884
T H E J O U R S A L O F IAVDLTSTRI.4L A X D E Y G I N E E R I ~ V GC H E M I S T R Y
of t h e coal from a given bed in a n y particular district, t h e engineer can determine whether t h e coal he receives comes from t h e bed a n d t h e district s t a t e d , a n d whether i t is being prepared for m a r k e t as carefully a s i t should be. Wide variations in t h e compositions a n d heating values of t h e coals from different districts a n d from different beds make comparable analyses almost indispensable t o engineers having t o install boiler or gas-producer plants in different cities, as well as t o railroads a n d steamboat companies, a n d t o t h e engineers a n d purchasing agents of t h e various d e p a r t m e n t s of t h e United States Government. T h e Bureau of Mines report is in t w o p a r t s : one gives t h e methods used in collecting a n d analyzing t h e samples a n d t h e results of t h e analyses; t h e other
f
COPPERCOVERED ORCOPPERCLADSTEEL‘ METAL MADE BY ALLOYING OR WELDING COPPER AND STEEL By J A M E S OTIS HA\DY
Of t h e metals in common use, copper is t h e only one which occurs naturally i n large quantities i n t h e metallic s t a t e . T h e great native copper deposits of Michigan are t h e best illustration TThich could be found of t h e extraordinary resistance of copper t o corrosion. c T h e metals zinc, t i n , a n d lead. which are commonly used as protective coatings for iron a n d steel, do n o t occur i n t h e native s t a t e a n d do not approach copper in durability under atmospheric conditions. Their use is comparatively modern a n d in t h e case of t i n (now selling a t 4 1 cents per pound) is commercially possible only because of t h e exceedingly t h i n coat which m a y be p u t on ( 2 lbs. per box means a 0.00012 inch t i n coating). T h e necessity of protecting iron from corrosion a n d t h e desirability of strengthening copper alloys used for primitive tools a n d weapons led, a t a very early period, t o t h e use of a combination of a n iron core a n d a bronze covering. Such articles h a v e been found in t h e ruins of S i m r o u d , a n ancient Assyrian city, a n d among t h e remains of t h e Swiss Lake Dwellers (Friend: “Corrosion of Iron a n d Steel,” page 8). For similar reasons, t h e union of iron or steel with copper i n such a manner a s t o obtain a composite metal retaining unimpaired t h e good qualities of both metals has long been a desideratum. 1 I a n y inventors have given their attention t o t h e difficult problem of securely uniting iron or steel with copper. a n d t h e p a t e n t literature of t h e United States a n d of European countries contains m a n y illustrations. chiefly of unworkable or imperfect processes. T h e earliest p a t e n t me have found m-as one issued t o Poole; of England, in t h e year 1821. Copper or brass was melted i n a shallow cast iron pan (British P a t e n t 4 j 9 8 of 1821). 1 Paper presented at the 48th meeting of the American Chemical Society Rochester, September 8-12. 1913.
SO. I I
gives t h e exact location from which each sample of coal was t a k e n , together with a description of t h e characteristic features of t h e coal bed a t t h e point of sampling, t h e nominal capacity of t h e mine a n d such notes on t h e preparation of t h e coal as might be useful t o consumers. T h e d a t a contained i n these t w o volumes are not equalled in scope a n d detail a n d i n value for comparative purposes b y t h e figures t h a t have been published b y a n y other coal-producing country in t h e world. T h e Governments of some of these countries have published analyses of coals from different mines a n d from different districts b u t , with few exceptions, t h e samples of coal were n o t collected a n d analyzed under a uniform system t h a t would make t h e results comparable in all respects.
ORIGINAL PAPERS
THE MANUFACTURE, PROPERTIES AND USES OF COMPOSITE
VOI. 5 ,
I
T h e commercial electro-deposition or electroplating of copper on iron or steel seems t o have s t a r t e d before 1840. in England. This process h a s been used chiefly for t h e production of t h i n coatings, b u t processes have been developed b y Elmore a n d others for making, b y electro-deposition, copper t u b e s of considerable thickness. Such a , process was undoubtedly used commercially by t h e manufacturers mho, in t h e year 1883, supplied t o t h e Postal Telegraph Cable Company bimetallic wire consisting of a h a r d steel core wire o 1 2 0 inch in diameter surrounded b y a seamless, electro-plated, closely-fitting copper t u b e having approximately 0 . 0 4 2 inch walls. T h e ratio of steel t o copper b y weight \\*as a b o u t I : 3 (Figs. I a n d 2). T h e same company used also a compound wire consisting of a steel core wire surrounded first b y a t h i n electro-deposited copper sheath 0.008 inch thick a n d t h e n b y a wrapping of heavier copper with walls 0 . 0 4 4 inch thick (Figs. 3 a n d 4 ) . I n neither t y p e of wire was there a bond between copper a n d steel other t h a n a close conformation. T h e workmanship was so excellent, however. t h a t samples t a k e n down after 28 years’ exposure in t h e line from New York t o Chicago showed absolutely no corrosion of t h e steel core. T h e copper sheaths retained t h e dimensions given above. Desire t o solve t h e problem of uniting copper with iron or steel would naturally lead t o a t t e m p t s t o make a weld b y processes analogous t o t h e one successful with iron or mild steel. When, however. heating a n d h a m mering iron a n d copper together failed. even with t h e use of various fluxes, t o produce a weld, i t became apparent t h a t t h e problem was a difficult one. Inventors tried t o follow established methods for coating one metal with a more fusible one, e . g., t h e coating of iron or steel with zinc, t i n , or tin-lead alloy b y passing t h e former metals, properly cleaned. through b a t h s of t h e melted, fusible metals. T h e high melting point of copper a n d i t s strong tendency t o oxidize when melted, offered great obstacles. Coatings made b y dipping steel into melted copper are always t h i n if t h e steel is h o t , a n d are usually imperfectly adherent.
Nov., '9'3
T H E JOUR.VAL O F 1NDUSTRI:I L . I S D E,VGILVEERIAVG C H E M I S T R Y
The ideas of casting melted copper in considerable a m o u n t against an iron o r steel base or around L: steel core, or of casting iron or steel inside a copper t.iihi., were conceired and tried b y the following inventors: "llolleis for Printing Fabrics," British Pat. 1924 of I Q G ; Tytherlcigh. Method:-Dipping iron in melted copper; then heating and casting more copper around it.
FiC
I
x
11
"Poilal Telcginph" Wire with Steel Core and Entirely Electrodeposited capper coverine
Fa;.2
FIG 4
x
44
"Postnl Tcleraph" Wire wit11 Steel Core.
Electrodeposited Tube, and Folded-on copper cover
"Improvement in Coating Xetals," U . S. Pat. 21.79, oi r8j8; Jl~lh,od.-Cssting iron or steel in a copper mold and Comi)letiw the rolling. Hiler.
.~pec?knlion.--"I
x
44
Fro. 3 x l 1 "Postal Telcgrsph" Wire with Steel Core, Electrodeposited Tube, and Folded-on copper Cover
tact until both met& have become cooled to B propcr tt,rnperaturf.". , , . . . . . . . . ."On the melted iron h e i q pnnred into the mold in Cmtncl with the coating m c t d tire letlei bfcomer fused b y the keet of the Jornzer; bot before it is so fused the iron has parted with so much of its ticat as tp be sufficiently chilled to prevent the coating rnctal from miring with thc iron, and also to prevent the said coating mctal from sinking down and occupying tlic bottom of thc mold." C/oim.-"What 1 claim as my invention, and desire to secure by Lctters Patent, is: The coiting iron or stcel with copper, silver, or brass, or alloys whew silver or copper is used, by bringing the iron or steel, while in B mdtcd siate, into contact with the coating metal, and allowing them to so rem3iii until the two
Fir.. i X SO Commercial Alloy Union of Coppci and Iron. Luiigitudinal Section of ".Moiinot-DiiPlrX" wire
material. I place B core or axis in the iron cylinder, and adjust it, by screws or otherwise, in the axis of the mould. I put the whole into a furnace, and when the coating on the iron cylinder begins to molt I pour fused copper 01 copper alloy into the mould, and I maintain the heat until the copper or alloy is thoroughly incorporated with the coating on the iron cylinder; the fir? is then slackcned, and the whole allowed to cool. The roller or cylinder is aiterwards finished by the process of turning."
have discovered that the best union oi the
iron or SteCI aith its coating metal can be made by heating the iron or steel until it is fused and bringing it w h i l e in that conditian in contact with the coating metal, and keeping it so in con-
"Portal Telegraph" Wire w i t b Steel Core end Entirely Elecfiodrporiirci CoPprr covering
Spcrificatiorr.--"In carryin8 m y invention into effect, I take a hollow iron cylinder, oi somem.hnt less diameter than the rollcr or cylinder to be manufactured. I prefer to prrforate the said cylinder with holes. I clean the said .iron cylinder by acids or otherwise. I afterwards cover the said iron cylinder with borax or other snitable flux, and heot it sufficiehtly to fuse the flux thereon. In another iurnace I heat a pan or vessel (made of 8 material cqnble of bParing thc requisite beat), the said pan or vessel containing copper, brass, or other alloy oi copper. When the copper or alloy is iused, I put the iron cylinder thercin and &irn it, so as to coat i t with thc metal. I take thc cylinder from the said pan or vessel, and while hot I p u t it in B hollow cylinder or mould, closed at bottom with clay or othrr suitable
885
P m 6 x 44 Coninieicial Alloy Iiaion of Copper and Iroil. crosa section of "Duplex copperClSd" wire
metals haw become hard by coolins, .ubstantinlly as ipecificd." Suggested User of Hder Produrt.-Stair rods. trunk nails and
bands, hittons, etc. "Improvement in Line-Wires for Telegraph," 1T. S . Pet. 47,940 oi 1865: Farmer and Sfillikcn. Method.-Casting copper around ail iron bar: then rolling and drawing into wire. . This seems to be the pioneer patent on telegraph wire composed of copper with a steel core.
