracks in the compacts, obviously in itself an important advantage from the standpoint of compact strength. In addition to this, however, the matrix performs another function in the process. The problem of product recovery is alleviated by the use of a matrix which has a lower degree of porosity than the hard-particle aggregates in the pockets therein and which undergoes no large overall volume decrease such as occurs in an irreversible phase change. Because such a matrix undergoes a smaller volume change than the particle aggregates, it is able to restrain the aggregates, i.e., to retard their forward motion (motion in the direction of shock travel). Furthermore, any residual particle velocity of this aggregate-laden matrix when the pressure has become ambient is considerably lower than would be that of the aggregates alone. Thus, retainment of the particle aggregates in the shocking assembly is facilitated.

When, as may often be the case, an intensification of the pressure produced in the particle aggregates is desired, sucn intensification can be achieved by employing a matrix for which the product of its initial density and the relative shock compression, AV/V is greater than that for the particle aggregates at a given shock velocity. V is the initial specific volume of the material in question and AV is the change in specific volume from V to that in the compressed state. AV/V is obtained from the Hugoniot curve for a material.

The shock impedance of the matrix relative to that of the aggregates is another parameter to be considered in a matrix selection. Shock impedance is equal to the initial density of a material times the velocity of the shock wave passed through it, and thus varies with pressure. In the present progressive-collision shock process, the shock velocity equals the axial collision velocity, as described in the aforementioned U.S. Pat. 3,667,911. When pressure uniformity throughout all of the aggregate pockets is of prime importance, it may be desirable to employ a matrix having a shock impedance and density about the same as those of the particle aggregates. On the other hand, when conditions favoring high shock-induced tempera tures, i.e., low initial aggregate density and high collision velocity, are employed, it may be desirable to employ a matrix having a higher shock impedance than that of the particle aggregates. If the latter situation prevails, the aggregates will attain their peak pressure cumulatively due to the reflection of compressive shocks back into the aggregates from the aggregate/matrix interface. In such a situation, the peak pressure in the aggregates can approach the pressure in the matrix. This cumulative shocking does not produce the high peak temperatures associated with one-step shocking to the same pressure, and therefore is desirable when conditions favoring high shock-induced temperatures are employed. If the shock impedance of the matrix is equal to, or less than, that of the aggregates, the latter will attain their peak pressure in one step, the pressure being transiently as high as, or higher than, that in the matrix, but rapidly attenuated when the shock impedance of the matrix is less than that of the aggregates, as a result of the shock reflections which are rarefactions.

It also may be desirable that the melting point of the matrix be sufiiciently high that gross melting of the matrix does not occur during the shock process. This contributes to easier containment after shock when there is some residual particle velocity, owing to the friction between an essentially solid matrix and a surrounding container wall.

Many materials are capable of being used as the car rier matrix, i.e., are weaker than the hard particle aggregates to be bonded and can be provided in forms needed to give different selected densities, shock impedances, and relative stock compressions. Metals as well as non-metals can be employed, metals being preferred on the basis of ductility and high tensile strength. Metals such as iron, nickel, copper, columbium, titanium, molybdenum, and alloyed tungsten can be used, for example, when a higherimpedance matrix is desired, and metals such as aluminum when impedance matching is desired.

To obtain the optimum benefit from the matrix, e.g., repress crack formation, reduce residual particle velocity, and intensify pressure throughout as much as possible of the total mass of hard particles to be bonded, the aggregates of particles should be well-dispersed or -scattered throughout the matrix. The matrix can be provided in the form of a nonparticulate body such as a wrought product or casting, or a particulate body such as a pressed and/or sintered powder compact. The matrix with particles amassed in interstices therein can be provided, for example, by machining or punching holes of the desired pocket size and shape in the pre-formed matrix body and thereafter loading the holes with particles; or by loading the particles into individual containers of the desired pocket size and shape, maintaining the loaded containers in desired positions relative to one another, and thereafter forming the matrix around the containers,

e.g., by compacting matrix-forming powder around them or casting them in position in molten metal, i.e., a metal which melts at a temperature which is sufliciently low so as not to cause decomposition of the particles in the containers. If required to achieve a desired bulk density, the mass of particles can be pressed after loading into the holes of containers.

