Aluminum-lithium welding alloys

Abstract

Welding alloys based upon aluminum, copper, mangesium, silver and lithium have been found to possess exceptional properties such as resistance to hot cracking, high strength and ductility, and an unexpected natural aging response. The aluminum-base alloys may contain 3.5 to 7 weight percent Cu, 0.05 to 1.5 weight percent Mg, 0.01 to 2 weight percent Ag, 0.1 to 4 weight percent Li, and 0.01 to 2 weight percent grain refiner.

Description

FIELD OF THE INVENTION

The present invention relates to aluminum-base welding alloys which are characterized by a unique combination of mechanical and physical characteristics. Specifically, the aluminum-copper-magnesium-silver-lithium welding alloys of this invention possess exceptionally high strength, ductility, low density, excellent weldability, and an exceptional natural aging response.

BACKGROUND OF THE INVENTION

Aluminum and its alloys have long been known to have desirable properties such as low cost, relatively low density, corrosion resistance, and the ability to be fabricated relatively easily into a wide variety of shapes and configurations for various applications. Aluminum, as a base metal for alloying purposes, is generally approximately 99 per cent pure, with iron and silicon comprising the major impurities.

A number of aluminum alloys have been found having good weldability, and characteristics such as medium to high strength. Several of these aluminum alloy compositions are based upon aluminum-copper systems, wherein significant strengthening is induced by the precipitation of CuAl.sub.2. In recent work, Polmear has disclosed an aluminum-copper-magnesium-silver alloy composition designed for elevated temperature service that has high room temperature strength. No weldability data for the alloy have been reported. Polmear reported that the presence of small amounts of magnesium and silver create a previously unreported novel precipitate (i.e. omega phase) which additionally strengthens the alloy; such precipitates being coherent platelets of very minimal thickness, and high stability at slightly elevated temperatures. The strengthening such alloys appears to be a function of the presence of both magnesium and silver. See "Development of an Experimental Wrought Aluminum Alloy for Use at Elevated Temperatures," Polmear, ALUMINUM ALLOYS: THEIR PHYSICAL AND MECHANICAL PROPERTIES. E. A. Starke, Jr. and T. H. Sanders, Jr., editors, Volume I of Conference Proceedings of International Conference, University of Virginia, Charlottesville, VA, 15-20 Jun. 1986, pages 661-674, Chameleon Press, London.

Lithium additions to aluminum alloys are known to produce a variety of low density, age-hardenable alloys, such as alloys of aluminum-magnesium-lithium, and aluminum-copper-lithium. Such alloys have seen limited use, due to difficulty in melting and casting and relatively low ductility and toughness in wrought products. Among conventional commercially available aluminum-lithium alloys, the only generally accepted weldable aluminum-lithium alloy is the Soviet alloy 01420, as disclosed in British patent 1,172,736, of Fridlyander et al., comprising Al-5Mg-2Li. This alloy is reported to have medium to high strength, low density, and a modulus of elasticity higher than standard aluminum alloys.

In U.S. Pat. No. 2,381,219, of I. M. LeBaron, assigned to Alcoa, aluminum base alloys containing Cu and Li are disclosed. These alloys are "Mg-free", i.e. the alloys have less than 0.01 percent Mg which is present as an impurity. In addition, the alloys disclosed by LeBaron require the presence of at least one element selected from Cd, Hg, Ag, Sn, In, and Zn. The surprising increase in strength observed in the Al-Li alloys of the present invention, containing both Ag and Mg, is not observed in LeBaron.

U.S. Pat. No. 2,915,391 to Criner, assigned to Alcoa, discloses Al-Cu-Mn base alloys containing Cu, Li, Mg, and Cd. Criner teaches that Mn is essential for developing high strength at elevated temperatures and that Cd, in combination with Mg and Li, is essential for strengthening the Al-Cu-Mn system. Criner does not achieve properties comparable to those of the present invention, i.e. ultra high strength, strong natural aging response, high ductility at several technologically useful strength levels, weldability, etc.

A paper appearing in the Journal of Japan Institute of Light Metals lists Al-Li-Mg base alloys to which one of the elements Ag, Cu, or Zn has been added (see "Aging Phenomena of Al-Li-Mg Alloy Affected by Additional Elements," Hayashi et al, Journal of Japan Institute of Light Metals, Vol. 32, No. 7, July 1982). While the reference does disclose the addition of Ag to an alloy consisting of Al, Li, and Mg, no disclosure is made of the addition of silver to a copper containing alloy. Further, the reference does not teach the addition of grain refining elements as required in the alloys of the present invention. Since the purpose of the paper is to study the effect of specific alloying additions on the aging behavior of ternary Al-Li-Mg alloys, the authors specifically desired to limit additions to the individual elements Ag, Cu, and Zn listed without combining such additional elements or adding further grain refining elements.

U.S. Pat. No. 3,306,717 to Lindstrand et al relates to filler metal for welding Al-Zn-Mg base alloys. The filler metal comprises Al with 2-8 weight percent Mg, 0.1-10 weight percent Ag, and up to 8 weight percent Si In addition, the filler metal may contain up to 1.0 weight percent each of Mn and Cr, up to 0.5 weight percent each of Cu, Ti and V, and up to 0.1 weight percent each of Li, Zr and B. The only example given by Lindstrand et al lists a filler metal composition of Al--5 Mg--0.9 Ag.

SUMMARY OF THE INVENTION

It is an object of the present invention to provide a novel aluminum-base alloy composition suitable for use as a welding alloy.

It is another object of the present invention to provide an aluminum-base welding alloy having improved properties such as high strength, good weldability, i.e., high resistance to hot cracking, and good resistance to weld corrosion.

By the addition of lithium to alloys similar to those suggested by Polmear, we have achieved surprisingly high strength, a degree of ductility unexpected in lithium-containing aluminum alloys, good weldability, and a natural aging response superior to any previously reported for an aluminum alloy.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1: Isochronal aging curves for Al-Cu-Mg-Ag-Li Alloy 049.

FIG. 2: Isochronal aging curves for Al-Cu-Mg-Ag-Li Alloy 051.

FIG. 3: Isochronal aging curves for Al-Mg-Ag-Li Alloy 050.

FIG. 4: Isothermal aging at 160.degree. C. for Alloy 049.

FIG. 5: Isothermal aging at 160.degree. C. for Alloy 050.

FIG. 6: Natural aging response of Alloy 049 extrusion with and without 3% stretch.

FIG. 7: Natural aging response of Alloy 051 extrusion with and without 3% stretch.

FIG. 8: Natural aging response of Alloy 050 extrusion with and without 3% stretch.

FIG. 9: Room Temperature tensile properties of Alloy 049 for various aging times at 160.degree. C.

FIG. 10: Compressive yield strength of Alloy 051 for various deformation temperatures.

FIG. 11: Tensile yield strength vs. temperature for alloy 049.

FIG. 12: Ultimate tensile strength vs. temperature for alloy 049.

