US3698863A - Fibrous metal filaments - Google Patents

Abstract

A METALLIC FILAMENT WHICH HAS AN EFFECTIVE DIAMETER OF LESS THAN 50 MICRONS AND IS FORMED WHILE SURROUNDED BY A SUBSEQUENTLY REMOVED SACRIFICIAL MATRIX. THE FILAMENT HAS A PRESELECTED PERIPHERAL SURFACE VARYING FROM SUBSTANTIALLY SMOOTH TO RE-ENTRANT AND A PRESELECTED SURFACE TO VOLUME RATIO. THE AREA OF THE FILAMENT ALSO HAS A CONTROLLED NON-UNIFORMITY ALONG THE LENGTH THEREOF WHICH PROVIDES AN ACCEPTABLE DIMENSIONAL TOLERANCE. THE METALLIC FILAMENT MAY BE SUBSTANTIALLY ONE METAL, BIMETALLIC OR TUBULAR.

Description

J. A. ROBERTS ET AL FIBROUS METAL FILAMENTS 9 Sheets-Sheet l Filed Jan. 29, 1970 O. 17, 1972 1 A ROBERTS E'I'AL 3,598,3

FIBROUS METAL FILAMENTS Filed Jan. 29, 1970 9 Sheets-Sheet 2 J. A. ROBERTS EVAL 3,698,863

FIBROUS METAL FILAMENTS Oct. 17, 1972 A9 Sheets-Sheet 4 Filed Jan. 29, 1970 o? oo.

N202 XFCRE L A S T R E B O R A. J.

FIBRoUs METAL FILAMENTS 9 Sheets-Sheet Filed Jan. 29, 1970 OCt. 17, 1972 L A, ROBERTS ETAL 3,698,863

FIBROUS METAL FILAMENTS 9 Sheets-Sheet 6 Filed Jan. 29, 1970 ,OGL 17, 1972 1 A RQBERTS ETAL 3,698,863

FIBROUS METAL FILAMENTS 9 Sheets-Sheet 7 Filed Jan. 29, 1970 Oct. 17, 1972 J. A. ROBERTS ETAL FIBRoUs METAL FILAMENTS 9 Sheets-Sheet 8 Filed Jan. 29. 1970 United States Patent O U.S. Cl. 29-183.5 28 Claims ABSTRACT F THE DISCLOSURE A metallic filament which has an effective diameter of less than 50 microns and is formed while surrounded by a subsequently removed sacrificial matrix. The filament has a preselected peripheral surface varying from substantially smooth to re-entrant and a preselected surface to volume ratio. The area of the filament also has a controlled non-uniformity along the length thereof which provides an acceptable dimensional tolerance. The metallic filament may be substantially one metal, bimetallic or tubular.

This application is a continuation-in-part of our copending application Ser. No. 742,010, filed July 2, 1968, now Pat. No. 3,505,039, a division of application Ser. No. 348,326, filed Mar. 2, 1964, and issued July 23, 1968, as U.S. Pat. No. 3,394,213.

BACKGROUND OF THE INVENTION Field of the invention This invention related to fine metal filaments and in particular to controlling the characteristics of solid and bimetallic filaments.

Prior art In our original application, of which this is a continuation-in-part, methods for simultaneously forming a plurality of metal filaments while surrounded by a matrix material were disclosed. Products made by our processing were also disclosed. In addition, it was disclosed that economic manufacture of filaments could be achieved by using high ratios of filament to matrix materials. Webber and Wilson in their U.S. Pats. Nos. 3,277,564 and 3,379,000 and owned by the assignee hereof, disclosed another fine metal filament and methods for making tows of fine metal filaments. However, it would be highly desirable to provide fine metal filaments with preselected characteristics not taught or disclosed by prior teachings.

SUMMARY OF THE INVENTION Introduction-As an introduction (and for ease of understanding) it is necessary to define the units of measurement used in this disclosure. In standard metal working, i.e., machining, drawing and the like, parts are made to a desired dimension. Because no piece of equipment is perfect, a variation in the exact dimension is always permitted. This variation is commonly referred to as total tolerance or allowance. If, for example, the final diameter of a cold drawn wire is 0.005 inch, then this dimension might have a tolerance of i0.0005 inch. Thus, the total tolerance or allowance would be 0.001 inch. However, since the filaments that are the subject of this invention are formed while surrounded by a sacrificial matrix material, the co-working of the filaments and the matrix create a filament wherein such a type of measurement is not applicable in many instances. The filaice ments that are provided can have a preselectable irregular geometric cross-sectional shape (regular geometric shape being defined as a square, circle, hexagon, etc.). Therefore, the size control measure used in this disclosure is the statistical term coefficient of variation, which is described below.

It is fully contemplated to be within the scope of this invention that a filament can have a controlled cross-sectional area that is non-uniform along the length thereof but is uniform within a specied range. This range is defined by the coefficient of variation; and the coefficient variation for the filaments which was the subject to this disclosure is approximately less than 25 percent. 'I'he coefficient of variation is Written as a/E and 5:2x/n

where x is the area of each cross section taken along the length of a filament and n is the number of areas taken. In order to find the coefficient of variation, the standard deviation (having the same units as x) is calculated by using the equation:

Therefore, the term coefficient of variation as used in this disclosure means the statistical coefficient of variation of cross sectional areas taken along the length of a filament substantially perpendicular to the axis of the filament.

The filaments which are the subject of this disclosure can be provided with an increased surface area and in describing the increase in surface area obtained, it is necessary to relate the increase to a standard. The standard is taken to be the surface area of a unit length cylinder having a cross section area equal to the cross section area of the filament. The enhancement in surface to volume ratio is defined as the difference between the surface area of a filament of unit length and the surface area of the equivalent cylinder of unit length divided by the surface area of the equivalent cylinder of unit length. This enhancement is expressed as a percentage.

Summary.-The advances in metal filament (or fiber) technology disclosed by Webber et al. and by Roberts, et al. were the basis for the formation of a commercial metal filament industry.

In this Idisclosure new and novel filament forms will be disclosed which function by providing preselected properties hitherto unobtainable and yet highly desirable.

This invention relates to fine metal filaments and is concerned with new and novel filaments that have preselected peripheral surfaces, controlled uniformity, preselected `urface to volume ratios and filaments that are bimetal- The present invention is concerned with providing highly desirable metal filaments by selecting the proper ratios of preselected matrix and filament materials to achieve: economical manufacture of such filaments; preselected surface volume ratios of such filaments; preselected surface periphery of such filaments; and, maintaining at low levels the coefficient of variation of the cross-sectional area.

Thus, a principal object and feature of the present invention is to provide a new and improved metal filament formed while in a sacrificial matrix material and having a preselected surface to volume ratio achieved by the proper selection of filament and matrix materials.

Another object of the invention is to provide a filament having smooth or rough peripheral surface determined by the metallurgical characteristics and physical propertes of the filament and matrix materials and their interaction during co-working.

Another object of the invention is to provide a metallic :filament wherein a series of cross-sectional areas taken along the length of said filament substantially perpendicular to the axis of said filament exhibits a coefficient of variation from 1% to less than 25 percent.

Still another object o-f the invention is to provide for utilizing a preselected ratio of matrix material to filament material to achieve an economically manufactured metal filament when the filament has an unburnished, unmachined outer surface, is fracture free and has a cyclical geometric variation along the length thereof, yet remaining under a 25 percent coefficient of variation.

And still another object of the invention is to provide a filament having a non-uniform cross-sectional area, but a controlled uniformity Within a coefficient of variation from 1 percent to less than 25 percent wherein the coefficient of variation is a function of the combination of: (a) the volume ratio of the matrix to filament material; (b) the amount and temperature of the hot working and/ or cold working between anneals (if desired) imparted to the starting filament and matrix materials; and, (c) the choice of matrix material for a preselected filament material.

Still a further object of the invention is to provide a bimetallic filament.

