Charles George Warnford Lock.

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

. (page 52 of 76)
Online LibraryCharles George Warnford LockEconomic mining: a practical handbook for the miner, the metallurgist and ... → online text (page 52 of 76)
Font size
QR-code for this ebook

to its place. When relining is necessary, the converter is landed <h
the floor, and the hood is detached by knocking out the wedges, whei
the sanded (or graphite may be tuBed) junction in the wall will giY<
an easy line of fracture. The crane lifts the hood and places i
upright on an iron frame 8 ft. high and hollow beneath. The pa
wedges are knocked out and the pins are driven inward through thi
lining with a heavy sledge, by which the lining is broken up and
detached. The hood lining lasts a long time, but must be removed
when it builds up and chokes the converter.

The body is next turned by the crane on its side, then completeh
upside down, and suspended at 1-2 in. from the floor. If the whoH
lining with the false sides and bottom together does not drop out, i
few taps on the sides will produce this result. The side and bottoi^
pieces are easily detached separately by jarring, and then the whoh
lining may be picked up by the crane and pitied in an iron mortal
to be broken up by a heavy chunk of iron being dropped on it. Thj
unbumed pieces may then be returned to the quartz pan for regrinl
ing and mixing with new composition. The pieoes badly coated a]||
impregnated with copper should be sent to the blast-furnace.

The time and labour saved by this construction of converter i
its manipulation can hardly be estimated by one who has not i
nessed the daily struggles of workmen with hard-baked linings
the old styles of converters. By dumping the old lining while ho

Digitized by



DO time need be lost in oooling by any syBtem* as the thin steel shell
maj be sprayed and cooled sufficiently for relining in a very few

Argning from the great advantages which ensue from the replace-
ment of the ordinary silicious linings of steel-melting furnaces by a
liiilDg composed of basic material, when phosphoric pig-iron has to
be converted, it might reasonably be expected that the substitution
of a basic lining for the silicious lining hitherto employed in copper-
smelting furnaces, should also be followed by similarly advantageous
results, especially when the cupriferous material to be treated contains
any notable percentage of arsenic or antimony. Operations in this
direction have been very successfully carried out at a large English
smeltine worl&, and described at length by P. C. Gilchnst.* The
roaster mmacee have cast-iron bottom plates, underneath which a free
current of air circulates ; by this means the bottom of the furnace is
kept cool ; it likewise prevents the possibility of any fusing action
taking place between the basic hearth and its support, which might
be the case were the basic hearth built directly upon the ordinary
silica arch. The basic material is ground and mixed with tar in the
nsoal way, and the furnace bottom is formed by throwing this material
into the hot furnace and burning it on in layers, well beating down
each separate layer, and giving it fire for some hours before applying
a fresh layer. It usually takes 4-5 days to bum on a bottom in this
way. When the bottom is properly shaped, it should be seasoned by
melting on it some rich copper precipitate or good blister copper.
It was at first considered that the basic hearth absorbed less copper
tban the ordinary sand ones, but there appears to be very little, if
UDy, diflferenoe between them, much depending on the way the
bottom is formed and seasoned. The tap-hole of the furnace is shut
by throwing a little basic material against it from the inside. In
)ther respects the working of the furnace is conducted in the usual
Mray. After each charge any slight repairs that the banks may
require are made by throwing some basic material against the place
leeding repair; the repairs required are, however, very slight in
XMDparison with an acid furnace, the tendency being for the furnace
janks to ^row rather than to cut away.

The naineral used is arsenical, and although in the crude ore the
rsenio is not high, yet when the ore is allowed to oxidise in heaps,
nd the copper is afterwards precipitated from the solution obtained
>j lixiviating the ore, arsenic is concentrated in the precipitated
opper to a considerable extent, the resulting precipitate usually con-
aining 3-3j^ per cent Very large quantities of tnis precipitate are
reatecL It is added to the mixture of slag and metal charged into
be smelting furnace, and according to the amount of precipitate so
dded to the charge, more or less of the product tapped from the
melter consists of impure copper, known as '* metallic bottoms." An
vera^ analysis of these bottoms gives 83-87 per cent ooppeo*, 5-7
rsenic, 1-3 sulphur, 3-5 lead, \ each iron and silica. It is in the
mversion of tnese bottoms into blister copper containing under
* Paper read before Soc Ghem. Ind., London section, Jan. 5, 1891.

