, . .^ ".^^^;v J% l^K^\ ^^'"^<^ °-^^^* /\ '.^P/ ^^^'^^ °"WP!^" /' V -1 
 
 **\.-'J^%\ C" -a^i'-"- /.tiiz^-^ oO^.i^I."^". /.•i;;^."-*' /.i^-.*". 
 
 
 .0*.. 
 
 
 
 
 0^ r ''^f^'; ^o v^ 
 
 
 
 
 
 ^^--^^ 
 
 
 A 
 
 
 
 
 I- O 
 
 - - //>55U,J«*^ ^-V ° ^^C^ 
 
 
 
 
 
 ii>\ ''W^/ /vl^-"**'^ "• 
 
 0" -..cfS^^^^ ^ 
 
 '. %./ /J 
 
 

 \/'.-^^V%^**'-*^ 
 
 . . • A 
 
 
 
 ^o. 
 
 
 cd 
 
 ,%' ... -^. ''".o^". ^-^^ ""* ^''^ -"- ^-^ 
 
 ''>^t 
 
 ^o-n^ 
 
 
 
 '•-..<* .-i^^-. **..s-* .•: 
 
 
 
 
 ^v^ %'?^\0^^ \-'o<^"a^ V-^^^\c,^^ "^"^o, 
 
 -o .^^'v^i.V /.^^^% .^^\.^:^\ ^0^ •-'- "° 
 
 <^ ^^. 
 
 
 
 
 
 > . s • • > *^ rt v y , 
 
 
 
 
 <c>^''^^.. V^ 
 
 
 ^°-^. 
 
 4 o 
 
 
 ♦^ : 
 
 
 o 
 
 C^ 
 
 z;^^: 
 
 
 
 % 
 
 
 >°-'*. V 
 
 ■• V„.' 'fm^. "^^.^'^ yM^: %.„. 
 
 ^0' \. •■•.■• .*^ 
 
 
 
 ■■^ 0«. ♦ • . 
 
 ■••;-^-.°- ..**.-i-.v.V .c»'.v;^.> .,**\.i^;i..v /..i:^:.*°o .**\.>;^-^-. 
 
 '-v 
 
Bureau of Mines Information Circular/1985 
 
 Metallurgical Effects of Impurities 
 in Recycled Copper Alloys 
 
 By Harry V. Makar and William D. Riley 
 
 1 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 
 m 
 
 T5I 
 
 ^^INES 75TH AV)''^ 
 
Information Circular 9033 
 
 Metallurgical Effects of Impurities 
 in Recycled Copper Alloys 
 
 By Harry V. Makar and William D. Riley 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 
 Donald Paul Hodel, Secretary 
 
 BUREAU OF MINES 
 Robert C. Horton, Director 
 
4^ 
 
 
 Library of Congress Cataloging in Publication Data; 
 
 Makar, H. V. (Harry V.) 
 
 Metallurgical effects of impurities in recycled copper alloys, 
 
 (Information circular ; 9033) 
 
 Bibliography: p. 18-19. 
 
 Supt. of Docs, no.: I 28.27:9033. 
 
 1. Copper alloy s— Metallurgy. 2. Copper alloy s— Inclusions. 3. 
 Scrap metals. I. Riley, W. D. {William D.). II. Title. III. Series: 
 Information circular (United States. Bureau of Mines) ; 9033. 
 
 TN295.U4 [TN693.C9] 622s [669'. 3] 85-600046 
 
^ 
 
 Page 
 
 'V^ CONTENTS 
 
 ^__^ Abstract 
 
 <^ Introduction 
 
 ■^Effects of impurities on copper alloys 
 
 >^ Wrought alloys 
 
 ^S-, Cast alloys 
 
 TjMetallurgical effects of tramp elements , 
 
 / ^ Hot workability 
 
 Effect of lead , 
 
 Effect of lead and iron , 
 
 Effect of lead and zirconium , 
 
 Fire cracking 
 
 Effect of chemistry and processing , 
 
 Liquid metal embrittlement , 
 
 Additional studies 
 
 Microstructural effects , 
 
 Tin bronzes , 
 
 Metallography 
 
 Base alloy , 
 
 Antimony , 
 
 Iron , 
 
 Lead , 
 
 Nickel , 
 
 Phosphorus , 
 
 Silicon , 
 
 Sulfur , 
 
 Microprobe analysis , 
 
 Iron , 
 
 Nickel and antimony , 
 
 Sulfur , 
 
 Tin , 
 
 Zinc , 
 
 Grain size and shape , 
 
 Eutectoid composition , 
 
 Microsegregation , 
 
 Failure analysis of bronze propellers , 
 
 Composition , 
 
 Results of analysis , 
 
 W>Summary , 
 
 t^ References , 
 
 ^ ILLUSTRATIONS 
 
 , 1. Typical sorting routine for mixed copper alloy scrap 4 
 
 2. Schematic representation of grain boundary voids due to dislocation 
 
 impingement on grain boundary impurities 9 
 
 3. Effect of lead, iron, and iron and lead on hot rolling of 60:40 brass 10 
 
 , > 4. Effect of iron on hot rolling of a complex brass containing 1.33 pet Mn.... 10 
 
 ^ 5. Effect of lead and iron on hot rolling of a complex brass at 650° C 10 
 
 6. Effect of lead and iron on hot rolling of a complex brass at 800° C 11 
 
 7. Effect of zirconium and lead on hot twist ductility of a copper-nickel 
 
 alloy 11 
 
 8. Effect of lead on tensile ductility (reduction in area) of nickel-silvers.. 13 
 
 9. Effect of lead on stress-strain behavior of nickel-silvers 13 
 
 1 
 
 2 
 
 4 
 
 6 
 
 6 
 
 8 
 
 8 
 
 9 
 
 9 
 
 11 
 
 11 
 
 12 
 
 12 
 
 13 
 
 14 
 
 14 
 
 14 
 
 14 
 
 14 
 
 14 
 
 14 
 
 14 
 
 15 
 
 15 
 
 15 
 
 15 
 
 15 
 
 15 
 
 15 
 
 15 
 
 15 
 
 15 
 
 15 
 
 16 
 
 16 
 
 16 
 
 16 
 
 17 
 
 18 
 
11 
 
 TABLES 
 
 Page 
 
 1. Typical copper alloys, nominal composition 3 
 
 2. Compositional variations for wrought and cast copper alloys 5 
 
 3. Limits for lead and bismuth in alpha copper alloys for hot rolling 6 
 
 4. Effect of impurities on phosphor bronzes 7 
 
 5. Effect of chemical composition on fire cracking 12 
 
 6. Composition of tin bronze alloy used for microstructure studies 14 
 
 7. Chemical composition of propeller blades 16 
 
 
 UNIT OF MEASURE ABBREVIATIONS 
 
 USED IN THIS REPORT 
 
 HB 
 
 Brinell hardness number 
 
 pet percent 
 
 in 
 
 inch 
 
 ton/in^ ton per square 
 inch 
 
 kg/mm^ 
 
 kilogram per square 
 
 
 
 millimeter 
 
 wt pet weight percent 
 
 mm 
 
 millimeter 
 
 
METALLURGICAL EFFECTS OF IMPURITIES IN RECYCLED COPPER ALLOYS 
 
 By Harry V. Makar ^ and William D. Riley^ 
 
 ABSTRACT 
 
 As part of a continuing research program for conserving domestic min- 
 eral resources, the Bureau of Mines is investigating new and improved 
 scrap identification techniques to improve sorted mixed scrap purity. 
 This report focuses on various classes of wrought and cast copper alloys 
 produced with recycled scrap. Based on a survey of the literature, 
 principal alloys, such as the brasses and bronzes, are examined with 
 respect to impurity elements and their effects on metallurgical be- 
 havior. Effects of elements such as lead, antimony, iron, chromium, and 
 aluminum are discussed. The metallurgical effects described include hot 
 shortness, fire cracking, and undesirable phase transformations. 
 
 'Branch Chief, Division of Ferrous Metals, Bureau of Mines, Washington, DC. 
 ^Physical science technician, Bureau of Mines, Avondale Research Center, Avondale, 
 MD. 
 
INTRODUCTION 
 
 Old scrap was 26 pet of the total 
 2,278,000 metric tons of U.S. copper sup- 
 ply in 1981 and is expected to be about 
 29 pet by 2000 ( 4_) . 3 The smelters and 
 refiners consumed 54 pet of the total, 
 brass mills 40 pet, and others 6 percent. 
 Overall, the secondary metals industry 
 consists of an estimated 4,000 dealers 
 and scrap processors. Details on second- 
 ary copper are available in other publi- 
 cations by the Bureau of Mines and the 
 National Association of Recycling Indus- 
 tries (NARI) (2, 2A, 23-24, 30-22). 
 
 Specifications for the various alloys 
 produced by the industry 'dictate the 
 quality and the extent of upgrading and 
 refining that may be required. In some 
 cases, refining may be too costly or im- 
 practical and the need for highly segre- 
 gated scrap is essential. It has been 
 estimated that at one time as many as 500 
 commercial copper alloys were made in the 
 United States. With such a large number 
 of alloys and the almost infinite varie- 
 ties of mixtures that can occur when 
 these alloys come back into the recycling 
 process, the task of proper upgrading is 
 indeed monumental. This task is greatly 
 lessened with the use of classifications 
 such as those published by the National 
 Association of Recycling Industries (24) . 
 These classifications provide a standard- 
 ized system for use by the industry to 
 segregate mixed copper alloy scrap into 
 recyclable categories. 
 
 Several routine techniques are tradi- 
 tionally used in the copper recycling in- 
 dustry to identify scrap for effective 
 segregation. These include identifica- 
 tion based on object recognition, color, 
 apparent density, magnetic attraction, 
 and chemical spot tests. Some of the 
 more sophisticated techniques commer- 
 cially available include fluorescent X- 
 ray spectroscopy, portable optical emis- 
 sion devices, and thermoelectric sorters. 
 Table 1 lists some common alloy groups 
 and typical compositions compiled by the 
 Institute of Scrap Iron and Steel (ISIS), 
 
 ■^Underlined numbers in parentheses re- 
 fer to items in the list of references at 
 the end of this report. 
 
 and figure 1 shows a typical scheme for 
 routine sorting of a mixture of copper 
 alloys. These techniques, when properly 
 applied by skilled sorters to mixtures of 
 these alloys, permit effective identifi- 
 cation and segregation into their respec- 
 tive specified categories , such as red 
 brass, yellow brass, manganese bronze, 
 and aluminum bronze. 
 
 Although traditional routine sorting 
 methods are effective with regard to 
 generic or descriptive specifications, 
 there are many opportunities for intro- 
 duction of impurities into new alloys 
 made from improperly segregated scrap. 
 Color distinctions may be obscured under 
 certain lighting conditions. Also, there 
 may be several specific alloys within a 
 certain group, differing significantly in 
 lead, antimony, aluminum, and other ele- 
 ments that cannot be distinguished in 
 routine sorting. In addition to the 
 large number of alloys that must be han- 
 dled, certain attachments to postconsumer 
 parts also introduce impurities. These 
 include iron fittings and plated parts. 
 Previous Bureau of Mines research (2, 19- 
 20 , 25 , 27 ) reported on improved tech- 
 niques for metal scrap identification. 
 The objective was to develop relatively 
 simple, rapid techniques requiring few 
 operator skills that would improve scrap 
 purity during sorting. 
 
