Journal of Research of the National Bureau of Standards Vol. 49, No. 3, September 1952 Research Paper 2354

Influence of Prior Strain History on the Tensile Propertiesand Structures of High-Purity Copper

William D. Jenkins and Thomas G. Digges

Tensile tests were made at room temperature on oxygen-free high-conductivity (OFHC)copper initially as annealed, as cold-drawn different amounts, and as prestrained in creepat 110°, 250°, and 300° F.

The shape of the true stress-strain curves obtained on copper cold-drawn 34-, 40-, and 70-percent reduction in area indicate a strain aging effect in the specimens when subjected tostresses in the vicinity of the maximum load. This phenomenon of strain aging is usuallymore closely associated with alloys of the ferrous type than with a high-purity nonferrousmetal. The present results show that the strength, ductility, and hardness of the copper ismarkedly affected by its prior strain history. The test conditions are correlated with thetensile properties, hardness, necking characteristics, formation of microcracks, and sub-structures.

1. IntroductionThe tensile and other properties at room tempera-

ture of both annealed and cold-drawn high-puritycopper were determined as a part of the Bureau'sresearch investigations of the mechanism of creepand of the effects of subzero temperatures on themechanical properties of metals and alloys. Thecreep behavior of this same lot of copper as annealedand as cold-drawn 40-percent reduction in area andthe influence of low temperatures on the true stress-strain relation in tension of the annealed copper havebeen discussed in previous publications [1, 2, 3.1]The resistance to creep and to fracture at moderatelyelevated temperatures was materially increased bycold-drawing the copper, but this superiority was ac-complished by a decrease in ductility. The strainhistory of the copper also affected the degree of dis-sociation of the parent grains into subsize grains

during creep and the type of fracture. However,the strength in tension of the annealed copper in-creased continuously with a decrease in temperatureto —320° P without any impairment of its ductility.

The present tests were made to evaluate the effectsof prestraining different amounts in creep undertension at 110°, 250°, and 300° F and of cold-draw-ing on the tensile properties at room temperatureand on the hardness and structures produced in thefractured specimens. The straining treatments usedprior to testing the specimens in tension at roomtemperature are summarized in table 1.

2. Material and Procedures

All the specimens were prepared from 13/16- or7/8-in. diameter bars processed from one lot ofoxygen-free high-conductivity (OFHC) copper con-taining 99.99+ percent of copper. The four bars

TABLE 1. Strain history of specimens prior to testing in tension at room temperature and reference to the figures used in the textfor summarizing the test data

Initial condition

Al" annealed 800° FDoDo -DoDo

Al; annealed 800° F; strained 9% in tension at room tempera-ture -

A2* cold-drawn 34% reduction in area- _

A3* oold-drawn 40% reduction in areaDoDo --Do -

Bl* cold-drawn 70% reduction in area

Bl; cold-drawn 70% reduction in area; then annealed 1300° F .

Creep test

Temper-ature

opNone

110110250250

250

None

Do250300

/no1300

None

do

Rate

%/tOOOhrNone2.751.0810.11.04

1.1

None

do0.20

.36

.63200+

None

do

Plastic ex-tension

None39.719.049.023.1

7.0

None

do0.73

.98

.7312.8 /

None

_ do

Figure reference

Tensile propertiesand hardness

1, 2, 3, 5, 7, 18, 193, 5, 191918, 193, 5, 18, 19

1, 2, 18, 19

1, 2, 4, 5, 18, 194, 5, 18, 194. 5, 18, 194, 5, 18, 19

1, 2, 18, 19

7, 8, 18, 19

Structures

6A9B, 11A, 12A, 14A, 17C9C12D, 14D9D, 10B, 11B, 12C, 14C, 16A, 17D.

12B, 14B

6B, 13C

11C, 15A, 17E9E, 11D, 13A, 15B, 16B9F, 10C, HE, 15C, 17F9G, 11F, 13B, 15D

6C, 13D

6D, 9A, 10A, 16C, 16D, 17A, 17B

-Figures in brackets indicate the literature references at the end of this paper.

217025—52 3 167

used were procured in conditions as follows:A. Bright annealed at 800° F for 1 hr:

1. As annealed 0.025-mm averagegrain diameter.

2. Cold-drawn 34-percent reductionin area.

3. Cold-drawn 40-percent reductionin area.

B. Bright annealed at 1150° F for 50 min (0.045-mmaverage grain diameter):

1. Cold-drawn___ 70-percent reductionin area.

Tensile specimens, 0.505 in. in diameter with 2-in.gage length, were prepared from the above bars.In addition, a portion of the bar cold-drawn 7Gpercent (condition Bl) was reannealed in the labora-tory at 1,300° F in air for 1 hr, thereby producingan average grain diameter of 0.120 mm, beforepreparing the tensile specimen. Some tensile speci-mens prepared from the bars as annealed at 800° Fand as cold-drawn 40 percent (conditions Al and 3,respectively) were strained in creep at varioustemperatures before fracturing them in tension atroom temperature.

The tensile tests were made at room temperatureat a rate of approximately 1-percent reduction ofarea per minute. A micrometer was used to followthe change in diameter of the tensile specimen duringthe testing. Simultaneous observations were madeof the minimum diameters and loads.

