Robert A. Taft Sanitary Engineering Center
TECHNICAL REPORT W62-17
PROCESS AND WASTE
CHARACTERISTICS
AT SELECTED URANIUM MILLS
•V
U. S. DEPARTMENT OF HEALTH
EDUCATION, AND WELFARE
Public Health Service
-------
SEC TR W62-17
PROCESS AND WASTE CHARACTERISTICS
AT SELECTED URANIUM MILLS
Prepared by
Radiological Pollution Activities Unit
Field Operations Section
Technical Services Branch
U. S. Department of Health, Education, and Welfare
Public Health Service
Division of Water Supply and Pollution Control
Robert A. Taft Sanitary Engineering Center
Cincinnati, Ohio
1962
-------
CENTER PUBLICATIONS
The Robert A. Taft Sanitary Engineering Center is a national
laboratory of the Public Health Service for research, training.
and technical consultation in problems of water and waste treat-
ment, milk and food safety, air pollution control, and radiologi-
cal health. Its technical reports and papers are available without
charge to professional users in government, education, and
industry. Lists of publications in selected fields may be obtained
on request to the Director, Robert A. Taft Sanitary Engineering
Center, Public Health Service, Cincinnati 26. Ohio
-------
CONTENTS
Page
FOREWORD v
THE RESIN-IN-PULP URANIUM EXTRACTION PROCESS . 1
Mines Development Company,
Edgemont, South Dakota 1
THE ACID LEACH-SOLVENT EXTRACTION URANIUM
REFINING PROCESS 19
I. Gunnison Mining Company,
Gunnison. Colorado 19
II. Climax Uranium Company.
Grand Junction, Colorado • 37
THE CARBONATE LEACH URANIUM EXTRACTION
PROCESS 55
I. Homestake-New Mexico Partners Company,
Grants. New Mexico 55
II. Homestake-Sapin Partners Company,
Grants. New Mexico 73
BIBLIOGRAPHY 93
in
-------
FOREWORD
This report contains the findings of detailed studies of process
and waste flows at five typical uranium mills. The studies were
initiated in 1957 by the Public Health Service for the purpose of
characterizing the liquid and solid wastes resulting from uranium
milling processes.
Uranium mills extract the naturally radioactive uranium from
its ores and produce a concentrated product that is subsequently
refined elsewhere for use in nuclear weapons and reactors. The
extraction and recovery processes employed are determined by
the character of the ore and its uranium content. The five mills
reported on here typify the processes normally encountered, i.e. ,
acid or alkaline leaching of the ore, concentration and purifica-
tion of the leach liquor by ion exchange or solvent extraction, and
chemical precipitation of the dissolved uranium.
Although the radioactive waste materials, especially Radium-
226. were of primary interest in these studies, useful information
regarding the chemical characteristics of milling wastes was also
obtained. The entire body of information thus developed forms an
excellent basis upon which to characterize the waste products
from the industry as a whole. This has resulted in an "Industrial
Waste Guide for the Uranium Milling Industry", which is pub-
lished as a separate SEC Technical Report.
The generous cooperation and assistance of many individuals
and agencies have contributed greatly to the successful comple-
tion of these studies. The work was supported in part by funds
made available through the Environmental and Sanitary Engineer-
ing Branch, Division of Reactor Development. U. S. Atomic
Energy Commission.
E.G. Tsivoglou. In Charge
Radiological Pollution Activities
Field Operations Section
Technical Services Branch
Division of Water Supply and Pollution Control
-------
THE RESIN-IN-PULP URANIUM EXTRACTION PROCESS.
MINES DEVELOPMENT COMPANY, EDGEMONT,
SOUTH DAKOTA.
E. C. Tsivoglou *
D. C. Kalda
J. R. Dearwater
Introduction and Background
During the summer of 1957, the Public Health Service car-
ried out a field study of liquid wastes resulting from the extrac-
tion of uranium from its ore in a typical refinery using the resin-
in-pulp extraction process. The study was performed in coopera-
tion with the South Dakota State Department of Health, the Mines
Development Company of Edgemont, South Dakota, and the United
States Atomic Energy Commission. It was the first of a series of
such surveys by the Public Health Service to develop detailed
knowledge of the characteristics of wastes, particularly radio-
actix'e wastes arising from the extraction of uranium from its
ores. Specific objectives of the studies include detailed analysis
of the extraction process, characterization of the resulting liquid
wastes, evaluation of their water pollution and public health
significance, and development of adequate and suitable waste
control measures. At the same time, parallel field studies of the
fate of these wastes in the water environment were carried out. *•
The uranium refinery at Edgemont, South Dakota, is located
on the banks of the Cheyenne River about 35 river miles above
Angustora Reservoir, a recreational lake (see Figure 1). This
refinery of intermediate capacity for ore processing, is a typical
example of the acid leach-resin-in-pulp process. At the time of
the field survey it was processing slightly more than 500 tons per
day of ore that assayed about 0.20 per cent U.,08. The mill had
been in operation about 1 year prior to this study. Virtually all
liquid wastes were delivered to tailings ponds for storage and
volume reduction by evaporation and seepage. There was a small
direct liquid discharge to Cottonwood Creek, a tributary of the
Cheyenne.
The Cheyenne River near Edgemont, South Dakota, is a re-
latively shallow stream with a sandy bottom. Flow is often tur-
bid, and at the time of the field survey biological life was rela-
tively sparse. River flows have been recorded from 1928 to 1933
and from 1947 to the present by the U. S. Geological Survey at a
*Respectively, Chief, Radiological Pollution Activities Unit,
Division of Water Supply and Pollution Control, Robert A. Taft
Sanitary Engineering Center, Cincinnati, Ohio: Chief, Water
Pollution Section, South Dakota State Department of Health; and
Senior Assistant Sanitary Engineer, Radiological Pollution
Activities Unit, (deceased)
-------
RESIN-IN-PULP PROCESS
Figure 1. Radioactivity monitoring stations, Cheyenne River, 1956.
gaging station located just upstream from Cottonwood Creek.
There are some 7, 143 square miles of drainage area above this
gage. From 1947 to 1955 the average annual flow in the river at
Edgemont ranged from 20.5 to 144 cubic feet per second (cfs); the
minimum monthly average flow for this 9 year period ranged as
low as 0.02 cfs and as high as 1.6 cfs. The flow drops to zero
every year for varying periods; in 1952 there were only 6 days of
zero flow, while in 1950 there were 62 days. Over the 9-year
period there were 296 days of zero flow; in other terms, the
records indicate that on the average the river flow at Edgemont
has been zero for 8.1 per cent of the days.
During February 1956. before the mill went into operation,
a radioactivity background survey of the Cheyenne River below
Edgemont was made by the South Dakota State Department of Health
and the Public Health Service. At that time samples of river water,
mud or sand, and aquatic life were collected at four locations and
analyzed for gross alpha and beta radioactivity. The sampling
stations are shown in Fibure 1. Biological samples included
plankton, algae, insects, and minnows. The water samples con-
tained 10 to 40 micromicrocuries per liter (jt/ic/1) of dissolved
alpha activity and 10 to 120/i^c/l of dissolved beta activity. Sus-
pended radioactivity was practically nil. River mud samples con-
tained an average of 10 and 15 micromicrocuries per gram (u/tc/g)
of dry solids of alpha and beta activity respectively, with no ob-
-------
MINES DEVELOPMENT COMPANY 3
servable variation between sampling stations. The biological
samples showed correspondingly low concentrations of gross
radioactivity. Ground water samples had dissolved alpha and beta
activity in the same range as was indicated by the Cheyenne River
water samples.
In 1957, a sample of slimes tails before entering the slimes
tailings pond yielded gross alpha and beta activities of 590, 000
and 780, QQQft/ic/1, respectively. Virtually all of this activity was
in suspended solids, and the dissolved activities were 980 and 930
/i/ic/l, respectively. A liquid sample from the slimes pond indi-
cated similar dissolved activities, with only slight suspended
radioactivity. Subsequent water samples (June 1957) indicated no
significant radioactivity above background in the Cheyenne River
or in Cottonwood Creek, although the small direct drainage from
the sands pond to this creek contained 1, 400 and 1, 800/t/ic/l,
respectively, of dissolved gross alpha and beta radioactivity. No
radium analyses were made prior to the mill survey of July 1957.
In terms of radioactive waste disposal, radium is the most
hazardous radioelement involved in the extraction of uranium from
its ores. Only the uranium is wanted, and all of its radioactive
daughters, including radium, are disposed of as waste products.
Of all of these decay products, radium has by far the lowest
maximum permissible concentration in water.2 Hence, the
amount involved, as well as the course of its passage through the
extraction process, is of considerable interest. Of special con-
cern is the question of how much radium becomes dissolved in
the processing of ore, and where this dissolved portion goes. One
of the prime aims of this study, therefore, was to perform a
radium balance through the mill, and in so doing to answer the
foregoing questions.
The Mill Process 3
Briefly, ore received at the mill is crushed and ground, and
leached with sulfuric acid to dissolve the uranium. The coarse
sands are separated and discharged to waste, and the remaining
slurry, or IX feed, containing the slimes or fine solids, proceeds
to ion exchange resin banks. Here the uranium is extracted from
the feed solution by resin beads, and is in turn stripped from the
beads, precipitated, filtered and dried to form yellowcake, which
is the final product of the process. This uranium concentrate is
then shipped to other facilities for further refining and process-
ing. The IX feed solution, having been stripped of its uranium,
becomes the slimes tails, and is neutralized with lime and sent
to the slimes pond. Figure 2 is a schematic flow diagram of the
entire process.
-------
4 RESIN-IN-PULP PROCESS
SAMPLING
Trucks are used to dump the uranium ore into either of two
ore bins from the ramp shown at the far left in Figure 2. The ore
is fed to the sampling plant from these bins by a conveyor belt. A
magnetic iron separator removes metal scraps, nuts, bolts, etc.
In the sampling plant the ore passes over a 1-1/2 inch vibrating
grizzly, then goes to a 1-1/2-inch jaw crusher, and falls on a con-
veyor belt. Ten per cent of the ore is removed from the belt by
calibrated rotating buckets and 90 per cent goes directly to the
mill. The 10 per cent passes over a 3/4-inch screen, goes to a
3/4-inch jaw crusher, and drops to a conveyor belt. Ten per cent
of this (one per cent of the original ore) is removed by calibrated
rotating buckets, and the remainder leaves for the mill. The
retained material (1 per cent) goes to a 1/4- inch screen and a
1/4-inch jaw crusher, and falls to a conveyor belt from which a
Vezin sampler removes 10 per cent. The remainder leaves for
the mill. The sample amounts to 0.1 per cent of the original ore
feed, or two pounds of ore per ton of ore fed. This representa-
tive sample is assayed for its UgOg content. After leaving the
sampling plant the crushed ore goes to either of two 250-ton ore
bins or to a 50-ton truck bin from which specification material
can be taken for blending.
GRINDING
The mill bins each have two bottom discharge hoppers. One
means of blending control consists of regulation of the feed rates
from these four chutes. Another method of blending involves
scheduling of various shippers lots through the mill. Factors
considered for primary blending purposes include the grade of
ore, particle size of sand grains, slimes content, and oxidizing
or reducing characteristics of the ore.
The grinding facility consists of a 4- by 8- foot rod mill in
open circuit, with a 42-inch spiral classifier. The feed to this
grinding section is very fine, not only because the sandstone ores
are poorly cemented but also because crushing and repeated
handling before the mill bins cause a large portion of the ore to
be broken down into individual grains. The resulting feed is eas-
ily ground to minus-12-mesh. Water is added at the spiral classi-
fier to make a slurry. Water for the mill is obtained from an
artesian well at a natural temperature of 53 °C.
LEACHING
The slurry from the grinding plant is pumped to a series of
four wood-stave tanks, each measuring 14 feet in diameter and
14 feet high, where leaching is carried out. Each tank is equip-
ped with a 48-inch rubber covered propellor mounted on a 4-1/2-
inch diameter rubber covered shaft; these tanks are operated in
series.
-------
SAIiPLIW AND CRUSHING
GRINDING
AND
LEACHING
SANOS-SLIMES SEPARATION
SCLuTICN
cj 3AN«S
TANK
LEGEND:
II - ION EXCHANGE
P - PREO»A»T LI l
3 - BANK
T - TAILS
S - SUSGf
E - EATING
• - RESIN
FCV - FLO» CONTROL VALVE
HIP - KESIN M
Figure 2. Flow diagram of resin-in-pulp process. Mines Development,
Inc., Edgemont, South Dakota, July 1957.
3FO q13—173—3
-------
MINES DEVELOPMENT COMPANY 5
Most of the ore received at this mill is easily leached. Most
of the uranium is in an oxidized state, and hydrocarbons or other
reducing elements are not present. Excessive sliming is not a
problem, because the ore is loosely cemented and contains very
little clay, bentonite. etc.
Sulfuric acid is automatically metered to the first two leach
tanks from a 30,000-gallon acid storage tank in quantity sufficient
to maintain pH from 0.9 to 1.4, depending upon the type of ore be-
ing processed. Two continuously recording pH meters with calo-
mel electrodes are used at the leach tanks and are coupled to
controllers that automatically regulate the acid feed. Leaching is
carried out at about 40°C, with no attempt to regulate heat in the
system. The retention time in each leach tank is about 3 hours
and the pH of the pulp overflow from the fourth tanks is usually
under 1.5.
SAND-SLIME SEPARATION
The slurry, or pulp, from leaching, with the uranium in solu-
tion, undergoes sand-slime separation in five 30-inch by 17-1/2-
foot spiral classifiers and two 10-inch cyclones. Three 2-inch
vertical sand pumps and two 3-inch pumps are used in this sec-
tion. Sands advance from the first to the fourth classifier, and
from the second cyclone to the fifth classifier, are washed, and
pumped to the sands tailings pond. To facilitate pumping, fresh
water is added at the pump. Water recycled from the sands tail-
ings pond isintroduced at the second and fifth classifiers to wash
the leached sands and to reduce pulp density for a sharper sepa-
ration in the cyclone. To minimize losses of uranium-bearing
solution, the spiral classifiers are set at a relatively steep slope,
which results in a small pool surface and long drainage deck.
The overflow from the first cyclone contains 5 to 10 per cent
solids that are minus-300-mesh in size. After screening to re-
move wood chips and other trash, this overflow goes to a 21- by
21-foot wooden tank for storage ahead of the ion exchange circuit.
Powdered iron is usually added at this tank to adjust the solution
EMF to 400. in order to keep vanadium in its tetravalent state and
thus prevent poisoning of the resin during ion exchange. The ion
exchange feed has a pulp density of about 1.05; entrained solids
are minus-300-mesh, pH is 1.7 to 1.9, EMF is 400, and the con-
centration of UoOfl in solution is about 1.0 gram per liter (1,000
parts per million). This pulp, with a small amount of recycle IX
feed from the banks, is pumped to a small elevated constant head
tank, from which it is metered by means of a weir to the distri-
butor for the resin tanks.
ION EXCHANGE
Basically, the uranium is absorbed from the IX feed solution
on anion exchange beads. The beads are then stripped of their
uranium bv an acidifier nitrate solution, and the uranium is sub-
-------
6 RESIN-IN-PULP PROCESS
sequently precipitated from the nitrate solution.
The resin-in-pulp (RIP) section consists of 14 rubber-lined
steel tanks, or banks, each containing two stainless steel baskets
which are 4-1/2 by 4-1/2 feet in cross section and 5 feet high and
have 30-mesh openings. The baskets hold a 10-inch bed of 20-
mesh resin beads. The bank is filled with pulp slurry, and the
motor-driven baskets oscillate up and down in the tank at 6.3 cy-
cles per minute, to ensure good contact of bead surfaces with the
pulp.
The distribution of solutions in the RIP section is accom-
plished by an ingenious device that eliminates all need for expen-
size valving in this section. The central unit is a distributor
wheel through which IX feed solution and eluate enter the banks.
The wheel is divided into 14 compartments, each connected by
pipe to one bank. A pH indicator and recorder continuously gives
pH of the IX feed and eluting solution, and meters indicate and
control the flows of these solutions. Full flexibility of operation
is obtained with the system, a detailed description of which is
given by Dayton. ^
The banks are alternated between an adsorption (loading) cy-
cle, during which bead surfaces are loaded with uranium from the
pump, and an elution (stripping) cycle, during which the uranium
is stripped from the beads by a stripping solution. In normal op-
eration. 7 banks are on adsorption, 5 are on the elution cycle, one
between these two cycles is being washed, and one bank is on
standby.
The countercurrent principle is used during loading and strip-
ping. The uranium-rich IX feed solution enters the bank contain-
ing the most completely loaded resin: fresh eluate is added to the
bank where the beads have the least amount of absorbed uranium.
Pregnant eluate is taken from the bank where the beads are most
heavily loaded with uranium, and the stripped IX feed leaves from
the bank containing the least loaded resin. The bank that is taken
off the loading cycle is the next bank placed in the circuit at the
end of the stripping cycle. Solution from the seventh bank on the
adsorption cycle is sent to the slimes tails neutralization tank.
Control is obtained by quick fluorimetric uranium analysis of
samples taken at the first and second banks from the discharge
ends of the adsorption and elution cycles. For instance, the lead
bank is taken off adsorption and another bank is added at the end
of the adsorption cycle when the uranium concentration in the IX
effluent to tails builds up to a predetermined level. Otherwise
uranium would be lost to the tailings.
The banks are drained by means of bottom discharge ports
fitted with a hollow vertical plunger. The solution level in the
banks is controlled as desired via large funnels connected by
means o£ flexible rubber hose to the hollow plunger. Excess
-------
MINES DEVELOPMENT COMPANY 7
solution is thus drained to the bank pumps. Control of the solu-
tion level provides smooth continuous flow from bank to bank, and
minimizes surging and overloading of any of the banks.
The beads must be hosed intermittently to keep them wet and
promote drainage, once a bank has been drained of solution. The
beads tend to swell and stick together when dry, but separate
readily if kept wet. The bank taken off the adsorption cycle is
washed to remove slimes adhering to the beads, in order not to
contaminate the eluting solution. After elution. the beads are
washed to remove excess nitrate. Wash water is kept to a min-
imum, and is recycled to the IX feed storage tank.
Eluting solution is made up in three wooden tanks. At any
time, one tank is delivering fresh eluate to the RIP banks, one is
receiving filtrate from yellowcake filtration, and the third, full of
filtrate, is being adjusted with acid and nitrate. Fresh eluate
entering the RIP section has about 56 grams per liter (g/'l) of ni-
trate ion and is acidified with sulfuric acid to a pH of 1.2.
Considerable purification and concentration of uranium re-
sults in the RIP circuit. The anion resin beads extract in the
neighborhood of 99.7 per cent of the uranium in solution, but only
a small fraction of the dissolved iron, vanadium and aluminum.
After elution, the pregnant eluate sent to precipitation assays 10
to 20 times the uranium assay of the IX feed solution, and contains
10 to 12 g/1 of uranium, expressed as U~0n. This uranium-rich
n. expressed as U.,0q. This
a pregnant eluate holaing ta
solution is pumped to a pregnant eluate holding tank.
URANIUM EXTRACTION
The pregnant eluate is next filtered and clarified, before
uranium precipitation, in order to obtain a clean concentrate un-
contaminated by slimes solids carried into the pregnant eluate.
Sufficient milk of lime is added to bring the pH up to 3.5 with
subsequent precipitation of calcium sulfate, or whitecake. which
is returned to the IX feed tank. The pregnant eluate is then clari-
fied by a 38-frame plate and frame filter press. This prelimi-
nary filtration also controls filtrate buildup. Sulfates are con-
trolled also by bleeding off about 10 per cent of the yellowcake
filtrate to slimes tails.
The clarified pregnant eluate is next cycled through a yellow-
cake dust collector, which strips it of dust from the yellowcake
dryers. It then proceeds to one of two 12- by 14-foot precipita-
tion tanks, where magnesium oxide is added in dry form to pro-
duce a diuranate precipitate. Enough MgO is added to produce a
solution pH of 6.8, The magnesium oxide is rather slow-acting
but produce a large floe as compared to other precipitation agents.
Precipitation required from 4 to 10 hours.
All of the chemical reactions involved are indicated on Figure
2, as well as in the process described by Dayton.
-------
8 RESIN-IN-PULP PROCESS
The precipitated slurry is pumped to one of two frame filter
presses for yellowcake filtration. A layer of filter paper and a
layer of nylon filter fabric are used on the press frames. After
filtration the cake is washed with water and given an air blow. Be-
low the filter presses are removable drip pans, which are instal-
led over a paddle re-pulper that extends the full length of the fil-
ter. The drip pan is removed and the filter opened; the precipi-
tate is scraped off the frame and drops to the re-pulper, which
keeps the cake fluid while it is delivered to an agitator. The con-
centrate that forms on the drum is scraped off and drops to a
hopper, from which it is drummed for shipment.
The drum dryer operates under a slight vacuum, and the ex-
haust is pulled through a dust separator fitted with ceramic baf-
fles. Clarified eluate is introduced at the top of the separator and
percolates down, stripping the uranium concentrate from the dryer
exhaust. The scrubbed air is vented to atmosphere.
The filtrate from the yellowcake presses goes to the eluting
solution tanks, where it is adjusted with acid and nitrate to make
fresh eluant. Ten per cent of the filtrate is delivered to the
slimes tails to prevent sulfate buildup.
The slimes tails from the RIP banks are sent to a neutraliza-
tion tank, where lime is added to bring the pH to about 9.5. After
neutralization, these wastes are delivered to the slimes tailings
pond, a large lagoon located near the Cheyenne River.
The Mill Survey
For purposes of analyzing the mill process and characteri-
zing the resulting liquid wastes, eight sampling stations were set
up within the mill (Table 1); the field survey was carried out from
July 25 through July 30, 1957. Six other sample types were col-
lected outside the mill. These included:
a. Liquid from the slimes tailings pond.
b. Wet solids from the slimes tailings pond.
c. Liquid from the sands tailings pond.
d. Dry sand from the sands tailings pile.
e- Silt and wet solids from the slimes pond outlet to the
Cheyenne River.
f- Direct liquid drainage to Cottonwood Creek.
Sampling stations 1, 2, 4, 5, and 8 were main mill process
streams, whereas 3, 6. and 7 represented liquid waste streams.