 The purpose of this paper is to examine 
 some of the reported metallurgical ef- 
 fects of impurities in copper alloys. 
 Based on the literature, a general over- 
 view of reported effects is given for 
 various impurities. This is followed by 
 specific case studies in which the metal- 
 lurgical effects of selected impurities 
 are examined in more detail. The effects 
 discussed include hot shortness, fire 
 cracking, and unwanted phase transforma- 
 tions. It should be noted that not all 
 of the metallurgical effects are attrib- 
 utable solely to composition. In some 
 cases , the mechanisms of the problems en- 
 countered are still not fully understood, 
 and in other cases improper metallurgical 
 processing in terms of foundry control 
 and/or heat treatment may be more to 
 blame than the presence of impurities. 
 
TABLE 1. - Typical copper alloys, nominal composition, weight percent 
 
 Allov 
 
 Cu 
 
 Sn 
 
 Pb 
 
 Zn 
 
 Others 
 
 Common uses 
 
 Silicon bronze. . 
 
 82.5-97 
 
 0-1 
 
 Trace 
 
 13 
 
 1.35-4.50 Si 
 
 Rod, bar, plate sheet, 
 bearings, impellers, pump 
 parts, valve stems, corro- 
 sion-resistant castings. 
 
 Red brass 
 
 85 
 
 5 
 
 5 
 
 5 
 
 NAp 
 
 General casting for good 
 machining qualities, low- 
 pressure valves, pipe fit- 
 tings, small pump tast- 
 ings, ornamental fixtures. 
 
 Phosphorous 
 bronze. 
 
 95 
 
 5 
 
 NAp 
 
 NAp 
 
 NAp 
 
 Bars, spings, rods. 
 
 Navy brass 
 
 88 
 
 6-8 
 
 0-2 
 
 4 
 
 NAp 
 
 Steam pressure castings, 
 valve bodies, pipe flange, 
 gears and bushings, pump 
 impellers, steam fittings, 
 valves and parts. 
 
 Yellow brass. . . . 
 
 66 
 
 1 
 
 3 
 
 30 
 
 NAp 
 
 Light casting not subject 
 to high internal pressure, 
 hardware fittings, orna- 
 mental castings, gas 
 cocks, radiator fittings. 
 
 Muntz metal 
 
 60 
 
 NAp 
 
 NAp 
 
 40 
 
 NAp 
 
 Large nuts and bolts, braz- 
 ing rods, condenser tubes. 
 
 Admiralty brass. 
 
 70 
 
 1 
 
 NAp 
 
 29 
 
 NAp 
 
 Condenser tubes. 
 
 Manganese bronze 
 
 59 
 
 .25 
 
 .25 
 
 38 
 
 1 Fe 
 1 Al 
 0.5 Mn 
 
 Valve stems, marine cast- 
 ings and propellers , 
 gears , bushings and 
 bearings. 
 
 Aluminum bronze. 
 
 88 
 
 NAp 
 
 NAp 
 
 NAp 
 
 2 Fe 
 
 10 Al 
 
 Acid-resistant pumps, valve 
 seats, gears, bushings and 
 bearings. 
 
 Copper-nickel. . . 
 
 70 -90 
 
 NAp 
 
 NAp 
 
 NAp 
 
 10-30 Ni 
 
 Boiler tubes, saltwater 
 equipment. 
 
 Nickel-silver. . . 
 
 60 -65 
 
 0-5 
 
 0-7 
 
 NAp 
 
 5-25 Ni 
 
 Hardware fittings, valves, 
 plumbing fixtures, orna- 
 mental castings, dairy and 
 citrus machinery. 
 
 NAp Not applicable. 
 
 Source: Institute of Scrap Iron and Steel, Inc. 
 
Copper 
 (red brown) 
 
 Mixed scrap 
 
 Visual 
 
 (object recognition 
 
 and color) 
 
 Bronze and some brasses 
 (dark yellow) 
 
 Magnet 
 
 Brasses (other than red brass) 
 (light yellow) 
 
 Aluminum bronze, 
 manganese bronze 
 
 (magnetic) 
 
 Tin bronze, silicon bronze, 
 
 leaded bronze, red brass 
 
 (nonmagnetic) 
 
 Admiralty metal, 
 
 naval brass 
 
 (white precipitate) 
 
 Yellow brass, 
 aluminum brass 
 (no precipitate) 
 
 AgNO, 
 
 Manganese bronze 
 (dark spot) 
 
 Aluminum bronze 
 (no reaction) 
 
 Tin bronze 
 (white precipitate) 
 
 Silicon bronze 
 (gelatinous material) 
 
 Leaded bronze, 
 red brass 
 
 H,SO. 
 
 HCI 
 
 HNO, 
 
 KOH 
 
 Alizarin S 
 
 Yellow brass 
 (no color) 
 
 Leaded bronze 
 (white precipitate) 
 
 Red brass 
 
 (no white precipitate) 
 
 FIGURE 1. - Typical sorting routine for mixed copper alloy scrap. 
 
 EFFECTS OF IMPURITIES ON COPPER ALLOYS 
 
 Aluminum brass 
 (red color) 
 
 A standard designation system for 
 copper and copper alloys has been pub- 
 lished by the Copper Development Associa- 
 tion, Inc. (7^). Approximately 12 pages 
 of wrought and cast copper alloys are 
 listed under various families within the 
 following four general classes: Brass- 
 es, bronzes, copper-nickels, and nickel- 
 silvers. Table 2 lists typical alloy 
 families and their compositions. Many of 
 these alloy families also have high-lead 
 versions. Even within given families of 
 alloys (e.g., brasses, bronzes, etc.), 
 there are wide ranges for many elements 
 that can cause metallurgical problems , 
 singly or through interaction effects, if 
 
 recycled scrap is improperly mixed. Most 
 of these elements are intentionally added 
 to impart certain desired properties in 
 the original alloys. For example, anti- 
 mony, arsenic, and phosphorus are added 
 in small amounts to wrought alloys to 
 inhibit dezincif ication; lead to enhance 
 machinability and to promote smooth edges 
 during shearing and blanking operations; 
 tellurium, selenium, and sulfur for im- 
 proved machinability; aluminum for re- 
 sistance to impingement corrosion in 
 condenser tubing alloys ; and chromium 
 to form a heat-treatable alloy ( 18) . 
 Some of these elements are also used in 
 cast alloys for the same reasons. In 
 
-I 
 
 c 
 a 
 
 V 
 
 a. 
 
 x: 
 
 m 
 o 
 
 u 
 
 OJ 
 Q. 
 
 a. 
 o 
 a 
 
 4.1 
 
 CO 
 
 CO 
 
 C 
 
 x: 
 
 3 
 
 ce 
 o 
 o. 
 
 e 
 o 
 u 
 
 J 
 
 aa 
 < 
 
 
 
 
 • 
 
 
 
 
 
 
 o 
 
 
 
 
 
 ^ 
 
 
 
 
 
 • a. 
 
 
 
 
 
 
 • 
 
 
 
 
 • 
 
 P- 
 
 
 
 
 
 a, 
 
 in 
 
 
 «s 
 
 
 
 vO 
 
 
 
 
 0) 
 
 
 in 
 
 
 
 o 
 
 o 
 
 • 
 
 • 
 
 0) 
 
 
 
 1 
 
 
 •^ 
 
 
 fa. 
 
 o 
 
 CO 
 
 
 
 •—1 
 
 o — 
 
 in 
 
 1-1 
 
 fa. 
 
 *-. 
 
 
 in 
 
 ^ 
 
 CO 
 
 
 " 
 
 • 
 
 • 
 
 
 
 • 
 
 f— 4 • 
 
 1 
 
 z 
 
 
 <v 
 
 
 CN 
 
 XI 
 
 < 
 
 
 O CO 
 
 •-H 
 
 o 
 
 
 
 o 
 
 . o 
 
 in 
 
 
 o 
 
 fa. 
 
 
 * 
 
 C/5 
 
 
 
 . < 
 
 1 
 
 • 1 
 
 
 
 
 o 
 
 • 
 
 CN 
 
 • 
 
 
 
 o 
 
 
 in 
 
 
 <f 
 
 in 
 
 •H O 
 
 
 
 0> 
 
 ». 
 
 o 
 
 • 
 
 <r 
 
 ro 
 
 
 
 in 
 
 o 
 
 
 1 o 
 
 o 
 
 2 
 
 
 
 05 
 
 « CD 
 
 
 CN 
 
 1 
 
 • 
 
 
 ^ 
 
 CN 
 
 • 
 
 
 ■<r o 
 
 • 
 
 #* 
 
 
 
 < 
 
 •H < 
 
 «^ 
 
 1 
 
 o 
 
 CN 
 
 
 c 
 
 • 
 
 o 
 
 
 • • 
 
 o 
 
 in c 
 
 
 
 
 z 
 
 c 
 
 o 
 
 
 1 
 
 
 )!§ 
 
 O 
 
 1 
 
 • 
 
 o o 
 
 
 • ^ 
 
 
 
 o 
 
 o 
 
 s 
 
 
 *v 
 
 o 
 
 
 
 1 
 
 o 
 
 c 
 
 1 
 
 ^ 
 
 in • 
 
 
 
 ^H 
 
 Pvl — < 
 
 
 t^ 
 
 c 
 
 
 
 O 
 
 o 
 
 
 s 
 
 » o 
 
 • XI 
 
 1 O QJ 
 
 a; 
 
 
 • 
 
 • • 
 
 in 
 
 c 
 
 2 
 
 *K 
 
 • 
 
 • 
 
 
 *l 
 
 
 iH 
 
 a. CO 
 
 o . fa- 
 
 i-i 
 
 
 C 
 
 — " o 
 
 • 
 
 s 
 
 
 c 
 
 (U 
 
 t-H 
 
 ** 
 
 XI 
 
 in 
 
 Z ' 
 
 
 . — 1 
 
 <u 
 
 
 
 
 r^ 
 
 
 o 
 
 s 
 
 CO 
 
 I 
 
 <u 
 
 en 
 
 CN 
 
 c 
 
 vD o 
 
 - 1 in 
 
 s: 
 
 
 #^ 
 
 ^ #t 
 
 1 
 
 in 
 
 • 
 
 
 
 in 
 
 fa-. 
 