FIG. 13: Ultimate tensile strength vs. weight percent lithium of Al-Cu based alloys in the SUA temper.

FIG. 14: Yield strength vs. weight percent lithium of Al-Cu based alloys in the SUA temper.

FIG. 15: Natural aged strength of three percent stretched Al-Cu based alloys as a function of lithium content.

FIG. 16: Natural aged strength of unstretched Al-Cu based alloys as a function of lithium content.

FIG. 17: Natural aging response of unstretched 049 Alloy after quenching from 504.degree. C. into various media.

FIG. 18: Strength vs. elongation for alloy 049.

DETAILED DESCRIPTION OF THE INVENTION

The primary elements of the welding alloys of the present invention are aluminum, copper, magnesium, silver and lithium, with grain refining additive elements such as zirconium. Although emphasis herein shall be placed upon use of zirconium for grain refinement, conventional grain refiners such as chromium, manganese, titanium, boron, vanadium, hafnium, titanium diboride, and mixtures thereof, may be used. Additionally, elements such as Be, Co and Bi may be included to aid weldability.

In general, the welding alloys of this invention comprise from about 2.0 percent to about 9.8 percent by weight for the total of copper plus magnesium, preferably from about 2.5 to about 8.5 percent, and more preferably from about 3.1 to about 7.5 percent. Silver may generally comprise from about 0.01 to about 2.0 percent by weight of the alloy, preferably 0.05 to 1.2 percent, and more preferably 0.10 to 1.0 percent. The highest strengths are achieved with Ag concentrations of between about 0.3 and 0.5 percent. Lithium may comprise from about 0.05 to 4.1 percent by weight of the alloy, preferably 0.20 to 3.1 percent, and most preferably 0.50 to 2.7 percent. Below about 0.05 percent Li, significant reductions in density are not realized while above about 4.1 percent the solubility limit of Li is passed. Similarly, the total amount of grain refiner present may comprise from about 0.05 to 1.0 percent by weight of the alloy, preferably 0.05 to 0.6 percent, and most preferably 0.05 to 0.5 percent.

The alloys may be broadly characterized as either aluminum-copper based, or aluminum-magnesium based, with compositions varying as set forth in Table I. In those alloys classified as aluminum-copper based, the copper may comprise from about 3.5 to about 7 percent by weight, preferably from about 4.0 to about 6.5 percent, more preferably from about 4.5 to about 5.5 percent, and most preferably from about 4.8 to 5.2 percent. Below about 3.5 or 4 percent Cu, weldability decreases to an unacceptable degree. Above 7 percent Cu, density increases while corrosion resistance and fracture toughness decrease. In such alloys, the magnesium may comprise from about 0.05 to about 1.5 weight percent, preferably 0.05 to 1 percent, more preferably 0.10 to 0.8 percent, and most preferably 0.2 to 0.5 percent. The highest strengths are achieved in alloys containing 0.3 to 0.5 percent Mg, with gradual decreases in tensile properties below and above this range. Tensile properties are highly dependent upon Li content in the Al-Cu based alloys. The highest strengths are attained with Li levels of from about 1 to about 1.5 percent with decreases below and above these percentages.

In those alloys classified as aluminum-magnesium based, the magnesium may comprise from about 2.0 to about 9.8 percent by weight, preferably from 2.5 to 7.0 percent, and most preferably from about 3.0 to 5.5 percent. In such alloys, the copper may comprise from 0 to 1.5 percent by weight, preferably from 0.01 to 1.0 percent, and most preferably from 0.10 to 1.0 percent.

The alloys are presently disclosed, particularly the aluminum-copper based alloys, stronger by far than conventional alloys of the closest known compositions, and stronger than would be predicted based on the scientific literature. Furthermore, they have greater specific properties than conventional weldable Al alloys. They also have a surprisingly strong natural aging response, which is extremely useful for welding. Their weldability, defined as the resistance to hot cracking that can occur during welding, is excellent. This is surprising, in view of the disappointing weldability of recently commercialized Al-Li alloys.

Three alloy melts having the compositions as set forth hereinafter in Table II were prepared. Each alloy was cast, homogenized, extruded, solutionized, quenched, and stretched, using the parameters set forth in Table III.

TABLE I______________________________________COMPOSITIONAL RANGES More Most Broad Preferred Preferred Preferred______________________________________Al--Cu BasedCu 3.5-7 4-6.5 4.5-5.5 4.8-5.2Mg 0.05-1.5 0.05-1 0.1-0.8 0.2-0.5Ag 0.01-2 0.05-1.2 0.1-1 0.3-0.5Li 0.1-4 0.5-2.6 0.8-1.8 1.0-1.6Grain Refiners 0.01-2 0.05-1 0.05-0.5 0.08-0.2Al Balance Balance Balance BalanceAl--Mg BasedCu 0-1.5 0.01-1 0.1-1 0.1-1Mg 2-9.8 2.5-7 3-5.5 3-5.5Ag 0.01-2 0.05-1.2 0.1-1 0.3-0.5Li 0.05-4.1 0.2-3.1 0.5-2.7 0.5-2.7Grain Refiners 0.05-1.0 0.05-0.6 0.05-0.5 0.05-0.5Al Balance Balance Balance Balance______________________________________ TABLE II__________________________________________________________________________COMPOSITIONS (Weight Percent) H DensityHeat # Cu Mg Ag Li Zr (ppm) Al g/cm.sup.3 (lb/in.sup.3)__________________________________________________________________________049 Planned 6.30 0.40 0.40 1.40 0.14 -- Balance Actual 6.20 0.37 0.39 1.21 0.17 0.201 Balance 2.73 (0.098)050 Planned 0 5.00 0.40 2.00 0.14 -- Balance Actual -- 4.90 0.39 1.91 0.15 0.541 Balance 2.52 (0.091)051 Planned 6.30 0.40 0.40 1.70 0.14 -- Balance Actual 6.51 0.40 0.38 1.45 0.16 0.622 Balance 2.72 (0.098)__________________________________________________________________________ TABLE III__________________________________________________________________________PROCESSING PARAMETERSPour Homogenization Extrusion Ram*Heat Temperature Temp. Time Temp. Speed SHT Stretch# (.degree.C.) (.degree.C.) (h) (.degree.C.) (in/min) (.degree.C.) %__________________________________________________________________________049 740 450 (16 h) + 510 (8 h) 370 0.10 504 3.6050 740 450 (16 h) + 510 (8 h) 375 0.10 465 3.0051 740 450 (16 h) + 510 (8 h) 373 0.12 504 3.0__________________________________________________________________________ *for 6inch diameter billets in a 6.375 inch chamber extruding into 0.375 .times. 4.0 inch flat plate

AGING RESPONSE

Isochronal aging curves were developed for the three alloys to identify advantageous aging temperatures, i. e., those which would produce high hardness in a practical period of time (FIGS. 1-3). Then, artificial aging curves were developed at selected temperatures and natural aging studies were performed with different amounts of stretch (see FIGS. 4 through 8). The copper-containing alloys, e.g., alloy 049, display a rapid and extremely strong aging response at 160.degree. C. (FIG. 4), a temperature where Al-Cu alloys normally respond sluggishly, although a decrease in hardness is noted during the early stages of aging (i.e., "reversion").