886
T H E J O r R S A L OF ISDL-STRIAL i l S D ESGISEERI,VG CHEMISTRY
Specification.-”In the construction of aerial telegraph lines the general practice, as is well known, is to use iron wire for transmission of the electric current, this wire being galvanized or covered with a thin coating of zinc to prevent oxidation. I n the earlier constructions of lines it was sometimes attempted to use copper wire, but for want of tensile strength in such wire its use had to be abandoned. Yet the employment of copper wire is a desideratum on account of its great superiority over iron as a conductor, and could its tensile strength be made equal to that of iron it would, to a great extent, supersede the iron wire.” “Our invention, therefore, consists in a copper telegraph wire cored or covered, for the purpose of strength, with iron or steel.” “In carrying out the invention or constructing this compound wire we prefer t o core the copper with iron, first casting the copper around an iron bar and then by the processes of rolling and drawing attenuating the same into wire by the well-known methods of wire-drawing. It will be obvious, however, that the iron can be placed around the copper and the resultant bar drawn out into a wire, leaving the copper in the center; but the first-named construction we consider preferable, as the iron is protected from oxidation, and the manufacture of this wire is the more practicable.” Claim.--“As a new article of manufacture, a telegraph wire reenforced, for the purpose of strength, with a core or cover of iron or steel, the wire being made by drawing a compound bar of the two metals.” “Improvement in Combining Copper with Cast Steel,” U. S. Pat. 73,375 of 1868; Park. ,$fethod.-Casting steel in a hollow copper ingot, previously heated. Specification.-“A combination or union of copper with some harder and stronger metal has been long needed for many purposes in the arts and manufactures, such as locomotive fire-box sheets, wire for ships’ rigging, rods, bolts, spikes, and various other articles for which copper alone cannot be employed, owing to its softness and want of strength.” “My invention consists of combining copper uTith cast steel, by casting the molten steel on or around the copper, which is previously heated, thus forming an ingot which may be hammered, rolled, or otherwise worked into any desired shape, the steel being tempered or hardened as may be desired after it is worked.” “If it is desired to have steel in the center and copper all around it, as is required for making copper-coated steel wire, or copper-coated spikes or rods, I prepare a hollow ingot of copper, into the center or cavity of which, after it has been heated to a good red heat, I teem or pour the molten cast steel. In this case no mould would be necessary, but may be used if preferred. Such ingot may be drawn out into wire, or worked into other articles, such as bolts, spikes, etc.” Claim.-“Combining copper and cast steel by heating the copper to a good red heat, and teeming or pouring thereupon liquid molten cast steel, substantially as and for the purposes hereinbefore described.” “Process of Manufacturing Compound Telegraph Wire,” U. S. Pat. 310,995 of 1885; Farmer. Method.-Copper plated iron wire drawn through melted copper. claim.-"^. The method herein described, of manufacturing compound telegraph wire, which consists in electroplating a steel or iron wire with a thin film of copper, then drawing the same through molten copper and shaping the coating in substantially the manner set forth.” “Improvements in the Llanufacture of \Vires, French Pat. 168,133 of 1885; Llartin. hfethod.-(a) Casting copper in iron tube. ( b ) Casting copper around medial part of iron bar and then reheating and rolling, etc.” In an 1888 supplement t o t h e a b o v e p a t e n t he described casting copper a r o u n d t h e medial p a r t of a
T‘ol. j, NO. I I
steel bar. H e t h e n rolled a n d drew i t . M a r t i n seems t o h a r e introduced bimetallic wire very extensively i n France a n d elsewhere i n Europe. His work appears t o be t h e commercial development of t h e ideas of F a r m e r a n d Milliken. Claim.-“I claim in this supplement to my principal patent my perfected method of making bimetallic wires in which the metal which forms the exterior envelope is more ductile than that which forms the interior core, this method consisting essentially in casting the metal intended to form the envelope around the medial portion of the rod which must form the heart of the wire, and in extending the rod so partially covered by the metal of the envelope by alternate passes first in one direction and then in the other between the rolls of a rolling mill.” “Process of Producing Compound Metal Bodies,” V. S . Pat, 853,716 of 1907; and “Copper Clad Iron and Steel,” 893,932 of 1908; Llonnot. Monnot develops f u r t h e r t h e plan suggested b y Tytherleigh, F a r m e r and Milliken, a n d hlartin. H e uses supermolten, i. e . , very h o t me‘lted copper t o give a t h i n initial coating and t h e n casts copper around t h e coated steel billet. H e finally reheats a n d rolls. Specification.-(U. S. Pat. 853,716.) “When a t the supermolten temperature many metals are very sensitive to flame, gases and other bodies, readily becoming impure ; and there is reason to believe that a t such temperatures suck metals have a solvent action on other solid metals placed in them. “By applying only a thin film coating by the action of supermolten metal, and then applying the main coating by means of metal nearer its point of solidification, there is less probability of reduction of quality of the coating metal. However, the m a i n coating m a y be and frequently is formed entirely f r o m supermolten metal, this method having the important advantages of requiring less manipulation and fewer baths of molten metal.” Claims.-(U. S. Pat. 893,932.) “ r s t . As a new article of
manufacture, a ferrous metal base, having a welded-on, continuous, poreless, dense coating of copper united thereto by a union resisting temperature changes, cutting tools, and mechanical stresses, said coating hauing the properties of metal set f r o m a liquid state. “2nd. As a new article of manufacture, an extended metal
article comprising a ferrous metal base and a continuous, poreless, dense coating of copper weld-united thereto, by a union resisting temperature changes, cutting tools, and mechanical stresses; said ferrous base and copper coating having been extended together.” In all of t h e above processes except those of T y t h e r leigh a n d N o n n o t , heating and rolling are essential t o complete t h e union between t h e t w o metals. T h e unions are more or less perfect according t o whether t h e operator reads i n t o t h e p a t e n t s effective means of keeping t h e a b u t t i n g surfaces clean at t h e critical i n s t a n t . T h e products of most of these processes cont a i n copper a n d iron alloys because steel dissolves readily i n melted copper a n d melted steel dissolves copper with great speed. Copper i n steel or iron i n copper are disadvantageous electrically because t h e y h a r d e n t h e principal metal a n d diminish t h e conductivity of composite wire. Copper i n steel retards its corrosion (Buck, THIS J O U R S A L , 5 , 4 4 7 ) . It is possible b y skilful regulation of temperatures of casting a n d b y proportioning correctly t h e a m o u n t s of melted metal a n d of t h e solid metal core or t u b e , t o avoid alloy formation, except i n t h e Tytherleigh, P a r k . and hIonnot procedures. In t h e latter, an
Nov., ' 0 ' 3
T H E J O U R N A L O F I.VDCSTRI.1L
alloy is intentionally made, a n d t h e f a c t i s considered a n i m p o r t a n t mechanical advantage, counterbalancing, i n t h c inventor's opinion, t h e highcr electrical rcsistance. There h a r e becn a number of a t t e m p t s t o utilize hydraulic pressure and means other t h a n rolling pressure t o weld cast copper while still h o t t o steel. As examples of these are t h e folloming: "New Process of Manufacturing Bimetallic Plates and \Vires," Belgian Pat. 171,442 of 1903, Ma&z. Claim.-"A Iroccss of making bimetallic plates and wire, which consists in applying on the inside walls of a conical ingot mold, widened a t the bottom, a lining or sheet of copper, of silver, of brass, etc., covering all or a part of the surface of the ingot mold, then mnning the steel into the mold, and (finally) forcing out the ingot through the small end of the mold by the use oi a hydraulic press whose piston presses against the large end of the ingot, in such a way as to weld t h e two metals and to suppress all flaws between them." "Process of Making Bimetallic Products," U. S. Pat. 853,932 of 1907,Monfrot and Martin. The chief f e a t u r e of these patents, which both cover t h e same processes. is t h e forcing of a cast, composite