-------
MINES DEVELOPMENT COMPANY
Table 1. SAMPLING STATIONS WITHIN THE MILL
Station
number
1
2
3a
4
Description
Raw ore slurry before entering leach tanks.
Leached ore slurry before sand-slime separation.
Sand slurry proceeding to sands tailings pond.
IX feed solution.
Loaded eluting solution from RIP circuit.
Slimes tails from IX banks, after absorption or uranium
and before neutralization.
Neutralized slimes tails proceeding to slimes pond.
Yellow cake.
a Additional dilution water is added to this process stream beyond the
sampling point, to facilitate pumping.
Sampling inside and outside the uranium mill was performed
by personnel of the Public Health Service and the South Dakota
State Department of Health. Sampling within the mill commenced
on July 25 and was completed on July 30, 1957. During four days
of this period, samples at each station within the mill were col-
lected hourly for eight hours and composited; in the middle of the
survey the mill samples were collected hourly for an uninterrup-
ted 24-hour cycle and were composited into 3-hour samples. This
program resulted in four 8-hour composite and eight 3-hour com-
posite samples at each of the eight sampling stations within the
mill. A full set of samples was shipped to the Robert A. Taft
Sanitary Engineering Center of the Public Health Service, in Cin-
cinnati, Ohio, for gross radioassay and chemical analysis; iden-
tical samples were sent to the Occupational Health Field Station,
Public Health Service. Salt Lake City, Utah, for radium analysis.
A primary purpose of the study was to make a complete bal-
ance of all radium entering and leaving the mill in both dissolved
and undissolved form, and separate radium analyses were per-
formed on the solid and liquid phases of the samples. For this
purpose, the hourly samples collected during the 8-hour periods
on July 25 and 26 were composited into a single sample for radium
analysis for each mill station. The hourly samples collected over
the 8-hour periods on July 27 and 30 were similarly composited.
The hourly samples taken for the 24-hour period during July 28
and 29 were composited into a single sample for each station for
radium analysis.
Samples of the direct drainage to Cottonwood Creek were
collected on each of the latter three days of the survey. A single
representative sample of dry sands was obtained from the sands
pile, and a single representative sample of wet solids was taken
Irom the slimes pond. Liquid samples were collected from the
GPO 813-173-3
-------
10 RESIN-IN-PULP PROCESS
sands and slimes ponds on each of three days. One representa-
tive mud sample was collected at the slimes pond outfall to the
Cheyenne River. Duplicate sets of these samples were shipped to
the Public Health Service laboratories in Ohio and Utah for the
analyses noted above.
A record of the tonnage of ore processed was obtained fre-
quently during the survey from the weightometer preceding the rod
mill. Weir discharge records for flow were kept regularly at the
weir box preceding the RIP section, and the quantity of eluate
flow was obtained from plant equipment. Water flows at the var-
ious points of addition were obtained from plant personnel, as
were estimates of chemicals used in the process. These in-
cluded ammonium nitrate, magnesium oxide, lime, iron, and
sulfuric acid. Water from Cottonwood Creek was pumped to the
sands pond during the survey period in order to provide suf-
ficient water in the sand-slime separation and for diluting the
sand slurry. Estimates of solids concentrations, pulp density, etc.,
at various points in the mill were obtained from the operating
personnel.
Laboratory Procedures
RADIOACTIVITY
Radium was determined generally by coprecipitation with
barium sulfate. Following pretreatment of the various types of
samples to put the radium in solution, the procedure consisted
essentially of evaporation with sulfuric acid, removal of polonium,
coprecipitation of radium with barium sulfate, purification, and
alpha counting of the precipitate. Rather extensive pretreatment
of the undissolved solids of the mill samples was necessary to
entirely dissolve them. Those samples that contained large quan-
tities of undissolved solids were centrifuged for separate radium
analysis of liquid and of solid phases. The solids were then
washed with water to wash out the liquid not removed by centri-
fuging, and this wash water was added to the liquid portion to be
analyzed for radium.
Most of the samples had high concentrations of suspended and
settleable solids. The gross radioactivity analyses of the suspend-
ed and dissolved solids were performed independently on separate
representative portions. This procedure eliminated the need for
a large number of absolute quantitative transfers for the radio-
activity determinations. A known volume of sample was filtered
through a membrane filter and washed. The filter and filtered
solids were removed and ashed at 600°C to constant weight. The
filtrate was evaporated to dryness and ashed at 600°C to constant
weight. Solids concentrations were based upon the total sample
volume.
-------
MINES DEVELOPMENT COMPANY 11
The general procedure for gross radioactivity determination
is described elsewhere. Dissolved and undissolved solids were
analyzed separately, and a self-absorption analysis was perform-
ed for a representative sample from each station. This procedure
also is described elsewhere. *
CHEMICAL
Chemical analysis of the various samples was performed by
standard methods outlined in detail elsewhere.5 Nitrates were
determined by the phenoldisulfonic acid method, and sulfates,
calcium, and magnesium by the gravimetric methods outlined.
Iron was determined by the phenanthroline method, manganese by
the periodate method, and chlorides by the mercuric nitrate
method. The results for manganese may have some error due to
interferences caused by iron and chlorides, and in future work it
is planned to determine manganese colorimetrically by the ammon-
ium persulfate method to correct for these problems.
Analytical Results
During the survey period the average rate of ore processing
was 517 tons per day. Table 2 shows the average slurry flow at
each mill sampling station, as well as the specific gravity and
suspended solids content. Slurry flows were based upon observa-
tions within the plant and on computations that accounted for
specific gravity of the dry solids, dry solids flows estimated from
ore processed, specific gravity of the liquid phase, and observed
specific gravity of the slurry. The slurry flow at Station 3 is not
the entire flow going to the sands pond, as considerably more
dilution water was added just following the sample collectoin point
at this location.
Table 3 presents radium concentrations, dissolved and un-
dissolved, in the various mill process streams as well as at the
several locations outside the mill. These are the average re-
sults for the survey period, and are given in stet of slurry, and
in stet dry suspended solids. As can be seen, there is good
agreement between the dissolved radium concentrations at Sta-
tions 2, 4, 5, and 6, and between Station 7 and the slimes pond
liquid. Similarly, the concentrations of radium per gram of dry
solids agree well at Stations 1 and 2, at Stations 4, 5, and 6. at
Station 7 and slimes pond, and at Station 3 and sands pond (dry
sands).
Gross alpha and beta radioactivity, dissolved and undis-
solved, for the mill stations is given in Table 4, in/i/ic/1. The
dissolved gross alpha and beta radioactivity shows a large gain as
the result of acid leaching, and, as expected, slimes neutraliza-
tion results in a major reduction of the dissolved gross activities
in the slimes liquid at Station 6. The results for Station 1 (Tables
3 and 4) indicate that the radium constitutes about 16 per cent of
the gross alpha activity of the ore.
-------
12
RESIN-IN-PULP PROCESS
Table 2. PROCESS STREAM CHARACTERISTICS
Mill Sampling
Station
1
2
3
4
5
6
7
Slurry flow
gaL min
91
91
91
145
12
142
151
Specific Gravity
of slurry
1.52
1.53
1.45
1.05
1.13
1.05
1.05
Dry suspen
by weight.
60
59
64
7,2
~ 0.07
7.2
8.2
ded solids
per cent
Table 5 indicates the results of the chemical analyses of sam-
ples of waste flows and pond contents. These results are given in
milligrams per liter (mg/1) of the liquid portion of the samples,
and represent only dissolved chemicals. Nitrates are expressed
as nitrate nitrogen, rather than nitrate ion. It should be noted that
the sands pond was also receiving water pumped from Cottonwood
Creek during the survey.
The general agreement between Station 7 and the slimes pond
and Station 3 and the sands pond is evident. Nitrate nitrogen was
Table 3. RADIUM CONCENTRATIONS
Station
1
2
3
4
5
6
7
8
Liquid from
Slimes Pond
Liquid fro-,
Sar.ds Por.d
Drainace to
Co;t-,-r.-.vood Creek
Dry Sar.ds
Solids from
Sliir.es Pond
Siirres Pond
Outlet Ditch
Radium in/i/ii: 1
Dissolved
93
2. 150
114
2.450
2.290
2.450
350
-
270
a
17
-
—
—
Undissolvcd
605.000
640. 000
163.000
273.000
2.050
233.000
250.000
-
--
—
—
-
—
—
Radium in dry
suspended solids, ftfiz g
650
710
170
3.640
~ 2.600
3.760
2.930
ISO
--
--
—
150
2.470
53
-------
MINES DEVELOPMENT COMPANY
Table 4. GROSS RADIOACTIVITY.
13
Station
1
2
3
4
5
5
7
Gross Alpha, ppc 1
Undissolved
R
3.66 x 10
R
2.86 x 10
R
1.08 x 10
0.880 x 106
R
0.010 x 10
R
0.810 x 10
0.725 x 106
Dissolved
•3
1.71 X 10
T
473 X 10
T
3.03 x 10
388 x 103
•i
2.930 x 10
T
94.5 x 10
1.03 x 103
Gross Beta, /j/ic, 1
Undissolved
R
4.79 x 10
R
3.96 X 10
R
0 992 x 10
0.962 x 106
f-
0.017 x 10
R
0.866 x 10
0.610 x 106
Dissolved
•3
1.02 x 10J
•3
613 x 10
T
2.05 x 10°
555 x 103
T
5.740 x 10
1
47.5 x 1015
1.23 x 103
also determined for the liquid portions of samples from other mill
stations, and was essentially zero at liquid portions of samples
from other mill stations, and was essentially zero at Stations 1,
2, and 3. but was 15, 8, 200, and 230 mg/1, respectively, at
Stations 4, 5, and 6. Based on the flows in Table 2 (liquid por-
tion), the ore processing rate of 517 tons per day. and a chemical
estimate of the nitrate ion required for exchange with uranium,
the nitrate ion use during the survey was calculated to be about
15 pounds per ton of ore processed. This is in good agreement
with the mill records.
Table 5. QUALITY OF WASTE FLOWS AND PONDS CONTENTS.
Material dissolved in liquid sarrcle. me 1
Sampling
station
Slimes to Tails
(Station 7)
Liquid [rom
Slimes Pond
Sands to Tails
(Station 3)
Liquid from
Sands Pond
Drainage to
Cottor.'A'ood
Creek
Sulfate
2.330
2. 190
2. 180
1.970
1.090
Chloride
205
200
240
275
170
Calfiurn
730
820
570
440
360
Magnesium
75
SO
120
150
65
Iron
0.14
-0
Tract
~0
~0
Manganese
~0
1.3
7,0
-jo
~0
Nitrate
nitrogen
500
460
-0
-0
Trace
As has been indicated, one of the primary purposes of this
survey was to make a radium balance throughout the process, if
possible. Table 6, based upon the data in Tables 2 and 3. pre-
sents the radium balance for the mill, and indicates the various
paths by which specific quantities leave the mill. The acid leach-
ing process dissolves a certain amount of radium as well as the
uranium. Some of the suspended radium remains tied up with the
sands (about 170 micromicrograms of radium per gram of dry
-------
14
RESIN-IN-PULP PROCESS
sands) and is discharged with them to tails at Station 3. A portion
of the dissolved radium (150 microgram per day) present at Sta-
tion 4 goes with the loaded stripping solution at Station 5, although
the data of Table 3 indicate that dissolved radium is not extracted
or concentrated by the ion exchange section. As will be seen, vir-
tually all of this carried over radium becomes a part of the yellow-
cake that is shipped out of the mill. The neutralization of the
slimes tails with lime results in precipitation of a substantial
fraction (about 85 per cent) of the radium dissolved in the liquid
at Station 6, as shown by both Tables 3 and 6.
Table 6. RADIUM BALANCE
Station
1
2
3
4
5
6
7
Radium, mg/day
Undissoived
301
316
81
217
0.1
223
225
Dissolved
0.049
1.06
0.057
1.95
0.150
1.90
0.316
Total
301
317
81
219
0.25
225
225
As a check on the data in Table 6, the radium content of the
ore can be estimated within reasonable limits of error. If radio-
active equilibrium of radium with uranium is assumed on the
basis of 517 tons per day of ore that assays 0.20 per cent U.,0,,,
it has been calculated that 270 milligrams per day of radium
enter the mill with the ore. This is in satisfactory agreement
with the totals of Table 6.
Further computations indicate that virtually all of the radium
present at Station 5 (150/tg/day) becomes a part of the final uran-
ium concentrate. If 96 per cent over-all recovery of uranium is
assumed and notice is taken that the yellowcake is approximately
75 per cent uranium as U.,0 •*, calculations indicate that roughly
1.2 tons of yellowcake are produced per day. If the radium con-
centration of 150^/tg/g of dry solids in Table 3 is used about 160
micrograms per day ( g/day) of radium leave the mill with the
yellowcake. This is quite close to the figure of 150 /tg/day for
Station 6 (Table 6).
From Table 6 it is evident that about 300 milligrams per day
of radium enter the mill. About 80 milligrams per day leave with
the sands at Station 3, and 220 milligrams continue through the
process to the slimes pond. Some 60 micrograms per day of dis-
solved radium are discharged to the sands pond and about 300
micrograms per day of dissolved radium go to the slimes pond.
The data in Tables 3 and 4 for Station 7 indicate that of the dis-
-------
MINES DEVELOPMENT COMPANY 15
solved gross alpha activity of 1,080/i/iC/l., about 30 to 35 per
cent is due to dissolved radium.
In similar fashion rough balances of the gross alpha and beta
radioactivity can and have been made. They need not be reproduced
here, as radium is the specific radioelement of major interest, and
its balance has been shown. Of interest,'however, is the fact that
the undissolved radium varies between 15 and 38 per cent of the
undissolved gross alpha activity at Stations 1 through 7, while
dissolved radium constitutes from 0.1 to 35 per cent of the dis-
solved gross alpha activity. Details of these figures can be ob-
tained by a simple conputation with the data of Tables 3 and 4.
Although dissolved radium is not concentrated in the ion ex-
change section, dissolved gross radioactivity is. Table 4 indicates
that the dissolved gross alpha and beta radioactivity was concen-
trated by a factor of 8 to 10 in the RIP section, or between Sta-
tions 4 and 5. This is in good agreement with the estimated con-
centration of uranium by a factor of 10, as quoted by mill per-
sonnel and by Dayton. ^
Summary and Conclusions
About 17 per cent of the gross alpha activity of the ore pro-
cessed at the uranium refinery under study was due to the pre-
sence of radium. At the ore processing rate of 517 tons per day,
the liquid and solid wastes from ore processing contained about
0.6 milligrams of radium per ton of ore processed, or a total of
approximately 300 milligrams per day. The bulk of this radium
99.8 per cent, remained in undissolved from throughout the
process, and was effectively retained in the sands and slimes
tailings ponds.
About 0.2 per cent of the radium that entered the mill with the
ore either left the refinery in the final uranium concentrate or was
delivered in dissolved form to the slimes or sands pond. Specifi-
cally, it has been estimated that some 310/*g/day of dissolved
radium went to the slimes pond, about 60 ^g/day went to the sands
pond, and about 150/tg/day of radium left the mill in the dried
yellowcake.
The radium content of the dried sands was about 150 micro-
micrograms per gram (/*/ig/g) of sands, while the radium content
of dried slimes solids was 2, 500 or more /i/ig/g. Sands accumu-
lation was in the neighborhood of 440 tons per day, while slimes
accumulated at a rate of about 80 tons per day. Effective reten-
tion and confinement of these solids and of the tailings pond
liquids resulted in practically no radium leaving the plant site
during the survey, except for that contained in the uranium con-
centrate that was drummed and shipped. The very small amount
of direct drainage to Cottonwood Creek was from the sands pond
and contained little radium or nitrate nitrogen.
-------
16 RESIN-IN-PULP PROCESS
The data in Table 3 indicate that some of the radium becomes
dissolved during the acid leach process (see data for Stations 1
and 2). Radium was not concentrated, however, by the RIP sec-
tion the concentration of dissolved radium in the loaded stripping
solution (Station 5) was essentially no different from that in the
IX feed (Station 4) or the slimes to tails (Station 6). The opposite
occurred in regard to gross alpha and beta radioactivity: the
gross alpha activity in the IX feed was 3.88 x 10 puc/l, in the
loaded stripping solution it was 29.3 x 105/t/tc/l, or greater by a
factor of 7.5; and in the slimes to tails it was reduced to 0.95 x
105/i/jc/l. or a factor of 4.1 as against the IX feed. The gross
beta activity behaved similarly; the loaded stripping solution had
a concentration 10.3 times that of the IX feed, while the slimes to
tails showed only 8.6 per cent of the dissolved beta activity of the
IX feed. Hence, the data indicate that although uranium was con-
centrated as usual by the RIP section, radium was not concentra-
ted here.
Slimes neutralization before discharge to the tailings pond
reduced the dissolved radium concentration from 2,450 to 350
ftfig/l, or by about 85 per cent. Liquid from the slimes pond
showed a comparable concentration of dissolved radium about 270
j»/jg/l. The data for Station 7 indicate about one-third of the dis-
solved gross alpha activity in the slimes pond liquid was due to
radium.
The nitrate nitrogen concentration in slimes pond liquid was
460 mg/1. While this is probably of no public health significance
in this case, such concentrations could be of considerable im-
portance at other locations where public water supplies are in-
volved. These quantities of nitrogen could also provide sufficient
nutrient material to result in undesirable blooms of algae and
other biota in streams and reservoirs. Shortly after the field
survey was completed the use of sodium nitrate in place of am-
monium nitrate was instituted at the refinery, but it is doubtful
that this would appreciably alter results presented here.
WASTE DISPOSAL PRACTICE
As has been noted, there has been quite careful control of .
wastes at the Mines Development Company refinery at Edgemont,
South Dakota. Sand and slime solids have been effectively con-
fined at the plant site, and liquid waste releases have been at a
minimum.
Waste disposal from this mill and from mills employing a
similar process should continue to be carefully supervised, and
should in any instance be based upon available knowledge of down-
stream water uses and of the fate of the wastes in the water en-
vironment. The sand and slimes solids should generally be effec-
tively retained and confined, as they contain considerable radium
and other radioelements. Their release to a stream would result
-------
MINES DEVELOPMENT COMPANY 17
in long term contamination of the watercourse, and they could
significantly contaminate equipment at downstream domestic or
industrial water treatment plants.
Liquid wastes from the slimes ponds at this and similar mills
should be released to surface waters only in accordance with exist-
ing regulations and on the basis of detailed information as to local
downstream water uses. This liquid, free of suspended solids,
contains considerably more than allowable concentrations of radium
and of nitrate nitrogen. At the uranium refinery studied, such con-
trol is especially important in view of the zero flows and extended
low flow periods that occur in the Cheyenne River. As has been
noted, these flows are such that little or no dilution of the effluent
would occur for extended periods of time. Any susceptible ground
water supplies located near the slimes tailings ponds of such re-
fineries should be tested periodically for nitrate nitrogen and
radioactivity content, as infiltration of the slimes pond liquid into
the ground water may occur at the pond.
As regards measurement of radioactivity in the effluent from
the slimes tailings pond at Edgemont, South Dakota, it appears
reasonable to suppose from these studies that the dissolved radium
content will generally be in the neighborhood of 30 per cent of the
dissolved gross alpha activity. It seems feasible, for routine
measurement purposes, therefore, to analyze these samples for
dissolved gross alpha activity and apply the factor of 0.30 to com-
pute dissolved radium. An occasional analysis for radium itself
will then serve as a check, and refine the percentage figure to be
used. In this way, adequate control can be provided and the
costs and labor of sample analysis for routine control can be mini-
mized. This percentage figure might also prove adequate within
reasonable limits of error for application at other mills using
the same process, but this cannot be known until further studies
are carried out.
Increases or decreases in the rate of ore processing at the
Edgemont, South Dakota, refinery, or changes in the U.,0 con-
tent of the ore, should be proportionately reflected in many of the
waste quantities. This applies particularly in the case of radium.
Thus, if no important process change is assumed, a capacity
increase to, say, 750 tons of ore per day, at 0.20 per cent U«0
should result in about 440 milligrams per day (mg/day) of
radium entering the slimes tailings pond. Any basic change in
the type of ore received would, of course, affect these figures,
especially the latter.
It is not intended here to generalize very much beyond the
local situation studied. Each uranium refinery waste disposal
problem is individual, and must be interpreted in terms of speci-
fic local water uses such as domestic water supply, irrigation of
croplands, recreation, etc., as well as in terms of the specific
-------
18 RESIN-IN-PULP PROCESS
waste characteristics and mill process. Available dependable
dilution in the receiving waters is also a critical factor in control
of these wastes. No two problems are likely to be precisely the
same, and in the dual interest of radiological safety and economy,
such waste disposal problems should each be carefully analyzed
as individual cases insofar as necessary. It is hoped that the in-
formation presented herein will provide some insight into the
waste disposal problems at acid leach resin-in-pulp mills
generally, but the data should not be applied to other mills with-
out considerable caution.
Acknowledgment
The generous cooperation and assistance of the following
organizations and individuals is hereby gratefully acknowledged:
Harold Webb, Plant Superintendent, and personnel of the Miles
Development Company; the U. S. Atomic Energy Commission;
D. E. Rushing and W. J. Garcia, Public Health Service, Salt Lake
City, Utah; M. Nelson and R. Spicer, South Dakota State Depart-
ment of Health; J. K. Neel, Public Health Service, Kansas City,
missouri; A. F. Bartsch, N. D. Wastler, S. D. Shearer, J. T.Jones
R. C. Kroner, H. L. Faig, and R. J. Lishka, Public Health Service
Cincinnati, Ohio
-------
THE ACID LEACH-SOLVENT EXTRACTION URANIUM
REFINING PROCESS.
I. GUNNISON MINING COMPANY,
GUNNISON, COLORADO.
S. D. Shearer *
C. E. Sponagle
J. D. Jones
E. C. Tisvoglou
Introduction
This is a report of an in-plant survey of the Gunnison Mining
Company, a uranium refinery utilizing the acid leach solvent ex-
traction process; Part II describes a study performed at the
Climax Uranium Company mill at Grand Junction, Colorado, which
has a similar process. This study was performed by the Public
Health Service during August 1958 with the cooperation of the Colo-
rado Department of Public Health, the Gunnison Mining Company,
and the U. S. Atomic Energy Commission.
