SPE

Society of PetroIeun Engineers of AIME

SPE 12054
In-Situ Production Cost Model for Uranium
by G.W. Toth, NUS Corp., and W.C. Larson, USBM
Copyrighl 1983 Sociely of Pelroleum Engineers of AIME
This paper was presenled allhe 581h Annual Technical Conference and Exhibilion held in San Francisco, CA, Oclober 5-8, 1983. The male rial is subjecl
10 correclion by Ihe aUlhor. Permission 10 copy is reslricled 10 an abslracl of nol more Ihan 300 words. Wrile SPE, 6200 North Cenlral Expressway,
Drawer 64706, Dallas, Texas 75206 USA. Telex 730989 SPEDAL.

INTRODUCTION
In a recently completed research project for
the U.S. Bureau of Mines -Twin Cities Research Center, a computerized costing procedure for uranium
in situ leach mining was developed by the NUS Corporation.
This costing procedure, termed a cost
model, employs a process engineering approach for
estimating total project costs as well as equipment
and manpower requirements
for
uranium in situ
leaching operations in either Texas or Wyoming.
The Bureau recently added the capability to apply
the model for New Mexico deposits.
NUS also provided consulting support for this modification.
Both capital and operating costs are generated by
the model
along with
the
overall
production
cost/pound U308 subject to a known rate of return.
Conversely, rate of return on equity can be solved
for, subject to a given price of yellowcake.
During the course of the research effort, sens1t1vity tests were conducted for numerous key
parameters to determine the cost influence attributable to incremental changes in parameter values.
The results of these sensitivity tests along with a
description of the in situ leaching cost model have
been documented
as
Bureau
of Mines,
Mineral
Research Contract Report No. J0199l2, "Cost and
Sensitivities Analysis for Uranium In Situ Leach
Mining," dated March 1981.
An update of this
report was subsequently prepared by the Bureau
entitled "Improved Solution Mining Production Cost
Model," dated September 30, 1982.
This update contains the modifications required for application of
the model for New Mexico deposits.
In the remainder of this paper, the various
features and capabilities of the cost model will be
highlighted, along with a review of the sensitivity
analysis findings.

COST MODEL FEATURES

The design of the uranium in situ cost model
required the development and integration of a
References and illustrations at end of paper.

process engineering approach, a discounted cash

flow procedure and a regionalized cost data base.
The resulting model has also been designed to offer
the user maximum flexibility in specifying site
conditions.
The model accepts this site information, sizes both well field and extraction plant,
determines equipment and manpower requirements, and
assigns appropriate costs from the model data base.
The following points provide further
regarding features of this cost model.

detail

Develops detailed costs (both capital and

operating) and requirements for any userspecified project condition for the life
of the project.

Solves for minimum required sales revenue

per pound U308 (production cost) or rate
of return on equity.

Contains
regionalized
data
base
Texas,
Wyoming
and
New Mexico
applications.

Allows
for
cost analysis applications
when only minimal information is known,
as well as for cases in which detailed
project data are available.

Accepts and accounts for either static or

dynamic site conditions throughout project life.

Accepts
user-specified
structure options.

for
site

capitalization

Each of these points is further explained in

the following discussion.
Develops detailed cos ts.
Beginning with the point
in time when a decision is made to start a pilot
plant facility, all capital and operating costs are
estimated by this model.
The categories of costs
generated are listed below:

IN-SITU PRODUCTION COST MODEL FOR URANIUM

Capital Costs
Process equipment
Equipment installation
and/or site improvement
Building
Initial well field
Permitting
pilot plant
Restoration system
Engineering and/or
project management
Fixed capital
Contingency
Deferred capital

Operating Costs
Well field replacement
Manpower
Chemical (reagent)
Utility
Operating and maintenance supplies
Makeup water
General and administrative

Operating costs listed above are considered

direct operating costs. Other noncash costs, such
as depreciation and depletion, are also calculated
in the discounted cash flow analysis (DCF). Royalties and local taxes are likewise estimated in the
DCF analysis.
The costs presented above represent
those categories computed by submodels dealing with
capital and operating cost estimates which all feed
into the DCF analysis.
The process analysis component of the model
essentially sizes the project.
The basis userspecified site conditions are translated into
requirements for extraction plant size (gallons per
minute feed rate) and well field size (number of
patterns and well fields)
to meet the userspecified production level.
Manpower, horsepower,
and well field replacements are also computed.
These requirement calculations serve as the
basis for appropriate cost assignments from the
model data base. A simplified overview of the cost
model process is presented in Figure 1.
In
Figure 1,
submodels 1
and
2
develop
requirements and costs for the respective categories listed. Submodels 3 and 4 take both capital
and operating cost components of the well field and
extraction plant and generate total costs for each
of these categories.
The DCF analysis of submodel 5 solves for the sales revenue per pound U308
or the rate of return on equity.
Solves for minimum required sales revenue per pound
of U328 or rate of return on equity.
The discounted cash flow analysis submodel provides the
mechanism for either of the above solution options.
Solving for one of these options requires knowledge
of the other as a model input.
The DCF analysis
uses a profit and loss statement structure for
financial analysis of the project.
A user solving for sales revenue per pound of
U308 is interested in determining the minimum sales
price that is adequate to cover all operating costs
and capital recovery expenses and to provide a
specified rate of return on equity invested in the
project.
In this situation, sales revenue per
pound assumes an interpretation of economic cost of
production.
A user solving for rate of return on equity
will be employing a known market price of U308 as