With a particulate matrix, the matrix with the hard particles amassed in apertures therein can be provided by mixing the hard particles with larger-size matrix particles. The larger-size particles, e.g., metal particles, form a matrix, and the smaller particles of material to be shockbonded take up location in the interstices or pockets formed by adjacent matrix particles. When the matrix particles are ductile and fully dense, the density of the mass of particles in the interstices can be increased by compacting the mixture, whereby the matrix particles are forced to deform into the pocket spaces, reducing the volume thereof and therefore the density of the particle masses therein. Welbcompacted particle mixtures can be self-supporting.

While the carrier matrix and aggregate pockets therein may be formed as a single unit, e.g., by the differentialsize particle mixture technique described above, better control of the density of the particle aggregates amassed in the matrix, and of the distribution of the pockets in the matrix is achieved by building the matrix up from a number of adjacent or superimposed separate units or segments, e.g., a number of circular disks, or bars or strips.

In the present process, an explosively propelled projectile collides progressively with a body of solid material formed from the aggregate-laden carrier matrix and containment means for the matrix. The containment means can provide partial or complete containment of the matrix. For example, if the matrix is a self-supporting singleunit cylinder or a suitably supported segmented hollow cylinder, e.g., comprised of a number of adjacent annular disks fastened to, strung on, or otherwise attached to support rods, only partial containment of the matrix may be employed. In such an instance, the ends of the cylindrical matrix can be capped with solid plug closures. The end-capped cylinder collides with a surrounding hollow projectile cylinder, and the matrix, after collision, is held within a container formed by the projectile cylinder and the capping blocks or plugs at the end of the matrix cylindex.

In addition to endor edge-capping containment means, however, a surrounding container such as a circular cylinder for the matrix can be employed and is preferred because it affords an extra measure of protection against loss of particle aggregates during the shock process (which may be especially desirable as higher shock pressures are employed). If the matrix is initially held within a surrounding container, the matrix laden with the shockbonded bodies after the collision is held within a doublewalled container formed by the container, capping blocks, and projectile.

The concentration of aggregate pockets in the matrix, or the volume ratio of the particle aggregates to the matrix, can vary over a fairly wide range and depends chiefly on the shock conditions employed and the initial packing density of the aggregates since these factors affect the residual particle velocity and therefore determine how difiicult the compacted aggregates are to retain in the shocking assembly. Generally, the restraining eifect of the matrix can be observed with a ratio of matrix to particle aggregates, by volume, as low as about 0.25:1, although a ratio of at least about 1:1 usually is preferred to provide better shock uniformity. There is no upper limit on the matrix content beyond that imposed by economic considerations with respect to the minimum amount of material that one can afford to compact in a given shock operation. With more valuable products, it may be acceptable to employ a ratio as high as about 20:1 if the intensity of the shock conditions Warrant it.

The size and shape of the aggregate-laden interstices in the matrix are selected so as to conform substantially to the size and shape of the bonded bodies or compacts desired. The interstices can be substantially equi-axed, e.g., spherical, cubic, or discoid; or they can have one dimension greater than another, e.g., they can be in the form of cylinders or parallelepipeds. With elongated interstices the long dimensions preferably are in planes substantially normal to the direction of shock travel, e.g., normal to the axis of a cylindrical matrix. In one embodiment of the process, the particles can be amassed within coiled tubes, and the coils superimposed on one another in the axial direction (direction of shock travel) to form a multi-unit matrix, the matrix being composed of the walls of the coils and, if desired, a surrounding material, e.g., pressed powder, in which the coils may be embedded. In this case, the interstices and the pockets of aggregates in the matrix are continuous coils or spirals.

