Charles George Warnford Lock.

Economic mining: a practical handbook for the miner, the metallurgist and ... online

. (page 44 of 76)
Online LibraryCharles George Warnford LockEconomic mining: a practical handbook for the miner, the metallurgist and ... → online text (page 44 of 76)
Font size
QR-code for this ebook

Each electrode consists of a bundle of 9 carbons, each 2^ in. dianu
attached to a head of cast iron for ferro-aluminium, or cast copper
when cupro-aluminium is made. Each carbon rod weighs 20 lb.

Quite recently it has been possible to obtain carbons 3 in. diam.
weiffhiug 36 lb. ; these are employed in bundles of five. The ^'heid'*
of the electrode is screwed to the copper rods of leads which can be
readily connected with or disconnected from the flexible cabks
supplying the current.

Each cable is secured to slides running on omnibus bars of copper
overhead, so that it can be brought into position opposite the fumAoe
to be used.

The electrodes are arranged so that it is possible, by means of a
handle and screw, to advance or withdraw them from each other in
the furnace.

Lining the furnace is the first operation preparatory to smelting;

this is done by covering the bottom
of the trough with a layer of prepared
charcoal. Oak charcoal is usei
This is ground under edge runners
treated with milk of lime, and dried,
first in a steam-jacketed revolving
drum and afterwards on a hot iron
plate. Each particle of charcoal is
thus coated with an insulating fibn
of lime.




Digitized by



to oonnect the electrodes and start the current. The charge is then
covered with coarse charcoal, and the oast-iron cover, having a hole
in the centre for the escape of gases, is lifted into place ^nd luted so
as to prevent the entrance of air.

The commencing current is about 3000 amperes, gradually in-
creasing to 5000 amperes in the first half hour. The time occupied
by a " run " is about \\ hours. Soon after starting, the gas escaping
from the orifice in the cover takes fire and bums with a white flame.
Phis gas consists mainly of carbonic oxide, with small quantities of
the hydrocarbons and nitrogen.

Oq the conclusion of the run the furnace is allowed to cool, the
nelted alloy collecting in the bottom of the furnace. The next
furnace, ready charged, is connected, so that the process is a con-
Linuous onje, the furnaces being successively charged and connected.

A form of fumfiu)e in which the metal is tapped from the bottom
las given very satisfactory results.

The energy required to produce 1 lb. of contained aluminium
Lepends, of course, upon the grade of alloy, the average being 18 h.p.
>er hour.

The crude metal from the furnace is then refined and remelted in a
everberatory furnace, each ''run" is sampled, and the percentage
•f alaminium is ascertained by analysis.

The nature of the reaction that takes place in the electric furnace
s not very easy to ascertain. The reduction of the ore takes place
Q an air-tight box, and in the presence of an enormous excess of
arbon. It may be assumed that at the intense heat of the electric arc
be ore melts and gives up its oxygen to the carbon^

Al203+3C = 3CO + AI2.

The presence of copper or iron is immaterial, as the reaction takes
lace with ore alone, but in absence of the alloying metal the liberated
luminiam absorbs carbon and is converted into a carbide.

Prof. Hampe, however, contends that the reaction is in two stages,
be first being electrothermic, in which the ore melts ; the second,
lectrolytic, in which electrolysis of the molten oxide takes place,
luminium being liberated.

Snppoeing an alternating current is used in the furnace instead
r the continuous, and that we have a high number of alternations —
•om recent experiments with alternating currents we know that very
ight or no electrolytic action occurs — if then we get equal results,

would go to prove that the action is thermic, and not electro-

An experiment was made by the Cowles Ck>mpany, using an
Itemating current.

The fnmace used was 48 in. X 12 in. x 24 in., with current of
[K) amperes wiih E.M.F. of 50 volts, alternations per minute 11,000.
t the end of 1 J hours they obtained 45 lb. of 4 per cent, aluminium

This investigation was purely of an experimental character, and
; cannot be doubted that on the larger commercial scale the result
ronld be more satisfactory.

Digitized by



The results obtained were identical with those given by tin
continuous current in every way, and the resultant slags mmikr in

The slag from the electric furnace is really only melted tmreduoed
ore, and not a slag in the proper sense of the word. Its approximate
composition is AljO. 90, SiOj 2-00, FcjO, 4-00, CaO 3-9, P 0-10.

