US3492172A - Method for producing titanium strip - Google Patents

Description

United States Patent 3,420,172 METHOD FOR PRODUCING TITANIUM STRIP Adrian Burt Sauvageot, Toronto, and James M. Partridge,

Steuhenville, Ohio, assignors to Titanium Metals Corporation of America, New York, N.Y. No Drawing. Filed Nov. 9, 1966, Ser. No. 592,995 Int. Cl. C22f 1/18, 1/16 US. Cl. 148-115 25 Claims ABSTRACT OF THE DKSCLOSURE Method for producing strip having a substantially recrystallized microstructure from commercially pure titanium, alpha, stabilized alpha type titanium base alloys, and alpha stabilized alphabeta type titanium base alloys comprising unidirectionally hot rolling a metal body from a temperature requiring a substantial amount of reduction in the alpha-beta field, heating said rolled metal at a temperature above the beta transus of said metal to transform the crystal structure of said metal to the beta phase and rapidly cooling said metal to a temperature below the beta phase to produce an acicular type microstructure in the metal followed by rolling and annealing said metal at temperatures below the beta transus.

This invention relates to a method for the manufacture of titanium and titanium alloy strip which is particularly characterized by freedom from the tendency to either rib or ridge, good ductility and minimum anisotropy of mechanical properties. More particularly, the method of our invention permits the production of equiaxed substantially completely recrystallized elongated strip which may be rolled to finish gauge or fabricated. Our method may be successfully applied to commercially pure titanium and to alpha and alpha-beta type titanium alloys containing alpha stabilizers.

Ribbing and ridging are indicated by longitudinal striations which occur in titanium metal which has been unidirectionally rolled at temperatures high in the alphabeta field and which are oriented in the direction of rolling, Ribbing refers to the striations which develop during subsequent rolling of the strip to finish gauge, and ridging refers to the striations which develop during testing of rolled strip or during fabrication of products therefrom. Both of these effects are believed to be caused by crystallographic orientations imparted to the metal during the unidirectional rolling of bars and billets into a hot band. Anisotropy is also created in the metal by unidirectional rolling and is evidenced by a difference between the mechanical properties of the strip when measured parallel to the direction of rolling and when measured transversely of the direction of rolling.

Since ribbing and ridging in titanium and titanium base alloys are apparently created by unidirectional rolling of the metal in or through the two-phase alpha-beta field, they could be avoided by rolling at temperatures sufiiciently low to avoid deformation in the alpha-beta field. Additionally, ribbing, ridging and anisotropy of mechanical properties may be eliminated by cross rolling the metal body at right angles to the original direction of rolling to an extent to obtain approximately the same degree of reduction as obtained during the original rolling. However, low temperature rolling is not practical with titanium alloys since suflicient reduction cannot be obtained and cross rolling is not practical for the commercial production of either titanium or titanium alloy strip.

Initial rolling at temperatures in the beta field or relatively high in the alpha-beta field is essential for the production of a hot band of alpha and alpha-beta type titanium alloys for subsequent rolling to finish strip since reduction of a billet or bar to a hot band involves a substantial decrease in size. This reduction must be initiated at elevated temperatures since the alloys have a low specific heat and a consequent rapid temperature drop, and it is important that the metal be sufficiently plastic to per mit rapid and easy deformation. Although an elevated temperature is essential for easy deformation, the temperature should not be so high as to cause excessive oxidation and scaling of the metal surface during deformation. Practical temperatures from the standpoint of ease of deformation and avoidance of excessive contamination of the metal surface are high in the alphabeta field or low in the beta field; and substantial reduction at other temperatures is not practical.

Although cross rolling will eliminate ribbing and ridging and will substantially decrease directionality, cross rolling is limited to sheet products having a length not much greater than their width since the dimensions of the sheet must be such that it can pass crosswise through the rolling mill. Since cross rolling is of necessity limited to sheets having a length no greater than the roll width, it obviously cannot be applied to elongated hot bands having lengths of a hundred feet or more which are to be rolled into even longer strips.

Heat treatment of commercially pure titanium and alpha and alpha-beta type titanium alloys by annealing at a temperature near the beta transus of the metal prior to reduction to eliminate ribbing and ridging and to decrease anisotropy is disclosed in United States Patent No. 3,169,085. While the method disclosed in this patent will reduce the tendency toward ribbing and ridging and will reduce anisotropy, the resulting microstructure of commerically pure titanium and alpha and alpha-beta type alloys processed in accordance with the method is such that ductility is extremely poor; and, therefore, the metal cannot be further reduced or fabricated without catastrophic internal shear cracking of the metal. The extent of the shear cracking has been found to be dependent upon the annealing temperature and the rate of cooling from the annealing temperature. In order to produce a strip which will not have a tendency to rib or to ridge and which can be successfully rolled and fabricated, we have found it essential that the strip be annealed at a temperature above the beta transus of the metal and that the strip be rapidly cooled to a temperature below the beta transus. Treatment in this manner creates a substantially completely acicular type microstructure in the metal as opposed to the lamellar or to the Widmanstatten type structure which is created when the metal is cooled at a slower rate. Rapid cooling in accordance with our invention may be obtained by immersion, spraying or extremely fast air cooling; and the method of cooling will depend upon the specific material and the thickness of the strip being processed. Annealing at a temperature above the beta transus and rapid cooling to a temperature below the beta transus are termed a beta quench hereinafter.

Summarized briefly, our invention provides a method for producing strip of commercially pure titanium and alpha and alpha-beta type titanium alloys containing alpha stabilizers which is characterized by freedom from ribbing and ridging, minimum anisotropy of mechanical properties and good ductility. Strip produced according to our invention can be cold rolled to finish gauge and fabricated without failure due to internal shear cracking. Basically, our method comprises unidirectionally hot rolling the metal from a temperature in its beta field or high in its alpha-beta field; heating the rolled metal to a temperature above its beta transus and rapidly cooling to a temperature below the beta transus to create a substantially completely acicular microstructure and subsequently deforming and annealing the metal to substantially completely recrystallize the microstructure. The beta quench may be performed on the hot band immediately after the unidirectional rolling, or it may be performed later in the process after the hot band has been reduced. After the beta quench, it is necessary that the metal be subjected to deformation and annealing in a manner and to an extent to effect substantially complete recrystallization of the microstructure in the metal. It is essential that the microstructure be substantially completely recrystallized for the metal to have sufiicient ductility to permit subsequent cold rolling and fabrication Without failing due to internal shear cracking. The term cold rolling as used herein encompasses rolling between about room temperature and about 300 F.

