Embed Size (px)
Citation preview
Downhole XRF-Logging:
A new tool to explore borehole rock composition
Bachmann, C.1, Bachmann, J.1 , Harms, U.2 and Hawke, P.³
1 J&C Bachmann GmbH, Bad Wildbad, Germany 2 GFZ - German Research Centre for Geosciences, Potsdam, Germany,
[email protected] ³ Pilbara Wireline Services, Osborne Park , WA , Australia
Abstract: A novel type of downhole logging probe has been developed that is capable to register continuously the chemical composition of borehole walls by combining XRF techniques with downhole logging. The new tool obtains quantitatively the content of elements with atomic weights higher than sodium along a stripe of a borehole wall. In mineral exploration, geochemical drill core analysis is a standard to derive metal contents of an underground mineral resource. Drill core is extracted, the mineralogy of core is assessed by rapid field camp methods and sampling is performed to send samples to a lab that provides results later. In contrast wireline borehole measurements provide immediately after drilling uninterrupted geochemical results over the complete drilled section. The new XRF tool provides quantitative geochemical data along the borehole wall that can be tuned to major element composition, to specific metal contents, or to geochemical markers such as element-element ratios. Alternative in situ downhole tools require utilization of neutron activation which is often difficult to handle or expensive due to environmental legislation. Measurements in dry borehole show results for iron ore content in comparison to laboratory results derived from chemical analysis in the laboratory as well as handheld XRF analyses of the borehole debris. The results of this measuring campaign performed in Australia show reasonable correlation between the different measuring methods. The diagram shows as an example the correlation of the iron content measured with different methods. The downhole probe performed the measurement continuously while it was fed down respectively up. The results were averaged to increments of 2 meters each in order to match the other measurement´s resolutions. Travel speed of the probe was 1 meter per minute. The results are shown in figure 2. Measurements were taken above the water table since the probe was not certified as pressure proof at the time of the measuring campaign. However, due to the kind of the physical measurement different calibrations have to be applied above and below the water table. XRF logging requires a short distance between borehole wall (sample) and x-ray source as well as detector in order to allow most effective characteristic x-ray fluorescence radiation intake. For this purpose the probe is equipped with eccentric arms pressing the instruments x-ray window and sensors as close as possible against the borehole wall. The arms are adapted to borehole diameter. Currently oilfield service companies apply logging standard wise to decipher physical conditions of rock and fluids down the well continuously and in situ.
Motivation to develop downhole XRF • Geochemical whole rock analyses of drill core is a usual technique
in exploration and scientific drilling
• The characterization of drilled rock composition is usually optical and other low-tech on-site inspection as well as sampling and off drillsite lab analysis
• The latter procedures require several days to weeks while the drilling may be demobilized and no onsite decision can be made about further drilling
• For an immediate, continuous on-site geochemical profile only downhole logging provides rock composition, ore content peaks, fluid precipitations and other specifics which are crucial for on-site decisions
Mineral Exploration Drilling
diamond wireline drilling w/coring
• Core analyses are time
consuming and the
discontinuity of cores (low
recovery rate) does not cover
the full drilled section
• Cutting analysis is hampered
by mixing over large depth
ranges, loss of circulation and
incomplete sampling
• Several petrophysical tools exist and are applied but only Spectral Gamma (K, Th, U) is useful for geochemical analysis
• Geochemical loggings tools are available since long (e.g., GLT of SLB or GEM of Halliburton)
• Analytical base is neutron activated gamma-ray spectroscopy
• Radioactive sources, large size, high costs and a limited element spectrum available preclude a far-reaching application, especially in the mining industry
Available Logging Techniques
Analytical Range
Si, Fe, Ca, S, Ti, Gd, Cl, H and some others are directly analysed several other elements are calculated
Halliburtons GEM
Laboratory XRF Instruments • Excellent standard tool in most
geochemcial labs • Large-size instruments • Complex sample preparation
(glass discs for main elements, pressed powder tablets for traces)
• HQ analyses of most metal elements starting from Na
Present x-ray tube size
• Small • Low power
(10–50 kV, 5–200 µA)
• Fast • Stable output • Variable
current and voltage
• Continuous operation
• Runs w/ USB & AC adaptor
• Consume 10 W
XRF principles
New possibilities: Handheld XRF
New Possibilities: Raw Materials Online XRF
Online XRF raw materials conveyor belt analyses identify and measure the concentration of elements Al (Z=13) to U (Z=92) Online non-destructive analyses of material composition directly above the conveyor belt with HQ and stability in a hostile environment including impacts, dust, low/high temperature and humidity; measurements are independent of lump size, relative humidity and distance within a range of some cm
XRF Sonde Design, Construction, Test Slimhole Prototype
Slimhole Prototype Constraints
• 53 mm outer diameter
• Up to 1000 m water pressure resistent
• Up to 80° C environmental temperature
• Decentralized position close to borehole wall