The Gunnison Mining Company refinery is located at Gunnison,
Colorado, on the Gunnison River, about 130 miles upstream from
Grand Junction, Colorado, and the confluence of the Gunnison and
Colorado Rivers. At the time of the survey there were no direct
discharges to the Gunnison River. Figure 1 outlines the flow dia-
gram for this refinery and the process is described in detail in the
following sections.
The Gunnison uranium refinery began operation about 8 months
prior to this survey. At the time of survey, it was processing an
average of 330 dry tons per day of ore that assayed from 0.21 to
0.49 per cent U«0fi, and was producing uranium concentrate at a
rate of about 1, oOl) pounds per day. All wastes from the mill
were retained in a tailings pond, with no direct discharge to the
Gunnison River.
The Mill Process
ORE RECEIVING, SAMPLING. AND CRUSHING
Loaded ore-trucks are emptied into a 50-ton ore hopper,
from which ore is fed on a 30-inch conveyor belt to the sampling
plant. A magnet removes tramp iron from the ore as it dis-
charges from this conveyor onto a vibrating grizzly. Ore less
*Respectively, Senior Assistant Sanitary Engineer. Radiological
Pollution'Activities Unit; Sanitary Engineering Director, Colorado
River Basin Water Quality Control Project. Public Health Service,
Denver, Colorado; Health Physicist (present address: Radiation
Control Service, University of Michigan. Ann Arbor: and Chief,
Radiological Pollution Activities unit.
19
-------
Fllf WCJROML
<7>
SOLHBl
EX me no*
r*«
A-
I«*|C
SOLVfNT
(IIWCTIMI
T*»f
"?/F
i J niTt" 1 J^fcitn j riLTiti i fFWMCA
F_«_siuoer. n ""(r"i"" rn ""' rfn r.j
o-
Figure 1 . Flow diagram for fhe Gunnison Mining Compony uranium
,,,HI ' "• • ^ « I- * ,,o,,rt 1 O '-.R
-------
GUNNISON MINING COMPANY 21
than 3 inches in size drops through the grizzly onto a 24-inch con-
veyor, while larger pieces are crushed in a jaw crusher, and then
rejoin the ore flow. The 24-inch conveyor feeds a vibrating screen
through which ore less than 3/4-inch in size drops onto another 24-
inch conveyor belt; larger pieces are crushed in a gyratory crush-
er and then fall onto the belt. As the ore leaves the end of this
conveyor, a 10 per cent sample is taken by a chain and bucket
sampler, while the remainder of the ore is conveyed to one of two
250-ton fine-ore bins.
The sample is screened, the larger pieces are crushed in a
jaw crusher, and a 10 per cent sample is taken by a second chain
and bucket sampler; the balance of the original sample is convey-
ed to the fine-ore bins.
The remaining sample is again screened and crushed, and
enters a vezin sampler from which a 5, 10, 15. or 20 per cent
sample may be taken as required. The balance of the original
sample proceeds to the fine-ore bins. The final sample collect-
ed amounts to about 0.1 per cent of the original ore fed to the
process, or two pounds per ton. This representative sample is
assayed for its U000 content.
3 o
An additional 50-ton fine-ore bin is available for temporary
storage of special ores or excess fine ore. This is not ordinarily
used, however, and only a small stockpile is usually maintained.
Blending is not practiced; the ore is processed immediately upon
receipt at the plant.
GRINDING
The fine-ore bins feed a conveyor belt that transports the ore
to the rod mill. Feed tonnage is determined by a weightometer
connected to the belt. The ore goes to a 6- x 12-foot rod mill in
series with a 48-inch spiral classifier. As the ore enters the rod
mill, it is slurried by the addition of water and a small amount of
sodium carbonate solution (for corrosion prevention.)
Slurry from the classifier discharges into a sump where sod-
ium chlorate is added to oxidize ferrous iron to the ferric state and
to maintain an EMF of over 400 in the leach tanks. The slurry is
pumped to the operating floor, passes through a small cyclone
that returns plus-35-mesh particles to the rod mill feed, and
enters the leach tanks.
LEACHING
There are four acid leach tanks, each 16 feet in diameter and
16 feet deep, which are arranged for series flow. Steam is added
to the first tank to maintain a temperature of 85°F., and concen-
trated sulphuric acid is added to the first two tanks to maintain a
pH of 0.8 in the leach liquor leaving the last tank. Average leach-
ing time is about 17 hours. Constant agitation is provided by a
propellor-type mechanism in each tank.
-------
22 ACID LEACH - SOLVENT EXTRACTION PROCESS
SAND-SLIME SEPARATION
Slurry leaving the leach tanks is diluted with the overflow
from the number 3 thickener (see Figure 1), and the combined
flow enters a 30-inch classifier. This is the first of four class-
ifiers provided for sand-slime separation, the remaining three
being 24-inch size. The sands proceed through the four classi-
fiers and are discharged to a slurry tank. In this advance of the
sands, they are washed with thickener overflow; number 2 classi-
fier receives the overflow from number 2 thickener, and number
3 classifier, the overflow from number 1 thickener. Freshwater
is used in the number 4 classifier.
The overflow from each classifier carries the slimes into
the thickeners. Figure 1 shows the manner in which this is done,
each classifier discharging its overflow into the thickener that is
adjacent to it.
The slimes proceed through each of the four thickeners, are
washed counter-currently during their travel, and the spent
slimes are discharged to the slurry tank from the number 1
thickener. The washed sands and slimes are combined in the
slurry tank, repulped with raffinate from the solvent extraction
process, and discharged to tails. Equipment is available for
feeding lime to the slurry tank, but this is not done.
ACID LIQUOR STORAGE
Pregnant acid liquor from the number four thickener proceed;
to a 22-foot diameter by 10-foot deep storage tank. An EMF ad-
justing tank and two filter presses are provided following this stor
age tank. These units are incorporated into the plant for the pur-
poses of (a) reducing ferric iron to ferrous iron, in order to elim-
inate interference in the solvent extraction process, and (b) re-
moving the small amount of slimes remaining in the pregnant acid
liquor, so as to eliminate difficulties in the solvent extraction
process. Usual practice at the time of the survey, however, was
to bypass most of the acid liquor around these units into a second
storage tank. This practice at times caused emulsification of the
solvent due to the presence of slimes in the acid liquor entering
the solvent extraction process.
SOLVENT EXTRACTION
Pregnant acid liquor from the storage tank is pumped to a cor
stant head tank above the operating floor, from which it flows into
the first of two solvent extraction tanks. Uranium is extracted
from the acid liquor by the organic solvent by alternate cycles of
agitation and quiescence. There are five such mixing-settling
cycles in these tanks. Flow of the organic solvent is counter-
current to that of the acid liquor. The raffinate, or barren acid
liquor, is discharged into a holding tank from which a portion is
pumped directly to the tailings pond, while the remainder is used
-------
GUNNISON MINING COMPANY 23
to repulp the sand-slime slurry, as previously described. There
is recovery of some solvent that has been carried over, which
rises to the surface of the liquor in the raffinate tank.
SOLVENT STRIPPING AND YELLOWCAKE PRODUCTION
The pregnant organic liquor from the solvent extraction pro-
cess goes to a bank of three mixing and settling tanks, where the
uranium is stripped from it with a counter-current flow of a 10
per cent sodium carbonate solution. The barren organic is re-
turned to a holding tank for recycling through the solvent extrac-
tion system.
The prenant sodium carbonate solution is passed through a
filter press for removal of iron as ferrous carbonate. Sludge
from the press is returned as a slurry to the rod mill, while
the filtrate proceeds to one of two precipitation tanks, each of
which is 12 feet in diameter. Concentrated sulphuric acid is add-
ed to neutralize the sodium carbonate, and the solution is heated
with steam to about 165° F. to drive off any excess carbon dioxide
that may be present. Magnesium oxide is then added to bring the
pH to about 7. During this process the mixture is constantly stir-
red by a mechanism within the tank. Uranium is precipitated as a
sodium salt. When precipitation is complete, the tank contents are
filtered and the yellowcake recovered by use of a filter press.
Filtrate from the press is returned to the slimes thickeners, and
the yellowcake goes to a drying furnace. The dried produce is
drummed, weighed, and shipped to an Atomic Energy Commission
facility. Yellowcake is produced only during the day shift.
The Mill Survey
For purposes of analyzing the mill process and characteri-
zing the resulting liquid waste, ten sampling stations were selec-
ted. Samples were collected on an hourly basis from August 7 to
August 11, 1958 and were composited at each sampling point
over the periods shown:
Cycle 1 (24 hours): 4 PM. August 7 to 4 PM, August 8.
Cycle 2 (35 hours): 4 PM. August 8 to 4 AM. August 10.
Cycle 3 (36 hours): 4 AM. August 10 to 4 PM. August 11.
The sampling stations selected are described in Table 1.
All samples were collected inside the mill building, with the
exception of those at Stations 6 and 8, which were collected at
the points of discharge to the tailings pond.
During the survey plant flows were obtained at various stations
and locations from the operating records of the mill and directly
by the survey party. Mill records provided frequent data as to the
flow of barren organic pregnant acid liquor and Na?CO., stripping
solution. Installed flowmeters gave data as to the acid ilow to the
-------
24
ACID LEACH - SOLVENT EXTRACTION PROCESS
first leaching tank and the flow of wash water to the number 4
classifier. Flows at Stations 6 (raffinate to tails) and 8 (sand-
slime slurry to tails) were measured hourly in terms of the a-
mount of time required to fill a 55-gallon drum. Hourly readings
from the weightometer, together with plant records of the per
cent moisture in the belt feed, yielded accurate data regarding
tonnage of ore processed. The daily yellow-cake production was
obtained from plant records.
Table 1. GUNMSON MILL SAMPLING STATIONS
Station Number
1
2
3
4
5
6
7
8
9
10
Description
Classifier effluent
Acid leach tank effluent
Sands entering slurry tank
Slimes entering slurry tank
Pregnant acid liquor to solvent extraction
Raffinate to tails
Pregnant organic to stripping circuit
Sand-slime slurry to tails
Yellowcake
Acid leach tank feed
Plant equipment and records included hourly observations of
pulp density and per cent solids by weight at Stations 1, 2, 3, 4,
8, and 10. Notes were made of any process interruptions during
the survey.
Daily composite samples were collected and assayed by plant
personnel for the U«0 content of the mill heads (from the rod
mill), the tails (from the slurry tank), the leach tank effluent, the
pregnant acid liquor, and the raffinate, and these data were made
available for this survey. Records of chemical consumption (acid.
sodium carbonate, etc.) for the month preceding the survey were
also made available, and chemical use figures for the survey
period were obtained.
Sampling was performed by personnel of the Colorado De-
partment of Public Health and the Public Health Service. The
yellowcake samples were collected by mill personnel, a small
vial of about 30 grams being composited from the day's produc-
tion. All other samples were collected hourly during each cycle
indicated.
All samples collected during the survey were shipped to the
U. S. Public Health Service, Robert A. Taft Sanitary Engineering
Center, at Cincinnati, Ohio. Portions of selected samples were
then sent to a private laboratory for analysis of dissolved and un-
-------
GUNNISON MINING COMPANY 25
dissolved radium. All other analyses were performed at the Cin-
cinnati laboratory.
Yellowcake production amounted to 1, 258 pounds, 1, 250
pounds, and 3. 245 pounds during sampling cycles 1, 2, and 3 re-
spectively. As noted earlier, yellowcake production was carried
on only during the day shift and cycles 1 and 2 each included a
single day shift, whereas cycle 3 included two day shifts.
Analytical Results
Figure 2 is a schematic flow diagram that indicates the liquid,
solids, and slurry flows at the various process location. At each
station the slurry flow is given first, with solids and liquid flows
next in order. These figures were obtained on the basis of obser-
ved specific gravities, tonnages of ore, and other laboratory and
field data; computations treat the flow as being composed on two
separate streams, liquid and solids. Flows are given to the nearest
gallons per minute (gpm). These figures and all succeeding re-
sults represent combined computations for cycles 2 and 3 of the
survey period, as the samples for cycle 1 were not analyzed.
Late in cycle 2 and extending well into cycle 3 of the survey a
batch of custom ore was processed. This ore assayed 0.49 per
cent or more U^O.., compared to the more usual ore assaying
about 0.22 per cent U 0R. Because of this, and as a result of the
time lags at several points in the process (for instance, the acid
leach process took about 17 hours), it was necessary to combine
the data for cycles 2 and 3 in order to make balancing computa-
tions.
Table 2 presents certain process stream characteristics for
the stations sampled. Slurry flows in gallons per minute, speci-
fic gravity of slurry, specific gravity of dry solids, and the per
cent dry suspended solids by weight are shown for each sampling
station. Table 3 indicates the solids balance for the process dur-
ing cycles 2 and 3. Approximately 330 tons per day of ore.were
processed during the survey, as indicated at Station 10. Acid
leaching dissolved about 6 per cent of this total tonnage: i.e., 19.8
tons per day left the leach tanks in the dissolved state. The preg-
nant acid liquor (Station 5) contained about the same amount of
dissolved solids.
Table 3 shows that the total plant output (Stations 8,9, and 6)
was about 332 tons per day (318.5 + 0.7 + 12.6), which is in excel-
lent agreement with the 330 tons per day of ore processed. The
total solids into the slurry tank (Stations 3 + 4 + the raffinate
solids going to the slurry tank) total 318 tons per day, which
agrees with the output to tails (Station 8) of 318.5 tons per day.
The dissoK-ed solids into the slurry tank (Stations 3 + 4+ the raf-
finate solids to the slurry tank) total 10.0 tons per day, which is
in agreement with the dissolved solids output at Station 8. The
-------
WASH
WATER
135 gpm
FROM Cli/
PREGNANT
ORGANIC -
99 gpm
21
78
ACID
LEACH
SO
EXTF
i
BARRtM
"W29.6 gpm
S
CAR
EXT
LV
AC
i
:NT
TION
(
10IUM
BONATE
(ACTION
9 ,
97 gpm
21
76
PREGNANT (T)
ACID LIQUOR V'
176 gpm '
0
175
RAFFINATE (BARREN AC
PREGNANT
SODIUM
CARBONATE
CLASSIFIERS
SLIMES
f
THICKENERS
D LIQUOR)
Tfl THrrrrKFfli _rf
URANIUM
PRECIPI-
TATION
WASHED SANDS
WASHED SLIMES
63 gpm *
13 f
50
©-•
150 gpm
8
SLURRY .
TANK -4—*" TO TAILS
rr
V
22
I'm
RAFFINATE
53 gpm
0
£~
122 gum
0
122
,YELLOWCAKE
fi)
1500 LB./DAY
KEV:
EXAMPLE'
SLURRY, gpm 97
SOLIDS, gpm 21
LIQUID, gpm 76
'STATION-
Figure 2. Schematic flow diagram, Gunnison Mining Co., uranium mill,
August 1 958.
-------
GUNNISON MINING COMPANY
27
total dissolved solids output (Stations 6 plus 8) of 22.6 tons per
day isin good agreement with the dissolved solids after acid leach
and in the pregnant acid liquor.
Table 2. PROCESS STREAM CHARACTERISTICS
Station
1
2
3
4
5
6
7
8
9
10
Slurry Flow,
gals, min
-
97
50
63
175
122
29.5
165
3°
99
Specific
Gravity of
slurry
1.52
1.36
1.33
1.30
1.01
1.01
(2)
1.19
3C
1.35
Dry Susp.
Solids by-
Weight, ^c
55.4
39.6
38.7
38.1
~ 0.005
~ 0.004
,b
26.0
3°
41.2
Specific
Gravity of
dry solids
2.61
2.52
2.64
2.49
-
-
2b
2.37
n.d.
2.61
Average of cycles 2 and 3.
Liquid (negligible suspended solids).
c Solid sample.
Table 4 indicates the concentrations of radium 226 in dis-
solved and undissolved form at the several stations. Undissolved
radium is that portion retained on a millipore filter, while dis-
solved radium represents that contained in the filtrate. The data
are representative of cycles 2 and 3, and are given as micromi-
Table 3. SOLIDS BALANCE a
Station
Tons per Day
10
2
3
4
5
6
8
9
Raffinate to
slurry tank
Suspended
323. 0
309.5
120.8
187.5
-0
-o
308.5
0.7
-0
Dissolved
1.6
19.8
3.3
1.2
19.9
12.6
10.0
0.0
5.5
Total
329.6
329.3
124.1
183.7
19.9
12.6
318.5
0.7
5.5
' During cycles 2 and 3.
-------
28
ACID LEACH - SOLVENT EXTRACTION PROCESS
Table 4. RADIUM CONCENTRATIONS
Station
10
2
3
4
5
6
7
8
9
Radium 228 in total sample, /i>ig/l
Undissolved
271,000
345.000
94. 500
338,000
50
130
70
156.000
a
Dissolved
125
270
180
110
490
480
3
155
a.
Radium 226 in dry
undissolved solids.
PPS e
490
640
235
680
905
3,500
130
505
35
a Solid sample
crograms of radium-226 per liter of slurry, as well as per gram
of dry suspended solids.
Portions of the samples were also assayed for gross alpha
and beta radioactivity; these results are given in Table 5.
The dissolved alpha activities show clearly the effects of the
various steps in the process. A sharp increase in activity result-
ed from the acid leach (to 250, OOO^u/tc/1 at Station 2), and the
bulk of this activity was contained in the pregnant acid liquor from
the thickeners (Station 5). Most of it was removed by solvent ex-
traction (see Station 6, barren acid liquor), and appeared finally
in the yellowcake as uranium (Station 9, 306, 000/i/ic/g of dry sus-
Table 5. GROSS RADIOACTIVITY CONCENTRATIONS.
Station
10
2
3
4
5
6
7
8
9
Activity in total sample ftpc,- 1
Undissolved
Alpha
2.560,000
1.980.000
304.000
1. 810.000
1.400
730
n.d.
973.000
a
Beta
3.000.000
2.740.000
494.000
1.840.000
1.550
660
n.d
1.052.000
a
Dissolved
Alpha
5.900
250.000
17.400
4.400
272.000
5,900
n.d.
14. 030
a
Beta
15,300
830.000
50.600
33.200
475.000
11,500
n.d.
33.300
a
ActivHv in
undissolved solids.
we g
Alpha
4.633
3.6SO
760
3. 640
25.800
20.600
n.d.
3. 140
306.000
Beta
5.440
5.090
1.230
3,700
28.000
18.000
n.d.
3. 400
445. 000
Solid sample
-------
GUNNISON MINING COMPANY
Table 6. CHEMICAL QUALITY OF MILL EFFLUENTS
29
Determination
Total acidity as CaCO~
Mineral acidity as CaCO,
Hardness as CaCOj
Sulfate
Chloride
Iron
Manganese
Copper
Seler.ium
Sodium
Fluoride
Beryllium
Vanadium
Arsenic
Concentration, mg. 1
Station 6
Kaffinate
10.000
9.000
1.850
12.600
ISO
48
17
0.2
0
1.400
12
2.4
0.06
11
Station 8
Sands-slimes slurry
5.400
3.500
1.550
10.000
275
82
7
0.2
0
830
13
2.1
0.03
17
pended solids). Although the suspended solids in mg/'l of the preg-
nant and barren acid liquors were low. with correspondingly
slight alpha activity on a per liter basis, the suspended solids
that were present contained relatively high alpha activity on a per
gram basis.
Table 4 clearly shows that little radium-226 was dissolved by
the acid leach (see Stations 10 and 2, dissolved radium), and rad-
ium stayed mainly in the undissolved state through the entire mill
process. The uranium concentrate (Station 9) contained only 35
micromicrograms of radium-226 per gram, dry weight. While
the pregnant and barren acid liquors (Stations 5 and 6) contained
relatively high-radium suspended solids, in small quantities, the
dissolved radium content of these liquors was also relatively
high. This indicates that very little radium was extracted from
the pregnant acid liquor, and this is also borne out by the low
radium content of the yellowcake.
The radium-226 and gross alpha balances that follow clarify
these conclusions.
Samples of the barren acid liquor, or raffinate. (Station 6)
and of the sands-slimes slurry going to the tailings pond (Station
8) were analyzed for various chemical constituents of interest.
The results are shown in Table 6. Only the liquid portions of the
samples were analyzed, and the results, in mg/1, represent
only dissolved chemicals. The pH of the raffinate was 1.3.. and
that at Station 8 was 2.1.
-------
30
ACID LEACH - SOLVENT EXTRACTION PROCESS
Table 7. CHEMICAL CONSUMPTION
Chemical
H2S04
Mt-O
Kerosene
DEHPA
TBP
Separan
Na2CO3
NaC103
Decinol
Amount used per ton of ore
processed, Ibs.
70.2
0.85
0.225a
0.0054
0,0054
0.354
15.6
2.89
0.016
Gallons.
Chemicals used in the mill process are indicated in Table 7.
together with average consumptions per ton of ore processed.
These data are averages for the month preceding the survey, and
were supplied by company officials.
During the mill survey, about 36,000 Ibs H?SO,, about 900
Ibs of NaC103, and an average of about 5, 000 Ibs of Na^CO- were
used per day. MgO and Separan were used at rates of about 310
and 97 Ibs per day, respectively.
One of the primary purposes of this survey was to determine
the amounts of radium-226 in the process at various locations,
and the amounts in suspended and dissolved form in the effluents.
To that end, a radium balance for the process has been carried
out (Table 8).
Table 8. RADIUM BALANCE
Station
Radium - 226. me/day
10
2
3
4
5
6
7
8
9
naltinate to
s!-irrv tank
Undissolved
146
180
26
116
0.0-13
0.037
0.012
142
0.025
0.033
Dissolved
0.063
0.142
0.049
0.038
0.467
0.319
~0
0.141
-
0.139
Total
146
180
26
116
0.52
0.41
0.0!
142
0.025
0.18
-------
GUNNISON MINING COMPANY 31
The values in Table 8 were computed directly from the
radium concentrations of Table 4 and the slurry flows of Table 2;
a value of 1, 500 Ibs per day of concentrate for Station 9 was used.