Digitized by




1 per oent. arsenic that the basic linings have proved so superior, as
illustrated by the following example : —



tofw cwt. qr.

tuwi evt. qr.

Metallic bottoms osed during 12 weeks

400 10 U

400 10

AyeraKO analysis 84*52 per cent copper and

5*91 arsenic

59 charges made, averaging per charge

Blister copper prodaced from same

6 15 1

6 15 1

323 6 2

191 15

Ayerage produce per charge

5 9 2

3 5

Average arsenic 1*11 per cent

Slag made ftom above charges

102 16 3

221 15

Average weight per charge

1 14 3

3 15

Average copper percentage

Time occupied per charge, including fettling,

25 per cent.

55 percent

charging; &c

29} hours


Calculating from the above figures, we find that taking the real
copper in the ** metallic bottoms" at 84* 6 per cent., and the real
copper in blister at 98*5, there was obtained in the form of blister
94 per oent of the real copper from the basic furnace, and 56 per
oent. of the real copper from the acid furnace, showing a gain of
38 per cent, in favour of the basic furnace. The real copper in the
slag works out to 25 tons 14 cwt. from the basic fumaoe, and
121 tons 19 cwt. from the acid furnace.

At Deville, near Rouen, for refining arsenical copper, the usual
silicious lining is replaced by a basic bottom, for which a basque
of lime and tar is employed. Every operation of refining is per-
formed on a false bottom of limestone, mixed with manganese
peroxide, on which the ingots are placed. The false bottom gives off
carbonic acid and oxygen, and these traverse the half-melted copper,

Euddling and oxidising it When the bath is sufficiently liquid, the
me and manganese protoxide rise through the copper and dissolTo
the arsenic acid, which thus passes into the cinder. About 80 per
cent, of the arsenic is removed. To drive off the last trace, the copper
is left to become past^ under a current of air, and then again melted
with basic fluxes until it is completely purified. The following is an
example of an operation on a sample of copper from Eio Tinto : —

Copper charged in ..
After first melting ..
After second melting
After third melting ..




per oenC





This process does away with the roasting of the ore, and the abseno»|
of silica diminishes the loss of copper in the cinder.

The " direct " method * of producing refined copper is baoed upon
the well-known reaction between oxide and sulphide of copper

* 0. Yautin, Trans. Inst. Min. and Met., ii. 76.

Digitized by



npon their being melted together, sulphurous anhydride being given
on, and metallic copper remaining, which is directly refined in the
fnrnaoe in which it is produced. White metal is the product pre-
ferred for the operation, crushed through a f-in. screen ; part is cal-
cined nearly sweet, mixed with the requisite proportion of uncalcined,
and filled into a refinery furnace, melted, skimmed, refined, and ladled
into ingots, cake, «&c., as may be required, the operation of melting
and refining taking the same time as if a charge of pimple or blister
copper were operated on. This method has been working continuously
at the Cape Copper Co.'s Works, Briton Ferry, since December 1890.
The white metal treated is produced in the usual way in Welsh
reverberatory furnaces, and averages about 75-76 per cent, copper.
The whole is crushed in rolls, and part is calcined in old-type rever-
beratory calciners, which put through 3J tons every 36 hours, and
in a revolving calciner of the Oxland and Hockin type, which puts
through about 10 tons a day, running continuously, and calcining
much better.

The calcined metal is sampled, and the sample is mixed with a
portion of raw metal, is melted in a crucible, and, from the result
of this trial, the proportions of the refinery charge are determined.
If the button of copper is covered with slag, then the proportion of
raw material must be increased, or, if the button is coarse or covered
with regule, it requires more calcining, the pitch aimed at being just
below blister copper when the whole charge is melted. The weighed
proportions of raw and calcined metal are now mixed for a 15-ton
charge, say, 9 tons revolver calcined and 6 tons raw, or 11 tons rever-
beratory calcined and 4 tons raw. The mixed charge is filled into the
refinery whilst it is still hot from the previous ladling ; the furnace
.s closed up and fired. In 4-5 hours the charge softens and flattens
iown, copper forming on the surface, and, as this hot copper penetrates
he charge, a gentle yeast-like working takes place, dense fumes of SOj
ire given off, so much so that the furnace flue is often inadequate to
any them off quickly enough, and they force out through every
Tack, the heat produced by 3ie chemical reaction aiding the reduc-
ion. The whole charge is quickly melted down, and very little slag
3 made. The furnace is now skimmed, when it takes about ^ hour's
abbling to set \% and it is ready for poling as if ordinary blister had
•een used. Each refinery works a charge of 15 tons metal, making
1 tons copper every 24 hours. Trials made to ascertain the best
nd cheapest pitch to which the metal should be calcined showed
hat though, when the metal was calcined to black oxide, it would
ecom]>ose its own weight of raw material, and when calcined to red
xide it would only decompose about half, yet the cost of extra cal-
Lnation counteracted the saving. On an experience of 14,000 tons
f ingots it is claimed that the direct method gives 36 per cent,
lore ingot copper and 44 per cent, less slag to be re-treated, whilst
le actual cost per ton of ingots made was 13«. 6d. a ton in its favour,
esides which the extra product means a saving of a further 2L a
jh of ingots in making white metal. But it is not at all evident
lat really impure ores or mattes could be satisfactorily treated in
lis way.

