grain size was too large for use as wet ground mica the pool of the lagoon.

of the lagoon. Crude burnt lime fed at the

and too small for use as dry ground mica. The silica overflow of the cones effectively coagulates the dis-
sand tailings could not be transported profitably to persed slimes. The double diking method' of dam
market under prevailing freight rates. The mill as construction provides adequate pool area in the rela-
finally operated, therefore, is used solely to con- tively narrow and steep draws to permit winter
centrate spodumene. operation.
The plant constructed at Hill City, S. D., was put Insofar as possible, therefore, mill control has been
into operation early in 1952. It has operated success- made fully automatic so that operators may devote
fully without major change except that the classifier their full attention to pulp trends. Milling of peg-
in the grinding circuit has been replaced by Hummer matite is subject to even more intuitive control than
screens. The plant is comprised of standard metal- the mining operations, for added to the continual
lurgical equipment, but some of the equipment has variations in the mineral character and associations,
had to be modified to correct for the rapid pulp attention must be given to the dissolved mineral
settling rate and the extreme abrasiveness of the content of the pulp, pulp temperature, and pH as
pegmatite material. The generalized flowsheet is well as to changing mechanical and hydraulic con-
shown in Fig. 2. ditions related to nature of the feed.
Metallurgical control is based chiefly on analysis
Location of the plant in the heart of a National
by flame photometer, but grain estimates under the
Forest and in a widely publicized recreational area
microscope are also used. Microscope determination
makes it necessary that waste water returning to
applied to critical products in the circuit gives ade-
the drainage area be entirely free of contamination.
Mill waste is pumped some 200 ft vertically and quate information for proper adjustment of the
machines. Typical mill results on three ores are
2000 ft horizontally to dry draws where the solids
given in Table I.
are settled in normal tailing ponds equipped with
underdrains to bleed the clarified water from the References
1 L. R. Page et al.: Pegmatite Investigations 1942-1945.Black Hills.
back of the lagoons. The dams are constructed of S. D. Geological Survey Prof. Paper 241, 1953.
2 Fkemont F. Clarke et al.: Edison Spodumene Mine, Pennington
the tailing sands after dewatering in sand cones that County South Dakota. U. S. Bur. Mines R. I. 4234, 1948.
are located at the discharge end of the tailing pump :IS.A. Falconer and B. D. Crawford: Froth Flotation of Some
Non-sulphide Minerals of Strategic Importance. A I M E Trans., 1946,
lines and at the heads of the distributing launders. vol. 169.
V. A. Zanadvorf: Disposal of Mill Tailings Holden Concen-
The overflow liquid from the cones is directed to trator. A I M E Trans.. 1946. vol. 169.

Extraction of Lithium from I t s Ores

Lithium chemical plant extraction methods are discussed with reference to 1 )
base exchange with alkali sulphates; 2 ) processing based on roasting with lime; 3 )
miscellaneous methods; and 4 ) application of the Lithium Corp. process to extraction
of lithium from run-of-mine, low grade spodumene ore, or concentrates.