oi the plates from scnle, oxid, or other impurities, then coating or covering om or illore faces of this rlate with a mctal or a11alloy t h w d Ir:iving a luwcr fusing point. then applying to this coated surface a iilatc of rnctai or an alloy of thc same character as the coating, but irre from any coating. then applying heat t o the tl5-0 plates, and Finiilly uniting by pressure, substantially as described." The new mcthods introduced b y Griffith n.ere t h e plating of t h e iron with copper, either by chemical or electrical deposition, a n d t h e n placingcopperandcoppercoated iron in close contact, heating t h e m u p a n d rolling t h e m together a t t h e proper temperature, T h e copperplating of t h e iron prevents i t s oxidation a n d keeps i t s surface in condition t o unite firmly with t h e copper when t h e t w o metals receive a proper rolling pressure at t h e right temperature. T h e temperature recommended b y Griffith was a b o u t 1 7 0 0 " F. T h e writer h a s observed t h a t at 9jo" C. (1742' F.) welding of copper t o copper-plated iron, rolled as composite cylindrical billets, was complete a f t e r t h e first few passes. Welding b y rolling pressure is t h e most effective method. Forcing through a die or otherwise
FIG.8 X 44 Commcirial \Veld Union a1 Copper and Iron. Cross Section 01 "Staiiclaid C. C. C." mire
ingot through a die, o r t h e use of hydraulic pressure on properly heatcd, flat slabs. in order in both cases t o get a weld bcfore f u r t h c r rvorking. This practice, even if successful. seems expensive a n d is superfluous if heating a n d rolling are t o follow. I n t h e process of Gvi,flit/i (L-. S. P a t . j80,34.i of 1 x 9 7 ) for "Cniting or Welding hietal," there was a return t o t h e original idea of wclding without first melting either copper or iron. Specifica/iorr.-"To this end the invention consists in the process hwcinaiter drscribcd, in which the plate, bar or other article of iron or stzel is first cleansed from scale or oxid: its surface or surfaces thcn covered with B contiiig or deposit of copper or an alloy oi CoSPcr or any other suitable metal or alloy thcreoi capable of being deposited by chemical or electrical deposition or otherwise, in this instanre hy the action of a bath oi a solution of the salts oi copper; is then placed iace to face with a sheet or illate of coppcr or an alloy thereof; is thcn heated to the proper degree, and, finally, is passed through rolls or is suhjected to pressure requisite to elTect the perfect and intimate welding of the metals." Clain-"The herein described process of uniting or welding plates of metal of different kinds, mhich consists in cleansing one
1:xo 11 x / I Alloy Union of Coppci and iron ai 113V C.
using hydraulic pressure are very ciimbersomc methods. G~~~~~ pat. 152,042 1903, ~ ~ ~ h ~ i t ~ . This inventor uses aluminum as intermediary in effecting a union of copper a n d iron. H e rolls clcan aluminum a n d copper sheets together at a b o u t Son" F. H e rubs metaiiic aluminum into t h e surface of clcan iron. His first operation i n sheet mnnnfacturc is t o hind t h c copper-aluminum sheet t o t h e ironaluminum slab a n d h e a t a n d roll. These complicated operations are said t o produce a secure weld. v. s. Pat. I,O6S37'7 Of '9'3, Rockey Eldridgc. This is an to *lake copper.clatl steel b y dipping i n t o sllccessive baths of mcltcd copper ered ,\.itl1 ,nelted boric nnl,yilride, is melltjoned only t o show t h a t efforts are still being made along t h e lines indicated. It has noted that weided composite made b y t h e Griftith process contains little or n o copper-iron alloy. Spring, i n 1878, b y pressure alone (without heat), produced alloys of certain mixed metal
T H E J O r R N A L 0 F I N D C S T RI A L A N D E N G I N E E R I X G C H E M I S T R Y
888
filings (Law, “Alloys,” page 3). I t has also been asserted t h a t solid metals m a y be united in t h i n layers a t t h e point of contact b y t h e use of heavy pressure a n d t h a t t h e union is a n alloy. It might naturally be expected t h a t when forced together a t 9 j o ” C., copper a n d iron might superficially alloy. T h e fact remains, howzver, t h a t copper-clad steel wire made b y t h e “Colonial” a n d “ S t a n d a r d ” developments of t h e Griffith process,
v01. j , N O . 11
TABLE11-COIDUCTIVITIES BASED ON PERCEITAGES BY WEIGHT, COICOPPERBEING ASSUMEDAS 100 STEEL A S 1 4 . 3 C Conductivity :I Conductivity of the steel Yo copper total
DUCTIVITY O F
B
7csteel 0 10 20
30 40 50
55 60 65
io -_ I ,
80 90 100
100 90 80 TO 60 50 45 40 35 30 23 20 10 0
0 1.43 2.86 4.29 5.7’ 7.15 7.87 8.58 9.30 10.01 1 I .oo 11.44 12.87 14.3
TABLEIII-cOIDUCTI\rITY
“IVelded” type , , “Alloyed” t y p e . . .
, ,
SOFT
A N D OF
Ratio ‘4 :
c
100.0
91.43 82.86 74.29 65.72 5 i . 15 52.87 48.58 44.30 40.01 34.00 31.44 22.87 14.3
1 I 1 1
: 1 : 1.016
1 1 1 1 1
: 1,l7,5 : 1.215
1.036 1,060 1.095 1.143
: : 1 : 1 :
: 1.266
: 1.333 : 1.478 I : 1.570 I : 2.287
. ...
STEEL \ T I R E S C O S T - A I S E D I S COPPERCLADSTEEL L l a r i m u m Rlinimum ..\rerage
OF SOFT
.,, , .,. .
12 samples 9 samples
14.41cq 14.021;
14 0 4 r l 13.44(;
14 28‘;
13.745;
FIG. 9-wIRE-DRAU’ISG BENCH T h e Dies are a t t h e Left of Each Ree!. Each Die and Reel F o r m an Independent Unit
has a conductivity almost t h e same as t h e s u m of t h e conductivities of t h e iron a n d copper composing i t . This is n o t t r u e of copper alloyed with iron nor of t h e products of other processes with which t h e writer is familiar. TABLEI-COSDUCTIVITIES BAWD DUCTIVITY
OS PERCENTAGES BY VOLUME, CONof COPPERBEING ASSUMEDAS 100 AID of SOFT
STEELAS 13.5
B
% steel 0
IO 20 30 40 50 55 60
65
io 77 80 90 100
Conductivity of t h e steel 0 1.35 2.70 4.05 5.40 6.75 7.43 8.10 8.78 9.45 10.40 I O . 80 12.15 13.5
.4 7ccopper 100
90 80 70 60 50 45 40
35 30 23 20 IO 0
C Conductivity total 100.0
91.35 82.70 74.05 65.40 56. i 5 52.43 48.10 43,78 39.45 33.40 30.80 2?,15
13.5
Ratio
A : C 1: 1 : 1 : 1 : 1 : 1 : 1 : 1 : 1 : 1 : 1 : 1 : 1:
1 1.015
1.034
FIG. 10-CONTIIUOUS mrIRE-DR.4WISG hfACHINE Six Size Reductions in One Operation
1.058
1.090 1.135 1.165
TABLE IV-ACTUAL CONDUCTIVITY PARED
UITH
OF COPPER-CLAD U ‘ I K E SAXPLES C O M C.4LCULATED COSDUCTIVITY
Conductivity
1.202
1.251 1.315 1.452 1.540 2.215
.....
T h e accompanying tables show how remarkably uniform t h e “ w e l d e d ” copper-clad wire is a n d how closely t h e actual a n d t h e calculated conductivities correspond. I n t h e “ a l l o y e d ” t y p e occasional samples show close agreement, b u t i n general there is a noticeable deficiency a n d in several cases t h e low a c t u a l C O Y I d u c t i v i l i e s are remarkable. T h e b e t t e r showing of t h e “ w e l d e d ” t y p e is undoubtedly due t o t h e high p u r i t y of t h e copper coating, which is free from iron alloy or oxide of copper. I n t h i s connection i t is of interest t o endeavor t o define “ w e l d ” a n d t o compare welded with alloyed copper-clad steel. A weld, until t h e a d v e n t of electric a n d oxy-acetylene methods of uniting metals, meant simply t h e join-
Copper
’’ Welded ” type. , . , . . . , . . . . . “Welded” t y p e . , . . , , . , . . . . . “ IVelded” type. . . . . , . , . . . . . “ V’elded” t y p e . . . . . . . . . . . . , “XVelded” t y p e . . . . . . . . . . . , . “Welded” t y p e . . , . . . . . . , . . . “Welded” t y p e . . , . , , . , . . . . , “ Welded” t y p e . . . . . . . . . . . . . “\Velded” t y p e . . . . . . , . . . . . , “Alloyed” t y p e . . . . . . . , . . . . . “ Alloyed” t y p e . . . . . . . . . . . . . “Alloyed” t y p e . . . . . . . . . . . . . “.-Uloyed” t y p e . . . . . . . . . . . . . ‘’ Alloyed” t y p e . . . . . . . . . . . . . “Alloyed” t y p e . . . . , . . . . . . . . “Alloyed” t y p e . . . . . . . . . , . . . “Alloyed” type. . . . . . . . . . . . . “Alloyed” t y p e . . , . . . . . , . . , . “.4lloyed” t y p e . . . . . . . . . , , . . “Alloyed” t y p e . . . . . . . . . . . . .
