Charles George Warnford Lock.

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

. (page 70 of 76)
Online LibraryCharles George Warnford LockEconomic mining: a practical handbook for the miner, the metallurgist and ... → online text (page 70 of 76)
Font size
QR-code for this ebook

than when amalgamation is to follow, as in the latter event metallic
silver will be taken up by the mercury.

All forms * of furnace have admirers, showing that much depend
on the character of the ore. In rotating cylinders and reverberaton«
the operation cannot be hurried without great loss. Ores which btS
badly are not adapted to cylinder furnaces. Moreover, in this fam
of furnace, as pointed out by Aaron,! ^^® ^ admitted to the cylinclfl'
is not only largely deprived of its oxygen in passing through ^e fin.
but is still former rendered inoperative by the fact that owing to ^^
lightness it will lie above the stratum of heavy vapour (solplig:
dioxide, Ac."^ immediately resting upon the roasting ore ; and tk
admission oi cold air below the flames from the fireplace has bes
proposed as a partial remedy, and proved effective, reducing ec^
sumption of fuel and increasing capacity of furnace. The volatilitt-
tion and loss of silver in chloridising-roasting is a matter of tempeiv
ture and duration. Li this respect choice seems to lie betwe^
reverberatories and the Stetefeldt: in the former, the operation po^
grosses slowly and regularly, avoiding excess of heat at any time; m
the latter, the beat is intense but momentary, and the completion d.
chloridation is effected in the pit under moderate heat. Overheatof
may also decompose some of the chloride formed.

The most recent development of the Stetefeldt furnace in t^s
connection is the application of gaseous fuel. At the lixiviatioo ni3
of the Holden C0.4 Aspen, Colorado, coal gas is used for fuel in h^
drying and roasting. The gas plant consists of Taylor revdviif-
bottom producers, one 6 ft. diam. for the driers, and one 7 ft. dia&
for the Stetefeldt, using local inferior coals. The drying plant of»

* G. J. Rockwell, ** ObloridlsiDg-roastiDg and Lixiviation at Yedias Mine,* &
and Min. Jl., Feb. 1888.

t 0. H. Aaron, «• Notes on the Hydrometallargy of SiWer,** Bop. £tetolfi»
ralogist Calif., 1888, p. 847. % Willard 8. Mowe.

Digitized by



nsis of 4 Stetefeldt double-shelf driers, each fired by 2 gas-burners,
3^ being need for ore and the other \ for salt, the average moisture
being 6*13 per cent, in ore and 1 per cent in salt. The gross coal
x)n8umption, at 12f. 6<i. a ton, for drying, was 72*22 lb. per ton dried,
)r 5^. a ton; for roasting, 117*44 lb. per ton, costing 9d. The gas
% supplied to the furnace by 3 burners, 2 in the shaft and 1 in the
lue. Trouble was encountered by condensation of tar in the gas-
)ipe8, and Blauvelt suggests lining the pipes with fire-clay, so that
hey could readily be burned out, at the same time prolonging the
ife of the iron pipes. The average composition of the ore chloridised
January 1893 was: 21 -64 per cent. SiO,, 20*92 BaS04, 10*99 CaO,
0*02 Fe, 8*10 S. 4*24 MgO, 2*85 Zn, 2-27 Pb, and 2791 oz. silver
»er ton ; after r*iasting, *2 per cent, sulphur remains as sulphide.

The lixiviation processes are chiefly as follows : —

Augustine's : roasting the ore to drive off sulphur ; grinding the
oasted ore ; roasting with salt to form chlorides ; dissolving out the
blorides with a saturated solution of salt; precipitating the silver
y copper; compressing and melting the silver. It is mostly
mployed on copper mattes.

Ziervogel's: roasting mattes to produce sulphates; decomposing
11 sulphates except silver; dissolving out silver sulphate in hot
^ater. Jn practice, it is very difficult to determine the exact moment
hen all the other sulphates are decomposed and none of the silver
1; and to avoid loss by decomposition of silver sulphate to oxide,
>me is incurred by stopping short of complete decomposition of the
Jier sulphates, whereby some silver is precipitated, and mostly
•covered afterwards by Augustine treatment of the residues.