SPE 12054

input to the model. This situation will be testing

the viability of a project in terms of its rate of
return yield at the anticipated market value.
Either of the solution options can provide
valuable planning information for property screening
or for testing alternative production levels or
other project design factors for a given or body
configuration.
Contains regionalized data bases.
The distinction
between costs incurred in Texas, Wyoming and New
Mexico project sites has been incorporated into the
data base of the cost model. The primary variations
recognized by the data base account for equipment
price differences for similar equipment as well as
for equipment and/or process system preferences
typically associated with each region.
A summary of the key data base distinctions for
regional sites follows:
Extraction Process Equipment
o

Defines
ion-exchange
employed.

Allows user four options (Up flow fixed

bed,
Downflow,
Upflow
Porter,
Upflow
USBM).

Defines cost differences for each ionexchange system in each region for each
equipment item for three plant sizes (400,
1000, and 2000-gpm plant feed rates).

system

typically

Well Field Equipment

o

Defines drilling and casing cost differences for each region for depth categories
up to 2500 ft.

Defines surface piping cost differentials

for insulated piping versus standard PVC
piping.

Capital Costs
o

Defines schedule of expenditures for each

capital item according to site location.

Defines permitting cost differentials and

time involved according to region.

Restoration system selection and therefore

cost is based on region.

Operating Costs
o

Incorporates
differentials
in
reagent costs between regions.

Incorporates
preferences
for
leaching
solution
between
regions
(ammonium
carbonate-bicarbonate for Texas,
sodium
carbonate-bicarbonate for Wyoming).

Includes differences in
costs between regions.

power

chemical

and

labor

Incorporates differences
and royalty charges.

in

local

taxes

Allows for cost analysis applications under a wide

range of information availability conditions.
To
accommodate the broadest possible applications, the
input structure of the cost model has been organized into three categories:

1.
2.
3.

Required input
Optional input (default values)
Calculation override input

These three categories represent the range of

information
availability
regarding
a
uranium
in situ leach mining project.
Category 1 contains
the basic input parameters that must be known about
a project in order to initiate a model run.
There
are 13 input parameters in category 1, organized
according to physical, operating, and financial
characteristics.
Examples of these parameters are
depth, ore grade, ore thickness, annual production,
productive
life,
and capitalization parameters.
There is also a series of cost update factors for
adjusting the base year of the costs.
Category 2 input included more detailed characteristics of the project which may not always be
known.
These parameters are assigned default
values in most cases according to the site location.
These default values will be used by the
model calculation procedures unless the user specifies another value in the input sequence.
Examples
of parameters in this category include leach solution, pattern type, solution grade, oxidizer, IX
process, injection - production well ratio, pattern
spacing, production well flow rate and operating
schedule.
Category 3 includes those cost and requirement
parameters that are calculated by the model.
If
information is available, however, on the specific
costs of one or more of the parameters included in
this listing, the model user may input the value
when initiating the model run.
This procedure will
then negate any model calculation required for the
subject parameter and instead use the value established by the user input.
Accepts and accounts for either static or dynamic
site conditions throughout the project life.
This
feature
relates
to
the ore deposit
geometry,
chemistry, well field design, and anticipated flow
rates and solution grades.
The model user is given
the
option
of
specifying
constant
conditions
throughout the project life for 10 parameters or
varying the conditions for each succeeding set of
well field patterns.
When changing site conditions are more appropriate than using average values for the selected
input parameters, the model user may specify changing values in terms of absolute or percentage
values.
To further demonstrate this option, the
following example of ore body depth changs over the
life of the project is presented:
Depth

GEORGE W. TOTH

SPE 12054

= 400, 420, 480, 410

In this case, the input

field 1 is 400 feet deep, well
deep, and well fields 3 and
410 feet,
respectively.
A
increase (or decrease) from the
could have been input.

