Charles George Warnford Lock.

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

. (page 69 of 76)
Online LibraryCharles George Warnford LockEconomic mining: a practical handbook for the miner, the metallurgist and ... → online text (page 69 of 76)
Font size
QR-code for this ebook

)n the copper sulphate present. The quantity of magistral used will
.aiy with its quality, but is commonly i-2 per cent., depending
)artly, also, on whether the ore contains antimony, arsenic, sulphur, or
inc; it is applied as a hot solution. Mercury at the rate of 6-8
imes the silver contents is introduced by squeezing little showers
hrough canvas bags at intervals, and the whole mass is most
horoughly incorporated — ore, magistral, and mercury — by further
mle treading. Temperature and correct proportions of magistral
irgely influence the result, excess of either causing an exceptional
J6S of mercury, which is normally equal to the silver extracted. The
peration is very slow, occupying 2-3 weeks in summer and 6-7 weeks
1 winter. The theory of the process is much debated, but seems to
e as follows: — The mercury acting on the copper chloride makes
lb-chlorides of both ; the copper chloride absorbs oxygen, which acts
n the silver sulphide, makes sulphuric acid^ and leaves the silver in
metallic state to be absorbed by the mercury ; the sulphuric acid
)t A^e acts on the sodium chloride, and forms soda sulphate ; chlorine

given off, combines with the sub-chloride to make copper chloride,
hich is again decomposed, and so on. The action of the chemicals
I the pile is especially slow if silver sulphide is present, in which
i8e the loss of hiercury is also very large ; when the whole of the
Iver is in the state of sulphide, a large part of it, up to 40 per cent.,
lost. The mercury transforms the copper chloride into sub-chloride,
hich, like silver chloride, is soluble in an excess of salt ; the sub-
iloride in this state acts more energetically on the silver sulphide
lan the chloride ; copper sulphide is formed,
hile the silver is precipitated, and the
pper chloride is formed again by giving up
ilf the copper, which becomes a sulphide,
lis advantage is gained only at the expense
a very large quantity of mercury.

Test samples are repeatedly taken, and
len they show that not less than 75 per
at. of the silver is amalgamated, the mass
washed as quickly as possible, either in
fina (mechanical settler) or in a lavadero
>x settler). The latter (Fig. 171) is built
-ectly on the edge of the patio a, the walls
>eiiig of stone, lined with cement ; it mea-
res 6 ft. long, 3 ft. deep, and IJ ft. wide,
i has a platform c from which it is fed, and Fia. 171.— Lavadero.
Tont of plank d, perforated by 6 holes e,

in. diam., closed by wooden plugs, which let off the slimes into a
-tical partition /, whence they pass to the inclined trough ^,
nished with a series of mercury traps h. The material is carried
to the platform c by means of the stairs t, and is fed slowly into
; settler, with a stream of water, while two men get into the settler

Digitized by



and keep the contents agitated by their feet, the fines flowing off by
the various holes e as fast as the water runs in. Much skill is needed
to prevent loss of floured mercury with the slimes, and the work is
done more quickly and easily by the modem mechanical setthng pan.
The heavy valuable portions (amalgam, mercury, and onatUcked
Bulphurets) collected in the settler and in the traps are concentiated
in various contrivances, and the silver is recovered by retorting the
amalgam as usual. The whole process is much more wastefol than
the Freiberg barrel, and is only adapted to hot dry climates, bat it
needs no capital to speak of.

The Cazo or Fondo process is very simple and rapid, and consists
in boiling the argentiferous matters, with addition of salt, copper
sulphate and mercury, under constant agitation, in order to (^ed
amalgamation. Originally the cazo or cauldron was entirely of copper,
but now the sides are generally of wood. Water is added in suffid^kcr
to make the pulp thin, and it is brought to the boiling point before
any salt is added ; 5-15 per cent of the latter is used. ConstaGt
agitation to prevent adhesion to the copper bottom is most essential
Sometimes large pieces (over 1 cwt.) of copper are suspended in tbc
cazo and rotated as mullers. The loss of mercury is only about 2 pe:
cent. Silver sulphides are not acted on, and any present will be imsA
in the tailings. When the ore contains much bromide or chloride,
lead to the extent of 25 per cent, of the weight of silver present k
first added to the mercury, and thus, by its greater ^finitj for
bromine and chlorine, saves the mercury from attack, and reduce*
the loss of mercury £h)m 150 per cent, aown to 25 per cent on tie^
weight of silver extracted.