Yon Patera's: roasting to drive off sulphur, &o.; leaching with
[>t water to remove any soluble salts ; r'>asting with salt ; dissolving
ilorides by sodium hyposulphite (thiosulphate : these salts are
^mmonly <^ed " hypo " in everyday language) ; precipitating silver
f sodium polysulphide ; reducing silver sulphide.

Bliss's : same as Yon Patera's, but using calcium hypo and poly->
ilphide instead of the sodium salts.

Bossell's: roasted and washed ore is leached with sodium hypo
id iRrith sodiocuprous hypo ; metallic silver and various silver salts
keljr to occur in imperfectly roasted ores are soluble in presenoe of
e copper salt, and hence a certain source of loss is avoided. Among
dviation processes this has by far the widest application, and is the
ost satisfactory. In many cases it would compete successfully with
oalgamation on free-milling ores, in cost of plant, cost of working,
td percentage of extraction, besides avoiding the heavy dead capital
presented by the stock of mercury ; and is well adapted for dealing
ith tailings either from amalgamation or previous lixiviation. The
ossell process is« in &ct, a modification of or supplement to the
iters and the Kiss processes. In these methods, the extraction of
e silver is based upon the fact that silver chloride is easily soluble
solutions of sodium or calcium hypo, and that silver is precipitated
>ni such solutions by an alkaline »ulphide, with regeneration of the
rpo salts. In case the ore contains lead, a portion of the latter is
10 dissolved, lead sulphate being soluble in hypo solutions. If, at

Digitized by VjOOQIC


the same time, copper is present in the roasted ore in the fonn of
ouproiis chloride, the sulphides precipitated from the lixiviAtion
solution contain silver, copper, and lead. From auriferous silver ores,
gold is obtained together with the silver, but the percentage of its
extraction varies, and depends upon many circumstances. A high
ohloridation of the silver cannot always be obtained, especiaUy in case
the ore contains calcspar, which is converted bv roasting, in put
into caustic lime. The caustic lime not only reduces silver chloride
to metallic silver, but also greatly diminishes the R>lubility of moet
silver oompoundis in hypo solutions. Finally, though Bodium or
calcium hypo solution cussolves (besides silver chloride) silver anti-
moniate and arseniate, and, more or less, metallic or native silver, it
does not attack at all either silver sulphide or silver glance, or tk
group of silver minerals known as antimonial and arsenical sulphides,
like polybasite, stephanite, ruby silver, and fahl ore. But these irc
attacked by BusseU's sodio-cuprous hypo. The character of the ore
determines the strength and temperature as well as the order in which
the two hypos — the simple sodium or calcium hypo, called ** ordinarr
solution,** and the compound or cuprous hypo, called '* extra solution"
— are used. The compound hypo is made as required for each charge
of ore by dissolving 1 lb. bluestone (copper sulphate) to 2 IK
crystallised sodium thiosulphate in water. Further, it is loiown tiiat
cailstic alkalies in leaching solutions render insoluble certain con-
pounds of silver, and these are apt to occur in all calcareous or«,
oesides which, if sodium hypo is used, part of it will be converts!
into caustic soda. The remedy is an acidulated wash before leadunf .
Bussell uses sulphuric acid when his solution contains caustic sotk
If an ore containing much antimony or arsenic has been roasted so as
to have formed antimoniate, &a, acidulation may be dispensed with,
as these compounds are more soluble in caustic solutions. Carbonat&i
alkali is said to be hnrmless. Neutralising caustic by addition f
soda bicarbonate may hinder recovery of lei^. Boasted ore is ooc^i
before leaching, moistened to lay the dust, and sifted if lumpy ; !«•
it should not be quenched while red hot, or some of the chloride mar
be reduced to the metallic state by generation of steam.