Two flats were prepared approximately 0.2 in.apart, diametrically opposite, symmetrical to andparallel to the longitudinal axis of the fracturedspecimen. Rockwell hardness (F scale, 60-kg load,i/16-in.-diameter ball) readings were made at variouspoints along the center line of each flat. Specimenswere also cut from the bar cold-drawn 70 percentfor use in determining the effect of annealing tem-perature on its grain size and hardness.

3. Results and Discussion3.1. Effect of prestraining on tensile properties

a. Cold-drawing at room temperature

The relations between true stress and the totaltrue strain 2 (due to cold-drawing and tension) forspecimens of copper as annealed and as cold-drawndifferent amounts are shown in figure 1. The truestress at the beginning of plastic deformation intension, and the true stress and true strain at maxi-mum load increased with an increase in the amountof cold-drawing. The values for true stress at thebeginning of fracture in each specimen as annealed,or as cold-drawn 40 or 70 percent, were nearly alikebut cracking started at a somewhat lower valueof stress in the specimen cold-drawn only 34 percent.However, the total true strain at the beginning offracture of this specimen was about the same asthat of the specimen as annealed or as cold-drawn40 percent. In this series of tests the total strain

2 "True strain" is defined as the natural logarithm of the ratio of the initialcross-sectional area (AQ) of the specimen to its current minimum cross-sectionalarea 04).

at the beginning of fracture attained a maximumin the specimen cold-drawn 70 percent.

The observed differences in the relative positionsand slopes of the true stress-strain curves can bepartly ascribed to variations in tbe stress systemduring plastic deformation in tension in additionto variations in tbe degree of strain hardening in-duced by cold-drawing. For the annealed specimen,the deformation was essentially by unidirectionaltension for stresses up to the maximum load. Asthe specimen began to neck at this point, furtherdeformation was under a triaxial stress system.This triaxiality results in modifying the positionof the curve to higher values of stresses for similarstrains than would have been the case had deforma-tion to complete fracture been under a unidirectionalsystem. Similarly, necking began at maximum loadin each of the cold-drawn specimens and thereafterdeformation was by triaxiality. The effect of.tri-axiality, however, varied with the amount of cold-working and thereby affected the stress-strainrelationship. Furthermore, some triaxiality alsoexisted during cold-drawing, and this is a factorthat would be expected to decrease the value ofthe stress at the maximum load. Obviously, somestrain hardening occurred during cold-drawing, andthe capacity of the copper to further strain hardeningwas thereby reduced.

The effect of cold-drawing on the hardness andtensile properties of the copper is shown by the re-sults given in figure 2. The yield strength, tensilestrength and hardness increased at a decreasing rate,with an increase in the amount of cold-drawing. .The plastic extension in tension, however, was mater-ially decreased by cold-drawing, but the magnitudesof the decrease in plastic extension and of the increasein hardness were not significantly affected by varyingthe degree of cold-working from 34 to 70 percent.For example, the plastic extension at maximum loaddid not exceed 1 percent in any of the cold-drawnspecimens but was about 30 percent in the annealedspecimen. Thus the improvements in strengths"and in hardness by cold-drawing the copper wereaccompanied by an impairment in its plastic exten-sion in tension at room temperature.

b. Cold-working in creep at different temperatures

(1) Initially as annealed. The influence of strain-ing specimens of the annealed copper in creep at 110°and 250° F on the true stress-true strain relation atroom temperature is shown by the curves in figure 3.The curves for the specimens prestrained in creep fallsomewhat below that of the annealed specimen.However, the true stress at maximum load (point M)of the annealed copper was increased slightly by pre-straining 23.1 percent in creep at 250° F and further *increased in another specimen strained 39.7 percentin creep at 110° F before fracturing in tension atroom temperature. The curves for stresses from themaximum load to the beginning of fracture for thetwo specimens prestrained in creep were nearly alike,but the specimen prestrained at 250° F began tofracture at considerably lower stress and strain than

168

did the specimen prestrained in creep at 110° F.The stress required to initiate fracture in the latterspecimen was only slightly less than that of theannealed specimen and the corresponding totalstrains at this point were nearly alike for the twospecimens. Although the specimen prestrained at110° F was strained into the third stage of creep, thecurve indicates that its mode of deformation in ten-sion at room temperature was similar to that of theannealed specimen; however, straining the specimeninto the second stage of creep at 250° F was believedto be sufficient to initiate cracks of submicroscopicdimensions, and these possibly are partly responsiblefor its restricted ductility at room temperature.These assumptions are supported by the evidenceobtained from a rhetallographic examination of thefractured specimens. In the specimen prestrained23.1 percent at 250° F, numerous cracks were ob-served in the region adjacent to and at some distance(unnecked portion of the specimen) from completefracture (fig. 12C), whereas in the specimens pre-strained 39.7 percent at 110° F, the microcrackingwas confined to the region of complete fracture(fig. 12A); the mode of fracture of the latter specimenwas similar to that previously described for an an-nealed specimen [1].

Straining the specimen 23.1 percent in creep at250° F also lowered appreciably both the true stressand the strain at complete fracture in relation tothose of the specimens as annealed or as prestrained39.7 percent at 110° F.