Metal which may be advantageously treated by the method of our invention is selected from the group consisting of commercially pure titanium, alpha type titanium alloys and alpha-beta type titanium alloys. The alloys must be alpha stabilized since beta stabilized alpha-beta alloys are susceptible to the formation of the brittle omega phase during rapid cooling from the beta field. Commercially pure titanium is available in various grades depending upon hardness, and the compositions of the grades differ principally in the content of the interstitials, oxygen, nitrogen and carbon. Generally speaking, commercially pure titanium has alpha type microstructure at room temperature and will convert to the beta crystal form at temperatures above its beta transus. Alpha type titanium alloys have an alpha microstructure at room temperature and contain alpha stabilizers of which aluminum, tin and columbium are examples. An alloy containing 5% aluminum, 2.5% tin, balance titanium, is typical of the alpha type alloys. The alpha-beta type titanium alloys have a mixture of alpha type microstructure and beta type microstructure at room temperature and are exemplified by an alloy containing 6% aluminum, 4% vanadium, balance titanium. A group of alloys generally designated alpha lean beta type contain a substantial amount of alpha stabilizing agent and a minor proportion of beta stabilizing agent and are subject to treatment by our method since the alpha stabilizer predominate. An alloy containing 8% aluminum, 1% vanadium, 1% molybdenum, balance titanium is a typical alpha lean beta type.

The initial hot rolling step to reduce the bar, billet or other shape to hot band is initiated at a temperature in the beta field or high in the alpha-beta field of the metal being processed. Any type of rolling mill may be employed, and the dimensions of the starting shape are not material so long as the shape will fit into the mill. Conveniently, the initial rolling is carried out by heating the metal body to the starting temperature heat or above the beta transus and then passing it through the rolling mill, rolling being continued as the temperature of the metal drops. A large portion of the reduction and certainly the later part thereof is accomplished in the alpha-beta field of the metal being rolled due to the low specific heat of titanium and titanium base alloys.

After rolling, the metal is heated to a temperature above its beta transus. This is accomplished in any suitable manner by conventional apparatus and may be carried out by placing a coil in a furnace although it is preferable to continuously heat successive portions of a hot band or strip. Maintenance of titanium and titanium base alloys at a temperature above the beta transus for an extended period of time is undesirable because it coarsens the grain structure; and, if the heating is carried out without the protection of an inert atmosphere, surface oxidation and scaling occur. It is, however, essential that the metal be held above the beta transus for a suflicient time to in- I sure complete transformation of the entire metal body to beta structure; and, therefore, the duration of the anneal will depend upon the temperature of the heating furnace and the size and shape of the body being heated. For example, in the case of a closely wound coil, sufiicient time must be allowed for the innermost convolutions to be raised to temperature. When a continuous anneal is utilized to raise successive portions of hot band or strip to temperature the time required at temperature will be very short since the hot band or strip will be rapidly heated throughout. The precise temperature employed for the anneal is not critical providing it is above the beta transus of the metal being treated. A preferred maximum is about F. above the beta transus since this provides a margin of safety insuring complete heating above the beta transus and is not so high as to coarsen the grain size and produce excessive surface oxidation and scaling.

After the metal has been completely transformed to beta structure, it is rapidly cooled to produce the required substantially completely acicular microstructure for subsequent recrystallization. Rapid cooling of the metal to a temperature below the beta transus is absolutely essential for the success of our process as it is not possible to obtain recrystallization by subsequent deformation and annealing unless the metal is beta quenched to create the acicular microstructure necessary for recrystallization. As will be shown hereinafter, the ductility of unrecrystallized strip as determined by bend tests is low. Cooling may be accomplished by either immersion or spraying, and any coolant which will rapidly decrease the temperature of the metal may be used. Additionally, air cooling is suflicient to rapidly lower the temperature of very thin strip. As stated heretofore, the beta quench may be performed immediately after hot rolling, or it may be carried out as an intermediate step between reductions.

The precise temperature at which the microstructure will become substantiallycompletely acicular in the rapidly cooled or beta quenched metal will vary depending largely on the particular composition of the metal, as will be appreciated by those skilled in the art. However, in all instances, this temperature will be lower than the beta transus of the metal. For practical application of the method, it is preferable to cool the heated metal to room temperature since the formation of the necessary acicular microstructure is insured and a minimum of control is required. However, it is not essential that the metal be cooled to room temperature; and if control conditions permit, it may be cooled to a temperature above room temperature. In addition to cooling the metal to a temperature below 1ts beta transus, it is necessary that the rate of cooling be rapid to produce acicular microstructure in the metal. ThlS microstructure will be readily recognized by those skilled in the art by the appearance of a section of the metal cut, polished and etched by conventional techniques and examined under a microscope. The microstructure consists principally of needle-like grains more or less randomly oriented. The aciculas may have more than three habit planes, that is, they may be oriented or grouped in more than three directions. Such a microstructure is readily distinguishable from that obtained by slower cooling from the beta field which will consist principally of platelet alpha titanium in basketweave or Widmanstatten array. Usually the platelets will lie in no more than three directions in blocks smaller than the original beta grains. The acicular microstructure, formed by the rapid cooling from the beta field according to our invention, is readily broken up and recrystallized by subsequent deformation and annealing; and this substantially completely recrystallized microstructure provides good ductility in the final strip.