Bussell's method of applying the ooprous hypo is to cause it t
pass through the ore in the vat several times, by means of an eject*?,
after which the liquid is passed to the precipitating tubs, and tiie ^rt
is washed, first with sodium hypo, in some cases hot, and then, ^
usual, with water ; there are cases, however, in which the cup^of^
h} po is allowed to stand on the ore during 12 hours. The cuprr&i^
hypo is compatible with the Kiss process. Bussell discovered tb:
while either hypo extracts the lead from lead sulphate in the roaetK
ore, they do not dissolve lead carbonate ; consequently, sodium car
bouate precipitates the lead from such a solution, leaving silver *£'
copper dissolved. The same discovery was made by Aaron tc:
Befl^sley, in 1882. But sodium carbonate would also preciptit:
lime from a calcium hypo ; hence the necessity for using the Faten
process when this pai-t of BusselFs process is to be employed : althoo^.
even when sodium hypo is used, it sometimes happens that Hnie a
present (dissolved from the ore), and then it must r-nntwimnatrr tti

Digitized by



lead carbcmate; in &ct, the lead prooees is not applicable in such
cases. This part of the Bussell process necessitates an extra set of
precipitating tnbs, in which the lead carbonate is allowed to settle,
after which the liqnid is transferred to other tubs in which the sUver
(and copper) is precipitated in the usual manner. It is claimed that
the lead carbonate obtained in this way is nearly free from silver, and
otherwise pure in the absence of lime in the ore.

In working the Kiss process, the lead may be thrown down by
milk of lime, as hydroxide, and this is done at the Mount Cory Mill,
but the product seems to be very impure and to contain much silver.
1 1 is contended by advocates of the Kids process that the precipitated
sulphides of silver, ^., settle much better than when sodium poly-
sulphide is used. It is probable that the precipitate might settle
badly in an alkaline solution, and the sodium hypo seems much more
liable to become strongly alkaline than the calcium hypo, because of
the much greater solubility of caustic soda than of caustic lime, and
of the complete insolubility of calcium carbonate. It is not easy to
make sodium polysulphide quite free &om caustic ; and, where the
precipitant is caustic, the hypo becomes so necessarily, unless means
are taken to counteract that effect. It may be that some of the
difficulty experienced in the Patera process was due to excessive
alkalinity of the solution. Certainly the less liability of the calcium
hypo to becoming strongly alkaline is a point in its favour. The
choice of a hypo for praotioil work resolves itself into the choice of a
polysulphide. Sodium hypo is easily procured, and one of the
strong^ arguments in favour of the Patera process is that the
sodium polysulphide is made with the expenditure of vastly less time,
labour, and fuel than the other. Also, in the use of calcium sulphide,
about 8 times as much sulphur is consumed in recovering a eiven
quantity of silver as when the sodium sulphide is used. Tne dioice
must depend mainly on tiie price and purity of the substances used,
but the many advantages of sodium sulphide will compensate a con-
siderable difference (to its disadvantage) between the price of caustic
soda and that of lime.* The hypo can be used over again, indefinitely
almost, as, though it is decomposed in dissolving (and decomposing^
silver chloride, it is reproduced in the precipitation of the dissolved
silver by the polysulphide. Without this result the hypo would be
weakened every time it was used, a portion of it being converted into
silver hyxx); but when the silver is removed by combination with
sulphur from the sulphide, the sodium or calcium from the sulphide
takes its place, thus regenerating the original hypo.

A brief account t of the practical operation and plant of the
Bussell process will sufficiently illustrate the conduct of all lixiviation
methods. All crushing (raw or roasted ore) should be dry. Fineness
has varied from a maximtmi of 8-mesh on raw ores to a minimum of
150-me8h in tailings ; experiment must determine in each case, and the
maximum size be adopted, for economy sake in crushing and facility

• O. H. Aaron, op. cit

t For most minute details see paper by Ellsworth Daggett, **The Bossell
Process In its Pmotical Application and Economic Besultd,^ Trans. Amcr. lusL
lliii. Engs., xvi 362.

