48o 



NATURE 



[September 15, 1892 



lated for a minimum conductivity of 95 per cent, of Matthies- 

 sen's standard of pure copper. Now, chiefly owing to electro- 

 lytic refining, a conductivity of loo per cent, is demanded by 

 the buyer and conceded by the manufacturer. 



To show the difference between the past and present state of 

 things in relation to the commercial conductivity of copper, I 

 am going to exhibit on the screen measurements of the resistance 

 of six pieces of wire of equal length and equal cross section — 

 they have been drawn through the same drawplate. Three of 

 the pieces are new, and three are old. The three new pieces 

 are made from electrolytic copper, and are representative of the 

 present state of things. The three old pieces are taken from 

 three well-known old submarine telegraph cables, and they 

 show how very bad the copper was when it was first employed 

 for telegraphic purposes, and how great has been the improve- 

 ment. I will take No. i wire as the standard of comparison. 

 It is a piece of the wire about to be supplied to the Post Office 

 Telegraph Department for trunk telephone lines. It will show 

 the very high standard of conductivity that has been reached in 

 the copper of commerce. I am indebted for it, and for two out 

 of three of the old cable wires, to Mr. Precce. No 2 wire is 

 made from electrolytic copper, deposited in my own laboratory. 

 No. 3 is also electrolytic copper, but such as is commercially 

 produced in electrolytic copper refining ; it has been supplied to 

 me by Mr. Bolton, to whom I am also indebted for wire No. 6 

 — a particularly interesting specimen : it is from the first Trans- 

 Atlantic cable — the cable of 58. No. 4 wire is from the Ostend 

 cable of i860, and No. 5 wire is from the old Dutch cable. 

 These wires are so arranged that I can send a small and con- 

 stant current partly through any one of them, and partly through 

 a galvonometer. When this is done the result will be a deflec- 

 tion of the spot of light on the scale from the zero point to an 

 extent corresponding to the resistance of the particular wire in 

 the circuit. The worse the wire is, the greater will be the 

 deflection. We will begin with the Post Office sample first. 

 I connect the galvanometer terminals to wire No. i ; you see 

 there is a deflection of ten degrees. I will now shift the con- 

 tacts to wire No. 2 — exactly the same length of wire is included 

 — but now you see there is a deflection of slightly less than ten 

 degrees, showing that this wire has a little lower resistance than 

 No. I. The difference is very small — it may be 2 per cent. — 

 and 2 per cent, less of it would be required to conduct as well 

 as the No. i wire. The next is No. 3. This is Mr, Bolton's 

 wire, and shows a resistance almost equal to the last. 



No. I, 2, and 3 are, therefore, nearly alike, and have a degree 

 of conductivity almost as high as it can possibly be. 



Now we come to the three old wires. 



We will take No. 4 (the Ostend cable). There, you see, is 

 a great difference. Instead of the spot of light being on the tenth 

 degree, it is upon the eleventh. 



We will now try No. 5 (the Dutch cable). That drives the 

 index to 17. 



Now I change to No. 6 (the old Atlantic cable), and we have 

 a deflection of no less than 25 degrees. I suppose we may 

 assume that this wire fairly represents the commercial con- 

 ductivity of copper in 1858, for it is highly probable that for a 

 work so important as the first Atlantic cable every care would 

 be taken in the selection of the copper. 



The result of this experiment shows that the copper of that 

 cable was extremely bad as a conductor — that, in fact, it is 150 

 per cent, worse than the best commercial copper of to-day. In 

 other words, it shows that, in point of electrical conductivity, 

 one ton of the copper of to-day will go as far as two-and-a-half 

 tons of such copper as was used for the cable of '58. 



This change is largely due to electrolytic copper refining. 



The process of electrolytic copper refining is the same in 

 principle as that which produced the thickening of one of the 

 wires and the thinning of the other in my first experiment. To 

 prepare the crude copper for the refining process it is cast into 

 slabs ; these form the anodes, and correspond to the wire which 

 in my first experiment became thin. The cathodes, correspond- 

 ing to the wire which became thick, are formed of thin plates of 

 pure copper. Here are plates such as are used in electrolytic 

 copper refining works. They are portions of actual cathodes 

 and anodes, and represent the state of things at the commence- 

 ment, and at the end, of the depositing operation — an operation 

 that takes several weeks to complete, and effect the great 

 change these plates show. In copper refining works an 

 immense number of these plates, each having 6 to 10 square 

 feet of superficial area, are operated upon together in a great 



NO- II 94, VOL. 46] 