After the beta quench, the metal is treated by a sequence of rolling and annealing steps designed to create recrystallized microstructure in the metal. The deformation of the metal during rolling imparts strain to the metal, and subsequent annealing tends to relieve the strain and recrystallizes the grain structure. Partial stress relief is obtained by intermediate anneals, and the final anneal results in complete stress relief and substantially complete recrystallization. As would be expected, the precise sequence of rolling and annealing applied to the beta quenched metal and the temperature of these steps will vary with the material and thickness being treated, the desired finish gauge of the strip and the equipment available. For certain materials a stress relief anneal is beneficial prior to any deformation immediately after the beta quench. Generally speaking, smaller reductions and longer intermediate anneals must be used with lower rolling temperatures to prevent internal shear cracking while larger reductions are required at higher rolling temperatures to impart the required strain to the metal for subsequent recrystallization.

It is possible to roll some titanium strip products, such as commercially pure titanium and an alloy containing aluminum, 2.5% tin, balance titanium, at temperatures which permit recrystallization simultaneous with rolling. The rolling temperature for simultaneous recrystallization will be between about 1000 F. and 1200 F. for beta quenched commercial pure titanium and between about 1200 F. and 1400 F. for the alloy containing 5% aluminum, 2.5% tin, balance titanium. When recrystallization occurs during rolling, the final anneal will result in stress relief and grain growth and the resulting microstructure, which is substantially completely recrystallized, is large grained and less equiaxed than metal which is recrystallized by a final anneal only. Due to the grain size and orientation in metal which has partially recrystallized during rolling, the metal is less susceptible to further cold rolling than metal which is recrystallized by a final anneal. However, if no further reduction of the strip is required, rolling at temperatures which permit recrystallization is satisfactory as the strip produced has no tendency to rib or ridge, has low anisotropy and high ductility. For this reason, rolling temperatures for commercially pure titanium and the titanium alloy containing 5% aluminum and 2.5% tin should be based in part on the work to be done on the material after the rolling recrystallization cycle.

It is necessary that the rolling and annealing temperatures be kept below the beta transus of the metal since reheating to above the beta transus will require subsequent rapid cooling followed by rolling annealing to obtain the recrystallized structure. It has been found that temperatures ranging from room temperature to about 1600 F. are acceptable for deforming the beta quenched metal de pending upon the metal and amount of reduction taken in each rolling cycle. Annealing temperatures from about 1400 F. to about 1700 F. are acceptable, and the prethickness. The temperature decreased during rolling to a finish temperature of 1650 F., and a majority of the rolling was carried out in the alpha-beta field. A test panel was cut from the hot band and heated at a temperature of 1950 F. for three minutes. The panel was then quenched by rapid immersion in a cold water bath. The beta quench produced an acicular type microstructure in the metal.

The beta quenched panel was then subjected to five 15% reduction rolling cycles at room temperature. The panel was line annealed for five minutes at 1 600 F. followed by air cooling after each rolling cycle. The final anneal effected substantially complete recrystallization, and the finished cold rolled strip was free from internal cracking and showed no ridging when a 2-inch by 10-inch test piece was cold stretched to an elongation of about 10%. No ribbing was discernible either in the final product or at any stage after the beta quench. A specimen of the finished cold rolled strip was sectioned and polished and its microstructure found to be fine-grained and substantially completely recrystallized.

Another bar of Ti-5Al-2.5Sn alloy was unidirectionally hot rolled to 0.130 inch thickness from a temperature in the beta field with a majority of the reduction taking place in the alpha-beta field. A panel from the hot band was then heated at 1950 F. for five minutes and water quenched. The beta quenched panel was sandblasted and pickled and rolled at 1000 F. to effect a reduction to 0.051 inch. The test piece was again sandblasted and pickled and was then annealed for ten minutes at 1650 F. and air cooled. No ribbing occurred during the rolling at 1000 F.; and, as indicated in Table A, the mechanical properties of the resulting strip were good. The microstructure was found to be completely recrystallized after the final anneal.

Three other panels from this hot band were processed in the same way and then subjected to one, two and three 20% reduction cold rolling cycles. Each cold rolling cycle was followed by an anneal at 1600 F. for a period of three to five minutes and air cooling. No ridging occurred when specimens were stretched to an elongation of at least 10%. The material did not rib during cold rolling, and good mechanical properties and low anisotropy were obtained as indicated in Table A. All bends in Table A and in the other tables appearing hereinafter were rated at 20X magnification with no observable rupture.

TABLE A Test E, 20X gauge Ftu, Fty, Percent p.s.i. MBR Cold rolling inch Dir. K s.i. K s.i. E? X 10 R/T None 051 L 139. 3 125. 0 20. 0 16.3 3. 1 T 134. 3 128. 2 21. 0 17. 6 3. 1 One 20% cycle .040 L 134.1 118. 4 20. 0 16.5 3. 8 T 134. 4 124. 3 20. 2 17.2 3. 1 Two 20% cycles.... 030 L 134. 8 122.1 17. 0 16. 3 3. 0 T 133. 3 123. 3 19. 5 17. 7 2'. 6 Three 20% cycles .024 L 137. 7 125. 4 17.0 16. 9 3. 6 T 136. 2 123. 7 19.2 17. 3 2. 7

*One inch gauge.

cise temperature will be selected in accordance with the metal and the type of anneal utilized. The duration of the anneal will also vary with the metal being treated and the annealing temperature. Line anneals ranging from five to ten minutes have been used, and box anneals varying from sixteen to forty-eight hours have also been used.

The following non-limiting examples illustrate specific embodiments of our process as applied to commercially pure titanium and to the representative titanium alloys, Ti-5Al-2.5Sn; Ti-6Al-4V and Ti-8Al-1Mo-1V.

EXAMPLE I A sheet bar of an alloy containing 5% aluminum, 2.5% tin, balance titanium having a beta transus of 1900 F. :t25 was unidirectionally hot rolled from a temperature The maximum acceptable bend radius for Ti-5A1-2.5Sn is 4.0 for .070 inch gauge and below and it is apparent that all specimens met this specification.