The above and other further objectives and features will be more readily understood by reference to the following detailed descriptions and the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a transverse cross-section of a metal wire from which a filament may be formed in following the method embodying the invention;

FIG. 2 is a transverse cross-section of the wire disposed within a coaxial sheath as in a first step of the method;

FIG. 3 is a perspective view of a plurality of sheathed wires disposed in a cylindrical housing in a subsequent step, the housing being broken away to illustrate the bottom portion thereof;

FIG. 4 is a reduced vertical section of a compact means illustrating a compaction of the assembly of sheathed wires in the cylindrical housing to define a compacted billet;

FIG. 5 is a reduced vertical section of another form of compacting means illustrating another method of reducing the diameter of the assembly to define a compacted billet;

FIG. 6 is a top plan view of a plurality of sheathed Wires in a modified housing having a hexagonal internal cross-section as by the provision therein of sector shaped spacers;

FIG. 7 is a top plan view of a modified arrangement of the sheathed elements in a cylindrical housing with spacers disposed between the sheathed elements to effectively minimize the voids therein;

FIG. 8 is an exploded vertical section illustrating the arrangement of the housing subsequent to the installation of the sheathed wires therein and prior to the securing of the end plug across the open end thereof;

FIG. 9 is a vertical section illustrating the arrangement of the sheathed wires in the housing with the end plug secured across the open end of the housing;

FIG. l0 is a vertical section illustrating the step of evacuating and sealing of the housed sheathed wires to define the primary billet;

FIG. 11 is a fragmentary diagrammatic elevation of the billet during a subsequent hot extrusion step;

FIG. 12 is a diagrammatic elevation of the extruded billet with suitable cutting means acting to remove the opposite ends of the extruded bundle;

FIG. 13 is a diagrammatic elevation illustrating the cutting of the extruded billet into a plurality of shorter lengths;

FIG. 14 is an elevation of one of the short lengths being provided with a replacement plug at each of its opposite ends;

FIG. 15 is a fragmentary side elevation of a short length being further hot extruded to reduce the diameter thereof;

FIG. 16 is a side elevation of the original billet being reduced in diameter as by hot rolling means in lieu of or subsequent to the hot extrusion means of FIG. 1l;

FIG. 17 is a fragmentary diagrammatic vertical section illustrating the cold drawing of the hot formed billet in a subsequent step;

FIG. 18 is a vertical cross section of a tank wherein the drawn billet of FIG. 17 is disposed to be acted upon by a suitable fluid within the tank to remove the sheathing and housing material from the billet;

FIG. 19 is a resultant tow of filaments embodying the invention;

FIG. 20 is a transverse cross section of a metallic tubular element from which a tubular filament may be formed in accordance with the invention, a filler being disposed within the tubular element as in a first step of the method;

FIG. 21 is a transverse cross-section of the filled tubular element disposed within a tubular sheath as in a second step of the method;

FIG. 22 is a transverse cross section thereof constricted to provide a tight assembly of the filler, tubular element, and sheath as in a third step of the method;

FIG. 23 is a perspective top view of the filled and sheathed tubular elements disposed within a cylindrical housing for subsequent sealing, hot forming, and drawing steps as illustrated in FIGS. 7 through 17;

FIG. 24 is a fragmentary enlarged perspective view of a tubular filament formed by the method;

FIG. 25 is a fragmentary enlarged perspective view of a bimetallic filament;

FIG. 26 is a transverse cross section of a can with sheathed elements packed in a hexagonal array having spacing shims between the faces of the hex and the inside diameter of the can;

FIG. 27 is a fragmentary enlarged longitudinal cross section of two non-uniform filaments in a matrix material that exceed a 25% coefiicient of variation;

FIG. 28 is a fragmentary enlarged longitudinal cross section of discontinuous filaments in a matrix material;

FIG. 29 is a graph showing the regions of gross nonuniformity and background non-uniformity when plotted on a chart wherein the coefficient of variation is the ordinate and the degree of deformation is the abscissa with Wc defined as the critical deformation and the start of the gross non-uniformity region;

FIG. 30 is an array of three graphs showing gross and background non-uniformity for different matrix material having the same ratio of matrix to filament material.

FIG. 3l is a graph with different curves for different ratios of the same matrix material to the same filament material wherein the coefficient of variation is the ordinate, the degree of deformation is the abscissa;

FIG. 32 is a graph showing regions for stainless steel filament material and different matrix materials wherein the ordinate is the volume ratio of matrix to total composit material and the abscissa is the critical degree of deformation;

FIG. 33 is a photomicrograph at 605x magnification of a composite showing filaments of substantially Ti- 55A (alpha-titanium) material still surrounded by Monel metal matrix material; the composite is as extruded;

FIG. 34 is a photomicrograph at 245 magnification of the composite of FIG. 33 wherein the composite has been cold worked;

FIG. 35 is a photomicrograph at 125 magnification of a composite showing filaments of substantially Ti-6Al- 4V (alpha and beta titanium) material still surrounded by an AISI 1010 steel matrix material; the composite is as extruded;

FIG. 36 is a photomicrograph at 240x magnification of a composite showing filaments of substantially Ti-6Al- 4V (alpha and beta titanium) material still surrounded by an AISI 1010 steel matrix material; the composite is as extruded;

FIG. 37 is a photomicrograph at 240 magnification of the composite of FIGS. 35 and 36 wherein the composite has been cold worked;

FIG. 38 is a photomicrograph at 240 magnification of a composite showing filaments of substantially Til3V-l1Cr-3Al (beta titanium) material still surrounded by an AISI 1010 steel matrix material; the composite is as extruded;

FIG. 39 is a photomicrograph at 240x magnification of the composite of FIG. 38 wherein the composite has been cold worked;

FIG. 40 is a photomicrograph at 590x magnification of the composite of FIG. 38 wherein the composite has been cold worked;

FIG. 41 is a photomicrograph at 245x magnification of a composite showing filaments of substantially Nickel 270 material still surrounded by an AISI 1010 steel matrix material; the composite is in a cold worked condition;

FIG. 42 is a photomicrograph at 605x magnification of a composite showing filaments of substantially Nickel 270 material still surrounded by an AISI 1010 steel matrix material; the composite is in a cold worked condition;

FIG. 43 is a photomicrograph at 240x magnification of a composite showing filaments of substantially Nickel 270 material still surrounded by an Aluminum 1100 matrix material; the composite is in a cold worked condition;

FIG. 44 is a photomicrograph at 590x magnification of a composite showing filaments of substantially Nickel 270 material still surrounded by an Aluminum 1100 matrix material; the composite is in a cold worked condition;

FIG. 45 is a photomicrograph at 240x magnification of a composite showing laments of substantially Nickel 270 material still surrounded by an Aluminum 5052 matrix material; the composite is in a cold worked condition;

FIG. 46 is a photomicrograph at 600)( magnification of a composite showing the same filaments of FIG. 45 wherein the composite is in a cold worked condition; and

FIG. 47 is a dual graph wherein part A is a plot of the average effective diameter of filaments in a bundle shown as the ordinate and the degree (percentage) of deformation is the abscissa while part B is the coefficient of variation for the same filaments also plotted vs. the degree of deformation.

'PREFERRED EMBODIMENTS OF THE INVENTION In an exemplary embodiment of the invention, a tow generally designated of filaments 11, as shown in FIG. 19, is formed by a process wherein a plurality of elongated elements, or wires, 12 are bundled in side-byside relationship and, when so bundled, reduced in diameter by a transverse, or radial, constriction of the wires in the bundle to provide a resultant filament of extremely small diameter and great length. In addition, the invention comprehends the forming of the filament as a tubular filament 13, as shown in FIG. 24, and a bimetallic filament 113 as shown in FIG. 25, the original elongated element in this process comprising a tubular element 14 as shown in FIG. 20.