An initial test of the results was made on the basis of the
uranium assays, yellowcake production, and the assumption of
radioactive equilibrium between uranium and radium in the raw
ore. The yellowcake production rate during the survey was 1, 500
Ibs/day, containing about 1.0 per cent moisture and 82 per cent
tLOp,. If radium-226 were in equilibrium with this much uranium
there would be 156 mg/day of radium-226 involved in the process.
The uranium assays of raw ore ("heads") and tailings indicated an
over-all efficiency of about 90 per cent for uranium recovery;
hence it is estimated that about 173 mg/day of radium-226 en-
tered the mill with the 330 tons per day of ore. This is in good
agreement with the data of Table 8.
A separate computation, based on the assumption of 0.22 per
cent U«0_ in the raw ore (because of the time lag between raw ore
and concentrate in the mill), the tonnage processed, and radio-
active equilibrium between uranium and radium, indicates about
180 milligrams of radium-226 enter the mill daily.
The radium balance is in generally good agreement through-
out the process with the exception of some discrepancy between
the undissolved radium results for Stations 2 and 10, before and
after the acid leach. Due to the presence part of the time of the
smaller batch of higher grade ore. and to the large time delay
during the acid leach process step, the result for Station 2 may
not be fully representative for the survey period and may be some-
what high. Other than this, the undissolved radium data indicate
146 mg/day entering from the cyclone, 142 mg/day at Stations 3
and 4 combined (separate washed sands and slimes slurries) and
142 mg/day at Station 8 (the combined slurries). Total mill input
and output therefore agree adequately so far as undissolved
radium is concerned.
The dissolved radium data also indicate good balances. The
46 7/jg/day in the pregnant acid liquor, together with the outputs
of 49 and 38/ig/day in the washed sands and slimes, respectively.
yield 554^g/day. This is accounted for adequately by the 319
/tg, day in the raffinate to tails (Station 6). the 25 fig/day in the
yellowcake. and the 141 /tg/day at Station 8 a total of 485 /tg/day.
The dissolved radium entering and leaving the slurry tank totals
226 /ig, day (Stations 3. 4. and the portion of raffinate to the slurry
tank) as against 141 /ig/day at Station 8. It appears possible here
that some of the radium initially dissolved in the raffinate preci-
pitated on mixing with the washed sands and slimes slurry in the
slurry tank.
It is also of interest to note that the dissolved radium leaving
the acid leach tanks (142/ig, day) does not account for the 554/ig/
day at Stations 3. 4. and 5. This, together with the observed con-
-------
32
ACID LEACH - SOLVENT EXTRACTION PROCESS
Table 9. GROSS ALPHA BALANCE
Station
10
2
3
4
5
6
7
8
9
Raffinate to
slurry tank
Gross alpha Radioactivity, me, day
Undissolved
1.380
1.040
85
620
1.4
0.5
n.d.
680
210
1
Dissolved
3.2
130
•4.7
2.9
260
3.9
n.d.
13
a
1.7
Total
1.380
1.170
90
620
260
. S. Geological Survey almost adjacent to the mi
property.
Discharge records for this station are available from Octobe
1911 through September 1958: continuous daily discharge measur
ments are available for the period October 1, 1945, through Sep-
tember 30, 1958. Only these continuous records were analyzed.
In order to obtain reasonable estimates of flow frequencies ex-
pected, arithmetic-probability and Gumbel type 8 curve fitting
techniques were employed. In general, the two types of analyses
-------
GUNNISON MINING COMPANY
33
were in agreement. Figure 3 presents the arithmetic-probability
analysis of the minimum daily and minimum monthly average
flows for the continuous years of record. This curve shows that
half the time the minimum daily flow has been equal to or greater
than 135 cfs, while the minimum monthly average flow has been
equal to or greater than 150 cfs; the minimum daily flows ranged
from 96 to 200 cfs, with an average of 136 cfs, and the minimum
monthly flows ranged from 111 to 252 cfs, with an average of 160
cfs. During the survey period. August 7 through August 11, 1958,
the flow ranged from 790 to 825 cfs, with an average of 803 cfs.
A hydrograph of daily flows for this station for the period Octo-
ber 1. 1957, through September 30, 1958, is shown in Fugure 4.
The discharge records from which this hydrograph was platted
show that the average flow for the water year was 838 cfs, while
the average flow for the six-month period October 1957 through
March 1958 was 355 cfs. This is a typical hydrograph of rivers
in the western United States, in which the spring snow-melts pro-
duce high runoffs during the three spring months and the flows for
the remainder of the year remain relatively steady.
300
250 -\—
200
en
u.
o
MINIMUM MONTHLY AVERAGE
(1945-58)
50
20 40 60 80 100
% OF TIME *THAN STATED VALUE
Figure 3. Occurrence of minimum daily and monthly average flows,
Gunnison River near Gunnison, Colorado.
It must be pointed out that the probability methods utilized
are statistical ones and are subject to variations such as length
-------
11 TTITTTTTTTTTTTT'TTTTTI
--uJ JVN" AI|
LLLLLLLLL J.I LI 1 |..1_1 U.LJJJ LI..I ill
.'* ''*,''' \ * '° '^ ?''
Figure 4. Daily flows for Gunnison River near Gunnison, Colorado.
Water year 1958.
-------
GUNNISON MINING COMPANY 35
of record, stream regulation, amount of upstream irrigation, etc.
For this particular gage the years of continuous record include
only a 14-year period. Since September 1937, the flow at this
station has been partly regulated by Taylor Park Reservoir about
37 miles upstream. There are also about 22.000 acres of irriga-
ted land above the station.
So far as can be ascertained, no public water supplies are
taken from the Gunnison River below Gunnison, Colorado. The
extent of other uses of the stream is not known in detail, but ex-
tensive downstream diversion for irrigation purposes occurs, with
the possibility that individual ranch families use the river water
in its raw state for domestic purposes.
WASTE DISPOSAL
During the mill survey a large unused tailings pond area was
available, and all wastes from the mill process were retained in-
definitely in this pond area.
About 12 per cent of the gross alpha activity of the ore that
was being processed at the uranium refinery under study was due
to the presence of radium. The liquid and solid wastes from ore
processed at the rate of 330 tons per day contained about 0.6 milli-
grams of radium per ton of ore processed, or a total of approxi-
mateily 150 mg/day. Some 99.6 per cent of this radium remained
in undissolved form throughout the process and was effectively
retained in the tailings ponds.
About 0.4 per cent of the radium that entered the mill with the
ore either left the refinery in the final uranium concentrate or was
delivered in dissolved form to the pond. Specifically, an estimated
480 micrograms of dissolved radium per day went to the tailings
pond, and about 25 micrograms of radium per day left the mill in
the dried yellowcake.
These studies indicate that two waste constituents in particular
are present in potentially hazardous quantities: radium-226 and
arsenic (see Tables 6 and 8).
As indicated earlier, the effluents from the Gunnison Mining
Company uranium mill contained per day about 500 micrograms
of dissolved radium-226. The records of flow of the Gunnison
River near the refinery indicate that if all of this dissolved rad-
ium were released routinely to the river, the dissolved radium
content of the river water would show an increase from essenti-
ally zero at high or flood flows to about 1.2^/ig/l at average or
usual flows, and to about 2.0/i/ig/l at low flows, which are rela-
tively rare.
Thus, by itself, the quantity of dissolved radium regularly
produced as waste from this refinery, while detectable in the
river, would not constitute a major hazard in terms of existing
standards/ In practice, of course.it is not usually released to the
-------
36 ACID LEACH - SOLVENT EXTRACTION PROCESS
river directly, although the extent of possible seepage from the
tailings pond is not known.
The undissolved radium-226 wasted daily from the mill to the
tailings pond constitutes a much more significant source of po-
tential environmental contamination. As has been shown else-
where. 10 the initially undissolved spent ore solids can result in
a relatively high degree of water pollution if discharged to a
river and permitted to accumulate on the stream bed. It is also
true here (see Table 4), as at other mills, that the lighter sus-
pended solids that are carried by the effluents contain relatively
high radium concentrations. As a result of these considerations,
it is quite important that tailings or spent ore solids should be re-
tained at the plant site effectively and regularly, and should not
be released to the Gunnison River in any regular or significant
quantity.
The problem of arsenic is somewhat different. From the data
of Table 6 and Figure 2 (liquid flow rates) it can be shown that the
two main effluent streams. Stations 6 and 8-, carry about 20,000
grams of dissolved arsenic per day to the tailings pond. No esti-
mate of the arsenic content of undissolved tailings solids has been
made.
If this quantity of arsenic were released routinely and the
available dilution in the Gunnison River, was considered the ar-
senic content of the river would, for a considerable portion of the
year, equal or slightly exceed the allowable concentration based
upon the Public Health Service Drinking Water Standard. ^ For
instance, the minimum monthly average flow, which would be ex-
ceeded only 50 per cent of the years, is 150 cfs (see Figure 3). At
this flow, the arsenic concentration would be about 0.057 mg/1 or
slightly more than the allowable 0.05 mg/1.
From the standpoint of chemical pollution the mill effluents
clearly should continue to be retained indefinitely, as has been the
practice in the past. This retention, however, raises the question
of the potential accumulation of arsenic at the rate of 20, 000 g/dav
in the soil near and surrounding the tailings pond. The possibility
that the accumulating quantities of arsenic may leach to the river
in increasing amounts, or to nearby well supplies (such as that
used by the Gunnison Mining Company) should not be ignored. For
a time, a minimal amount of monitoring of the river and any near-
by wells for arsenic appears to be desirable.
Acknowledgment
The generous cooperation and assistance of the following are
gratefully acknowledged: personnel of the Gunnison Mining Com-
pany; the U. S. Atomic Energy Commission; Stan May, Colorado
Department of Public Health; Bill Fixen, Public Health Service
Region VTH, Denver; and E. A. Pash, G. Harlow, Carl Shadix and
Carl Hirth, Public Health Service, Cincinnati, Ohio.
-------
THE ACID LEACH-SOLVENT EXTRACTION
URANIUM REFINING PROCESS
II. CLIMAX URANIUM COMPANY,
GRAND JUNCTION, COLORADO
J. B. Cohen *
C. E. Sponagle
R. M. Shaw
J. D. Jones
S. D. Shearer
Introduction
An in-plant survey of the Climax Uranium Company uranium
refinery at Grand Junction, Colorado, was performed by the Public
Health Service during August 1958 with the cooperation cf the Colo-
rado Department of Public Health, the Climax Uranium Company,
and the U. S. Atomic Energy Commission.
The Climax Refinery is located on the Colorado River, about
30 miles upstream from the Utah-Colorado State Line. It provides
an example of the acid-leach solvent extraction process for uranium
recover. Vanadium is also recovered at this plant. Figures 1 and
2 present the process flow diagram for this refinery and the waste
pond arrangement respectively, which are discussed in detail in
the following sections.
Mill Process
ORE RECEIVING, SAMPLING AND CRUSHING
Ore delivered to the plant by truck is weighed and unloaded
into truck bins. Each load is run separately through a jaw crush-
er, which reduces it to a maximum size of two inches. A sample,
varying in size from two pounds per ton on large lots (20 tons or
more) to about four pounds per ton on small lots (8 to 10 tons), is
taken automatically during crushing. High grade ore is hand sam-
pled or specially sampled, according to the size of the lot. Crush-
ed and sampled ore goes either to the stockpile or to fine-ore
storage bins. From these bins it proceeds to the rod mill where
it is ground to less than 14-mesh size (2 to 4 per cent retained on
14-mesh screen). During this process water is added to the ore
so that the finely ground material leaves as a slurry for the next
phase.
^Respectively, Sanitary Engineer, Radiological Pollution Activities
Unit; Sanitary Engineering Director, Colorado River Basin Water
Quality Contro. Project, Public Health Service, Denver, Colorado;
Captain, U. S. Army, Fort McPherson, Atlanta, Georgia; Health
Physicist: and Senior Assistant Sanitary Engineer, Radiological
Pollution Activities Unit.
37
-------
Figure 1. Flow diagram for the Climax Uranium Company, Grand
Junction, Colorado, Augusf 1958.
-------
CLIMAX URANIUM COMPANY
39
OEBQTE! OLD TAILINGS POM (ROT III USE)
Q DEMOTE! SiWtlKG STATIC*
HOTt: USUAL OPERATIC OF POMS I. 2, «nd 3
DESCRIBED 111 THT
Figure 2. Mill area and pond arrangement of Climax Uranium Co.,
Grand Junction, Colorado, August 15-18, 1958.
CONDITIONING AND CLASSIFICATION
Slurry from the rod mill is pumped to acid-conditioning tanks,
of which there are six arranged in series. A strong H?SO. solu-
tion (2-1/2 to 7 per cent acid) is added in the first tank: most of
this solution is recirculated from storage tanks following the sand
leach. A more dilute solution of HC1 and H?SO, is also added, this
being recirculated from the roaster gas scrubbing unit. The pH in
the first three conditioning tanks is about 1.0 to 1.5, rising in the
last three conditioning tanks in the range of 1.5 to 4.0. Some raf-
finate from the solvent extraction process is also returned to the
No. 1 conditioning tank along with the acid liquor from the sand
leach.
The purpose of this conditioning is to destroy the lime in the
ore, and change CaCO., to insoluble CaSO4. 13 This prevents
formation of the water-insoluble calcium vanadate during the
roasting process; instead, the water-soluble sodium vanadate is
formed. Acid conditioning takes about 1-1/2 hours.
Upon leaving the acid conditioning tanks, the slurry is sub-
jected to a second conditioning with ammonia to neutralize the re-
maining acid and to raise the pH to about 6.5. The principal pur-
poses of this neutralization are (a) to avoid corrosion, and (b) to
precipitate any uranium and vanadium dissolved during acid con-
ditioning or entering in the recirculated acid solutions. Precipi-
tation will occur in the pH range of 5.3 to 7.5.
Upon completion of conditioning the slurry is partially de-
watered in a cyclone separator before proceeding to a hydraulic
sizer. Liquor from the cyclone goes to the thickeners. In the
sizer, sands and slimes are separated hydraulically. An attempt
is made to maintain the slimes coming out of this unit in such a
-------
40 ACID LEACH - SOLVENT EXTRACTION PROCESS
manner that 85 per cent of the solids are less than 200-mesh in
size. All larger particles are removed as sands.
SAND LEACH
Sands from the sizer go to a spiral classifier for dewatering.
Overflow from the classifier goes to the thickeners, while the
underflow is discharged to one of ten acid leach tanks where the
uranium and some vanadium are leached from the sands. As the
sands leave the classifier, concentrated H?SO. is added at a rate
of about 105 pounds H?SO. per ton of dry solids. The acid leach
process is conducted as a catch operation; a tank is filled with the
sands-acid slurry and then handled as a unif
The leach tanks are constructed with false bottoms so that the
sands will be retained, while the liquor content of the tanks can be
drained off as underflow. When a tank is filled, the underflow is
recycled through the sands for about 2 hours. At the start of the
recycling operation, NaC103 is added to the liquor in an amount
equal to 3-1/2 pounds per ton of dry sands in the tank. At the sarr.e
time the liquor is heated to 90 degrees Fahrenheit by a heat exchan-
ger, using steam as the source of heat. At the end of 2 hours, re-
cycling is stopped, and the tank contents are allowed to "cure"
for about 8 hours. At the end of the curing period the acid liquor
is drained into a storage tank. Fresh water is then flushed
through the sands and run into the same storage tank until the pH
is in the range of 1.8 to 2.3. The water flow into the leach tank
is then stopped and the liquor remaining in the tank is diverted
into a second small (4, 500 gal.) storage tank to be used for roas-
ter gas scrubbing. The spent sands are then removed from the
leach tank by repulping with water and discharged to the tailings
pond.
Acid liquor from the first storage tank, which contains uran-
ium and vanadium leached from the sands, is returned to the No.l
acid conditioning tank as described in the previous section.
THICKENING AND ROASTING
The components comprising the thickener feed are slimes
from the classifier receiving sands from the sizing operation, and
filtrate from the disc filters following the thickeners. As this cm:;
bined flow enters the thickening tanks. Separan (200 pounds per da;
is added to improve sedimentation. Overflow from the thickeners
is discharged to Settling Pond No. 1, while the underflow, about
30 per cent solids, is filtered on disc filters. The filtrate is re-
turned to the thickener feed as mentioned above.
The filter cake drops into a screw conveyor and proceeds to
a mixer where Nad is added at the rate of 15 per cent by weight
of dry solids. This material with about 40 per cent water is still
too wet to serve as dryer feed, so it is mixed in a pug mill with
the previously dried filter cake-NaCl mixture in an amount suf-
ficient to reduce the moisture content to about 25 per cent. Thi?
-------
CLIMAX URANIUM COMPANY 41
material is fed to a gas-fired rotary dryer, which reduces the
moisture content to about 13 per cent. The dried material passes
through a 50-ton storage bin and thence to a gas-fired 10-hearth
roaster, operating at 1400°F. to 1600°F.
The roasting process converts insoluble vanadium compounds
in the ore to water-soluble sodium vanadates. This reaction can
be written
V0O..+ 2NaCl -f H0O-^2NaVO0+ 2HC1.
25 2 3
The hydrochloric acid gas is recovered for use in the acid condi-
tioners. by passing the roaster gas through the scrubbing units.
The uranium compounds remain insoluble in water. Calcines are
-split into two "streams" on leaving the roaster. One "stream" is
carried on a vibrating conveyor to a Baker cooler in which cool-
ing takes place by heat transfer into cooling water circulated a-
round the outside of the mechanism. In this process there is no
contact between calcines and water. The second "stream" is car-
ried on a vibrating conveyor to a quench tank where it is quenched
with filtrate and wash water from an Oliver filter that follows the
thickeners.
Cooled calcines from the Baker cooler are reground in a ball
mill: quenched calcines go to a spiral classifier. Sands from this
classifier join the cooled calcines entering the ball mill. The
classifier overflow is thickened in two 32-foot thickeners. Thick-
ener overflow is processed for vanadium, underflow for uranium.
VANADIUM EXTRACTION
The thickener overflow feeds five vanadium precipitation tanks
of 7, 100-gallon capacity each. Precipitation is carried out as a
batch process: the typical operation in any tank is as follows:
The tank is filled with thickener overflow, agitation begun,
and the contents brought almost to the boiling point by injection
of steam. Concentrated H?SO. is then added in an amount suf-
ficient to lower the pH to below 4.0. The tank contents are agi-
tated until complete precipitation of vanadium as sodium poly-
vanadate (redcake) is effected. The time required for this may
vary from 1/2 to 3 hours, depending upon the "grade" and charac-
teristics of the liquor. The tank contents are then run onto a fil-
ter tray and the liquor is drained to Pond No. 3. The redcake is
washed with water to leach sulfates from the cake. The amount of
water used is roughly equal to the volume of the tank, e.c., about
7, 500 gallons. Wash water is also drained to Pond No. 3. Washed
redcake is fused at 650 3C in a fusion furnace, and emerges as
V^O product, which is then cooled and drummed for shipment.
URANIUM EXTRACTION
Underflow from the thickener is filtered on an Oliver filter
and the filtrate, together with water used to wash the filter cake.
-------
42 ACID LEACH - SOLVENT EXTRACTION PROCESS
is returned to a storage tank from which it is used for quenching
roaster calcines. Washed filter cake is repulped with water and
goes to three acid conditioning tanks, in series. Here concentra-
ted HpSO. is added at about 230 pounds per ton of dry solids. The
uranium in the cake, together with vanadium that was not water
soluble, is dissolved in the acid during the contact time (about one
hour) in these tanks.
Upon leaving the last of these tanks the slurry goes through
three additional cycles of conditioning, filtration and repulping to
extract the maximum amount of uranium. Cake from the last fil-
ter is repulped with water and pumped as "slime leach tails" to the
tailings pond. Filtrate from the last filter is returned to the
process.
The pregnant acid liquor filtrate from the first and second fil-
ters is settled for removal of slimes. The liquor then goes to onu
of three 7500-gallon storage tanks and then to the solvent extrac-
tion tanks. Here, uranium is stripped from the acid liquor with
an organic solvent. The raffinate is returned to the No. 1 condi-
tioning tank, while the pregnant organic goes to a storage tank,
then into a circuit where the uranium is stripped from the solvent
with a 10 per cent Na~CO,, solution. Barren organic is returned
to the solvent extraction circuit, while the loaded Na^CO,, proceed;
to one of two precipitation tanks, operated as a parallel batch pro-
cess. When one of these tanks is filled, about 800 pounds of con-
centrated H?SO. is added to neutralize the NapCO,,, and the con-
tents are brougnt to a boil with steam to drive off excess carbon
dioxide. Ammonia is added to precipitate the uranium as an am-
monium di-uranate. When precipitation is complete, the tank
contents go to a filter press. Filtrate is returned to the No. 4
conditioning tank. Filter cake is fed to a gas-fired dryer operatin;
at a temperature of 900°F. where drying takes place, and the feed
is converted to U«0R (yellowcake), which is then drummed and
shipped to an Atomic Energy Commission facility.
OPERATION OF SETTLING PONDS NOS. 1, 2, and 3
As shown in Figure 2, in addition to a large tailings pond, thiv
smaller settling ponds, each approximately 150 feet square, are
provided for settling thickener overflow and vanadium tank filtrate
and wash water prior to their discharge to the river. Ordinarily
thickener overflow enters Pond No. 1. This pond discharges to
Pond No. 2, and thence to Pond No. 3. where the waste vanadiun;
liquors enter. The only discharge from this series of ponds is
from Pond No. 3 to the river.
During Cycle I of the mill survey, Pond No. 2 was being
cleaned out. During Cycle II, Pond No. 2 was still out of opera-
tion and cleaning operations were underway in Pond No. 3. Be-
cause of this situation, the contents of Pond No. 1 were dischar-
ged directly to the river, as were those of Pond No. 3. Conse-
GFO 813—1"'-
-------
CLIMAX URANIUM COMPANY
43
quently, the effluent samples collected from these ponds did not
reflect the discharge which would have occurred if the ponds were
being operated in their normal manner.
Since the survey in 1958, an additional tailings pond has been
constructed. The large tailings pond, shown discharging to the
drainage ditch on Figure 2, now discharges to the new pond. No
direct discharge to the river is anticipated from this new pond.
In addition, the settling pond arrangement has been changed so
that the only discharge to the river is from the vanadium extrac-
tion circuit. All other discharges now enter the tailings ponds.