number of large wooden vats containing sulphate of copper 

 solution and a small proportion of sulphuric acid. Electric 

 current from a dynamo, driven by a steam-engine or water- 

 power, is conveyed by massive copper conductors to the vats, 

 arranged in long lines of 50 or ico or more in series. Thick 

 copper bars connect adjoining vats, and provide a positive and 

 negative support for the plates, which hang in the solution 

 opposite each other, two or three inches apart. During the 

 process the impure slabs dissolve, and at the same time pure 

 copper is deposited from the solution upon the thin plates. The 

 deposition and dissolving go on slowly, in some cases very 

 slowly, for a slow action takes less power, and gives purer 

 copper than a more rapid one. The usual rate is one to ten 

 amperes per square foot of cathode surface. You will better 

 realise what these rates of deposit mean, when I say that one 

 ampere per square foot rate of deposition gives for each foot of 

 cathode surface, nearly one ounce of copper in twenty-four 

 hours, and a thickness of one-eighth hundred of an inch ; and 

 therefore the production of one ton of copper at that rate in 

 twenty-four hours would require a cathode surface in the vats, 

 in round numbers, of 36,000 square feet. At the higher rate of 

 ten amperes per square foot, which is used where coal is cheap, 

 one-tenth of this area would be required. 



The importance of the electrolytic copper refining industry, 

 and the extent of the plant connected with it, may be inferred 

 from the fact that, reckoning the united production of all the 

 electrolytic copper works in the world, nearly one ton of copper 

 is deposited every quarter of an hour. 



Very little power is required for copper deposition if the ex- 

 tent of the dissolving and depositing surfaces is large, relatively 

 to the quantity of copper deposited in a given time. 



Some of the impurities ordinarily found in crude copper are 

 valuable. Silver and gold are common impurities, and these 

 and some other impurities do not enter into solution, but fall 

 down as black mud, are recovered, and go to diminish the 

 cost of the process or increase the profit ; and even those im- 

 purities which enter into solution are, under ordinary conditions, 

 almost completely separated. 



Electrolytic copper refining is both an economical and an 

 effective process. The deposited copper is exceptionally pure. 

 At one time it was supposed that it must necessarily be quite 

 pure, but this is not the case ; other melals can be deposited 

 with the copper, but it is not difficult to realise in practice a 

 close approximation to absolute purity in the deposited copper. 

 Here is an example of the deposition of a mixed metal — brass, 

 that is, copper and zinc deposited together, and there are in the 

 Library a number of interesting specimens of mixed metal de- 

 position. These deposits of brass and other allo)s show that 

 more than one metal can be deposited at the same time. The 

 great enemy to conductivity in copper is arsenic, and the depo- 

 sition of arsenic as well as copper is one of the things to be 

 guarded against in electrolytic copper refining. Not only are 

 the chemical characteristics of electrolytically refined copper 

 generally good, but its mechanical properties are largely con- 

 trollable. Usually electrolytic copper is melted down and cast 

 into billets of the form required for rolling and wire-drawing. 

 This treatment not only involves cost, but the copper is apt to 

 imbibe impurity during fusion ; though, if the process is care- 

 fully conducted, the deterioration is slight. 



But it is evident that the re-melting of the deposited copper 

 is a thing to be avoided if possible, and the question naturally 

 arises, why, now that deposition costs so little, may not the 

 beautiful principle which comes into play in electrotype, and 

 which enables the most complicated forms to be fahhfully copied 

 be taken advantage of to give to plainer and heavier objects 

 their ultimate form ? 



There are several reasons why this idea is not more fre- 

 quently acted upon. One is that the process of electrolytic 

 deposition is slow ; another, that knowledge of the conditions 

 necessary for obtaining a deposit having the required strength 

 and other qualities, is not very m idespread. Moreover, in the 

 electrolytic deposition of copper, and indeed of all metals, there 

 is a strong tendency to roughness on the outside of the deposit, 

 and to excrescent growths, the removal of which involve waste 

 of labour and material. These tendencies can to a very great 

 extent be counteracted by careful manipulation and the use of 

 suitable solutions, and they can also be counteracted by 

 mechanical means. This has been done by Mr. Elmore. He 

 rernedies the faults I have mentioned by causing a burnisher of 

 agate (arranged after the manner of a tool in a screw-cutting 