Broadly, the invention comprehends the constriction of the bundled wires, or tubular elements, by firstly, forming the bundled wires or elements into a billet, and subjecting the billet successively to a hot forming constriction and a subsequent drawing constriction. The hot forming constriction may be alternatively by hot extrusion or hot rolling of the billet. The drawing operation may comprise a plurality of cold drawing steps with intermediate annealing steps.

Referring now to FIGS. 1 and 2, the wire 12 is firstly enclosed in a suitable sheath 15 formed of a material having metallurgical and physical properties differing from those of the wire 12 to permit separation of the sheath material from the resultant filaments when desired. However, the mechanical flow properties of the lament and matrix material may be similar, as desired. As illustrated in FIG. 2, the original internal diameter of the sheath may be slightly larger than the external diameter of the wire 12 to permit facilitated insertion of the wire into the sheath. Alternatively, the wires 12 and the sheaths 15 may =be co-reduced, such as by cold drawing through a die, to form a composite wherein the Wire 12 is tightly encased by the sheath 15 (but not illustrated). The thusly, loosely or tightly sheathed wires may then be installed in a can, or `housing 16 having a bottom closure wall, or nose plug, 17,

with the sheathed wires extending in parallel side-by-side` upright relationship, as shown in FIG. 3.

For improved uniform constriction of the wires in the subsequent constricting steps, it is desirable to closely pack the sheathed wires within the housing 16 as by suitably compacting the assembly. Referring to FIG. 4, one method of effecting the desired compaction is by placing the assembly in a press generally designed 18 having a liner 18a defining a cylindrical cavity closely fitting the cylindrical housing 16. The lower end of the cavity is closed by a blind die 18b and the liner 18a and the blind die 18b are supported on a suitable anvil 18C. A ram 18d is provided to apply pressure on the top of the assembly whereby the assembly is axially shortened and thereby laterally or radially compacted. Such compacting apparatuses are well known in the art and need no further description herein.

Turning now to FIG. 5, an alternate method of effecting the desired compaction of the assembly is shown to comprise the compaction of the assembly by means of an extrusion apparatus generally designated 118. In apparatus 118 an extrusion die 118g is provided through which the assembly is longitudinally forced by means of a suitable pressure applying element 118b. Only a small amount of constriction of the assembly is effected by die 118a so that only an elimination of the voids in the assembly is effected in this step.

IReferring now to FIG. `6, a method of facilitating the compaction of the assembly is illustrated. More specifically, the internal configuration of the housing 16 is made to be hexagonal in transverse cross section by means of a plurality of spacers 19 comprising chordal sector pieces or fiat shimstock stacked spacers 19a shown in FIG. 26.

FIG. 7 illustrates a further method of facilitating the compaction of the assembly. More specifically, in FIG. 7 the sheathed wires 12 are shown installed within the housing 16 with a plurality of spacers, or suitable particulate materials such as metal powder, 21 disposed between the wires. Thus, with the arrangements illustrated in FIGS. 6 and 7, less compaction of the assembly by the steps illustrated in FIGS. 4 and 5 is required to provide the desired compacted billet.

Prior to the compaction steps discussed above, the sheathed wires 12 are sealingly enclosed within the housing 16 by means of an end plug 23 installed across the open end 24 of the housing 16. As illustrated in FIG. 8 the end plug comprises a generally cylindrical disk which may have a notched portion 23a adapted to engage the upper end of the housing in the housing closing arrangement. The end plug 23 is further provided with an evacuation pipe 26 which opens through a hole 23b in the end plug, being secured to the end plug by suitable means such as weld 26a.

With the sheathed wires 12 or composites 112 installed in the housing 16, as shown in FIG. 10, the end plug 23 is secured across the open upper end 24 of the housing by suitable means such as weld 23C. The evacuation pipe or tube 26 may be utilized during the welding of the end plug to the housing end 24 to fiush the interior of the housing during the welding of the plug. Upon completion of the installation of the end plug on the housing, a vacuum is applied to the pipe 26 by suitable means (not shown) to withdraw substantially all gas from the interior of the housing.

As shown in FIG. 10, when the desired vacuum condition is obtained within the housing 16, the pipe 26 is pinched and welded closed as at 26b to complete the sealing of the wires 12 within the housing 16. To provide an improved vacuum within the housing 16, the housing may be disposed within a suitable conventional heater 27.

The resultant housed bundle comprising a packed billet 31 is next subjected to a hot forming process to reduce the diameter thereof in one or more passes. As illustrated in FIG. 1l, the billet 31 may be reduced in diameter by a hot extrusion step wherein the `billet is forced through heated extrusion dies 32 by a suitable pressure device 33. It is desirable that the billet 31 be preheated to a preselected suitable temperature and suitably lubricated for facilitated extrusion in this step. The rate and pressure of the extrusion is preselected to provide optimum diametric reduction of the billet in conformity with the nature of the materials involved. In the event that a second extrusion step is employed, the opposite ends 34 of the reduced diameter billet 35 are trimmed (see FIG. 12) as by suitable cutters 36. Any non-uniform extruded end portions of the billet as determined by observation thereof may be included in the ends 34 so cut from the billet. The trimmed billet 35 is then divided into a plurality of short lengths 37, as shown in FIG. 13, by suitable means such as cutting wheels 38. Each of the short length billets 37 is then provided with a nose plug 39 and a tail plug 40 as by welding as illustrated in FIG. 14. The short length billet 37 is then reheated and passed through heated extrusion dies 41 (see FIG. for further diametric constriction thereof to a final formed billet 42.

As indicated briefly above, the hot forming of the billet 31 may be effected by hot rolling the billet in lieu of the extrusion thereof. Thus, as shown in FIG. 16, the billet 31 may be suitably heated and passed between suitable rolls 43. The billet may be firstly hot formed by the extrusion step illustrated in FIG. l1 and subsequently hot forming effected by hot rolling as desired. The rolls 43 are preferably arranged to produce a hot forming constriction of the billet wherein the elements therein can be maintained in a substantially smooth circular crosssectional configuration.

Alternatively, the elements or filaments can be formed to have a rough re-entrant periphery, if desired, such as shown in the photomicrograph, FIG. 38.

Subsequent to the hot forming steps discussed above, as shown in FIG. 17, the resultant final formed billet 42 is drawn through a suitable conventional cold drawing die 44 by a suitable conventional drawing device 45. The billet may be successively drawn down to smaller and smaller diameters by means of successively smaller dies 44 to produce the final composite diameter containing the desired small diameter filaments. Annealing may be effected between the successive drawing steps in conformity with the requirements of the metal of which the filaments are formed. The cold drawing of the billet may be conducted suitably to develop texture in the filaments and to work-harden the filaments for providing improved mechanical properties thereto.

Where the hot rolling steps are employed, the final cold drawing of the billet may be dispensed with, such as where the physical characteristics provided by the cold drawing are not required. Thus, the hot rolling steps may be continued with successively smaller rolls 43 providing the desired ultimate constrictive deforming of the billet whereby the filaments may be made to have the desired diameter of approximately 50 microns or less.

The filaments are released from the final constricted billet 46 by suitable means such as selective chemical attack of the sheating 15 and can 16. Thus, as shown in FIG. 18, the final billet 46 may be disposed within a tank 47 holding a suitable acid 48 to dissolve the sheathing and can material. Obviously, other methods of removal of the can and sheathing material may be employed; illustratively, the sheathing and can material may be removed by electrochemical dissolution, thermal removal, selective oxidation, mechanical removal, etc. In the final tow 10 of filaments 11, as illustrated in FIG. 19, the filaments have an extremely small diameter, for example, down to approximately 0.002 inch or 50 microns or less. Where the filaments are formed by utilizing the hot extrusion process With subsequent cold drawing, the tow filaments may have a length of up to 50,000 feet or more, and where the tow filaments are formed by the hot rolling process, the length may be up to 300,000 feet or more. It is to be understood that a continuous filament means a single filament having a length of at least 5 feet as compared to a staple filament, which is dened to mean any filament having a length of from approximately 1454, inch to approximately 24 inches and having an aspect ratio of at least 10 to 1.