Mill Survey
During the survey period, August 15 through August 18, 1958,
nineteen sampling stations were established for the purpose of ob-
taining representative samples from the mill process for analysis.
Table 2 describes the stations sampled; Figure 1 gives the loca-
tions. Sampling Stations 1, 2, 3, 4, 8,9,10,13,14,16, and 17 were
main mill process streams, whereas Stations 5,6, 7,11,12,15,18,
and 19 were representative of waste streams.
Sampling was performed by personnel of the Public Health
Service and the Colorado Department of Public Health. Two sepa-
Table 1. CLIMAX MILL SAMPLING STATIONS
Station
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Description
Classifier discharge to conditioning tanks
Conditioning tank effluent to hydraulic classifier
Thickener feed (65-foot tanks)
Sands to acid leach
Spent sands to tails
Influent to Pond No. 1
Effluent from Pond No. 1
Roaster feed
Roaster calcines
Thickener feed - vanadium circuit
Influent to Pond No. 3
Effluent from Pond No. 3
Vanadium product
Filter cake to solvent-extraction circuit
Slime leach tails
Filtrate to solvent extraction tanks
Uranium product
Combined effluent from Ponds Nos. 1 and 3
Sand-slimes tailings pond discharge
-------
44 ACID LEACH - SOLVENT EXTRACTION PROCESS
rate sampling cycles were selected during which samples were
collected continuously. The sampling periods were:
Cycle I - 3:00 PM, August 15 - 11:00 PM, August 16, 1958
(32 hours)
Cycle H - 3:00 PM, August 17 - 3:00 PM, August 18, 1958
(24 hours)
Compositing of samples during these periods varied with the type
and duration of operation. Adequate sampling was difficult at
several locations due to the batch type operations described pre-
viously. Also, flow at several other stations was not continuous
over the entire length of the sampling cycles. Detailed records
were maintained for each sampling station in order that data de-
rived from laboratory analyses could be adequately interpreted.
All samples collected during the survey were shipped to the
U. S. Public Health Service Robert A. Taft Sanitary Engineering
Center at Cincinnati, Ohio. Portions of all samples were sent to
a private laboratory for determinations of dissolved and undis-
solved radium; all other determinations were performed at the
Cincinnati laboratory.
Analyses and Results
To adequately determine the operating characteristics of the
individual units of the mill process, a flow balance was calculated
for the mill. This was accomplished by combining the laboratory
analyses with the flow measurements taken on the effluent streams.
To estimate the quantities of liquid, suspended solids, and dis-
solved solids at the various stages of the process during each cy-
cle a solids balance throughout the plant was assumed. Because
of this assumption the solids for each cycle necessarily balanced.
The input to the plant on the basis of a balance was calculated as
540 and 583 tons of ore per day for Cycles I and H, respectively.
Results from the two cycles as presented in this section of the re-
port have been combined and adjusted to represent the mill pro-
cessing 540 tons of ore per day; the solids flow at each station is
presented in Table 2.
Figure 3 illustrates the average flow for the two cycles in
terms of gallons per minute of liquid and solids at the stations sam-
pled during the process. The flow at Stations 4, 5, 7,11, 12.13, 15.
16, and 17 were measured during the survey while the flows at the
remaining stations were calculated from the solids balance.
Table 3 presents process stream characteristics, both mea-
sured and calculated. These and all other process stream char-
acteristics are reported on the basis of uninterrupted operation
at all stations and steady flow conditions for batch operations such
as the sand leach and the vanadium precipitation operations.
-------
CLIMAX URANIUM COMPANY
45
Table 2. SOLIDS IN PROCESS STREAM
Station
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Tons per day
Suspended
537
448
120
394
390
1.7
0.3
123
123
97
0.04
0.04
7.9
97
88
0.1
2.0
0.2
0
Dissolved
3.0
72
71
1.0
1.6
56
50
a
a
74
6.2
6.3
a
12
0.9
20
a
49
0.2
Total
540
560
191
395
392
57
50
123
123
171
6.2
6.3
7.9
109
89
20
2.0
49
0.2
•* Solid Sample
From Figures 1 and 3 comparisons between stations can be
made of the flows entering and leaving various sections of the mill.
For the over-all plant balance, the input to Station 1 can be com-
pared with the total output of the spent sands, Station 5; the in-
fluent to Pond No. 1. Station 6: the influent to Pond No. 3. Station
11; the vanadium product. Station 13; the slime leach tails. Sta-
tion 15, and the yellowcake product, Station 17. The effluent from
the ammonia conditioners, Station 2, should be accounted for at
the feed to the acid leach tanks. Station 4; Pond No. 1, Station 6;
and the roaster feed. Station 8. The roaster calcines at Station 9
should divide between the filter cake at Station 14. and Stations
11 and 13. The roaster feed at Station 8 should compare with the
calcines at Station 9. Lastly, the filter cake at Station 14 should
be measurable in the filtrate to the solvent extraction tanks.
Station 16, and in the slime leach tails, Station 15. These com-
parisons can also be used for estimating the quantities of water
added between sampling stations.
Table 4 shows the gross alpha and beta radioactivity concen-
trations. The figures are an average for the two sampling cycles.
-------
46
ACID LEACH - SOLVENT EXTRACTION PROCESS
4—
K
• a
ORE
\
CLASSIFIER
36
61
— <
CC»niTIC»-
C«S
31
JIL
262
— C
DOWrLr-F
'I8
SIZE'
5,
\
S»I»AL
CLASSIFIER
SO.^.
103 .
~>
— »
_^
*
^
— CD
A: o
LEACH
TA.»«S
i
F
Si»r
, TO TA
-
TANKS
"3
-C
rfj
•v— ..-
(5)«!
26
SCR'J33ERS
P
MA
J
PCI
V£»
TFR
»3
-^1. 1 ^
N.
~^l
«' Fl(^
©! ~c
9C
7) S°
~ C
•|J2L^
THIC.E.FR-;
PISC
FILTERS
^*-x
ROASTER
,_>
-*•
'«^_
)f F r :
sours FU
till! IS FL
SLL'^'r FU
MLi»"ILL FILT"
CLASSIFIER
•-0 Hf
TBn.EBEI1 ™ERFL»
)W -
V -
V -
WLINS
PE
1C gp«"
CC go
-------
CLIMAX URANIUM COMPANY
Table 3, PROCESS STREAM CHARACTERISTICS3
47
Station
1
2
3
4
5
6
7
a
9
10
11
12b
13
14
15
16
17
18b
19
Slurry
flow, gpm
97
282
562
58
82
469
408
c
c
133
90
145
c
35
41
50
c
545
8
Specific
gravity
of slurry
1.56
1.23
1.02
J.70b
1.49
1.01
1.01
-
-
1.14
1.01
1.01
-
1.33
1.22
1.08
-
1.01
1.00
Dry suspended
solids by
weight, %
59.2
23.9
35.1
67.3b
53.9
d
d
- 87
-100
11.0
d
d
- 100
35.0
29.5
d
- 100
d
d
Specific
gravity of
dry solids
2.47
2.63
2.63
2.56
2.63
-
-
2.13
2.30
2.21
-
-
2.73
2.60
2.66
-
5.81
-
-
Average for two cycles
One cycle only
Liquid (negligible suspended solids)
cines at Station 8, 1, 626 me/day, can be equated to the 1, 499
(16 + 5 + 1, 478) me/day calculated for Station n, the influent to
Pond No. 3; Station 13, the vanadium product; and Station 14. the
filter cake going to the uranium extraction process. In turn the
activity at Station 14, 1478 me/day, agrees well with the 1472
(1066 + 406) me/day encountered at Station 15, the slime leach
tails; and Station 16, the filtrate to the solvent extraction tanks.
For the over-all plant balance, 1778 me/day enter the plant as
compared with 1990 (273 + 31+16+5+ 1066 + 599) me/day that
leave at Stations 5, 6, 11, 13. 15, and 17. It should be noted here
that the concentrations of alpha and beta activity present in Table
4 are laboratory determinations and as such could be different
from the actual activity that might be measured at the plant, if
this had been possible. Material that was not in radioactive equil-
ibrium during the process would tend to return to equilibrium in
-------
48
ACID LEACH - SOLVENT EXTRACTION PROCESS
Table 4. GROSS RADIOACTIVITY CONCENTRATIONS
Station
1
2
3
4
5
6
7
8
9
10
11
12 a
13
14
15
16
17
18 a
19
Gross radioactivity of total sarr.ple (slurry), >i/ic/l
Undissolved
Alpha
3,350.000
1, 180.000
533.000
1.500.000
631.000
11.900
3,270
b
b
1.780.000
19.800
2.560
b
7.900,000
4.800.000
15.300
b
452
235
Beta
4.150.000
1.640.000
753.000
1. 500,000
923.000
20. 700
3.830
b
b
1,660.000
29.900
1.320
b
7.960,000
3.520.000
8.320
b
1.030
335
Dissolved
Alpha
329
3.030
1,880
418 a
I,i70
260
4.130
b
b
5.380
10. 700
5. 100
b
3.250
725
1.520,000
b
4.300
320
Beta
905
7.320
2.760
1.5C03
5.770
700
5. 460
b
b
14.700
48,200
26.800
b
8.600
1.090
2.970.000
b
10. 150
330
Gross radioactivity of dry
undissolved solids. >*^ic, g
Alpha
3.640
4.050
15.300
1, 180
770
22.600
27.700
16.000
14.700
14,300
291.000
42. 100
690
16,800
13,300
33.000
329,000
70,500
3, 190
Beta
4.530
5.670
21,600
1.330
1. 150
35.000
33,200
16,300
13, 100
13,400
451,000
21.800
808
16,800
9.750
18.000
436, 000
168. 000
3.730
Single cycle.
jj
Solid sample.
the sample before being assayed for radioactivity.
Radium-226 analyses were performed by a private laboratory
after sample preparation at the Taft Center. Solids were separated
from the liquid by means of a membrane filter and then ground t-~>
less than 100 mesh. The solids and liquid portions were then ana-
lyzed; results were reported in terms of/i/ig/1 of liquid sample and
/*Ag/g of solid sample. These results were then converted to the
units presented in Table 6. As in the case of the alpha activity,
the sand-slime separation apparently tends to concentrate the un-
dissolved radium in the slimes. The effect of Pond No. 1 is also
evident from Table 6. Here, the undissolved radium is settled,
going from 1370 to 142npg/l of slurry, while the dissolved activ-
ity increases from 125 to 475^/ig/l of slurry. This is not as evi-
dent in Pond No. 3 where river water is returned from the Baker
cooler.
-------
CLIMAX URANIUM COMPANY
Table 5. ALPHA ACTIVITY IN PROCESS STREAMS
49
Station
1
22
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Undissolved
1,778
1.790
1.653
423
272
30
7
1.778
1.626
1 257
10
2
5
1.477
1.066
4
599
1
< 1
Dissolved
< 1
4
6
< 1
1
1
9
a
a
4
6
4
a
1
< 1
402
a
13
< I
Total
1.778
1.794
1.659
423
273
31
16
1.778
1.626
1.251
16
6
5
1.478
1.066
406
599
14
< 1
The radium concentrations in Table 6 have been combined with
the flows in Table 3 (by cycle) to estimate the quantity of radium-
226 in the process stream (Table 7 and Figure 4). Determination
of this radium distribution was one of the main objectives of the
mill survey. In balancing the radium input against output at the
locations indicated previously, we see that the 256 me/day at
Station 2, the conditioner tank effluent, compare favorably with
the 270 (59 + 4 + 207) me/day at Station 4. the sand to the acid
leach; Station 6, the influent to Pond No. 1; and Station 8. the
roaster feed, the 219 me/day of radium in the roaster calcines.
Station 9, are accounted for in the 184 (2.1 + 182) me/day at
Station 13, the vanadium product, and Station 14, the filter cake
from the Oliver filter. The 182 me/day at Station 14 are in
agreement with the 176 me/day at Station 15. the slime leach
tails. The over-all plant balance equates 331 me/day at Station 1
to 232 (50 + 4 + 2 + 176) me/day, leaving the plant at Stations 5,
6, 13, and 15, respectively. An examination of the results from
Stations 1 and 2 in Tables 6 and 7, and consideration of the man-
ner in which Station 2 balances when compared with other stations
-------
50 ACID LEACH - SOLVENT EXTRACTION PROCESS
Table 6. RADIUM - 226 CONCENTRATIONS.
Station
1
2
3
4
5
6
7
8
9
10
11
12a
13
14
15
16
17
18a
19
Concentration in slurry, Itpg/'l
Undissolved
625,000
168,000
75.800
194,000a
112.000
1.370
142
b
b
305.000
127
7.3
b
963.000
739.000
600
b
35
645
Dissolved
187
660
345
72a
138
125
475
b
b
4.950
56
5.8
b
2.710
65
1.250
b
490
74
Concentration in dry
undissolved solids,
Wg/g
680
575
2,100
165
140
2,450
1,200
1,850
1,950
2,550
1,950
120
300
2.050
2.200
1,300
26
550
690
a Single cycle.
b Solid sample
tend to indicate that the laboratory analysis for the undissolved
radium was higher than would normally be expected at Station 1.
On the basis of these comparisons, the radium-226 passing
Station 1 should more probably be about 240 me/day in order to
produce more over-all agreement throughout the process.
Chemical analyses were performed on several of the effluent
samples collected during the survey; Table 8 presents these re-
sults. No reason is apparent for the large variations between the
analyses from Cycles I and n at Station 19.
Chemical utilization was reported as follows:
HpSO. - 105 Ib. per ton of sand, for sand leach
- 250 Ib. per ton of slime, uranium extraction
- 400 Ib. per tank, vanadium precipitation
- 800 Ib. per tank, uranium precipitation
-------
CLIMAX URANIUM COMPANY
51
NH,
NaC103
WASTE DISPOSAL
- 548,000 Ib. per month
- 3.5 Ib. per ton of sands
Results of the survey indicated that Radium-226 was entering
the river at Stations 18 and 19 at rates of approximately 0.16 and
0.005 mg/day, respectively. Of this quantity 0.11 mg was sus-
pended radium and the remainder, dissolved. These quantities may
have since decreased, due to tailings ponds rearrangement pre-
viously mentioned. A survey of the Colorado River in the vicinity
of Grand Junction was conducted during August 1960 for the Colo-
rado River Basin Water Quality Control Project. *4 At that time
samples of river water and sediment were collected about 0.3
miles above the Climax Mill (Station G-l at the lower end of a
diversion channel at the mill (Station G-M), and below this di-
version channel, about one-fourth mile above the mouth of the
Gunnison River (Station G-2). The results of this survey are pre-
sented in Table 9. Each of the two consecutive cycles was of
Table 7. RADIUM - 226 IN PROCESS STREAM
Station
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Racium - 226 mg/day
Undissolved
331
255
230
59
50
3.5
0.32
207
219
220
0.06
0.01
2.1
181
176
0.16
0.05
0.11
0.01
Dissolved
0.10
0.92
1.07
0.02
0.06
0.33
1.05
a
a
3.5
0.02
< 0.01
a
0.50
0.01
0.33
a
1.45
< 0.01
Total
331
256
231
53
50
3.8
1.4
207
219
224
O.OB
0.01
2.1
162
176
0.49
0.05
1.6
<0.01
Solid sample.
-------
52
ACID LEACH - SOLVENT EXTRACTION PROCESS
KEY: UK3ISSOLVED - 3CC
-------
CLIMAX URANIUM COMPANY
Table 8. CHEMICAL ANALYSES OF EFFLUENT SAMPLES
53
Determination
Total acidity, as CaCO,
Mineral acidity as CaCOj
Salfate
Hardness, as CaCOj
Chloride
Sodium
Fluoride
Vanadium
Arsenic
PH
Concentration mg/1 station - Cycle
12-1
500
250
2.400
2.500
1.340
1.800
12
.03
0
2.7
15-1
250
20
1.300
2.100
140
135
13
.03
0.2
3.2
19-1
310
10
7.600
3.600
2.280
1,350
13
.OS
3
3.8
i9-n
150
0
1.300
1,300
415
420
6
.007
0
6.9
would be from seepage, pond overflow if this should occur and
spills due to washout of pond walls. The latter is not very likely
although not without precedent in the industry.
The increase in the alpha activity activity and radium and
uranium concentrations at Station G-M is clearly detectable. In
terms of the maximum permissible concentrations (MFC) of radio-
nuclides in water outside of a controlled area, as specified by
NBS Handbook 69 ^ the uranium concentration is well below the
allowable 20 mg/1 while the radium-226 is definitely above the
allowable 3.3/i^c/l; however, due to the limited access to the
diversion channel before reaching the main body of the river
where the radium concentration is below the MFC, this cannot
be considered of important public health significance.
Table 9. ANALYSES OF RADIOACTIVITY IN MUD AND WATER SAMPLES
OF THE COLORADO RIVER - SURVEY RESULTS. AUGUST I960 a
G-] Cycle I
Cycle II
G-M Cycle I
Cycle II
G-2 Cycle I
Cycle II
Mud Sa.rr.ple b
Alpha
activitv.
we f
14.6
26.4
386
653
249
343
Beta
activity
Itlic e
39.6
58.9
478
615
509
546
Ra-226
A^c g
3.2
3.4
13
19
3.7
4.8
Water Sarr.ple l
Alpha
activity
W*c 1
4.4
10.7
7 0
Beta
activity.
,11/lC 1
13.1
-
31.6
Ra-226
we 1
0.3
1.6
3.9
4.8
1.1
1.5
Uranium
MS 1
12
14
44
35
22
21
-------
54 ACID LEACH - SOLVENT EXTRACTION PROCESS
Acknowledgment
The generous cooperation and assistance of the following are
gratefully acknowledged: Ralph Musgrave, Chief Metallurgist;
and other personnel of the Climax Uranium Company; the U. S.
Atomic Energy Commission; S. May, Colorado Department of
Public Health; and G. Harlow and E. A. Pash, U. S. Public Health
Service.
-------
THE CARBONATE LEACH URANIUM EXTRACTION
PROCESS
L HOMESTAKE-NEW MEXICO PARTNERS COMPANY,
GRANTS, NEW MEXICO
J. B. Cohen*
H. R. Pahren
M. W. Lammering
Introduction
This report presents the results of an inplant survey of a uran-
ium refinery that utili/es the carbonate leach extraction process for
the separation of uranium from its ore. The study was performed
by the Public Health Service during September 1959 with the co-
operation of the New Mexico Department of Public Health, the
Homestake-New Mexico Partners Company, and the U. S. Atomic
Energy Commission.
The Homestake-New Mexico mill is located about 10 miles
northeast of Grants, New Mexico. Operation began during 1958.
The mill is rated at 750 tons per day: although between Septem-
ber 22 and September 28, 1959, the dates of the survey, about
900 tons of ore per day were processed. The ore assayed from
0.173 to 0.195 per cent U.,0 and yeilded about 4,000 pounds per
day of yellowcake. Wastes Trom the mill were discharged to a
tailings pond, and the liquid portion not lost by seepage and eva-
poration was recycled for use as process water. There was no
surface water in the vicinity likely to receive any of these wastes.
Figure 1 is a flow diagram of the process, which is descri-
bed in detail in the following sections.
Process Description
ORE PREPARATION
Ore is brought from the nearby uranium mines by truck and
stored outdoors until used. The ore is transferred by means of a
bulldozer to a feed hopper, from which the ore is conveyed to a
jaw crusher. Here the lumps are crushed to less than 3/4 inch.
A 10 per cent sample of the crushed ore is obtained by moving
buckets as they pass through the stream of ore falling from a con-
veyor into a hopper. A second 10 per cent sample is obtained from
the original 10 per cent, cut in a. similar manner, to give a one
per cent portion of the original ore. The stream representing one
per cent of the ore is further crushed and then impinges on a ro-
*Sanitary Engineer, Radiological Pollution Activities Unit: Sanitary
Engineer. Colorado River Basin vVater Quality Control Project.
Public Health Service, Denver. Colorado: and Senior Assistant
Sanitary Engineer. Radiological Pollution Activities Unit.
55
-------
56
CARBONATE LEACH PROCESS
Q S»MPLING POINTS
•SS"
Figure 1. Flow diagram of Homestake-New Mexico Partners uranium
mill, Grants, New Mexico, September 1959.
tating disk containing slots. The number of open slots determines
whether the disk takes a 2.5, 5.0, 7.5, or a 10 per cent sample of
the ore stream. Thus, the final sample may contain from 0.025
to 0.10 per cent of the ore passing through the plant. The remain-
ing ore is stored in a fine ore storage tank of 3, 000 ton capacity.
GRINDING
After storage there are parallel circuits for the grinding,
classification and cyclone separator steps. The remainder of the
process is essentially a series circuit.
Ore is conveyed from the fine ore storage bin to a ball mill
where it is ground to less than 65 mesh. Carbonate solutions will
-------
HOMESTAKE - NEW MEXICO PARTNERS COMPANY 57
not react with many ore components other than the uranium mine-
rals, and it is necessary to grind the ore to this size to provide
the necessary amount of surface area for efficient leaching. The
necessary particle size is determined by the type of ore processed.
Limestone ores in which the uranium is finely dispersed throughout
the matrix must be ground finer than sandstone type ores in which
the uranium material is part of the bond between the sand grains.
Mill solution is added to the ball mill with the ore. A spiral
classifier separates any oversize particles and returns them to
the ball mill. The specific gravity of the slurry from the classi-
fier is automatically controlled at 1.20 by means of an automatic
valve. The flow of additional mill solution added to the classifier
is either increased or decreased, depending on the amount neces-
sary to maintain the desired specific gravity in the classifier
overflow. This effluent is pumped to a small cyclone separator
where the coarse solids are separated from the slime.
The cyclone overflow is sent to a 75-foot thickener where the
solids are further concentrated. About 20 pounds per day of Sepa-
ran are added here to aid in solids separation. Overflow from the
thickener is recycled to the mill solution storage tank. The thick-
ener underflow is combined with the underflow from the two cy-
clones in an agitated tank. Combined effluent from the agitated
tank flows to a sump, from which it is pumped to the leaching
section.
LEACHING
Leaching of the uranium from the solid particles is accom-
plished with a sodium carbonate-bicarbonate liquor in six Pachuca
tanks, 19 feet in diameter and 48 feet high, operating in series.