The dimensions of the pockets (in the case of the coiled pockets, the inner diameter of the coiled tube) should be small relative to the dimension of the matrix normal to the direction of shock travel, e.g., the diameter of a cylindrical matrix, larger pockets generally requiring larger-size matrices. As a rule, the dimension of the matrix normal to the direction of shock travel should be at least about 3 times the maximum dimension of the pockets (or the inner diameter in the case of coiled tubes). If the pockets in a given matrix become overly large, there is an increased possibility that distortion or cracking of the bonded bodies during pressure release will occur. Inasmuch as the present process finds especially great utility in the production of stone-size polycrystalline bodies, i.e., bodies at least about 0.5 millimeter in minimum dimension, the aggregate pockets often will have a minimum dimension of at least about 0.5 millimeter. As is shown in the following examples, much larger interstices can be employed, and larger bodies formed, however. Aggregateladen interstices of any shape can. be used. Best control of the size and shape of the bonded bodies formed is obtained by the use of a non-particulate matrix, and this is preferred. However, with a particulate matrix the size of the matrix particles can be adjusted to provide the desired size of interstices and, therefore, of the particle aggregates therein and of the bonded bodies formed by the process.

The selection of such process variables as the particle size and packing density of the particle masses in the interstices, and the variables associated with the explosively propelled projectile system, depends on the particular powder composition employed, and specifically on the manner in which the composition responds to high pressure and temperature. All particles bonded in the present process are hard, in the sense that the particles have a high resistance to permanent deformation. They include, for example, diamond, cubic and wurtzitic boron nitride, and carbides such as silicon carbide and boron carbide. Mixtures of these materials also can be employed. As a rule, particle aggregates of this nature, for bonding, require shock pressures of at least about kilobars at low-to-moclerate packing densities, higher pressures being required with higher-density aggregates. The average or peak shock pressure which will be induced in the matrix and the particle masses depends on the velocity with which the projectile and the aggregate-laden body collide (i.e., the detonation velocity of the explosive), higher pressures being induced at higher collision velocities. The initial shock pressure in the aggregates and the temperature induced therein depend on the relative shock impedances and densities of the aggregates and matrix as described previously. The manner in which the shock pressure can be estimated for a given powder and explosive is described in the aforementioned US. Pat. 3,667,911.

At low packing densities, the plastic flow of the particles may be facilitated as a result of the larger volume change occurring in the aggregates during the shock process. Therefore, when the particles to be shock-bonded are amassed in the interstices in the matrix, e.g., by loading into apertures in a disk, with little or no pressing to increase the packing density of the particle aggregates, shock pressures at the lower end of the operable pressure range can be employed.

When diamond particles are to be bonded according to the present process, a peak pressure of at least about 300 kilobars should be employed. The peak pressure is the maximum pressure induced in the particle aggregates, and, as stated previously, is reached in a single step when the aggregate/matrix shock impedances and densities match, or cumulatively when these properties are higher for the matrix than for the aggregates. Much higher peak pressures can be employed to shock-bond diamond particles by the present process, e.g., pressures of 1000 kilobars or more, provided that such pressures are reached cumulatively in order to avoid the high shock-induced temperatures associated with single-step shocking to high pressures. In the cumulative shocking technique, the initial pressure induced in the diamond aggregates can be as low as about 200 kilobars.

In the shock bonding of diamond particles by the present process, it generally is desirable to employ a diamond packing, or bulk density in the range of about 50 to 75% of the crystal density, densities in this range being high enough to avoid unduly high shock-induced temperatures at commonly attainable collision velocities, yet insufficiently high as to require extraordinary pressing conditions for their attainment. Collision velocities of about from 4500 to 7000 meters per second are required to achieve peak pressures of at least about 300 kilobars in diamond particles amassed to a bulk density in the specified range, the collision velocity required to achieve a given peak pressure being higher as the density of the diamond aggregate matrix mixture is higher. To achieve a higher peak shock pressure in a given system, one can increase the bulk density of the aggregates or the collision velocity, or both. The peak shock pressure can be estimated from the Hugoniot relationship for the aggregate/matrix mixture, as described in U.S. Pat. 3,401,019.