 
 • 
 
 o S 
 
 O CSI 
 
 C in 
 
 u 
 
 
 c 
 
 CO XI 
 
 in 
 
 » 
 
 in 
 
 in 
 
 r^ 
 
 o 
 
 
 o 
 
 o 
 
 • 
 
 • • 
 
 ^ • *— ' 
 
 O 
 
 
 s 
 
 < w 
 
 • 
 
 --< 
 
 1 
 
 • 
 
 • 
 
 
 <f 
 
 CM 
 
 1 
 
 -* in 
 
 o o 
 
 o 
 
 
 
 
 
 O 
 
 1 
 
 in 
 
 CN 
 
 o 
 
 o 
 
 • 
 
 • 
 
 o 
 
 1 
 
 1 
 
 <r 
 
 
 
 in 
 
 in o 
 
 
 • O 
 
 o 
 
 1 
 
 1 
 
 
 o 
 
 o 
 
 
 O -H 
 
 O ' 
 
 _l .s PL, 
 
 
 
 r-i • 
 
 CN — ■ 
 
 ** 
 
 CO 
 
 • 
 
 o 
 
 o 
 
 ^ 
 
 1 
 
 1 
 
 ^ 
 
 1 
 
 Q) 
 
 1 -H 
 
 
 
 • a- 
 
 
 Q) 
 
 < - 
 
 o 
 
 
 
 •H 
 
 o 
 
 o 
 
 •H 
 
 ' o 
 
 <- fa. 
 
 o z in 
 
 
 
 c 
 
 c o 
 
 Cb 
 
 0) 
 
 • 
 
 •^ 
 
 ^ 
 
 z 
 
 
 • 
 
 Z 
 
 c 
 
 XI 
 
 o 
 
 
 
 m 
 
 
 
 in fa- 
 
 - CO 
 
 1-1 
 
 )-l 
 
 
 ^ 
 
 CL, « 
 
 
 )s ** 
 
 CO in 
 
 C '-0 • 
 
 
 
 • » rn 
 
 ^ ^ 
 
 O 
 
 m 
 
 1-1 <: 
 
 z 
 
 o 
 
 o 
 
 1-1 
 
 OJ 
 
 O 
 
 0) 
 
 CN 
 
 S QJ • o 
 
 
 
 <D 0) • 
 
 Q) (U 
 
 
 • 00 
 
 z 
 
 
 
 • 
 
 z 
 
 in fa- 
 
 • 
 
 O fa- 
 
 vD • 
 
 fa. r^ 1 
 
 
 
 6u Cr. O 
 
 fa- Ck 
 
 in 
 
 o • 
 
 00 
 
 o 
 
 o 
 
 vo 
 
 
 o 
 
 r— 4 
 
 • 
 
 O O 
 
 ^ CSI O 
 
 
 
 
 
 1 
 
 o 
 
 m • 
 
 
 a 
 
 CM 
 
 o 
 
 • r^ 
 
 1 
 
 in in 
 
 • 1 
 
 • O 1 
 
 
 
 r-~ in 
 
 in o 
 
 o 
 
 - 1 
 
 . O 
 
 ro 
 
 o-i 
 
 1 
 
 
 O • 
 
 CN 
 
 1 
 
 o o 
 
 o . o - 
 
 
 
 O O J3 
 
 en -H 
 
 ,—1 
 
 T-( -H 
 
 CN 1 
 
 en 
 
 1 
 
 O 0) 
 
 1—t 
 
 1 o 
 
 • 
 
 ^ CN 
 
 1 -1 
 
 1 in . X2 
 
 
 
 • • C/2 
 
 • • 
 
 • 
 
 z • 
 
 1 O 
 
 1 
 
 o 
 
 • fa. 
 
 1 
 
 O 1 
 
 O 
 
 • 1 
 
 o • 
 
 O 1 — 1 CO 
 
 
 
 c o 
 
 c o 
 
 o 
 
 o 
 
 O 
 
 CSI 
 
 
 r~~ 
 
 o 
 
 o 
 
 
 o o 
 
 o 
 
 O -H 
 
 
 
 ex D. 
 
 a D. 
 
 
 
 
 O. 
 
 
 o. 
 
 in 
 
 
 
 o. 
 
 in 
 
 
 
 
 < < 
 
 < < 
 
 
 
 
 <: 
 
 
 <; 
 
 o 
 
 in 
 
 
 < 
 
 o 
 
 in 
 
 
 
 z z 
 
 z z 
 
 CN 
 
 o 
 
 in 
 
 z 
 
 
 z 
 
 o 
 
 CN 
 
 
 Z in 
 
 o 
 
 CO ^ 
 
 
 
 
 
 • 
 
 • 
 
 • 
 
 
 
 
 • 
 
 • 
 
 
 • 
 
 • 
 
 • • 
 
 •H 
 
 
 
 
 CN 
 
 <f 
 
 ,-H 
 
 
 
 
 
 
 
 in 
 
 
 CO 
 
 I/l 
 
 
 
 
 1 
 
 1 
 
 1 
 
 c 
 
 
 
 
 
 1 
 
 o 
 
 
 1 
 o 
 
 
 1 1 
 
 o o 
 
 
 
 ^ o 
 
 F-l rH 
 
 C» 
 
 in 
 
 .H 
 
 o 
 
 
 iH 
 
 o 
 
 o 
 
 
 o 
 
 o 
 
 D.O 
 
 
 
 to • 
 
 CO CO 
 
 • 
 
 • 
 
 CO 
 
 • 
 
 
 CO 
 
 • 
 
 • 
 
 
 • 
 
 • 
 
 <: • 
 
 
 
 CO o 
 
 CO CQ 
 
 
 t—t 
 
 CO 
 
 »-H 
 
 
 CO 
 
 r^ 
 
 I-- 
 
 
 CN vO 
 
 in 
 
 2 en 
 
 
 
 CN 
 
 
 
 
 
 
 
 
 ^H 
 
 <r 
 
 
 <r -H 
 
 
 CN 
 
 c 
 
 
 1 
 
 
 1 
 
 1 
 
 
 1 
 
 
 
 1 
 
 1 
 
 
 1 1 
 
 1 
 
 1 
 
 E< 
 
 
 o 
 
 
 
 
 
 
 
 
 o 
 
 o 
 
 • 
 
 
 o 
 
 • 
 
 in 
 
 CN 
 
 o 
 
 
 
 CNi 
 
 
 o 
 
 o 
 
 
 o 
 
 
 
 "^ 
 
 o 
 
 CN 
 
 
 CN in 
 
 CN 
 
 
 " 
 
 
 
 a. o 
 
 D. O 
 
 o 
 
 vD 
 
 in 
 
 CI. 
 
 
 -.D 
 
 in 
 
 o 
 
 
 in o 
 
 O 
 
 o- in 
 
 
 
 < • 
 
 <d . 
 
 
 
 • 
 
 <c 
 
 
 
 • 
 
 • 
 
 
 • • 
 
 • 
 
 <: . 
 
 
 
 z rn 
 
 Z — 1 
 
 CN 
 
 .-H 
 
 t-H 
 
 z 
 
 
 
 vO 
 
 CN 
 
 
 i — 1 ^H 
 
 o 
 
 Z in 
 
 C 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CN 
 
 
 w: 
 
 
 1 
 
 1 
 
 1 
 
 1 
 
 1 
 
 
 
 1 
 
 1 
 
 1 
 
 
 1 1 
 
 1 
 
 1 
 
 
 
 fM 
 
 in 
 
 CN 
 
 
 
 
 
 
 CN 
 
 CN 
 
 
 CN 
 
 O 
 
 in 
 
 
 
 « 
 
 • 
 
 • 
 
 
 
 
 
 
 « 
 
 • 
 
 
 • 
 
 • 
 
 • 
 
 
 
 o 
 
 
 
 o 
 
 o 
 
 
 
 O 
 
 
 
 
 O 
 
 v£> 
 
 1 — t 
 
 
 
 D. a. 
 
 C CL 
 
 
 
 
 a. 
 
 
 D. 
 
 in 
 
 
 
 
 in 
 
 m 
 
 
 
 < < 
 
 < < 
 
 
 
 
 < 
 
 
 < 
 
 o 
 
 in 
 
 
 
 o 
 
 o 
 
 
 
 z z 
 
 z z 
 
 in 
 
 .—4 
 
 o 
 
 z 
 
 
 Z 
 
 o 
 
 in 
 
 
 in in 
 
 o 
 
 in o 
 
 
 
 
 
 • 
 
 • 
 
 • 
 
 
 
 
 • 
 
 • 
 
 
 • • 
 
 • 
 
 • • 
 
 ^H 
 
 
 
 
 en 
 
 
 vO 
 
 
 
 
 
 
 
 r^ ^^ 
 
 
 ^-H 
 
 < 
 
 
 
 
 1—1 
 
 CO 
 
 • 
 
 
 O 
 
 
 
 
 
 O 
 
 
 1 1 
 
 in 
 
 o 
 
 
 1 1 
 
 o 
 
 • 
 
 vO o 
 
 
 
 in 
 
 r~- rsi 
 
 ^ 
 
 
 
 in 
 
 
 
 
 
 
 
 o 
 
 IT) 
 
 
 
 ^^ ^H 
 
 • • 
 
 , — 1 
 
 00 
 
 o 
 
 o 
 
 
 o 
 
 o 
 
 00 
 
 
 in o 
 
 in 
 
 o o 
 
 
 
 « • 
 
 m CNl 
 
 • 
 
 • 
 
 • 
 
 • 
 
 
 • 
 
 • 
 
 • 
 
 
 • • 
 
 • 
 
 • • 
 
 
 
 ^ 
 
 
 
 
 n 
 
 
 
 CN 
 
 00 
 
 en 
 
 
 -H . — 1 
 
 
 1-H 
 
 XI 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 r-H 
 
 a. 
 
 
 1 1 
 
 r^^ in 
 
 1 1 
 
 1 
 
 1 
 
 1 
 
 1 
 
 
 1 
 
 1 
 
 1 
 
 
 1 1 
 
 in 
 
 1 
 
 o 
 
 1 1 
 
 
 
 O O 
 
 m CN 
 
 
 CN 
 
 CN 
 
 
 
 
 in 
 
 CSI 
 
 
 CM —1 
 
 rsj 
 
 o 
 
 
 
 a * 
 
 • • 
 
 
 • 
 
 • 
 
 
 
 
 • 
 
 • 
 
 
 • • 
 
 • 
 
 • 
 
 
 
 O 
 
 
 o 
 
 
 
 o 
 
 
 O 
 
 
 
 
 
 
 O -H 
 
 
 
 in 
 
 in 
 
 
 
 
 .-H 
 
 
 
 
 
 
 
 
 
 
 
 • 
 
 • 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 00 v£> 
 
 O m 
 
 vO 
 
 p^ 
 
 00 
 
 CO 
 
 
 o- 
 
 <r 
 
 <J- 
 
 
 00 as 
 
 <r 
 
 vO r~- 
 
 3 
 
 
 vC a^ 
 
 CT^ vO 
 
 a^ 
 
 0^ 
 
 vX3 
 
 
 
 p^ 
 
 C7^ 
 
 r^ 
 
 
 \o 00 
 
 CTS 
 
 00 vO 
 
 O 
 
 
 1 1 
 
 1 1 
 
 1 
 
 1 
 
 1 
 
 
 
 1 
 
 1 
 
 1 
 
 
 1 1 
 
 1 
 
 1 1 
 
 
 
 ~> o> 
 
 CJv CJ> 
 
 CJN 
 
 <r 
 
 in 
 
 
 
 CN 
 
 in 
 
 r^ 
 
 
 CO <J\ 
 
 as 
 
 -H cn 
 
 
 
 vO r~~ 
 
 in in, 
 
 r^ 
 
 Cjv 
 
 in 
 
 
 
 o- 
 
 i-~ 
 
 in 
 
 
 in r^ 
 
 r-~ 
 
 r-~ in 
 rsi 
 
 
 
 
 
 • 
 
 • 
 
 , 
 
 
 
 
 , 
 
 
 
 • • 
 
 
 . . 
 
 
 
 
 
 • 
 
 • 
 
 • 
 
 
 
 
 • 
 
 
 
 • • 
 
 
 • • 
 
 
 
 
 
 • 
 
 « 
 
 (U 
 
 
 
 
 T3 
 
 
 
 0) . 
 