There is a 7-hour retention time in each tank or 42 hours total
retention in the leaching circuit. The Pachuca tanks are operated
at 179°F. and at atmospheric pressure.
The insoluble quadrivalent uranium must be converted to the
soluble hexavalent form. Soluble uranyl tricarbonate then forms
in the carbonate solution, under the leaching conditions. The solu-
bilization of the hexavalent form, such as the uranium mineral
carnotite in order to produce the uranyl tricarbonate ion may be
represented as follows: ^- 17
+ H00
z
The Pachuca tanks are aerated to provide for the oxidation of
reduced uranium compounds. Copper sulfate and ammonia are nor-
mally added to the ore slurry to catalyze the oxidation reaction.
During the period of the survey however, a test run was made
with cyanide instead of the copper sulfate and ammonia. Crude
cyanide that contained about 50 per cent NaCN equivalent was
-------
58 CARBONATE LEACH PROCESS
added to the ore at the rate of 0.8 pounds per ton of ore as it was
conveyed to the ball mills. The cyanide reacts with the iron balls
in the ball mills to form the complex ferricyanide ion, a strong
oxidizing agent for uranium.
PREGNANT LIQUOR PREPARATION
The slurry from the final leaching tank flows to the first-
stage rotary filters, where the solids are separated from the
liquid containing the uranium or the pregnant solution. There are
three filters in each stage of filtration. The filter cake is washed
with recarbonated barren solution and this wash water becomes a
part of the filtrate.
The pregnant solution is then pumped to an aerated flotation
tank where any hydrocarbons in the solution are removed. If these
hydrocarbons are not removed they interfere with the next step by
preventing the complete precipitation of uranium with the caustic.
A small amount of organic chemical is continuously added to pro-
mote frothing and to aid in removing chemical is continuously adde:
to promote frothing and to aid in removing the hydrocarbons. The
froth overflows to the floor sump, from which it goes to the pri-
mary thickener. The flotation tank underflow is sent to a second
75-foot settling tank for further clarification. Overflow from this
settling tank flows to the pregnant liquor storage tank while the
underflow is pumped to the primary thickener.
PRECIPITATION AND PREPARATION OF YELLOWCAKE
Pregnant solution is pumped to the first of three 20-foot dia-
meter precipitation tanks operating in series. A 50 per cent
sodium hydroxide solution is added at a rate of about 5,000
pounds per day to precipitate the sodium diuranate yellowcake.
The reaction is as follows:
3 + 6NaOH—*Na2U207+ 6Na2C03 + 3H20
The precipitated yellowcake is concentrated in a 12-foot diameter
thickener. Underflow from the thickener passes through a Burwell
filter press that removes the solids. After the yellowcake is re-
moved from the filter, it is dried in a gas fired drier, ground in
a hammermill, and drummed for shipment to the U. S. Atomic
Energy Commission.
Overflow from the yellowcake thickener and the filtrate from
the Burwell presses are combined. This barren solution is padded
through a Sperry filter press to remove any remaining yellowcake
particles before it is pumped to the recarbonation tower where the
boiler plant flue gas is passed countercurrent to the barren solu-
tion. Carbon dioxide in the flue gas neutralizes the excess caustic
alkalinity thus forming additional carbonate and bicarbonate. The
recarbonated barren solution is then reintroduced into the process
at the first stage filters, following the leach process.
GFO 813—17'-
-------
HOMESTAKE - NEW MEXICO PARTNERS COMPANY 59
FILTRATION OF TAILINGS
The solids that are separated from the pie gnant solution by the
first stage filters are washed by a portion of the recarbonated
barren solution. The remaining recarbonated solution is used to
repulp the first-stage filter cake. This repulped slurry is then
filtered by the second stage filters. The water used to wash the
filter cake from the second stage filters is return water decanted
from the tailings pond. Filtrate from the second stage filters is
returned to the primary thickener. About 50 pounds per day of a
flocculating agent, guar gum solution, is added to the filter feed
to aid in filtration.
The filter cake from the second stage filters is repulped so
that it may be pumped to the tailings pond. Part of the repulping
water consists of fresh water from the plant well. The balance of
the water is return water from the tailings pond.
Normally the water returned from the tailings pond passes
through an ion exchange system to recover any dissolved uranium,
and a flotation unit to remove any hydrocarbons before the water
is used. During the period of the survey, however, the ion ex-
change system was not in operation and the return tailings water
by-passed this step. The flotation unit, however, remained in
operation. A gas flame was used to burn off any hydrocarbons in
the froth from the flotation unit, before the froth flowed to the
floor sump and the primary thickener.
TAILINGS POND
The tailings pond contained approximately 40 acres within the
dikes, of which about 10 acres were occupied by clear water and
about 10 acres by moist tailings. The remaining area was dry. At
one end, an area is partially separated from the remainder of the
pond. Water overflows from the main portion of the pond into this
area, and from here is returned to the mill process.
The Mill Survey
For the purpose of analyzing the individual components that
make up the entire extraction process, samples were obtained
during two sampling cycles of 72 hours each. The sampling per-
iods were as follows:
Cycle 1-7 AM. Sept. 22, 1959. to 7 AM, Sept. 25, 1959.
Cycle 2-7 AM, Sept. 25, 1959. to 7 AM, Sept. 28, 1959.
Table 1 gives a list of the sampling stations, together with a
brief description of each..
A sample of the ore being fed to each ball mill was collected
by plant personnel every hour and composited over a 24-hour
-------
60
CARBONATE LEACH PROCESS
Table 1. SAMPLING STATIONS
Station number
1
2
34
5
6
9
10
11
12
13
14
IS
17
Description
Raw ore feed to ball mill
Mill solution
Combined overflow from classifiers
Slurry to leach taris
Leach tank effluent
Waste to tailings pond
Pregnant filtrate from first stage filters
Pregnant solution to precipitation tanks
Yellowcake product
Barren solution from Burwell Press
Filtrate from second stage filters
Raw well water
Return water from tailings pond
Recarbonated barren solution
Combined barren solution to Sperry Press
period. The samples from the two ball mills were combined, pul-
verized, blended and resampled. Portions of the daily composite
samples were weighted according to tonnage of ore fed to process
and combined for each cycle (Station 1).
Equal portions were composited from each drum of yellow-
cake packaged during each three-day sampling period to give the
yellowcake sample representative of Station 10.
Every 2 hours, samples were collected at both classifier
overflows (Stations 3 and 4); combined slurries were pumped to the
leach tanks (Station 5), and from the final leach tank (Station 6).
These samples were composited for each cycle.
During Cycle 1 samples were collected from the drinking
water fountain every two hours and composited (the plant raw wa-
ter sample, Station 13). For the second cycle, one grab sample
of water was obtained to determine if there was any difference in
the two methods of collection, such as possible contamination of
the composite sample by plant dust.
All other plant samples were collected by means of automatic
samplers, and daily samples were composited for each 3-day
period of the cycle.
The samples from this survey were sent to the Public Health
Service's Robert A. Taft Sanitary Engineering Center, at Cincinnat:
Ohio, where all physical, chemical, and radiological determination
except radium analyses, were conducted. Duplicate portions of the
liquid and solid samples were sent to a private laboratory for
determination of dissolved and undissolved radium.
-------
HOMESTAKE - NEW MEXICO PARTNERS COMPANY
61
Analyses and Results
The results presented here are a combination of field data,
laboratory analyses and flow balances; together they present an
over-all picture of the units composing the carbonate-leach pro-
cess. With physical data from the field and laboratory, a flow
balance for the mill was obtained. This was done through the use
of a solids balance and a total weight balance between the various
mill process units. These results enabled the preparation of Fig-
ure 2, a schematic flow diagram giving the estimated slurry flow,
separated in terms of solids and liquid, at each of the sampling
points. Table 2 presents the physical characteristics of the process
stream at the various stations. The data, including all balances,
were measured or computed separately for each cycle and then
averaged because of their close agreement. As previously men-
tioned, the 42-hour time lag between Stations 5 and 6 (the leach-
ing process) tends to make a mill balance difficult, but averaging
the two 72-hour cycles minimized this problem.
In Table 2 the data in colums (3) through (6) are laboratory
determinations, while the slurry flows in column (2) were Calcu-
lable 2. PROCESS STREAM CHARACTERISTICS a
Station
(1)
1
2
34
5
6
7
8
9
10
11
12
13
14
16
17
Slurry flow.
cpnl
h)
b
508
580
183
178
285
117
80
b
d
73
70
150
82
81
Specific
gravity cf
slurrv
(3) "
-
1.09
1.23
1.48
1.54
1.32
1.10
1. 10
-
1.10
1.07
1.00
1.01
1.03
1. 10
Dry suspended
solids by
weicht. ^~
(4)
94.1
c
25.4
54.6
52. B
39.1
c
c
99.4
c
c
c
c
c
c
Specific
gravity of
dry solids
' (5)
-
-
2.64
2.59
2.66
2.72
-
-
-
-
-
-
-
-
-
PH
(6)
-
10.1
10.1
10. 1
10.0
9.6
10.1
10.1
-
12.0
10.2
-
9.8
10.3
12.0
Average of cycles 1 and 2
Solid sample.
Liquid (neirlitrible suspended solids).
Flows not calculated.
-------
62
CARBONATE LEACH PROCESS
lated. Although the fresh water flow, used for repulping the final
tails, at Station 13 was calculated as 70 gpm, additional fresh
water entered the process at other points, including water for
chemical feed preparation, water used for plant housekeeping, and
makeup water needed because of evaporation and other losses.
Table 3 presents the average solids flow (average of cycles
1 and 2). Because of the selectivity of the leaching process for
the uranium compounds and the low concentration of uranium in
the ore, there was little detectable change through the mill as re-
gards tonnages of undissolved ore solids.
Table 3. SOLIDS IN PROCESS STREAMS
Station
1
2
34
5
6
7
8
9
10
12
13
14
16
17
Tons per day
Suspended
331
0.8
834
633
663
633
0.6
0.2
2.0
0.9
"0
-0
0.2
0.1
Dissolved
-
302
299
83
111
12
64
57
-
35
0.5
12
47
55
Total
881
303
1183
9?6
979
895
85
57
2.0
36
0.5
12
47
55
Figure 2 shows that there are locations throughout the mill
where quantities entering and leaving sections of the process can
be directly balanced. The mill solution. Station 2, and the ore
from storage. Station 1, entering the ball mill and classifier
should be quantitatively accounted for at the exit from the classi-
fiers. Station 34. Quantities present in the pregnant liquor at
Station 9 should be measurable in the yellowcake product at Sta-
tion 10 and in the recarbonated barren solution at Station 16. For
the over-all plant balance, the ore entering at Station 1, the fresh
water at Station 13, and makeup water at Station 14 should approxi-
mate the yellowcake produced at Station 10 and the slurry flow to
the tailings pond at Station 7. A balance at the primary thickeners
would equate the input from the classifiers at Station 34 and the
primary and secondary filtrates at Stations 8 and 12 to the output to
-------
HOMESTAKE - NEW MEXICO PARTNERS COMPANY
63
Figure 2. Schematic flow diagram of Homestake-New Mexico Partners
mill, September 1959.
leaching at Station 5, the mill solution at Station 2 and the preg
nant liquor storage at Station 9 respectively.
Results for the stations mentioned above show that the
solids flows of Table 3 are fairly good. At the ball mill and
classifier, 1184 (881 + 303) tons of solids per day enter as raw
ore (Station 1) and mill solution (Station 2) compared to the 1183
tons per day leaving these units. In the yellowcake precipitation
stage of the process, 57 tons per day of solids enter in the pregnant
liquor (Station 9) compared to the 49 (2 + 47) tons per day leaving as
rellowcake (Station 10) and recarbonated barren solution (Station 16). In
terms of theover-all mill balance 894 (881 + 12 + 0.5) tons per day
enter the plant as raw ore (Station 1), return water (Station 14)
and well water (Station 13) compared to the 897 (895 + 2) leaving
the plant via the tailings pond (Station 7) and as yellowcake pro-
duct (Station 10). A balance at the thickeners shows 1304 (1183 +
85 * 36) tons per day entering from the classifiers (Station 34)
and the primary and secondary filters (Stations 8 and 12). This is
in good agreement with the 1326 (966 + 57 +303) tons comprising
-------
CARBONATE LEACH PROCESS
the slurry to leach tanks (Station 5), the pregnant liquor (Station 9)
and the mill solution (Station 2) in that order.
The gross radioactivity concentrations determined in the lab-
oratory are listed in Table 4. These data demonstrate how the
leaching process affects the activity of the sample. Between
Stations 5 and 6 (the leaching process) the undissolved activity
decreases while the dissolved activity increases. The effect of
particle size is also shown in this table. At those stations where
the large solids had already been filtered and only the fine par-
ticles remained, the activity was several times greater than that
of the unfiltered solids when considered in terms of ppc/g of dry
undissolved solids.
Table 4. GROSS RADIOACTIVITY CONCENTRATION'S
Station
1
2
34
5
6
7
8
9
10
11
12
13
14
16
17
Total Sample (Slurry) Radioactivity, ppc 1
Undissolved
Alpha
2.580
662.000
2.300.000
2.200,000
1. 100.000
4,830
3. 630
-
b
16. 500
b
b
4.420
25.200
Beta
-
5.100
676.100
2,810.000
2.660.000
l.OBO. 000
9.940
7.690
-
b
26. 500
-
-
8,430
66.100
Dissolved
Alpha
i
393,000
456.000
386.000
554.000
5.720
902.000
637.000
a
12.600
41.400
2.6
11.600
27.700
5.430
Beta
d
I, 110.000
1. 190.000
1.030.000
1.760.000
17,500
2.120.000
1,730.000
•\
28,400
122.000
41
41.600
98.400
23.400
Radioactivitv in dry
undissolved solids, ^/*c-g
Alpha
3.300
9.910
2.630
2.870
2.700
2.150
7.780
10.500
164.000
-
8. EDO
-
-
10.200
112.000
Beta
3.300
19.400
2.660
3.520
3.230
2.100
15,700
21,400
374,000
-
14,200
-
-
19.300
281,000
a Solid Sample.
b Liquid Sample
Gross alpha quantities are presented in Table 5. The flows
computed in Table 2 have been combined with the assay of Table 4
to trace the alpha activity through the various mill processes.
The figures for the alpha balance are a function of flow and as
such would be expected to vary with production.
The balance at the ball mill was good with 3727 (2630-i- 1097)
me/day entering as raw ore (Station 1) and mill solution (Station 2)
compared to the 3500 me/day at Station 34. The effect of the
leaching process is illustrated between Stations 5 and 6 where the
-------
HOMESTAKE - NEW MEXICO PARTNERS COMPANY
65
Table 5. ALPHA ACTIVITY IN PROCESS STREAMS
Station
1
2
34
5
6
7
8
9
20
12
13
14
16
17
Gross Alpha Radioactivity, me, day
Undissolved
2.630
7
2.060
2.310
2. 140
1.730
3
2
300
-
-
2
12
Dissolved
-
1 090
1.440
450
550
11
600
273
-
17
- 0
9
12
2
Total
2.630
1.097
3.500
2,760
2.690
1.741
603
280
300
24
0
9
14
14
dissolved alpha activity increased as the undissolved activity en-
tered solution. A comparison of these two stations shows 2760
me/day entering the leaching stage and 2690 me/day leaving at
Station 6. Table 5 also shows that 280 me/day entered the uran-
ium extraction stage of the process while 314 (300 + 14) me/day
were accounted for in the yellowcake product and the barren solu
tion. A total of 4127 (3500 + 603 + 24) me/day entered the thick-
ening stages from the classifier (Station 34), and the filters
(Stations 8 and 12), which is in excellent agreement with the 4137
(2760 + 1097 + 280) me/day leaving for the leaching process(Sta-
tion 5) as mill solution (Station 2) and as the pregnant liquor
(Station 9).
The over-all plant balance required further study since the
2639 (2630 + 9) me/day entering with the raw ore and the return
water was in relatively poor agreement with the 2041 (1741 •*•
300) me/day in the slurry to the tailings pond (Station 7) and in the
yellowcake product (Station 10). A close investigation of the pro-
cedures used in collecting and analyzing the samples revealed that
the discrepance could have been due to the manner in which the
samples are collected at Station 7. At this station a portion of the
flow is collected in a large drum. To obtain a sample for analysis
the contents are mixed and the tank is completely drained from
below. A sample bottle is passed through the stream every few
seconds until the drum is emptied. Because the heavier particles
would tend to settle first at a higher flow, the sample would con-
tain a disproportionately greater quantity of large solids than the
-------
66
CARBONATE LEACH PROCESS
Table 6. RADIUM - 226 CONCENTRATIONS
Station
1
2
34
5
6
7
8
9
10
11
12
13
14
16
17
Radium - 226 in slurry ,u>ig, I
Undissolved
a
890
124.000
427.000
463.000
225.000
1.410
270
a
2.800
7.900
b
b
170
530
Dissolved
a
8, 160
9.820
5.620
7.830
100
19,600
15.200
a
2.4
2.400
0.4
160
990
5.4
Radium - 226 in dry
undissolved solids.
WS £
530
3.450
490
540
570
490
2.150
740
3.490
26.400
4.250
-
-
390
2.260
Solid Sample.
b .Vecligible Solids Content.
actual flow. The larger particles usually contain less activity per
gram than the smaller particles, and we would, therefore, expect
to find a discrepancy of the type involved. This was evident pre-
viously in Table 4. Thus, while the activity at Station 7 was cal-
culated at 1741 me/day, it was more probably in the neighborhood
of 2200 me/day. Table 5 also illustrates that the majority of the
alpha activity remained undissolved through the process, and was
discharged to the tailings pond with the solids.
Samples to be analyzed for Ra-226 were prepared at the Taft
Center and forwarded to a private laboratory for these determina-
tions. Preparation entailed separating the solids from the liquid
with membrane filters, and grinding the solids portion to less than
100-mesh. The undissolved activity is in the solids remaining on
the membrane filter: the dissolved activity is in the liquid portion
of the sample passing through the filter. Table 6 presents the
results of the laboratory analyses for radium per liter of slurry
and per gram of dry undissolved solids. The finer particles ap-
parently have a greater activity than the coarser particles.
When the laboratory results of Table 6 are combined with the
flow balance of Table 2 a radium balance throughout the mill may
be computed. This is shown in Table 7 and Figure 3. A balance
-------
HOMESTAKE - NEW MEXICO PARTNERS COMPANY
67
OPE
STOCK
PILF
CffUSHIKG
AND
SAMPLING
Figure 3. Schematic flow diagram for Radium-226 at the Homestake-
New Mexico Partners miJI, September 1959.
Table 7. RADIUM - 226 IN PROCESS STREAM
Statira
1
2
34
5
fi
7
S
9
10
12
13
14
16
17
Ridi'-m-226 rng day
Undissclvcd
420
2.5
394
432
449
395
0.90
0.12
6.4
3.2
-
-
0.08
0.23
Dissolved
a
22.5
31
6.0
7.7
0.16
12.5
6.6
a
- 1.0
0
0.18
0.44
- 0
Total
423
25
425
438
457
335
13
6,7
6.4
4.2
- 0
0.18
0.52
0.23
S.-.lid San-pie
-------
68 CARBONATE LEACH PROCESS
Table 8. CHEMICAL CONSUMPTION
Chemical
Cyanide (50^ CN~)
Caustic (100^ NaOH)
Soda Ash
Jaguar
Separan
Pounds added per ton
of ore processed
0.82
2.84
2.65
0.052
0.023
was calculated for each cycle and then averaged at each station.
Thus the figures represent the quantities of radium in the process
while the mill was processing an average of 880 tons of ore per day.
Table 7 and Figure 3 indicate the radium balance between
stations is good in most cases. The 445 (420 + 25) mg/day of
radium entering the ball mill with the raw ore and the mill solu-
tion agrees well with the 425 mg/day at Station 34. The uranium
precipitation stage only contains a small amount of Ra 226- the
6.7 mg/day entering at Station 9 are accounted for as 6.9 (6.4 +
0.5) mg/day in the yellowcake and the recarbonated barren solu-
tion. The 457 mg/day of Ra 22^ leaving the leaching stage at Sta-
tion 6 is slightly higher than expected though still in line with the
other stations having high solids content. A total of 442 (425 + 13
-t- 4.2) mg/day enter the thickener units from the classifiers and
the filters as compared to the 470 (438 + 25 -t- 6.7) mg/day leaving
these units for the leaching stage (Station 5) as mill solution
(Station 2) and pregnant liquor solution (Station 9). The over-all
plant balance shows 420 mg/day in the raw ore (Station 1) as a-
gainst 402 (396 -1-6.4) mg/day in the slurry to tails (Station 7) and
the yellowcake product (Station 10). Because the Ra 22^ at Stations
13 and 14 is negligible for balancing purposes, again it would
appear that Station 7 has less activity than would actually be ex-
pected. In all cases mentioned above the discrepancies in the
radium balance were less than 10 per cent of the quantity being
considered.
Most of the radium remains undissolved through the mill, al-
though dissolved radium concentrations build up at several process
locations, due to recycling of mill solution. If the ore entering
the process at Station 1 is considered to be in radioactive equili-
brium, the amount of radium entering with the ore each day can
be computed from the assay for UgO On the basis of the average
assay of 0.186 per cent V^o during the survey period, the rad-
ium entering was calculatedas 420 me/day, in exact agreement
with the value calculated from the radium analysis. On this same
basis, if it could be assumed that all of the daughters of U 238 are
in radioactive equilibrium with the parent isotope, we would expect
the gross alpha activity entering the process to be about 3, 300 me
day. Radon-222t however, is a gas; hence a portion of it will be
lost to the atmosphere in the ore body during mining and handling
-------
HOMESTAKE - NEW MEXICO PARTNERS COMPANY 69
and milling, thus partially breaking the chain. As a result, the
observed value of 2,630 me/day in Table 5 is within the range
that would be expected.
Of the 420 mg/day of Ra ^26 entering the mill in the raw ore,
6.4 mg/day, or 1.5 per cent, leave with the yellowcake. This is
substantially more than has been found in mills using the acid
leach process.