The working of the process of chloridising-roast and amalgaxnatksi
at the Colquechaca mines,* Bolivia, is as follows : — The rich ore, abort
300 oz. per ton, is shipped to Europe ; the rest is treated loo^y. A
battery of 11 400-lb. stamps crushes 8 tons per 24 hours in the vet
season and ^ in the dry. Single-hearth reverberatories roast 6 charge^
and double-hearth 7 charges, of 400 lb. per 24 hours; and V^ p^-
cent, salt is added 2^^ hours after the ore. Fuel is yareta or taqnk
350 lb. for each charge, costing 9(i. per 100 lb. Labour costs U. 6d
and salt 1«. 8c2. per (marge. The fondos take 15U lb. ore and 50 II
salt every 3 hours ; labour costs I9., fuel I9., and salt la. 8d. per charge;
and 10-12 lb. mercury are used for each 250 oz. silver in the ore, t^^
loss being \ oz. per oz. silver produced, or 8 lb. mercury, costing ^^
per ton of ore. Wooden barrels sometimes replace the fondos, aci
take a charge of 600 lb. ore and 200 lb. tailings, requiiinf^ 6-8 husis
for 120-150 oz, ore, and 15 hours for 250 oz. upwards; each barrJ
has 25-30 IJ-lb. copper balls to mix the charge ; salt used per chaip^i
125 lb., costing 4*. ; mercury, 1-1 J oz. per oz. silver in ore ; labos:
costs 1«. 6(2. per charge, fuel 5(f., salt 38. 9(2. ; total cost, indadlB:
mercury loss aud wear and tear, 7«. per 600 lb. The ore genenUy
carries 250-300 oz. silver, and the loss in tailings is 25-33 per oeit
The actual cost per ton of 2000 lb. is approximately : —

Crushing : labour, 6«. ; repairs, 3«. ; total, 9«.

Boasting: labour, 7«.; salt, 8«.; fuel, 12«.; repairs, dc, 2#.; total, ^*«
♦ R. Peele, op. cit.

Digitized by



Barrel amalgamation : labour, 59. ; salt, 12«. ; fuel, 1«. 6<2. ; mercury
loss, 14«. ; repairs and copper loss, 69. ; total, 38^. 6c2.

Fondo amalgamation : labour, 18«. ; salt, 2I9. ; fuel, 12«. ; repairs,
2«. ; mercury loss, 80«. ; total, 78«.

Betorting, &o., h%. ; superintendence and office expenses on 150 tons
a month, 12«. ; general expenses and interest, 8«. ; total, 25^.

Grand total with barrels, hi, \%. 6c2. per ton ; grand total with
fondos, 11. 1«. per ton.

Francke*s Una process* is simply a fondo heated by steam and
agitated by steam power. The fondo extracts * 80-85 per cent, of
the silver in the very base ^galena, blende, &c.) Bolivian ores, while
the tina does better; and tne cost of the fondo process at Potosi,
running the stamps by water power, is not quite 8/. a ton (2000 lb.).
Wendt replaced the local form of calcining furnace by lump-roasting
kilns, using no fuel, and reduced the cost to I9. Sd. a ton, attaining also
volatilisation of the antimony and arsenic, and rendering subsequent
stamping a very easy matter, but unpleasant owing to excessive dust.
Pile-roasting removed more sulphur, but less antimony and arsenic,
rhe sweet ore then goes to the ohloridising roast. Eotary roasters
jave 10-15 per cent, loss by volatilisation, while 3-floor gas-fired
reverberatories reduced it to 5 per cent. Complete oxidation before
ohloridising is the remedy. Dealing with the 75-80 oz. ores of Potosi,
^Vendt got an extraction of 80 per cent. (10 per oent. lost in roasting
md 10 per cent, in the tailings) silver, 900 fine, costing, on 8 tons a
lay, about 3Z. a ton. The tin oxide encountered in the ore is mostly
cashed out and concentrated for market before charging into the
bndo9 ; it carries some silver when smelted.