Digitized by



of leaching. All oharges mtist be aoourately weighed. Bate of
leaching is not diminished by increase in depth of charge in tanks ;
deep charges are most economical, nsing less water and less salt in
solution, as well as less labour. The consumption of hypo will vazy
between about Ij^ lb. a ton for alkaline and arsenical ^es^ and ^7
lb. for acid ores ; of bluestone^ from %\ lb. a ton raw to l\ lb. roasted,
and from 4J^ lb. a ton alkaline to 6f lb. acid ; of sulphuric aotd, 1-2 lb.
a ton. The use of soda ash as a precipitant for lend saves ohemicak
in precipitating the precious metals, produces purer bullion, and
affords lead carbonate as a marketable bye-product. The first wash-
water is admitted to the tank either above the ore ^if little silyor is
extracted by it, or if a larger extraction is not objectionableX or below
it (if water is scarce, or if silver exti:^tion is to be kept down), the
consumption being 25-40 per cent, less in the latter way ; the average
is about 40 cub. ft per ton of ore. The application of water to roai^ed
ore above 150^ F. results in solution of about 3 times as much silTer
as when the ore is below 120^ F. ; as the wash-water runs to waste, it
is highly essential to secure perfect precipitation of all silver it may
contain—the other solutions only circulate. This precipitation is
often accomplished by iron and addulation (1-2 lb. acid per too),
heating the water first to about 175^ F. ; it is thorough, but ooOT|Hes
12 hours ; sodium sulphide acts more quickly but makes a bulky pre-
cipitate ; dilution with water is simple but uncertain. The seocmd
wash' water is only to restore the volume of stock solution ; it averages
about 5^ cub. ft. per ton. The stock solution is usually niade up with
\\ per cent, hypo (94 lb. per 100 cub. ft. water), the amount required
for 100 tons (3500 cub. ft.) being 3281 lb. ; for 50 tons (2000 cub. ft),
1875 lb. ; for 25 tons (1500 cub. ft.), 1406 lb. The strength can
generally be reduced 25-50 per cent, after some experience with the
ore. The strength of the cuprous solution varies from * 7 to 1 * 1 p^
cent, bluestone, and from 1 *5 to 2* 3 per cent. hypo. The treatment
of roasted alkaline ores only needs 4*5 lb. bluestone and 2*9 lb. hypo,

while acid ores take 6*1 lb. Uue-
stone and 4*9 lb. hypo on the





The leaching and storage
tanks (Fig. 172) are of wood,
with straight sides ; 3-in. staves
a, dressed to sweep of tank, and
long enough to allow a 6-in. diime
h ; gaining e 1 in. deep, done bj
hand; bottom planks d 3 in.,
Fio. 172.— LsAcmxo Tahk Dstaua grooved and joined with tightly-
fitting tongue 6, I by 1^ im
plugged with white lead; no nails cr screws permissible; painted
inside and out; must be tight enough to withstand not on!/
weight of charge, but also pressure of ejectors used to hasten leaching,
and of heat applied to wash-waters. False bottoms are wooden
slats /, 1^ in. high, 1 in. wide, and 1 in. apart, fastened by screws
embedded in thick white lead, and leaving an annular space If ii^
wide all round the tank, which is partially occupied \^ a ring

Digitized by



if bent wood g^ 1 in. wide, and the balance by a J-in. tope A, which
lecnres the filter cloth t in place. The latter is of No. 8 canvas duck,
nt 6 in. wider than the inside diameter of the tank, to allow for
iredging down by the rope, and lies immediately on a sheet of stiff
)iioonnt matting I;. The outlet is in the centre, of the tank, and
xmsists of a threaded cast-iron flange bolted to the bottom, all bolts
ind heads being embedded in white lead.

The working cost of mnning a Bnssell plant on 100 tons per 24
Krars of pulverised and roasted ore is given by Daggett as follows :—

Per diem. Per ton.

(a) Hazimnm. — £ «. d. «. d

Labour, 13 men at 12f 7 16 17

Fuel, 3^ cords wood at 24a 4 4 10

ChemioaU 20 4

Repairs, 902. a month 3 7

Asnyiog 1 12 4

Total 86 12 7 4

(6) Minhnmn. —

Labour, 11 men at 12«. 6 12 . 1 5

Fuel, 2) cords wood at 16« 2 5

Chemicals 11 4 2 4

Bepairs, 30Z. a month 10 2|

Assaying 18 3|

Total 22 4 4 8

Yarions actual costs per ton of ore are given as below : —

Cnsihniriaobic^ Mexico: including crashing, roasting, re-

fining sulphides, &0. 48<. 4<2.