22.8 % 24,06 20.15 23.44 23.42 23.65 23,06 23.66
24.32 23.59 44.00
43.50 34.20 22,53 33,65 41.20 22.75 31.90 38.86 40.35 35.20 35.10 38,80
Actual 3 5 ,20cc 35.80 32.24 35.25‘ 35,45
1
35.11 35.43 35.13 36.12 35.42, 34.00 45.72 43,80 33.92 41 ..57 47.42 33.52 38,49 46.81 48.91 43.52 45.47 45.06
Calculated 34.60% 35.68 32.31
35.28
52.57 5 2 . 15
44.27 34.38 43.80 50.20 34.57 42.32 48.21 49.48 45.11 45.03 48.16
ing securely b y pressure of t w o pieces of heated, plastic metal. Usually both pieces of metal were of iron
T H E JOURNAL OF INDCSTRIAL A N D ENGINEERIXG CHEMISTRY
8 90
copper-clad steel is described b y Tassin.’ I t consists essentially i n preparing round steel billets approximately 5 inches i n diameter a n d 2 6 inches long, which a r e cleaned b y pickling a n d t h e n heated u p in molds which serve t h e double purpose of exclud n g t h e air a t first a n d finally of receiving t h e annular copper jacket which is cast a r o u n d t h e steel core. Producer gas is r u n t h r o u g h t h e mold t o hinder oxidation of t h e steel a n d copper. T h e h e a t e d billet a n d mold are lowered i n t o a b a t h of w h a t is called b y M o n n o t , t h e i n v e n t o r , ‘ ( s u p e r molten c o p p e r ” having a t e m p e r a t u r e approximating 2 s j O o F., for t h e purpose of forming a film of copperi r o n alloy on t h e surface of t h e steel billet. F o r t h i s purpose t h e mold is opened a f t e r i t enters t h e copper b a t h a n d is closed again as i t leaves i t . T h e mold a n d billet are t h e n lowered i n t o a second copper b a t h which has a very much lower t e m p e r a t u r e a n d t h e mold is t h i s t i m e filled with copper a n d is closed a n d withdrawn from t h e b a t h i n t h a t condition a n d allowed t o cool. T h e copper-covered steel billet is t h e n pushed o u t of t h e mold a n d heated a n d rolled i n t o rods which are afterlT-ards drawn i n t o wire. T h e object i n having t h e first copper b a t h a t so high a t e m p e r a t u r e is s t a t e d t o be t h e hastening of t h e f o r m a t i o n of a film of adherent copper-iron alloy. T h e reason for casting t h e greater p a r t of t h e copper a t t h e ordinary casting t e m p e r a t u r e . m-hich is a b o u t 2 0 0 ’ above i t s melting point, is t h a t i t is t h e n possible t o keep t h e copper f r o m rapid oxidation a n d also, which is more i m p o r t a n t , t o prevent t h e speedy contamination of t h e b a t h with copper a n d iron alloy. Kumerous experiments made b y t h e writer have shown t h a t copper a t z j 5 0 ° F. dissolves solid steel, or alloys Kith i t , with remarkable rapidity. t h u s making t h e copper unfit for use as a p a r t of an electrical conductor. T h e welding process of uniting copper a n d steel is (2) carried o u t as follows: Soft steel billets, approximately 3 I l 4 inches i n diameter b y 7 2 inches long. are carefully freed from scale a n d are electro-plated with copper. T h e y are t h e n enclosed i n neatly fitted cast copper t u b e s having walls of such thickness as t o f u r nish approximately z j per cent of copper i n t h e composite blank, t h e t o t a l weight of each blank being approximately 2 0 0 pounds. T h e use of copper t u b e s gives a uniform p r o d u c t not otherwise obtainable. I n o r d e r t o exclude t h e air t h e blanks are coated a t t h e ends with a fire-resistant paste. T h e y are t h e n h e a t e d t o a t e m p e r a t u r e approximating 9 j o ” C. ( 1 7 3 2 ~ F.),t h i s being sufficient t o render b o t h metals plastic. E a c h blank is t h e n rolled i n such a way t h a t it is giyen a strong a n d uniform rolling pressure a t all points, while t h e metals still retain a large p a r t of their original heat. It has been f o u n d t h a t t h e m-elding operation is completed r i t h i n jo seconds from t h e t i m e t h e composite blank leaves t h e heating furnace a n d t h a t t h e original 6 foot billet is extended i n I ~ minutes t o a z j f o o t wire b a r 18,4 inches square. During t h e first f e w passes there is a slight extrusion of copper f r o m t h e ends of t h e composite b l a n k ; t h e remainder. howeirer, is securely welded a n d t h e small THISJOURXAL,
1, 6iO.
1’01. 5 , KO. 1.1
extruded portions are sheared off from t h e finished wire bar. These wire b a r s are reheated a n d rolled i n t o coils of 3 , inch wire rods, These a r e freed from oxide b y pickling a n d are subsequently d r a w n cold i n t o wires of a n y desired fineness (Fig. 13).
FIG 13-IN
STORE-ROOM A T WIRE
Copper-Clad Wire Bars in Foreground Wire and Cable
?IIIILL
Coils of Copper-Clad
I t is remarkable t h a t t h e copper after i t is welded
t o t h e steel b y t h e preliminary rolling process adheres firmly t o it a n d extends evenly with i t through t h e various processes of h o t rolling, cold drawing, annealing, e t c . T h e r e is a strong stripping force in wire drawing which would i n s t a n t l y remove t h e copper jacket if i t v,-ere n o t securely united t o t h e steel. T h e \Triter h a d t h e o p p o r t u n i t y of observing t h e drawing of copper-clad wire of m a n y sizes, from 0.30 inch i n diameter t o 0.003 inch, t h e l a t t e r being as fine a s hair (samples s h o w n ) . K O stripping took place i n a n y case a n d t h e finest wire showed by chemical analysis t h a t t h e copper covering still maintained nearly t h e s a m e ratio t o t h e steel core which existed in t h e original wire b a r . Wire dran-ing m a y be likened i n stripping tendency t o t h e drawing of a lead pencil through a hole of t h e size of t h e graphite core only. Supplementing t h i s evidence of secure union of t h e t w o metals b y t h e “ w e l d i n g ” process is t h e fact t h a t numerous heating, quenching, a n d twisting tests of welded copper-clad wire of all sizes produced no separation of copper from steel. P R O P E R T I E S A S D r S E 5 O F COPPER-CLAD S T E E L
T h e principal use of copper-clad steel a t t h e prese n t t i m e is for electrical conductor wire of which over I O , O O O . O O O pounds h a v e been made b y one manufact u r e r . I t is used under conditions where a combination of high tensile s t r e n g t h a n d elasticity with sufficient conductivity is needed. T h i s is particularly t h e case with uninsulated line mires for telegraph, signal, a n d telephone service. a n d with insulated “ d r o p wires” running f r o m poles t o subscribers’ houses i n telephone practice. E v e n h a r d - d r a w n copper wire h a s n o t sufficient mechanical strength for such uses unless i t is employed i n larger sizes t h a n c a n well be afforded. I r o n or steel conductors h a v e been used, ,b u t~ t h e y rapidly deteriorate when i n contact with t h e air, whereas copper-clad steel is as durable a s t h e copper itself. T h e elastic limit or point a t which p e r m a n e n t stretching begins i n t h e case of pure copper wire is so low t h a t wind strains cause a sagging which must be t a k e n u p
I913
T H E J O C R S - I L O F I S D I ' S T R I d L . 4 S D E-VGISEERI-YG C H E M I S T R Y
a n d LThich eventually causes breakage. T h e most rapid destruction occurs, however, when copper wires are exposed t o snow a n d ice i n winter. G r e a t losses occur t h r o u g h breakage of t h e wires a n d i n t e r r u p t i o n of communication. (Sagging a n d breaking of copperclad x-ire shown b y slides ) T h e judicious substitution of copper-clad steel wires for copper wires has done am-ay m5th much of t h e trouble formerly experienced i n t h e maintenance of telegraph, telephone, signal, power, a n d lighting wire circuits. T h e grade of copper-clad wire most used has a cond u c t i v i t y averaging 30 per cent of t h e conductivity of p u r e copper wire; a t t h e s a m e time i t s tensile s t r e n g t h is ;o per cent greater a n d it is possible t o a p ply ~ j oper cen't greater tension before p e r m a n e n t stretching occurs. I t s weight is I O per cent less t h a n t h a t of copper wire of t h e s a m e size a n d i t h a s IOO per cent greater conductivity t h a n either steel or iron (Table T). T4BLE
V-PROPERTIES
O F STAPDARD COLOhIAL COPPER-CLAD H A R D - D R A W COPPER N WIRE
Bare, hard drawn Average resistance a t 60' F. in Weight per M. ohms per M. ft. ft. in Ihs. Size B. and S. G.