The charcoal from the old linings, which has been partially ad-
verted into graphite, is recrusbed and treated with milk of lime, dried
as before, and thus used many times over. The crude metal from tlu
" runs '* is remelted with fluorspar as a flux, and tapped &om the
reverberatory furnace into ingots or plates ; the refinery slag is pal
into a revolving drum for breaking into pieces, and then 'washed tc
remove carbon ; the entangled particles of alloy are thus recovered,
and are worked up with ouier cmar^es. These operations are carried
out in a separate building, the reclaiming house. |

When prep€a*ing aluminium from cryolite, at starting the carbcaj
cylinders are lowered till they touch the carbon bottom. A pooi
contact is thus formed, and the ground cryolite, which is then pilea
around the carbons, is melted by the resistance. When enough \m
been melted to form a good bath, the carbons are raised, and th^
melted cryolite carries the current and becomes the electrolyte. Tb^
resistance is very high until alumina is added to the bath and did
solved there, when it falls suddenly, and the difference of pot^itia^
between the electrodes becomes constant at 6 to 10 volts. A sudden
rise in voltage indicates, therefore, very clearly when the alumini
has all been reduced, and when more must be added. The process i^
continuous, day and night, seven days a week. The metal whid|
collects in the bottom of the furnace is removed every 24 hours. Tb<
metal made by this process is very pure, most of it running abovi
99 per cent, aluminium, and by using extra-pure materials it may b^
made to approximate very nearly to 100 per cent.

Hall's process differs chiefly in using a bath composed of alnminimij
fluoride and the fluoride of a metal more electro-positive thai
aluminium, passing the current through and adding fresh alumina w
intervals to regenerate the salt.

The Heroult process is established at Neuhausen, Switzerland, aoc
at Froges, near Grenoble, France, and by it nearly all the alnmininn
produced in continental Europe is made.

In the Heroult furnace the melted mass of oxide takes the place <m
the electrolyte in an ordinary voltaic couple, the reduced metal at tlM
bottom of the furnace being one electrode and a carbon bar inserted ixm
the top the other. Upon a current passing from the carbon bar througll
the ore to the metal, the ore is decomposed, oxygen travelling upward^
and attacking the carbon, whilst the molecules of the reduced meta^
travel downwards, and are merged in the metal bath. The oonstrooi
tion of the furnace will be understood from Figs. 106 and 107, th^
former being a sectional elevation and the latter a plan. The bodyc
the furnace, or to speak more correctly, of the electric crucible, oon '
of a block of carbon B contained in an iron box A. In the
crucible made, the iron box was cast round the carbon block in order t
obtain a vei^ intimate contact between the surfaces, and thus facilitate
eduction of electricity from the electrodes C to the interior. To»^

Digitized by



ron, by oontraction upon cooling, would securely grip the surface of
lie box from all sidcH, thus ensuring perfect contact. This method
)f constraction is, however, only suitable for small crucibles, and the
)re8ent large furnace is built up of carbon slabs held together by a
m)nght-iron casing, no difficulty having been experienced in making
;ood contact between the two materials. At the bottom of the cavity

Figs. 106, 107.— Heboult ALumNiUM Fusnace.

the carbon block is a tap hole D, closed by a plug E, which is
Ithdrawn from time to time to allow the liquid alloy to run into
.e ladle L, from whence it is cast into ingots. The current enters
e crucible by the upper carbon electrode, consisting of a bundle of
rbon slabs F suspended by a chain. The chain passes overhead
mad the drum of a small winch, which is controlled by the attendant,
id as the lower end of the electrode is being consumed, the latter is

Digitized by VjOOQIC


allowed to desoend so as to keep the distance between the snr&ce i
the molten liquid in the crucible and the end of the carbon electan
as near as possible constant. This distance is preferabljr made Tei
small, and should in practice not exceed 3 mm.