by Reuben B. Ellestad and Fremont F. Clarke

IandNthethe early days of the lithium industry most of

production was from lepidolite, zinnwaldite,
amblygonite. Nearly all the early extraction
and von Girsewalt. In these methods the finely
ground silicate ore (spodumene or lepidolite) is in-
timately mixed with an excess of alkali sulphate
processes described in the literature involve heating (usually K,SO,) in at least a 1 to 1 proportion, and
the finely ground mineral with sulphuric or hydro- the mixture was heated to a relatively high tem-
chloric acid. On subsequent water leaching most of perature. Base exchange results, with the forma-
the bases present in the mineral (especially alumi- tion of lithium sulphate. A water leach dissolves
num) are dissolved as sulphates. As a result, the the lithium sulphate, together with the excess
leach solution required extensive chemical purifica- potassium sulphate. Successful operation of this
tion before the lithium could be precipitated as car- type of process requires very thorough grinding and
bonate. Following the remarkable growth of the mixing, as well as careful temperature control. The
lithium industry to its present size, zinnwaldite and use of K,SO, is objectionable from cost consider-
amblygonite ores must be considered of minor im- ations since purification of lithium carbonate re-
portance only. Attention is now focused on spodu- quires the use of potassium carbonate, if the K,SO,
mene, abundant enough in North America to be a is to be recovered and recycled. The lower solubil-
major source of supply, and there are important ity of K,SO, as compared with Na2S04is also objec-
supplies of lepidolite and petalite in Africa. The tionable, since it limits the concentration of the
extraction processes described below all apply to Li2S0, solution to be precipitated by K,CO,. Early
spodumene, although several will also operate on laboratory-scale investigation of this process by
other lithium minerals, such as petalite. Lithium Corp. was not encouraging.
Base Exchange with Alkali Sulphates: A distinct Other related base exchange processes are those
advance was made with the disclosures of Wadman of Lindblad, Wallden, and Sivander3 and Sivander,
- - - -
Gard, Villestad, and Wallen'. The former covers the
R. B. ELLESTAD is Director of Research and F. F. CLARKE, Mem- reaction of lithium silicate minerals with a sodium
ber AIME, is Vice President and Production Manager, Lithium sulphate solution, at 100" to 300°C (under pres-
Corp. of America Inc., Minneapolis. sure), while the latter involves the extraction of
Discussion of this paper, TP 4079BH, may be sent (2 copies) to
silicate minerals with molten sodium sulphate. Both
AIME before Jan. 31, 1956. Manuscript, Feb. 9, 1955. Chicago
Meeting, February 1955.
these processes would seem to be difficult and ex-
pensive to operate.

