Planning a Tailing Investigation by using XRF - CPT

Fugro Workshop: Innovative Tailings Characterization and Remining

Content

 Case Study: Homogeneity of a Tailing

 Optimizing Performance

 Planning the grid

 Szenario 1: Single source delivery – distal distribution

 Szenario 2: Multiple source delivery

 Szenario 3: Multiple source delivery – 3 patterns

2 www.fugro.com
Content

 Case Study: Homogeneity of a Tailing

 Optimizing Performance

 Planning the grid

 Szenario 1: Single source delivery – distal distribution

 Szenario 2: Multiple source delivery

 Szenario 3: Multiple source delivery – 3 patterns

3 www.fugro.com
Proximal - Distal: Pb-Zn Tailing Bad Grund, Germany

Tailing from Pb-Zn-Cu-Ag Mine

operated until 1992

16 Mt Ore (1 Mt Pb, 0.7 Mt Zn und 2.5 kt Ag)

Tailing operated from 1961 - 1987

4 www.fugro.com
Proximal - Distal: Pb-Zn Tailing Bad Grund, Germany

Standort 1 Standort 2

XRF 1.1 XRF 1.2 M5

0.7m M1 M2 XRF 2.1 XRF 2.2 0.5m
M3 M4
XRF 1.3
ca.1.5m ca.1.5m
proximal distal proximal distal

5 www.fugro.com
Case Study: Pb-Zn Tailing Bad Grund, Germany

MOSTAP Liner Sampling System

Static Push System, socket for reducing friction
High Quality soil sampling for XRF data calibration

6 www.fugro.com
Case Study: Pb-Zn Tailing Bad Grund, Germany

MOSTAP Liner for calibration lab testing

ID of Liner = 36mm, L = 1m
System for discrete sampling / 0,5 h per liner

7 www.fugro.com
Case Study: Pb-Zn Tailing Bad Grund, Germany
Vertical Scale 1:125 (DIN A3)
XRF 1.3

Cone Resistance [MPa] Friction Ratio [%] Arsenic [ppm] Lead [ppm] Copper [ppm] Zinc (ppm)

[m] 0 10 20 30 40 50 60 70 0 2 4 6 8 10 0 500 1000 0 5000 10000 0 500 1000 0 20000 40000 [m]
0 0
1 1 1 1 1 1

2 2 2 2 2 2

3 3 3 3 3 3

4 4 4 4 4 4

5 5 5 5 5 5

6 6 6 6 6 6

7 7 7 7 7 7

8 8 8 8 8 8

9 9 9 9 9 9

10 10 10 10 10 10

11 11 11 11 11 11

12 12 12 12 12 12

13 13 13 13 13 13

14 14 14 14 14 14

15 15 15 15 15 15

16 16 16 16 16 16

17 17 17 17 17 17

18 18 18 18 18 18

Notes:
Robertson CPT Soil Classification 1986 (modified) Soil Density and Consistency (interpreted) Legend Cone Resistance Profile Legend
100,0
Legend (Colors in Friction Ratio Profile) very loose
100
10 Contaminant Distribution measured with XRF Probe [ppm]
1 Sensitive, fine grained loose
12
9 11 2 Organic soils, peat
medium dense
Cone Resistance [MPa]

8 3 Clay
Cone Resistance [MPa]

10,0
7 4 Clay to silty clay 10 dense
6 5 Clayey silt to silty clay
very dense
5 6 Sandy silt to clayey silt
3
7 Silty sand to sandy silt very soft Project: 510-16-608 Bad Grund
1,0 4 1
8 Sand to silty sand soft
Test Location: XRF 1.3
9 Coarse to medium sand
1 firm
2 10 Gravel to gravelly sand Client:
0,1 11 Very stiff, fine grained 0,1 stiff
Bohrfirma: Fugro Consult GmbH
12 Very stiff sand to clayey sand
very stiff
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 Contractor: ST
Soil types 11 and 12 are heavily
Friction Ratio [%] Friction Ratio [%] hard
overconsolidaded or cemented. Test Date: 21.09.2016 Depth (CPT): 18,66 m bgs

8 www.fugro.com
Case Study: Distal logs

9 www.fugro.com
Case Study: Proximal logs
Vertical Scale 1:125 (DIN A3)
XRF 1

Cone Resistance [MPa] Friction Ratio [%] Arsenic [ppm] Lead [ppm] Copper [ppm] Zinc [ppm]

[m] 0 10 20 30 40 50 60 70 0 2 4 6 8 10 0 500 1000 0 5000 10000 0 500 1000 0 20000 40000 [m]
0 0
1 1 1 1 1 1