The Al-Cu-Mg-Ag-Li alloys show a natural aging response that is extraordinary, and particularly significant in terms of utility (FIGS. 6, 7). The response is so rapid for alloy 049 that within ten hours after quenching, hardness increases from 48 to 72 R.sub.B, both with a 3% cold stretch prior to aging, and also without prior cold work. This is extremely important because all existing pilot-commercial Al-Li alloys have insufficient strength to replace existing high technology alloys unless they are cold worked prior to aging. This is especially problematical for forging alloys, where it is most often not practical to induce cold work prior to aging. It is significant that the natural aging response without cold work is ultimately greater than that with cold work, an extremely beneficial attribute for forging. This natural aging response is observed for both heats 049 and 051. Alloy 050 displayed a slight natural aging response with no stretch, and a more rapid response with stretch (FIG. 8).

In contrast, known 2xxx alloys have a very slight, insignificant natural aging response without prior cold work. Ultimate tensile strengths such as were obtained with the subject alloys were totally unanticipated on the basis of 2xxx alloys.

TENSILE PROPERTIES

The alloys were tested in tension in various aging conditions in both the longitudinal (L) and long transverse (LT) directions. In Table IV, one can see that the properties are clearly exceptional and substantially stronger than any conventional weldable aluminum alloy, and also stronger than Soviet alloy 01420, the only commercial Al-Li alloy designed to be weldable.

The highest strengths reported are generally in the "slightly underaged" (SUA) temper, where toughness is expected to be higher than in the peak-aged temper. One can look at the artificial aging curves (FIGS. 4 and 5) to assess the degree of underaging in the 160.degree. C. (24 h) temper reported.

The new alloys also demonstrate unexpectedly good properties in the naturally aged condition (Table V). In fact, Al-Cu-Mg-Ag-Li alloy 049 is stronger in the naturally aged condition than any conventional weldable aluminum alloy is in its peak strength condition. This has important implications for welding, where a strong natural aging response provides the potential to restore strength in weldments. Moreover, it is extremely significant that this exceptional natural aging response exists without prior cold work. Consequently, a useful vehicle exists to increase the strength of weldments and forgings in which inducing cold work is often not practical. Even better naturally aged properties were obtained in subsequently prepared alloys, as described hereinafter (Table XIV). Alloy 050, an Al-Mg-Ag-Li alloy, in the stretched and naturally aged condition (T3 temper) is essentially as strong as alloy 01420 in the artificially aged condition. The alloys of the present invention also have superb specific properties, with the specific strength of 049 almost twice that of 2219 T81 (Table VI).

With artificial aging, the experimental alloys 049 and 051 are actually as strong as, or stronger than, any ingot metallurgy aluminum base alloy, both weldable and non-weldable, as can be seen by comparison of Tables V and VII, and FIG. 9.

In contrast, the Ag-containing, Mg-free, Al-Li alloy developed by LeBaron did not show the surprising strength increases contained in the present invention. In fact, the presence of Ag decreased strength slightly (see Table VIII).

The present Al-Cu-Mg-Ag-Li alloys (049 and 051) show a significant strength increase with respect to the experimental Al-Cu-Mg-Ag-Mn-Zr-Ti-Fe-Si alloy developed by Polmear. As may be seen, improvements of 34 percent and 23 percent were observed for yield strength and ultimate tensile strength, respectively, for alloy 049 when compared to the Polmear alloy (Table VIII).

TABLE IV__________________________________________________________________________TENSILE PROPERTIES OF EXPERIMENTAL ANDCONVENTIONAL WELDABLE ALLOYS Aging Temp. (time) .sup.E Measured YS UTS E1 DensityID # Orientation .degree.C. (h) (.times. 10.sup.3 ksi) (ksi) (ksi) % (lb/in.sup.3)__________________________________________________________________________Novel Weldable Alloys (SUA)049 L 160 (24) 11.3 100.4 103.3 5.1 0.098 LT 160 (24) 11.3 94.1 97.8 1.9051 L 160 (24) 11.5 89.8 94.9 5.2 0.098 LT 160 (24) 11.3 90.2 92.8 1.3050 L 160 (14) 11.1 53.5 72.5 7.7 0.091 LT 160 (14) 11.1 52.5 76.2 14.1Conventional Weldable Alloys2219 L 10.6** 44.0** 61.0** 6.0** 0.102T815083 L 10.3*** 31.0*** 44.0*** 12.0*** 0.096***H3217039 L 10.4*** 50.0*** 60.0*** 14.0*** 0.099***T6101420 L 11.1* 42.2* 61.7* 16.9* 0.0908*__________________________________________________________________________ *Typical actual data from 50lb heats. **Minima for plate. The minima for extrusion are 42.0 ksi YS, 58.0 ksi UTS. ***Handbook Minima TABLE V______________________________________TENSILE PROPERTIES OF ALLOYS Aging Condition Temp. (time) YS UTS E1Alloy # Orientation .degree.C. (h) (ksi) (ksi) (%)______________________________________Novel Alloys in Various Tempers049 L NA* (1000 h) 55.6** 74.0** 16.5 LT NA* (3000 h) 52.0** 71.5** 21.7 L 160 (14 h) 94.6 98.2 5.2 L 160 (24 h) 100.4 103.2 5.1050 L NA* (1000 h) 44.9** 59.4** 11.9 LT NA* (3000 h) 38.6** 59.4** 16.9 L 160 (14 h) 53.5 72.5 7.7 LT 160 (14 h) 52.5 76.2 14.1Handbook Property Minima (for extrustions)2219 T81 42.0 58.0 6.0______________________________________ *NA = naturally aged with aging time following **Cold stretched three percent prior to natural aging. TABLE VI______________________________________SPECIFIC PROPERTIES OF EXPERIMENTAL ALLOYS(Slightly Underaged Temper) UTS/.rho. E/.rho.Alloy # (.times. 10.sup.3 in) (.times. 10.sup.6 in)______________________________________Novel Al--Li Weldable Alloys049 1053 115050 798 122051 967 117Handbook Property Minima2219 T81 592 103______________________________________ TABLE VII__________________________________________________________________________Highest Strength Non-WeldableAluminum I/M Alloys (Extrusions, Typical Values) YS UTS E1 E DensityAlloy System (ksi) (ksi) % (.times. 10.sup.3 ksi) (lb/in.sup.3)__________________________________________________________________________7001 T6 Al--Zn--Mg 91 98 9 10.3 0.1037178 T6 Al--Zn--Mg 78 88 11 10.3 0.1028091 T651 Al--Li--Cu--Mg 88.5 91.4 4 11.7 0.0917__________________________________________________________________________ TABLE VIII______________________________________ YS UTS E1Alloy (ksi) (ksi) %______________________________________LeBaronAl--4.5 Cu--0.7 Mn--0.25 Si--0.15 Fe-- 75.9 82.1 7.81.0 Li--0.15 CdLeBaronAl--4.5 Cu--0.8 Mn--0.25 Si--0.15 Fe-- 74.5 80.6 7.51.0 Li--0.15 Cd--0.15 AgPolmear experimental alloyAl--6.7 Cu--0.46 Mg--0.50 Ag--0.48 Mn-- 75.0 84.1 7.50.18 Zr--0.07 Si--0.1 Fe--0.06 Ti049Al--6.2 Cu--0.37 Mg--0.39 Ag--1.21 Li-- 100.4 103.3 5.10.17 Zr______________________________________