Digitized by VjOOQIC


In order to minimise the coet for renewal of alag- and matte-pots,
B. H. Terhnne has adopted a form with a movable bottom, whidi
overoomes the cracking tendency. The pot is cast with a 6-in. hole
cored in the bottom, in which is inserted a flanged bottom, 4 coonter-
sunk rivets making a tight and permanent joint.

P|fd-|>roc{tfcto. — ^These are chiefly bismnth, cobalt, gold, lead, man*
ganese, nickel, seleninm, and silver ; iron oxides, such as ochre uA
*' blue-billy " or pnrple ore; iron and line sulphates; tin oxide;
arsenic and sulphur. In many cases, only the copper, say 2-3 per
cent, of the nuneral mined, is saved, while the remaining 97-88
per cent, is wasted, and this is likely to continue in a great measme,
owing to the absence of a market for the majority of the products at
the seat of production, and the cost of transportation. Bat the
utilisation of the sulphur and the ferruginous matters is of primary
importance, and will surely be aocompliiuied in the near foture. Aii
improved method of roasting would enable the sulphur to be ocm-
verted at once into sulphuric acid, and that again might be used to
make sulphate of iron from such of the iron residues as were not fit
for paint-making.

Commerce. — The following table shows at a glanoe the approximate
relative importance of the principal copper-producing countries, stated
in English tons of fine copper : —

Algiers 120

Argentine »• 150

Australia —

Wallaroo 5,200

Other mines 2,500

Austria 1,200

Boliyia 4,500

Canada 4,000


Gape Copper Go. 5,350

Namiiqua 1,375

Chili .. 20,000

England 1,000

Qermauv —

Mansfeld 14,700

Other 2,000

Hungary 300

Italy 2,000

Japan .. 18,000


Boleo 6,200

Other 1,480

Newfoundland 1,900

Norway —

Yigsnaee 925

Other 450

Peru 150

Russia 5,000

Spain and Portugal —

RioTinto 30,200

Tharsis 10,800

Mason & Barry 4,400

Sevilla 1,000

Other 200

Sweden 800

Digitized by



United Statee—

Lake Superior 50,000

MontaDa 70,000

Arizona 20,000

Other States 6,000

Yenezoela 3,000

Total 274.900

Thus the United States afford more than half the whole output,
md Spain more than one-sixth.

Samjpling Copper. — ^English and American practice are at variance
n this important particular. The Cornish and Swansea methods of
xriving at the value of a parcel of copper or copper ore are fully
leecribed in the author's * Miner's Pocket Book,' p. 363. In America
lothing is left to judgment, everything depends on actual assay of a
lechanically taken sample, sampUng machines being of various kinds.
Coet sampling works have now adopted a form of sampler (e. e. the
trunton or the Constant) which diverts the entire stream during a
iven and definite interval of time. They can be so adjusted that the
rushed ore, elevated to a shute immediately above the sampling-
tachine, descends upon it, and during, say, 10 seconds the descending
:ream is diverted to the right, and during, say, 20 seconds it is
i verted to the left, so that whatever passes during the first 10 seconds,
hether it be fine or coarse, rich or poor, light or heavy, is all deflected
> a set of rolls. There it is pulverised, and is again raised to a
cond sampling-device, which again automatically diverts \ (now \
* the original lot), the discarded § in each case returning to the
mpling-noor, where it is bagged and returned to the car. The
oisture sample is taken from uie discarded f at the time when the
aterial is weighed into the car before proceeding on its journey,
lie sample, now representing, say, ^ of the original lot, which has
^Aeed the rolls, is returned to the floor, where upon an iron-covered
atform it is mixed together by hand and divided down by quarter-
g, which mast be the ultimate termination of all sampling. But
ere is no selection in this quartering, and it is always proceeded
itli bv the same rule and in the same maimer. The sampling of
pper oars is performed by boring, say, every fifth bar twice, half-way
rough on opposite sides. In sampling argentiferous bars of variable
tn position every bar is bored twice. K they carry gold also in any
antity (and always in the case of anodes), the borings are melted
d granulated, or recast into a sample bar, which is again bored.
lese are the only ways to secure uniformity in the laboratory
xiple and assays. Electrolytic assay is preferred for reliability,
Lug an Edison current. There are usually considerable differences
returns between American shippers and English buyers. Loss of
ig,ht will often average 2 per cent, on matte shipped in sacks, but
icli less with casks (e. g. old petroleum barrels). With ordinary
Xte, a common difference in copper assay is * 75 to 1 per cent, loss;
tlx argentiferous matte, l^lf per cent, copper and '3 oz. silver per
^lish ton. So that importers buying in the United States and
ling in England must make an average allowance of 4(2. per unit to
rer such differences. The market value of a parcel of copper ore is