When the hard particles are amassed so as to form a circular cylinder, the particle-laden cylindrical matrix is positioned coaxially Within a hollow circular metal cylinder with a spacing between the facing surfaces of the cylinders, and a layer of explosive de'tonates progressively along the outer surface of the outer cylinder in the general direction of the cylinder axis, causing progressive collision of the cylinders and shocking of the particle aggregates in the matrix interstices. In most instances, for practical reasons the facing surfaces of the cylinders are substantially parallel to one another, i.e., the spacing is substantially uniform throughout the length of the cylinders, and this is a preferred arrangement. However, if desired, one of the facing cylinder surfaces can be tapered from one end to the other so that the surfaces are positioned at an angle to each other, i.e., the spacing is nonuniform in an axial direction. In the parallel arrangement, the velocity with which the cylinders collide is equal to the detonation velocity of the explosive. If the detonation velocity of a given explosive is higher than the collison velocity needed to produce a desired pressure in the diamond aggregates, and a lower-velocity explosive is not available, the angle arrangement of cylinder surfaces can be used, inasmuch as the collison velocity in this case is lower than the detonation velocity of the explosive.

To recover the shock-bonded bodies formed by the present process, they are separated mechanically from the matrix in which they are contained, if that is possible, or the matrix is removed by a treatment which selectively converts the matrix into a readily separable form. A metal matrix, for example, can be dissolved in a medium which does not attack the shock-bonded bodies, or melted if the shock-bonded bodies are stable at the melting temperature of the matrix. Melting of the matrix and segregation of the shock-bonded bodies therefrom can occur during the shock process itself, requiring no subsequent melting step, although, as stated previously, gross melting of the matrix is not preferred. With bonded diamond products, an iron or copper matrix can be removed by dissolution in nitric acid, and a columbium matrix by dissolution in a warm nitric acid-hydrofluoric acid mixture.

The following examples illustrate various embodiments of the present invention.

Example 1 (a) A cylindrical mild steel matrix having apertures scattered therethrough laden with masses of diamond particles is constructed from twenty-one apertured steel disks and twenty blank steel disks stacked together coaxially, the apertured disks alternating with blank disks in the stack. The apertured disks, 0.125-inch-thick and 1.5 inches in diameter, have twenty 0.125-inch-diarneter circular holes bored through them in the thickness direction on axes normal to their 1.5-inch surfaces, the holes being arranged in five parallel rows: a central row having four holes with their centers in line with the center of the disks; two rows adjacent to the central row having five holes; and two rows farthest from center having three holes. The distance between the centers of any two adjacent holes in each row is 0.250 inch, and the centers of any pair of adjacent holes in a given row, together with the center of a hole which is adjacent to them in an adjacent row, form the points of an equilateral triangle having 0.25'0-inch sides. The center of the disk is located between the second and third holes in the central row, and spaced from the center of one of these holes a distance which is one-third of the distance between the centers of the two holes. The twenty holes are to have diamond powder pressed into them. All of the disks, apertured as well as blank, have three 0.125-inch-diameter mounting holes bored through them near the periphery, one on each of three radii which divide the disk into three 120 segments.

The twenty holes are filled with synthetic diamond powder, which is a blend of 33 percent by weight of nominal 4 micron powder and 67 percent by weight of nominal 40 micron powder. The diamond powder is a product of the process described in U.S. Pat. 3,608,014. The average amount of diamond powder loaded per disk is 1.133 grams (total weight of diamond=23.79 grams). The loading density of the diamond powder is 2.25 g./cc. (64.1% of theoretical). The diamond-laden disks are assembled as follows: A 0.125-inch-thick steel mounting disk 1.50 inches in diameter has three 5-inch-long, 0.125- inch-diameter parallel steel pins mounted to it on end, the pins being normal to the plane of the disk and spaced 120 apart on a 1.25-inch circle. The diamond-laden disks are stacked onto the mounting disk with the pins passing through the mounting holes in the disks. Starting with a stack of disks held with their holes all in alignment, each disk is rotated 120 (from the position of the previous disk, in the same direction) before it is placed into the pinned stack. A 0.125-inch-thick blank disk is mounted between each pair of loaded disks.