WELDABILITY

Fusion Weldability

Tungsten Inert Gas

Alloy 049 was manually TIG (tungsten inert gas) welded with both conventional 2319 filler and parent metal filler. Alloy 050 was manually TIG welded with both conventional 5356 filler and parent metal filler. V-notch butt weldments with a fair amount of constraint were made. TIG (tungsten inert gas) welding. The V-notch weldment does not produce weldments of particularly high joint efficiency (i.e. strength weldment/strength parent alloy). However, constrained V-notch butt weldments are instructive for demonstrating weldability when welding parameters are not known for a new material. In addition, automatic TIG square butt weldments were made on alloys 049, 050, and 051 using conventional fillers.

Whereas weldability is defined as the resistance to hot cracking, none of the weldments displayed any hot cracking susceptibility despite the constraint imposed. Furthermore, the welding parameters used (Table IX), which were selected based on analogy to conventional alloys, led to good metal flow during welding and caused the welders to claim that the alloys welded as easily as conventional alloys 2219 and 5083.

The alloys had low hydrogen content (Table II), yet weldments did display gas-formed pores at the weld bead-parent alloy interface. The porosity is believed to have resulted from improper handling during welding. This porosity caused the weldments to fail in tension at that interface. Nevertheless, the strengths of several weldments are quite good (see Table X) and suggest that very good weld strengths are possible. Corrective care to weldment pretreatment was exercised in subsequent welding tests, and no significant porosity was detected by radiographic techniques. In fact, TIG weldments of an 049-type alloy (059, described hereinafter) using parent metal filler, passed the very stringent quality control production test for porosity which is used on the Titan missile.

Electron Beam Welding

Autogenous electron beam (EB) weldments were performed on an alloy similar to alloy 049, alloy 061 described hereinafter. A square butt weld was fabricated along the longitudinal axes of two 100 mm wide plates of alloy 061. The welding was performed at 140 kv potential and 80 mA current at a travel speed of 500 mm/min. in a single pass.

The alloy proved to be readily weldable by this technique, and the two 18 mm thick plates were easily joined in the full penetration pass.

The weldments were radiographed and showed negligible porosity, and long transverse tensile properties were determined. The tensile strengths were exceptionally high, and are from material in the as-welded condition with only 12 days natural aging (see Table XI). Although the apparent elongation over the 4 inch length is 0.3 percent, the deformation was localized to the narrow (about 5 mm) weld zone, so the true strain-to-failure ratio in this localized region is much higher. This is supported by the much higher 9% reduction in area.

Friction Weldability

Several specimens of 049-type alloy 061 were stretched to 0.5 inch diameter, 2 inch cylindrical rods, and were friction welded using parameters in Table XII. Although a detailed study was not undertaken to develop optimized friction welding parameters, good strengths were obtained (Table XII). For example, the tensile strength of joints were as high as 57.9 ksi with no post-welding heat treatment, demonstrating that this alloy is amenable to friction welding.

TABLE IX__________________________________________________________________________Welding Parameters Used forNovel Alloys__________________________________________________________________________ Filler Voltage Current # of Time per pass LengthAlloy Material (volts) (amps) Passes second(s) (inches)__________________________________________________________________________MANUAL TIG PROCEDURE049 2319 18-24 190-250 4 155, 120, 121, 148 (10 in.)049 Parent 20-24 150-205 4 71, 61, 41, 53 (5 in.)050 5356 18-24 150-210 4 161, 138, 112, 115 (10 in.)050 Parent 19-22 160-210 4 63, 52, 46, 48 (5 in.)3/16 in. diameter, 2% thoriated tungsten electrodeCoverages: 100% He at 8 cu ft/h or 100% Ar at 25 cu ft/h.75.degree. V-NotchPreheat base metal to 170.degree. F.Gas Cup size: 3/8 in.__________________________________________________________________________Weld Voltage Current Travel Speed Wire SpeedPass (Volts) (Amps) (in/min) (in/min)__________________________________________________________________________AUTOMATIC TIG PROCEDURESeal Pass 11.2 190 12.0 --Penetration Pass 11.4 245 9.8 --Filler Pass 11.7 195-200 9.8 46-55DC current straight polarity049 and 051 used 2319 filler wire050 used 5356 filler wire, square butt weldsCoverages: 100% He at 8 cu ft/h. or 100% ar at 25 cu ft/h.__________________________________________________________________________ TABLE X__________________________________________________________________________Tensile Properties from TIG WeldmentsParent Post-Welding Weld YS UTS E1* Type of TIGAlloy Filler Temper Bead (ksi) (ksi) (%) Welding__________________________________________________________________________049 2319 NA (500 h) On 31.6 38.9 3.0 Manual049 2319 NA (500 h) Flush 28.8 39.0 3.0 Manual049 2319 NA (500 h) Flush 31.8 47.5 5.7 Auto049 2319 SHT 504.degree. C./WQ/NA(770 h) Flush 42.8 61.0 9.5 Auto049 2319 SHT 504.degree. C./WQ/160.degree. C.(24 h) Flush 56.8 58.6 0.4 Auto049 2319 SHT 504.degree. C./WQ/NA(770 h) Flush 33.0 51.3 6.0 Manual049 049 NA (500 h) On 38.4 44.0 0.5 Manual049 049 NA (500 h) Flush 38.5 40.6 0.5 Manual050 5356 NA (500 h) On 22.3 40.6 6.5 Manual050 5356 NA (500 h) Flush 22.5 43.2 14 Manual050 5356 NA (500 h) Flush 24.6 48.0 13 Auto050 050 NA (500 h) On 27.1 42.8 6 Manual050 050 NA (500 h) Flush 22.5 51.0 17 Flush__________________________________________________________________________ All alloys were in the SUA temper before welding *Elongation is from a 1inch gauge length. TABLE XI______________________________________Long Transverse Electron Beam Weldment Propertiesof Alloy 061Temper Before ApparentWelding Apparent UTS Elongation Reduction in AreaWelding YS ksi % %______________________________________SUA 60.5 62.4 0.3 9SUA 60.4 63.0 0.3 9______________________________________ As welded, NA (12 days). TABLE XII______________________________________Friction Welding of Alloy 061 in SUA TemperRotational Friction Forging Condition TensileSpeed Force Force After Strength(rpm) (lbs) (lbs) Welding (ksi)______________________________________3000 1200 2400 NA (4 h) 43.83000 2000 3000 NA (4 h) 44.53000 4000 5000 NA (24 h) 40.53000 4000 5000 NA (192 h) 57.93000 4000 5000 AW-160.degree. C. (20 h) 59.6______________________________________ *Water quenched 15 s after friction welding.