(2) Initially as annealed and then cold-drawn 40percent. The effect of prestraining the cold-drawncopper in creep at different temperatures on sometensile properties at room temperature is shown infigure 4. The strain in creep at moderately elevatedtemperatures resulted in some modification of thepositions of the true stress-strain curves at roomtemperature for values of true stresses below 65,000psi. In this range of stresses, the curves for thespecimens prestrained in creep fall somewhat belowthat of the specimen as cold-drawn. At true stressesof 65,000 psi, or higher, the true stress-strain curveof the specimen strained 0.73 percent at 110° F, andthen 2.8 percent at 300° F nearly coincided with thatof the specimen as cold-drawn. The values for truestress at maximum load, and true stress and strainat the beginning of fracture were also similar forthese two specimens, but the values for stress andstrain at complete fracture were lower in the specimenstrained in creep.

Prestraining a specimen 0.73 percent at 250° F andanother specimen 0.98 percent at 300° F loweredsomewhat the true stresses at maximum load andappreciably decreased the true stresses and strainsat the beginning and at complete fracture. Thedecrease in stresses and strains at fracture was con-siderably more pronounced in the specimen pre-strained at 300° F than in the specimen prestrainedat 250° F. The values of stress and strain at com-plete fracture of the former specimen were lower thanthe corresponding values at the beginning of fracturefor the latter specimen. This deterioration, as shown

in the tensile properties at room temperature, isbelieved to be due partly to the initiation of cracksof submicroscopic or possibly microscopic dimension(fig. 15, C) by the presence of tensile stresses inducedin creep. Moreover,, the tendency towards brittlefracture was increased by increasing the temperatureof the creep test from 250° to 300° F and adjustingthe stresses to produced strain rates of 0.2 to 0.4percent/1,000 hr. (figs. 11, D and E).

The true stress-strain curves at small strains ofthe annealed and cold-drawn copper specimens arereproduced on an enlarged scale in figure 5. The-reversal of curvature of the curve for the cold-drawn specimen not prestrained in creep indicatesthat an aging effect took place subsequent to thecold-drawing operation (fig. 5, A). This initialeffect is in contrast to the recovery behavior usuallyassociated with nonferrous metals such as aluminum[4] after cold-extending and appears to be moreclosely associated with the strain-aging behavior ofalloys of the ferrous type [5]. The residual stressesdue to cold-drawing and aging of the copper arepartly relieved after the specimen has been deformedslightly past its maximum load. This caused alowering of the true stress values over a limitedrange of increasing strains. This effect of aging isless pronounced but still present, as is shown by thepositions of the true stress-strain curves for the twocold-drawn specimens prestrained in creep at 250°and 300° F to true strains of less than 0.01. Therelief of the causes of the aging phenomenon appearsto be more pronounced for the specimen prestrainedat the higher temperature even though the creepstrain was slightly less. The true stress-strain valuesfor the specimen prestrained in creep at 110° andat 300° F to a higher strain value show a pronouncedrecovery characteristic, as evidenced by a decreasein true stress values at small strains. This effect,however, was followed by rapid strain-hardening incontrast to the other two specimens whose rate ofstrain-hardening was approximately equal to thatof the cold-drawn specimen in the region from theinflection range up to a true strain value of about0.15. A relief of internal stress followed by rapidstrain-hardening is also shown by the position of thetrue stress-strain curve for the annealed specimenthat was prestrained in creep at 250° F, whereasaging appeared to predominate for the specimenprestrained in creep at the lower temperature (fig. 5,B). The recovery phenomenon associated withstraining in creep above the temperature that wasused in the tensile tests is further illustrated by thevalues labeled ei, e2, etc. in figure 5. The brokenlines intersecting the true stress-strain curves at ex,e3, e5, . . . are in each case the extrapolation of thelinear portion of the corresponding true stress-straincurve for a specimen prestrained in creep. Thepoints e2, e4, e6 represent the strains actually obtainedat the corresponding stress. In spite of the factthat the true stress-strain curve was lowered con-siderably for the specimens prestrained at 110° and300° F to about 0.09 strain (fig. 5, A), the attainmentof the stress at ei0, corresponding to the stress of the

169

cold-drawn copper at e9, took place at a much smallerchange in strain than similar attainments for theother two specimens (e8 and e7, and e6 and e5). How-ever, prestraining in creep to different strains at.different temperatures had no appreciable effect onthe initial recovery characteristics of the annealedspecimens when the relationships are evaluated bythis means. A comparison of the properties at smallstrains with the true stress-strain relationships,shown by figures 3 and 4, indicates that the instanta-neous strain state of the material does not neces-sarily predict the subsequent flow properties.

Burghoff and Blank [6] extended cold-drawn spec-imens of oxygen-free copper in creep at 300° and400° F prior to testing in tension at room temperature.The tensile strength was materially decreased andthe elongation was increased in specimens initiallycold-drawn 84 percent after exposure of 6,500 hr at300° or 400° F, with zero or small stresses. Partialor complete recrystallization occurred in the spec-imens exposed at 300° or 400° F.

Some strain-aging effects, associated with OFHCcopper have recently been discussed by Lubahn [7].From results of room-temperature tensile tests, inter-rupted by various heat treatments, he suggests thatthe aging process may be associated with (a) yield-point phenomenon, (b) increase in flow stress, (c)discontinuous yielding, and (d) abnormally low ratesensitivity.