The drawing shows the relative positions of the holes in any three consecutive diamond-laden disks in the stack, the surfaces of the apertured disks being superimposed in the schematic representation. Line AA' passes through the centers of the central row of four holes in one of the three disks, the holes in this disk being the cross-hatched holes. The alignment of the next consecutive apertured disk is such that the centers of its central row of four holes lie on a line, AA, located by rotating the previous disk 120 in a counter-clockwise direction. The holes in this disk are dotted. The alignment of the next consecutive apertured disk is such that the central row of four holes lie on a line, AA, located by rotating the previous disk 120 in a counter-clockwise direction. The holes in this disk are unmarked. The holes designated 1, 2, and 3 occupy the same position when the holes in the three disks are all in alignment, but are 120 apart when the disks are rotated as described. 4a, 4b, and 4c designate the mounting holes in the three disks. Rotation of the disks in this manner affords a better dispersion of the powder pocket throughout the matrix.

(b) A cylindrical container for the stack disks is loaded as follows:

The container is a Type 3'04 stainless steel hollow cylinder 53.5 inches long, and having a 2-inch outer diameter and a 0.095-inch wall thickness. The 5.125-inch-long stacked diamond-laden steel disk assembly described in (a) is first positioned in a 1.5l-inch ID. x 1.75-inch O.D. stainless steel tube, which in turn is positioned at the center of the 5 3.5-inch-long cylinder between two cylindrical plugs of an 8-inch-long, 1.8-inch-diameter cast iron cylinder and a 14-inch-long iron disk section consisting of 19 superimposed 1.8-inch-diameter, 0.75-inch-thick disks of 100-mesh low-carbon steel powder pressed to tapering densities, gradually increasing from 50% (of the theoretical) in the outermost disk to 84% in the innermost disk, which abuts the cast iron cylinder. The density and shock impedance of the cast iron are approximately matched to the density and shock impedance of the diamond-laden steel disks, thereby transmitting the shock pressure with no substantial wave reflection; and the decreasing density of the iron disk assembly allows the momentum associated with the shock wave to be carried ofi and causes the container cylinder to contract to a smaller radius, thereby restricting the forward motion of the cast iron plug and diamond-containing sections. This type of plug closure is described in US. Pat. 3,568,248.

The cylindrical container is positioned coaxially inside a 54-inch-long Type 1015 steel projectile cylinder having a 4.5-inch outer diameter and a 0.375-inch wall thickness (spacing between cylinders is 0.875 inch). Surrounding the outer steel cylinder is a 60-inch-high cylindrical layer ofgrai ned 80/20 amatol (80% ammonium nitrate/20% trinitrotoluene) contained in a thin-walled cardboard cylinder having an inner diameter of 16 inches (explosive layer thickness is 5.75 inches). The loading density of the explosive is 1.22 g./cc., and the total explosive Weight is 510 pounds. The explosive extends across the top of the cylindrical assembly, and is initiated axially at the top end, five inches above a 2-inch-thick steel cover welded to the outer cylinder, by means of three superimposed l-inch-thick, 1.5-inch-diameter 95/5 RDX/wax pellets, initiated axially by means of a detonating cord. The detonation velocity of the explosive is 5960 meters per second.

Detonation of the detonating cord initiates the RDX booster and, in turn, the cylindrical explosive layer. The axially progressive collision of the steel cylinders caused by the detonation of the explosive subjects the diamond powder to an estimated peak pressure of about 700 kilobars.

