New Projects: Conventional or High Density Tailings?

badwaterreflectiveΠολεοδομικά Έργα

29 Νοε 2013 (πριν από 4 χρόνια και 1 μήνα)

91 εμφανίσεις

Davi d Romo

Paterson
& Cooke, Chi l e

Ray
Marti nson

Paterson
& Cooke,
Chi l e

New
P
rojects
:
C
onventional

or

H
igh
D
ensity

T
ailings
?

Every new mining project has to define the most important aspects for the business:


Mineral resources model


Mine
explotation

design


Methallurgicall

process



The challenges of new mining projects

After that, the main consumptions of the process have to be evaluated:


Water consumption


Energy consumption



At the end, the projects have to define what to do with the waste of the
process:


Sterile material


Tailings



In this case Paterson & Cooke office in Santiago was in charge of the
definition of tailings management system of the project.


Aim of the study

The aim of this presentation is to show the issues we had to face and
what we did to deal with them, in order to support the definitions of
the Client.

“What we have to do is to look at the whole picture of the problem”

1.
Project overview

2.
Tailings deposit location

3.
Bench scale tests

4.
Conceptual trade off: conventional or thickened tailings?

5.
Semi pilot test work

6.
Reviewing the solution

7.
Conclusion




Content

The

study

adresses

a

new

project

of

Sociedad

Punta

del

Cobre
,

the

greatest

medium

copper

mining

company

in

Chile
.


It is a gold
-
copper mine located at the IV region of Chile, around 300 km
north of Santiago


It is located in a desert area, the annual rainfall is less than 200 mm.


The project considers a plant to process over 15,000
tpd


The tailings deposit considers a capacity over 100 million tonnes


1. Project
O
verview

Paterson & Cooke led a consulting team to design the tailings system for the project

1. Project
o
verview

2010

2011

2012

2013

2014
-
15

2016

TRANSITION

ENGINEERING

PREFEASIBILITY
STUDY

CONSTRUCTION

FEASIBILITY

STUDY

PROFILE

STUDY

COMMISIONING

DEPOSIT
LOCATION


PRELIMINARY
ARRANGEMENT

BENCH SCALE

TESTS


CONCEPTUAL
TRADE OFF

SEMI PILOT

TESTS


UPDATE THE

SOLUTION

PATERSON & COOKE LED THE TAILINGS
CONSULTING TEAM

PUMPING

TAILINGS

SYSTEM

DESIGN

SUB
-
CONTRATOR

OF AMEC CHILE


Based on satellite survey,
11 potential sites were
identified closer than 30
km to the plant.


The storage target was 150
million tonnes.


A comparison was done
considering:


Conventional tailings pumping
head


Recovered water pumping
head


Dam wall volume


Existing human activity


Three alternatives were
selected to complete the
trade off comparison


2.
Tailings

Deposit

Location

Alt. 1
Alt. 2
Alt. 3
Alt. 4
Alt. 5
Alt. 9
Alt. 8
Alt. 10
Alt. 11
Alt. 6
Alt. 7
PLANT
The evaluation of each alternative
considered:


Storage capacity of 80 million
tonnes


Processing plant of 17,000
tpd


Conventional tailings with 55%cw


Sand wall from
cycloned

tailings

The results were the following:

2.
Tailings

Deposit

Location


Al ternati ve

2

Al ternati ve

4

Al ternati ve

5

PLANT

LOCATI
ON

Parameter

Unit

Alternative

2

Alternative

4

Alternative

5

Starting

wall

volume

Mm³

1.5

1.3

0.8

Final

wall

volume

Mm³

11.1

5.0

4.4

Final
wall

height

m


102

81

85

Slurries

and
water

pipeline
length

m


8.4

36.2

56.1

Total
installed

power

kW

1,249

6,669

7,759

Plant

make
-
up
water

m³/t

0.54

0.55

0.64

Due

to

lower

Capex

and

Opex
,

lower

installed

power,

best

plant

make
-
up

water

and

lower

environmental

impact,

Alternative

2

was

selected
.