It has been found that when the filaments are formed in a metal matrix and then released therefrom that the filaments have a scale free outer surface that is longitudinally fracture free (as compared with steel wool shaving formed by a cutting or chipping action), unburnished (as compared with single end wire drawing) and unmachined. It is believed that during the reducing steps the filament material is subjected to a triaxial stress system of reducing forces that is substantially different from that imposed on a single solid material deformed by the same reducing method or process (i.e., wire drawing, rod rolling, swaging, extruding and the like).

More specifically, the wire 12 of which the filaments are ultimately formed may comprise a metal wire formed of a suitable material such as nickel, type 304 stainless steel, titanium and its alloys and the like. The sheathing 15 may comprise of mild steel, copper, aluminum and alloys thereof, Monel metal and the like. Several examples of material of which the can 16 may be formed are mild steel, Monel metal, copper and the like. Thus, in the final step the can 16 and the sheathing material 15 may be removed from the filament 12 by the use of an acid fluid 48 or other means as described above.

The following specific examples of the filament forming process and the filaments formed thereby are intended only to illustrate the invention and not to limit it in any way.

EXAMPLE 1 Type 304 stainless steel wires having a 0.250 inch diameter and an 18 inch length were inserted into Monel 400 sheaths (tubes) having a `0.293 inch outside diameter and a 0.253 inch inside diameter and an 18 inch length. The lower end of an AISI 1010 (mild) steel can having a 5.950 inch outside diameter, a 5.25 inch inside diameter and a 22 inch length was fitted with a partially slip-fit included angle (45 on a side) frusto-conical nose plug and welded thereto; thereby defining a closed end cavity. The can cavity was packed with two hundred sixtyeight (268) sheathed wires. An end plug was fitted and welded to the upper end of the can; the end plug having a 1A inch evacuation tube secured thereto. The composite can billet was evacuated to 0.1 micron of mercury (this is 10-4 torr) at 800 F. and then sealed off. The billet was heated to a temperature of 1800 F. for six hours in a graphite container. The heated billet was extruded in a press using a heated extrusion die (900 F.) to a diameter of 2.925 inches whereby an extrusion ratio of 4.3 times was obtained with the billet being quenched in water immediately thereafter. The press was operated at a linear ram speed of approximately 500 inches per minute with a 1340 ton upset force and a 1200 ton running force. The resultant first extrusion was cut into pieces 10 inches long. A new 90 included angle nose plug was welded to the lower end of the new inch long billet and a 1/2 inch thick end plug was welded to the upper end thereof. The new 10 inch billet was heated to a temperature of 1800 F. for approximately three hours in a graphite container. The billet was extruded in a press using a heated extrusion die to a diameter of approximately `0.625 inch providing an extrusion ratio of 22.8 times in area. The billet was extruded at a rate of 145 inches per minute with an upset force of 590 tons and a running force of 560 tons with the billet again being immediately quenched in water. The 0.625 inch billet was then cold drawn by 20% area reduction per pass for four (4) passes, or v60% area reduction, between annealing steps. The billet was annealed several times at a temperature of 1700 F. and at a rate of approximately two seconds per 0.001 inch of billet diameter during the cold drawing schedules. The final billet (composite) diameter was 0.016 inch. The can and the sheath were removed by a nitric acid leach providing filaments of approximately 0.0007 inch in diameter. These filaments had an average ultimate tensile strength of approximately 250,000 p.s.i. and an average elongation of 2.1% in the cold worked condition. In the annealed condition these filaments had an approximate ultimate tensile strength of 109,000 p.s.i. with an average elongation of approximately 11% in the example.

EXAMPLE 2 Type 304 stainless steel wires having a 0.080 inch diameter were inserted into Monel 400 sheaths (tubes) having a 0.097 inch outside diameter and a 0.085 inch inside diameter. The sheathed Wires were drawn through a 0.091 inch diameter die to facilitate mating of the surfaces of the stainless steel and Monel and were then straightened and cut into three inch lengths. An AISI 1010 (mild) steel can having a 1.970 inch outside diameter, a 1.740 inch inside diameter and an overall length of 6 inches was fitted with a partially slip-fit 90 included angle (45 on a side) frusto-conical nose plug which was welded to the lower end of the can. Two hundred and forty-two (242) of the sheathed wires were placed in the can with an end cap slip fitted into the can and welded thereto, thereby forming a billet. The end cap having a 1A inch diameter evacuation secured thereto. The billet was evacuated to less than 0.1 micron of mercury at 800 F. and then sealed off. The evacuated billet was then heated to 1800 F. for two hours in a graphite container. The hot billet was extruded (in a press using a heated extrusion die) to a diameter of approximately 0.500 inch, whereby an extrusion ratio of 16 times in area was obtained. The billet was extruded at the rate of approximately 65 inches per minute with an upset force of 272 tons and a running force of 260 tons.

The 0.500 inch diameter extruded billet was cut into 4 portions. One portion of the billet was drawn to a final composite diameter of 0.008 inch with annealing treatments of 1700 F. for 2 seconds per 0.001 inch of diameter interposed at substantially 60% area reduction intervals.

The final diameter of the resultant 242 filaments was found to be approximately 0.00034 inch with a coefficient of variation of cross-section area of 7.75%.

The remaining three portions of the extruded billet were all drawn to a final composite diameter of 0.016 inch in such a manner -as to impart 36%, 75% and 90% reduction of area in the respective portions after the last anneal. In all three portions the filament diameter was found to be approximately 0.0007 inch or 17.5 microns.

The coefficient of variation of the cross-sectional area and the average effective diameter of the filaments of the three composites are shown in FIG. 47 wherein part A of the graph shows the effective diameter as the ordinate and the degree of deformation as the abscissa and p-art B the coefficient of variation (t1/2E) also as the ordinate. As shown on this graph, the coefficient of variation of the filament area is well within the self-imposed 25% limitation. The average effective diameter of the filament was calculated by equating the average filament area to a circle having the same area and then solving for the diameter of the circle. This diameter is the Kaverage effective diameter of the filaments. It can be seen from this graph that the effective diameter for the 36%, 75 and deformably worked filaments does not 'vary more than 0.02 mil (0.0002 inch).

By using the same method the coefficient of variation of the filaments from Examples 3, 4, 5 `and 6, obtained and exhibited respectively coefficients of variation of 8.1%, 4.46%, 11.7%, and 10%. For Example 7 the 2 portions had coefficients of variation of 6.3% (Ni/Al-ll00) and 13% (Ni/Al-5052).

EXAMPLE 3 Titanium-55A (commercially pure type titanium) wires having a 0.080 inch diameter were inserted into Monel 400 sheaths (tubes) having a 0.115 inch outside diameter and a 0.100 inch inside diameter. The wires and sheaths were in an annealed condition and then drawn to a composite diameter of 0.085 inch with the titanium core wires being 0.073 inch. The drawn combination was then straightened and cut into three inch lengths. A unitary OFHC copper can with an integral nose cone was machined wherein the outside diameter of the can was 1.630 inch, the inside diameter of the can was 0.950 inch and the overall length of the cavity of the can was 3%. inches. Ninety-one sheathed titanium wires were packed in the can in an hexagonal array with 11 sheathed wires constituting the vertex to vertex spacing with six sheathed wires comprising any and all faces. Three inches long by 0.015 inch thick copper shims were placed adjacent to the hexagonal faces of the packed rods. Additional back-up shims were stacked between the inside diameter of the copper can and the primary shims in order to reduce the open or void area within the can. An end cap having a 1A inch evacuation tube secured thereto was slip fitted into the can and welded in place so that there was substantially no longitudinal movement of the rods. The billet was evacuated to 10-5 torr at 800 F. and sealed thereafter. The billet was then heated to 1400 F. and extruded to a diameter of 0.270 inch from a 1.100 inch extrusion liner. A photomicrograph of a portion of the extruded billet is shown in FIG. 33 wherein the filaments have an approximate effective diameter of approximately 0.020 inch. The billet was extruded with an upset force of 56 tons. The billet was cold worked from 0.270 inch diameter through a series of drawing dies with intermediate anneals at 1000 F. employed after different drawing diameters. The resultant composite was drawn to 0.0126 inch having been reduced 50% by cold work after the last anneal. The can and surrounding Monel tubing were dissolved in nitric acid. The final diameter of the resultant filaments was found to be approximately 0.0009 inch. A photomicrograph of a portion of the final cold drawn composite showing filaments with reentrant peripheries is shown in FIG. 34. The enhancement of surface to volume ratio of these filaments is 38.8%. The coefficient of Variation of these filaments was found to be 8.1%.