 
 • I-I 
 
 
 
 
 
 cu 
 
 • 
 
 N 
 
 
 
 
 OJ 
 
 
 
 N • 
 
 
 OJ QJ 
 
 
 •• 
 
 
 
 .. N 
 
 0) 
 
 C 
 
 
 
 
 I-I 
 
 
 
 C 0) 
 
 
 N > 
 
 
 t» 
 
 
 
 CO C 
 
 N 
 
 o 
 
 
 
 
 1 
 
 
 
 O N 
 
 
 C -H 
 
 
 V 
 
 CO • 
 
 CO • 
 
 QJ O 
 
 C 
 
 Ui 
 
 
 
 
 1-1 
 
 CO 
 
 
 I-I C 
 
 
 O -H 
 
 
 « 
 
 CO • 
 
 CO v> 
 
 N 1-1 
 
 o 
 
 XI 
 
 .— ( 
 
 
 U 
 
 .. s 
 
 CO 
 
 
 •• J3 O 
 
 
 IJ to 
 
 >» 
 
 » 
 
 C8 • 
 
 CO CO 
 
 E X3 
 
 Ui 
 
 
 0) 
 
 
 0) 
 
 CO qj 
 
 CO 
 
 
 CO I-I 
 
 QJ 
 
 XI 1 
 
 o 
 
 « 
 
 U CO 
 
 Ui CO 
 
 O 
 
 XI 
 
 <u 
 
 ^ 
 
 
 > 
 
 QJ CO 
 
 I-I 
 
 
 QJ Q) XI 
 
 N 
 
 .-1 
 
 -H 
 
 u 
 
 X3 CO 
 
 XI W 
 
 i-i e 
 
 
 CO 
 
 u 
 
 
 — ( 
 
 CO 
 
 X3 
 
 
 N CO 
 
 c 
 
 S QJ 
 
 »H 
 
 ^ 
 
 « 
 
 X) 
 
 X3 3 
 
 c 
 
 (U 
 
 1-1 
 
 
 1-1 
 
 CO T3 
 
 • 
 
 
 C Q) C 
 
 o 
 
 3 j»; 
 
 <: 
 
 
 3 u 
 
 -o 
 
 c 
 
 o 
 
 c 
 
 c 
 
 
 CO 
 
 CO C 
 
 CO 3 
 
 
 O C O 
 
 I-I 
 
 c o 
 
 
 u 
 
 o -o 
 
 01 —1 
 
 U 1-1 
 
 o 
 
 CO 
 
 1 
 
 
 1 
 
 U CO 
 
 CO o 
 
 
 1-1 CO O 
 
 XI 
 
 •H -H 
 
 
 rr 
 
 ^H 
 
 -O CO 
 
 x: B 
 
 -^ 
 
 bO 
 
 Ui 
 
 
 •-I 
 
 X5 
 
 CO -1 
 
 
 ^ bO 1-1 
 
 
 B C 
 
 
 50 
 
 — 1 c 
 
 CO > 
 
 «. 3 
 
 -^ 
 
 c 
 
 <u 
 
 
 01 
 
 T3 
 
 IJ ^ 
 
 
 c ^ 
 
 c 
 
 3 
 
 
 3 
 
 0) 1-1 
 
 V CO 
 
 3 ^H 
 
 1-4 
 
 CO 
 
 o. 
 
 
 ^ 
 
 4-1 0) 
 
 X5 V 
 
 
 4-1 CO -H 
 
 •H 
 
 M iJ 
 
 
 O 
 
 >« H 
 
 J z 
 
 o <: 
 
 'J1 
 
 X. 
 
 D. 
 
 
 o 
 
 CO oi 
 
 >• 
 
 
 CO Z C/5 
 
 H 
 
 <; to 
 
 
 k< 
 
 
 
 u 
 
 
 
 O 
 
 
 iH 
 
 CO 
 
 
 
 CO 
 
 
 CO 
 
 
 3 
 
 
 
 :x 
 
 
 
 O 
 
 
 z 
 
 u 
 
 
 
 U 
 
 
 u 
 
 c 
 
 QJ 
 CO 
 QJ 
 1-1 
 (X 
 QJ 
 1.1 
 
 CO 
 QJ 
 
 to 
 
 c 
 
 CO 
 I-I 
 
 c 
 o 
 
 •H 
 
 ■I-I 
 
 •H 
 CO 
 
 o 
 a. 
 S 
 o 
 
 T3 
 0) 
 
 •H 
 O 
 QJ 
 O- 
 CO 
 
 QJ 
 
 x: 
 
 4-1 
 O 
 
 1-1 
 
 o 
 
 QJ 
 bO 
 
 c 
 to 
 
 1-1 
 
 
 
 to 
 
 
 
 
 CO 
 
 
 
 
 c 
 
 
 
 
 3 
 
 
 
 
 o 
 
 
 
 
 x; 
 
 
 
 
 CD 
 
 
 
 
 QJ 
 
 
 
 
 1-1 
 
 
 
 
 QJ 
 
 
 
 
 x; 
 
 
 
 
 s 
 
 
 
 
 4-1 
 
 
 
 
 O. 
 
 • 
 
 
 
 0) 
 
 Pu 
 
 
 
 u 
 
 3 
 
 
 
 X 
 
 O 
 
 
 
 QJ 
 
 1-1 
 bO 
 
 
 
 CO 
 
 
 
 
 
 
 to 
 
 
 
 3 
 
 
 
 
 
 
 C 
 
 
 
 •H 
 
 •H 
 
 
 
 X 
 
 x: 
 
 
 
 CO 
 
 4J 
 
 
 
 B 
 
 •H 
 
 3 
 
 
 
 0) 
 
 
 
 
 l-i 
 
 3 
 
 
 
 to 
 
 o 
 
 rH 
 rH 
 
 
 * 
 
 to 
 
 to 
 
 
 QJ 
 
 Q) 
 
 
 
 rH 
 
 1-1 
 
 o 
 
 
 XI 
 
 3 
 
 •H 
 
 
 to 
 
 bO t4-i 
 
 
 O 
 
 •H 
 
 •H 
 
 
 •H 
 
 U-l 
 
 O 
 
 • 
 
 H 
 
 
 QJ 
 
 QJ 
 
 D. 
 
 4-1 
 
 Q- 
 
 U 
 
 CX 
 
 c 
 
 to 
 
 C 
 
 to 
 
 QJ 
 
 • 
 
 to 
 
 
 3 
 
 — 1 CO 
 
 rH 
 
 4-1 
 
 4J 
 
 a 
 
 to 
 
 O 
 
 ■r-l 
 
 C 3 
 
 CQ 
 
 z 
 
 4-1 
 
 to S 
 
 
 
 CO 
 
 x: -H 
 
 
 
 C 
 
 4-t C 
 
 pH 
 
 to 
 
 Ci. 
 
 < 
 
 o 
 c_) 
 
 QJ 2 
 
 CQ 
 
 z 
 
 
 I-I (N 
 
 o 
 
 s 
 
addition, lead is added to produce bear- 
 ing alloys; iron, silicon, and aluminum 
 may be used to provide improved mechani- 
 cal properties; phosphorus and boron (as 
 well as silicon) are used as deoxidizers 
 ( 18) . Mixing high-lead scrap with an 
 otherwise similar alloy (dilution with 
 cathode copper) would be particularly 
 costly in alloy applications where hot 
 workability demands very low lead con- 
 centrations. In general, cast copper 
 alloys can tolerate higher impurity lev- 
 els than wrought alloys since they are 
 not subjected to mechanical working. 
 Typically, impurities in wrought alloys 
 affect work response (e.g., hot and cold 
 shortness) or thermal response (e.g., 
 recrystallization, fire cracking), while 
 in cast alloys impurities generally af- 
 fect castability, soundness, and physical 
 properties. 
 
 WROUGHT ALLOYS 
 
 Wrought alloys are generally hot-worked 
 above the recrystallization temperature 
 of the alloy and thus are free from the 
 strain-hardening effects associated with 
 cold working. Cold working is normally 
 done below recrystallization tempera- 
 tures. Foulger (12) has discussed im- 
 purity effects on hot and cold worka- 
 bility in copper alloys. In cold work- 
 ing, impurities that affect the amount 
 of the relatively hard and brittle beta 
 phase must be carefully controlled. As 
 a general rule, beta and alpha plus beta 
 alloys are more easily hot worked, even 
 in the presence of impurities, while 
 single-phase alpha alloys require a much 
 higher purity level. Hot shortness in 
 wrought alloys is caused by such low- 
 melting elements as lead, antimony, 
 and bismuth and in some cases iron and 
 silicon. Even very low levels of impuri- 
 ties can be detrimental. Bismuth, for 
 example, has been considered detrimental 
 in amounts just sufficient to form a 
 single atom layer along grain boundaries 
 (6^, _12, 18) . Limits for lead and bismuth 
 based on commercial hot-rolling experi- 
 ence were reported by Jackson (17) for 
 four common copper alloys as listed in 
 table 3. 
 
 TABLE 3. - Limits for lead and bismuth 
 in alpha copper alloys for hot roll- 
 ing, weight percent (17) 
 
 Alloy 
 
 Pb 
 
 Bi 
 
 70:30 brass 
 
 0.02 
 .015 
 .004 
 
 0.002 
 
 Nickel-silver copper 
 
 Phosphor bronze (95:5) 
 
 .001 
 .0004 
 
 Cold shortness is a loss in ductil- 
 ity in metals worked at temperatures 
 below the recrystallization temperature, 
 Glickman ( 14 ) has shown that cold short- 
 ness is another intergranual phenomenon 
 caused by extremely low amounts of bis- 
 muth, tellurium, antimony, and boron seg- 
 regated along grain boundaries. 
 
 File cracking has also been attributed 
 to low-melting impurities, such as lead, 
 and those that form brittle intercrystal- 
 line films, such as bismuth (6^, 18) . 
 Copper alloys may alternately undergo 
 heat-affected-zone (HAZ) cracking when 
 lead, tellurium, bismuth, or sulfur is 
 present, even though actual fire cracking 
 does not take place (13) . 
 
 Recrystallization and grain growth are 
 also affected by impurities. For exam- 
 ple, Fargette ( 11 ) has shown that, at 
 least for relatively simple copper al- 
 loys, a few hundred parts per million 
 or less of silver, phosphorus, cadmium, 
 manganese, or tin can increase the re- 
 crystallization temperature. Recrystal- 
 lization is inhibited because impuri- 
 ties segregate in dislocations and grain 
 boundaries, reducing their mobility (^, 
 10). Impurities have also been shown to 
 affect grain size and orientation ( 22 , 
 ^). 
 
 CAST ALLOYS 
 
 Cast copper alloys are used for sand 
 casting or chill casting of various com- 
 ponents. Typically, alloy additions are 
 made to improve characteristics such as 
 hardness or corrosion resistance. Usual- 
 ly one characteristic is improved at the 
 expense of another. For example, iron 
 is added to some yellow brasses to in- 
 crease strength at the expense of reduced 
 elongation. Table 2 lists the range of 
 compositions for some principle cast- 
 ing alloys. Elements such as lead and 
 
antimony, which are in relatively high 
 concentrations in some alloys, could not 
 be tolerated in others. Cast alloys gen- 
 erally can tolerate greater amounts of 
 impurities than wrought alloys. The ef- 
 fect of impurities on cast alloys may al- 
 so depend on the method of casting, i.e., 
 whether sand cast or chill cast (32) . 
 Mechanical properties can be impaired in 
 certain tin bronzes by magnesium, sili- 
 con, aluminum, and even zinc. These ele- 
 ments have similar properties in that 
 they form oxide films on the surface of 
 molten bronzes which break up on pouring 
 and cause rough surfaces on the resulting 
 chill-cast products. Because film forma- 
 tion interferes with feeding within 
 the mold, fissure porosity can result, 
 greatly reducing mechanical properties 
 (33) . Proper segregation of scrap alloys 
 is also essential in that some silicon 
 bronzes contain small amounts of arsenic 
 and antimony. In bronze castings, arsen- 
 ic and antimony decrease the solubility 
 of tin, thus increasing and coarsening 
 the brittle delta phase, which impairs 
 the mechanical properties (1-^). Arsenic 
 and antimony concentrations as low as 
 0.12 and 0.18 wt pet were found to se- 
 verely reduce the tensile elongation of 
 tin bronzes (32) . 
 