In

order

to

get

slurry

design

parameters,

bench

scale

tests

were

done

on

samples

of

each

UGM

and

on

a

composite
.


The

samples

were

got

from

metallurgical

bench

scale

tests
.


The

average

p
80

was

148

μ
m,

61
%

less

than

75

μ
m

and

Sg

of

3
.
03

Sedimentation

tests

were

done

by

Delkor
:


Magnafloc

338

was

the

best

floculant
,

floc

dose

between

6
-
12

g/t

and

dilution

of

13
%
cw



Depending

on

the

sample,

a

paste

thickener

could

produce

tailings

between

68
.
5
%

-

74
.
1
%
cw,

being

the

expected

average

equals

to

71
.
5
%
cw
.

3.
Bench

S
cale

Tailings

T
ests

0
10
20
30
40
50
60
70
80
90
100
1
10
100
1000
% Passing
Particle size (micron)
Tailings Particle Size
Distribution
SF UGM1
SF UGM2
SF UGM4
SF UGM5
SF UGM6
Composite Sample

A

range

of

rheology

was

measured

in

rotational

viscometer
.



The

unsheared

yield

stress,

for

the

expected

operational

thickening

points
,

was

determined

between

15

and

40

Pa
,

which

is

considered

a

low

rheology
.


The

rheology

reduction

was

measured

at

30
%
,

so

the

expected

fully

sheared

yield

stress

was

between

10



28

Pa
.

At

the

average

thickening

point

the

expected

fully

sheared

yield

stress

was

19

Pa
.

3.
Bench

S
cale

T
ailings

T
ests

0
10
20
30
40
50
60
70
80
66%
67%
68%
69%
70%
71%
72%
73%
74%
75%
76%
Yield Stress (Pa)
Solids Concentration (Cw)
Bingham Yield Stress vs Solids Concentration
UGM1 (F)
UGM2 (F)
UGM4 (F)
UGM5 (F)
UGM 6 (F)
Composite Sample (F)
Design Curve
0
10
20
30
40
50
60
70
80
66%
68%
70%
72%
74%
76%
Yield Stress (Pa)
Solids Concentration (Cw)
Bingham Yield Stress vs Solids Concentration
Minimum
Rheology
Maximum
Rheology
In

order

to

estimate

the

beach

slope

at

the

deposit,

a

flume

test

was

done

on

the

composite

sample
.

The

limitations

of

flume

tests

for

predicting

beach

slopes

are

well

known,

so

the

measured

values

were

reduced

by

considering

two

reduction

factors
:

evaluating

the

average

slope

in

the

middle

of

the

flume

and

also

considering

a

reduction

of

30
%
.


3.
Bench

S
cale

T
ailings

T
ests

5.1%
2.2%
1.5%
3.6%
0%
1%
2%
3%
4%
5%
6%
0
5
10
15
20
25
30
35
40
45
50
Estimated Beach Slope (%)
Yield Stress (Pa)
Estimated Beach Slope vs Yield Stress (Fully Sheared)
Middle Profile
Slope
Recommended
Curve
With

those

considerations,

at

the

average

fully

sheared

yield

stress

of

19

to

20

Pa

at

the

deposit,

a

beach

slope

of

1
.
5
%

was

estimated

for

the

purpose

of

pre
-
feasibility

engineering

The

main

aspects

to

figure

out

in

the

project

in

terms

of

the

prefeasibility

study

were

as

follows
:


Deposit

location


Tailings

management

system


Dam

wall

and

storm

water

system

design

The

dam

wall

and

storm

water

system

design

were

undertaken

by

Knight

Piesold

Chile
.

Two

different

options

for

tailings

management

system

were

evaluated
:


Conventional

tailings

option
:

low

density

tailings,

cyclons

operation,

slimes

transport,

tailings

sand

wall

and

tailings

discharge

at

the

deposit

from

the

wall
.