3.2. Effect of grain size

A change from 0.025 to 0.120 mm in the averagediameter of the grains in the annealed copper (fig. 6,A and D) caused a decrease in the yield strength(0.2% offset) from about 11,000 to 6,000 psi, aslight lowering of the true stress-strain curve forstresses in the region of strain below 0.15, and a de-crease in the stess and the strain at the beginning offracture (fig. 7). However, the tensile strength andthe strain at this stress were not affected by theincrease in grain size.

The effect of annealing temperature (1 hr at tem-perature) on the grain size and hardness of the coppercold-drawn 70 percent from an average grain diam-eter of 0.045 mm (condition Bl), is shown in figure8. Increasing the annealing temperature from 800°to 1,100° F resulted in a slight increase in grain size.The size of the grains continued to increase graduallyas the temperature was further increased to 1,200° Fand then increased markedly at an annealing tem-perature of 1,300° F. This grain growth and recoverywas accompanied by a corresponding decrease inhardness of the recrystallized specimens.

The features observed in necking and fracturingthe specimens in tension at room temperature can beconveniently classified into three groups, as illustratedin figure 10, A, B, and C. All the specimens frac-tured at room temperature were of the ductile, orfibrous, type, but the values for strain and reductionof area varied appreciably with the prior-strain his-tory. Roughing of the surface, necking down tonearly a point and then fracturing at approximatelyright angles to the longitudinal axis of the specimen

(fig. 10, A) were characteristics of all specimens notstrained prior to testing in tension at room temper-ature. Possibly the relatively large grains contrib-uted to the excessive roughing of the surface inthe region of the fracture of this particular speci-men. The appearance of the surface at and near thefracture of the specimens prestrained by cold-drawingand in creep is represented by the specimen shownin figure 10, C. Some roughening of the surface isevident in this specimen, but it is considerably lesspronounced than that of the specimen not prestrainedin creep (fig. 10, A). The fractures of the prestrainedspecimens were usually in a plane about 45° to 60°to the longitudinal axis. No macrocracks were ob-served in the surfaces near complete fracture of thespecimens shown in fig. 10, A and C, but there wassome evidence of their presence in B. Moreover,macrocracks were prominent in an initially annealedspecimen after being strained to complete fracturein creep at 300° F (figure 10, D).

An end view of some of the fractured specimens isgiven in figure 11. The cavities located in the centerregions and their relative absence near the outer sur-face support the belief that cracking was initiatednear the axis and the outer surface was the position offinal rupture.

3.3. Effect of prestraining on structures

a. Necking and fracturing characteristics

Figures 9, 10, and 11 are photographs of some ofthe specimens selected as representative of the neck-ing and fracture characteristics as affected by prior-strain history. As all of the specimens showed somenecking (fig. 9), it is believed that cracking eventuallyleading to failure started at the axis of each specimenand progressed outward, and the final fracture wasthat of the "rim" at the surface. However, the de-gree of necking and the magnitude of the rim effectof the specimens were influenced by the amount,rate, and temperature of prestraining. For example,an initially annealed specimen prestrained 39.7 per-cent extension in creep at 110° F developed both arelatively pronounced neck (fig. 9, B) and rim (fig.11, A), whereas the specimen cold-drawn 40-percentreduction of area and then strained in creep only 0.98percent at 300° F developed a relatively small neck(fig. 9, F), a small rim (fig. 10, C and fig. 11, E), andfractured in a relatively brittle manner. As thesetwo specimens were processed from the same lot ofcopper and tested in tension at room temperature atthe same strain rate, the defference in ductility atfracture must be attributed to changes brought aboutby the different methods of prestraining.

b. Rim formation and microcracks ^

The influence of prestraining on the formation ofmicrocracks and the flow characteristics at the rimin the region of complete fracture in tension at roomtemperature is shown by the photomicrographs offigures 12 and 13. (No microcracks were evidentnear the surface or axis of the annealed specimensnot prestrained in creep.) Furthermore, it was

170

possible to strain a specimen at 110°F into the thirdstage of creep without the development of numerousmicrocracks, except in the vicinity of complete frac-ture, in subsequent fracturing at room temperature(fig. 12, A). " Microcracks were not observed inanother specimen (fig. 12, B) prestrained only intothe second stage of creep at a somewhat highertemperature (250° F). However, a general disinteg-ration is shown in two other specimens that wereprestrained at 250° F into either the second or thirdstage of creep (fig. 12, C and D). As previouslypointed out [2], the presence or absence of micro-cracks in specimens fractured solely in creep dependedupon the initial condition of the copper, the creeprate, temperature, and amount of extension in creep;relatively slow creep rates, high temperatures, andincrease in amount of extension into the third stageof creep favored a general disintegration.