7
C. C. C.
0000
0 . 1603
000
0.2021 0.2549 0.3214 0.4052 0.5110 0,6443 0.8121 1 ,025 1.292 1.679 2.054 2.590 3.267 4.118 5.195 6.548 8.258 10.41 13.13 16.56 20.88 26.33 33.22
00
0 1 7
3 4
5 6 7 8 9 10 11
12 13 14 15
16 17 18 19 20
H. D. Copper 0.04906 0.06189 0.0;803 0.0983 1
0.1241 0.1565 0.1972 0.2488 0.3138 0.3955 0.4986 0.6288 0.7934 0,9996 1,262 1.591 2.003 2.527 3.185 4,022 5.059 6.392 8.057 10.14
,
C. C. C. 584 463 366 291 230 154 145
114.5 91 . 0 i2.0 5i.7
45.5 35.5 29.0 22.8 18.2 14.3 11.3 9.0 7.2 5.6 4,4
3.5 2.82
.
H. D. Copper 64 1 509 403 320 253 202 I59 126 100 79 64 50 39 32 23.5 20.0 16.5 12.4 10.0 i.9 6.2 4.8 3.9 3.1
\vIRE
891
work in military. telephone, a n d telegraph service. I t s use h a s been suggested for small angle irons i n skylights a n d other windon-s. I t has n o t yet been extensively applied in building construction. b u t i t s great durability entitles i t t o wider use, especially for roofs, cornices, rain-water conductors. etc. T h e results of exposure t e s t s of copper-clad sheet steel in t h e P i t t s b u r g h atmosphere h a v e been most encouraging. \VEATHER
RESISTASCE
OF
COPPER
AliD
COPPER-CLAD
SHEET STEEL
Samples of sheet of copper a n d of steel coated with
4hD
Average breaking weight in lhs. , -
C. C. C. 9805 82.50 6830 5680 4500 3900 3200 2630 2160 17iO 1450
1180 Y65 790 645 52.5 430 350 280 230 185 150 120 100
H. D. Copper 7~14 65.13 5365 4386 3565 2892 2338 1890 I520 1221 984 i58 630 506 403 318 257 202
TABLE VI-COllP.\RlSON
OF COSTS FOR EQGAI. STRESGTH ( B A R E~YIRE-HARD D R A W K ) For equal strength values, this diagram gives the approximate relative costs per mile of t h e tTvo materials, in bare form, hard drawn
165
128 102 83 64 51
T h e tensile s t r e n g t h of copper-clad wire varies f r o m 60,000 pounds per s q u a r e inch i n 0000 wire ( 0 . 4 6 inch i n d i a m e t e r ) t o 98.000 pounds per s q u a r e inch i n t h e case of 1-0. I O wire (0.1 0 2 inch in d i a m e t e r ) , t h e a c t u a l breaking weights i n t h e t w o cases being 9 , S o j p o u n d s a n d 7 9 0 pounds, respectively. I t is a n interesting t h i n g t o know t h a t t h e average breaking weight of copper-clad wire is f r o m 3 t o 6 times as great a s t h e weight of a mile of t h e s a m e wire, a n d t h e elastic limit is sufficiently high so t h a t if wind a n d ice m-ere n o t t o b e allowed for, t h e r e could be s p a n s of a mile a n d more i n length i n t h e case of river crossings, etc. Tables V I a n d 1-11 show relative costs for equal s t r e n g t h s a n d relative s t r e n g t h s for equal size. Copper-clad wire is also used for trolley wire. springs, anchor bolts, p u m p rods. cotter pins, etc. Copperclad wire (bare) has been successfully used for field
OF STRENGTHS FOR EQUAL SIZE (BAREWIRE-HARD D R A W N ) For equal sizes, this diagram gives t h e actual strengths of t h e t w o materials, in bare form, hard drawn
TABLEVII-COMPARISON
T H E JOLiRLVAL O F I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y
892
t h e same grade of sheet copper b y t h e Griffith process, a n d a sample of sheet made from a Colonial coppercovered wire b a r were exposed on t h e roof of t h e P i t t s burgh Testing Laboratory building from November 21st, 1 9 1 1 . t o August 1 2 t h , 1913. T h e sheets were held in a flat position on a wooden base b y means of screw hooks. This arrangement permitted t h e retention of moisture under t h e sheets a n d favored t h e s t a r t i n g of decay of t h e wood a t those points. T h e sheets were in no case held so tightly t h a t dust was excluded a n d slag d u s t from t h e roof coating blew in a n d cemented itself t o t h e sheets in some cases. T h e sheets were measured a n d weighed before exposure. T h e samples exposed were as follows: A B C
D
E F
Copper-clad, one side. 1st lot Griffith. Copper-clad, both sides. 1st lot Griffith. " L a k e " copper sheet used f o r welding t o steel in making 4 . B, D, and F. Copper-clad, one side. 2nd lot Griffith. Copper-clad sheet made from one a n d three-fourths inch wire bar C-17. Duplicate of A, b u t exposed with steel side up.
Yol. 5 , No.
11
made specimens of copper-clad steel made by t h e Griffith process, A being intentionally left with exposed steel projecting a t edges a n d ends in order t o show t h e manner in which t h e steel corroded a t a n d near t h e contact of copper a n d steel. T h e loss of weight of b o t h sheets was practically t h e same although A was exposed with t h e copper side u p , a n d F with t h e steel side up. This seems to show t h a t moist air r a t h e r t h a n rain controlled corrosion of both steel surfaces. N o perforation took place a t a n y point of t h e copper surfaces on .-I or F . There was no special grooving t o indicate electrolytic corrosion such as would be expected if t h e copper were not securely welded t o t h e steel. T h e large losses in weight were due t o t h e rusting a w a y of t h e exposed steel (Fig. 14).
WEIGHT A N D SIZES Thickness, inch Weight Grams
... , . . ... ... E , .. . . F ...
A B C D
Loss
OF
3730.6 5317.7 274.58 2853.8 1067.42 3773.1
Iron
Copper
0.060 0.071
...
0.005 0.006 0.04
0.026 0.060 0.059
(inci. C u ) 0.004
?