Digitized by



arrived at by reckoning the " settled produce" or fine copper yielded
by it (say 4'55ner cent.^ at standard or current price of Chili Imrs
(say 70Z. a ton), and dedncting the "retnming charges." Th€«e
latter vary. In Cornwall they are fixed at 55f . per ton of ore, whether
rich or poor. In Swansea there is a fixed rate of \2». 2<f. per ton, kA
an additional sliding rate of 3«. 9cl. per unit of metal in the ore. The
presence of antimony, arsenic, bismuth, lead, or snlphur depreciAte&
the value, and a sensible percentage of either element may render the
parcel unsaleable.

Digitized by





[te geographical distribution * of gold is very wide. Approximate
gures rating to its production in 1891, stated in kilos, (of 2*2 lb.),
re given below : —

United States 49,917

Australasia 47,245

Bnada 36,310

Aftica 21,366

China 8,020

Colombia 5,224

British India 3,754



British Gniana


Mexico 1,505



French Guiana 825

Japan 775

Brazil 670

Dutch Guiana 668

Central America 226

France 200

Italy 160

Uruguay 140

Argentine 123

Peru 113

Bolivia 101

Sweden 88

Great Britain 4

Australian gold is still largely derived from plaoers (33 per cent.
Victoria), but the bulk of it is obtained from quartz reefs, mainly
free gold. There are enormous deposits of refractory ores whioa
ve not yet been developed.

In Queensland, the most notable auriferous deposit is at Mount
[>igan, 26 miles S.W. of Eockhampton, where about 75,000 tons of
3 up to the end of November 1889, had yielded 323,000 oz. of gold
line nearly 1,332,0002.), or over 4^ oz. per ton, at a working cost of
ly 17 per oent. on the value. This mine is equally remarkable for
3 various opinions expressed as to its genesis and geology, the most
tusible of which,t founded on features developed by extensive work-
^, seems to be that the ore deposit represents an altered mass of
ittered country rock, readily acted upon by mineral solutions, which
»laoed the basic and felsj^thic portions by acidic and quartzose
riferous material Its permeable and quartzose character has saved
frunx disintegration, and preserved it as an ore mass on the summit
a low hill, as seen in Fig. 125. The "country" consists, in
ending series, of dolerite, quartzite, greywack, and "desert*'
idstone, intersected by numerous dykes of rbyolite and dolerite.
e deposit itself overlies the quartzite, which is usually very pyritic,
e greywack (not seen on the mountain itself) underlies the sand-
ue in other parts of the district, and is usually so much metamor-
>sed as to be scarcely distinguishable from a true eruptive rock ;

•* replacement" probably occurred in this formation.
Surface cuttings show a mass of silicious iron-stained ore, which
the one hand shades off into crushed quartz and on the other into
• A. O. Lock, < Gold : its Occurrence and Extraction.' f I"- A. Itickard.