EXAMPLE 4 Ti-6Al-4V (titanium alloy) wires having a 0.079 inch diameter were inserted into AISI 1010 (mild) steel sheaths (tubes) having a 0.115 inch outside diameter and a 0.100 inch inside diameter. The sheathed wires were then swaged to have an outside diameter of 0.0875 inch with the titanium alloy wires having a 0.076 inch diameter; the swaged sheathed wires were straightened and were cut to 3inch lengths. An AISI 1010 (mild) steel can having a 1.250 inch outside diameter, a 1.150 inch inside diameter and an overall length of 3`1/2 inches was fitted with a 1A inch thick slip lit lower end cap with the cap welded thereto. Adjacent to this end cap a 90 included angle nose plug was secured to the can by means of welding. One hundred twenty-seven sheathed rods were packed into the can cavity in an hexagonal array. In between each of five sides of the hexagon and the circular interior of the can, two additional sheathed rods and two 0.080 inch diameter (mild) steel rods were inserted. In between the sixth side of the hexagon and the interior surface of the can shims were packed in order to fill the remaining void space; therefore, the packed can contains 137 sheathed titanium alloy wires. An upper end plate 1A inch thick and having a 1A inch evacuation tube secured thereto was slip fitted into the can cavity and welded thereto; thereby forming a billet. The billet was evacuated to -5 torr. at 800 F. and then sealed off. To enhance the extruding capabilities of the evacuated billet, a mild steel cylinder having a 1% inch outside diameter, a 1/2 inch inside diameter and 11A inch length was placed over the protruding sealed evacuation tube with the cylinder then being tack Welded to the back of the billet. The billet was then heated to 1600 F. and extruded from a 1.280 inch extrusion liner through a 0.370 inch die (12 times reduction in area) when an upset force of 99 tons and a running force of 76 tons were employed. Photomicrographs of a portion of the extruded billet are shown in FIGS. 35 and 36 wherein the filaments have an effective diameter of approximately 0.024 inch. The extruded billet was then cold drawn through a series of cold drawing steps to a final diameter of 0.018 inch. The billet was annealed between several of the drawing steps at a temperature of 1450 F. The can and the sheathing (matrix) material were then removed by chemical dissolution in nitric acid. The final diameter of the resultant 137 filaments was found to -be approximately 0.00116 inch. A photomicrograph of a portion of the final cold drawn composite showing filaments with re-entrant peripheries is shown in FIG. 37. The enhancement of surface to volume ratio of these filaments is 33.8%. The coefficient of variation was observed to be 4.46%.

EXAMPLE 5 Ti-13V11Cr-3Al (titanium alloy) wires having a 0.079 inch diameter were inserted into AISI 1010 (mild) steel sheaths (tubes) having a 0.115 inch outside diameter and a 0.100 inch inside diameter. The sheathed wires were swaged to have an outside diameter of 0.086 inch and were then straightened and cut to 31/2 inch lengths.

An AISI 1010 (mild) steel can having a 1.493 inch outside diameter, a 1.370 inch inside diameter and an overall length of 4% inches was fitted with a li inch thick lower end cap inserted into the bore of the can and welded thereto. A frusto-conical nose plug having a 3A; inch flat portion and a 90 included angle was welded to the lower end of the can adjacent to the previously welded flush end cap. Thecan was packed with 199 sheathed wires in a substantially hexagonal array with mild steel Wires having diameters of 0.071 inch, 0.041 inch, and 0.028 inch inserted around the periphery of the hexagonal array to provide a tightly packed can. 4In order to improve the extrusion characteristics, a 1.365 inch outside diameter by 1 inch long (mild) steel cylinder was slip-fitted into the back of the can. A 1A; inch thick end cap was then slipped into the upper end of the can and welded thereto, thereby forming a billet; the end cap having a '1A inch diameter evacuation tube secured thereto. The billet was evacuated to l0F5 torr at 800 F. and then sealed off. The evacuated billet was then heated to l650 F. The hot billet was extruded (in a press using a heated extrusion die) from a 1.530 inch diameter extrusion liner through a 0.382 inch diameter die. The billet was extruded with an upset force of 155 tons. A photomicrograph of a portion of the extruded billet is shown in FIG. 38 wherein the filaments have an effective diameter of approximately 0.0185 inch. The 0.382 inch diameter extruded billet was then cold drawn down to 0.036 inch With intermediate anneals using an annealing temperature of l500 The can and sheathing (matrix) material were then removed by chemical dissolution in nitric acid. The final diameter of the 199 filaments was found to be approximately 0.0017 inch. Photomicrographs of a portion of the final cold drawn composite showing filaments with highly re-entrant peripheries are shown in FIGS. 39 and 40. The enhancement of surface to volume ratio of these filaments is 56%. The coefficient of variation was observed to be 11.7%.

EXAMPLE 6 Nickel 270 wires having a 0.080 inch diameter were inserted into AISI 1010 (mild) steel sheaths (tubes) having a 0.115 inch outside diameter and a 0.100 inch inside diameter. The sheathed wires were drawn to a 0.085 inch outside diameter with the nickel wires reduced to approximately 0.070 inch diameter. An AISI 1010 (mild) steel can having a 1.063 inches outside diameter, a 0.950 inch inside diameter and a 3% inch length was fitted with a -lz inch thick lower end cap inserted into the bore of the can and welded thereto. A frusto-conical nose plug with a included angle was tack welded to the lower end of the can adjacent to the previously welded end cap. The can was packed with 9'1 sheathed wires disposed in an hexagonal array giving 11 sheathed wires from vertex to vertex and 6 sheathed Wires on any one side of the hexagon. Several layers of mild steel shim stock were placed between the hexagonally packed rods and the interior surface of the can. A 1A; inch thick endcap was slip fitted into the can and welded thereto thereby forming a billet; the end cap having a 2% inch diameter evacuation tube secured thereto. The billet was evacuated to 10-5 torr at 800 F. and sealed. The evacuated billet was heated to 1600 F. and extruded to a diameter of 0.270 inch from a 1.10() inch extrusion liner at a rate of inches per minute. The billet was cold Worked from the 0.270 inch diameter through a series of drawing dies with intermediate anneals at 1650 F. When the cold drawn composite had a 0.014 inch diameter it was cut into two portions. The first portion was subjected to a leaching operation using phosphoric acid (H3PO4) to dissolve the can and matrix material. The final diameter of the resultant nickel filaments was found to be approximately 0.001 inch diameter. The second portion was cold drawn through another series of reducing steps to a final composite diameter of .0071 inch. The can and surrounding matrix material were dissolved in phosphoric acid. The final diameter of these second resultant filaments was found to be approximately 0.0005 inch. The coefficient of variation of the 0.001 inch and 0.0005 inch filaments was found to be 9.45% and 7.5% respectively. FIGS. 41 and 42 are photomicrographs of filaments in a 0.022 inch diameter composite. This composite subtends filaments with an eective diameter of 0.0016 inch and the filaments have a coefficient of variation of 10% and an enhancement of surface to volume ratio of 31%.