At many mines there is considerable ore that is too low grade to
ustify smelting or preliminary roasting, and not free enough to
kmalgamate raw. Concentration may then be advisable. As an
ixample of this, the figures obtained at Black Pine,! Montana, may
)e quoted. The concentrates, 20 into 1, contained 82^ per cent, silica,
► lead, 8i copper, 1 • 19 sulphur, • 81 zinc, and • 54 silver. A year s
working to May 31, 1889, gave 9062 tons crushed ; average per stamp,
!'612 tons per 24 hours; average assay of ore, 22*67 oz. silver per
on; concentrates produced, 542 tons; average assay, 136*17 oz.
ilver ; percentage saved, 83 * 45 ; total cost per ton, 17#. 6(i., viz. 10«. 2d.
or labour and superintendence, 2%, for mercury. Is. Id, for castings
.nd iron, \%. 3(2. for salt and other chemicals, lOd. for fuel, 5d. for
ubricants and illuminants, and I9. 3(2. miscellaneous. The escaping
limes from the mill were found to be carrying 52 oz. silver per ton,
r double the contents of the ore going into the battery. This was
^rtially remedied by using the slime waters in the battery instead
•f clean water, and at the heads of the concentrators. The best size
o crush to was found to be 40 mesh. Pan amalgamation is used, the
aoet suitable charge proving to be 50 lb. salt, 2 lb. sulphuric acid,
nd \ lb. potassium cyanide, with 100 lb. mercury strained in after
he pan has been running 4 hours; the pulp is steam-heated to

♦ A F. Wendt, op. cit.

t C. W. Goodale and W, A. Akers, " Concentration before Amalgamation, &c.,"
ranfl. Amer. Inst Miu. Engs., June 1889.

Digitized by



IdO"" F., and the charge is run at 65 rev. for 8 boors ; settlerB are nm
14 rev., and give good agitation with 8-in. shoes.

The ores of the Tombstone district,* Arizona, contain mod)
manganese, the " milling " ore carrying 43 oz. silver per \xm having
41-74 per cent, manganese peroxide, and the smelting ore wortli
23*3 oz. silver, 47*7 per cent, manganese. Free milling aftw hand
sorting to remove as much manganese as possible only gave 60 per
cent, silver recovery with a loss of 7 lb. mercury per ton. Opinions
differ as to the precise action of the manganese on the mercury.
Goodale says that only the earthy oxides of manganese are tronble-
some, and tiiat hard granular pyrolusite is harmless ; that expulsion
of water of composition from wad and psilomelane in roasting destroys
their earthv character, and with it their tendency to " foul ^ mercury.
This would indicate a mechanical effect only. Pearce belieres that
MnOj may have an oxidising action on mercury, as the flouring can
generally be avoided by adding metal (lead, copper, &c.) more oxi-
disable than mercury, and per contra, that when amalgamating base
silver ores, addition of manganese oxide helps to prevent reduction
and amalgamation of the base metals and produces a finer quality
bullion. Further, Clark, at the Moulton mill, found that whenever
much zinc was present in the ore, loss of silver by volatilisation was
greatly reduced by plentiful admixture of oxidised mannnese ore.
The iron of the pan plays a most important part in pan amalgamation,
and great losses may occur in the tailings if for any reason (such as
a coating of oxide or slimes) the iron oi the pan cannot be attacked
by the charge, so that sometimes wrought-iron bands are added inside
the pan to ensure a sufficient supply of iron. Lead occurring as
carbonate or sulphide was not found (by Church, at Tombstone) to
debase the bullion from the pan, but tellurides had a marked e£fect
"With freely amalgamating ore, the pan charge was h\ lb. salt and
1 lb. bluestone per ton, vnlh a consumption of 1 lb. mercury ; as the
ores became more sulphuretted, 5*22 lb. salt and 1*2 lb. bluestoDe,
and 1 * 258 lb. mercury. Subsequent re-working of tailings showed
that the mercury was consumed mainly by chemical combiQation,and
was finally lost. Lime in excess, while not preventing reductioD
and amalgamation of silver chloride, does interfere seriously in pan
amalgamation, and the use of lime to prevent amalgamation of
base metal will cause loss of silver in the tailings.

