Ion Beam Analyis on the Upgraded CSIRO Nuclear Microprobe

Transcription

1 Ion Beam Analyis on the Upgraded CSIRO Nuclear Microprobe Jamie S Laird, Chris G Ryan, Robin Kirkham, Peter Siddons, David Parry, Takahiro Satoh, Stephen Gregory and Roland Szymanski MINERAL RESOURCE FLAGSHIP SNEAP 50, Sydney 2016

2 Outline of Talk Summary of CSIRO Nuclear Microprobe Main Ion Beam Applications in Geosciences Moving to Higher count rates for larger areas New chamber design Accelerator control upgrades needed for smoother operation 2 Presentation title Presenter name

3 CSIRO Nuclear Microprobe (NMP) DAQ VAC NEC 5U Vertical Pelletron Accelerator RF ion source (H +, He +, ++ ) Premier beamline at the University of Melbourne, 90% beam usage last calendar year. Roughly 50+ users per annum all using PIXE and/or PIGE. 3 SNEAP 50, Sydney 2016

4 Original CSIRO Nuclear Microprobe (NMP) ~1999 Unique separated quintuplet magnetic quadrupole lens system designed to accept high beam currents necessary for some IBA methods but with an acceptable spot size of 1-2mm. Designed by Chris Ryan. Primarily 3MeV H + beams are used with target current ranging from 50pA thru to 5nA. 4 SNEAP 50, Sydney 2016

5 NMP Labview DAQ System: NI RIO, Vision etc. PXI FPGA & CRIO 5 SNEAP 50 SYDNEY 2016

6 NMP Optical Microscopy Driven Scanning Low mag front view PubSub Variables Scan ROI (Shape driven) Scan Modes, dwell, interlacing, pixel dose, etc. 6 SNEAP Sydney 2016

7 Main course: PIXE Elemental Mapping of Minerals Non-invasive, ppm level mapping Proton induced excitation followed by return to ground state via x-ray emission (fluorescence). x-ray energies are elemental signatures PIXE typically uses 3MeV protons focussed to 1-2 mm scanned across the sample with spectra recorded at each position. 7 SNEAP 50, Sydney 2016

8 Key Benefits of PIXE for Geosciences Rapid High Sensitivity Elemental Mapping of Minerals covering a higher fraction of the periodic table than does SXRF Z>13. An order of magnitude faster than before for trace elements. Access to elements not easy on the AS XFM. Complements XFM at the Australian Synchrotron Simultaneous Proton Induced Gamma-Ray Emission (PIGE) to lower Z elements (e.g. Li, Be, B, Fl, Na and Al) previously deemed unsafe due to high neutron radiation levels 8 SNEAP 50 SYDNEY 2016

9 NMP Elemental Coverage using PIXE/PIGE PIXE - Proton Induced X-ray Emission: Z > 13 up to 35 elements detectable SXRF F F Na Mg Mg Al Mg Al Si Mg Na PIGE - Proton Induced g-ray Emission: K Si Mg Mg B Th Conventional detector dead-time limitation means filter used to cut majors Need multiple detectors to cover the full range PIGE detectors mostly light elements missed by PIXE. 9 SNEAP 50 SYDNEY 2016

10 Geological samples are Complex! Geological / environmental samples: Highly specialised Need to unfold software complex to unravel overlaps quantitative results e.g. Au L a overlaps Fe Fe Au As As Rb 1. W L b1,2 2. Yb L g1 3. Zn K b 4. Ge K a 5. As tails Yb Ni Cu Ga Yb Au Cu W Ni W Pb Yb Ge Pb Se Au W Pb Se Au Need full fit to all spectral components Yb Au 10 SNEAP 50 SYDNEY 2016

11 DORIRI APATITE: STANDARD ANALYSIS These types of data sets takes around 4 hours to collect for ppm sensitivity. Expensive and limited to surveying a chosen heterogeneity across a core. Easy to miss key features or traces used for geochemical exploration vectors. 11 SNEAP 50 SYDNEY 2016