Thickened

tailings

option
:

high

density

tailings,

wall

constructed

with

borrowed

material

and

tailings

discharge

at

the

end

of

the

deposit
.

4
. Conceptual
Trade

off:
C
onventional

or

T
hickened

T
ailings
?

4
. Conceptual
T
rade

off:

Parameter

Conventional
tailings

system

Thickened tailings system

Solids concentration

64%
CW

70%
CW

Thickener

(1)
×

Ø48 m high rate thickener, yield
stress < 15 pa

(1)
×

Ø48 m paste thickener,
yield stress < 80 Pa

Tailings transport

Gravitational flow through 4,300 m
HDPE pipeline

Gravitational

flow through

700 m HDPE pipeline

Tailings classification

3

+

3 cyclones


No
cyclones

or

pumping

slurries

systems

were

considered

Sand transport

2

+

2 slurry pumps TDH 20 m

Slimes transport

Gravitational flow through 1,475 m
steel pipeline

Recovered water transport

2

+

2 barge pumps and 4

+

2 water
pumps TDH 230 m

Starting dam wall

1.8 Mm³

1.0 Mm³

Final dam wall

12.5 Mm³

6.5 Mm³

Beach slope

0.3%

1.5%

Plant make
-
up water

0.46 m³/t

0.40 m³/t

Installed power

3,100 kW

1,700 kW

4
. Conceptual
T
rade

off

The

thickened

tailings

system

was

selected

because

of
:


Lower

capex


Lower

installed

power


Lower

plant

make

up

water




Lower

volume

of

water

within

the

deposit


Greater

opportunities

to

improve

the

using

of

the

deposit

facilities




Wall

Thickener

Plant

Location

Tailings

discharge

Semi

pilot

tailings

tests

were

done

in

order

to

improve

the

knowledge

of

tailings

on

a

larger

scale
.



The

p
80

was

218

μ
m

-

192

μ
m


Fine

fraction

less

than

75

μ
m

was

56
%
-
60
%

Slurry

behaviour

tests

were

done

by

P&C

Johannesburg
:


Natural

pH

7
.
8


Slurry

conductivity

over

40

mS
/cm

(sea

water)


Slurry

naturally

settling



5.
Semi

Pilot

T
ailings

T
ests

0
10
20
30
40
50
60
70
80
90
100
1
10
100
1000
% Passing
Particle size (micron)
Particle Size Distribution
-
Semi Pilot Samples
M1
M2
Semi

pilot

thickening

tests

were

done

by

Outotec

in

a

Ø
190
mm

x

4

m

thickener
:


Magnafloc

338

was

the

one

of

the

best

floculants
,

floc

dose

grew

up

to

17

g/t

and

dilution

was

12
%
cw



Optimum

flux

rate

was

0
.
4

t/m²
.
h


Expected

U/F

solids

concentration

65
%
-
73
%
cw


Unsheared

yield

stress

between

50
-
180

Pa


Two

options

were

proposed
:


One

55
Øm

high

compression

thickener


Two

38
Øm

high

compression/
paste

thickeners



5.
Semi

Pilot

T
ailings

T
ests


A

range

of

rheology

was

measured

in

rotational

and

tube

viscometer
.



A

great

difference

between

the

rheology

of

both

samples

was

measured
.

Even

the

rheology

was

6

times

greater

than

the

measured

one

at

the

PFS

study
.


A

rheology

reduction

of

60
%

was

measured

on

these

samples

5.
Semi

P
ilot

T
ailings

T
ests

0
20
40
60
80
100
120
140
160
180
200
50%
52%
54%
56%
58%
60%
62%
64%
66%
68%
70%
72%
74%
Bingham Yield Stress (Pa)
Solids Concentration (Cw)
Bingham Yield Stress vs Cw
Sample 1 Fully Sheared
Sample 2 Fully Sheared
Sample 1 Unsheared
Sample 2 Unsheared
Paste
Thickened
Conventional

4

m

flume

tests

were

done

to

estimate

the

beach

slope
.