TRANSACTIONS AlME NOVEMBER 1955, M I N I N G ENGINEERING1045

 Processes Based on Roasting with Lime: Several and LiCl recovered (96.5 pct recovery) by alcohol
processes have been described in which the finely extraction.
ground ore is mixed with either CaO or CaCO, and Lithium Corp. Process: The process used a t pres-
heated. On water leaching, LiOH is dissolved and ent by Lithium Corp. for the extraction of lithium
separated from the insoluble residue. from spodumene is that of Ellestad and Leute." The
The first of these is the patent of Rosett and first step in this method is the conversion of natural
Bichowsky," which covers mixing the ore with alpha-spodumene to beta-spodumene by heating in
CaCO, in a 1:2 ratio and heating to 800" to 1000°C a kiln to about 1100°C. After grinding, the beta-
until it is clinkered. Grinding and leaching follow. spodumene is mixed with an excess of 66" Be sul-
The process of Coltono is very similar. Next is the phuric acid and heated in a smaller kiln to about
patent of Nicholson' in which beta-spodumene* is 250°C, with a retention time of about 10 min at
This is natural, or alpha-spodumene, which has been heated to temperature. This results in a true exchange reac-
approximately 1100°C. tion, in which lithium ions in the beta-spodumene
autoclaved with Ca(OH), and water. An alternate structure are replaced by hydrogen ions, with the
procedure is also covered, in which alpha-spodu- formation of lithium sulphate, which is leached and
mene is mixed with CaCO,, in a 1:2, or 1:2.7 ratio, separated from the ore residue. Details of these op-
and heated to a temperature of 1000" to 1230°C, erations are given by Hader, Nielsen, and Herre.'"
followed by leaching. The process of StaufferQn- Some of the advantages of this process are:
volves heating a mixture of spodumene and CaO, 1) It is not necessary to mix the ore with solid
in a 1:3 ratio, in a vacuum of 0.01 mm at 1150°C, reagents. Adequate mixing with sulphuric acid is a
with the distillation and condensation of lithium much easier operation.
oxide, Li,O, which dissolves in water to form LiOH. 2) Extreme fine grinding of the ore is not neces-
Finally there is the process of Kroll-n which sary. This follows from the fact that due to the
ground spodumene is mixed with CaCO, in a 1:O.Z porous nature of beta-spodumene, a relatively large
ratio and fired at high temperature, followed by particle will effectively absorb sulphuric acid. This
leaching in an autoclave under pressure with vari- is in contrast to most of the other processes, which
ous salt solutions, such as Na,SO,. are based on solid phase reactions requiring very
Strictly speaking, the reactions involved in these fine grinding of both ore and reagent, as well as in-
lime processes are probably not simple base ex- timate mixing, if maximum recovery is to result.
change reactions, in which calcium replaces lithium. 3) Both the kiln operations used in this process
Except in the process of Stauffer, it is doubtful if require heating at temperature for only a relatively
free Li,O is present after the sintering operation. It short time. In the first kiln operation, in which
seems more likely that a lithium calcium silicate or alpha-spodumene is converted to beta-spodumene,
aluminosilicate is formed, which on hydrolysis the inversion is practically instantaneous a t llOO°C,
yields LiOH. In principle these methods are all at- and only a matter of minutes at 1075°C. Also, in
tractive, in that the lithium compound first obtained the actual extraction operation, only a relatively
on leaching is LiOH, which can be converted to low temperature of 250°C is needed, with retention
other salts without the necessity of going through time of only about 10 min at temperature. This is
the carbonate stage. in contrast to most of the processes described. One
Miscellaneous Processes: Several processes have advantage of this low reaction temperature is that
been proposed in which spodumene and sometimes fusion of lithium sulphate does not take place, thus
amblygonite is mixed with CaCOa and either CaSO, eliminating one of the troubles of the process using
or CaCl,, followed by heating. Sternberg, Hayes, a lime-gypsum roast.
and Williams'kix spodumene, CaCO,, and CaSO, 4 ) After the heating with sulphuric acid at
in the weight ratios 1:1:0.6, heat to 1100°C for 2 to 250°C, no grinding is needed prior to leaching with
3 hr, and leach out lithium sulphate. Similarly water, unlike most of the other processes.
Kalenowski and Runke" mix spodumene, or mix- 5) Water leaching of the acid roast ore is rapid
tures of spodumene and amblygonite, with Ca(OH), and is not affected by leaching with moderately
and gypsum (CaS04.2H,0) in the weight ratios of strong lithium sulphate solutions.
1:2:1, heat to 1050°C for 2 hr, and leach out lithium 6) It is the authors' opinion that proper utiliza-
sulphate. The process of Sternberg, Hayes, and tion of this process results in an efficiency of lithium
Williams was tried out on a plant scale by Lithium recovery not achieved by any of the other methods.
Corp. and found to be very difficult to operate. In 7) It is also their opinion that this process is the
the kiln heating operation, the lithium sulphate and only one that lends itself to efficient direct extrac-
the excess CaSO, present formed low melting mix- tion of lithium from low grade, run-of-mine ore.
tures which caused serious problems. Chemical Plants a t Minneapolis and at Bessemer
Fraas and Ralston" describe a method in which City, N. C.: The Lithium Corp. of America Inc.
spodumene is mixed with CaCO, and CaCl,, fol- chemical plant at Minneapolis was designed for the
lowed by heating in a kiln to a temperature suffi- use of spodumene concentrates. These are chiefly
ciently high to volatilize LiCl, which is condensed flotation concentrates, although coarse ore concen-
and collected. This process has been tried on a large trated by either hand-picking or sink-float methods
scale, but was not successful. has been used. During the past year a considerable
Another process is that of Kepfer and Pfanstiel" tonnage of petalite was also processed.
in which spodumene and CaC1, are mixed in a ratio The method of extraction used at Minneapolis is
of 3: 1 to 1: 1 and heated at 760" to 925°C for 2 hr, the sulphuric acid roast of beta-spodumene, follow-
followed by leaching out lithium chloride. ing the patent of Ellestad and Leute referred to
Finally there is the process of Erasmus" in which previously .'"
spodumene ore (1.2 pct Li) is mixed with CaCl, in In brief, the first step is the conversion of natural
the weight ratio of 4 : l and heated under reduced alpha-spodumene to beta-spodumene in a kiln op-
pressure (2.5 to 5 mm) at 1050" to 1150°C for 3 to 5 eration. Coarse ore is crushed to 1 to 2 in. size
hr. The volatilized reaction products are condensed before being fed into the kiln, whereas flotation