An additional bar of Ti-5A1-2.5Sn alloy was unidirectionally hot rolled to 0.149 inch thickness from a temperature in the beta field with the majority of the reduction taking place in the alpha-beta field. Two panels from the hot band were then heated at 1950 F. for five minutes and water quenched. Both beta quenched panels were then rolled to 0.057 inch, a total reduction of 60%. One panel was rolled at 1400 F. and the other at 1600 F. No ribbing occurred during rolling and the microstructure of each panel showed complete recrystallization during rolling. Mechanical testing after line annealing for ten minutes at 1650 F. and air cooling showed low anisotropy, high tensile and bend ductility as indicated in of about 2100 F. to produce a hot band of 0.135 inch Table B. No ridging occurred after 10% elongation.

7 8 TABLE B A third bar of this alloy was unidirectionally hot rolled 20X from a temperature in the beta field to a thickness of 0.175 Rolling Ftu Fty Percent E, p.s.l. MBR inch. Two panels were cut from the hot band and heated temp-I 131* X 108 R/T for five minutes at 1950 F. followed by water quenching. 1,400 L 124.3 111.9 20.2 16.6 3.5 The beta quenched panels were sandblasted and pickled 1 600 133% igg to remove contamination. The panels were then rolled T 121.5 114.5 19.2 17.2 3.5 at 1400 F. and 1600 F. to 0.065 inch, a 60% reduction. inch gauge Each panel was annealed at 1450 F. for forty-eight hours i followed by vacuum cooling and pickling. The final micro- EXAMPLE H structure of the panels was essentially completely recry- A sheet bar of an alloy containing 6% aluminum, 4% stallized. Mechanical tests showed low anisotropy and vanadium, balance titanium having a beta transus of good tensile and bend ductility as indicated in Table D. 1820 F. i25 was unidirectionally hot rolled from a No ridging occurred on specimens pulled to 10% elontemperature of 1900 F. to produce a hot band of 0.140 gation. inch thickness. The rolling finished at 1500 F.; and, TABLE D therefore, a major portion of the reduction was accom- 0X plished in the alpha-beta field. A panel from the hot band Rolling Fm Fty Percent E, MBR was heated to a temperature of 1925 F. for five minutes temp, F. D11. Ks.i. Ks.i. El X 10 and then rapidly quenched by immersion in water. The 1,400 130,3 114 4 ltaletanigzpch produced an acicular type microstructure in 1,6O0 g 35 9 55 g g igg 5:? e

' EXAMPLE III Roll-157 reduction Anneal7 -minutes at 1500 F. AC A sheet bar of alloy containing 8% aluminum, 1% Ro11 15% reduction molybdenum, 1% vanadium, balance tltanlurn havlng a a beta transus of 1900 F. i25 was unidirectionally hot Anneal 7 minutes at 1500 F. AC

rolled from a temperature of 2100 F. to produce a hot Roll-45% reduction h fi h d hours at 14505 Rvcl band of 0.120 inch thickness. T e r0 mg n1s e at 1650 F., and the ma or portlon thereof was carried out Roll 20% reduction Annea1 6 minutes at 16005 R Ac 1n the alpha-beta field as the bar cooled during rolling. A lvacuum c001 7 slower than air c001 parsigl flrq om the hot Sand) was heated for five mmlpteg at 19 and queue ed y immersion in water. T e eta The cold rolled P5111e1 W21S free from internal Cracking quench produced an acicular type microstructure in the and showed no ridging when a 3-inch by 10-inch test piece metaL Was stfotohofl to an elongation of about ribbing The beta quenched panel was then subjected to cold was discernible at any stage after the beta quench. A rolling and annealing according to the following sequence: specimen of the finished cold rolled test piece was sec- R n 107 d tioned and polished and its microstructure found to be i He 2 F Vc 1 equiaxed fine-grained and substantially completely re- 40 nnea ours? Roll15% reduction crystallized. A 1 7 t t F AC A second bar of this alloy was unidirectionally hot g i g a rolled from a temperature in the beta field to thickness A t F Ac of 0.135 inch. A plurality of panels were cut from the hot g g 233;? a band, and each panel was heated at 1950 F. for five gg i gi F Vcl minutes after which it was water quenched. Following Rol1 15(7 reducfo the beta quench the panels were sandblasted and pickled Ann minute; F AC in order to clean the surface of the metal. The panels were then rolled at 1000 F. to efiect reductions of 15%, 20%, 1 Vacuum Slower than C001- 25% and 30% after which each panel was further reduced The fi i h d o a panel was free from internal cracklng and 40% at 1000 F. A five mlnute anneal at 1650 F. and showed no id i g when a 2-inch by 10-inch plece Was CQOImg by sandblasting n o o Was cold stretched to an elongation of about 6%. No ribbing 323.? $533??? l fiiifiity iiii ifii rs iiftl ii fiiii loil' If, dimmible final piodlflct w at uy stag. (1. .5. e processing 0 e pane a ter t e eta quenc 25,531 fizz/ gli i ggi-rgg 2 1:22:3 g gg specimeln 1trfdthe finished cold rolled panel was sectioned and p0 is e and its microstructure found to be finer gno i k In 2 ih b t g obszrvod grained and substantially completely recrystallized. mg q 0 o o a quono an 110 Another bar of Ti-8Al-1Mo-1V alloy was unidirectiong g gzggg 3 :5 w g eq-ah 511 hct T3116: from a itlemperalture in the beta field to o a 1115 r 1S a ct an o 0.140 inc Pane s were cut from the hot for 0.070 inch gauge and below and 5.0 for strip thicker band and annealed at 1950 F. for five minutes followed than 0.070 inch. All speciments met these specifications. by quenching in water. Each beta quenched panel was TABLE 0 Test E, 20 gauge Ftu, Fty, Percent p.s.i. MBR 1,000 F. cycles inch Dir. Ks.i. Ks.i. E1 X R/T 15%,-40% .075 L 132.5 121.8 18.7 15.7 3.3 T 133.5 123.9 18.5 17.0 2.5 20%-40% .071 L 133.6 120.4 19.0 10.0 3.5 T 132.6 125.1 19. 5 18. 1 3. 5 25%-40% .055 L 134.5 121.2 18.5 15.8 3.8 T 133.5 125.4 19.0 17.7 3.0 30%-40% .060 L 132.4 110.0 19.3 15.9 3.9 T 130. 7 121. 5 18. 3 17. 3. 5

"One inch gauge.