The

experience

and

test

data

analysis

were

done

with

the

support

of

Professor

Andy

Fourie
.


Both

samples

showed

different

behaviour
,

due

to

the

different

lithology
.

The

beach

slope

estimation

considered

a

safety

factor

of

50
%

over

the

measured

slope
.


At

the

average

solids

content

for

the

thickener

operation

of

70
%
CW
,

the

average

beach

slope

was

estimated

on

1
.
6
%

5.
Semi

P
ilot

T
ailings

T
ests

1,6%
0,0%
0,5%
1,0%
1,5%
2,0%
2,5%
3,0%
64%
66%
68%
70%
72%
74%
76%
Estimated Beach Slope %
Solids Concentration Cw %
Estimated Beach Slope vs Solids Concentration
Estimated Beach
Slope M1
Estimated Beach
Slope M2
Operational Point
6.1 Tailings Pumping

Because

of

the

higher

rheologies

measured

at

semi

pilot

test

work,

pump

stations

are

required

for

tailings

transportation

from

thickener(s)

to

the

deposit
.

Depending

on

the

option

of

thickening

circuit

the

pumping

system

is

as

follows
:

6
.
Reviewing

the

S
olution

Pump

station

(1)
×

Ø55 m
thickener

(2)
×

Ø38 m
thickener

Underflow discharge

(1

+

1) 20 m/15 m
250

kW
centrifugal

pumps

2
×

(1

+

1) 20 m /15
m 200 kW
centrifugal

pumps

Transport to deposit

(1

+

1) 25 m/20 m 200 kW
centrifugal

pumps

6.2 Filtered tailings

Bench

top

filtering

tests

were

done

by

Delkor
,

but

the

filtration

rates

were

very

poor

for

vacuum

and

pressure

technologies
.

In

spite

of

those

results,

an

evaluation

of

filtered

tailings

system

was

done
.

The

area

of

the

deposit

is

not

suitable

to

dispose

tailings

in

stock

piles
.

This

is

because

of

the

maximum

expected

value

for

the

hydrographic

basin

storm

flow

which

reaches

211

m³/s
.

As

a

result,

a

water

dam

as

well

as

an

evacuation

tunnel

had

to

be

considered

in

this

evaluation
.

Finally

the

filtered

tailings

option

resulted

into

a

Capex

over

2
.
4

times

the

thickened

tailings

option
.

6
.
Reviewing

the

solution

The

study

developed

for

this

new

project

of

Sociedad

Punta

del

Cobre

ended

into

the

following

conclusions
:


The

site

conditions

are

the

most

important

aspects

that

define

which

type

of

dewatering

system

is

the

best
.

The

topography

and

the

hydrographic

basin

storm

determine

the

design

of

the

deposit

and

the

tailings

management

system
.

The

other

very

important

aspect

are

the

pumping

distances
.


If

the

material

presents

good

conditions

to

be

thickened

or

filtrated,

a

complete

test

plan

must

be

done

to

support

every

engineering

evaluation
.

Bench

scale

or

pilot

tests

are

recommended

to

be

done

for

each

stage

of

engineering
.


In

this

case
,

thickened

tailings

clearly

has

technical

advantages

over

conventional

and

filtered

tailings

systems
.

The

expected

plant

make
-
up

water

is

0
.
4

m³/t

and

the

installed

power

is

2
,
700

kW


For

feasibility

engineering,

the

final

selection

of

the

thickening

circuit

and

the

study

of

disposal

strategies

will

be

important

issues

to

address
.

7
.
Conclusion




We

want

to

acknowledge

to

all

people

who

worked

with

us
,
from

Knight

Piesold

Chile,
Multical
,
Delkor
,
Outotec
, Mr. Andy Fourie and
to

Pucobre

S.A.

Acknowledgements