indicated that well

field 2 is 420 feet
4 are 480 feet and
constant
percentage
starting value .also

Values for up to 30 well fields may be input in

this manner.
Other parameter inputs that may be
varied in a similar fashion include:
Depth of Deposit
Thickness of Deposit
Ore Grade
Solution Grade
Well field pattern
type

Injection to production
well ratio
Injection to production
well scrubbing
Production well flow
rate
Monitor well fraction of
total wells
Recovery or percent of
contained reserves

Accepts
user-specified
capitalization
structure
options.
The model user may indicate any debtequity capitalization structure for the project
being analyzed. Further, the length of loan payback
as well as the debt serv1c1ng rate may also be
established by user input.
This flexibility offers obvious advantages for
testing the effects of alternative project financing
arrangements and the sensitivity of rate of return
or cost per pound of U308 to variations in any of
the capitalization parameters.
All of the above-mentioned features are indicative of a costing tool which has been designed for
maximum user flexibility.
The model will facilitate rapid sensitivity
testing of the effect of the project parameter
changes on cost results for a specific site.
It
will serve a useful function in preliminary screening of properties for economic viability.
Alternative well field designs or extraction plant systems
may likewise be quickly examined.
The model is not designed to predict key project parameters such as production well flow rate or
solution grade based on permeability, depth, or
other influencing factors.
Many parameters, however, have been assigned default values based on
regional location which may be used or overridden by
model users.
The primary value of this model is its ability
to quickly translate any user-dictated values for
such parameters into overall project cost and design
implications.

MODEL APPLICATION
When applying the model to hypothetical but
representative
site
conditions
found
in Texas,
Wyoming, and New Mexico, the project parameters are
as shown in Table 1.
These conditions simulate
common mining situations in the three regions.
All
three are designed to recover 5,000,000 lbs U308

IN-SITU PRODUCTION COST MODEL FOR URANIUM

over an S year production life.

The
these cases are presented in Figure 2.

results

of

The results of these case runs are surprisingly

close.
New Mexico costs are highest at $34.34/lb
U30S followed by Texas costs at $34.l3/lb and Wyoming at $33.l6/lb.
All cost categories are relatively commensurate in each region with the exception of state/local taxes, royalties and annual well
field replacement costs.
The Wyoming well field costs are highest due
to the higher injection to production well ratio as
well as the closer spacing of wells.
The New
Mexico case exhibits the lowest cost/lb for wellfield replacement even though the operating depth
is 2000 ft versus 400 ft for Texas and Wyoming.
This extreme depth variation is offset by a higher
solution grade, higher production well flow rates
(30 gpm versus 20 gpm for Texas and 10 gpm for
Wyoming),
and greater injection/production well
spacing.
All of these factors contribute to a
requirement for significantly fewer wells in the
New Mexico case which offsets the greater expense
of drilling and casing at 2000 ft depths.
The higher costs for well field replacement in
Wyoming are more than offset by the more favorable
state/local taxes as well as the lower royalty payments.
In Wyoming, these rates are based on the
former AEC Circular 5 guidelines which computes on
adjusted project value (for tax purposes) based on
the average ore grade.
For testing the sensitivity of production cost
to changes in project conditions, any of the input
parameters may be analyzed.
The results of two
such tests are persented in Figures 3 and 4.
These
graphs are computer generated outputs of sensitivity analyses on the New Mexico base case.
The two
parameters tested were well spacing (injection to
production well spacing) and depth.
By making
interactive runs of the model, varying the parameter in question and holding all others constant,
the sensitivity patterns of Figures 3 and 4 were
generated.
Well spacing exhibits the greater influence on
production cost when compared to the depth sensitivity tests.
Overall, Figure 3 indicates that a
2.5 percent decrease in well spacing created a
In the
1 percent lowering of the production cost.
depth analysis, each 3 percent increase in depth
created a cost increase of 1 percent.
The pattern

SPE 12054

of sensitivity affects for other parameters tested

are listed below in order of decreasing influence:
1.

2.
3.

4.
5.
6.
7.

S.