1046MINING ENGINEERING, NOVEMBER 1955 TRANSACTIONS AlME

concentrates are used directly. The kiln, 40 f t long Li?O, is about 85 pct. If this figure is coupled with
and 4 f t ID, is lined with fire brick and gas-fired a realistic figure of 60 pct for the recovery of lith-
countercurrent to the flow of ore, rotating a t about ium in flotation concentration, the result is an over-
1 rpm. The hot zone is at a temperature of 1050" to all efficiency of only 45 to 50 pct. In view of this,
1100°C. Capacity is 2 tph of concentrate feed. After when Lithium Corp. investigated the utilization of
cooling in a rotary cooler the ore is ground in a Wil- its North Carolina ore holdings, attention was given
liams mill. The ground ore is then mixed with 66" to the possibility of using its sulphuric acid process
Be sulphuric acid, using approximately 40 pct ex- directly on r u n of mine ore. Continuous large-scale
cess above the theoretical requirement. The ore is tests have shown that an overall recovery of 80 pct
then fed into a small unlined rotary kiln, 26x3 ft, is very feasible, in contrast with the 45 to 50 pct
which is gas-fired concurrent to the ore flow. Dis- overall recovery when preconcentration methods
charge temperature is 250°C. The lithium sulphate coupled with chemical plant processing are utilized.
from this operation is water-leached from the ore As a result, the chemical plant at Bessemer City,
residue in air-agitated Pachuca tanks of 8000-gal N. C., treats r u n of mine ore direct without precon-
capacity. Ground limestone is added to the ore slurry centration. Except for capacity, and the use of more
in the leach tanks to neutralize the excess sulphuric appropriate equipment for performing the unit
acid and to precipitate the small amount of soluble steps in the process, especially with regard to the
iron and aluminum salts present. The slurry is fil- larger amount of material involved, the process at
tered and washed in a vacuum drum filter. The Bessemer City is essentially the same as that used
solids are discarded, and the filtrate, containing at Minneapolis.
about 100 g per liter of lithium sulphate, is treated
for removal of calcium, magnesium, iron, and alu- References
minum. The solution is concentrated by evapora- 1 U. S. Patent 847,856;1906.
tion to about 200 g per liter of lithium sulphate. The a U. S. Patent 1,110,556;1929.
8 U. S . Patent 2.331.R38: 1943.
primary product of lithium carbonate is produced patent 2;23o:i671 i941.
Patent 2,020,854;1935.
by reaction of this solution with soda-ash. Owing Patent 2,021,987; 1935.
to the appreciable solubility of lithium carbonate in Patent 2,413,644; 1946.
Patent 2.424.512: 1947.
the mother liquor, it is necessary to remove sodium Patent 2:622:809: 1953.
Bureau o f rciznes R.I.3848. 1946. Patent 2,533,246.
sulphate from the lithium carbonate filtrate and re-
turn the solution to the tank room system.
The lithium recovery of the above process, when
operating on spodumene concentrates of 4 to 5 pct Is Industrial and' Engineering Chemistry. vol. 43, p. 2636.

Errors in Underground A i r Measurements

Stefan Boshkov and Malcolm T. Wane

T HE validity and accuracy of velocity measure-

ments underground have been questioned re-
peatedly by those in mine ventilation work. The
general disagreement on the subject is well illus-
trated in an AIME publication.' Although the pres-
ence of errors is readily admitted, their magnitudes
are not known and often are described by practical
operators merely as "great," "small," or "of no im-
portance." The belief that errors are of academic
interest only prevails because of a justifiable claim
that actual volumes of air circulated are well in
excess of those required by law and safe mining
practices.
Poor understanding of the source and nature of
errors inherent in velocity measurements, coupled
with questionable determination of cross-sectional
area, result in faulty techniques for the purpose of
expediting ventilation surveys and render doubtful
the calculated volume. Often consistency of read-
ings is confused with accuracy. A succession of check Experimental wind tunnel for laboratory testing is shown above.
readings showing a maximum deviation of 5 pct in
the measured value of velocity usually signifies This article discusses the errors involved in air
nothing more than diligent application of technique measurements underground. Where instrument
in a stable ventilating current. errors are presented, the treatment is confined solely
to the rotating-vane anemometer, recognized as a
5. BOSHKOV, Member AIME, is Assistant Professor of Mining most useful ventilation measurement tool because
Engineering and M . T. W A N E is Associate in Mining, School of of its simple design and easy operation. In the fol-
Mines, Columbia University, N . Y. C. lowing pages a positive deviation is defined as one
Discussion of this paper, TP 4124A, may be sent (2 copies) t o which has to be added to the registered air velocity
A I M E before Jan. 31, 1956. to obtain the true air velocity.

TRANSACTIONS A l M E NOVEMBER 1955, M I N I N G ENGINEERING-1047