In the Boss process, handling of the pulp is avoided, as it flows
successively tlm)ugh all the pans and settlers. Obviously this »
possible only with pulp which can be worked very thin. At the
Garfield mill, Calico, with very free milling or^, the plant consista
of a battery of 15 stamps (weight 850 lb., fall 8 in.), fed by the itenal
Hendy feeder, and the pulp discharged through a SO-mesh screen;
2000 gal. water are used per ton of ore, and about 33 tons are crashed
a day. Hot water is used in the battery. From the battery the
thin pulp is conducted directly to the first pan. The pans and settlers
are all on the same level, and are so connected that the pulp may
flow freely from one to another. All are driven directly ficom the

• C. W. Goodale, " Arj^cntiferous Manganese Ores of Tombstone," Tt!mi& Abct.
t. Min. Engs,, xvii. 767.

Digitized by



main shaft by means of bevel gearing and fHction olntcbes, so that
any pan may be thrown out, if repairs, &c., are necessary. In that
case connection is established between the others, so that no interrup-
tion of the work takes place. The pulp in the isolated pan is then
pumped nut by means of a siphon-shaped steam-injector. There are
8 pans, 5^ ft. diam., and making 65 rev. a minute ; also 3 settlers
8 fL diam^ making 20 rev. a minute. A little mercury is added
in the first pan; uie muller is let down, and the pulp is ground.
Experiments have been made with grinding in tiie second and
third pan also. Exhaust-steam is introduced from below into the
false bottoms and cones of all the pans, and a thermometer is
attached giving the temperature of the pulp. In the third pan
copper sulphate and sodium chloride are added — of the former alK)ut
1-1 J lb., of the latter, 8-15 lb. |>er ton of ore. This is done by means
of a self-feeder : a slowly revolving disk witii about 10 small buckets
filled with bluestone and salt, which are autcmiatically discharged in
the pans at certain intervals. In the third pan, the last portion of
the mercury is added. In the fifth pan a little caustic lime is fed —
also by a revolving automatic feeder — ^in order to clean the mercury.
From the last of the pans, the pulp flows into the settlers, where the
remaining amaleam is collected. Sprays of water from pipes, radi^v
arranged, aid the settling. About 95 per cent^ of the amalgam is
drawn firom the first three pann. It is estimated that the purest
chloride ores are worked up to 90-95 per cent, of the assay-value ;
baser ores do not yield more than 75-80 per oent.

Smelting is resorted to when the silver is accompanied by much
lead or copper. Indeed, in the United States, where by-products of
silver-bearing ores, such as arsenic, bismuth, sulphur and zinc, are
disregarded, smelting is the most popular method of treatment for all
except the free milling ores, no matter what the gangue. The opera*
tioQ IS really a concentration of the precious metal in metallic lead or
copper matte, and has been fully described under the respective heads
of copper (p. 415) and lead (p. 515). A suggested * modification of the
copper process is to concentrate the precious metals in pure copper
instead of matte. The extraction is said to he quite as good as with
metallio lead, and no loss arises from volatilisation ; but the loss of
copper in slags is a more serious matter by reason of its greater value
(2^-4 times that of lead), though this is partly neutndised by the
fact that conoentration may be carried further (20 to 1 as against
10 to 1), so that charges can be dealt with canying only 5 per cent,
copper as compared with 10 per cent. lead. The method, however,
can have but limited application, being adapted only to the uncommon
non-amalgamable ores froe from lead and containing less than 1^ per
cent, sulphur.

The smelting of tailings from the pan-amalgamation mills work-
ing on manganese-silver ores at Tombstone,! Arizona, is of special
interest. Concentration was efiected by trommels, jigs, and buddies ;
the last-named gave exceedingly good results on the fine, slimy
tailings, which Frue vanners quite failed to treat. The buddies were

• H. F. Collins, ** Smelting Processes for the Extinction of Silver and GoW from
their Ores," Proc. Inst. C.E., Paper No. 2665. f J. A. Ohoroh, op. cit.