ELEVATED TEMPERATURE DATA

Elevated temperature properties were determined using hot compression tests on alloy 051 and conventional weldable alloy 2219, the leading conventional alloy for 120.degree. C. service. (Table XIII, FIG. 10).

The compression tests were conducted at a strain rate of 8.times.10.sup.-3 s.sup.-1 after holding at temperature for 30 minutes. The novel alloy showed exceptional warm temperature strength, well above that of 2219 (see FIG. 10), throughout the entire temperature range investigated, which is considered to be beyond the useable temperature range of 2219. In fact, alloy 051 has considerably higher yield strength at 200.degree. C. than 2219 has at room temperature, and very nearly the same yield strength at 250.degree. C. that 2219 has at room temperature.

Several tensile tests were performed at elevated temperatures (Tables XIII, FIGS. 11 and 12) and strength values were found to be exceptional. Both tensile and yield strengths are superior to alloy 2014.

TABLE XIII______________________________________Elevated Temperature Data for Novel Alloys and 2219______________________________________Elevated Temperature Compression Dataof Novel Alloy and 2219 Deformation Temperature YS*Alloy Orientation (.degree.C.) (ksi)______________________________________051 L 24 94.3 100 92.0 150 84.8 200 74.1 250 52.12219-T87 L 24 56.0 150 49.4 250 31.3______________________________________Elevated Temperature Tensile Data for Alloy 049 Deformation Temperature YS UTS E1 ETemper .degree.C. ksi ksi (%) (.times. 10.sup.6______________________________________ psi)SUA 93 95.3 99.0 8.1 12.1SUA 93 94.4 94.8 6.11 11.5NA 93 55.9 70.6 NA 11.5SUA 150 86.7 90.1 6.2 11.2SUA 150 86.1 89.2 7.2 11.4NA 150 79.6 83.4 10.5 10.8______________________________________ *0.2 percent off set yield strength.

ADDITIONAL ALLOYS

Having obtained the exceptional properties demonstrated for alloys 049-051, ten additional 50 lb. heats were fabricated in accordance with the compositions set forth in Table XIV. Several tests were near-duplicates of 049 and 050, to demonstrate reproducibility, and others were attempts to improve specific properties. For example, Al-Cu based alloys similar to 049 were fabricated with higher Li levels to decrease density and increase modulus. Other Al-Cu based alloys were made with lower Cu content to decrease density and to determine whether hot cracking susceptibility increases at the Cu level investigated, as predicted by data from Al-Cu binary alloys. Meister and Martin, "Welding of Aluminum and Aluminum Alloys," Defense Metals Information Center, Battelle Memorial Institute, Columbus, Ohio, 1967.) Additionally, two Al-Cu based Alloys similar to 049 were fabricated with lower Cu levels of 4.5 and 5.0 weight percent. The compositions and tensile properties of extrusions of these alloys in the T8 temper are given in Table XIVa.

Two Al-Mg based alloys similar to 050 were prepared, one of the same nominal composition (062), and one fabricated with 0.2 percent Cu to assess the effect of Cu on strength. The alloys were cast and homogenized using parameters similar to those in Table II. Extrusion and processing parameters are in Table XIV, and comparative tensile properties are shown in Tables XV.

The near duplicate tests of alloy 049 (i.e. alloys 059, 060, 061) reproduced the unexpectedly high properties of alloy 049 superbly (see Table XVI). The increased addition of Li to Al-Cu based alloy (064, 065, 066, 067) did indeed decrease density (Table XIV) but had an unexpected effect on strength. Both tensile and yield strength decreased, in the slightly underaged condition, with increasing Li content in the range studied (see FIGS. 13, 14). The same trend was observed for the naturally aged data both with and without cold stretch (see FIGS. 15, 16). This strength decrease is the opposite effect for virtually all practical Al-Cu-Li alloy compositions. In fact, very recent Japanese work (Yo Kojima et al., "Effect of Lithium Addition on the Elevated Temperature Properties of 2219 Alloy," Journal of Japan Institute of Light Metals. Vol. 36, No. 11, Nov. 1986, pp. 737-743.) related to Li-modified 2219 alloys (Al--6 wt % Cu) shows a directly related increase in yield strength with increasing Li content from 0 to 2 weight percent (see Table XVII).

The natural aging tensile data for alloys similar to type 049 alloys are higher than any reported in the literature for aluminum alloys. In fact, the tensile strengths without cold work, 84.2 to 86.1 ksi, are significantly higher than the properties of 2219 in the fully artificially aged condition. (See Table XVIII.)

Despite the fact that strength decreases with Li content from about 1.2 percent to about 1.9 weight percent, the strengths at the higher lithium level alloys are still competitive with other high strength alloys as well as conventional alloys. Thus, there are several technologically useful families of alloys within the scope of the present invention; for example, an ultra high-strength alloy (049) with a four percent decrease in density with respect to 2219, and a high strength alloy (065) with a six percent or higher density decrease.

Attempts to decrease density by lowering the copper content were successful in that strength did not change, for example, with a decrease in copper content from 6.3 to 5.4 weight percent at both the 1.4 weight percent and 1.7 weight percent lithium levels.