 In general, the impurities tend to de- 
 crease the mechanical properties and 
 castability. It is also possible that 
 impurities such as aluminum and silicon 
 can cause casting voids, which serve as 
 sites for stress corrosion cracking. 
 
 Similar effects of impurities on phos- 
 phor bronzes , tin bronzes , and their 
 leaded alloys have been reported by Win- 
 terton ( 33 ) and Stolarczyk (32). For 
 example, table 4 illustrates the effects 
 of aluminum, silicon, iron, lead, and 
 bismuth on a typical phosphor bronze. In 
 general, an increase in impurities de- 
 creases tensile strength and elongation, 
 while hardness is variable. 
 
 Impurities can also cause microstruc- 
 tural changes in cast copper alloys. For 
 example, grain shape or size can be 
 changed, as in the case of iron in brass- 
 es (10). In other cases, solid solutions 
 or compounds are formed. It has also 
 been shown (26) that chromium, although 
 below specified maximum concentrations, 
 can lead to fatigue failure in a 
 manganese-nickel-aluminum bronze alloy 
 (75Cu-12Mn-8Al-2Ni-3Fe). Failure results 
 from chromium segregation and eventual 
 formation of a dendritic phase, leading 
 to coarse grains and weak structure. 
 
 TABLE 4. - Effect of impurities on phosphor bronzes (33-34) 
 
 
 Concentration, 
 wt pet 
 
 Tensile strength, 
 ton/in^ 
 
 Elongation, 
 pet 
 
 Hardness , 
 HB 
 
 Standard alloy. . 
 Aluminum 
 
 Bismuth 
 
 NAp 
 
 0.005 
 .01 
 .47 
 .32 
 
 1.25 
 .28 
 .80 
 
 1.69 
 .24 
 .70 
 
 1.61 
 .01 
 .07 
 .38 
 
 27.2 
 25.3 
 21.8 
 23.2 
 25.6 
 23.8 
 28.5 
 26.1 
 22.2 
 26.3 
 25.4 
 23.8 
 26.3 
 24.3 
 24.2 
 
 18 
 
 15 
 
 5 
 
 6 
 
 14 
 
 10 
 
 15 
 
 11 
 
 4 
 
 15 
 
 15 
 
 10 
 
 18 
 
 9 
 
 7 
 
 138 
 123 
 133 
 130 
 127 
 
 Iron 
 
 126 
 136 
 
 Lead 
 
 132 
 139 
 128 
 
 Silicon 
 
 125 
 126 
 119 
 
 
 131 
 138 
 
 NAp Not applicable. 
 
METALLURGICAL EFFECTS OF TRAMP ELEMENTS 
 
 The practical aspect of secondary cop- 
 per alloy production is presumably based 
 on achieving the best corrosion resist- 
 ance at the lowest cost. Cost in this 
 context refers to both processing and raw 
 material cost. The extent to which rela- 
 tively low-cost scrap can be utilized to 
 achieve required final quality without 
 excessive processing cost is well under- 
 stood by the industry. Millions of tons 
 of high-grade products testify to the 
 metallurgical acceptability of recycling. 
 As noted in the above overview section, 
 the effects of certain undesired elements 
 are well understood, and for the most 
 part they are effectively avoided. Some 
 metallurgical problems have arisen, how- 
 ever, where interaction effects among 
 various elements and/or certain micro- 
 structural behaviors were not well under- 
 stood. The following case studies were 
 selected as examples of the perhaps less 
 well known metallurgical effects of im- 
 purities in copper alloys and to empha- 
 size the importance of careful scrap 
 sorting and segregation. Some of the 
 examples also illustrate that corrective 
 metallurgical treatments may counteract 
 certain impurities. It is important to 
 note that although impurities are often 
 major problems, certain metallurgical 
 problems are due to factors other than 
 composition. 
 
 HOT WORKABILITY 
 
 Metallurgical and processing factors 
 associated with the hot workability of 
 brasses, bronzes, copper-nickels, and 
 nickel-silvers have been examined by 
 a number of researchers. Studies by 
 Cook (6^) , Davies (^) , Foulger ( 12 ) , and 
 Heslop (15) were selected for most of the 
 following discussion. The preceding dis- 
 cussion described various effects of al- 
 loying elements and impurities on copper 
 alloys. It is important to note that the 
 literature from which these data were 
 obtained also mentions the use of metal- 
 lurgical controls (e,g, , special alloy 
 additions or processing techniques) to 
 
 enhance hot workability. Even without 
 harmful levels of impurity elements, im- 
 proper metallurgical processing can cause 
 hot shortness. 
 
 Factors affecting hot workability in- 
 clude crystal structure, soundness, seg- 
 regation, and composition. The effect of 
 composition is highlighted here to illus- 
 trate some of the direct or potential 
 effects of certain impurity elements. 
 Alloys such as brasses, aluminum bronzes, 
 copper-nickels, and silicon bronzes may 
 contain up to 1 or 2 pet of various al- 
 loying elements to achieve desired 
 mechanical properties. Except for lead 
 and other low-melting elements, the al- 
 loying elements enter into solid solution 
 and generally have little or no effect on 
 hot workability. Interaction effects 
 from impurities at relatively low concen- 
 trations can have severe detrimental ef- 
 fects, however, although there seems to 
 be little published on this subject. It 
 is important to note that the quality of 
 the cast ingot prior to working and 
 response to homogenizing soak treatments 
 are extremely important factors affecting 
 hot working, A fine, uniform cast grain 
 structure, for example, generally results 
 in better hot workability and tolerance 
 for impurities than a coarse, nonuniform 
 grain structure. The severity of hot 
 working also determines hot workability, 
 such that a given alloy may fail during 
 hot piercing for tubemaking but form sat- 
 isfactorily during hot rolling or extru- 
 sion. Regarding workability of metals in 
 general, Semiatin (29) gives an excellent 
 description of the action of dislocations 
 and intergranular impurities on the for- 
 mation of voids at grain boundaries. 
 Secondary tensile stresses develop at the 
 voids during metal working operations, 
 eventually leading to fracture. Figure 2 
 is a schematic description by Semiatin 
 showing grain boundary voids and disloca- 
 tion impingement (inverted T's). Figure 
 2 also depicts triple-junction cracks, 
 which appear at the junction of three 
 grains and subsequently lead to fracture 
 during hot working. 
 
Triple-junction 
 crack 
 
 Grain-boundary voids 
 
 FIGURE 2. - Schematic representation of grain 
 boundary voids due to dislocation impingement on 
 grain boundary impurities. Also shown is a triple 
 junction crack (29). 
 
 Effect of Lead 
 
 Lead is among the most prevalent and 
 deleterious elements affecting hot worka- 
 bility. The maximum that can be toler- 
 ated depends on such factors as overall 
 composition, structure, and processing 
 techniques. 
 
 In brasses, lead is harmful at levels 
 as low as 0.01 wt pet. In phosphor 
 bronzes, Jackson (J_7) showed that 0.004 
 wt pet lead is considered the maximum 
 tolerable for hot rolling (table 3). In 
 other alloys, lead may be tolerated up to 
 0.02 wt pet, and even 0.05 wt pet or 
 higher. Extreme care is needed if alloys 
 containing lead for machinability are to 
 be hot-worked. For copper-nickel alloys, 
 bismuth, tellurium, and selenium must be 
 limited to levels as low as 0.001, 0.003, 
 and 0.006 wt pet, respectively (12). The 
 harmful effects of lead (and the other 
 low-melting elements) are attributed to 
 the general characteristics of low melt- 
 ing point and limited solid solubility in 
 the parent alloy. The lead and/or its 
 low-melting compounds segregate along 
 grain boundaries and are liquid at normal 
 hot-working temperatures. Hot shortness 
 may or may not occur, depending on inter- 
 facial tensions between the liquid phase 
 
 and the solid grains of the parent alloy 
 (_5, j7, 33.). The details of this pro- 
 posed mechanism are beyond the scope of 
 this paper. As stated earlier, the ef- 
 fect of such impurities may be enhanced 
 or diminished, depending on other factors 
 such as grain size control, other alloy- 
 ing elements, initial cast structure, and 
 processing techniques. 
 
 Effect of Lead and Iron 
 
 Figure 3 shows the effect of high lev- 
 els of lead on edge cracking of a 60:40 
 brass during hot rolling. The alloy used 
 in this study was relatively pure except 
 for lead and showed appreciable cracking 
 at 0.3 wt pet Pb and catastrophic crack- 
 ing at 1.0 wt pet Pb. The addition of 
 1.0 wt pet Fe increases the tolerance for 
 lead. 
 
 Studies by Davies ( 10 ) described the 
 beneficial effects of iron in lead- 
 containing brasses and offered a possible 
 explanation. Figure 4 shows the effect 
 of iron on hot rolling of brass alloys 
 containing approximately 0.8 wt pet Pb 
 (plus 1.3 wt pet Mn and 1.5 wt pet Al). 
 Figures 5 and 6 show the effect of iron 
 in a somewhat different manner, i.e., two 
 iron levels and varying concentrations of 
 lead. The actual mechanism by which the 
 beneficial effect of iron occurs is 
 apparently not fully understood. Davies 
 refers to other investigators who studied 
 cast beta brasses containing 3 wt pet or 
 more aluminum and related the intererys- 
 talline fraction to the segregation of 
 aluminum atoms along grain boundary ar- 
 eas. They felt that aluminum was being 
 replaced by iron atoms, which were less 
 harmful. In the alloys studied by Davies 
 and depicted in figures 4, 5, and 6, the 
 iron was thought to replace the lead 
 atoms. A typical composition of these 
 brass alloys is, in weight percent, 60 
 Cu, 1 Sn, 1.5 Al, 0.2 Ni , 0.8 Pb, balance 
 Zn. Iron, manganese, and lead concentra- 
 tions ranged between 0.04 and 1.34, 0.44 
 and 1.54, and 0.05 and 1.16 wt pet, 
 respectively. 
 
10 
 
 pet Pb 
 
 1.0 pet Fe 
 
 0.3 pet Pb 
 
 1.0 pet Fe and 1.0 pet Pb 
 
 1 .0 pet Pb 
 
 FIGURE 3. - Effect of lead, iron, and iron and lead on hot rolling of 60:40 brass (6). 
 
 80 
 
 70 
 
 60 
 
 50 
 
 40 
 
 30 
 
 20 
 
 10 
 
 Not cracked CD DD 
 
 KEY 
 
 Rolling temperature 
 
 650° C 800° C 
 
 ■ D 
 
 0.2 0.4 0.6 0.8 
 
 IRON, wt pet 
 
 1.2 
 
 1.4 
 
 FIGURE 4. - Effect of iron on hot rolling of a 
 complex brass containing 1.33 pet Mn at 650° and 
 800° C (10). 
 
 0.4 0.6 0.8 1.0 
 
 LEAD, wt pet 
 
 FIGURE 5. - Effect of lead and iron on hot 
 rollingof acomplexbrassat 650°C (JO). 
 
 1.2 
 
11 
 
 80 
 
 70 
 
 60 - 
 
 50 - 
 
 40 - 
 
 I- 30 
 
 20 - 
 
 10 
 
 0.2 
 
 0.4 
 
 0.6 0.8 
 
 LEAD. w1 pet 
 
 1.4 
 
 FIGURE 6. - Effect of lead and iron on hot 
 roiling of a complex brass at 800 C (10). 
 