Chemicals added to the process at the time of the survey are
listed in Table 8. On a yearly basis, the plant uses caustic at a
rate of 17 to 22 pounds per ton of ore processed and soda ash at
a rate of 0 to 5 pounds per ton, with an average of 20 pounds and
1 pound per ton. respectively. In addition, the use in the recar-
bonization tower of carbon dioxide from the flue gas must be
recognized as part of the chemical consumption. Analysis of the
filtrate at Station 7 provided the following data in terms of mg/1
of filtrate:
Arsenic - 0.20 mg, I
Chlorjda - 275 mg 1 Cl"
Phenol Aklalinity - 900 mg 1 as CaCO.
Total Alkalinity - 4050 rcg; 1 as CaCOj
Sodium - 2950 rr.g/1 Na+
pH - 9.6
These figures, with the exception of pH, could be expressed
in terms of slurry flow by reducing the concentrations by 20 per
cent.
Waste Disposal
As the tabular data indicate, 285 gpm of slurry entered the
tailings ponds during the survey period. Of this, 54 gpm were
undissolved solids and the remaining 231 gpm were liquid. The
solid portion contributed an estimated 396 milligrams of Ra 226
per day to the pond and the liquid phase only 0.16 mg/day. The
return water from the pond (Station 14) was slightly higher in
Ra-226 concentration, probably due to leaching and evaporation
in the pond. As the flow diagrams indicate, the liquid returned at
Station 7 is mainly a combination of repulping water from Station
14 and fresh water from Station 13.
The wastes are contained within the tailings pond and there
are no surface waters within the vicinity: hence, surface water
pollution is not a problem. The possibility of ground water pollu-
tion, however, cannot be ignored.
A number of prixrate wells are in use in the area and ground
water is used for domestic consumption, for livestock, and for ir-
rigation. In addition, the towns of Milan and Grants, a few miles
south of the mills, have municipal supply wells.
Three aquifers are present in the area: Silt, sand and gravel
of the alluvium: interbedded clay, siltstone and sandstone beds in
-------
70 CARBONATE LEACH PROCESS j
the Chinle formation: and the San Andres limestone. Domestic
and stock wells usually obtain sufficient water from the alluvium or
from the Chinle formation. Water levels in the general area indi-
cate that the water in the alluvium occurs under water table condi-
tions and moves southwestward from the mill sites into the valley
of the Rio San Jose. It then moves down the valley in a southeast-
ward direction. The alluvium is in contact with the San Andres
limestone along the southwest edge of the valley, west of Milan.
and at these points water in the alluvium can enter the limestone
formation. Water in the San Andres limestone is under artesian
pressure and moves in a general easterly direction from the mills.
The level and movement of water in the Chinle formation has not
been determined definitely. At the mills the depth to water in the
alluvium is thought to be about 80 feet. The depth to the San Andres
limestone is 600 feet but the artesian pressure raises the water
level in this formation to 130 feet below the land surface. ^
Samples from test wells in the vicinity of the mill were analy- ;
zed for radium-226. The results (0.8, 1.8, 0.7, 4.5. 9.5, and 3.1 <
ftpc/l) are shown on Figure 4, a map of the area. Radioassays •
on the solids in the samples from the north and the east test wells <
indicated radium 226 concentrations of 551 and 1, 685/i/ig/g of j
solids, respectively. These concentrations are higher than in the j
ore feed to the mill. Additional samples were analyzed from wells j
several miles to the west of the mills and from wells between i
Grants and San Rafael. Results of these analyses indicate a radiuir
content of from 0.1 to 0.4/i/zc/l. or the usually expected natural
concentrations. •
If monitoring of the test wells at the mill should indicate a
buildup or spread of the radium-226 consideration may have to be
given to ways and means of sealing the ponds and preventing fur-
ther buildup. This is suggested in view of the facts that the radiur
concentration of the discharge to the pond (Station 7) was measured
as 124/i^ug/l of liquid and, at the time the wells were sampled, the
mill had been in operation less than two years.
Acknowledgment
The generous cooperation and technical assistance of the fol-
lowing are gratefully acknowledged: John Hernandez, New Mexico
Department of Public Health: personnel of the Homestake-New
Mexico Partners Mill: the U. S. Atomic Energy Commission:
Charles E. Sponagle. Public Health Service, Region VIII, Denver.
Colorado: E. A. Pash, Carl Hirth, Carl Shadix. and H. D. Nash.
Public Health Service, Cincinnati, Ohio: and D. E. Rushing. Publit
Health Service. Salt Lake City, Utah. The technical advice and
guidance of Dr. E. C. Tsivoglou. Public Health Service, Cincinnati.
Ohio, was greatly appreciated. This study was supported by funds
made available through the Environmental and Sanitary Engineer-
ing Branch, Division of Reactor Development, U. S. Atomic Energy
Commission.
-------
HOMESTAKE - NEW MEXICO PARTNERS COMPANY
71
-- *,-«
NOSI"_::-,V.t^4- —, - iSr
••• • "-'^i!L^i:-r£\ > <
v-^7 ^^^ -m v-'^, /»
'tr.l'.^W^'"
5ai --S^^ 'as
WflTE"
*ELL
,-\'xi^;,/
""' "^ * •''
TA-L'iGs OCND ' '*•; »//
^"*.j
• ,G*rc
L^ ^O
BALL fw*v
HOMESTflKE-PARTNERS
MILL-SITE
ezc ADC c
Figure 4. Radiym-226 in water samples from test wells at Homesfake-
Partners mill-site.
-------
THE CARBONATE LEACH URANIUM EXTRACTION
PROCESS
II. HOMESTAKE-SAPIN PARTNERS, GRANTS,
NEW MEXICO
H. R. Pahren*
M. W. Lam me ring
J. Hernandez
Introduction
This report presents the results of a study conducted at a
second uranium refinery employing the "carbonate leach extrac-
tion process. The study was performed by the U. S. Public Health
Service and the New Mexico Department of Public Health during
September 1959, with the cooperation of personnel of the Home-
stake-Sapin Partners mill and the U. S. Atomic Energy Commis-
sion.
The Homestake-Sapin mill with a design capacity of 1500
tons of ore per day was placed in operation during the latter half
of 1953. Located about ten miles northeast of Grants. New
Mexico, (adjacent to the Homestake-New Mexico Partners mill)
it lies in an arid area with no surface water in the near vicinity.
At the time of the survey, the mill was processing about 1640
tons per day of dry ore. which assayed from 0.143 per cent to
0.235 per cent IL08. Recovery of U^O,, averaged 90 per cent
with a yield of 5uOO pounds per day of yellowcake. The plant ef-
fluents were discharged to a tailings pond, where all but a small
liquid portion (recycled as process water) was retained for dis-
sipation by evaporation and possibly seepage. Figure 1 is a de-
tailed flow diagram of the refinery. The units are described in
detail in the following section.
Process Description
ORE PREPARATION
Ore is brought from the mines by truck and is initially stored
outdoors on a storage pad. A bulldozer transfers the ore to a
feeder from which it is carried by belt conveyor to a jaw crusher.
If the ore contains more than about 10 per cent moisture it is
passed through a gas fired rotary kiln dryer: ores with less than
10 per cent moisture by-pass the dryer.
"Respectively. Sanitary Engineer. Colorado River Basin Water
Quality Control Project. Public Health Service. Denver, Colorado:
Senior Assistant Sanitary Engineer, Radiological Pollution Activi-
ties Unit: and Associate Engineer. Environmental Sanitation Ser-
vices. New Mexico Department of Public Health. Santa Fe.
73
-------
74
CARBONATE LEACH PROCESS
Figure 1. Flow diagram of Horrtesrake-Sapin Partners uranium mill.
Grants, New Mexico, September 1959.
Next, the ore is sampled to ascertain the uranium content for
each lot. The ore passes through a sample cutter which diverts
10 per cent of the ore stream. The 10 per cent sample then passe-
through a second cutter which takes another 10 per cent sample.
After four such cuts, a sample of 0.2 pounds per ton of ore pro-
cessed is obtained. The balance of the ore is approximately
equally distributed in four fine-ore bins of 1500 tons capacity
each.
In the process following the fine ore storage there is a dupli-
cate circuit through the plant. This duplication of facilities con-
tinues until the pregnant solution streams from each circuit are
combined.
Two of the fine ore bins feed to each circuit. Ore fed to
process is first weighed by an automatic weightometer and then
is added to a ball mill along with mill solution to make a relatively
thick slurry. Following this, the slurry flows to a spiral classi-
fier where oversize material is separated and returned to the
ball mill. Additional mill solution is added to the classifier to
give an effluent with a specific gravity of 1.20. Approximately 45
per cent of the solids in the classifier effluent are minus 200 mesh
and 12 per cent are plus 65 mesh.
The classifier effluent is pumped to a 100-foot diameter pri-
mary thickener where the solids are concentrated in the bottom
and the overflow is returned to mill solution storage. Separan is
added in a one per cent solution to facilitate settling.
GFO 813—I 73-"
-------
HOMESTAKE - SAPIN PARTNERS 75
LEACHING
The underflow from the primary thickener is pumped by a
diaphragm pump to the first of eight leach tanks in series. The
retention time in each leach tank is about one hour. Under leach-
ing conditions of approximately 225° F. and 60 pounds pressure,
most of the uranium remaining in the solids is leached out.
Leaching solutions contain carbonate and bicarbonate ions,
and under the process conditions, the soluble uranyl tricarbonate
ion, UOr.(COo)Q , is formed.
^ o o
After leaching, the slurry flows to a 100-foot diameter
secondary thickener where additional separan is added to"aid in
settling the suspended solids. The secondary thickener overflow,
known as pregnant solution, from each circuit is combined. The
pregnant solution is clarified by means of a rotary filter, and then
is stored in the clarified-pregnant-solution storage tank. The
small amount of solids removed by the filter is slurried and re-
turned to the secondary thickener. Underflow the secondary
thickeners is pumped to the tailings filtration area.
PRECIPITATION AND PREPARATION OF YELLOWCAKE
Clarified pregnant solution is pumped to the first of seven
agitated precipitation tanks operated in series. Here, sodium
hydroxide is added as a 50 per cent solution until the pregnant
solution contains 8 grams of NaOH per liter. With the increase of
pH, the uranium precipitates as sodium diuranate or yellowcake,
as follows: ^
2 Na4U02(C03)3+ 6 NaOH—>Na2U20?+ 6 NagCOg-s- 3 HgO
The yellowcake slurry is transferred to a thickener to which
a solution of locust bean gum, a natural polysaccharide, is added
to aid in settling the yellowcake. Overflow from the thickener is
filtered through a plate and frame filter press to remove the small
amount of yellowcake that is carried over, and the clarified barren
solution istthen passed through a recarbonation tower. Flue gas
from the boiler plant is passed countercurrent to the barren solu-
tion in the tower, and the carbon dioxide in the flue gas neutralizes
the caustic alkalinity and forms additional carbonate. The recar- ,
bonated barren solution is then available for re-use in the process.
Underflow from the thickener contains the yellowcake, which
is filtered out with a rotary filter. The yellowcake is repulped
with fresh, softened water and refiltered, and the filtrate is added
to the yellowcake thickener. The resulting filter cake is washed,
dried, pulverized and then drummed for shipment to plants of the
U. S. Atomic Energy Commission. Yellowcake is filtered and
packaged only during the day shift.
-------
76 CARBONATE LEACH PROCESS
FILTRATION OF TAILINGS
The tailings filtration area consists of three stages of five
rotary filters each. The combined underflow from the two secon-
dary thickeners passes through the three filtration stages and the
pregnant solution that carries over is separated from the solids
before the solids are discharged to the tailings pond. Filtrate
from the first stage filters is added to the secondary thickener,
while filtrate from the second stage filters is pumped to the mill
solution storage tank.
Filter cake from the first stage filters is washed by part of
the third stage filtrate. The balance of the third stage filtrate is
used to repulp the filter cake prior to the second stage filtration.
About one-third of the recarbonated barren solution is used to
wash the filter cake on the second stage filters. The remaining
recarbonated barren solution is used for repulping between second
and third stage filtration. Third stage filter cake is washed by
reclaimed water from the tailings pond and, after removal from
the filter drum, is repuloed with unsoftened raw water. This
slurry passes through an automatic sampler. After passing throus;.
the sampler, additional water is added before the slurry is pumpen
to the tailings pond. Mill personnel indicated that the additional
water consisted of approximately 10 gpm from the reagent build-
ing, 40 gpm of cooling water from the vacuum pumps and 25 gpm
of excess return water from the tailings pond.
Tailings Pond
The tailings pond is circular in shape and has 60 acres within
the dikes. Water covered only 39 acres of this area at the time of
the survey. In addition to the main pond there is a smaller pond
used as a surge pond for the tailings pond water that is returned
to process. Return water is decanted from the main pond. In
addition to the process water, boiler plant blowdown water and the
plant domestic sewage are added to the tailings pond. The sewage
is first treated in a septic tank.
Sampling Procedures
Samples were obtained during two 72-hour cycles as follows:
Cycle I - 8 AM Sept. 17, 1959, to 8 AM Sept. 20. 1959:
Cycle II - 8 AM Sept. 20, 1959, to 8 AM Sept. 23, 1959
At all but three sampling stations, a sample was obtained for
each cycle by compositing equal volumes every 2 hours for the dur-
ation of the cycle. Thus, 36 portions were used to make up the
single composite sample.
The wastes going to the tailings pond were sampled by an
automatic sampler operated by mill personnel. The Public Health
Service obtained a portion of the sample collected by this auto-
-------
HOMESTAKE - SAPIN PARTNERS 77
matic sampler each day and composited the three equal portions
for each cycle.
A 1-day lead time was used for the raw ore samples. A por-
tion of the official sample from each ore lot was obtained from
that collected by the plant sampling equipment before the ore was
discharged to the fine ore storage bins. To make the composite
sample for each cycle, the portion taken from each ore lot pro-
cessed was weighted according to the tonnage of the lot. For the
ore samples, the period from 8 AM Sept. 16 to 8 AM Sept. 19 was
Cycle I and the period from 8 AM Sept. 19 to 8 AM Sept. 22 was
Cycle II.
Composite yellowcake samples were obtained for each of two
consecutive 3-day periods beginning September 18. thus giving a
1-day lag period. Representative samples of each day's produc-
tion were weighted according to tonnage of production for the day
and composited for the cycle. Yellowcake was not packaged on
September 20, the third day of Cycle I. Thus, the production on
September 21 was equally divided between both cycles to compen-
sate for this variation.
As mentioned in the process description, there are duplicate
circuits in the plant from the beginning of the process through the
secondary thickener step. During the survey individual samples
were obtained at significant points for both the north and south
circuit. Samples from the same point in each circuit vvere then
combined. A list of all the sampling stations is shown in Table 1.
Sample Processing
The samples collected during the survey were shipped to the
Public Health Service's Robert A. Taft Sanitary Engineering
Center, where all chemical and physical analyses were performed,
except the analysis for radium-226.
Although the results presented in the following section are
average values for Cycles I and II, analyses for dissolved and un-
dissolved gross alpha and beta activity, dissolved and undissolved
radium-226, sodium (Na+), and pH were performed on individual
samples from each cycle. Thus, each station is associated with
two results for each applicable analysis. Supplementary analyses
for arsenic, chloride, phenolphthalein alkalinity, and total alka-
linity were performed on the liquid portions of Station 11 and 13
samples.
The radium analyses were performed by a private laboratory
under contract to the Public Health Service. Pretreatment of the
samples by the Taft Center laboratory consisted of liquid-solid
separation by filtration through a membrane filter, and grinding
the dried suspended material to less than 100 mesh.
-------
78
CARBONATE LEACH PROCESS
Table 1. SAMPLING STATIONS
Station Number
9
10
11
12
13
14
Description
Raw ore
Mill solution
Overflow from classifiers
Overflow from primary thickeners
Underflow from primary thickeners
Effluent from digesters
Underflow from secondary thickeners
Pregnant solution to precipitation tanks
Recarbonated barren solutions
Yellowcake product
Repulped third stage filter cake to tailings pond
Return water from tailings pond
Unsoftened well water
Softened water
Discussion of Results
Table 2 through 8 show average values obtained from the re-
sults of Cycle I and II. This technique of averaging over the two
72-hour periods was used because of the close agreement between
analytical results for the two cycles, and to further minimize
possible errors associated with representative sample collection
and retention times, such as the 8-hour retention across the
leaching circuit. The material balances were developed from a
combination of field data (ore tonnage, yellowcake production,
flows, etc.) extracted from mill records and laboratory analyses.
The balances characterize each unit in the carbonate-leach pro-
cess and form the basis for comparison between varying ore feed
rates and ore quality.
Throughout the survey period, the mill operated somewhat in
excess of design capacity. The ore was fed at the rate of 1590 dry
tons per day during Cycle I, and 5130 pounds per day of yellowcake
was packaged. Corresponding figures for Cycle n were 1686 dry
tons per day and 4250 pounds, respective. The general physical
characteristics and the concentration of sodium ion for the in-plant
process streams, waste effluent, and raw water supply are pre-
sented in Table 2. Concentrations of dissolved and suspended solids
are presented in Table 3.
In Table 2, the data in Columns (3) through (7) are laboratory
determinations. The slurry flows (Column 2) except for Stations
8, 12, and 13, are calculated values. The flow at Station 9 was
metered at an average value of 232 gpm for the survey period,
GPO 813-173-8
-------
HOMESTAKE - SAPIN PARTNERS
Table 2. PROCESS STREAM CHARACTERISTICS a
79
Station
(1)
1
2
3
4
5
5
7
j,
9
10
II
12
13
14
Slurry flo*.
gpm
(2)
b
1501
1665
1291
357
361
316
236
241
b
625
eo
399
3.5
Slurry specific
gravity
(3)
_
1.12
1.19
1.12
1.50
1.50
1.56
1.11
1.13
-
1.26 d
1. 01
1.00
1.00
Dry suspended
solids by weight ^
(4)
_
c
13.9
c
51.2
50.2
54.6
c
c
-
34.6 d
C
c
c
Specific gravi
of dry solids
(5)
.
-
2.60
-
2.46
2.59
2.50
-
-
-
2.40
-
-
-
ty / Na
/ rr.g; 1 in slurrv
' (6)
_
50.3 x I03
39.2 x 103
44.5 x 103
24.2 x 103
25.1 x 103
23.6 x 103
51.5 x 103
53.3 x 103
-
4.45 x 103
5.15 x I03
3. TO x I03
4.45 x 103
pH
(7)
-
10.5
10.4
10.3
10.3
10.2
10.1
10.1
10.6
-
10.1
9,9
7.5
6.9
Average "f Cycles I and II.
S">hd sample.
Liquid sarrple (negligible solids).
Based on Cycle I only.
Table 3. CONCENTRATION OF SLURRY SOLIDS
Station
(1>
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Dissolved solids.
mg 1 slurry
(2)
b
138.000
12S.OOO
137.000
87.900
94.500
91.300
159.000
162.000
b
12.200
14.500
1.860
1.700
Suspended solids.
rr.g 1 slurry
(3)
b
850
165.000
7SO
765.000
752.000
854.000
145
50
b
435.000
193
0
56
-------
80
CARBONATE LEACH PROCESS
which is in good agreement with the calculated value of 241 gpm.
The field notes indicated, however, that the average metered
flow would probably be low because of marked fluctuations in line
flow for a period of about 8 hours during Cycle II. In the same
8-hour period softened water was added to the recarbonation
barren tank to maintain suction. Fresh water was substituted
for recycled water (Station 12) during the last half of Cycle II.
10 gfl
no llm
10.- O
TO TAILUGS PCND
Figure 2. Schematic flow diagram of the Homestake-Sapin mi!!,
September 1959.
Figure 2 illustrates the basic flow diagram of the mill in
schematic form. The small differences in estimated slurry flows
across process units are considered to be insignificant. The
relatively small discrepancies in solids flow across a unit pro-
bably resulted from the limits of sensitivity of the dry-solids
specific gravity analysis. Theoretically, if a constant feed rate
and a steady state process throughout the mill are assumed the
solids flow would be expected to remain unchanged throughout the
mill because the decrease in undissolved solid tonnage is slight.
Discrepancies in slurry flow are the result of the experimental
errors that entered into the total weight and solid balances in the
laboratory determinations. As will be seen, these discrepancies
are not of major importance in this study.
-------
HOMESTAKE - SAPIN PARTNERS
81
Figure 2 also indicates that material balances can be made
across grouped process units. Proceeding from the ore storage
bin, (Figure 2) the first balance quantitatively equates the input to
the ball mill and classifier circuits of ore (Station 1) and mill
solution (Station 2) with the output from the classifiers (Station 3).
The output from the classifiers should also be accounted for in the
overflow (Station 4) and in the underflow (Station 5) from the pri
mary thickeners. Quantities present in the underflow from the
primary thickener (Station 5) should be measurable in the effluent
from the leaching circuit (Station 6). The recarbonated barren
solution (Station 9) and yellowcake production (Station 10) should
approximate the pregnant solution to the precipitation tanks
(Station 8) and the caustic stream. As an over-all plant balance,
the ore feed (Station 1), caustic stream, return water at the third-
stage filters (Station 12), and untreated well water for tails re-
pulping (Station 13) should approximate yellowcake production
(Station 10) and the effluent to the tailings pond (Station 11).
Table 4. SOLIDS QUANTITIES a
Station b
(1)
1
2
3
4
5
6
7
e
9
10
11
12
13
14
Dissolved solids.
tons dav
(2)
-
1.250
1.260
1.050
189
204
174
224
235
-
46
5.8
5.0
- 0
Suspended solids.
tons dav
(3)
1.643
8.0
1.640
6.5
1,640
1.620
1.620
-0
-o
2.5
1.620
-0
-0
-0
Total solids.
tons, day
(4)
1.640
1,253
2.900
1.067
1.829
1.S24
1.794
224
235
2.5
1.666
5.8
5.0
-0
Average of Cycles I and II.
Approximately 18 tons per day of dissolved solids are added to the precipitation
tanks (caustic stream)
Table 4 presents the average solids quantities of both cycles.