9 10 then sandblasted and pickled and rolled at 1000 F. to TABLE G effect a 30% reduction, annealed at 1650 F. for five minutes followed by air cooling rolled at 1000 F. to R m 20X F effect a 40% reduction, sandblasted and plckled and an teinpf F. Dir. Ksii K 53 1 Qla fg/ nealed for forty-eight hours at 1450 F. followed by 5 1,400 L 1381 12% m7 16.9 as vacuum cooling. The microstructure of the panels was T 132.5 127.5 20.5 13.1 3.7 fine grained and substantially completely recrystallized L600 "g 13%? i 7.9 3.7 after the forty-eight hour anneal. Individual panels were then cold rolled through one, two, three and four 20% gaugereduction cycles with a five minute anneal at 1650 F. EXAMPLE 1V followed by an coolmg between each cycle. Subsequent to the last rolling cycle, each panel was annealed at 1450" A Panel of mmer ia ly Pure titanium, Ti-75A, was F. for forty-eight hours followed by vacuum cooling to cut from a unidirectionally hot rolled band having 3. effect a second recrystallization. Each of the panels was thickness of inch containing 0-3O4% Oxygen and successfully cold rolled without internal shear cracking. 15 havlng a beta tfanslls of 1725-1750 The Panel Was A sh wn i T bl E, th t l was du til d i cold rolled to 0.10 inch and was heated at a temperature tropy was low. No ribbing was observed during rolling of 1800 F. for 2.5 minutes. The panel was then quenched and specimens were elongated without ridging. The acy YaPId immersion in a Cold Water bath- The beta quench ceptable bend radius for Ti-8Al-1Mo-1V is 4.0 for 0.070 Produced all aciclllaf yp microstructure in e me alinch gauge and below and all samples tested met this The beta quenched Panel was then Subjected to f0111' specification. cycles of 15% reduction at room temperature usin TABLE E Test E, 20 gauge Ftu, Fty, Percent p.s.i. MBR Cold rolling inch Dir. Ks.i. KS1. El" 10 RI'I None .055 L 138.6 129.5 20.0 16.9 3.7 T 134.5 127.5 22.0 17.9 2.9 One 20% cycle .044 L 137.9 127.4 20.0 16.8 3.9 T 135.0 124.7 19.5 13.0 3.5 Two 20% cycles .033 L 134.6 122.0 19.5 16.7 3.7 T 129.0 120.4 21.0 17.7 3.1 Three 20% cycles .024 L 131.8 122.3 21.8 18.2 N.T. T 125.9 121.5 21.5 13.4 3.2 Four 20% cycles .018 L 131.8 118.5 20.5 15.7 3.9 T 127.3 118.5 23.3 17.2 3.2

One inch gauge. N.T.-Tot tested. N

A pair of Ti-8Al-1Mo-1V alloy panels were treated 1400 F. seven minute intermediate anneals. The final in the same way as the panels in the immediately preanneal effected substantially complete recrystallization, ceding group except that after the first reduction 40 and the finished cold rolled strip was free from internal at 1000 F. and the following anneal, the panels were cracking and showed no n'dging when a 2-inch by 10-inch rerolled at 1000 F. through two and three 30% reductest piece was cold stretched to an elongation of about tion cycles at 1000 F. Each rolling cycle was followed 10%. No ribbing was discernible either in the final by a five minute anneal at 1650 F. and air cooling, and product or at any stage after the beta quench. A specimen the last rolling cycle was followed by sandblasting, pickof the finished cold rolled strip was sectioned and polling and a forty-eight hour recrystallization anneal at 1450 F. and vacuum cooling. These panels also exhibited substantially complete recrystallization and direc tionality was low as shown in Table F. No ribbing occurred during rolling and the panels were elongated without ridging.

ished and its microstructure found to be recrystallized.

Six additional panels from the same 0.150 inch hot band were heated at 1850 F. for five minutes and then rapidly cooled by immersion in water. The beta quenched panels were sandblasted and pickled and rolled a total of at 600 F., 800 F., 1000 F., 1200 F., 1400 F.

*One inch gauge.

and 1600 F. respectively. The panels rolled at 1000 F. and below were completely recrystallized into equiaxed grains by a 1650 F. ten minute anneal followed by air cooling. The tests performed on these panels showed no ridging, low anisotropy and high tensile and bend ductility as shown in Table H. All specimens met the bend specifications of 2.5 for 0.060 inch and below and 3.0 for strip thicker than 0.060 inch. The panels rolled at 1200 F. and above recrystallized simultaneously with rolling. These panels demonstrated good mechanical properties and high ductility without an anneal. A 1650 F. ten minute anneal followed by air cooling further 1 1 increased the ductility by completely stress relieving the hot rolled structure.

12 60% at 1000 F. and annealed for ten minutes at 1650 F. followed by air cooling in an attempt to recrystalllze TABLE H Test 20X gauge, Ftu Fty Percent E, p.s.i. MB R Rolling temp., F. inch Dir. K at K s.i. El" X 10 RT 054 L 97. 1 72. 8 31. 14. 9 2. 1 600 T 96. 2 80. 1 29. 0 17. 3 2. 0 800 059 L 94. 2 66. 33. 0 15. 5 2. 0

'1 96. 1 80. 9 3 055 L 96. 4 73. 0 5. L000 T 93. 0 77. 5 2s. 5 16. a 2. 1 1,200 055 L 96. 0 70. 4 32. 0 l5. 6 2. 1 T 104. 7 82. 0 29. 0 19. 5 2. 1 1,400 054 L 97. 5 71. 4 33. 0 15. 7 2. 1 T 96. 9 78. 8 26. 5 17. 0 2. 1 1,600 055 L 94. 6 64. 5 36. 0 l5. 1 2. 1 T 100. 0 81. 8 29. 0 17. 4 2. 1

*One inch gauge.