The inventor lays particular stress upon the necessity of keepiu
the distance between the electrodes small. The intervening spac
being filled with a layer of badly conducting molten ore, offers
resistance which increases with the distunce ; and although reeistan
is necessary in order that the current should produce heat, it is n
economical to have more heat than is just sufficient to melt the oi
the work of separating the metal from the oxygen being chieflj doi
by the electrolytic action of the current, and not by high tempen
ture. The electrode consists of carbon slabs F, held together by!
metal clamp G, so as to form one huge compound carbon prism, whid
before insertion into the furnace, is 10 ft. long by 17 in. wide and 9 J ij
deep. As the slabs are only obtainable in lengths of about 3 ft^ ti{
prism is built up of slabs laid upon each other so as to hreak join
the whole being held together with stout copper pins, and protectJ
on the outside by copper plates f in. thick. An electrode of that siJ
is consumed in the produciion of about Jton of contained aln mining
The crucible is closed by a cover H, also made of carbon slabs, b^
insulated from the body of the crucible. Holes J are providt?d \
this cover through which the ore and scrap metal are intix>duoed, an
these openings can be closed by shutters K. The ore used is alomiiu
free from silicon and other impurities, and the scrap metal is eith€
iron or copper, according as the desired product is a feixo or a bron
Bauxite could be used instead of alumina ; but as the former minei
contains many impurities, Heroult prefers the absolutely pure thouj
more expensive alumina. The cost of the raw materials is, howe
in either case not a very important item as compared with the
of power, carbons, and wear and tear of plant. The process of smell
is continuous, and need only be interrupted when the upper elecl
has been consumed, and then only for the short time required to ii
a new electrode. The crucible is periodically charged through cT
holes, and the molten alloy is tapped off at D from time to tii
The production of aluminium per horse-power hour varies somewj
with the percentage of the metal contained in the alloy, the avei
being 30 grm. of contained aluminium per horse power hour,
the maximum 40 grm. Reduced to English measure, this works oi
to an expenditure of 15 horse-power hours per poimd of oontain<
aluminium under averas^e, and 11 horso-puwer hours under very favoui
able conditions. The present production of the crucible is 4 cwt. d
contained aluminium m 24 hours. The crucible can also be used ft
the production of silicon bronze, in which case the scrap copper \
charged not with alumina, but with clean white sand.

As to the general arrangement of the works, power is supplied bj
a turbine, to the shaft of which are direct coupled the two dynamd
supplying the current to the crucible. These large dynamos an
excited by a smaller machine driven by a belt from the turbine shaft
and the speed of the turbine and strength of the main current art
controlled by an automatic regulator acting upon a throttle valv^
placed in the inlet pipe of the turbine. Withm certain limits ti^

Digitized by



Ain current can alflo be regulated by hand, for which purpose one of
ix>wn*8 electrical regulators is inserted in the exciting circuit of the
nail dynamo. The main conductprs are naked copper cables or bars,
id no special precautions are employed to insulate them, as in compari-
m with the huge working current a leakage of 100 amperes, more or
68, ia too insignificant a matter to trouble about. The current is
easured by a large ampere meter. The main current passes once
irou^h this circle, in the centre of which is pivoted an electro-magnet,
rovided with a pointer and counter-weight. The electro-magnet is
ccited from a primary battery, giving a constant E.M.F., and l£e cali-
ration of the instrument can be varied by fixing the counter-weight
Tther from or nearer to the centre of suspension of the magnet, so as
^ obtain a fairly large deflection for the usual working current. The
or king current is about 12,000 amperes; but sometimes short-
rcniting occurs, when the current suddenly rises to 20,000 ampdres
id more. These short circuits are generally due to the fact that one

other of the carbon slabs composing the electrode projects beyond
le others, and touches the surface of the metal bath in the crucible,
he projecting portion is then immediately burned off, but during the
me that this takes place the current is considerably increased beyond
% normal value. It might be supposed that a short circuit of such
^nitade would be likely to damage the electric machinery, but
iparently this is not the case. The dynamos do not appear to suffer

the slightest degree from the short circuit, and the only indications

its taking place are a slight sparking at the brushes and a peculiar
imbling noise in the turbine.

The field is of the multipolar type and made in one casting, so
At there are no joints of any kind in the magnetic circuits. The
mature is of the usual Brown type, with embedded wires, but con-
ins t^o distinct circuits, each provided with its own commutator,
be armature is 38 in. diam. and 24 in. long. The conductor consists
L the outside of round copper bars f in. diam., and on the inside of
kt copper plates placed in grooves planed out of a wood lining, by
hiofa Uiese plates are firmly held and at the same time insulated,
tie ourrent is taken off each commutator by 6 sets Of brushes, there
ing in all 72 brushes on each machine. Although cross connection
tween equipotential coils on the armature are not absolutely neces-
ry when the number of brushes employed equals the number of field
les, the designer of the machine has thought such connections
rvertheless advisable as tending to fairly distribute the work between
I the armature wires.