Relatively few microcracks are evident near thesurface of the copper that was cold-drawn beforetesting in tension (fig. 13) even when the specimenswere prestrained into third stage of creep (fig. 13, Aand B). The effect of prestraining at the highertemperature (fig. 13, B) seems to promote a grainrecovery effect causing a healing of the broken bondsto take place and thus a restoration of ductility andstrength (fig. 4). Limited grain fragmentation andthe alinement of the grains in a wavy manner resultsfrom cold-drawing the copper 34-percent reductionof area prior to testing in tension (fig. 13, C), whereasfurther cold-drawing to 70-percent reduction in area(fig. 13, D) appeared to result in considerable grainfragmentation and an increase in tensile strength andin total ductility at fracture (fig. 1).

c. Microcracks and structures at axis

Structures at the surface do not necessarily reflectthe flow characteristics in the interior of these speci-mens as the stress systems in these regions are dif-ferent. However, as shown in figures 14 and 15,microcracks are evident near the axis of all the speci-mens that showed the tendency toward crack forma-tion near the surface (figs. 12 and 13). Relativelyfew cracks are visible in the annealed specimen pre-strained at 110° F to a large strain (fig. 14, A) or at250° F to a small strain (fig. 14, B). A number ofof cracks are evident at the axis of the specimen pre-strained in creep into the second stage at 250° Fwith a slow creep rate (fig. 14, C) and of another spec-imen prestrained into the third stage of creep at thesame temperature with a faster creep rate (fig. 14, D).However, the contour of the cracks is different, prob-ably due to the different mechanisms of crack nucle-ation in creep and subsequent growth in the tensiletest.

The effect of prior straining on the formation ofmicrocracks at the axis of the cold-drawn copper isshown in figure 15. Some microcracks of ellipticalshape are evident in specimens not strained in creep(fig. 15, A), but these are much smaller than thecracks in a specimen prestrained in creep at 250° Fto a small strain (fig. 15, B). The type of fracture

and the shape of the microcracks appear to be af-fected significantly by raising the creep test tempera-ture (fig. 15, C) without appreciable change in creeprate or in plastic strain. Straining in creep at ahigher temperature (300° F) and at a faster rateafter prestraining at 110° F to a small strain valueseems to cause an effect that is reflected in the ab-sence of microcracks even in the region of completefracture at room temperature (fig. 15, D).

A further indication of the effects of prior historyon the internal structure and the initiation andgrowth of microcracks is shown in figure 16. Thepresence of microcracks in the regions representativeof structure at the limit of uniform strain (fig. 16,A and B, several cracks are enclosed in circles) indi-cates that a partial disintegration was evidentthroughout these specimens. As no microcrackshave been found in any of the copper specimenspreviously strained into the second stage of creep [1,2], it is believed that the conditions for crack forma-tion are initiated during the second stage of creep,and with a continued straining in tension at roomtemperature the cracks grow to such a size that theflow properties of the copper are affected. Repre-sentative structure in the unnecked portion of thespecimens not tested in creep is shown in figure 16,C and D. No microcracks were evident in thisregion of either this specimen or of any of the speci-mens not prestrained under creep conditions. Therelative straightness of portions of the twins and thedifference in orientation between the twins and thesurrounding material (fig. 16, D) indicate that theseare annealing twins that were nucleated at a grainboundary (lower arrow) and grew toward anotherboundary (upper arrow) of the grains. Duringdeformation portions of the annealing twins werebent, thus indicating that they are ductile (fig. 16,C).

d. Substructures

The effect of prestraining on the formation ofsubstructures is shown in figure 17. Etch patternsrepresentative of the initially annealed structure areshown in figure 17, A. It has been previouslypointed out [8] that the size of these substructures isinfluenced by the temperature of annealing and thatthe tendency toward formation of substructures ofsmaller size during deformation increased withdecrease in test temperature and with increase increep rate [1, 2]. However, the latter observationswere based on tests carried to completion in creeptests only. The present tests show that the sizes ofthe substructures in the copper were materiallyreduced by testing in tension at room temperatureonly (fig. 17, A and B). It is also evident that boththe size and distribution of the substructures areaffected by the temperature, creep rate, or strain increep prior to the tensile test (figs. 17, C, D, E, andF). Both high temperatures and slow strain ratesappear to increase the tendency toward the forma-tion of a large number of these substructures.

171

3.4. Effect of prestraining on specimen contour andon hardness at room temperature

The relation of reduction of area to distancefrom the fractured surface and to hardness at roomtemperature, of specimens tested in tension at roomtemperature, as affected by prestraining, is shownin figures 18 and 19. The distribution of ductilityas measured by the reduction of area values alongthe specimen was materially affected by cold-drawing the copper prior to testing in tension(fig. 18, A) but increasing the amount of cold-draw-ing from 34- to 70-percent reduction of area had noappreciable effects on the final contour of the surfaceof the specimen. Annealing at 1,300° F appearedto restore the original necking characteristics sothat i:he curve nearly coincided with the curveobtained with the specimen annealed at 800° F(fig. 18, B). However, when specimens cut fromthe latter bar were strained in creep before theroom-temperature tensile test, a more brittle typeof fracture, as shown by decrease in sharpness ofthe neck (fig. 18, C), was obtained: This tendencytoward brittle fracture was also evident in the cold-drawn specimens after straining in creep (fig. 18, D),and was promoted by an increase in creep tempera-ture or by a decrease in creep rate.