Pattern spacing
Solution grade
Production level
Rate of return
Ore thickness
Ore grade
Ore depth
Production well flow rate

SUMMARY AND CONCLUSIONS

Over the last several years numerous companies,
state and federal agencies have used the model as a
predictive tool in deposit evaluations or in sensitivities analyses to answer the "what if" questions
such as "what if I change the well spacing," or
"what if I increase the pregnant solution grade."
It is expected that the model will find even wider
applications now that the capability to analyze the
New Mexico region has been added. The primary value
of the model has been, and will continue to be, it's
use as a preliminary screening tool to provide rapid
comparisons of alternative sites or site conditions
based on a consistent analysis framework.
In the last several years the Bureau of Mines
has provided, free of charge, copies of the in situ
mining cost model after the requestor has sent a
tape with the following specifications: 9 track,
160 bpi, ASCII code, no label, fixed SO characters
per record, and blocked 10 records per block.
During the last IS months sixteen companies, one state
agency and one federal agency have requested copies
of the model.
Several companies have provided the
Bureau with validation information or with suggestions on how to improve the model.
Current and future work on the cost model
focuses on converting the data base and program
structure to include other commodities. For example
a data base is being generated and equations modified to develop a gold in situ mining cost model.
Gold is probably the next commodity to be seriously
considered for either a "true" or modified in situ
mining scenario.
Gold in situ mining has a number
of inherent advantages due to its high unit value
and for its appeal to the small operator.
Other
commodities that may be considered for in situ
mining (and therefore in situ model application) are
silver, manganese, cobalt, nickel, lead, and zinc.

TABLE 1

MODEL APPLICATION PROJECT PARAHTERS

Category 1 Input

Texas Values

Wyoming Values

Hew Mexico Values

Depth
Ore Thickness
Ore Grade
Annual Production
Productive Life
Rate of Return
Proj. Start/Base Year
Debt Financed Portion

400 ft
10 ft
.1%
625,000 lbs/yr
8 yrs
20%
1980
0

400 ft
10 ft
.1%
625,000 lbs/yr
8 yrs
20%
1980
0

2000 ft
12 ft
.11%
625,000 lbs/yr
8 yrs
20%
1980
0

70%
Anononia
2
Downflow
50 ppm
5 spot
2:1
50
20 gpm
350 days/yr
7.5% Sales Rev.
10% Sales Rev.

70%
Sodium
2
Upflow-USBM
80 ppm
7 spot
3:1
40
10 gpm
340 days/yr
Per AEC Circ.
Per AEC Circ.

70%
Sodium
02
Upflow-USBM
85 ppm
5 spot
2.5:1
140
30 gpm
340 days/yr
7.03% Sales Rev.
10% Sales Rev.

Category 2 Input

Recovery
Leachant
Oxidizer
Extraction Process
Solution Grade
Pattern Type
Inj/Prod. Well Ratio
Inj/Prod. Well Dist.
Prod. Well Flow
Operating Schedule
State/Local Tax Rate
Royalty Payments

PROCESS ANALYSIS
SUBMODELS

MODEL
INITIALIZATION
User and

1.

Well field analysis

default input

2.

Extraction plant
analysis

3.

Capital cost
analysis

4.

Operating cost
analysis

FINANCIAL ANALYSIS
SUBMODEL

5.

Discounted
cash flow
analysis

parameters

!f
MODEL
DATA
BASES

Fig. 1-Simplified overview of the cost model process.

50

45

40

35
NET PROFIT
FEDERAL TAXES

30
=

~ 25

$34.34 ._-

$34.13
$33.16

2.42
2.02

_._--STATE!LOCAL
TAXES

2.56

ROYALTIES

3.41

20

3.43

4.34

"~

--

.3/

/
./

4.06

-v/

15

10

.90

./

4.40

-DEPRECIATION

/4.51

2.06

./

/'

DEPLETION
ALLOWANCE

------

-- --

//'

='
.....

--

----- ---

2.65

2.45

1",,-

4.84

5.01

10.24

/'

1----

/
5

8.49

ANNUAL
WELLFIELD
COSTS

2.40

"'-

"" ""

10.05

DIRECT
OPERATING
COSTS

"'"

2.20

5.02

9.90

"" ""

3.91

o
TEXAS

WYOMING

NEW MEXICO

COST BREAKDOWN COMPARISON

OF TEXAS, WYOMING, AND NEW MEXICO PROJECT CONDITIONS
Fig. 2-Cost breakdown comparison of Texas, Wyoming, and New Mexico project conditions.

40

35

30

25

=
=
crJ

=
=
-'

--tI'3-

~20

en

=
=
=
........
l=
=
=
=
t....:)

t....:)

n...15

10

BASE CASE SPACING

1I
I

r
:

100

110

120

130

140
150
WELL SPACING (FT)

160

170

180

BASE CASE SENSITIVITY TO WELL SPACING

Fig. 3-Sase-case sensitivity to well spacing.

 40

35

30

25
=
=
CT"J
=

-=
-..J

......

~20

I~

CJ"J

=
<-.:>
=
r=
=
=
=
0...15

=
~

IO..)

BASE CASE DEPTH

10

1800

1900

2000
ORE DEPTH [FT)

2100

BASE CASE SENSITIVITY TO ORE DEPTH

Fig. 4-8ase-case sensitivity to ore depth.

2200