The effect of the beta quench on the ribbing and ridging characteristics of titanium alloy is clearly shown by the results of a test in which hot band panels of Ti-5Al-2.5Sn alloy were rolled at 1000 F. to effect a 60% reduction and box annealed at 1450 F. for forty-eight hours followed by vacuum cooling. One panel had been beta quenched prior to rolling and annealing and the second panel was rolled as received. The microstructures of both panels were recrystallized to the same extent. However, when the panel which was not beta quenched was pulled to 10% elongation, ridging occurred while no ridging took place in the beta quenched panel.

Panels of Ti-8Al-lMo-1V alloy and Ti-6Al-4V alloy were also cold rolled from non-beta quenched hot bands, and in each instance ridging occurred when the rolled panels were stretched at less than 10% elongation.

In order to ascertain the necessity of rapid cooling from the beta field to create the acicular type microstructure, comparative tests were conducted. The tests and the results are set forth hereinafter.

Test A The test panels of Ti-8Al-1Mo-1V alloy were taken from a unidirectionally hot rolled hot band at 0.150 inch gauge and were heated at a temperature of 1950 F. for five minutes. One of the panels was water quenched below the beta transus, and the other was air cooled. Metallographic examination showed that the quenched panel had an acicular alpha structure, whereas the air cooled panel had a lamellar alpha structure. Both panels were then rolled at 1000 F. to effect a 30% reduction, annealed at 1650 F. for five minutes followed by air cooling and rolled again at 1000 F. to effect a 40% reduction. Both panels were then box annealed at 1450 F. for forty-eight hours followed by vacuum cooling. Examination of the panels showed that the micro-structure of the air cooled panel was not completely recrystallized and still showed remnants of the beta heated structure while substantially complete recrystallization occurred in the quenched panel. In order to determine the cold rollability of the metal, both panels were cold rolled through three cycles of reduction with a five minute line anneal at 1650 F. with air cooling after each reduction. Internal shear cracking occurred in the air cooled panel during the first cold reduction, and the cracking increased during the second and third reductions. No cracks occurred in the panel subjected to the beta quench even after the third cycle of 15% reduction.

Test B Two panels of Ti-5Al-2.5Sn alloy were taken from a hot band at 0.130 gauge and heated at 1950F. for five minutes. One of the panels was air cooled to below the beta transus and the other was water quenched. Metallographic examination of the panels showed an acicular alpha structure in the quenched panel and a lamellar alpha structure in the air cooled panel. Both panels were then reduced the micro-structure. The microstructure of the air cooled panel was not completely recrystallized, whereas that of the quenched panel was. The panels were then cold rolled through three cycles of 15%, 20% and 25% reductlon, and serious cracking occurred in the slowly cooled panel during the 15% reduction cycle and the cracking increased during the next two cycles. No cracking occurred in the panel subjected to the beta quench.

The results of tests A and B clearly show that cooling to a temperature below the beta transus at a rate sufiiciently rapid to create an acicular alpha structure prevents internal shear cracking and is essential to obtain complete recrystallization.

As Will be appreciated by those skilled in the art, the degree of reduction must be considered in conjunction with the rolling temperature and the desired final gauge in order to produce recrystallized strip having good ductility and minimum anisotropy. For example. Ti-8Al-1Mo- 1V alloy can be recrystallized by two 20% reductions and a 30% reduction at 600 F. with intermediate five minute anueals at 1600 F. followed by a final recrystallization anneal at 1450 F. for forty-eight hours or by a single 60% reduction at 1200 F. followed by a 1450 F. anneal for forty-eight hours.

A temperature which has proved advantageous for rolling Ti-5Al-2.5Sn, Ti-6Al-4V and Ti-8Al-1Mo-1V alloys is 1000 F. The Ti-5Al-2.5Sn can be directly reduced 60% at this temperature without any intermediate stress relief anneal as is shown in Example I. The Ti-6Al-4V and Ti- 8Al-lMo-1V alloys must be reduced in at least two cycles at 1000 F. with an intermediate partial stress relief anneal. The initial reduction for Ti-6Al-4V may vary from about 15 to 35% and for Ti-SAl-lMo-IV from about 20 to 35%. Initial reductions of these two alloys much in excess of 35% at 1000 F. will result in internal shear cracking, and reductions much less than the above indicated minimum will impart insufficient strain to the metal at this temperature to permit recrystallization. If only two reduction cycles are used at 1000 F., the second reduction cycle for Ti-6Al-4V and Ti-8Al-1'Mo-1V must be about 40% although it may be slightly lower depending upon the amount of reduction accomplished during the preceding reduction cycle. The degree of reduction necessary in the second cycle is a function of the degree of reduction of the first cycle since a portion of the strain energy imparted to the metal during the first cycle carries over and is added to the strain created by the second cycle if the intermediate anneal is properly designed. Hence, a larger first reduction and a relatively low temperature short duration anneal will permit recrystallization with less reduction in the second cycle since less strain energy will be required during the second cycle. It is possible to utilize more than two reduction cycles if desired, and as many as five 30% reduction cycles with intermediate anneals have been used to produce quality strip. Since larger reductrons are required at the elevated temperatures to impart the necessary strain to the metal, it is important to beta quench the hot band directly with no preliminar rolling so that a recrystallized strip of reasonable thickness may be obtained for subsequent cold finishing.

After recrystallizing by rolling at elevated temperatures and annealing, the titanium strip alloys can be cold rolled without internally shear cracking up to 100% more than is possible by directly cold rolling beta quenched material. Subsequent to recrystallization they can also be cold rolled through several cycles of reduction with intermediate anneals down to very thin gauge. The proper anneal after cold rolling will result in a second complete recrystallization. For example, a fifteen minute anneal at 1650 F. will recrystallize Ti-5A1-2.S Sn alloy for the second time after it has been cold-rolled through one or more 20% reductions; and a forty-eight hour anneal at 1450 F. followed by vacuum cooling will recrystallize Ti-8A1-1Mo-1V and Ti-6A1-4V alloys for the second time after they have been cold rolled through one or more 2 reductions.