 Effect of Lead and Zirconiutn 
 
 Figure 7 illustrates the severe effect 
 of lead on hot ductility of a copper- 
 nickel, and also shows how this effect 
 can be offset by zirconium additions 
 ( 16) . Other studies indicate that ap- 
 proximately 0.05 wt pet Ce and other rare 
 earths can improve the hot workability of 
 copper alloys. The beneficial effects 
 of zirconium are presumably due to pro- 
 nounced grain refinement, while rare 
 earths, and certain other elements (e.g., 
 thorium, uranium, and lithium) form 
 high-melting intermetallic compounds with 
 lead. 
 
 FIRE CRACKING 
 
 Fire cracking is a for^ of embrittle- 
 ment (cracking) that occurs during rapid 
 heatup to annealing temperatures. This 
 type of cracking has been the subject of 
 a number of studies, yet the mechanism is 
 not fully understood. In one study by 
 Sato (28) fire cracking was investigated 
 for aluminum brass, copper-nickel, and 
 
 700 
 
 FIGURE 7. 
 
 800 
 
 1,000 
 
 900 
 
 TEST TEMPERATURE, "C 
 
 Effect of zirconium and lead on hot twist ductility of a copper-nickel alloy (16) 
 
 1,100 
 
12 
 
 chromium-copper. It was concluded that 
 embrlttlement occurred as a result of 
 voids along grain boundaries, similar to 
 the behavior observed for alloys subject- 
 ed to high-temperature creep and tensile 
 tests. The embrlttlement due to void 
 formation was further found to be closely 
 related to several factors, including — 
 
 1. Magnitude of the residual stress 
 from cold working. 
 
 2. Heating rate. 
 
 3. Time at temperature. 
 
 4. Annealing temperature. 
 
 5. Grain size. 
 
 6. Reduction ratio during cold working. 
 
 The effects of impurities were not 
 described, implying that they were 
 minor compared with the above-mentioned 
 factors. 
 
 Effect of Chemistry and Processing 
 
 Isler (16) undertook a detailed study 
 to define the mechanism of fire cracking. 
 Table 5 shows results of preliminary 
 tests relating chemical composition to 
 fire cracking. The alpha alloys with 
 lead were clearly prone to fire crack- 
 ing, whereas the presence of beta phase 
 reduced the tendency. In the absence 
 of nickel, cracking could not be in- 
 duced. All cracks were intercrystalline. 
 A large number of tests conducted on 
 alloy A revealed that this alloy could 
 exist in a sensitive (to fire cracking) 
 as well as a nonsensitive state after be- 
 ing cold-worked and heat-treated in pre- 
 sumably identical fashion. This observa- 
 tion led to the conclusion that in addi- 
 tion to lead concentration and type of 
 crystal structure, other factors governed 
 sensitivity to fire cracking. Further 
 detailed studies were conducted to exam- 
 ine effects of residual stresses from 
 cold work, the rate of heating, grain 
 size, the role of porosity, and the role 
 of lead. Casting porosity was clearly a 
 major factor rendering alloys sensitive 
 to fire cracking. The effect of lead was 
 studied in considerable detail, providing 
 the major subsequent observations on 
 which a proposed mechanism was based. 
 
 TABLE 5. - Effect of chemical composition 
 on fire cracking (16) 
 
 
 Composition, 
 
 Phase(s) 
 
 Sensitive 
 
 Alloy 
 
 wt pet 
 
 present 
 
 to fire 
 
 
 Cu 
 
 Zn 
 
 Ni 
 
 Pb 
 
 cracking 
 
 A 
 
 62 
 
 19 
 
 18 
 
 1 
 
 a 
 
 Yes 
 
 B 
 
 62 
 
 20 
 
 18 
 
 
 
 a 
 
 No 
 
 C 
 
 62 
 
 24 
 
 13 
 
 1 
 
 a 
 
 Yes 
 
 D 
 
 53 
 
 38 
 
 8 
 
 1 
 
 a 
 
 Yes 
 
 E 
 
 47 
 
 41 
 
 10 
 
 2 
 
 a and p 
 
 No 
 
 F 
 
 51 
 
 42 
 
 6 
 
 1 
 
 a and p 
 
 No 
 
 G 
 
 71 
 
 28 
 
 ND 
 
 1 
 
 a 
 
 No 
 
 Liquid Metal Embrlttlement 
 
 Hearing tests on alloys A and D, table 
 5, revealed that both alloys cracked at 
 318° C, slightly lower than the melting 
 point of lead (327° C) . Microprobe anal- 
 ysis showed that lead particles contained 
 the main alloying elements of the matrix 
 (copper, zinc, and nickel). This and the 
 fact that lead is known to form low- 
 melting eutectics with these elements 
 (e.g., 318° C with 0.5 wt pet Zn) sup- 
 ported the conclusion that the fire- 
 cracking temperature coincided with the 
 melting point of the lead particles. The 
 fire-cracking temperature was found to be 
 independent of the matrix composition. 
 Liquid metal embrlttlement was thus sus- 
 pected as being responsible for fire 
 cracking, and a series of tensile tests 
 were run to show whether alpha nickel- 
 silver and lead formed an embrittling 
 couple. Figure 8 shows tensile ductility 
 (reduction of area) for alloys A, B, and 
 F at various temperatures. These results 
 (and supporting Charpy impact tests) 
 showed a marked drop in ductility at 
 about 300° C for alloy A, but not for 
 alloy B (which is the same as alloy A but 
 without lead). A ductility minimum also 
 occurred at about 300° C for all alloys, 
 owing to spontaneous strain aging for 
 the alpha-phase alloys (A and B) and 
 to intercrystalline cavity formation for 
 the alpha-beta alloy F. Figure 9 shows 
 strain-aging embrlttlement as evidenced 
 by the serrations in the stress-strain 
 curve of alloy B but shows no serrations 
 for alloy F. 
 
13 
 
 z 
 o 
 
 t- 
 
 o 
 
 z> 
 a 
 111 
 
 CO 
 
 80 
 
 70 
 
 60 - 
 
 50 
 
 40 
 
 30 - 
 
 Q~~ 
 
 \:>^ 
 
 XK 
 
 .^ 
 
 \ 
 
 20 
 
 10 
 
 \ 
 \ 
 \ 
 \ 
 
 KEY 
 
 Alloy A.awith Pb 
 □ Alloy B. a without Pb 
 ■ Alloy F, a* /3 with Pb 
 
 ^1 
 
 
 c 
 "5 
 
 Q. 
 O) 
 
 c 
 
 / 
 
 100 200 300 400 
 
 TEST TEMPERATURE. «C 
 
 500 
 
 FIGURE 8. - Effect of lead on tensile ductility 
 (reduction in area) of nickel-silvers (16). 
 
 Microprobe analysis was also conducted 
 to show that there is a marked concentra- 
 tion of lead along the fissures of the 
 advancing crack front during fire crack- 
 ing. This further supported the liquid- 
 metal embrittlement model. 
 
 Additional Studies 
 
 Other tests were conducted to study 
 stress relaxation and desensitizing phe- 
 nomena, the latter being related to dis- 
 locations piling against grain boundar- 
 ies. The influence of storage time after 
 cold work was also examined, as was the 
 effect of lead on alloys of alloy A 
 composition but with 0,2 and 2.5 wt 
 pet Pb. Both these and the previously 
 discussed results were related to the 
 Griff ith-Orowan fracture theory, leading 
 
 
 
 I- 
 
 V) 
 
 ou 
 
 
 \ 
 
 
 
 
 
 / 
 
 20 
 
 1 
 
 / 
 
 A 
 
 Alloy F 
 
 
 10 
 
 H 
 
 
 
 r^--^^_^___^ 
 
 - 
 
 
 lo 
 
 O 
 
 o 
 
 O ~~^ N^ 
 
 
 n 
 
 1 o 
 
 / o 
 
 f ^ 
 
 o 
 
 o 
 
 o 
 
 o 
 t 
 
 CO 
 
 "" 1 
 
 
 STRAIN 
 
 FIGURE 9. - Effect of lead on stress-strain be- 
 havior of nickel-silvers (alloy B; alpha alloy plus 
 lead; alloy F; alpha plus beta alloy plus lead) 
 (16). 
 
 to the following summary characteristics 
 associated with alloys sensitive to fire 
 cracking: 
 
 1. A second phase is present that 
 melts during heat up, forming an embrit- 
 tling couple with the alloy. 
 
 2. The melting point of the second 
 phase must be low enough so that the re- 
 sidual stresses are relaxed only slightly 
 when melting occurs. 
 
 3. High yield strength of the matrix 
 alloy permits substantial buildup of the 
 residual stresses. 
 
 4. The alloy must contain voids along 
 the potential fracture path, thus elim- 
 inating the need for crack initiation. 
 
 It was not clear from Isler's study 
 why the alloy with lead but no nickel 
 (alloy G) was not sensitive to fire 
 cracking. 
 
14 
 
 MICROSTRUCTURAL EFFECTS 
 
 Micros true tural effects, such as solute 
 element segregation at grain boundaries 
 and phase relationships , were considered 
 to some extent in the above discussions 
 on hot workability and fire cracking. 
 Other reported studies concentrated on 
 microstructural evaluations to deter- 
 mine the effects of various elements on 
 the micros true ture of tin bronze and 
 to determine the cause of failure of 
 manganese-nickel-aluminum bronze cast 
 propeller blades. 
 
 TIN BRONZES 
 
 Couture (9^) studied a tin-bronze alloy 
 (88 Cu-10Sn-2Zn) to determine the effects 
 of nickel, phosphorus, iron, lead, anti- 
 mony, sulfur, and silicon on microstruc- 
 ture. The alloy base composition and 
 ranges of additions studied are listed 
 in table 6. The purpose was to produce 
 microstructures that would help explain 
 the influence of these elements on the 
 properties of tin bronzes investigated 
 by other researchers. The observations 
 made from micrographs are highlighted 
 here. 
 
 TABLE 6. - Composition of tin bronze 
 alloy used for microstructure 
 studies, weight percent (8) 
 
 Element 
 
 Cu. 
 Sn. 
 Zn. 
 Fe. 
 Ni. 
 P.. 
 Pb. 
 S., 
 Sb. 
 Si. 
 
 Master alloy 
 
 Range of 
 
 
 impurities 
 
 87.0 
 
 (') 
 
 10.4 
 
 (') 
 
 2.4 
 
 (0 
 
 .017 -0.018 
 
 0.02- 1.88 
 
 .0015- .0018 
 
 .002-3.38 
 
 .003 - .004 
 
 .004-1.05 
 
 .004 - .005 
 
 .004-1.84 
 
 .004 - .005 
 
 .004- .030 
 
 .0016 
 
 .002-2.19 
 
 .002 
 
 .002- .035 
 
 'The master alloy was used to make all 
 subsequent melts with the various levels 
 of impurities. 
 
 Metallography 
 
 Base Alloy 
 
 Microstructure consisted of the contin- 
 uous copper-rich alpha phase (tin and 
 zinc in solution) and pools of the alpha- 
 delta eutectoid of tin bronzes. 
 
 Antimony 
 
 At 0.78 and 2.19 pet, antimony caused 
 an increase in the amount and size of the 
 delta phase. This was attributed to de- 
 creased solubility of tin as antimony 
 goes into solid solution. Grain size was 
 not affected. 
 