Size, feet 2.97 2.42 1.09 4.59 2.05 3.0
X 1.09 X 1.525 X 0.31 X 0.92 X 0.40 X 1.1
Area Square inches 466.56 531.36 48.66 680.11 144.65 475.2
WEIGHTDURING EXPOSURE OF 21 MONTHSA N D 20 DAYS
Original weight Final weight Loss i n weight Grams Grams Grams A .................... 3730.6 3317.0 413.6 B.................... 5317.7 5277.0 40. 7 C .................... 274.58 274.3 0.28 D .................... 2853.8 2038.0 815.8 E.................... 1067.42 1067.0 0.42 F ..................... 3773.1 3355 . 0 418.1
These results show t h e remarkable durability of copper a n d of copper-clad steel. (Specimens shown.) Sample E is t h e only one which really represents t h e protecting efficiency of copper a t t h e edges as well as on b o t h surfaces. T h e long edges were copper clad b u t t h e ends were sheared. There was no extraordinary penetration of iron r u s t a t t h e sheared ends. T h e only defects were some blisters developed by t h e crude way of working t h e wire b a r into sheets. Slight rusting occurred at those exposed points b u t nowhere else, a n d t h e t o t a l a m o u n t of rusting was very insignificant. N o tendency was observed for t h e rusting t o p e n e t r a t e under t h e copper a n d push i t u p . T h e same was t r u e a s , t o t h e places where t h e steel was exposed in drilling holes for holding t h e sheet in position. T h e t o t a l loss i n weight in t w o years was insignificant, a m o u n t i n g t o only a fraction of a g r a m , or less t h a n 0 .o j per cent. T h e durability of Sample C, t h e copper sheet, is especially remarkable considering t h e acidity of t h e P i t t s b u r g h atmosphere. T h e loss amounted t o only 0 . 2 8 gyam, approximately 0 .I per cent. T h e determination of t o t a l oxygen showed t h a t t h e oxidation h a d penetrated less t h a n 0 . 0 0 0 2 inch or 0 .j per cent of t h e thickness of t h e sheet. Samples A a n d F represented r a t h e r imperfectly
B FIb
14.-DIFFERENCE I N CORROSIOSS O F STEEL SHEETS, COPPER-CLAD 2's. COPPER-CASED
(-4) Cross-section Showing Corrosion of Steel a t Edges Only-Copper Well Joined t o Steel ( B ) Cross-section Showing Corrosion of Steel on Sides -Copper Separated from Steel
Sample B was coated with copper on both sides b y t h e Griffith process, t h e short ends being sheared, b u t t h e steel extended 0 . 4 inch beyond t h e copper on one of t h e long sides. There were no perforations of t h e copper, a n d t h e loss i n weight was caused b y t h e rusting of t h e exposed steel a t t h e edge. Sample D.-During t h e first p a r t of t h e exposure of t h i s sheet a number of pin holes i n t h e copper near one e n d were observed. I n s t e a d of extending b y electrolytic action, these pin holes a p p a r e n t l y were soon sealed u p a n d were n o t observable a t all, either as blisters or pits a t t h e e n d of t h e period of exposure. This is quite contrary t o what might have been predicted b y one who considered t h e union between t h e copper a n d t h e steel a purely mechanical one. While these experiments were in a way quite crude, t h e y shorn conclusively t h e remarkably efficient protecting power of copper for steel t o which it is welded. It is all t h e more remarkable when it is considered t h a t t h e exposures were made in t h e moist a n d acid P i t t s b u r g h atmosphere. Attention is called in t h i s connection t o t h e absence of electrolytic corrosion phenomena when copper a n d iron were securely welded. T h e method of manufacturing these steel sheets was t h e Griffith process of welding copper t o steel, slabs of soft steel being cleaned a n d electro-plated with copper a n d fastened securely t o copper sheets of t h e desired thickness, t h e combined metals being t h e n heated u p
ru’OV.,
I913
T H E J O C R S A L O F ILVDCSTRI.4L A S D E S GI LVE E RI h’G C H E M I S T R Y
t o a proper welding t e m p e r a t u r e a n d united b y rolling t o €he required dimensions. Such sheets, of course, can be given a n y finish which is desired. T h e y m a y either h a v e a high polish a n d considerable stiffness or t h e y m a y be m a d e soft a n d pliable. If used for flat roofing sheets a higher t e m p e r a t u r e will be required for brazing together t h a n is required for soldering either terne or t i n plate. If used as corrugated sheets no brazing would be required. T h e r e is n o good reason, we believe, n-hy a process of coating sheets with copper should n o t be developed along t h e lines of t h e present processes of tinning a n d galvanizing, a n d such copper-clad sheets could be more cheaply m a n u f a c t u r e d , b u t i t would be n a t u r ally expected t h a t as t h e coating would consist almost entirely of copper-iron alloy. i t would be less durable t h a n t h e p u r e copper coating m a d e b y t h e process of n-elding w i t h o u t melting. POT.4SSIL JI C Y A S I D E S O L U T I O S A S AS E T C H I N G J I E D I C M ASD
SOLYEST
FOR
THE
COPPER O S COPPER-
CLAD S T E E L
A boiling Z j per cent solution of potassium cyanide i n distilled water was found t o be a fairly rapid solvent for copper a n d i t s action upon steel is negligible. I t was accordingly used b y t h e writer for inT-estigating differences i n coatings of copper-clad steel wires T h e purest coatings dissolved most slowly a n d most uniformly, remaining bright. I m p u r e coatings dissolved rapidly. developed a grooved s t r u c t u r e , a n d if iron w a s present i n material a m o u n t , t h e coating t u r n e d grayish black. I n t h e case of coatings high i n iron ( 1 2 per c e n t ) . t h e iron s e p a r a t e d as a felt-like substance Tyhen t h e copper dissolved. (Samples of wire stripped of C u b y K C N were shown.) T h e following experiments illustrate t h e solubility of iron a n d copper i n dilute potassium cyanide solution: SUBSTANCE Sheet s t e e l . , , , . , . . Copper f o i l . , , . . . Copper foil in contact with steel.
Weight Gram
Weight Time boiled dissolved Minutes Gram
0.7400 0.0950
30 30
0 0006 0 0106
SOLVEST 50 cc. lOYc K C S 50 cc. 10yc K C S
0.0854 0,7080
C u 30 F e 30
0 0782 0.0006
50 cc. IOc, K C S 50 cc. 10% K C N
T h e more rapid solution when iron was i n contact with copper was very striking. T h e fact t h a t a n i m p u r e copper coating dissolves more rapidly is shown b y t h e following comparison of t w o copper-clad wires. COPPEH DISSOLVED
“Welded” “Alloyed”
....... .......
Weight Grams 299.0 302.0
1st boil Grams 9.0 11.0
2nd boil Grams 11.0 17.0
3rd boil Grams 14.0 16.0
4th boil Grams 15.5 18.0
5th boil Grams 14.5 23 0
As a n etching m e d i u m , potassium cyanide solution a t o r d i n a r y t e m p e r a t u r e developed differences i n copper coatings with a n d without iron, b u t i t was p a r ticularly successful in bringing out t h e coarse s t r u c t u r e produced i n cast copper b y t h e presence of arsenic. I n t h e following comparison KO. 3 showed a fine grain a n d bright color when etched, KO. I a distinct
8 93
coarse a n d slightly darker one, a n d K O . 2 , with 0.41 per cent arsenic, w a s very d a r k a n d very coarsely crystalline : 2
3
99.48 5; 0.412 0,024 44.75
99.90 c-; 0,005 0.065 97.9
1
Copper . . . . . . . . . . . , , , , , , . , Arsenic. , . . , , , . , , , . , , . , , . Oxygen. . . . . , , , . Electriral conductivity.. . . .
99.76 0.062 0 085 52. i 5
F o r these analyses a n d t h e specimens of cast copper, t h e writer is indebted t o 11r. G. L. H e a t h , Chief Chemist of t h e C a l u m e t a n d Hecla Copper C o m p a n y . S U JI 11X R 1
Copper m a y be united t o steel b y alloying or welding. I n t h e former m e t h o d t h e copper is melted. I n t h e l a t t e r it is plastic only: 9joo C. is a favorable t e m p e r a t u r e for welding. B o t h methods. if properly carried o u t . give mechanically secure unions. T h e welding process is most satisfactory for t h e manufact u r e of copper-clad wire because t h e conductivity is n o t diminished b y t h e presence of copper a n d iron alloys. Melted copper dissolves iron very rapidly. T h e resultant alloys are poor conductors of electricity. I r o n a n d copper form a n alloy if heated together i n R non-oxidizing atmosphere t o a point j u s t abol-e t h e melting point of copper. If heated h o t t e r . alloy formation proceeds more rapidly. Copper is extraordinarily resistant t o atmospheric corroding influences. Tests of copper-clad sheet steel i n P i t t s b u r g h showed very little effect on t h e copper in t w e n t y - t w o months. Copper is more expensive t h a n zinc b u t costs only 40 per cent as m u c h as t i n . I t is more durable t h a n either. Combinations of copper a n d steel should be more widely used. Steel welded t o copper does not corrode more rapidly t h a n does t h e s a m e steel entirely o u t of contact with copper. Potassium cyanide in z j per cent aqueous solution dissolves copper a n d barely a t t a c k s iron. It is useful as a solvent a n d as a n etching medium for copper-clad steels a n d as an etching medium for cast copper containing arsenic. BIBLIOGRAPHY OF P A T E S T S COVERISG
UKIOSS
O F COP-
P E R O R I T S ALLOYS TVITH IROh- OR S T E E L , B Y T H E AID O F H E A T L T I T E D STATES P d T E N T S Method of Coating Iron with Brass or Copper. 11.319. July 18. 1854. H . Burgess. Improvements in Coating hletals. 21,79i. 1858. Hiler. Improved Proccss of Uniting Steel with Copper. Brass, Etc. 39,531. August 11, 1863. E . Sarary. Improvenient in Line-wires for Telegraph. 4i.940. 1865. Farmer a n d Milliken. Improvement in Line \Tire, for Telegraphs. 5 9 . i 6 3 . 1866. Farmer and Milliken. Improved Mode of Uniting Steel or Iron with Copper. 69,001. September l i , 1867. E. T . Ligon. Improvement in Combining Copper with Cast Steel. 73,375. 1868. Park. Improvement i n Telegraph XVire. 91,416. June 15, 1869. A . C a w . Nachines for Above Process. 91,417. Cary. Plating Iron for t h e Manufacture of Hinges. 58,354. December. 1869. J . J. and L. Crooke. Improvement in \\-elding Brasi or Alloy of Copper to Iron or Steel. 126,594. hIay 21. 1872. G. R . LIeneely. Improvement in Plating or Coating Metals. 130,362. August 1.1. 1872. E . E. d e 1.obstein. Improvement in Coating Metals n i t h Copper. 141.132. July 2 2 , 1873. 0. Gauduin, J. B. J . Mignon, and S. H. Rouart.