Tests have shown that when rolling at room temperature and annealing are used to effect recrystallization, the maximum reduction possible per cycle without internal shear cracking of the beta quenched hot band is about 10 to 15% for the Ti-6A1-4V and Ti-8A1-1Mo-1V alloys and about 15 to 20% for Ti-A1-2.5 Sn alloy. Heavier cold reductions than these after the beta quench cause internal shear cracking of the metal. However, it is possible to obtain a finished strip with a recrystallized microstructure by effecting larger cold reductions prior to the beta quench and rolling to gauge in one or two cycles after the beta quench depending upon the alloy. As pointed out heretofore, it is not necessary that the metal be beta quenched immediately after unidirectionally hot rolling to hot band so long as suflicient reduction is effected after the beta quench to impart strain to the metal so that substantially complete recrystallization will occur during the final anneal. The desired finish gauge is, therefore, a determining factor in respect of the degree of reduction taken after the beta quench and the reduction temperature, as the metal must be reduced by an amount sufiicient to create the required strain after the beta quench. The temperature of rolling will determine the degree of reduction required to impart strain to the metal as less reduction is required at lower temperatures to create the strain necessary for subsequent recrystallization.

The recrystallization anneal provides the driving energy necessary to recrystallize the strained microstructures into v strain free equiaxed grains; therefore, the exact temperatures and times used are dependent on the rolling cycles used. The recrystallization temperature selected must not be above the beta transus since the object is to recrystallize the alpha structure and not transform alpha into beta. For alpha type titanium alloys a relatively short time line anneal will result in complete recrystallization. For example, a ten minute anneal at 1650 F. followed by air cooling is satisfactory for Ti-5Al-2.5 Sn. Much longer anneals are required for the alpha-beta type alloys and Ti-8Al-lMo-1V and Ti-6A1-4V require fortyeight hour anneals at 1450 F. for effecting substantially complete recrystallization. The longer anneals are termed box anneals and are performed on coils generally in a vacuum furnace to prevent contamination of the metal surface, although the use of a vacuum furnace is not a requirement for our process. Temperatures for vacuum anneals are limited to about 1450 F. because at higher temperatures titanium tends to stick together when under a vacuum; and such is, of course, detrimental to coil annealing.

It should be understood that descaling and pickling of the strip is not a requirement of our process. Descaling and pickling clean the surface and reduce further contamination, and removal of oxides and scale prior to a rolling pass prevents enlarging and spreading these defects.

The method of our invention is advantageous in the production of elongated strip titanium and certain titaniurn base alloys from unidirectionally hot rolled hot band. The strip produced by our method is completely free from ribbing, has no tendency toward ridging and has minimum anisotropy. The beta quench and subsequent deformation and annealing provide equiaxed recrystallized grain structure, and the material can readily be finish cold rolled without internal cracking. Additionally, the material has good ductility and can, therefore, meet bend specifications.

While preferred embodiments of the invention have been described, it may be otherwise embodied within the scope of the appended claims.

We claim:

1. A method for producing strip of a metal selected from the group consisting of commercially pure titanium, alpha stabilized alpha type titanium base alloys and alpha stabilized alpha-beta type titanium base alloys, which cornprises:

(1) unidirectionally hot rolling a body of said metal to reduce said body to an elongated hot band, said rolling being initiated at a temperature requiring a substantial amount of said reduction to occur in the alpha-beta field of said metal,

(2) heaing said hot band at a temperature above the beta transus of said metal to completely transform the crystal structure of said metal to the beta phase,

(3) rapidly cooling said hot band from said temperature above the beta transus of said metal to a temperature below said beta transus to produce an acicular type microstructure in the metal and (4) subjecting said rapidly cooled hot band to the steps of rolling and annealing at temperatures below said beta transus to produce an elongated strip having a substantially completely recrystallized microstructure.

2. A method as set forth in claim 1 including rolling said hot band prior to heating said hot band at a tem perature above said beta transus of said metal to effect a reduction in the thickness of said hot band prior to said heating.

3. A method as set forth in claim 1 including stress relief annealing said rapidly cooled hot band at a temperature below the beta transus of said metal prior to rolling said hot band.

4. A method as set forth in claim 1 wherein said unidirectional hot rolling of said body is initiated at a temperature above the beta transus of said metal.

5. A method as set forth in claim 1 wherein said unidirectional hot rolling of said body is initiated at a temperature sufi'iciently high in the alpha-beta field of said metal that the beta phase of said metal is unstable.

6. A method as set forth in claim 1 wherein said heating of said hot band is at a temperature between the beta transus of said metal and about F. above the beta transus of said metal.

7. A method as set forth in claim 1 wherein said heated hot band is rapidly cooled from said temperature above the beta transus of said metal by immersion.

8. A method as set forth in claim 1 wherein said heated hot band is rapidly cooled from said temperature above the beta transus of said metal to room temperature.

9. A method as set forth in claim 1 wherein said rolling of said rapidly cooled hot band is at a temperature between room temperature and about 1600 F.

10. A method as set forth in claim 1 wherein said rolling of said rapidly cooled hot band is at about 1000 F.

11. A method as set forth in claim 1 wherein said rolling of said rapidly cooled hot band consists of a plurality of rolling cycles between room temperature and about 300 F. and wherein said annealing consists of an anneal after each rolling cycle.

12. A method as set forth in claim 11 including rolling said hot hand between room temperature and about 300 F. to effect a substantial reduction prior to heating 15 said hot band at a temperature above the beta transus of said metal.

13. A method as set forth in claim 1 wherein said metal is an alloy consisting essentially of 5% aluminum, 2.5% tin, balance titanium, and wherein said rolling of said rapidly cooled hot band consists of a plurality of cycles at room temperature with a maximum reduction of 15% to 20% in each cycle and said annealing consists of heating at about 1600" F. for about five minutes after each rolling cycle whereby the final anneal produces said substantially completely recrystallized microstructure in said elongated strip.

14. A method as set forth in claim 1 wherein said metal is an alloy consisting essentially of 5% aluminum, 2.5% tin, balance titanium, and wherein said rolling of said rapidly cooled hot band consists of a single cycle at about 1000 F. to effect a reduction of about 60% and said annealing consists of heating at about 1650 F. for about ten minutes to produce said substantially completely recrystallized microstructure in said elongated strip.