After detonation of the explosive, the composite cylindrical assembly is recovered intact. To recover the diamond, the composite cylinder is cut open so as to expose the diamond-laden steel disk section, and the latter is removed from the cylinder by machining down the steel composite cylinder, leaving a thin wall around the diamond-laden steel disk section. Treatment of the latter with nitric acid selectively dissolves the steel, leaving inch-diameter, 0.09-inch-high diamonds having a density of 95% of the crystal density of diamond, as determined by measurement of pore volume by tetrachloroethylene absorption. Treatment of the diamonds with a mixture of CrO (84 grams), H 80 (350 milliliters), and H 0 (157 milliliters) for 48 hours at 90-100 C. oxidizes the small amount of graphite in the product (4.6%

Example 2 The procedure described in Example 1 is repeated with the exception that the cylindrical container is made of Type 1015 steel, the matrix disks are 1.80 inches in diameter and are made of columbium, the 1.51-inch ID. x 1.75-inch O.D. stainless steel tube is omitted, and the tapered density steel powder disk section above the central diamond-containing section is replaced by a solid carbon steel section (above denotes in the direction of the initiation end of the cylinder assembly). The cast iron cylinder below the same is replaced by a low-carbon steel cylinder. The center of each matrix disk is mid-way between the centers of the second and third holes in the central row. The holes in the matrix disk are 0.15-inch in diameter. The steel pins are 0.188 inch in diameter. The central, matrix section of the cylindrical container consists of several l5-disk, 1.875-inch long assemblies each separated by a 0.250-inch-thick nickel disk. One of the 1.5- disk assemblies is loaded with natural diamond powder, which is a blend of 35 percent by weight of nominal 6 micron powder and 65 percent by weight of nominal 60 micron powder. The loading density of the diamond powder is 72%. After detonation, the diamond is recovered by machining away most of the steel, and dissolving the remaining metal in a mixture of 60% HNO (2000 milliliters), 52% HF (4000 milliliters), and Water (1700 milliliters) at -100 C. over a period of 4 hours. The 0.15-inch-diameter diamonds recovered have a density of 90%. Graphite (3.2%) is removed by chromic acid oxidation (as described in Example 1).

Example 3 The procedure described in Example 2 is repeated with the exception that the matrix disks are made of carbon steel, and the projectile cylinder is '55 inches long. The explosive surrounding the projectile cylinder is a coaxial 5 6- inch-high stack of disks of RDX coated with 5% wax. The disks have been pressed to 1.2 g./cc. density, and have a 16-inch outside diameter, 5-inch inside diameter and 2-inch height. The space between the outer wall of the projectile cylinder and the inner wall of the stack of RDX disks is filled with granular RDX, which has been tamped to increase its packing density. The explosive is initiated with a IO-inch-diameter, 0.5-inch-thick disk of EL506-D sheet explosive placed on top of the stack of RDX disks. The sheet explosive is initiated by means of a detonating cord and a small booster. Steel powder (250 pounds) is laid on top of the explosive/initiator assembly to ensure good contact between the RDX disks. The detonation velocity of the explosive is 6650 meters per second.

The diamonds are separated from the steel matrix as described in Example 1. The 0.15-inch diameter, 0.09- inch-high diamond compacts have a density of 94% of the crystal density of diamond. Graphite (5.2 weight percent) is removed by chromic acid oxidation.

Example 4 Pressed disks containing diamond powder amassed in apertures in a copper matrix are made by slurrying 9 grams of nominal 9 micron diamond powder (particle density of and 51 grams of 100% dense 10+12 mesh copper particles (between 1.68 and 2.00 mm.) together in water, and pressing the slurry into 0.5-inchthick, 1.25-inch-diameter disks while damp. Disks of the same size and composition are made in the same manner with diamond powder consisting of 60% by Weight of nominal 36-54 micron particles and 40% by weight of nominal 2-4 micron particles. The diamond used is synthetic diamond made by the procedure described in the aforementioned US. patent. The density of the diamond aggregates in the disks containing 9 micron diamond is 48.4% of crystal density, that in the pellets containing the mixed diamond sizes is 55.1% of crystal density, these densities being computed from a diamond volume obtained by subtracting the volume of the copper, assumed to be at theoretical density, from the volume of the disk. The disks are loaded into a cylindrical container while damp, in the following manner:

The container is a Type 1015 steel hollow cylinder 53.5 inches long, and having a 1.5-inch outer diameter and a 0.125-inch wall thickness. Positioned at one end of this cylinder is a 13.5-inch-long iron disk section serving as an end plug and consisting of 18 superimposed 1.125-inchdiameter, 0.75-inch-thick disks of 100-mesh low-carbon steel powder pressed to tapering densities, gradually decreasing from 86% (of the theoretical) in the outermost disk to 70% in the innermost disk. Eight of the abovedescribed disks containing 9-micron diamond powder are loaded into the cylinder so as to form a 4-inch section (A) abutting the 70% dense end of the iron plug section. Section A is followed by a 9-inch-long iron disk section consisting of 12 superimposed 1.125-inch-diameter, 0.75-inchthick disks of l-mesh low-carbon steel powder pressed to tapering densities, gradually increasing from 71% (of the theoretical) in the end disk abutting Section A to 86% in the other end disk. Eight of the above-described disks containing mixed sizes of diamond powder are loaded into the cylinder so as to form a 4-inch section (B) abutting the 86% dense end of the 9-inch-long iron disk section. Abutting Section B is a 23-inch-long iron disk section which serves as an end plug and consists of 31 superimposed 1.125-inch-diameter disks are 0.75 inch thick and one is 0.50 inch thick) of 100-mesh low-carbon steel powder pressed to the following densities: a first portion, 7.5 inches long, has a tapering density, increasing gradually from 79% in the end disk abutting Section B to 86%; the next portion, 8.25 inches long has a uniform density of 86%; the remaining portion, 7.25 inches long, has a tapering density, decreasing gradually from 86% to the latter density in the outermost disk. The purpose of the tapered density in the plug is to match the shock impedance and density of the diamond/ copper disks to prevent shock reflections into the latter, and to permit the momentum to be carried off and forward motion of the plug to be restricted.

The loaded cylinder is heated for 18 hours at 200 F. and 6 hours at 750 F. to dry the disks and de-gas the diamond particles. The interior of the cylinder is evacuated to about 50 microns mercury pressure by means of a 0.25-inch copper evacuation tube passing through the cylinder wall. After the evacuation tube has been sealed off, the loaded cylinder is positioned coaxially within a mild carbon steel hollow cylinder 54.5 inches long, and having a 3.5-inch outer diameter and a 0.375-inch wall thickness, the coaxial position being maintained by means of steel end plates welded to both cylinders. The space between the cylinders also is evacuated.

Surrounding the outer steel cylinder is a -inch-high cylindrical layer of grained 80/20 amatol (80% ammonium nitrate/ 20% trinitrotoluene) contained in a thinwalled cardboard cylinder having an inner diameter of 16 inches. The explosive is initiated as described in Example 1. The detonation velocity is about 5900 meters per second. Detonation of the explosive causes the steel cylinders to collide progressively in an axial direction.

When the composite cylinder is out open so as to expose the diamond/copper disk sections, the copper is found to have melted completely and re-solidified, thereby allowing the diamond particles to become segregated therefrom and recoverable simply by pouring out of the container cylinder. After treatment of the diamond particles with CrO H SO solution for 15 hours to selectively oxidize any graphite present, the following particle size distribution is found by passing the particles through progressively s zed sieves:

Percent from Percent from 9 micron mixed-size Size range, in microns powder powder Larger than 590 0. 9 4. 7 250-590 15. l 30. 3 ll. 2 10. 9 12. 9 8. 9 9. 6 6. 1 5. 8 3. 2 8. O 5. 2 l0. 7 7. 5 10. 2 7. 5 l1. 5 9. 3 Smaller than 20 3. 9 6. 2

I claim:

1. In a process for shocking hard particles so as to bond them together by explosively propelling a projectile toward a body of solid material containing the particles in a manner such that the projectile collides with said body of solid material in an axially progressive manner at conforming collision surfaces, thereby introducing a shock wave into the body of solid material, the improvement which comprises (a) forming said body of solid material from (1) a carrier matrix having dispersed therethrough interstices in which the hard particles are amassed into aggregates and (2) a matrix-containment means, said matrix having a post-shock deformability which is greater than that of said particles, and having a lower degree of porosity than said aggregates, whereby the motion of said aggregates is retarded in the direction of shock travel, the peak pressure of the shock wave induced in the hard-particle aggregates being at least about kilobars; and

(b) after the collision, separating the bonded bodies present in the interstices from the matrix.