The effect of prestraining on the relation betweenhardness at room temperature and reduction ofarea is shown in figure 19. The trend was for thehardness to increase with increase in reduction ofarea up to values of about 35 percent for the ini-tially cold-drawn and 50 percent for the initiallyannealed specimens. Thereafter, the hardness eitherremained constant or decreased with an increasein reduction of area. The specimen initially cold-drawn 40 percent (fig. 19, A) had numerous cracksin the region corresponding to 50-percent reductionof area (fig. 15, A). Reannealing at 1,300 °Fafter cold-drawing 70 percent caused a loweringof the hardness, at all strains, to values approxi-mately the same as those of the specimen initiallyannealed at 800° F (fig. 19, B). The averagegrain diameter of the specimen annealed at 1,300° Fwas about five times that of the specimen annealedat 800° F (0.120 and .025 mm, respectively). Strain-ing in creep at 110° F to an extension of 19 percenthad no appreciable effect in lowering the hardness-reduction of area curve of the initially annealedspecimens (fig. 19, C). However, the relativepositions of the hardness-reduction of area curvesand the peak-hardness values were altered bystraining in creep at higher temperatures or tohigher strain values (fig. 19, C and D). Severalfactors, such as differences in the stress systems, sizeand distribution of the microcracks and substruc-tures, appear to influence the hardness values ofthe copper.

4. Summary

Tensile tests were made at room temperature onOFHC copper initially as annealed, as cold-drawn,and as prestrained under creep conditions. Micro-

scopic examinations and hardness tests were madeat room temperature to ascertain the effects of theprior thermal-strain history on the structure andhardness.

Yield strength, tensile strength, initial hardness,true stress at maximum load, increased with increasein prestraining by cold-drawing; this superiority inthe strength and hardness properties was accom-panied by a decrease in ductility at maximum loadand at fracture.

The general trend was for a decrease in the flowstress, fracture stress, and ductility as the pre-straining temperature in creep increased and thecreep rate decreased.

Aging, due to prestraining, and recovery wereevidenced by the shape of the true stress-straincurves.

The degree of necking and the tendency for theformation of a rim at fracture increased with adecrease in prestraining temperature or with anincrease in prestraining rate. However, the tenden-cy for recovery, disintegration by microcracking, andthe size of substructure were increased as the pre-straining temperature was increased. Microcrackingalso was increased by decreasing the prestrainingrate in creep.

An increase in grain size tended to lower both thehardness and yield strength. The temperature andrate of prestraining also materially affected thehardness.

The authors gratefully acknowledge their indebt-edness to C. R. Johnson for his assistance in thetest program.

5. References

[1] W. D. Jenkins and T. G. Digges, Creep of high-puritycopper, J. Research NBS 45, 153 (1950) RP2121.

[2] W. D. Jenkins and T. G. Digges, Creep of annealed andcold-drawn high-purity copper, J. Research NBS 47,272 (1951) RP2254.

[3] G. W. Geil and N. L. Carwile, Tensile properties ofcopper, nickel, and some copper-nickel alloys at lowtemperatures, NBS Circular 520 (1952).

[4] W. D. Jenkins, Creep of high-purity aluminum, J.Research, NBS 46, 310 (1951) RP2201.

[5] D. J. McAdam, Jr., G. W. Geil, D. H. Woodard andW. D. Jenkins, Influence of strain aging on the fracturestress of low carbon steel, Trans. AIME 176, 436 (1948).

[6] H. L. Burghoff and A. I. Blank, The creep characteristicsof copper and some copper alloys at 300, 400, and500° F, Proc. Am. Soc. Testing Materials 47, 725(1947).

[7] J. D. Lubahri, Strain aging effects, Trans. Am. Soc.Metals 44, 643 (1952).

[8] P. Lacombe and L. Beaujard, The application of etchfigures on pure aluminum to the study of some micro-graphic problems, J. Inst. Metals LXXIV, 1 (1948).

172

"A

t

Afy

I

•ft*.

\ //

4

p'RT

M--MAXIMUM LOADRT-BEGINNING OF FRACTURE

SYMBOL

oA••

D

REDUCTION OF AREA BYCOLD-DRAWING, PERCENT

344070

0.5 1.0 1.5TRUE STRAIN, LOGe Ao/A

FIGURE 1. Effect of cold-drawing on the true stress-strainrelation at room temperature of copper.

The strains corresponding to cold-drawing are included in the plotted values.

Y--YIELD STRENGTH (.2 PERCENT

OFFSET)U--TENSILE STRENGTHM--MAXIMUM LOADRT-BEGINNING OF FRACTURER--COMPLETE FRACTURE

60000 -CO

in<n

40000 ^

20000 |

20 40 60 20 40 60 80REDUCTION OF AREA BY COLD-DRAWING. PERCENT

FIGURE 2. Effect of cold-drawing on the hardness and tensileproperties at room temperature of copper.

The hardness tests were made oh cross sections of the bars initially as annealedand as cold-drawn.

4 0 z h

173

90000

80000

o: 60000

40000

M ^

/ A

I

1 A

/

% - • • •

•

•

/

A

"̂ M

/ A "

/ •

A

j

AI'm

ANNEALEDM MAXIMUM

R,R,,R2-COMPLETE

R

/

i

/

LOAD

FRACTURER, = STRESS, I5OOOO LB/ lN2 ; STRAINR2* STRESS, 139000 LB/lN2; STRAIN

SYMBOL

O

•A

CREEP TEST

• F

NONE

110

250

2nd. STAGE

%/lOOO HR

NONE

2.75

1.04

2.18

2.11

CREEP

t

NONE

39.7

23.1

0.5 1.0

TRUE STRAIN, LOGe Ao/A

1.5 2.0

FIGURE 3. Effect of prestraining in creep on the true stress-strain relation at room temperature of annealed copper.