15. A method as set forth in claim 1 wherein said metal is an alloy consisting essentially of 8% aluminum, 1% molybdenum, 1% vanadium, balance titanium, and wherein said rolling of said rapidly cooled hot band consists of a plurality of cycles at room temperature with a maximum reduction of 10% to in each cycle and said annealing consists of heating after each rolling cycle, the final heating being at about 1450 F. for about fortyeight hours to produce said substantially completely recrystallized microstructure in said elongated strip.

16. A method as set forth in claim 1 wherein said metal is an alloy consisting essentially of 8% aluminum, 1% vanadium, balance titanium, and wherein said rolling of said rapidly cooled hot band consists of a plurality of cycles at about 1000 F. with a reduction of to 35% in each cycle and said annealing consists of heating at about 1650 F. for above five minutes between rolling cycles and heating at about 1450 F. for about fortyeight hours after the final rolling cycle to produce said substantially recrystallized microstructure in said elongated strip.

17. A method as set forth in claim 1 wherein said metal is an alloy consisting essentially of 8% aluminum, 1% molybdenum, 1% vanadium, balance titanium, and wherein said rolling of said rapidly cooled hot band consists of a first cycle at about 1000 F. to effect a reduction of about and a second cycle at about 1000 F. to effect a further reduction of about 40% and said annealing consists of heating at about 1650 F. for above five minutes after said first cycle and heating at about 1450 F. for about forty-eight hours after said second cycle to produce said substantially completely recrystallized microstructure in said elongated strip.

18. A method as set forth in claim 1 wherein said metal is an alloy consisting essentially of 6% aluminum, 4% vanadium, balance titanium, and wherein said rolling of said rapidly cooled hot band consists of a plurality of cycles at room temperature to effect a maximum reduction of 10% to 15% in each cycle and said annealing consists of heating after each rolling cycle, the final heating being at about 1450 F. for about forty-eight hours to produce said substantially completely recrystallized microstructure in said elongated strip.

19. A method as set forth in claim 1 wherein said metal is an alloy consisting essentially of 6% aluminum, 4% vanadium, balance titanium, and wherein said rolling of said rapidly cooled hot band consists of a plurality of cycles at about 1000 F. to effect a reduction of 15 to in each cycle and said annealing consists of heating at about 1650 F. for five minutes between rolling cycles and heating at about 1450 fQ about fo y g h urs after the final rolling cycle to produce said substantially completely recrystallized microstructure in said elongated strip.

20. A method as set forth in claim 1 wherein said metal is an alloy consisting essentially of 6% aluminum, 4% vanadium, balance titanium, and wherein said rolling of said rapidly cooled hot band consists of a first cycle at about 1000 F. to effect a reduction of about 25% and a second cycle at about 1000 F. to effect a reduction of about 40% and said annealing consists of heating at about l650 F. for about five minutes after said first cycle and heating at about 1450 F. for about forty-eight hours after said second cycle to produce said substantially completely recrystallized microstructure in said elongated strip.

21. A method as set forth in claim 1 wherein said metal is commercially pure titanium, and wherein said rolling of said rapidly cooled hot band consists of a plurality of cycles at room temperature with a reduction of 15% in each cycle and said annealing consists of heating at about 1400 F. for about seven minutes after each rolling cycle whereby the final anneal produces said substantially completely recrystallized microstructure in said elongated strip.

22. A method as set forth in claim 1 wherein said metal is commercially pure titanium, and wherein said rolling of said rap-idly cooled hot band consists of a single cycle at about 1000 F. to effect a reduction of about 60% and said annealing consists of heating at about 1650 F. for about ten minutes to produce said substantially completely recrystallized microstructure in said elongated strip.

23. A method for producing strip of a metal selected from the group consisting of commercially pure titanium and a titanium alloy consisting essentially of 5% aluminum, 2.5% tin, balance titanium, which comprises:

(1) unidirectionally hot rolling a body of said metal to reduce said body to an elongated hot band, said rolling being initiated at a temperature requiring a substantial amount of said reduction to occur in the alpha-beta field of said metal,

(2) heating said hot band at a temperature above the beta transus of said metal to completely transform the crystal structure of said metal to the beta phase,

(3) rapidly cooling said hot band from said temperature above the beta transus of said metal to a temperature below said beta transus to produce an acicular type microstructure in the metal and (4) rolling said rapidly cooled hot band at a temperature below the beta transus of said metal and sufficiently high to cause recrystallization of the microstructure during said rolling, whereby said rolling produces an elongated strip having a substantially completely recrystallized microstructure.

24. A method as set forth in claim 23 wherein said metal is an alloy consisting essentially of 5% aluminum, 2.5 tin, balance titanium, and wherein said rolling of said rapidly cooled hot band is at a temperature between about 1200 F. and 1400 F.

25. A method as set forth in claim 23 wherein said metal is commercially pure titanium, and wherein said rolling of said rapidly cooled hot band is at a temperature between about 1000 F. and 1200 F.

References Cited UNITED STATES PATENTS 7/1968 Parris 148-ll.5 2/1965 Newman 148-115 UNITED STATES PATENT OFFICE CERTIFICATE OF CORRECTION Patent No 3 ,492 ,172 January 27 1970 Adrian Burt Sauvageot et al It is certified that error appears in the above identified patent and that said Letters Patent are hereby corrected as shown below:

Column 3 line 51 "heat" should read near Column 5, line 40, "rolling annealing" should read rolling and annealing Column 9 at the bottom of Table E, "Tot tested." should read Not tested. Column 11 in the last heading in Table H, "RT" should read R/T same column ll line 41 "The" should read Two Column 14', line 25 "heaing" should read heating Column 15 line 32 molybdenum,"

should be inserted after 'aluminum,".

Signed and sealed this 23rd day of June 1970.

(SEAL) Attest:

EDWARD M.FLE CHER, JR. WILLIAM E. SCHUYLER, JR. Attesting Of icer Commissioner of Patents