If the balance points listed above are used, it is apparent that the
materials balance (Figure 3) is quite good. The raw ore feed
(Station 1) and mill solution (Station 2) comprise an input of 2. 898
(1. 258 + 1, 640) tons per day to the classifier circuit as compared
-------
82
CARBONATE LEACH PROCESS
DISSOLVED SOLID? - 1C
SISPEKPED SOLUS - IOC
TOTAL SOLIDS - IIC
SAMPLING STATION »0. -
TO»S'P»Y
TINS'OY
Figure 3. Schematic flow diagram of solids balance, Homestake-Sapin
mill, September 1959.
to the measured output of 2, 900 tons per day (Station 3). The out-
put of 2,900 ton's per day from the classifiers is also balanced by
2,896 (1,067+ 1,829) tons per day from the primary thickener
overflow (Station 4) and underflow (Station 5).
In good agreement with the 1, 829 tons of solids per day in the
slurry to the digesters (Station 5) is the 1, 824 tons per day leaving
the leaching circuits (Station 6). Further examination of the data
shows an increase in dissolved solids of 15 tons per day across
the leaching circuit and a corresponding decrease in suspended or
undissolved solids of 20 tons per day. Pregnant solution (Station
8) and caustic at the precipitation tanks yield 242 (224 -f- 18) tons
of solids per day as compared to 238 (235 + 2.5) tons per day in
the recarbonated barren liquor (Station 9) and yellowcake product
(Station 10). For the process as a whole. 1,669 (1,640 -f- 18+ 5.8
+ 5.0) tons of solids per day enter the refinery as raw ore (Sta-
tion 1), caustic, return water (Station 12), and well water (Station
13) in that order. This is in exact agreement with the 1, 669 +
(1,666 + 2.5) tons per day leaving the refinery as waste to the
tailings pond (Station 11) and yellowcake product (Station 10).
-------
HOMESTAKE - SAPIN PARTNERS
Table 5 GROSS RADIOACTIVITY CONCENTRATIONS a
83
Station
(I)
1
2
3
4
5
R
7
3
9
10
11
12
13
H
Suspended
Alpha
(2)
-
3.813
423.000
7.550
1.843,000
1.710.000
1.640.000
850
b
-
7 13. GOO
430
0 5
1.3
Beta
(3}
-
6.410
545.000
9.240
2.050.000
1.700.000
1.710.000
2.150
b
-
642.000
740
0.7
2.5
Dissolved
Alpha.
(4)
-
191.000
227.000
217.000
111.000
225.000
327.000
353.000
53.400
-
7,900
8.500
5.S
4.5
Beta
(5)
-
575.000
562,000
555.000
341.000
763.000
775 . 000
1,030,000
153.000
-
22. 100
23. 100
30
20
Dry suspended solids
Alpha
>>MC g
(6)
3.070
4.430
2.590
9.6JO
2,400
2.280
1.923
5. 670 c
b
317.000
1.650
2.120
b
23 C
Beta
•MfC- 8
(7)
3.340
7,540
3,310
11.600
2.670
2.260
2.000
14. 300 c
b
420.000
1.930
3.720°
b
45 r
Average rf Cycles I and II.
Gross radioactivity concentrations are presented in Table 5.
These data show a consistent decrease in the gross alpha activity
of one gram of undissolved ore solids as it moves from the ore
storage bins through the digesters and ultimately to the tailings
pond. Leaching of the uranium from the ore solids at the digesters
is illustrated by the twofold increase in dissolved alpha activity
between Stations 5 and 6 and the small decrease in the suspended
solids radioactivity. The comparison of the alpha activity (u/ic/g)
of the heavy settleable solids (Stations 3, 5, 6, 7, and 11) with
that of the smaller size suspended particles that are characteris-
tic of thickener overflows (Stations 2, 4, and 8) was in agreement
with observations from previous mill surveys; that is, the alpha
activity of the smaller particles (slimes) exceeded that of the
sands by factors of about 2 to 4.
If radioactive equilibrium in the ore among uranium-238,
radium-226 and the other daughter products is assumed, the alpha
activity of the dry suspended solids should remain constant across
the ball mill and classifier circuit. The observed decrease at
Station 3 compared to Station 1 may be due in part to leaching and
a resulting loss of the third member in the decay chain, radon-
222, to the atmosphere as a gas. That further leaching occurs as
far as the secondary thickener is illustrated by the increase in the
dissolved alpha activity of the pregnant liquor (Station 8) in com-
parison to the dissolved alpha activity of the digester effluent
(Station 6)
-------
84
CARBONATE LEACH PROCESS
The gross alpha concentrations of Table 5 and the balanced
slurry flows of Table 2 form the basis for the gross alpha quanti-
ties presented in Table 6. If previously mentioned balance points
are used, the gross alpha balance appears to be adequate. The
6,170 (4, 570+1, 600) me/day input to the classifiers from the ore
(Station 1) and mill solution (Station 2) is in good agreement with
Table 6. GROSS ALPHA QUANTITIES3
Station
(1)
1
2
3
4
5
6
7
a
9
10
n
12
13
14
Gross Alpha Radioactivity, me, day
Dissolved
(2)
b
1 570
2.070
1.550
216
445
562
456
73
b
27
3
d
d
Suspended
(3)
4.570
30
3. 860
53
3.570
3.340
2.830
- 0
c
631
2.430
~0
d
d
Total
(4)
4. 570
1.600
5.930
1,603
3. 786
3.785
3.392
456
73
681
2.457
3
d
d
Average of Cycles I and 11.
Solid sample.
Not determined.
Negligible.
the 5.930 me/day leaving the classifiers (Station 3). The output of
5, 389 (1, 603 + 3, 736) me/day from the primary thickeners is
split between the overflow (Station 4) and underflow (Station 5).
This agrees sufficiently well with the 5,930 me/day input to the
thickeners (Station 3), although a definite loss of alpha activity is
noted. The balance across the digesters is excellent: 3, 786 me/
day enter at Station 5 and the same amount leave at Station 6.
Gross alpha activity in the recarbonated barren solution (Station 9)
that leaves the process as yellowcake product (Station 10) totaled
759 (78 + 681) me/day. This does not agree well with the 456 me/
day in the pregnant liquor (Station 8) and the negligible amount
contributed by the caustic. The explanation for this discrepancy
is not readily apparent, especially in view of the fact that the
balances for solids and slurry flows at these stations were very
good.
-------
HOMESTAKE - SAPIN PARTNERS
Table 7. RADIUM-226 CONCENTRATION'S '
85
Station
(1)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Radium-226 in slurry, pug, I
Dissolved
(2)
-
9.560
Suspended
(3)
-
1.220
13.200 (8. 300)° 83.900
8.790
5,160
6.620
5.280
18.900
59
-
34
35
5.4
4.2
b
323.000
370,000
380,000
286
5.4
-
222,000
bb
b
b
Radium-226 in
dry suspended
solids. nn« g
(41
497
1.440
508
b
429 (500) c
492
445
1.970
107
7,190
510
b
b
b
Average of Cycles I and n.
Not determined (negligible solids).
c The parenthetical result is most probable. See discussion in text.
The over-all plant balance also showed considerable disagree-
ment; 4, 573 (4, 570 + 3) me/day entered the process as ore (Sta-
tion 1) and recycled process water (Station 12), whereas the effluent
to the tailings pond (Station 11) and the yellowcake product (Sta-
tion 10) accounted for only 69 per cent of the input, or 3,138 me/
day. The over-all plant balance in the similar Homestake-New
Mexico Partners Company mill showed a 77 per cent accounting.
In that case it was suspected that the semiautomatic sampling de-
vice for collecting a tailings sample actually collected a non-
representative excess quantity of large versus fine solids, there-
by lowering the gross alpha activity; this may in part explain the
discrepancy noted above. It appears probable that the alpha acti-
vity of representative dry suspended solids at Station 11 was with-
in the range of 1,900 to 2, 300M/*c/g (Stations 7 and 6) as compared
to the 1,650/i/ic/g determined for the sample obtained. Based on
2,100/t/jc/g. the alpha activity at Station 11 would be increased to
about 3,100 me/day and the over-all recovery to about 83 per
cent.
Concentrations of radium-226 per liter of slurry and per gram
of suspended matter are presented in Table 7. The dissolved rad-
ium concentrations reflect those solids passing through a membrane
filter, and the suspended concentrations, that portion retained on
-------
86
CARBONATE LEACH PROCESS
the filter. From the data for dry suspended solids, it is apparent
that most of the radium remained in the undissolved form through-
out the process; 497/i/ig/g entered in the ore feed and about the
same concentration left in the spent ore solids (Station 11). At
intermediate steps in the process (Stations 3, 5, 6, and 7) the
range of observed concentrations in the ore solids varied between
429 and 508/t/tg/g. As was the case with gross alpha activity, the
fine-grained particles exhibited the highest activity.
Table 8. RADIUM-226 QUANTITIES a
Station
(D
1
2
3
4
C
6
7
8
9
10
11
12
13
14
Radium-226 mg/ day
Dissolved
(2)
b
78
e
130 (81)
61
14
18
13
25
- 0
b
-0
c
c
c
Suspended
(3)
739
10
755
9
e
639 (744)
724
655
-0
- 0
16
750
d
d
d
Total
(4)
739
88
e
885 (836)
70
£
653 (748)
742
668
25
- 0
16
750
c
c
c
Average of Cycles I and II.
Solid sample.
Negligible.
Not determined.
6 The parenthetical result is more probable. See discussion in text.
Table 7 also indicates certain discrepancies that are impor-
tant to interpretation. The concentration of dissoh'ed radium-226
at Station 3 (overflow from the classifiers) is clearly not consis-
tent with the dissolved radium-226 data for Station 2 (mill solution
added at the classifiers) or with the data for Stations 4 and 5,
while are locations immediately following Station 3. Specifically,
the average dissolved radium was 9, 560/*x*g/l of liquid in mill
solution entering the classifiers (Station 2), 8, 800/i/ig/l in the
overflow from the primary thickeners (Station 4) that follow the
classifiers, and 10, 100/i/ig/l in the liquid portion of the underflow
-------
HOMESTAKE - SAPIN PARTNERS
87
< E Y:
DISSOLVED PAOU'M-226 - 10 mg/d»
SUSPEKKD R»DIW<-22« -ICO »a'da»
TO T»IU»GJ PC«D
Figure 4. Schematic flow diagram of Radium-226, Homestake-Sapin
mill, September 1959.
slurry leaving the primary thickeners (Station 5). In contrast, a
value was reported of 15,300 micromicrograms of radium-226 per
liter of liquid in the slurry leaving the classifiers (Station 3) prior
to entering the primary thickeners. This latter value is clearly
inconsistent with the surrounding data; between 8. 800 and 10, 100
A/*g/l of liquid are required for Station 3. As will be seen, a mean
value of 9, 600/
-------
88
CARBONATE LEACH PROCESS
Table 9. CHEMICAL CHARACTERISTICS OF THE FILTRATE AT
STATIONS 11 AND 13
Station
11
13
Arsenic
mg/1
0.49
-
Chlorides^
mg/1 Cl
286
250
Phenolphthalein
alkalinity
mg/1 as CaCO
1,720
5
Total
alkalinity
mg/1 as CaCO,
3,560
690
somewhat low, although this is less directly interpretable because
of the side^-routes between Station 6 and 7 (see Figure 2). As a re-
sult, the radium-226 concentration of undissolved solids at Station
5 was probably about 500^/ig/g, rather than the reported value of
429/i/ug/g. This would also perfect the radium-226 quantity bal-
ance, as noted below.
No explanation for the foregoing two discrepancies can be of-
fered at this time, although they appear to be quite definite.
The assumption of equilibrium in the ore between uranium-238
and its daughter, radium-226, allows a separate check to be made on
the concentration in the raw ore. If the average ore assay of 0.167
per cent UoCL for the survey period is used, a concentration of 474
micromicrograms of radium-226 per gram of ore is derived. This
agrees quite favorably with the analyses average of 497^i/tg/g. If
the possibility of Radon-222 loss did not exist, it could be assumed
that all the daughters of uranium-238 were in equilibrium with the
parent during the survey period.
Table 10. CHEMICAL CONSUMPTION
Product
Guar Gum
NaOH (100%)
Separan
Quantity
25 Ibs/each three days
30.82 Ibs/ton of ore processed
0.05 Ibs/ton of ore processed
Table 8 presents the radium quantities at the various process
stations at an ore processing rate of 1, 640 tons per day. The
values in Table 8 were obtained with the concentrations of Table 7
and the slurry flows of Table 2. As indicated earlier, the reported
dissolved radium-226 concentration at Station 3 and the undis-
solved radium-226 concentration for Station 5 are probably erron-
eous. The more probable results are given as parenthetical fig-
ures in Tables 7 and 8. If these figures are used, the radium
balance throughout the process (Figure 4) is generally quite good,
as follows:
Mill solution combines with the raw ore feed for 827 (88 + 739)
mg/day of radium-226 entering the ball mill and classifiers. This
is in good agreement with the estimated 836 mg/day at Station 3.
Overflow (Station 4) and underflow (Station 5) from the primary
thickeners account for 828 (70+ 758) mg/day. The balance across
-------
HOMESTAKE - SAPIN PARTNERS 89
the leaching circuit consists of an estimated 758 mg/day entering
at Station 5 and 742 mg/day leaving at Station 6. In comparison to
the 836 mg/day at Station 3 is the 812 (70 + 742) mg/day from the
primary thickener overflow (Station 4) and the digester effluent
(Station 6). Pregnant liquor carries 25 mg/day to the precipita-
tion circuit. This is in acceptable agreement with the 16 (16 + 0)
mg/day accounted for as yellowcake product (Station 10) and re-
carbonated barren solution (Station 9).
The total plant balance shows 739 mg/day entering as raw ore
(Station 1) and 766 leaving via the waste effluent to the tailings pond
(Station 11) and as yellowcake product (Station 10): plant input is
in good agreement with plant output.
Based on the 739 mg/day of radium-226 entering the mill in
the raw ore, 16 mg/day, or 2.2 per cent leave with the yellowcake
product. This percentage is in general agreement with the 1.5 per
cent for the Homestake-New Mexico mill but in substantially higher
than that found in mills using the acid leach process and other ores.
To delineate further the characteristics of the main plant
effluent to the tailings pond and the raw well water, a series of
five chemical tests was run on the filtrate portions from Stations
11 and 13. The average values based on 1 liter of slurry, are
tabulated in Table 9.
The rate of consumption of chemical additives as reported by
mill personnel is presented in Table 10. Carbon dioxide, which
was used to partially neutralize the caustic alkalinity in the recar-
bonation tower, was not listed because it was generated within the
process as flue gas. Oxidizing agents were not added during the
survey and had not been added to the digesters for 3 months pre-
ceding the survey. In addition to the products listed in Table 10,
lime, a small amount of soda ash, and sulphuric acid were used
in the treatment of the mill's raw water supply.
Waste Disposal
Slurry flow to the tailings pond, as shown in Figure 2, aver-
aged about 700 gpm, of which 112 gpm was solid flow, and 588 gpm,
liquid. The spent ore solids are retained in the tailings pond, but
the liquid phase, in addition to being retained for concentration
by evaporation, may be recycled as return filter wash water or
may drain by seepage. The opinion of mill personnel was that the
soil at this location provided an excellent seal, thereby minimiz-
ing the problem of seepage into the ground water.
The 0.12 mg/day of dissolved radium entering the tailings pond
at a concentration of 35/i/tc/l does not constitute the only source
of possible ground water contamination. The insoluble radium ac-
cumulating at the rate of 750 mg/day provides a radium reservoir
for leaching which, if coupled with seepage, could also produce a
-------
90
CARBONATE LEACH PROCESS
significant ground water contamination problem. That leaching of
radium from spent ore solids does occur has been discussed in the
reports on the study of the Animas River by the U. S. Public Health
Service. 10> 12
Table 11.
RADIUM CONCENTRATIONS IN SAMPLES FROM TEST WELLS
ON HOMESTAKE - SAFIN" MILL PROPERTY.
Depth of well. ft.
70
80
95
-
60
Description
Cased
Cased
Cased
Observation Well
Cased
Dissolved Radium
1959
0.8
1.8
0.7
0.2
1
- 225. fipK 1
1961
0.22
0.13
0.24
-
1.79
The Homestake-New Mexico Partners report points out the
need for considering ground water contamination inasmuch as
ranchers in the general area of the mills use well supplies for
domestic consumption, for watering livestock, and for irrigation.
In addition, the communities of Milan and Grants, located a few
miles south of the mills, take their domestic supplies from wells.
The Homestake-New Mexico Partners report also describes the
water-bearing strata in the area, and the direction of ground
water movement, and summarized the radium results for num-
erous well samples. Table 11 presents the radium concentrations
found in samples from test wells located around the periphery of
the tailings pond on the Homestake-Sapin mill property. These
were collected in 1959 and 1961.
These results may be compared to the radium content of wells
located several miles to the west of the mills and that of wells be-
tween Grants and San Rafael. The concentration range for these
wells is 0.1 to 0.4/t/ic/l. a natural background concentration range.
An evaluation of the Homestake-Sapin pond must take into
account the presence of the Homestake-New Mexico Partners
pond and their combined effect. Therefore, it seems advisable,
in view of the reservoir of radium in both tailings ponds and the
variability of results to date, that the monitoring of well supplies,
particularly in the near-vicinity of the ponds, be continued until a
firm conclusion on the presence or absence of seepage can be
reached. If a significant buildup of radium in the test wells is ob-
served over a period of time, remedial measures may be neces-
sary. The need for such measures should be based on the obser-
ved rate of buildup, the possibility of peaking at a given concen-
tration, and the human exposure potential in accordance with
applicable radiation protection criteria.
-------
HOMESTAKE - SAPIN PARTNERS 91
Acknowledgment
The generous cooperation and technical assistance of the
following are gratefully acknowledged: Personnel of the Homestake-
Sapin Partners Mill; the U. S. Atomic Energy Commission; Charles E.
E. Sponagle, Public Health Service, Region VIII, Denver, Colo-
rado; E. A. Pash, Carl Hirth, Carl Shadix and H. D. Nash,
Public Health Service. Cincinnati, Ohio; and D. E. Rushing,
Public Health Service, Salt Lake City, Utah. The technical advice
and guidance of Dr. E. C. Tsivoglou, Public Health Service,
Cincinnati, Ohio, was greatly appreciate. This study was suppor-
ted by funds made available through the Environmental and Sani-
tary Engineering Branch, Division of Reactor Development, U. S.
Atomic Energy Commission.
-------
Bibliography
l.Tsivoglou, E. C., Bartsch, A. F., Rushing, D. E., and Holaday,
D. A. Effects of Uranium Ore Refinery Wastes on Reciving
Waters. Proc. 1958 Nuclear Science and Engrg. Conf., Re-
print 40, Am. Inst. Chem. Engrs., New York, N.Y.
(March 17, 1958).
2. National Bureau of Standards. Maximum Permissible Amounts of
Radioisotopes in the Human Body and Maximum Permissible
Concentrations in Air and vVater. Handbook 52. U. S. Gov't.
Printing Office, Washington, D. C. (1953).
S.Dayton, S. UgOq Processing Made More Efficient by Plant-Con-
trols at Edgemont Mill. Mining World, 19. No. 1, 36-43
(January 1957).
4. Public Health Service. Determination of Gross Radioactivity in
Water. Silt, and Biological Materials. Robert A. Taft Sani-
tary Engineering Center, Cincinnati, Ohio. (1957).
5. Standard Methods for the Examination of vVater. Sewage and Indus-
trial Wastes. Am. Publ. Hlth. Assn., New York, N.Y., 522 p.,
10th Ed. (1955).
6. Part 20 - Standards for Protection Against Radiation. U. S. At.
Energy Comm., Federal Register. 22. No. 19. (Tuesday,
Jan. 29. 1957).
7. Public Health Service. Report on an Investigation of Ground Water
Pollution. Grants-Bluewater. New Mexico. Robert A. Taft
Sanitary Engineering Center, Cincinnati, Ohio. (Aug. 1957).
S.Velz. C. J., Graphical Approach to Statistics. Water and Sewage
Works, 97. (1950).
9.Recommendations of the International Commission on Radiological
Protection. Main Commission Report, Sept. 9, 1958: Perga-
mon Press, London, England. (1959).
lO.Tsivoglou. E. C. Shearer. S. D.. Shaw. R.M.. Jr., Jones, J. D.,
Anderson, J. B., Sponagle, C. E., and Clark, D. A., Survey
of Interstate Pollution of the Animas River. Colorado-New
Mexico. U. S. Pablic Health Service. Robt. A. Taft San.
Engrg. Ctr., Cincinnati, Ohio. (May 1959).
H.Public Health Service Drinking Water Standards. 1946. Public
' Health Reports. 61, No. 11, 371-384: Reprint No. 2697.
(March 15, 1946: revised March 1956).
93
-------
12.Tsivoglou, E. C., Shearer, S. D., Jones, J. D. Sponagle, C. E.,
Pahren, H. R., Anderson, J. B., and Clark, D. A. Survey of
Interstate Pollution of the Animas River. Colorado-New
Mexico. II. 1959 Surveys. U. S. Public Health Service, Robt.
A. Taft San. Engrg. Ctr., Cincinnati, Ohio. (Jan. 1960).
IS.Knott, Woodrow, The Processing of Uranium Ores of the Colorado
Plateau by the Climax Uranium Company. The Mines Maga-
zine, p. 29. (January 1955).
14.Survey of the Colorado and Gunnison Rivers in the Vicinity of Grand
Junction. Colorado. Technical Report No. 2, Colorado River
Basin Water Quality Control Project, U. S. Public Health
Service. (May 1961).
15 National Bureau of Standards. Maximum Permissible Body Bur-
dens and Maximum Permissible Concentrations of Radio-
nuclides in Air and Water for Occupational Exposure.
Handbook 69, U. S. Gov't. Printing Office, Washington, D.C.
(June 5 1959).
16.McClaine, L. A., Bullwinkel, E. P., and Huggins, J. C. The
Carbonate Chemistry of Uranium: Theory and Applications.
Internat. Conf. of Peaseful Uses of Atomic Energy, A/CONF.
8/P/525, USA. (July 18, 1955).
17-Clegg, J. W.. and Foley, D. D. Uranium Ore Processing. Atoms
for Peace, USA. Geneva 1958. Prepared for USAEC. Addison-
Wesley Publ. Co., Reading. Mass. (Sept. 1958).
lg Review of Waste Disposal Problem at Homestake-Partners Mill-
site (N.M.) Communication from the New Mexico Department
of Public Health.
94
------- |