2 2 2 2 2 2

3 3 3 3 3 3

4 4 4 4 4 4

5 5 5 5 5 5

6 6 6 6 6 6

7 7 7 7 7 7

8 8 8 8 8 8

9 9 9 9 9 9

10 10 10 10 10 10

11 11 11 11 11 11

12 12 12 12 12 12

13 13 13 13 13 13

14 14 14 14 14 14

15 15 15 15 15 15

16 16 16 16 16 16

17 17 17 17 17 17

18 18 18 18 18 18

Robertson CPT Soil Classification 1986 (modified) Soil Density and Consistency (interpreted) Legend Cone Resistance Profile Legend Notes:
100.0
Legend (Colors in Friction Ratio Profile) very loose Contaminant Distribution measured with XRF Probe [ppm]
100
10 X.1
1 Sensitive, fine grained loose
12
9 11 2 Organic soils, peat X.2
medium dense
Cone Resistance [MPa]

8 3 Clay X.3
Cone Resistance [MPa]

10.0
7 4 Clay to silty clay 10 dense
6 5 Clayey silt to silty clay very dense
5 6 Sandy silt to clayey silt
3
7 Silty sand to sandy silt very soft Project: 510-16-608 Bad Grund
1.0 4 1
8 Sand to silty sand soft
Test Location: XRF 1
9 Coarse to medium sand
1 firm
2 10 Gravel to gravelly sand Client:
0.1 11 Very stiff, fine grained 0.1 stiff
Bohrfirma: Fugro Consult GmbH
12 Very stiff sand to clayey sand
very stiff
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 Contractor: ST
Soil types 11 and 12 are heavily
Friction Ratio [%] Friction Ratio [%] hard
overconsolidaded or cemented. Test Date: 21.09.2016 Depth (CPT): 18.66 m bgs

10 www.fugro.com
 Case Study: Proximal Logs (Details)

XRF 1

Lead [ppm] Copper [ppm] Zinc [ppm]

1000 0 5000 10000 0 500 1000 0 20000

1 1 1

2 2 2

3 3 3

4 4 4

5 5 5

6 6 6

7 7 7

8 8 8

9 9 9

10 10 10

11 www.fugro.com
11 11 11
Content

 Case Study: Homogeneity of a Tailing

 Optimizing Performance

 Planning the grid

 Szenario 1: Single source delivery – distal distribution

 Szenario 2: Multiple source delivery

 Szenario 3: Multiple source delivery – 3 patterns

12 www.fugro.com
How to plan a high resolution tailings investigation

 Gathering as much historical information as possible

 Understanding the history of the tailing (time of operation, design, mines, ores, etc.)

 Getting an idea about the morphology (basis, surface)

 Designing the approach (grid / transects) - planning the grid

 Adjusting the grid depending on real time results (staged approach)

 Geotechnical and soil parameter – what is required ?

 Resolution – what details do we need ?

 Planning the samples (criteria, number, intervals, rule of thumb: 10% ratio)

 Parameter for lab testing (elements, minerals)

The aim of the CPT-XRF investigation is a most accurate 3D model

13 www.fugro.com
Production versus Results

XRF CPT speed

measuring time detection limit 2 1 0,5
[sec] [ppm] [cm/sec] [cm/sec] [cm/sec]
resulting vertical resolution [cm]
30 100 - 250 60-70 30-35 15-18
15 150 - 400 30-35 15-18 7-9
5 > 300 10-13 5-7 2-5

Time per 25m [min] 20 40 80

Performance / day [m] 150 120 90

 Lower speed = Higher resolution (shorter intervals) but lower performance

 Higher measuring time = Lower detection limit but lower performance
 Lower Measuring time = High detection limit but high performance

14 www.fugro.com
The Challange of Optimization

Quality
(CPT + XRF)

Time Performance
(CPT + XRF)

Optimizing

Resolution Detection Limit

(XRF) (XRF)

Costs

15 www.fugro.com
Sampling Technique - MOSTAP

• Discrete sampling
• Obtain as nearly “undisturbed”samples
• Porous nylon stocking minimizes
friction during sampling
• 35mm OD sample
• 1 m length of each liner

16 www.fugro.com
Sampling Technique - MOSTAP

• Static push
• Pushed by CPT unit

Operation:
• Push to top of interval
• Unlock the tip from surface
• Push further for 1m
• Pull up rods with sampler

Time ca. 0,5h per liner

Not for continuous coring !

17 www.fugro.com
Sampling and Lab Testing

Sampling:
• Use the high resolution real time data for identifying the ideal sampling intervals
• Choose intervals from homogeneous soil and similar concentration ranges for calibration
• Choose intervals of high concentrations and from different soil for mineralogy

1. Samples for verification (XRF calibration)

 Understanding matrix effects (different effects depending on grain size and minerals)
 Ideal are homogeneous soil type areas with different concentrations levels
 Methodology: Chemical assays
 Also possible: Composite samples (1m length) for avoiding uncertainties based on
heterogeneity

2. Samples based an elevated metal concentrations in the XRF logs (Mineralogy)

 Understanding the mineralogy (grain size, minerals, liberation)
 Primary or secondary minerals ? – important for leaching
 Methodology: Microscope, SEM, Quemscan, …

18 www.fugro.com
Sampling Intervals

19 www.fugro.com
Lab Testing

Automated Mineralogy

• integrating largely automated measurement techniques

based on Scanning Electron Microscopy (SEM) and
Energy-dispersive X-ray spectroscopy (EDX).