?^ I

Digitized by VjOOQlC


15 ft. diam^ tamed 105 times in 100 minutes, and had a dopeTtiying
from 7 in. in 7^ ft. fur coarse slimes to 4^ in. for fine slimes, the bed
being covered with Akron cement ; water jets were nsed entirelj
instead of brushes. They treated a ton an hour each constantlj, tad
saved 77^ per cent, of the lead, 55j^ of the gold, and 53 of the iflfer,
in addition to which the slimes flowing from them, after settling, ga^
8-10 per cent, lead and 12-15 oz. silver per ton, at a total cost of sboat
5<. a ton of tailings, using steam power and hand labour. The con-
centrates were made into bricks by binding with pan slimes cootaiikiog
85 per cent, quartz, 2^ per cent, dav, and some calcite, manganese,
iron oxides, various sulphides, and lead carbonate. The flux wed
was manganese ore, which presented two peculiarities. The fluiditj
of the slag allowed less fusible impurities to settle rapidly and oom-
pletely out of it, and the furnace would accumulate crusts in the
hearth with great suddenness. This tendency was increased by the
absence of matte-forming materials. The manganese sulphide is di»-
sociated readily by heat, and the small quantity of iron, copper,
nickel, and antimony present were just sufficient to make a speiw
with the arsenic present. Usually the speiss ran out with theedag;
but if anything occurred to stop the flow of materials through th<
hearth, even for a short time, a crust was almost sure to form, tnd
once formed it was very hard to melt. When the charge was stroo^ Ij
basic, the furnace would melt 50-55 tons a day, but there was a
strong tendencnr to accumulate crusts. With a more acid diarge, tiie
work was much more regular, and the furnace melted about ^ toD»
a day. Though the composition of the slag varied daily, owing to
the unfavourable conditions for fluxing, the slags were always Teiy
clean and remarkably free from combined lead and silver. Hieir
extreme fluidity and the tenacity with which manganese retains its
oxygen, and the readiness with which it gives up sulphur, are pro-
bably the causes which contribute to this freedom from lead. Tba
experience obtained indicated that manganese would form an excellent
flux in matting-furnaces. The charges consisted of 35 * 2 per cent
concentrates, cdimee, and flue dust, 13*1 ore, 41*1 manganese, I'i
limestone, 9*2 slag and cleanings; fuel, chiefly American coke, 21*33
per cent.

At the Eelsey mine, California, the assorted ore contains 10(K^

1400 oz. silver per ton, 7-15 per cent, cobalt, and 2-3 nickel Dr.

Endlich adopts the following method : — The ore is crushed throngb t

20-mesh sieve, mixed with sufficient lithai^ to produce an 8 per

cent, charge, and enough borax is added to take up the gangue (qiurtx^

heavy spar, lime carbonate, magnesia, and iron). Soda carbonate and

flour are mixed with the charge. If the percentage of arsenic in tiie

ore is sufficiently high to produce speiss, none is added ; otherwisi

some metallic arsenic is mixed in. Some sulphides in the ore vA

reduced sulphur from the heavy spar are utilised to produce mattes.

The mixture is melted in large Dixon crucibles ; the slag is ponrtd

off, and the metallic product is allowed to cool. The bars obtan^

imposed of lead, silver, cobalt, nickel, arsenic, and sulpb^

pally ; the lead being in the form of sulphide, the cobalt ar

in the form of arsenides. The baiB contain 4500-7000

Digitized by VjOOQIC



silVdr per tfiH. The slag eontains a trace of silver, and averages
ahoQt * 75 per cent, cobalt, which can be worked over by arsenising,
if desired, and the cobalt obtained in the resulting speiss.

A Mexican method, applicable when the ores contain much
snlphnr and are easily fusible, consists in liquating the ores in an
adobe furnace 10-15 ft. long, with an inclined chimney* A cord of
wood will liquate several tons of ore. These chimneys are surrounded
with cla^ to retain the heat. When the fluxes are easily fusible, the
fumaoe is built as a square chimney 10 fL high and 8-10 in. wide at
the base, with a small hole at the bottom for the bellows, and a
tap-hole on the opposite side. The charging hole is about 5 ft. from
the ground. A quart of ore and 2 quarts charcoal are thrown in
iltemately until the furnace is full. The furnace is lighted, and
icon acts like a blow-pipe on a large scale. An iron bar to skim the
(lag from the metal is tne only tod necessary. The metal is run out
nto clav moulds. The furnaces have a capacity of about 5 tons a
lay, and cost 62. for charcoal. A plant of this size can be constructed
or about 60/.