TABLE XIV__________________________________________________________________________EXTRUSIONS OF ADDITIONAL COMPOSITIONS Ext. Ext. Ram ExtrusionAlloy Planned Compositions (wt. %) Temp Speed Dimension DensityID # Cu Mg Ag Li Zr (.degree.C.) (in/min) (in) g/cm.sup.3 (lb/in.sup.3)__________________________________________________________________________058 Planned 6.30 0.40 0.40 1.40 0.15 358 0.10 2 DIA rod 2.71 (0.098) Measured 5.79 0.38 0.40 1.34 0.17059 Planned 6.30 0.40 0.40 1.40 0.15 364 0.12 4 .times. 3/8 2.72 (0.098) Measured 6.21 0.40 0.43 1.39 0.16060 Planned 6.30 0.40 0.40 1.40 0.15 355 0.20 4 .times. 3/8 2.72 (0.098) Measured 6.18 0.40 0.42 1.32 0.155061 Planned 6.30 0.40 0.40 1.40 0.15 364 0.12 2 .times. 3/4 2.72 (0.098) Measured 6.33 0.41 0.42 1.34 0.16062 Planned 0.00 5.00 0.40 2.00 0.15 349 0.12 2 .times. 3/4 2.53 (0.091) Measured -- 5.40 0.42 2.10 0.135063 Planned 0.20 5.00 0.40 2.00 0.15 327 0.08 4 .times. 3/8 2.50 (0.090) Measured 0.22 5.25 0.41 2.02 0.14064 Planned 6.30 0.40 0.40 1.70 0.15 356 0.12 4 .times. 3/8 2.69 (0.097) Measured 6.30 0.41 0.41 1.62 0.155065 Planned 6.30 0.40 0.40 2.00 0.15 344 0.12 4 .times. 3/8 2.68 (0.097) Measured 6.31 0.46 0.41 1.86 0.145066 Planned 5.40 0.40 0.40 1.40 0.15 328 0.12 4 .times. 3/8 2.70 (0.098) Measured 5.56 0.46 0.41 1.38 0.165067 Planned 5.40 0.40 0.40 1.70 0.15 334 0.12 4 .times. 3/8 2.67 (0.096) Measured 5.55 0.44 0.40 1.66 0.16__________________________________________________________________________ TABLE XIVa______________________________________Composition and Tensile Properties of ADditional AlloysAlloy YS UTS E1ID # Cu Mg Ag Li Zr Temper (ksi) (ksi) (%)______________________________________068 5.0 0.4 0.4 1.3 0.14 T8 100.5 103.5 8.8069 4.5 0.4 0.4 1.3 0.14 T8 100.8 103.3 8.5______________________________________ TABLE XV______________________________________Tensile Properties of Al--Mg Based Alloys YS UTS E1Alloy Temper Orientation (ksi) (ksi) (%)______________________________________050 SUA L 53.5 72.5 7.7 SUA LT 52.5 76.2 14.1 T3* L 44.9 59.4 11.9062 SUA L 55.3 75.2 9.2 T3* L 45.5 61.8 15.1063 SUA L 57.9 75.5 6.7 SUA LT 51.9 73.3 9.2 T3* L 46.1 60.0 12.401420 SUA L 42.2 61.7 16.6 SUA LT 38.2 61.7 16.9______________________________________ *T3 = stretched 3%, natural aged 1200 h. TABLE XVI______________________________________Slightly Underaged Tensile Propertiesof Near Duplicate Heats of Alloy 049 AgingAlloy Temp (time) YS UTS E1ID # Orientation .degree.C. (h) (ksi) (ksi) %______________________________________049 L 160 (24) 100.4 103.3 5.1 LT 160 (24) 94.1 97.8 1.9059 L 160 (20) 99.8 102.4 5 LT 160 (20) 95.8 99.4 3060 L 160 (24) 98.7 101.5 5.4 LT 160 (24) 94.1 99.0 3.0061 L 160 (24) 101.4 104.0 4.7______________________________________ TABLE XVII______________________________________Room Temperature Yield Strength of Li-modified 2219Increases with Increasing Li ContentAlloy YS (ksi)______________________________________2219 46.92219 + 1% Li 67.52219 + 2% Li 81.0Al + 2% Li 24.2______________________________________ TABLE XVIII______________________________________Natural Aging Response of Al--Cu Based Alloys(NA 1200 h) YS UTS E1Alloy Stretch (ksi) (ksi) %______________________________________060 3% 57.5 74.9 14.4 0% 62.2 86.1 17.6061 3% 59.0 77.0 17.4 0% 64.5 85.2 14.4066 3% 59.0 77.1 14.6 0% 62.9 84.2 13.42219 T81 yes 42.0 58.0 6 Handbook Minimum for Extrusions 51.0 66.0 10 Typical______________________________________

Alloys 059, 066 and 069 were used as welding alloys to weld together 0.375 inch thick plates of various alloys presently disclosed. Table XIX lists the parent alloy and welding filler alloy, along with the welding technique and weld geometry utilized for each welding operation. The resultant weldments were subjected to various post-welding tempers as listed in Table XIX, and strength and elongation measurements were then made. It is apparent from the tensile properties given in Table XIX that the welding alloys of the present invention produce weldments having extremely favorable properties. The welding filler alloys used in the welding operations described in Table XIX were provided in the form of cut rod having a square cross section of 0.125 by 0.125 inch. The cut rod is produced by cutting strips from the extruded 0.375 by 4 inch plate of each of the alloys, followed by milling to yield a smooth surface. Alternatively, the welding alloys of the present invention may be provided in the form of drawn wire. The wire may be produced, for example, by extruding the ingot through a multiple hole die to a diameter of 0.375 inches with an extrusion ratio of 72:1 at a temperature of 690.degree. F. and a ram speed of 0.03 inches per second. The 0.375 inch extruded rod may then be drawn into wire using multiple drawing steps. A reduction in area of about 60 percent is suitable. Thus, for example, the 0.375 inch rod may be drawn in subsequent steps to approximately 0.25 inch, 0.15 inch, 0.09 inch and 0.625 inch diameters with an anneal at 800.degree. F. between each step. The 0.625 inch wire can also be drawn to smaller diameters, i.e., 0.05 and 0.035 inches. Thus, the production of weld wire having alloy compositions within the ranges presently disclosed may be accomplished.

While the use of the present welding alloys in welding parent alloys having compositions within the ranges presently disclosed is preferred, the present welding alloys may also be used to weld conventional aluminum-base alloys such as Al-Li alloys 2090 and 01420.

Forgeability

A casting of alloy 058 was extruded to 2 inch diameter rod at 358.degree. C. Sections of this rod were used as forging stock material to assess the forgeability of Al-Cu based alloys. Ten 4-inch long sections were cut from the rod, and were forged using a closed die for the arm torque shaft used on a fighter aircraft. The alloy exhibited excellent forgeability in this complex forging in that all ten forgings showed complete die filling and no tendency toward cracking over a range of forging temperatures from 275.degree. to 410.degree. C. A forging was subsequently solution heat treated at 504.degree. C. for 90 minutes, and water quenched at 20.degree. C. After 24 hours of natural aging, the forgings had a hardness of 77 R.sub.B, similar to that of the extruded bar. Furthermore, after aging at 160.degree. C. for 20 hours the hardness readings on the forgings ranged from 98 to 104 R.sub.B (ultimate tensile strength in ksi and hardness in R.sub.B roughly correspond one to one). This experiment demonstrates the exceptional potential of this alloy to achieve high strength in forgings without the benefit of cold work to stimulate precipitation during artificial aging. Other Al-Li alloys fail to meet strength goals in complex forgings because many forgings are not amenable to cold working, and these other alloys need cold work to reach high strength.

Quench Sensitivity

Quench sensitivity refers to the inability of a material to attain its maximum strength characteristics from precipitation processes when its cooling rate after solution heat treatment is decreased during quenching, i.e., effected by various liquid media or from heat transfer limitations through the bulk of the quenched piece. It is advantageous to have a material with low quench sensitivity where cooling rates are not critical. Quench "insensitivity" is valuable for weldments where it may not be feasible to re-heat-treat the alloy or where the weld bead and heat affected zones cool relatively slowly after welding.

As a preliminary assessment of quench sensitivity in the Al-Cu based alloy, a hardness profile was measured across a quenched two inch diameter rod (four foot length of extrusion 058). After solution heat treatment at 910.degree. F. for 1.5 hour, 20.degree. C. water quenching, the material was natural aged for 1400 hours, and Rockwell B measurements were taken in increments of 1/8 inch from the perifery toward the center of the extrusion as shown in Table XX. The maximum deviation in hardness is less than 4 R.sub.B indicating low quench sensitivity. Similar hardness profiles across the 2 in..times.3/4 in. extrusion 061 showed less than 2 R.sub.B points deviation through the thickness.