According to A. E. Hunt* of the Pittsburg Reduction Co., the
obable cost in the near future of producing aluminium on a large
lie will be as follows : —


2 lb. alnmina (Al^O cootalDing 52*94 per cent Al) at 8r 3

1 lb carbon electrode at 26 1

ChemlcalB, carbon dost, and pots \

22 electrical b.p. exerted 1 hour (asing water power) 2}

Labour and superintendenoe l}

General ezpenc>e, interest, and repairs l|

Total cost of lib. alominiam l(Ki.

• Mass. Inst Technology, Boston, 1891.

Digitized by VjOOQIC


The total cost of aluminium at the Pittsburg works at the tii
this estimate was published was much greater than this figure, wil
an output of 375 lb. a day. Other authorities think the limit i
cheapness of production has been nearly reached, and that so faur i
regards its future, unless it can be made in large quantities, juiit \
lead or copper or zinc, it cannot hope to enter as an important ^Kt
in the great industries. It must be smelted in large quantities dire
from its ores, or obtained as a bye-product in the preparation of sea
widely consumed substance, ere it will take in trade the position i
qualities command. It is doubtful if the further prosecution of ti
electrical methods, by which alone aluminium is now made, ^
bring the cost of it to the point at which it will become a prominfl
metcJ, unless they proceed along the line of direct reduction. £vi
here it is by no means certain that they can make it cheap enougL

The most notable characteristic of aluminium is its low speol
gravity — 2*6 for castings and 2*74 for wire. No other m^al
common use approaches it for lightness. Aluminium is not a riTsl
steel, and may never replace iron as a structural material ; but it ii
rival of brass, copper, tin, nickel, and white alloys, and is replacn
them in many directions. In tensile strength it ranks with cast-iro
breaking at 15,000 to 20,000 lb. per sq. in., but in malleability ai
ductility it ranks with the noble metals. Like gold and silver,
hardens very rapidly in working, and rods and wire vary in str^g
from 26,000 to 62,000 lb. per sq. in. Its elastic limit is about half i
tensile strength, and its elongation 10 to 20 per cent, in 1 in.

The electrical conductivity of aluminium is abont 50, with oopp
at 90 and silver at 100. Its thermal conductivity is abont 38, copp
73*6, and silver 100. Undoubtedly the conductivity of chemioid^
pure aluminium is much higher, the sample tested containing aba
1 * 5 per cent, impurities.

Aluminium is a little softer than silver, but its ductility allows
to be drawn, punched, or spun into almost any form. A great deal
sheet metal is being used for this purpose, and much more would
used if it were not for the fact that a slightly bluish tint detrad
from its resemblance to silver. Practically, aluminium is na
tamishable ; but strictly speaking, after long exposure to the atoi
sphere, its polish becomes dulled by a very thin film of white oxid
which seems to protect it from further deterioration. Sulphuretv
hydrogen has no action on it.

It is inert in most of the acids. Hydrochloric acid and the fi»
alkalies are its true solvents. It has been claimed by certain Germi
chemists that aluminium is attacked by almost all vegetable ado
even tea and beer having a solvent action upon it. This statem^
has repeatedly been prov^ to be untrue. Even strong brine has b«
shown to have a very inconsiderable action upon it The mistaJ
was due to the use of aluminium foil : it is well known that almost a
metals when in a line state of division are attacked by reagents tlu
have no action on them in bulk.

The specific heat of aluminium is very high (0*2253), being aboi
double that of iron ; consequently, although its melting point is l«fi
about 1290'' F., it takes a long time for it to become iiuid, and i

Digitized by



quallj long time for it to set. Founders do not usually realise this,
iod the melted metal appearing rather thick, they genenJly make the
Qistake of pouring it too hot.

Alaminium is not volatile at any heat obtainable by ordinary
omboBtion of carbon ; but a thin, tenacious film of oxide forms on the
urfaoe of the melted metal, which protects it from further oxidation.
Uumininm is very sonorous in the form of a bar, but not when cast
nto a bell. This is piobably explained by the fact, noticed by
''araday, that a note is compound, being formed of a tone audible in
he longitudinal direction and another in the transverse.

The alloys of aluminium are divisible into two classes —those con-
kin ing less than 12 per cent, of aluminium and the remainder some
ther metal, and those containing less than 15 per cent, of some other
letal and the rest aluminium. In the first class, aluminium bronze, con-
linrng 90 per cent, copper and 10 per cent aluminium, is the most
aloable. Its tensile strength is 90,000 ro 100,000 lb. per sq. in., and
rben a little silicon i8 also present its strength sometimes rises as
igh as 135,000 lb. per sq. in. To this class belongs the steel alloy,
freatly increased quantities of aluminium have lately been used by
leel workers. A faction of a per cent, of aluminium is added to the
teel, generally for the purpose of obtaining sound clean castings, and
> is probable that half the steel castings made at the present day are
sodered free from *' blow-holes " by the addition of small amounts of
InmiDium. To the second class belong numerous alloys of aluminium
ith small percentages of copper, tin, zinc, silver, or titanium added
ff hardening and stiffening purposes.