8 91
T H E J O r R N d L O F I A V D 1 7 S T R I d L d,VD ELVGINEERIAVGC H E M I S T R Y
Coating Iron Surfaces. 227,268. M a y 4, 1880. J . Kintz. Method of Plating Iron and Steel. 242,194. M a y 31, 1881. T. Fleitman. Manufacture of Copper-Plated Sheet-Iron. 244,230. July 12, 1881. C. Haegele. Mode of Plating Iron or Steel. 258,119. M a y 16, 1882. H . Reusch. Plating Metals. 267,879. November 21, 1882. C. Haegele. Mode of Plating Iron and Steel. Reissue 10,367, .August 14, 1883, of P a t e n t 258,119 of M a y 16, 1882. H . Reusch. Process of Making Electric Conductors. 286,796. October 16, 1883. T.Egleston. Manufacture of Metal. 286,904. October 16, 1883. G. H . Chinnock. Compound Electrical Conductor. 288,443. November 13, 1883. I,. Johnson. Manufacture of Electrical Conductors. 296,074. April 1, 1884. T.Shaw. Method of Plating Metals. 303,025. 1884. McCleane and Grey. Manufacture of Compound Wire. 309.468. December 16, 1884. I. A. and M . D. Kilner. Apparatus for Manufacturing Compound Telegraph Wire. 310,993. January 20, 1885. Farmer. 310,994. Modification of 310,993. Process of Manufacturing Compound Telegraph \Tire. 310,995. 1885. Farmer. Electrical Conductor. 312,673. February 24, 1885. C. M . Thompson and C. B. Eberle. Method of Making Electrical Conductor. 320,684. June 23, 1885. W.S.Platt. Method of Constructing Compound Metallic Bodies. 361,799. April 26, 1887. E. Wheeler. Process for Coating Iron with Tin or its Alloys or Other hletals. 371,248. October 11, 1887. E. I. Braddock. Ingut for the Manufacture of Compound Metallic Tubes. 371,719. October 18, 1887. T.S . Very. Telegraph Wire. 379,535. March 13, 1888. W.Hewitt. Manufacture of Wire. 410,368. 1889. Martin. Plated Wire. 441,885. December 2, 1890. G. U. bfeyer. DeProcess of Making Plates, Etc., of Combined Metals. 530,719. cember 11, 1894. A. Rodig. 543,192. Extension of 530,719 to cover over-hanging projection or holding fast t o cast material. Method of Making Compound Wire. 5 5 0 , 7 0 5 . December 3, 1895. H. E. Williams. Process or Method of Welding Copper upon Steel. 5 7 7 , 8 1 7 . February 23, 1897. J. Burns. Uniting or Welding hletal. 580,344. April 6, 1897. Griffith. Current-Conducting Rail for Electric Railways. 588,541. August 1 7 , 1897. L. E. Walkins. Method of and Apparatus for Welding Metals. 593,534. Xovember 9, 1897. J. W.Comley. Manufacture of Electric Conductors. 628,770. July 1 1 , 1899. S.0. Cowper-Cowles. Process of Welding Steel and Copper. 673,664. May 7 , 1901. T. Smith and F. G. Sherry. Tube. 674,394. M a y 21, 1901. A. E. Beck and G. Townsend. Uniting or Welding Metals. 685,758. Sovember 5. 1901. Griffith. 688,162. December 3, 1901. J. W., R . W., and C. H . L. Comley. Plating Metal. 704,793. July 15. 1902. Griffith. Process of Coating Tubes. 708,788. 1902. \\'achwitz. Anderson. I. Process of Uniting Metals. 729,113. M a y 26, 1903. J. & S2letal-Welding. 776,706. December 6, 1904. U'achwitz. Electrical Conductor. 750.509. 1904. Wherry. Process of Uniting Two Metals. 750,511. 1904. Wherry. Flux. 750,512. 1904. Wherry. Art of Uniting Metals. 789,530. M a y 9, 1905. VI. L. Fenn and J. W.Conway. Art of Making Copper-Coated Iron or Steel Sheets. 827,378. July 31, 1906. IN. P. Lewis. Ship's Plate. 851,069. April 23, 1907. J. Craig. Process of Making Compound Metal Bodies. 851,684. April 30, 1907. hlonnot. Process of Producing Compound Metal Bodies. 853.716. 1907. Monnot. Process of Making Bimetallic Products. 853,932. 1907. hlonnot and Martin. Manufacture of Compound Pipes and Tubes. 860,232. July 16, 1907. J. W. Offutt. Electric Conductor. 867,659. October 8, 1907. W ,Hoopes. 878,984. Extension of 851,684. Monnot. 893,932. July 21, 1908. Monnot. 894,163. Extension of 894,162 t o cover two coatings on the core. Monnot. Same as above, t o include silver. 894,164. Monnot.
Val. j, S o .
II
Process of and Apparatus for Making Compound Metal Objects. 905,558. December 1, 1908. hlonnot. Process and Apparatus for Producing Compound hletal Objects. 910,405. January 19, 1909. Monnot. Process of Making Clad Metal Articles. 927,371. 1909. Monnot. Clad Metal. 927,372. 1909. Rlonnot. Clad &fetal and Process of Producing Same. 929,687. 1909. Xlonnot. hIold for Rail Bonds. 918,108. 1909. Wherry. Method of Uniting Metals. 959,517. M a y 31. 1910. Griffith. Compound Metal Object. 960,372. June 7 . 1910. Monnot. Electrolyte for Depositing Cop,per. 976,454. 1910. Grey and Griffith. X e t h o d of Uniting Metals. 976,455. Sovember 22, 1910. Grey and Griffith. Method of Making Homogeneous Mechanical Juncture. 1.012 . 0 7 7 . 1911. Herrick. 1,065,727. 1913. Rockey and Eldridge. BRITISH PATENTS X Method of Covering or Coating Iron, Steel or other Metals, or
hlixtures of Metals. 4,053. 1816. J. Dayman. Plating Iron and Steel with Brass and Copper, and Forming the Same into Plates, Bars, Etc. 4.598. ld21. J. Poole. 5,111. 1825. D. Gordon and W. Bowser. Coating Iron and Other Metals. 8.403. 1840. J. B. A-eilson. Manufacture of bletals and Coating Iron and Steel. 11,971, 1847. A. Parkes. Coating Iron and Other Metals. 12,993. 1850. E. G. Pomeroy. Coating Iron with Copper and Brass. 421. 1857. Burgess and Watt. Covering Iron with Copper or Copper Alloys. 709. 1855. Tytherleigh. Coating Iron with Copper or its Alloys. 923. 1856. Tytherleigh. Rollers for Printing Fabrics. 1,924. 1856. Tytherleigh. (Provisional) Preparing and Coating Metallic Surfaces. 2,472. 1856. R. D. Atkinson. Improvement in Coating or Amalgamating Metals. 1,100, 1858. Hiler. Improvements in Coating Iron and Steel. 3,475. 1862. W.and H . Bowser. Improved Process for Coating Iron. 841. 1863. .W. hlitchell. Coating Ship Surfaces with Copper, Brass, Etc. 351. 1864. XI. C. deC. Sinibaldi. Coating Iron and Steel. 598. 1869. G. J. Hinde. Improvements in Coppered Iron or Sheet Steel. 3.466. 1871. Lake. Copper Covered and Copper Cored Wire. 1,847. 1872. W. R . Lake. Depositing Copper on Iron and Steel. 3,970. 1872. Johnson. Improvements in hlaterials for Casks, Etc. 1,887. 1875. A. B. Walker. Coating Metals. 1,836. 1881. F. C. Glaser. Coating or Plating Certain Metal Surfaces. 3,122. 1882. A . 11. Clark. Coating Iron and Steel Sheets, with Copper and Copper Alloys. 10,886. 1884. J. and J . Taylor. Method and Apparatus for Manufacturing Compound Telegraph \Tire. 798. 1885. 4llison. Process for Plating Metals and ?rletallic Alloys. 3.985. 1886. C. E. Steinwea. Improvements in the Manufacture of Wire. 6,834. 1 8 8 i . A. Mannesmann. Coating Iron and Steel Plates with Copper. 17,269. 1888. G. Prout and D. Murray. Coating of Metals. 3,149. 1889. -4.J. Boult. Coating Sheet Iron with Copper or Brass. 4,335. 1889. 1%'.E. Everitt. Flux for \\'elding Copper and Steel. 9.187. 1889. H . H . Chandler. N a n u f a c t u r e of Band, Plates, Sheets, Etc., of Combined Metals. 17,009. 1891. Martin. Improvement in the Manufacture of Plates Having a Partial Coating of Other Metal. 21.467. 1892. L. Grambow. Coating Iron and Steel with Brass and Other Metallic A4110ys. 1,906. 1893. A. V. C. and J. B. Fenby and G. Moore. Coating Iron and Steel with Copper or an Alloy of Copper. 29,251. 1896. R . D. Burnie and VV', T.Lougher. Process and Apparatus for Uniting Metal Sheets, Bars, Etc. 26,030. 1897. Comley. Improvement Relating to Uniting or Welding Metals. 8.780. 1897. Griffith. Protecting Tubes of Steam Generators by Application Of Silver. 9.057. 1899. E . J. M. la Combe. Uniting Copper and Copper Alloys with Another Metal (Iron or Steel) 10,763. 1899. W. G. Clark. Coating Metals. 11,603. 1900. Martin.