For suitable softer materials, such as aluminum alloys, used as the sheathing 15 or matrix and nickel elements, it has been found that the hot extrusion can be eliminated and only cold drawing steps performed on the billet. Thus, the following example is an illustration thereof.

EXAMPLE 7 Nickel 2704 wires having a 0.060 inch diameter were inserted into aluminum 5052 sheaths (tubing) having a 0.115 inch outside diameter and a 0.100 inch inside diameter. The sheathed wires were drawn to a 0.0475 inch diameter and were then straightened by stretching, and cut into 31/2 foot lengths. An aluminum 5052 can having a 0.625 inch outside diameter, a 0.555 inch inside diameter and a length of 3% feet was packed with 97 sheathed wires. The packed can underwent cold drawing by 20% reduction in area per pass. Protracted cold reduction reduced the packed can to a 0.0071 inch outside diameter. The can and sheathing (matrix) material were then removed by chemical dissolution in caustic soda (NaOH) solution. The final diameter of the resultant 97 nickel filaments was found to be approximately 0.0005 inch. During the cold drawing steps photomicrographs (see FIGS. 45 and 46) of the cross section of this composite were taken when the composite diameter was 0.0202 inch and the filaments were 0.0015 inch. These filaments exhibited substantially smooth peripheries with an enhancement of surface to volume ratio of only 6% and a coeflicient of variation of 13%. A similar composite was made using nickel 270 wires sheathed in aluminum 1100 which were encased in a Monel 400 tube. Sixty-one such sheathed Wires were placed in the tube. The packed Monel tube was cold drawn to a diameter of 0.020 inch with the can and sheathing (matrix) material being removed by chemical dissolution. The final diameter of the resultant 61 filaments of this example was found to be approximately 0.0019 inch and a portion thereof appeared in the photomicrographs of FIGS. 43 and 44. These filaments exhibited extremely smooth peripheries with an enhancement of surface to volume ratio of only 1.7%. The coefiicient of variation of filament cross-section area was found to be 6.3%.

As indicated brieliy above, the present invention comprehends the forming of tubular and/or bimetallic filaments by a process similar to that used in forming solid metal filaments. As illustrated in FIGS. 20-24 in the forming of tubular or bimetallic filaments, the starting elongated element comprises a solid wire core 49 inserted into a suitable sheath 50 made from a material different from the Wire. The wire 49-sheath 50 combination is then inserted into a tube 14 made from a material different from the sheath 49 material. The material used for the wires 49, sheaths 50 and tubes 14 may have different characteristics so that the tubes 14 can be removed from the sheath 50 and the wires (cores) 49 can also be removed from the sheath 50, as desired.

Illustratively, the wire-sheath-tube arrangement may be assembled into a composite 51 by constrictively drawing the three components together to form intimate contact therebetween. The composites 51 are cut to suitable length and packed in a desired array in can S2. Can 52 is then capped at both ends, evacuated and sealed in a similar manner to the method described above to form a packed and evacuated billet. The billet is then heated and extruded with subsequent cold drawing thereafter, thereby providing the desired tubular or bimetallic filament size. The cold drawn billet can then be cut into either short or long lengths, as desired. The can, tubes (matrix) and' cores may be removed from the short length composites thereby forming a plurality of tubular elements (the original sheaths). Alternatively, either the short length composites or longer length composites may have the can and tubes (matrix) removed thereform thereby forming either short or long bimetallic filaments (original Wire and original sheath). The following specific examples of tubular or bimetallic filament forming processes and the tubular or bimetallic elements formed thereby are intended only to illustrate the invention and not limit it in any Way.

EXAMPLE 8 Copper Wires having a diameter of 0.046 inch were inserted into Type 304 stainless steel sheaths having an outside diameter of 0.088 inch and an inside diameter of 0.048 inch which were then inserted into Monel 400 tubes having an outside diameter of 0.105 inch and an inside diameter of 0.090 inch. The wire-sheath-tube combination was cold drawn through drawing dies to give a resultant composite with an outside diameter of 0.100 inch, the stainless steel sheath had an outside diameter of 0.085 inch and the copper wire had an outside diameter of 0.045 inch. Ther composite was straightened and cut into three inch lengths. An AIS-I 1010 (mild) steel can having an outside diameter of 1.063 inch and an inside diameter of 0.920 inch and a Sil/z inch length was fitted with a partially slip-fit included angle frustoconical nose plug with the nose plug welded to the lower end of the can. The can was packed with 61 wire-sheathtube composites and an end-plug was fitted into and welded to the upper end of the can; the end cap having a 1A inch diameter evacuation tube secured thereto. The composite can or billet was evacuated to l05 torr at 800 F. and then sealed off. The evacuated billet was heated to a temperature of 1800 F. The hot billet was extruded (in a press using a heated extrusion die) to a diameter of 0.266 inch whereby an extrusion ratio of 16 times in area was obtained. The press was operated with a 70 ton upset force and a 65 ton running force. The extruded billet was then cold drawn to approximately a 0.172 inch outside diameter. The resultant billet was then cut into suitable short lengths of approximately 3 inches and the can, tubes, and wires were chemically dissolved leaving tubular filaments. These filaments had an average transverse dimension (i.e. between vertices) of approximately 0.015 inch. In addition, a long portion of the cold drawn billet was treated to dissolve the mild steel can and the yMonel tubing resulting in bimetallic filaments having the same size as the tubular filaments. These bimetallic filaments comprised a Type 304 stainless steel exterior and a copper core with a small or trace amount of solid state diffusion occuring at the copper-stainless steel interface. The copper core diameter was approximately 0.008 inch.

EXAMPLE 9 Monel 400 Wires having an outside diameter of 0.150 inch were inserted into Type 304 stainless steel sheaths having an outside diameter of 0.228 inch and an inside diameter of 0.152 inch which were in turn inserted into a Monel tube having a 0.250 inch outside diameter and a 0.230 inch inside diameter. These Wire-sheath-ttlbe composites were straightened and cut to 9 inches in length. An AISI 1010 (mild) steel can having a 2.950 inches outside diameter, 2.840 inches inside diameter and a 91/2 inch length was fitted With a 1A inch thick lower end-cap inserted into the bore of the can and Welded thereto. A truste-conical nose plug having a 2.750 inch diameter, a 90 included angle nose with an approximately 5A; inch circular fiat portion thereon was welded to the lofwer end of the can adjacent to the previously welded fiush end cap. Ninety-one composites were packed into the can in hexagonal array with mild steel metal shims placed between the hexagonal faces of the packed rods and theinternal walls of the can in order to decrease the amount of open space therein. An upper end cap 1A; inch thick with a 1A inch diameter evacuation tube thereon was fitted into the packed can and welded thereto. The billet was then evacuated to less than 10-5 torr at 800 F. and then sealed off. The billet was then heated to 1800 F. and extruded (in a press using a heated extrusion die) to a diameter of 0.625 inch. The billet was extruded with an upset force of 650 tons. The 0.625 inch billet was then cold drawn through successive dies with intermediate anneals to a composite diameter of 0.057 inch. The composite was cut in half with the first portion thereof being subsequently drawn to a 93.8% area reduction wherein the composite had a 0.0143 inch diameter. The can and Monel tubes were then removed by chemical dissolution in nitric acid. The resultant bimetallic, stainless steel- Monel iilament had an overall outside diameter of approximately 0.001 inch and a core diameter of approximately 0.00066 inch. During the processing of this bimetallic filament solid state diffusion took place between the Monel core and the stainless steel sheaths at the interface thereof. The second portion of the billet was further cold reduced in diameter with intermediate anneals at 1800 F. to a composite diameter of 0.028 inch and given a final anneal at 1800 F. The annealed composite was then drawn down to 0.007 inch diameter having a cold work reduction of area of 93.8%. The can and tubing material were then removed by chemical dissolution in nitric acid. The final diameter of the resultant bimetallic filaments (stainless steel-Monel) was found to be approximately 0.0005 inch wherein the Monel core had a diameter of approximately 0.00033 inch. It was found that a small or trace degree of solid state diffusion took place at the interface between the Monel and stainless steel. It is to be understood that the process for making these bimetallic filaments can be stopped after the extrusion step or any of the cold drawing steps with the bimetallic filaments released from the can and tubing (matrix) in order to obtain any desired bimetallic filament diameter.