Parral, Mexico : on roasted ore at 10 tons a day 36«. 7d.

Do. tailings, 10 tons a day 10s.

Silver Reef^ Utah : raw tailings, 40 tons a day .. .. 4. 6«. 7d.

Do. raw ore do. .. .. .. 120.-16«.

Lake Valley, New Mexico : total expenses, 60 tons a day .. 18«. 7d.

Blue Bird, Butte, Montana: from May 24 to December 8, 1893,
seated 15,797 tons (dry) of tailings, the cost being : — Hauling from
)it8 to tanks, 1«. 6^. ; labour, 1«. 5^. ; chemicals, 1«. lOd. ; pumping
^ter, 3d. ; superintendenca and miscellaneous, 1«. 4(1. ; total, 6«. 1(1.
The plant was designed for 60 tons roasted ore daily, and treats
100 tons tailings; cost 4000Z. to erect; tanks \1\ ft. diam., 9J^ ft.
leep, charge 70 tons (dry).

Broken Hill, New South Wales : concentrating mill recovers mo>t
>f the lead at a total cost of about 10«. Gd. a ton of crude ore ; tailings
)ontain much silver, as native, chloride, chloro-bromide, and iodide,
>ften enveloped in silica, which hinders action of leaching fluid ; |-1 j^
)er cent, sodium hypo used, with and without bluestone, giving, to
klay 31, 1892, 304,798 oz. silver from 40,794 tons, at a cost of about
»«. Id. a ton. Estimated that crushing, chloridising-roast, and
caching will amount to 20«. a ton.

Hofmann's proposal* to replace tanks by troughs for the lixivia-

^ Ottokar Hofmann, *' Trough-Lixiviatiou,'* Trani). Amer. Inst. Min. Engs., x?i.

Digitized by



tion with hypo solatioiis was based on the ohservaiion that speed of
extraction is in direct proportion to rapidity of moyement of Uie
•solvent through the ore. But the benefit thns arising appeals to be
neutralised by the stratification which ensues in the ^nks, a layer of
slimes forming on the top and resisting the leaching filtration ; beside
other objectionable features.

Commerce, — A few statistics concerning the cost of producing Bilker
(1891-92) may be interesting.

Small Hopes, Colorado : total mining cost per ton, 28«. lIcL; doj
per oz. silver sold in the ore, 28,

Alice, Montana : total mining cost per ton, 41«. 6d. ; milling, 28e.
grand total cost per ton mined, 69«. 6d. ; do. per oz. silver produced
4«. O^d.

Elkhom, Montana : total mining cost per ton, 42«. Sd, ; milling.
S7«. ; grand total cost per ton mined, 80«. lOd. ; do. per oz. siiv^
produ^, Is. 8Jd.

Granite Mountain, Montana: total mining cost per ton, 40r.;
hauling, lOd.; milling, S9«. M,; grand total cost pw ton mined
80«. Id. ; do. per oz. sUver produced, 28, Syt.

Daly, Utah : total mining cost per ton, 91«. 6<2. ; hauling, St. M,
milling, 27«. 7d, ; grand total cost per ton mined, 115». W. ; do. pei
oz. silver produced, 28, lOd,

Ontario, Utah : total mining cost per ton, 69«. fid. ; hauling, 2t. id.
milling, 35«. 9d. ; grand total cost per ton mined, 108«. Sd. ; do. pes
oz. silver produced, 28, 8^.

The world's production of silver in 1891, stated in kiloe (of 2 * 2 lb.
was estimated as follows : —