12 1.5 mm WATO: GOLD TRACES AWAY FROM FLUID PATHWAYS: MEDIUM SIZE Fe, Rb, Sr 5 mm Fe, Au, Sr Au inclusions These types of data sets takes around 12 hours to collect for ppm sensitivity. Expensive and limited to surveying a chosen heterogeneity across a core. Easy to miss key features or traces used for geochemical exploration vectors. 12 SNEAP 50 SYDNEY 2016

13 2mm OKINAWAN HYDROTHERMAL SMOKERS: ENORMOUS 48 hour ANALYSIS 15 mm Fe,Ba,Zn Obvious Need for Rapid Large Area Mapping Fe,Ag,Ba Precious metals inclusions can be sparse requiring large sampling areas. Each image is ~48 hours!! Not practical for high turn around work. Success depends on machine performance which can be sporadic depending on accelerator ion source conditioning. Need to collect higher count rates by using larger solid angles and multiple detectors. 13 SNEAP 50 SYDNEY 2016

14 The Road to Higher Count Rates Group Single Elements Planar semiconductivity lithography leads to large array numbers x4 OR SDD array Best E-Res Complex design x6 Si detector array Large channels reduce dead-time issues. Worse E-Res Simpler design Highly customized to geometry: not great gains Individual detectors still suffer dead-time issues physical crosstalk effects become more serious as individual detectors reduce in size. Energy resolution issues 14 SNEAP 50 SYDNEY 2016

15 BNL/CSIRO Maia X-ray Detector Developed for the AS XFM 20x20 array of detectors X10 Increase in trace/minor elements Need new target chamber with internal stages for large area mapping since detector response depends on beam location via solid angle i.e. no electrostatic scanning for pixel detectors 15 SNEAP 50 SYDNEY 2016

16 SXRF Maia on X-ray Focused Microbeam at AS PIXE Maia will result in much improved image quality 16 SNEAP 50 SYDNEY 2016

17 Engineering Was Very Challenging: Constraints on all sides Inside Target Chamber Cu cooling / support Peltier coolers block Al N detector support frame Mo incident beam tube 0.8 mm ID Mo charge-sharing suppression mask 384-element Si detector in 20 x 20 mm array 17 SNEAP 50 SYDNEY 2016

18 Water cooling, vacuum Fibreoptic cable NMP End Station Maia 384 detector head Microscope port MAIA Port Fibre probe port (spectroscopy, reflected light illumination) Motor drive/ encoder ports 18 SNEAP 50 SYDNEY 2016

19 Internal Stage for High Speed and Reproducibility Maia Lifter (DUAL MODES) Sample Stage (stubs + slides) Z X Internal FEMTO I2V amplifier measure beam current. V2f converts to an RS422 signal (twist pair diff) and exits chamber as digital pulses which are counted by the DAQ. Y Result is ~pa resolution for beam current even when motors are driving. 19 SNEAP 50 SYDNEY 2016

20 XY Stage Driven Scan Tests Are Very Positive 20 SNEAP 50, SYDNEY 2016

21 PIXE Maia Brookhaven National Labs PIXE Maia has a unique design to cater for the stringent chamber design. Complex Cu brazing for UHV operation Detector array in later generation is better quality than its predecessor Resolution ~ 240eV 21 SNEAP 50 SYDNEY 2016

22 Coupled Beamline and Accelerator Control for Reliable Operation NMP Beam Transport feedback Cu TEM grid (1000) Accelerator Control, BPM Internal telemetry Resolution ~1.5mm Understanding TPS stability (close to understanding) so no oscillation or drift leading to non-linear dead-time correction models (complex) and image artefacts Installation of bottom electrostatic lens to couple column to slits to fill out our divergence apertures on the NMP reducing spot drift. Taken out in the 90 s. 22 SNEAP 50 SYDNEY 2016