By this method it has been found that it is possible to form a Ibimetallic filament wherein the core can vary from less than one percent to over 80 percent of the total filament area. The bimetallic filaments can have any desired diameter. By re-bundling the constrictively reduced composites containing the bimetallic filaments, it is provided that such filaments may be formed concomitantly thereby producing from as few as 2 to as many as thousands of bimetallic filaments, as desired. The basic method also provides means for making bimetallic filaments of any desired size having an outside diameter from l5 mils to less than one micron. Illustratively, many bimetallic filament material combinations can be provided, such as: stainless steel over low alloy steel or mild steel, e.g. 1010, niobium over Monel, tantalum over Monel, nickel base super alloy over low alloy steel or mild steel, e.g. 1010, copper over aluminum, mild steel, e.g. 1010 over aluminum, aluminum over mild steel, e.g. 1010, gold over copper, platinum over copper, nickel over mild steel, e.g. 1010, copper over Molypermalloy, titanium over mild steel, c g. 1010, Monel overy beryllium, aluminum over magnesium, Hastalloy X over molybdenum, and the like. It is also contemplated that the core material does not need to be completely surrounded by the sheath (exterior) thus forming a bimetallic filament wherein a first portion material is adjacent to a second portion wherein the geometric cross section of the bimetallic filament can be substantially semi-circular shapes, rectangular shapes, circular segmented shapes and the like. Another embodiment of this invention is the ability to form filaments with either high or low peripheral surface area to volume ratios. When nickel filaments are made in an AISI 1010 (mild) steel matrix as shown in FIGS. 41 and 42 wherein the effective diameter of the filaments are approximately 0.0016 inch the cross sectional configurations of the filaments are extremely rough and the cross sectional area of the filament can be best described as having a highly reentrant surface. It was found that the coefficient of variation of these filaments was less than 25%. The peripheral surface area to the volume of the filament was found to be enhanced by 31.6% over the surface area to volume of a circular sectioned element having the same crosssection area. This highly re-entrant filament provides certain desirable characteristics when desired, such as extreme roughness as well as the high ratio of peripheral surface area to volume. By changing the matrix material for the nickel filaments from mild steel to Aluminum 1100 as is shown in FIG. 43 and FIG. 44, a substantially smooth nickel filament was formed with a much lower peripheral surface area to volume ratio wherein the enhancement was only 1.7% over the surface area to volume of a circular sectioned element having the same cross-section area. In addition, it has been found that for the same filament material it is possible to vary the surface to volume ratio by using different matrix materials. The nickel filaments of Examples 6 and 7 are illustrative of the ability to preselect the matrix to provide different surface to volume ratios for the same filament material, having suitably adjusted the processing. Thus, it has been found that the enhancement of the surface to volume ratio of the filament may be preselected.

In addition the use of the same matrix material for different filament materials also causes either smooth or rough peripheral .surfaces of the filaments, `as desired. When a mild steel matrix was used to form filaments of Ti-13Vl 1Cr-3Al (titanium alloy) a rough peripheral surface on the filaments was produced which may be seen in FIGS. 39 and 40. This material corresponds to the filaments described in Example 5 above. When an AISI 1010 (mild) steel matrix was used with Ti6Al4V to form titanium alloy filaments as described in Example 4 above, the peripheral surface of the laments also exhibited reentrant features. However, the enhancement in surface to volume ratio is 70% greater for the filaments in Example 5 compared to those in Example 4. These examples exhibit that it is possible to preselect the filament material for a given matrix material in order to obtain various peripheral surface to volume ratios. It is therefore contemplated that this disclosure provides for preselecting the filament and matrix materials to produce a pre-selected cross-sectional configuration of the filaments wherein either a highly reentrant cross-sectional configuration is formed, or a substantially smooth and circular cross-sectional filament is formed, as desired. It has been found that the metallurgical characteristics of the filament material and the matrix material can be varied to preselect the peripheral surface. It has been found that when using a high ratio of relatively soft matrix material such as copper to a hard filament material such as type 304 stainless steel that extensive hot and/or cold reduction causes the formation of discontinuous filaments or cyclic filaments. This discontinuity or cyclic effect is shown in FIGS. 27 and 28 wherein a section of the hard filament 112 has elongated crest portions 211 and elongated valley or root portions 212 while surrounded by the soft matrix material 115 and wherein the hard filaments 112A resembles elongated droplets surrounded by the soft matrix 115A.

It may be readily observed that if a series of cross-sections were taken along the length of the filament 112 or the droplets 112A perpendicular to the axis 110 or 110A and the areas thereof obtained, that the coefficient of variation of these areas would exceed 25% for both examples. Obviously, filaments 112 and droplets 112A would be unsatisfactory as continuous filaments or even staple filament. However, by the use of proper preselected ratios of soft or hard matrix material to soft or hard filament material it has been found that the coefficient of variation can be controlled to a level of less than 25 thereby providing a useable cyclic filament. In order to keep nonuniformity of the cross-sectional area of the filament within the self-imposed coefficient of variation of 25 several factors are predominant in controlling the coefficient of variation; (1) the volume ratio of matrix material to the filament material, (2) the amount and temperature of the hot working and/ or the amount of cold working between intermediate anneals, (3) the choice of matrix material for a preselected filament material and (4) the distribution of the filament material in the matrix. There are basically two classes of non-uniformity in fine filaments produced by multiple end reduction; background non-uniformity and gross non-uniformity. Background non-unformity can be characterized and defined as resulting from (l) the grain structure of the matrix and filament materials which because anistrophy (where the physical properties of the metals are not the same in all crystallograpln'c directions) of the metals will produce irregularities on the filament surface during the reduction processes and (2) from the presence of non-ductile inclusions in the filament and/or matrix materials which are not infiuenced by the reduction processes.

Gross non-uniformity is the region where the coefficient of variation has exceeded the background non-uniformity at higher levels of deformation. The magnitude of the gross non-uniformity depends upon (l) the degree of deformation performed on the specic composite between successive recrystallization heat treatments, (2) the magnitude of difference in mechanical flow properties of the various matrix and filament materials in the composite,