80000

70000

60000

50000

40000

/ a

A

A

A

/ ' *

/ A/ TD

/ A COLA M-"

A*

D-DRAWN-MAXIMUM L

40%OAD

/

ApR

RE DUC

I

TION OF AF

R 3

' -

EA

Rr STRESS, 90 00 LB/lN? STRAIN, 1.033R2= STRESS, 125400 LB/lN2;STRAIN, 1.479

SYMBOL

O

A

T

a

CREEP TEST

TEMPERATURE,

NONE

3 0 0

2 5 0

3 0 0

2nd STAGE

CREEP RATE.

%/lOOO MR

NONE

0.63 a>200

0.20

0.36

CREEP

EXTENSION.

NONE

0.73 82.8

0.73

0.98

0 0-2 0.4 0.6 0.8 1.0 12TRUE STRAIN, LOGe Ao/A

FIGURE 4. Effect of prestraining in creep on the true stress-strain relation at room temperature of cold-drawn copper.

' The strain corresponding to cold-drawing 40-percent reduction in area is notincluded in the plotted value.

TRUE STRAIN, LOGe Ao/A

FIGURE 5. Effect of prestraining in creep on the true stress-strain relation, at small strains, at room temperature; ofboth cold-drawn and annealed copper.

174

£"

^^t$

FIGURE (>. Representative structure?, of copper in different initial conditions.

Longitudinal sections etched In equal parts x i i i O i i and BaOa (:i'';,).X75.A, Annealed at six )"!•'; 0.02S nun average grain diameter; B, annealed at 800° F, then cold-drawn 34-percent reduction in area; C, annealed at 1,150'

i<\ then cold-drawn 70-percent reduction In area; i>, cold-drawn 70-percent reduction in area and then annealed al 1,300° F; 0.120-mm average graindiameter

2 1 7 0 2 5 175

O---ANNEALEA—ANNEALE70% REDUCTM — - MAXIMUMRT - -BEGINNIN

O

G

AA

2o

A0

k

D AT 800*F FOR 1 HRD AT II5O°F FOR 50 MINUTES; THEN COLD-DRAWNON OF AREA; THEN ANNEALED AT I3OO°F FOR 1 HR

LOADG OF FRACTURE

£

o

° 40000A

30000

20000

10000

c

C

o

>

o

o A

/ /

A

0.

AO

o£>

o

. 0.2 0.3

0 0.5 1.0 1.5 2.0 2.5TRUE STRAIN, UOGe Ao/A

FIGURE 7. Effect of grain size on the true stress-strain relationat room temperature of annealed copper.

Annealed at 800° F, 0.025 mm average grain diameter. Annealed at 1,300° F,0.120-mm average grain diameter.

35

30

25

20

HARDNESS OF COPPER AS COLD-DRAWN 70 PERCENT REDUC-TION OF AREA ROCKWELL F 91

900 1000 1100 1200

ANNEALING TEMPERATURE, °F

. 0 3

FIGURE 8. Effect of annealing temperature on the grain sizeand hardness of cold-drawn copper.

176

FIGUJRE 9. Some of the copper specimens after fracturing al room temperature,[ J n e t c h e d , x i .

A

ItC1)EF

(i

[nitlal condition

Cold-drawn 70%; an-nealed 1,300° F.

Vnnealed, 800° Fdodo

Cold-drawn, 40',dodo

(Ireep test

Temper-ature

°F\Diii'

mino2.SO260300

1 llo

Rate

%/l,0OOhr\<>Hc

2. n1.08l . d l(i. 20

86. 68

200+

Plastlcextension

None

39. 719. (i23. 10. 73

,98.731

2 8 1

Truestrain alcompletefracture

1.96

2. is1.94i.:(.r)1.030. 68

1.48

177

FlOURE 10. Representative copper specimens showing the surface conditions after fracture.

Qnetched, XT.

A

BC1)

Initial condition

( 'n l ( l - ( l r ; iw II 7 ( i ' , ; a nA nealed 1,800° F.Annealed 800° FCold-drawn W,Annealed 800° F

Creep test

Temper-ature

• • / • •

None

280300300

Rate

' ,11,1X1(1In

None

I . I I I

0.388.3

Plastic

tension

x,

23. I0. 98

38. :<

Remarks

Fractured at roomtemperature.

Do.Do.

Fractured In creepat 300° F.

178

FIGURE 11. Ends of copper specimens after fracturing at room temperature.I n, idied, X7.

ABc1)EF

Initial condition

Aimcnlrd 800° 1''do

Cold-drawn 40%dododo -

Creep test

Temper-ature

" / • •

no250

None260.'ton

| i ID\300

Elate

%Hfi00hr•2. 7.r>1 . 1 ) 1

None0.20

:R200 |

Plastioextension

' 039. 723. INonei). 73

98,7. ' f

2.8

179

FIGURE 12. Structures at the surf aces of specimens of annealed copper fractured intension at room temperaturtLongitudinal sections at the Intersection of the outer (original) and Fracture

surfaces. Etched in equal parts NH4OH and HaOj(3%). xioo.