The ores treated at Las Trojes,* Michoaran, Mexico, are of a
lass generally deemed unsuitable for smelting, their composition
eing approximately: 15-30 per cent, silica, 18-25 sulphur, 10-30
ron, 2-12 zinc, and traces of alumina, antimony, copper, lead, and
ime. Ore carrying less than 28 oz. silver per ton is not considered
rorth mining. Heap-roasting as a preliminary loses considerable
ilver, and leaves 5-8 per cent, sulphur in the ore. The smelting
lixture consists of 40 per cent, roasted ore, 35 slag, 10 roasted matte,
-10 litharge, and 5 lime ; or 15-20 oz. silver per ton, 12 per cent,
^ad, and 2-8 per cent. zinc. A charge is 600 lb. mixture and 80 lb.
lel. The lime cents 199. 6d. a ton. The fuel is 31 per cent, coke
leting 62«. a ton, and 69 per cent, charcoal (principidly pinewood,
jry inferior, and only 50-70 per cent, serviceable) costing 21«. 7d. a
»n. The furnace products are : — (o) " base bullion " carrying -4 per
nt. silver, or 75 per cent, of the silver in the original charge ; (b)
on roatte containing 20 per cent, of the original silver, or 20 oz. per
n, which is heap-roasted and re-smelted ; .(e) wa^te slag holding the
maining 5 per cent, of original silver, or 1-1 J oz. per ton. The total
sr, including refining base bullion, is 16«. per ton of mixture, or 37«.—
3, per ton of ore.

At Casapalca,! Peru, the great elevation (nearly 14,000 ft.) inr
•Ivee the oondition that owing to lessened density, the available air
nd oxygen) is only 54 per cent, of what it is at sea-level. Hence
Bbbting furnaces have only half the normal capacity, and smelting
maces require double the blowing power. Most of the ores carry
timony, arsenic or sulphur, and need prelinunary roasting, which

universally done in reverberatories when amalgamation is to
low. ^ Lead does not pay to ship so far, so the silver is extracted by

A kind of pyritic smelting is bometimes adopted^ e. g. at Mineral,

• J. W. Malcolmflon, «« Erection of SiWerLcad SmeUiag Works in Mexico,"
IT*. Inst Civ. Knv'fl., Paper No. 2583.
f A. L. Penroo, ** Mining in Peru,*' TrenB. Inst. Min. and Met

2 B 2

Digitized by



Idaho, on an ore resembling that of Las Yedras, Mexioo ^25 mliea,
46 calcite, 9*8 iron, 12*5 siuphuri 2*5 arsenic^, except for the eieen
of lime carbonate. According to H. Lang, this ore can be nm down
at one operation into a hi^h-grade matte at less than the cost of tba
salt for chloridising-roasting, and the matte can then be r^ned tod
its total silver extracted at an additional cost not exceeding that of
the chemicalB nsed in Bussell leaching. Lang nsee no flax, and about
7 per cent, fuel, adapting the fnmaoe and blast so as to bum off ma^
of the sulphnr and arsenic, and slag off the corresponding portion of
the iron and zinc, utilising the heat of combustion of uie elemeDts

Wet ways of extracting silver are founded on the solubilitj of the
metal and some of its salts in certain rengents. Experience bs
shown that the best results are obtained when the silver is pttfient
chiefly as chloride, hence a chloridising-roast almost always preoeda
lixiviation. But highly oxidised ores, where salt and fuel ara t^
dear, may economidSly be treated raw, though the extraction will be
less complete ; and in other cases an oxidising-roast (without alt)
may be admissible. As a general rule, however, very effideni
chloridising-roasting is an essential to high extraction, even moie m

Online LibraryCharles George Warnford LockEconomic mining: a practical handbook for the miner, the metallurgist and ... → online text (page 69 of 76)