N o v . . I913
T H E JOC‘R-VAL O F I N D C S T R I A L A N D E-VGINEERING C H E M I S T R Y
Improvements of Boiler a n d Other Metal Tubes. 11,981. 1900. A. E. Beck and G. Townsend. Manufacture of Steel Ingots Plated with Copper. 16,993. 1902. S. Vanstone. Improvements in Manufacture of H o t Water Tanks, Etc. 15,383. 1903. D. P. Menzies. Process of Uniting Metals. 18,454. 1903. J. D. Prince. Production of Metallic Protective Depo-its on hletals. 9,836. 1904. A. LPvy. Manufacture of Bimetallic Ingots. 12,000. 1904. H. Harmet. Uniting Iron a n d Steel with other Metals. 17,660. 1904. Davies a n d Clark. Improvement in Manufacture of Compound Metal Ingots, Etc. 8,913. 1906. Monnot. GERMAS PATESTS Method of Manufacture of Conductor Wire with Metallic Covering. 47.950. 1888. Martin a n d Martin?. Method for Coating One Metal with Another. 124,898. January 4, 1899. S. H. Thurston. Slanufacture of Sheet Steel w.ith Copper Covering. 124,387. 1900. Martin. Process for Welding Baser hletals for Purposes of Plating. 137,017. 1902. IVachwitz. Process for Uniting Steel a n d Other hletal Plates with Aluminum, Etc. 15?,04?. 1903. Wachwitz. F R E X C H I’ATEXTS Improvements in the Manufacture of Wire. 168,133. 1885. Martin. Supplement. Perfected Method of Slaking Bimetallic Wires. 168,133. 1888. Bi-Netal Sheets and Plates, Process of IIanufacture of. 206,789. 1890. 3lartin. 213,109. 1891. Perfected Process for Making Bimetallic \Vires. Slartin. Manufacture of Bimetallic Bands, Plates, Sheet Iron a n d Hoops. 216,565. 1891. Martin. BELGI.\X h-ew 171,442.
PATESTS
Process of Manufacturing 1903. Alartin.
Bimetallic
Plates
and
\Vires.
DEPARTMEST OF RESEARCH A N D CHEMICAL ENGINEERISG PITTSBURGH TESTISGLABORATORY
THE CONDENSATION OF GASOLINE F R O M NATURAL GAS’ B y GEORGE4 . BCRRELL.WD
FRANK A l . SEIBERT
I n t h i s p a p e r are given some results of work performed b y t h e Bureau of Mines having t o d o with t h e condensation of gasoline f r o m n a t u r a l gas. CHEMISTRY O F S.1TURAL
GAS
K i t h t h e growth of t h e n a t u r a l gas gasoline i n d u s t r y n a t u r a l gases h a v e been classified i n t o t\vo divisions so called “ w e t ” a n d “ d r y ” gases, depending upon whether or n o t gasoline can be commercially condensed from t h e m . T h e classification is exceedingly loose because n a t u r a l gas mixtures m a y range f r o m those containing only m e t h a n e as t h e combustible cons t i t u e n t ( a gas difficult t o liquefy) t o those i n which t h e hydrocarbon vapors predominate a n d which liquefy easily. Bet\%-eent h e t w o extremes t h e r e a r e n a t u r a l gases containing t h e different constituents, m e t h a n e , e t h a n e , propane, b u t a n e s , pentanes, etc., in m a n y different combinations. Some of these m a y not contain enough of t h e desirable gasoline constituents for commercial purposes, others m a y . N a t u r a l gases n o t f o u n d i n t i m a t e l y associated with oil a r e t h e so-called “ d r y ” gases. Those f o u n d i n t h e s a m e s t r a t a with oil a n d in i n t i m a t e contact with t h e s a m e are those f r o m which ,gasoline is obtained i n t h e n a t u r a l gas gasoline i n d u s t r y . T h e Bureau of Mines finds as t h e result of m a n y analyses t h a t n a t u r a l gases 1 Paper presented a t the 48th meeting of the American Chemical Society, Rochester, September 8-1 ? , 1913. Published by permission of t h e Director of the Bureau of 3Iines.
895
a r e mixtures i n which hydrocarbons of t h e paraffin series predominate a n d t h a t small quantities of nitrogen, carbon dioxide a n d water vapor are present. Hydrogen sulfide is sometimes p r e s e n t ; perhaps o t h e r sulfur compounds too. F. C. Phillips f o u n d n a t u r a l gases of Western Pennsylvania, which h e worked with, t o c o n t a i n paraffin hydrocarbons, carbon dioxide a n d nitrogen. Other investigators invariably report a t least small proportions of carbon monoxide, hydrogen a n d ethylene. Experimental errors i n t h e work easily accounts for these errors. T h e a u t h o r s of t h i s p a p e r believe t h e work of S. A. F o r d a s showing very large percentages of hydrogen t o be in error. His analyses h a v e been q u o t e d m a n y times i n different t e x t books. T h e y were m a d e in 1885. T h e a u t h o r s of t h i s p a p e r i n looking over t h e analyses m a d e b y t h e m of t h i r t y n a t u r a l gas samples collected from different p a r t s of t h e c o u n t r y find t h e heating value ranging from 6 8 j B. t. u. t o I j i 7 B. t. u. per cubic foot a t 60’ F . a n d 760 m m . pressure. These analyses will be incorporat e d n-ith m a n y others i n a government publication. These gases range from marsh gas issuing from t h e marsh beds, a n d containing only inethane as t h e combustible gas t o casing h e a d gases t h a t are used for lighting a n d heating towns. Only t w o of t h e gases, those of t h e highest heating value, are probably a d a p t e d for gasoline condensation. One sample contained ( a s shown b y combustion analysis) i n addition t o m e t h a n e , carbon dioxide a n d nitrogen, 7 5.16 per cent of e t h a n e . T h e n a t u r a l gas of P i t t s b u r g h has a gross heating value of a b o u t 1 1 7 7 B. t . u. per cubic foot a t o o C. a n d ;60 m m . pressure. S I G S I F I C A S C E 0 F 0 R D I S A R Y A S A L Y T I C A L R E S L-L T S
I n t h e analysis of n a t u r a l gases b y t h e slow combustion m e t h o d , t h e d a t a obtained a d m i t of t h e calculation of only t w o of t h e chief constituents. T h e mixture, however, m a y contain all of t h e gaseous paraffins a n d considerable quantities of t h e vapors of t h e liquid hydrocarbons. When t h e lower members of t h e paraffin hydrocarbons predominate, t h e results obtained are more accurate t h a n when t h e higher members predominate. N a t u r a l gases from v h i c h gasoline c a n be extracted contain appreciable q u a n t i ties of t h e liquid hydrocarbon vapors. I n t h e analyses of these mixtures t h e ordinary slow combustion analysis will give only approximate results for several reasons. First-The gas mixture often contains more t h a n tm-0 combustible constituents. Secoizd-Some of t h e gases a n d vapors deviate considerably f r o m t h e gas laws a n d their t r u e molecular volumes’ a r e unknown. Third-So small a n a m o u n t of t h e mixture must be used i n some cases t h a t experimental errors a r e greatly magnified i n calculating t o a percentage basis. Typical analyses of t w o different n a t u r a l gases follow which c o n t a i n small a m o u n t s of m e t h a n e a n d larger a m o u n t s of e t h a n e , propane a n d b u t a n e v-ith t h e vapors of t h e liquid hydrocarbons p e n t a n e , hexane, etc. These analyses serve t o show how approximate a com1 .4 government publication b y the authors which c o ~ e r sthis question is in press.