A

B0I)

Initial condition

Am led 800° FAnnealed 800°F; extended '•>', In ten-

sion at room temperature -Annealed 800°F

do

Creep test

Tem-per-

ature

°Fno2.r.(l2M2M

Kale

< ,,11,1)00 hr2 . 7.r>

1. 1l . d l

1(1. 1

Plasticexten-sion

39! 7

7.023. 119. 0

180

p

•n

III

»• «f • •

, •«> t ^ * "

- . - . y - ' • • •

' "Jim

;... - * $ »

• * • ' . • • . I r * •

FIGURE 13. Structures at the surfaces of specimens' of cold-drawn copper fractured in tension at room temperature.

Longitudinal sections at the intersection of the outer (original) and fracturesurfaces. Etched in equal parts of NH«OH and HJOJ(3%).. X100.

ABc1)

Initial condition

do

Cold-drawn 3495 reduction In area< 'old-drawn 7(i', reduction in area

Creep test.

Temper-ature

°F250

fllO

Nonedo . .

Rate

%/l,OOOKr0, 21). 03

200+None

do

Plasticexten-sion

0. 73..73.

2.8.None.

Do.

181

FIGURE 14. Structures at the axis of specimens of annealed copper fractured in tension at room temperature.Longitudinal sections at the intersection of the axis and fracture surface.

Etched in equal parts of NH«OE and 11,0.. (3%). xioo.

Ali

C1)

initial condition

Annealed 800° FAnnealed 800° F extended 9% in ten-

sion at room temperature.Annealed 800° F

do

Creep lest

Tem-pera-ture

°F110260

2M

H a l e

%/l,0O0hr2. vr.1.1

Mil10. 1

Plasticexten-sion

39. 77.0

23, 149.0

182

FlGTJRB 15. Structures at the axis of specimens of copper initially Cold-drawn 40% reduction in area afUfracturing in tension at room temperature.

Longitudinal sections at the Intersection ol the :ixis and fracture surface.Etched In equal parts of NH4OE and 11 L-< >i> (3%). Xioo.

AIt

c. .1)

Creep test

Tempera-ture

"FNone

2M800

/no1.800

Rate

%/lfiOOhr.None0. 20. 36.63

200+

Plasticextension

%None0. T.M

. 98

.712.8

183

FIGURE l(>. Structure in the un-necked region corresponding to the limit of uniform extension of specimens fracturedin tension at room temperature.

L o n g i t u d i n a l s e c t i o n s n e a r a x i s , e t c h e d in e q u a l p a r t s of N l l i O l I a n d I l a O s( 3 % ) .

A .It

( '

l>

in i t ia l condition

Annealed «(K)° P. -Cold-drawn 40% reduc-

tion in area.fold-drawn 70', reduc-

tion in area,; annealed1800° !•'.

- - . do. . . .

Creep lest

Tem-pera-ture

360380

None

(In

Kali'

%/l,0OOhr.1.010 2(1

None

do ^

Plasticextension

Of/o

28.1o 7:t

None

do

Magnifi-cation

xioo.Xioo.

xioo.

X600.

184

FIGURE 17. Effect of testing conditions on the structur, >l>pcr.

Longitudinal sections near axis, etched in 3 parts of glacial ucelic acid, 6 par t s ofni t r ic acid (cone.) and I part of elliylene glycol. X75O.

A

It

c1)E

lnitial condition

Cold-drawn 70' ,reduction in area:annealed 1,300° F.

do

Annealed 800° Fdo

Ccild d r a w n 1(1',;,,reduction In urea.

do

Creep test

Tem-pera-ture

/•'.None

do

110280None

800

Kale

%/lfi00 lir.None

do

2.751.01None

0.8fl

Plasticexten-sion

' i

None

do

89.723.1None

0.08

Remarks

As annealed.

Struct!1 II el0 0 inIracti

DoDoDo

re 0.10from

p 1 e 1 eire.

Do.

185

fe.

S80

O-ANNEALED AT 800*F FOR I HRO -COLD-DRAWN 34% RED. OF AREA

40% "70% •

% A - r> r-

ANNEALED AT 800* F FOR t HR

O-ANNEALED AT 800*F FOR I HR-COLD-DRAWN 70% RED. OF AREA

THEN ANNEALED AT 1300'F FOR I HR

COLD-DRAWN 40% RED. OF AREA

DISTANCE FROM FRACTURED SURFACE, INCHES

FIGURE 18. Effect of prestraining on the necking character-istics of specimens fractured in tension at room temperature.

O-ANNEALED AT 800*F FOR I HROF A

I- • . . . . ; THENNNEALED AT I3OO*F FOR I HR

7 O-ANNEALED AT 800*FO - COLD-DRAWN 34% R OF AA - " 40% • " "• - •• 70% « « •

COLD-DRAWN 40% R OF A

NUMBERS ADJACENT TO CURVES INDICATE

AVERAGE CREEP RATE, SECOND STAGE, INPERCENT PER ICOO HR; AND PLASTIC EXTENSIONINCURRED DURING CREEP TEST, IN PERCENT

50 75 25REDUCTION OF AREA, PERCENT

FIGURE 19. Effect of prestraining on the hardness-reductionof area relation of specimens fractured in tension at roomtemperature.

WASHINGTON. March 31, 1952.

186