• Data acquisition via wireline
• Logging-while-tripping capability with memory
data acquisition
Borehole wall contact is needed
Radiation Attenuation Coefficient
Be
Li
H
B C water, air
Cu
Energy (keV)
Ray path design
The X-ray window challenge
Design constraints A. Absorption, and distance • Water is strongly absorbing – the thickness must be max. 0.5 mm • The sonde must be pressed onto the borehole wall and a mudplow
must clear the borehole wall immediately before data acqusition • The X-ray window must be as thin as possible to minimize
absorption, either Be or Cdiamond need metal support
B. Geometry • White and characteristic X-ray paths must be as short as possible C. Window inset • The X-ray window must be pressure tight connected into the tool housing
D. Cooling • Detector must be cooled through Peltier element, copper block, heat pipe
• Prototype tool (IBERIA) built into a
100 mm diameter casing
• Data continuously collected at a
running speed of ~1-2 m/min. Each
sample represents a 2 m interval
• Up and down runs compared to get
some measure of data precision
Decentralising “spring”
XRF Sonde Design, Construction, Test Blasthole Field Prototype
• Data calibration and processing completed using
software developed by the tool manufacturer
• XRF element estimates made by fitting the
calculated response against laboratory assays
• Best result achieved for Fe was a 3.4% margin of
error
Fe calibration comparing IBERIA results to assays from NS676
Blasthole Prototype Calibration
• Key peaks which can be clearly identified in the data include: Fe(a), Fe(b), Mn(a), Si(a), Ar(a), Ti(a), Ti(b)
• Secondary peaks include: Ca(a), P(a), S(a), K(a)(all other elements derived from these peaks)
Sample spectrum curve showing XRF response from an interval in NS676
Fe(a) Fe(b)
Mn(a) Ti(b)
Ti(a)
Ar(a)
Si(a)
total count response
residual curve
(Air response)
(part of tool)
System response peaks
Blasthole Prototype XRF Spectrum
• Comparative measurements made
using a REFLEX InnovX handheld XRF
on cutting piles
• Calibrated by supplier (REFLEX)
• Each sample was recorded over a one
minute measurement window
• Two readings were made in separate
locations on each pile for NS676 to
gain some measure of data precision
Comparison with Handheld XRF
0
10
20
30
40
50
60
0 10 20 30 40 50 60 70 80
NS676 -‐ Fe log traces
Fe_lab Fe_hand1 Fe_hand2 Fe_BH_in FE_BH_out
y = 1.0064xR² = 0.7408std.err = 2.19
0.00
10.00
20.00
30.00
40.00
50.00
60.00
0.00 10.00 20.00 30.00 40.00 50.00 60.00
NS676 -‐ Fe handheld repeat comparison
y = 0.9095xR² = -‐0.621std.err=2.55
0.00
10.00
20.00
30.00
40.00
50.00
60.00
0.00 10.00 20.00 30.00 40.00 50.00 60.00
NS676 -‐ Fe borehole in and out run comparison
outliers (tool at water table)
Handheld
• Wider range of values predicted – trend agrees with assays
• Fe over-estimated by 30% compared with lab (calibration error)
• Absolute precision 2.2% Fe from repeats
Logging
• Very little variability in assay values over section of hole run
• Better agreement with lab assays
• Absolute precision 2.45% Fe from in and out runs Affected by high noise outliers at WT (50 m)
Blasthole Prototype Fe-Results
y = 1.0008xR² = 0.3551std.err = 1.40
0.00
10.00
20.00
30.00
40.00
50.00
60.00
0.00 10.00 20.00 30.00 40.00 50.00 60.00
NS676 -‐ Si handheld repeat comparison
y = 1.0337xR² = -‐2.68
std.err=2.11
0.00
10.00
20.00
30.00
40.00
50.00
60.00
0.00 10.00 20.00 30.00 40.00 50.00 60.00
NS676 -‐ Si borehole in and out run comparison
0
10
20
30
40
50
60
0 10 20 30 40 50 60 70 80
NS676 -‐ Si log traces
Si_lab Si_hand1 Si_hand2 Si_BH_in Si_BH_out
Handheld
• Si under-estimated by about 50% compared with lab (calibration error).
• Abs. measurement precision 1.4% Si from repeats.
Borehole • Very good correlation
with lab assays. • Absolute precision
1.8% Si from in and out runs.
Blasthole Prototype Si-Results
y = 0.7381xR² = 0.5811std.err = 0.03
0.00
0.10
0.20
0.30
0.40
0.50
0.00 0.10 0.20 0.30 0.40 0.50
NS676 -‐ P handheld repeat comparison
y = 1.0064xR² = 0.7408std.err = 2.19
0.00
0.10
0.20
0.30
0.40
0.50
0.00 0.10 0.20 0.30 0.40 0.50
NS676 -‐ P borehole in and out run comparison
0
0.2
0.4
0.6
0.8
1
1.2
0 10 20 30 40 50 60 70 80
NS676 -‐ Mn log traces
Mn_lab Mn_hand1 Mn_hand2 Mn_BH_in Mn_BH_out
Handheld
• Insensitive to low levels of P found in most of hole.
Borehole • Limited range of
values over logged interval.
• Similar assay values to those from lab.
• Absolute precision ~0.1%P (relative error about 20%).
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0 10 20 30 40 50 60 70 80
NS676 -‐ P log traces
P_lab P_hand1 P_hand2 P_BH_in P_BH_out
Blasthole Prototype P-Results
• Major elements Fe and Si, which have well defined XRF
peaks, were best defined by the borehole XRF tool. Repeat
logging suggests lower error limit ~2-3%
• Assay values of about the right order could be derived for
several minor elements: Al, P, perhaps Mn
• While not able to be calibrated for the downhole system
(no high assays), handheld XRF suggests it should be
possible to calibrate for Ca, Ti, S, K
• Mg was insensitive to the XRF measurements
• Similar results were achieved for data compiled in the
other holes
Blasthole Prototype Results Summary
Slimhole Prototype Design and Test
Handycap • X-ray penetration depth in borehole wall is micrometers • Only clean borehole walls will provide useful data • Borehole wall must be clean AND straight without breakouts • Utilization is limited to mining-style or other „clean“ drilling • X-ray window will illuminate only a small area of borehole wall
Benefit • X-rays will analyse all elements Mg – U • Density data will be provided as well • HQ analytics possible • Fast, simple and reliable technique • No international transport issue • No environmental issues
Thank you!