2. A process of Claim 1. wherein the body of solid material is a solid circular cylinder and the projectile is a hollow circular metal cylinder which surrounds the body of solid material.

3. A process of Claim 1 wherein the product of the initial density and the relative shock compression, AV/V is greater for the carrier matrix than for the particle aggregates at a given shock velocity, V being the initial specific volume and AV the change in specific volume from V to that in the compressed state.

4. A process of Claim 1 wherein the density and the shock impedance of the carrier matrix are higher than the density and shock impedance of the particle aggregates.

5. A process of Claim 1. wherein the carrier matrix is made of metal.

6. A process of Claim 1 wherein the carrier matrix is formed from adjacent apertured segments.

7. In a process for producing bonded diamond bodies by positioning a cylindrical body of solid material containing diamond particles coaxially within a hollow circular metal cylinder with a spacing between the facing surfaces of the cylinders and detonating a layer of explosive progressively along the outer surface of the outer cylinder so as to cause an axially progressive collision of the cylinders and introduce a shock wave into the cylindrical body of solid material, the improvement which comprises forming said body of solid material by amassing the diamond particles into aggregates within interstices dispersed throughout a circularly cylindrical carrier matrix having a greater post-shock deformability than the particles and a lower degree of porosity than the aggregates, capping the ends of the matrix with solid plug closures, and separating the resulting bonded diamond bodies from the matrix, the density of the diamond aggregates being at least about 50% of the crystal density of diamond and the peak pressure induced in the diamond aggregates being at least about 300 kilobars.

13' 14 8. A process of Claim 7 wherein the volume ratio of 1,951,174 3/1934 Simons 264-68 X the matrix to diamond aggregates is at least about 1:1. 2,225,193 12/ 1940 Benner 51-307 References Cited RICHARD R. KUCIA, Primary Examiner UNITED STATES PATENTS 5 Us C1 X R 3,667,911 6/1972 Balchan 264-84 X 423 446 3,401,019 9/1968 Cowan 423-446

Claims

1. IN A PROCESS FOR SHOCKNG HARD PARTICLES SO AS TO BOND THEM TOGETHER BY EXPLOSIVELY PROPELLING A PROJECTILE TOWARD A BODY OF SOLID MATERIAL CONTAINING THE PARTICLES IN A MANNER SUCH THAT THE PROJECTILE COLLIDES WITH SAID BODY OF SOLID MATERIAL IN AN AXIALLY PROGRESSIVE MANNER AT CONFORMING COLLISION SURFACES, THEREBY INTRODUCING A SHOCK WAVE INTO THE BODY OF SOLID MATERIAL, THE IMPROVEMENT WHICH COMPRISES (1) FORMING SAID BODY OF SOLID MATERIAL FROM (1) A CARRIER MATRI HAVING DISPERSED THERETHROUGH INTERSTICES IN WHICH THE HARD PARTICLES ARE AMASSED INTO AGGREGATES AND (2) A MATRIX-CONTAINMENT MEANS, SAID MATRIX HAVING A POST-SHOCK DEFORMABILITY WHICH IS GREATER THAN THAT OF SAID PARTICLES, AND HAVING A LOWER DEGREE OF POROSITY THAN SAID AGGREGATES, WHEREBY THE MOTION OF SAID AGGREGATES IS RETARDED IN THE DIRECTION OF SHOCK TRAVEL, THE PEAK PRESSURE OF THE SHOCK WAVE INDUCED IN THE HARD-PARTICLE AGGREGATES BEING AT LEAST ABOUT 100 KILOBARS; AND (B) AFTER THE COLLISION, SEPARATING THE BONDED BODIES PRESENT IN THE INTERSTICES FROM THE MATRIX.