 Iron 
 
 At 1.88 pet Fe , the eutectoid pools in 
 the cast structure were more abundant and 
 coarser. The as-cast high-iron alloys 
 also showed a fine precipitate around the 
 eutectoid and a starlike constituent in 
 the eutectoid, both presumably iron-rich. 
 Heat treatment at 700° C (1,292° F) dis- 
 solved the delta phase and resulted in 
 dense precipitation of a fine, iron-rich 
 constituent throughout the matrix. Some 
 grain refinement occurred, ranging from 
 radially columnar at 0.48 pet Fe to equi- 
 axed (0.5-mm-diameter) at 1,88 pet Fe, 
 
 Lead 
 
 Delta phase and grain size were not af- 
 fected by lead additions. However, be- 
 cause of low solubility, lead particles 
 concentrated in the eutectoid. 
 
 Nickel 
 
 Additions up to about 1 wt pet had no 
 significant effect. At the 1.33- and 
 3.38-wt pet levels, however, the micro- 
 structure showed a more abundant delta 
 phase in large clusters than was observed 
 in the base copper-tin-zinc alloy. Alpha 
 
15 
 
 pools within the eutectoid were larger. 
 Theta phase, though observed in nickel- 
 containing bronzes by others, was not 
 present in the alloys studied by Couture. 
 Possible explanations for this difference 
 are the absence of lead, lower nickel 
 contents, and lack of equilibrium in the 
 test specimens. Grain shape and size 
 were also modified by nickel. Grains 
 changed from radially columnar 3/8 in 
 (9.5 mm) long to equiaxed grains 1/64 in 
 (0.8 mm) in diameter at 3.38 wt pet 
 nickel. 
 
 Iron 
 
 Iron was concentrated in the matrix. 
 The iron concentration in the eutec- 
 toid was well below that of the sample 
 average. 
 
 Nickel and Antimony 
 
 These were primarily associated with 
 the eutectoid. 
 
 Sulfur 
 
 Phosphorus 
 
 Phosphorus contents of 
 pet resulted in an intermet 
 presumed to be copper pho 
 which at low concentrations 
 uted mainly on the outside 
 alpha and delta eutectoid, 
 concentrations was distribu 
 eutectoid and in adjacent 
 matrix alpha phase. Phos 
 effect on grain size. 
 
 .1 to 1.05 wt 
 
 allic compound 
 
 sphide (CU3P) , 
 
 was distrib- 
 
 edges of the 
 and at higher 
 ted within the 
 
 areas of the 
 phorus had no 
 
 Larger inclusions were confirmed as 
 sulfides. 
 
 Tin 
 
 Tin concentration in the matrix was be- 
 low the alloy average. In the cored ar- 
 eas, it was similar to the average compo- 
 sition. The eutectoid showed about three 
 times the average tin composition. 
 
 Zinc 
 
 Silicon 
 
 Silicon additions resulted in a more 
 abundant and coarser eutectoid. Grain 
 size changed from columnar to equiaxed 
 even at the low addition (0.005 pet). 
 Grain size was reduced only slightly with 
 further additions. 
 
 Sulfur 
 
 At all levels of addition, sulfur 
 formed a large number of translucent dark 
 gray, probably complex, sulfides. Grain 
 size was not affected. 
 
 Matrix and cored areas were about the 
 same, owing to rapid zinc diffusion, 
 which was greater than in the eutectoid. 
 
 Grain Size and Shape 
 
 Grain structure was changed from colum- 
 nar to equiaxed with additions of nickel, 
 iron, or silicon. The grain refinement 
 achieved was much less than that obtained 
 in other studies with zirconium. Grain 
 size and shape were not affected by phos- 
 phorus, lead, antimony, or sulfur. 
 
 Eutectoid Composition 
 
 Microprobe Analysis 
 
 Microprobe analyses were conducted to 
 determine the distribution of the impu- 
 rity elements. Areas examined included 
 the matrix (center of the dendrite arms), 
 coring (surrounding the eutectoid), delta 
 phase (or eutectoid pools in the cases of 
 very small delta areas), and inclusions. 
 The following qualitative results were 
 obtained. 
 
 The binary copper-tin phase diagram 
 predicts alpha plus epsilon as the equi- 
 librium phases at room temperature for 
 tin bronze alloys. The eutectoid decora- 
 position of delta to alpha plus epsilon 
 (350° C) is sluggish, however, and delta 
 is retained down to room temperature. 
 Although the delta would be expected to 
 contain 32.6 pet Sn, the presence of more 
 than 2 pet Zn and additional elements 
 will alter the delta composition. 
 
16 
 
 Microsegregatlon 
 
 Metallographic examination revealed 
 that phosphorus and lead segregated in or 
 near the eutectoid as compounds or solid 
 solutions and were present in the matrix 
 between dendrite arms, thus confirming 
 they segregated in the last-to-f reeze 
 liquid. 
 
 FAILURE ANALYSIS OF BRONZE PROPELLERS 
 
 As part of an extensive failure analy- 
 sis program conducted on large propeller 
 blades, Raymond (26) performed detailed 
 microstructural analysis to determine 
 cause of failure. Two 6-ton cast propel- 
 ler blades failed while in service on a 
 U.S. Coast Guard icebreaker. The alloy 
 was a manganese-nickel-aluminum bronze, 
 selected for excellent mechanical prop- 
 erties and good erosion and corrosion 
 resistance in high-velocity seawater. 
 Foundry and welding characteristics of 
 the alloy are superior to those of con- 
 ventional aluminum bronzes. When proper- 
 ly alloyed, a stable microstructure of 
 approximately 50 pet alpha and 50 pet 
 beta is obtained. 
 
 Composition 
 
 Chemical analyses were compared among 
 samples from the failed blades and a 
 good blade, and against the specification 
 for MIL-B-21230 A, alloy 2 (table 7). 
 The compositions all appeared to agree 
 rather closely. However, a slightly 
 higher concentration of chromium was 
 noted in the samples from the failed 
 blades. Although within the allowable 
 
 maximum for "others," this small amount 
 of chromium resulted in microsegregatlon, 
 which led to an unstable microstructure, 
 also causing a degradation of mechani- 
 cal properties. For example, tensile 
 strength, elongation, and Charpy V-notch 
 energy were respectively 25, 50, and 75 
 pet lower than typical values. 
 
 Results of Analysis 
 
 The study reported by Raymond ( 26 ) 
 involved a detailed analysis includ- 
 ing microstructural analysis, mechanical 
 property evaluation, thermal analysis, 
 and fracture mechanics. The overall ob- 
 jective was to devise a corrective weld 
 repair and heat treatment that would 
 permit salvaging additional propeller 
 blades. This was successfully accom- 
 plished. The overall analysis was rather 
 detailed and beyond the scope of this 
 paper. The studies to define the micro- 
 structural effects of chroDd-um are high- 
 lighted here. 
 
 Chromium was isolated as the deleteri- 
 ous contaminant which formed an iron-rich 
 dendritic phase ("sparkle") during solid- 
 ification from the melt. A detailed ion 
 microprobe mass analysis on a dendrite 
 showed — 
 
 1. Dendrite was composed of iron, 
 chromium, manganese, and only a small 
 amount of nickel. 
 
 2. Surrounding phases were discontinu- 
 ous alpha, richer in copper than the dis- 
 continuous beta, which was richer in man- 
 ganese and iron. 
 
 3. Aluminum and nickel were uniformly 
 distributed between the alpha and beta 
 phases. 
 
 TABLE 7. - Chemical composition of propeller blades, weight percent (26) 
 
 Element 
 
 Failed 
 blade 1 
 
 Failed 
 blade 2 
 
 Good 
 blade 
 
 MIL-B-21230A 
 (alloy 2) 
 
 Cu 
 
 74.09 
 
 11.90 
 
 7.77 
 
 .015 
 3.42 
 2.42 
 <.004 
 .027 
 
 
 74.11 
 
 11.80 
 
 7.82 
 
 .016 
 3.28 
 2.59 
 <.0078 
 
 .041 
 
 
 74.30 
 
 11.95 
 
 7.90 
 
 .005 
 3.10 
 2.51 
 
 
 .06 
 
 
 71 (min) 
 
 Mn 
 
 11-14 
 
 Al 
 
 7.0-8.5 
 
 Cr 
 
 Not specified 
 2.0-4,0 
 
 Fe 
 
 Ni 
 
 1.5-3.0 
 
 Pb 
 
 0.03 (max) 
 
 Si 
 
 0.10 (max) 
 
 Others 
 
 0.05 (max) 
 
17 
 
 Based on these results, it was con- 
 cluded that trace amounts of chromium 
 caused the formation of iron-chromium- 
 manganese dendrites at high temperatures. 
 The surrounding matrix was thus depleted 
 of (1) the manganese necessary to sup- 
 press eutectoid decomposition and (2) the 
 iron necessary for fine grain structure. 
 Manganese suppresses eutectoid decomposi- 
 tion because, being soluble in copper, it 
 
 lowers the melting point of the alloy. 
 The decomposition of eutectoid beta re- 
 sults in a brittle ternary phase, ob- 
 served as fine lamellar precipitates in 
 the beta phase. Two detailed heat treat- 
 ments were devised for restoration of 
 mechanical properties, both requiring 
 that the eutectoid temperature of 675° C 
 (1,250° F) be exceeded. 
 
 SUMMARY 
 
 The role of recycling copper alloys has 
 been examined with regard to known metal- 
 lurgical effects that result from exces- 
 sive concentrations of alloying elements 
 or impurities. A profile of the industry 
 is briefly presented, followed by an 
 overview of the reported metallurgical 
 problems. A few of the problems were ex- 
 amined in more detail (case studies) for 
 explanations of possible mechanisms and 
 corrective procedures. 
 
 Wrought copper alloys are generally 
 sensitive to even minor amounts of low- 
 melting elements such as lead, bismuth, 
 and antimony. Depending on the impurity, 
 harmful effects have been observed for 
 concentrations as low as 0.004 pet, al- 
 though considerably higher concentrations 
 can be tolerated for many of the alloys 
 under appropriate processing conditions. 
 The predominant effects are hot and cold 
 shortness and fire-cracking tendency due 
 to grain boundary segregation of such 
 elements. Sensitivity to impurities is 
 related to phase relationships, with 
 single-phase alpha alloys being most sen- 
 sitive. Certain other elements, includ- 
 ing iron and silicon, may produce harmful 
 effects, but in the case of iron can also 
 counteract the harmful effects of low- 
 melting impurities. Increased tolerance 
 for otherwise harmful impurities is 
 achieved also with additions of zirconi- 
 um, rare earths, or uraniijm. 
 
 Cast alloys are considerably more 
 tolerant of impurities than wrought al- 
 loys, although various impurities can 
 have pronounced effects on castability 
 and mechanical properties. Some of the 
 elements known to adversely affect prop- 
 erties if not closely controlled include 
 manganese, silicon, aluminum, iron, 
 arsenic, antimony, lead, and bismuth. 
 
 Small amounts of chromium (0.015 wt pet) 
 caused serious failures in cast aluminum 
 bronzes owing to preferential segregation 
 and subsequent effects on transformation 
 kinetics and brittle behavior. 
 
 Although the effects of impurities have 
 been well defined for many alloy systems 
 and adequate information seems to be 
 available to reevaluate alloy specifica- 
 tions, it is important to note that 
 definitive information is not available 
 for all cases, owing to the complexity of 
 some alloys and the interaction effects. 
 The many processing variables and service 
 conditions further prevent an understand- 
 ing of impurity effects in commercial 
 alloys. 
 