• synchronous acquisition of two types of information:

1) Mineral chemistry
2) Mineral texture (microstructure)

• Advanced software automation enables acquisition of

1) basic system data – e.g. modal mineralogy, grain
sizes and shapes, mineral associations and mineral
liberation
2) calculated data – e.g. elemental chemistry, matrix Processed MLA-Image of crushed and classified
density or grade-recovery curves copper ore showing particles composed of either
one, two or several intergrown minerals
• Robust statistics typically based on hundreds of thousands
of underlying individual measurements and mineral
classifications

• Systems available: QEMSCAN and Mineral Liberation

Analyzer (MLA) from FEI Company, Mineralogic from Zeiss

20 www.fugro.com
Lab Testing - Mineralogy

SEM Image

5 mm

Processed MLA Image

21 www.fugro.com
Content

 Case Study: Homogeneity of a Tailing

 Optimizing Performance

 Planning the grid

 Szenario 1: Single source delivery – distal distribution

 Szenario 2: Multiple source delivery

 Szenario 3: Multiple source delivery – 3 patterns

22 www.fugro.com
Planning the Grid

 The grid design depends on the history

 Processing plant history (how was the efficiency of processing evolved over time)
 Ore Bodies from the mine (how has the ore body been mined?)
 Delivery by a single mine / multiple mines
 Discharge pattern (single, multiple, complex)

 Is a single stage / multi stage approach best ?

 Transects or grid ?

 Grid design: Spacing / regular versus irregular / different grid patterns

 Flexible approach: Use the real time data for adjusting the grid

 Distance (correlation length) between grid points depends on the information

which is available
 Correlation length (how to identify?)
 Differs depending on the (expected) structure of the tailings deposits
 Experienced project manager sets up the x/y/z correlation parameters (Standard is 10/1)

23 www.fugro.com
3 D modeling

Deterministic interpolation methods

Inverse-Distance-Weighting Methode (IDW)

• The value of a variable at a not measured point will be estimated over a weigthed average of the
values at the neighbouring measured poins

• The weigths are proportional to the inverse distance between the measured and estimated points

1
 d  z (i )
z 0   i
1
d
i

z(i) - measured values at points „i“

z(0) - estimated value at not measured point „0“)
di - distance between measured points „i“ and not measured point „0“

24 www.fugro.com
Planning the Grid

3D-Model of uncertainties

The correlation length is the maximum tolerable distance

between data points

It varies depending on the homogenity of the tailing

x - y can vary slightly od significant (single source)

x/y - z veries very significant (factor 1:10 is rule of thumb)

min

max

25 www.fugro.com
Content

 Case Study: Homogeneity of a Tailing

 Optimizing Performance

 Planning the grid

 Szenario 1: Single source delivery – distal distribution

 Szenario 2: Multiple source delivery

 Szenario 3: Multiple source delivery – 3 patterns

26 www.fugro.com
History: Szenario 1 (single source – distal distribution)

27 www.fugro.com
History: Szenario 1 (single source – distal distribution)

28 www.fugro.com
History: Szenario 1 (single source – distal distribution)

Tailing dam

Tailing pond
(periodically dry)

Tailing dam

Single Point Discharge

29 www.fugro.com
History: Szenario 1 (single source – distal distribution)

distal

decreasing
grain size

proximal

30 www.fugro.com
History: Szenario 1 (single source – distal distribution)

Tailing dam

31 www.fugro.com
History: Szenario 1 (single source – distal distribution)

coarser material Tailing dam

fine grained deposits

32 www.fugro.com
How to plan the grid ?

33 www.fugro.com
Content

 Case Study: Homogeneity of a Tailing

 Optimizing Performance

 Planning the grid

 Szenario 1: Single source delivery – distal distribution

 Szenario 2: Multiple source delivery

 Szenario 3: Multiple source delivery – 3 patterns

34 www.fugro.com
History: Szenario 2 (multiple sources)

35 www.fugro.com
History: Szenario 2 (multiple sources)

36 www.fugro.com
History: Szenario 2 (multiple sources)

Fine Sand Silt / Clay

37 www.fugro.com
How to plan a grid?

38 www.fugro.com
Content

 Case Study: Homogeneity of a Tailing

 Optimizing Performance

 Planning the grid

 Szenario 1: Single source delivery – distal distribution

 Szenario 2: Multiple source delivery

 Szenario 3: Multiple source delivery – 3 patterns

39 www.fugro.com
History: Szenario 3 (multiple sources)

40 www.fugro.com
History: Szenario 3 (multiple sources)

3 different sources (mines) / 3 different patterns = correlation length

41 www.fugro.com
History: Szenario 3 (multiple sources)

Correlation length?

3 Areas

Different approaches
required

42 www.fugro.com
Thank you for your attention

m.wibbeling @fugro.com