As a further evaluation of quench sensitivity, sections of alloy 049 were quenched in either water at room temperature, ice water (0.degree. C.), or boiling water (100.degree. C.) after solution heat treatment at 504.degree. C. FIG. 17 shows the natural aging response of the alloy resulting after quenching from these three different quenching rates. A strong natural aging response is noted for all three quench rates. At the longest recorded aging times, no more than six Rockwell Points separate the slowest (boiling water) and fastest (ice water) quenching rates while there is little difference between standard room temperature water quench and boiling water quench. Thus, Al-Cu-Mg-Ag-Li alloys appear to be sufficiently quench insensitive to have potential for restoring strength in weldments and forgings.

This result also indicates that the exceptional strength of Al-Cu based alloys in the naturally aged condition, e.g. 62.2 ksi YS and 86.1 ksi UTS for alloy 061, could be increased further by quenching into ice water. Indeed, the natural aging curves in FIG. 17 indicate that tensile strengths greater than 85 ksi can be obtained after an ice water quench, in about 1000 hours. This is technologically significant because quenching at 0.degree. C. is a practical plant operation.

Given the unexpectedly high properties of the alloys of the present invention, it was decided to make additions of silver or silver plus magnesium to recently commercialized aluminum-lithium alloys 2090, 8090, and 8091. The compositions of these alloys are in Table XXI. Additions were made of 0.4 weight percent silver to 8090 and 8091, and 0.4 weight percent silver plus 0.4 weight percent magnesium to 2090. These alloys were successfully cast and found to have improved properties.

TABLE XIX__________________________________________________________________________TENSILE PROPERTIES FROM WELDMENTSParent Welding Weld Post-Welding Weld YS UTS E1Alloy Filler Technique Geometry Temper Bead (ksi) (ksi) (%)__________________________________________________________________________059 059 AC TIG Butt NA (1000+ h) Off 43.9 47.3 1.2059 059 AC TIG Butt NA (1000+ h) Off 38.6 40.4 1.0059 059 AC TIG Butt NA (1000+ h) Off 39.3 46.8 1.0059 059 AC TIG Butt 160.degree. C. (24 h) Off 27.7 42.4 4.3059 059 AC TIG Butt 160.degree. C. (24 h) Off 29.2 44.6 1.7059 059 AC TIG Butt 160.degree. C. (24 h) Off 32.1 45.8 2.5059 059 AC TIG V SHT/WQ/160.degree. C. (24 h) Off 65.8 71.7 1.5059 059 AC TIG V SHT/WQ/160.degree. C. (24 h) Off 65.3 72.0 1.1059 059 AC TIG V SHT/WQ/160.degree. C. (24 h) Off 69.9 76.2 1.9059 059 AC TIG V SHT/WQ/160.degree. C. (24 h) Off 66.3 71.3 1.2059 059 AC TIG V SHT/WQ/160.degree. C. (24 h) Off 62.4 69.8 1.2059 059 AC TIG V SHT/WQ/160.degree. C. (24 h) Off 72.2 73.2 1.0059 059 AC TIG V NA (1000+ h) Off 42.0 49.7 1.0059 059 AC TIG V NA (1000+ h) Off 41.2 45.7 1.5059 059 AC TIG V NA (1000+ h) Off 40.1 46.2 2.1059 059 AC TIG V 160.degree. C. (24 h) Off 29.2 40.4 2.0059 059 AC TIG V 160.degree. C. (24 h) Off 29.8 39.4 1.8059 059 AC TIG V 160.degree. C. (24 h) Off 27.6 39.7 2.1069 066 DC TIG Butt NA (500 h) Off 36.3 38.0 0.3069 066 DC TIG Butt NA (500 h) On -- 40.9 0.2069 066 DC TIG Butt NA (1400 h) Off 36.4 46.0 2.2069 066 DC TIG Butt NA (1400 h) Off 37.3 48.4 3.3069 066 DC TIG Butt NA (790 h) On 38.3 45.3 1.7069 066 DC TIG Butt NA (790 h) On 36.3 44.1 1.3069 066 DC TIG Butt NA (790 h) On 34.3 40.7 1.3069 069 DC TIG Butt NA (360 h) Off 34.7 45.7 3.1069 069 DC TIG Butt NA (600 h) Off 36.5 42.0 1.5069 069 DC TIG Butt NA (1200 h) Off 36.1 43.9 2.2069 069 DC TIG Butt 200.degree. C. (24 h) Off 43.1 53.0 2.8069 069 DC TIG Butt SHT/WQ/NA (930 h) Off 49.3 63.0 8.1__________________________________________________________________________ TABLE XX______________________________________Hardness Profiles Across a Two Inch Diameter Rodof Extrusion 058Distance FromSurface Hardness (R.sub.B)(in.) Reading 1 2 3______________________________________1/8 78.5 79 792/8 81.0 80.5 80.53/8 81.0 81.0 81.04/8 80.5 81.0 80.55/8 79.5 80.5 80.56/8 79.0 79.0 79.07/8 77.5 78.5 79.01 77.5______________________________________ TABLE XXI______________________________________Modified Compositions of Commercial AlloysAlloy Cu Mg Li Ag Zr Al______________________________________2090 Modified 2.7 0.4 2.2 0.4 0.14 Balance8090 Modified 1.3 0.7 2.5 0.4 0.14 Balance8091 Modified 1.9 0.9 2.6 0.4 0.14 Balance______________________________________

The alloys of the present invention also possess excellent cryogenic properties. Tests performed at cryogenic temperatures reveal improvements in tensile and yield strengths. For example, Alloy 049 in the T8 temper at -196.degree. C. (-320.degree. F.) displays tensile properties of 105.8 ksi YS, 108.6 ksi UTS and 3.6 percent elongation. Alloy 066 in the T8 temper displays tensile properties of 101.2 ksi YS, 105.2 ksi UTS and 4.0 percent elongation at -68.degree. C. (-80.degree. F.), and tensile properties of 118.9 ksi YS, 124.7 ksi UTS and 3.2 percent elongation at -196.degree. C. (-320.degree. F.). Alloy 068 in the T8 temper at -196.degree. C. (-320.degree. F.) displays tensile properties of 113.6 ksi YS, 121.8 ksi UTS and 7.8 percent elongation. In the T6 temper, Alloy 68 at -196.degree. C. (-320.degree. F.) possesses tensile properties of 114.4 ksi YS, 119.0 ksi UTS and 2.5 percent elongation. The excellent cryogenic properties attained with the alloys of the present invention give important implications for space applications where cryogenic alloys are often necessary for fuel and oxidizer tankage, and where the tankage is joined by welding.