 A major objective of the Bureau of 
 Mines recycling research has been to 
 improve the efficiency of recycling by 
 introducing new technologies to more 
 accurately identify and sort scrap metals 
 (2, 19-20, 27), thereby minimizing the 
 chance of introducing tramp elements. 
 The evidence clearly supports the often- 
 expressed concern that scrap must be 
 carefully and accurately segregated to 
 avoid harmful impurity effects when re- 
 cycled, but all metallurgical problems 
 are not attributable solely to impuri- 
 ties. Care must also be exercised to 
 assure that established metallurgical 
 treatments are followed when processing 
 recycled alloys and when putting them in- 
 to service. Unfounded claims that pro- 
 cessing or service failures are caused 
 solely by impurities from scrap do not 
 effectively serve the recycling industry, 
 the scrap user, or the Nation when con- 
 servation measures such as recycling are 
 so vitally needed in extending our min- 
 eral resources. 
 
18 
 
 REFEEIENCES 
 
 1, Bailey, A. R. The Intercrystal- 
 line Cracking of Binary and Complex Beta 
 Brasses Resulting From Moisture in the 
 Atmosphere, J, Inst. Met,, v, 87, 1959, 
 p, 380, 
 
 2, Bailey, R, R. The Effect of Com- 
 position on the Intercrystalline Cracking 
 of Sand Cast, Complex Beta-Brasses Under 
 Constant Tensile Load, J. Inst. Met,, v, 
 89, 1961, pp, 110-111, 
 
 3, Brown, R, D, , Jr,, W, D, Riley, 
 and C, A. Zieba, Rapid Identification of 
 Stainless Steel and Superalloy Scrap, 
 BuMines RI 8858, 1984, 23 pp, 
 
 4, Butterman, W, C, Copper, Bu- 
 Mines Mineral Commodity Profile, 1983, 
 18 pp, 
 
 5, Carrillo, F. V., M. H. Hibpshman, 
 and R, D, Rosenkranz, Recovery of Sec- 
 ondary Copper and Zinc in the United 
 States. BuMines IC 8622, 1974, 58 pp. 
 
 6, Cook, M, , and E, Davis, The Hot 
 Working of Copper and Copper Alloys, J, 
 Inst. Met., V. 76, 1950, p. 501. 
 
 7, Copper Development Association. 
 Application Data Sheet. New York, 1980, 
 36 pp. 
 
 8, Coutsouradis , D, , E, Diderrich, 
 J, Smets, G. Croco, and L, Pauwels. Ef- 
 fects of Trace Amounts of Impurities on 
 the Recrystallization Behavior of High- 
 Purity Tough-Pitch Copper. Metall. Rep. 
 CRM, V. 39, Jan. 1974, pp. 73-74. 
 
 9, Couture, A. Effect of Impurity 
 Elements on Microstructure of Copper-Base 
 Casting Alloys. Trans. Am. Foundrymen's 
 Soc. , V. 84, 1976, pp. 1-6. 
 
 10. Davies , D. W, Influence of Iron 
 Content on the Structure and Properties 
 of Wrought Complex Brasses (B,S, 2872: 
 CZ 114, 115, and 116), J, Inst, Met., v. 
 98, 1970, pp. 174-182. 
 
 11. Fargette, B. Interaction of Cold 
 Work, Recovery, Recrystallization and 
 Precipitation in Heat-Treatable Copper 
 Alloys. Met. Technol. , v. 6, No. 5, 
 1979, pp. 194-201. 
 
 12. Foulger, R. V., and E, Nicholls, 
 Influence of Composition and Microstruc- 
 ture on Mechanical Working Properties of 
 Copper-Base Alloys, Met, Technol, , v, 3, 
 No. 8, 1976, pp. 366-369. 
 
 13. Gavin, S, A,, J, Billingham, J, P, 
 Chubb, and P, Hancock, Effect of Trace 
 Impurities on Hot Ductility of As-Cast 
 Cupronickel Alloys, Met, Technol, v. 5, 
 No. 11, 1978, pp. 397-401. 
 
 14. Glickman, Ye E. , V. A. Likhachev, 
 V. V, Rybin, and Ue. N. Litvinov. Inter- 
 granual Cold Shortness Mechanisms of FCC 
 Solid Solutions of Copper-Antimony. 
 Phys. Met. and Metallog. , v. 46, No. 3, 
 1978, pp. 127-132. 
 
 15. Heslop, J., and A, R, Knott, Im- 
 purities and Workability, Met, and 
 Mater,, v, 5, No, 2, 1971, pp. 59-62. 
 
 16. Isler, P., and W. Form. The 
 Mechanism of Fire-Cracking, J. Inst. 
 Met., V. 100, 1972, pp. 100-113. 
 
 17, Jackson, R, J,, D, A, Edge, and 
 D, C, More, A Preliminary Assessment of 
 the Value of Minor Alloy Additions in 
 Counteracting the Harmful Effects of Im- 
 purities on the Hot Workability of Some 
 Copper Alloys, J, Inst, Met,, v, 98, 
 1970, pp, 193-198, 
 
 18, John Wiley & Sons. Kirk-Othmer 
 Encyclopedia of Chemical Technology, V. 
 7, 3d ed,, 1979, pp. 37-47, 84-86. 
 
 19. Marr, H. E. III. Rapid Identifi- 
 cation of Copper-Base Alloys by Energy 
 Dispersion X-Ray Analysis. BuMines RI 
 7878, 1974, 15 p. 
 
 20, Maynard, D, , and H. S, Caldwell, 
 Jr, Identification and Sorting of Non- 
 ferrous Scrap Materials, Paper in Pro- 
 ceedings of the Third Mineral Waste 
 Utilization Symposium, Chicago, IL, Mar, 
 14-16, 1972, IIT Res, Inst., Chicago, 
 IL, 1972, p. 255. 
 
 21, McMahon, A, D, Copper, A Ma- 
 terials Survey, BuMines IC 8225, 1965, 
 340 pp, 
 
 22, Murphy, S,, and C, J, Ball, The 
 Recrystallization of Tough Pitch Copper, 
 J, Inst, Met,, V, 100, 1972, pp, 225-232. 
 
 23. National Association of Recycling 
 Industries. Standard Classifications for 
 Nonferrous Scrap Metals. Circ. NF-80, 
 July 1980, 12 pp. 
 
 24, , Recycling Copper and 
 
 Brass, New York, 1980, 36 pp. 
 
19 
 
 25. Newell, R. , R. E Brown, D. M. 
 Soborof f , and H. V. Makar. A Review of 
 Methods for Identifying Scrap Metals. 
 BuMines IC 8902, 1982, 19 pp. 
 
 26. RaNTDond, L. Heat Treatment and 
 Weld Repair of Cast Mn-Ni-Al Bronze Pro- 
 peller Blades. Trans. Am. Foundrymen's 
 See. , V. 87, 1979, p. 537. 
 
 27. Riley, W. D. , R. E. Brown, and 
 D. M. Soborof f. Rapid Identification and 
 Sorting of Scrap Metal. Conserv. and Re- 
 cycling, V. 6, No. 4, 1983, pp. 181-192. 
 
 28. Sato, S. T. Otsu, and E. Rata. 
 Embrittlement of Copper Alloys During An- 
 nealing. J. Inst. Met., V. 99, 1971, 
 p. 118. 
 
 29. Semiatin, S. C, and G. D. Lahoti. 
 The Forging of Metals. Sci. Am., v. 245, 
 No. 2, 1980, pp. 98-107. 
 
 30. Smart, J. S. The Effect of Impu- 
 rities in Copper. Ch. in Copper, the 
 Science and Technology of the Metal, Its 
 Alloys and Compounds, ed. by A. Butts. 
 Reinhold, 1954, pp. 410-416. 
 
 31. Spendlove, M. Methods for Produc- 
 ing Secondary Copper. BuMines IC 8002, 
 1961, 41 pp. 
 
 32. Stolarczyk, J. E. The Influence 
 of Bismuth, Iron, Arsenic, and Antimony 
 in Sandcast Gun Metals. Br. Foundryman, 
 V. 53, Nov. 1960, p. 482. 
 
 33. Winterton, K. Tin Bronzes: Ef- 
 fects of Impurities in the Chill-Cast 
 Condition. Met. Ind. , v. 71, Dec. 12, 
 1947, p. 479. 
 
 iU S GPO: 1985-50>019/20,083 
 
 INT.-BU.OF MINES,PGH.,P A. 28048 
 
D DD 
 
 o 
 
 Q- 
 
 > ^ 
 
 a. o 
 
 9- 3 
 
 O 73 
 o n 
 
 o 
 
 o 
 
 O 
 
 3 
 
 (a 
 
 -a — 
 
 0) 
 CD 
 
 c: *: 
 O o c 
 
 H- « ^ 
 
 4! I -„ 33 ■-;: = 
 
 Q 
 
 3 
 
 a CD 
 m 2 
 
 u 
 
 -§ 
 
 m 
 
 ^ CD 
 
 i mo 
 
 < m z 
 > Z m 
 zc w 
 
 O 
 m 
 
 > 
 
 ii 
 
 m 
 
 > H 
 
 [Urn 
 
 m 
 3) 
 
 O 
 
 3 
 
 m 
 O 
 c 
 > 
 
 l— 
 
 O 
 
 TJ 
 TO 
 
 o 
 
 33 
 
 -\ 
 C 
 
 z 
 
 H 
 -< 
 
 m 
 
 O 
 
 -< 
 m 
 3} 
 
 H 52 
 
 ^ 55 
 
 4 
 
 c 
 v> 
 
 - m f" 
 z z > 
 
 I* -• z 
 h o o 
 
 o) "n -n 
 
 H m 
 
 I m 
 
 iti « 
 
 5> 
 
 m O 
 O 
 


 a5°^ 
 
 <i.^ . o « o 
 
 
 
 
 ■" *-,.# .-^-v \/ :^", \/ .•^\ \/ .•^^•. %,# •• 
 
 
 
 
 O. * e. » 
 
 
 
 
 n-i°^ 
 
 
 .^' 
 
 0- 
 
 

 oV 
 
 
 
 o 
 
 ""t.. -^a^.- ./°- 
 
 V^ 
 
 <^v. 
 
 O N O ' <^ 
 
 
 
 
 
 
 
 \/\ •.^^/ y '^^^ '-m^*' /'"^*^ .^,,^, 
 
 
 < O 
 
 .^*^°- 
 
 
 
 '^ -^^0^ ° 
 
 
 v-^;^ 
 
 0' s»'-% '^ 
 
 
 
 
 
 
 
 .^'\ 
 
 ^^ . 
 
 lO^ *»••' 
 
 
 
 'b V" 
 
 -^^0^ 
 
 ^^-•<^. 
 
 V ' ■ • ><:iiii:S : • ■ • ;v^* • ■ ■ ■ • V ■ • -^ 
 
 
 ^5* A. 
 
 • • 5 ^ > 
 
 i -.c,-^' 
 
 V 
 
 
 ^<^^^. ^^i- 
 
 ^s*^ "^E 
 
 % 
 
 g^x - V- c 
 
 
 
 > 
 
 MECKMAN I^Q 
 
 BtNOERY INC. ^^gff 
 
 ^, NOV 85 
 
 "*"^** N. MMCHESTOt 
 ^^^ IHOtANA 46962 
 
 
 
 
 •*bv^ . /;. -^^0^ ^^ 
 
 •j^5\^-k'^ »-? 
 
 .0' . - 
 
 
 0^ o»"«« "^o. 
 
 o > 
 
 ■>^- '^■°^' ' ■ >^ '■'^v*' ■■^^^^' 
 
■ ■ IHII HMWHinH 
 
 LIBRARY OF CONGRESS 
 
 
 mam