Digitized by VjOOQIC


siliciotis hematite ; the distribution is apparently irregular, except in
so far as the different varieties of material are bounded by dykes,
which throw off lateral branches, are irregular in direction (with a
general NJB. trend), and consist of kaolinised rock which is most pro-
bably rhyolite. There is no evidence of regularity in the bedding of
the material, nor are the joint planes persistent — the whole mas
appears to have been fissured and shattered by the violent dynunie
agency represented by the dykes. The outer portions of the ore body
show a greater proportion of iron oxides ; some portions carry small
crystals of pyrites, while the lowest edges lie upon the quartzite, the
boundary Ime being very distinct. "Where dolerite dykes form the
limit of the deposit, the contact is also well marked. The large masM
of highly pyritic auriferous quartzite suffice to account for the rich-
ness in gold. It is reported that the ground in immediate contact

Digitized by




bard tindeoomposed schists. At Cocoa Greek, the lodes occur in slates,
^ywacks, and quartzites, most probably Permo-carboniferous, and
}arry much stibnite and quartz, both auriferous.

In Victoria, dark-grey mineralised laminations are characteristic
>f the best and most permanent auriferous lodes.* Brown inclines to
he belief that '* all auriferous reefs have been deposited in the first
nstanoe at the time of fracture, and have in many instances been
enlarged by fresh eruptions of auriferous silica." E. J. Dunn describes
he occurrence in wash of small, white sandstone nodules, with a
erruginous kernel, and containing exceedingly minute particles of
;o]d disseminated through them, so that the average yield from many
ons was }-l oz. gold, and massive beds of soffc sandstone, which
umished tne nodules assay up to 9 dwt. Dunn quotes another in-
tance where fractured in sandstone and slate beds of Silurian aee
ave been filled witii quartz, forming a reef dipping from 30° to nearly
0° £. ; in some places in a horizontal displacement of 2 ft., besides
nearly vertical fault. The quartz occurs in the reef fiftult in an
nnsnal manner, only being found under the sandstone beds of the
anging wall, and along the strike of these beds even it makes af»
Tegular lenticular masses, ranging from 2-3 in. up to 2 ft in thick-
MS. Under the slate beds no quartz exists. Strike of the bedding
lanes, N. 36^-48'^ W. ; dip, nearly vertical. In some portions of the
indstone, thin veins of quartz are present, and a great deal of
irrites (arsenical). Gold in small quantities is found in such stone
he pyrites obtained in crushing the stone yields 2-3 oz. gold per

As to permanence there is abundant evidence that quartz does not
icessarily get poorer as one descends : richer and poorer zones may
t encountered, as shown by the following figures relating to the
iilway Tribute mine : —


>>300ft. .






OS. dwl gr.
846 16

8487 10

8740 6

Averafe per Ton.
1 1 17

16 21

1 8 20

In the Sandhurst district, two mines working below 2000 ft. are
tting excellent ore, thus : —

rkftb of BmC.




ATenge per too.




o«. dwt. gr.





1 11 16





1 9 16

B. A. F. Morraj.

2 H^2

Digitized by VjOOQIC


The subjoined table is interesting as showing the relations of depth

to yield in a number of Victorian mines : —

Width of R«et

Depth at which
Quarts waa got.

Qoaottty Croahed.

Arerafe Yield Of 0«ld
per Too.

OB. dvt gr.

ft Id.






13 15

8 6



13 22




11 9




4 9

16 ft to 20 ft



10 11




10 11

1 6



1 11 11

2 6



3 10

1 ft to 1 fL 6 io.



5 8

5 6



18 3









1 3




2 a to 12 ft



1 1 20

2 in. to 2 ft

1820 to 1920


1 1 8

7 6



10 9

3 in. to 2 ft. 6 in.



12 18




12 1

1 6



10 16

t ,,

1000 to 1200


10 5

1 6



4 5 22






200 and 300


10 3


• 570


2 16


880 to 1030


9 10




2 14

3 ft to 10 ft

630 to 700


8 10

' 2 ft to 8 ft

630 to 770


8 15

1 3



19 7




1 13




1 15 20

5 ft. to 8 ft

500 to 670


11 7


680 to 740


15 17

4 6



11 IS


800 to 1200


19 1




5 8




6 18




3 1 17

Some very rich parcels of quartz (9-10 oz. per ton) are got frd

^ squibs," or undefined reefs consisting mainly of small leadsa i

quartz mixed with slate and sandstone.

At Mount Doran (Fig. 126), the country rock a is a soft wiiii

sandstone, in which run bands of fine, soft slaty rock (. A qoil

leader c runs with the slaty rock ; it is about 4 in. wide, strikes!?. 1^

E., and dips 80*" W. Where it intersects the sandstone, it is barra

Online LibraryCharles George Warnford LockEconomic mining: a practical handbook for the miner, the metallurgist and ... → online text (page 52 of 76)