In fact, the chief demand for aluminium is at present — and pro-
ibly always will be — in connection with its alloys. The use of the
letal itself will be chiefly for light machinery, fittings for lighting
id water supply, astronomical and scientific instruments, and art
«tal work generally. With copper, aluminium forms a series of
lautifal alloys, varying in colour from bluish-white to red-gold and
de yellow, differing very widely in physical and mechanical proper-
B8. Those containing 60 to 70 per cent, aluminium are very brittle,
a88-hard; and beautifully crystalline ; with 50 per cent., the alloy

quite soft; but under 30 per cent, the hardness returns. Tne
I per cent, bronze has a whitish-yellow tint, somewhat resembling
smath. It is very brittle, and can be pulverised in a mortar. The
ittlenees of alloys above 11 per cent, prevents their use, but from

per cent, down to 1 per cent, they are most valuable. Their
tjBical and mechanical properties render them useful for every
^riety of metal work, and the high price of aluminium has been the
ily restriction to their use hitherto. In resistance to compression, a
^ per cent, alloy equals the best cast steel ; its transverse strength

rigidity is about 40 times greater than that of ordinary brass, and
9 limit of elasticity equals that of steel of the same tensile strength
id elongation. One of the most valuable properties of these alloys

the fact that they are forgeable, and can be worked at a bright red
At as easily as wrought iron ; all other bronzes are ** hot-short."
bo rolling of sheets, rods, bars, and wire is done at a bright heat ;
dj the finishing is done in the cold. A casting which had a tensile

Digitized by



resiBtanoe of 60,000 lb. to 65,000 lb. per sq. in. with 20 per ooii
elongation, when rolled, showed 83,000 lb. tensile resistance wi^
30 per cent, elongation. These alloys retain their strength thrcm^b t
high range of temperature, a qnality most valuable in a metal. Their
resistance to corrosion, which is only a degree short of that poGseflsed
by gold and platinum, makes them invaluable to shipbuilders, mA^iIl^
hydraulic, and sanitary engineers. This quality, combined with their
colour, render tbem far superior to ordinary bronze for art metal work
and statuary. The advent of smokeless powder, with its greater
corrosive action on ordinary gun-metal and steel, is already caosbg
military engineers to look round for a material to replace these metak
in the artillery and small arms of the immediate future, and, in the
peculiar combination of properties required, aluminium bronze is ns-
approachable by any other metal or alloy, and the solution of thisqiue-
tion will undoubtedly lead to a great demand for these alloys. Tbe
aluminium brasses are made by combining the metal with copper utd
zinc, yielding dose-grain homogeneous and remarkably tough alkvs.
They are made in two grades, and can be forged hot ; their special
casting properties, hieh corrosion resistance, and low specific gn^itj
render tnem most valuable foB hydraulic and engineering purposes
A propeller blade can be made of No. 2 brass at least one-third thiniMr
than cast iron. The high elastic limit, surpassing that of alloys used
hitherto for this purpose, would greatly lessen the chance of injury ts
the blades, in the event of contact with wreckage, "whales, Ac, it
present a frequent source of danger and delay. In these days of £ut
steaming, the weakening of blades by pitting and corrosion wonM be
reduced to a minimum if made of aluminium brass.

Digitized by




rms metal oocnrs in three forms : (1) the oxide, senarmostite, SbjOs,
ontAmmg 83*56 per cent, antimony; (2) the sulphide, stibnite,
(Dtimonite, or antimony glance, SbjSs, affording 71*8 per cent.;
3) a Hulph-oxide, kermesite, 2Sb2S3, SbjOj, giving 75' 72 per cent. ; in
iddition to some unimportant combinations with silver, dec. Beyond
he considerable quantities of oxide coming from Algeria, and of
:ermeeite from Tuscany, almost the entire output is in the form of
tibDite: and while it may be said that antimony is somewhat
ridely distributed in nature, yet, owing to cost and difficulties in

Online LibraryCharles George Warnford LockEconomic mining: a practical handbook for the miner, the metallurgist and ... → online text (page 44 of 76)