This invention thus provides a family of alloy compositions which are highly suitable for use as welding alloys and which provide exceptional strength--far greater than existing ingot metallurgy alloys of similar aluminum-based materials. The reasons for this ultra-high strength are not completely understood. Extensive transmission electron microscopy and selected area diffraction experiments failed to detect an omega phase in alloy 049, as might be expected from Polmear's work. It was observed that the T.sub.1 phase (Al.sub.2 CuLi) is exceptionally refined in alloy 049, which may help to explain the ultra-high strength. The alloys have lower density than similar alloys as a result of the addition of lithium. The alloys show surprisingly good ductility (in the naturally aged and underaged conditions), which is unexpected for an alloy containing lithium. Furthermore, the alloys have surprising and unexpectedly high strength in the naturally-aged condition, which makes them extremely promising for weldable and forging applications. The alloys have better combinations of strength and ductility than recently commercialized Al-Li alloys such as 2090, 2091, and 8090 (see FIG. 18). Welding trials indicate that the alloys of the present invention are extremely useful as welding alloys for joining together aluminum-base alloys including aluminum-lithium alloys.

It is to be understood that the above description of the present invention is susceptible to various modifications, changes, and adaptations by those skilled in the art, and that the same are to be considered to be within the spirit and scope of the invention as set forth by the claims which follow.

Claims

1. A welding alloy for welding aluminum-base alloys, the welding alloy consisting essentially of from about 3.5 to about 7 weight percent Cu, from about 0.05 to about 1.5 weight percent Mg, from about 0.01 to about 2 weight percent Ag, from about 0.1 to about 4 weight percent Li, from about 0.01 to about 2 weight percent grain refiner selected from the group consisting of Zr, Cr, Mn, Ti, B, V, Hf, TiB.sub.2 and combinations thereof, and the balance aluminum and incidental impurities.

2. A welding alloy according to claim 1, wherein the grain refiner is selected from Zr, Ti, or a combination thereof.

3. A welding alloy according to claim 1, wherein the Cu comprises from about 4 to about 6.5 weight percent.

4. A welding alloy according to claim 1, wherein the Mg comprises from about 0.05 to about 1 weight percent.

5. A welding alloy according to claim 1, wherein the Ag comprises from about 0.05 to about 1.2 weight percent.

6. A welding alloy according to claim 1, wherein the Li comprises from about 0.5 to about 2.6 weight percent.

7. A welding alloy according to claim 1, wherein the grain refiner comprises from about 0.05 to about 1 weight percent.

8. A welding alloy for welding aluminum-base alloys, the welding alloy consisting essentially of from about 4 to about 6.5 weight percent Cu, from about 0.05 to about 1 weight percent Mg, from about 0.05 to about 1.2 weight percent Ag, from about 0.5 to about 2.6 weight percent Li, from about 0.05 to about 1 weight percent grain refiner selected from the group consisting of Zr, Cr, Mn, Ti, B, V, Hf, TiB.sub.2 and combinations thereof, and the balance aluminum and incidental impurities.

9. A welding alloy according to claim 8, wherein the grain refiner is selected from Zr, Ti, or a combination thereof.

10. A welding alloy according to claim 8, wherein the Cu comprises from about 4.5 to about 5.5 weight percent.

11. A welding alloy according to claim 8, wherein the Mg comprises from about 0.1 to about 0.8 weight percent.

12. A welding alloy according to claim 8, wherein the Ag comprises from about 0.1 to about 1 weight percent.

13. A welding alloy according to claim 8, wherein the Li comprises from about 0.8 to about 1.8 weight percent.

14. A welding alloy according to claim 8, wherein the grain refiner comprises from about 0.05 to about 0.5 weight percent.

15. A welding alloy for welding aluminum-base alloys, the welding alloy consisting essentially of from about 4.5 to about 5.5 weight percent Cu, from about 0.1 to about 0.8 weight percent Mg, from about 0.1 to about 1 weight percent Ag, from about 0.8 to about 1.8 weight percent Li, from about 0.05 to about 0.5 weight percent grain refiner selected from the group consisting of Zr, Cr, Mn, Ti, B, V, Hf, TiB.sub.2 and combinations thereof, and the balance aluminum and incidental impurities.

16. A welding alloy according to claim 15, wherein the grain refiner is selected from Zr, Ti, or a combination thereof.

17. A welding alloy according to claim 15, wherein the Cu comprises from about 4.8 to about 5.2 weight percent.

18. A welding alloy according to claim 15, wherein the Mg comprises from about 0.2 to about 0.5 weight percent.

19. A welding alloy according to claim 15, wherein the Ag comprises from about 0.3 to about 0.5 weight percent.

20. A welding alloy according to claim 15, wherein the Li comprises from about 1.0 to about 1.6 weight percent.

21. A welding alloy according to claim 15, wherein the grain refiner comprises from about 0.08 to about 0.2 weight percent.

22. A welding alloy for welding aluminum-base alloys, the welding alloy consisting essentially of from about 4.8 to about 5.2 weight percent Cu, from about 0.2 to about 0.5 weight percent Mg, from about 0.3 to about 0.5 weight percent Ag, from about 1.0 to about 1.6 weight percent Li, from about 0.08 to about 0.2 weight percent grain refiner selected from the group consisting of Zr, Cr, Mn, Ti, B, V, Hf, TiB.sub.2 and combinations thereof, and the balance aluminum and incidental impurities.

23. A welding alloy according to claim 22, wherein the grain refiner is selected from Zr, Ti, or a combination thereof.

24. A weld wire having a composition consisting essentially of from about 3.5 to about 7 weight percent Cu, from about 0.05 to about 1.5 weight percent Mg, from about 0.01 to about 2 weight percent Ag, from about 0.1 to about 4 weight percent Li, from about 0.01 to about 2 weight percent grain refiner selected from the group consisting of Zr, Cr, Mn, Ti, B, V, Hf, TiB.sub.2 and combinations thereof, and the balance aluminum and incidental impurities.

25. A weld wire according to claim 24, wherein the Cu comprises from about 4 to about 6.5 weight percent, the Mg comprises from about 0.05 to about 1 weight percent, the Ag comprises from about 0.05 to about 1.2 weight percent, the Li comprises from about 0.5 to about to about 1 weight percent.

26. A weld wire according to claim 24, wherein the Cu comprises from about 4.5 to about 5.5 weight percent, the Mg comprises from about 0.1 to about 0.8 weight percent, the Ag comprises from about 0.1 to about 1 weight percent, the Li comprises from about 0.8 to about 1.8 weight percent, and the grain refiner comprises from about 0.05 to about 0.5 weight percent.

27. A weld wire according to claim 24, wherein the Cu comprises from about 4.8 to about 5.2 weight percent, the Mg comprises from about 0.2 to about 0.5 weight percent, the Ag comprises from about 0.3 to about 0.5 weight percent, the Li comprises from about 1.0 to about 1.6 weight percent, and the grain refiner comprises from about 0.08 to about 0.2 weight percent.

28. A weld wire according to claim 24, wherein the grain refiner comprises Zr, Ti, or a combination thereof.