l 17 (3) the volume ratio of the matrix and filament materials, and (4) the distribution of the filament material within the matrix. FIG. 29 graphically depicts background non-uniformity and gross non-uniformity when plotted on a chart wherein the coefficient of variation is the ordinate and the degree of deformation is the abscissa. It may be seen from FIG. 29` that there is a level of deformation at which the filament cross-section area non-uniformityr begins to show a rapid increase. This level of deformation is defined as the critical deformation (Wc). Once the filament and matrix. materials have been selected, the coefficient of variation does not change substantially with respect to the amount of deformation as long as there is no intrusion into the gross non-uniformity region. However, the magnitude of the background non-uniformity is cumulative and becomes progressively greater as the filaments become progressively smaller. The graph of FIG. 29 indicates that the gross non-uniformity is a function of the degree of deformation and the process steps may be chosen so as to avoid this region if desired. 'Ihe three graphs of FIG. 30 show the curves for type 304 stainless steel filaments drawn in a copper matrix (A); type 304 stainless steel filaments drawn in a mild steel matrix B); and, type 304 stainless steel filaments drawn in a Monel Matrix (C). The ratio of matrix to filament materials for all three curves is 2:3. These curves appear on the three graphs wherein the ordinate is the coefficient of variation and abscissa. is the degree of deformation. This graph indicates that for type 304 stainless steel filaments more recrystallization heat treatments are required during the constricting operations when a copper matrix is used in contrast to a Monel matrix in order to obtain equivalent uniformity. FIG. 31 is a graph wherein the ordinate is the coefficient of variation and the abscissa is the degree of deformation. Four curves are shown on this graph with curve A having a volume ratio of matrix to filament material greater than curve B; curve B greater than curve C; and curve C greater than curve D. This graph indicates that by the use of lower matrix to filament volume ratios it is possible to achieve higher degree of deformation without intermediate recrystallization steps in order to produce filaments having a cross-section wherein the coefficient of variation is within 25 Therefore, it has been found that by lowering the volume ratio of matrix to filament material it is possible to achieve a more economic manufacture of filament material because it is possible to put a much higher degree of deformation into the filaments between annealing operations without going into the gross non-uniformity region. Also, since the matrix is a sacrificial material, its use in reduced quantities constitutes an economic advantage. FIG. 32 is a graph wherein the volume ratio of matrix to total composite material is the ordinate and the critical degree of deformation (percent cold work) required to stay outside the gross nonuniformity region (We) is the abscissa. On the graph there are three curved area regions, shown as shaded regions; (a) defining a region of copper matrix-stainless steel filament, (b) defining a region of mild steel matrix-stainless steel filaments and (c) defining a region of Monel 400 matrix-stainless steel filaments. The different regions indicate that as the volume ratio of matrix to total composite material is decreased the ability to perform uninterrupted cold work to a greater degree increases. Thus it is evident that by the proper preselection of the matrix material, the filament material and the amount of cold work to be performed on the composite, it is possible to select the most desirable materials for the ultimate filaments to be formed. Obviously, these curves are only illustrative of the particular materials used, however, this same relationship will hold true for all matrix and filament materials. Thus it can be seen that the cross-sectional area uniformity of the filaments will depend on the relationship between: (a) the volume ratio of matrix to filament material, (b) the extent of deformation between the heat treating steps in the process, and (c) the mismatch in the mechanical flow properties of the filament and matrix materials. Further, the appreciation of this interdependency of the matrix and filament materials allows for a process or processes to be provided which results in a filament cross-section area non-uniformity of less than 25% coefficient of variation. In addition, when a filament-matrix combination of starting materials is hot worked it is necessary to select a proper volume ratio of matrix to filament material so that the mismatch of mechanical flow properties of the materials at elevated temperatures is consistent with the desired final filament uniformity. When using composites of equal volume ratio of Monel matrix and stainless steel filament materials and mild steel matrix and stainless steel filament materials, and when the same hot extrusion temperature and reduction ratios are used, it has been found that the filament cross-section area non-uniformity is of the order of l5% in terms of the coefficient of variation. However, for the same volume ratio of copper matrix to stainless steel filament material and under similar hot working conditions the filaments produced were discontinuous and the coefficient of variation approached which indicates that copper, having a greater mismatch in mechanical ow properties with respect to stainless steel at the hot working temperature, would not perform under the conditions that Monel and mild steel would operate. By the interplay of the parameters of the degree of hot and/or cold wor-king, the volume ratio of matrix to filament material :and their mechanical properties, 1a definite criterion has been provided for the economical and practical production of filaments that have a low level of cross-section area non-uniformity shown by a coefficient of variation of less than 25%.

It is to be understood that the preceding parameters are applicable to filaments formed of one metallic material or bimetallic filaments.

Thus, the present invention comprehends an improved method of forming solid, tubular and bimetallic filaments, preselecting the filament and matrix materials to achieve a coefficient of variation of less than 25%, a preselected peripheral geometry, with the added feature that the filaments may be formed in extremely long lengths.

While we have shown and described certain embodiments of our invention, it is to be understood that the invention is capable of many modifications. Changes, therefore, in the construction and arrangement may be made without departing from the spirit, scope and intent of the invention as described in the appended claims.

We claim:

1. A wrought metallic filament having an effective diameter of less than 50 microns formed while surrounded by a sacrificial metal matrix material with the matrix substantially removed therefrom, the filament comprising:

(l) an area coefficient of variation of less than 25% that is defined by a series of cross-sectional areas taken along the length of the filament;

(2) a pre-selected enhanced surface area to volume ratio, compared to the surface area to volume ratio of a right circular cylinder of substantially equal volume and equal cross-sectional area and defined by the ratio of the peripheries of the cross-sectional areas of the filament compared to the cylinder, the preselected surface area of the filament defined by the metallurgical characteristics of the filament material and the matrix material; and

(3) the filament having an unmachined, unburnished,

and substantially fracture free outer surface, the surface of the filament having a trace of the removed surrounding matrix material and having a texture developed by successive hot and cold working thereof while surrounded by the matrix.

2. The filament of 1 wherein the periphery of the filament is re-entrant.

3. The filament of claim 1 wherein the periphery of the filament is substantially smooth.

4. The filament of claim 1 wherein the filament is bimetallic.

5. The filament of claim 1 wherein the filament is continuous.

6. The filament of claim 1 wherein the effective cross section dimension of the filament is cyclic along the length thereof.

7. The filament of claim 1 wherein the filament is tubular.

8. The filament of claim 1 wherein the surface area to volume ratio excess 1.00 by a range of about 1% to over 60%.

9. The filament of claim 1 wherein said metallurgical characteristics of the filament material are similar to those of the matrix material.

10. The filament of claim 1 wherein the metallographic characteristics of the filament material and the matrix material are different.

11. The filament of claim 1 wherein the filament material exhibits a recrystallized structure.

12. The filament of claim 1 wherein the filament material is not a recrystallized structure.

13. The filament of claim 1 wherein the coefficient of variation is controlled by preselecting the ratio of the matrix material to the filament material.

14. The lament of claim 1 wherein the coefficient of variation is maintained substantially within the background non-uniformity region.

15. The filament of claim 1 wherein the coefficient of variation is controlled by the mechanical properties of the matrix and the filament materials.

16. The filament of claim 1 wherein the filament material is distributed within the matrix material in a preselected configuration.

17. The filament of claim 1 wherein the coefficient of variation is controlled by preselecting the ratio of matrix and filament material, maintained substantially within the background non-uniformity region, and the mechanical properties of the matrix and the filament materials.

18. A bi-metallic filament having an effective diameter of less than 5 mils comprising:

a first portion formed of a first metallic material and having a small amount of another material present on a portion of the exterior surface thereof; and

a second portion formed of a second metallic material adjacent to the first portion, the first portion and second portion having an interface therebetween, the filament having a texture developed by successive hot and cold working while surrounded by an after removed matrix material, the matrix material being the other material present on the first portion.

19. The filament of claim 18 wherein said second portion ranges between less than 1% to approximately 80% of said total filament.

20. The filament of claim 18 wherein said second portion has a trace amount of said first portion solid state diffused therein adjacent to said interface.

21. The filament of claim 18 wherein said first portion has a trace amount of said second portion solid state diffused therein adjacent to said interface.

22. The filament of claim 18 wherein there is solid state atomic interdiffusion between said first and second portions.

23. The filament of claim 18 wherein there is an absence of diffusion between the atoms of said first portion and said second portion.

24. The filament of claim 18 wherein said filament has a diameter of under l5 mils.

25. The filament of claim 18 wherein said second portion comprises a core substantially surrounded by said first portion.

26. The filament of claim 25 wherein said filament has an axis and a cross-sectional area substantially perpendicular to said axis, said area having a coefficient of variation of less than 25%.

27. The filament of claim 18 wherein said first portion material is an alloy.

28. The filament of claim 18 wherein said second portion material is an alloy.

References Cited UNITED STATES PATENTS 3,379,000 4/1968 Webber et al 29-193 X 3,098,723 7/'1963 Micks 29-`183.5 2,842,440 7/1958 Nachtman et al. 29-192 RX 3,113,376 12/1963 Pfiumm et al 29-l83.5 X 3,087,233 4/1963 Turnbull 29-193 UX 2,825,108 3/1958 Pond 29-193 ALLEN B. CURTIS, Primary Examiner

Images