United States .. .. 1,814,642

Mexico 1,275,265

Bolivia 372,666

AnstndasiA 311,100

Germany 180,000

Pera 74,879

Ohili 72,185

France 71,117

Spain 51,502

Anfltro-Hongary 50,618

Central America 48,128

Japan 43,282

Colombia 31,232

Argentine 14,680

Bufflia 13,847

Canada 12,464

Great Britain 9,075

Italy 8,108

Norway .. .. 5,539

Sweden 4,180

Total 4,464,499

Digitized by




Until the introduction of Castner's process in 1887, the manufacture
)r sodium was oonducted as follows : — An intimate mixture of 30 parts
loda carbonate, 13 charcoal, and 7 lime is calcined at red heat, to
-ender the mass more compact, thereby also expelling much carbonic
»xide. The calcined mixture is then introduced into wrought-iron
ylinders of small diameter, aud heated to a temperature of about
!550^ F., whereby the alkaline metal is reduced and distilled from
he cylinder containing the charge, through a small tube provided
or the gases and vapours, into the receptacle known as the condenser.
Pbrough a variety of causes, not more than 40 per cent, of the metal
ontained in the charge is obtained, and in the manufacture of potas-
ium very much less. The wear and tear on the metal cylinders is
normous, and forms a large proportion of the cost of manufacture.
^o carry out this process and arrive even at these results, requires —
a) most careful grinding and mixing of ingredients ; (5) addition
f lime to prevent fusion ; (c) excess of carbon to ensure contact
•etween the particles of soda and carbon in the refractory charge ;
i) previous calcination to make the charge less bulky ; (e) wrought-
ron in constructing the cylinders is the only practical metal that
nil stand the high temperature; (/) cylinders must be of small
iameter, so as to allow the heat to penetrate to the centre of the
efractory charge ; (^g) exit tubes from cylinders to condensers require
lost careful attention to keep them open, owing to formation of a
lack compound by the action of carbonic oxide upon the vapour of
he alkaline metal, which combination takes place at about the con-
ensing point of the metallic vapour. This is one of the most serious
btttacles in manufacturing so£um, not only causing a large loss of
letal, but interfering generally with the operation. In making
otassium, the formation of this compound, which is exceedingly
xplosive, and which is produced even more readily than when making
xlium, is the chief reason that potassium costs almost 10 times as
inch as sodium. The approximate cost of producing sodium by this
lethod is 4«. a lb., the chief items being wear and tear, 2«. ; materials^
9. ; labour, 8i2. ; fuel, 4<{.

The reactions by which sodium and potassium are prepared in
astner's method* vary somewhat, but may be generally expressed
y the formula

6NaHO+FeOa=: 2Na,C03 + 6H + Fe + 2Na
6KH0 + FeCj = 2K,C0, + 6H + Fe + 2K.

In place of using an actual chemical compound of iron and carbon, as
(pressed in the above equation, a substitute or equivalent is prepared

♦ J. Mactear, •* The Caetner Sodium ProceBS," 80c, Chem. Ind., Mar. 7, 1887.

Digitized by VjOOQIC


as follows : — To a given quantity of melted pitch is added a definitB
proportion of iron in a fine state of division. The mixture is oookd,
broken up into lumps, and ooked in large crucibles, giving a m^allie
coke, consisting of carbon and iron, the proportions of each depending
upon tbe relative quantities of pitch and iron used. This metallio
coke, after being finely ground, provides a substance having the iron
and carbon in a like proportion to an iron carbide, and from whidi
neither the iron nor carbon can be separated by mechanical meuv .
The fine iron is conveniently prepared by passing carbonic oxide and
hydrogen, in a heated state, as obtained from an ordinary gat pro-
ducer, over a mass of oxide of iron, commercially known as ** purple
ore,** heated to a temperature of about 930° F.

In producing sodium, caustic soda of the highest obtainable
strength is used, and with it is mixed sufficient so-called ** carbide"
to furnish the proper amount of carbon to cany out the reaction. The
crucibles in which this mixture is treated are made of cast steel,
and are capable of containing a charge of 15 lb. of caustic soda,
together with the proper proportion of the " carbide.**

After charging a crucible with the above mixture, it is placed in
a small famace, where it is kept at a low heat for about 30 minutes,
during which time the mass fuses, boils violentlv, and a large part
of the hydrogen is expelled by the combined action of the iron and
carbon, the *' carbide," owing to its gravity, remaining in suspensiim
throoghout the fused soda. At the end of the time stated, the con-
tents of the crucible have subsided to a quiet fusion. The crucible

Online LibraryCharles George Warnford LockEconomic mining: a practical handbook for the miner, the metallurgist and ... → online text (page 70 of 76)