Pitfalls in Depth Profiling II Joint ICTP/IAEA Workshop on Advanced Simulation a

Pitfalls in Depth Profiling II Joint ICTP/IAEA Workshop on Advanced Simulation a www.phwiki.com

Pitfalls in Depth Profiling II Joint ICTP/IAEA Workshop on Advanced Simulation a

Harris, Michael, Founder,President, Principal Analyst has reference to this Academic Journal, PHwiki organized this Journal Pitfalls in Depth Profiling II Joint ICTP/IAEA Workshop on Advanced Simulation in addition to Modelling as long as Ion Beam Analysis 23 – 27 February 2009, Miramare – Trieste, Italy Chris Jeynes University of Surrey Ion Beam Centre Guild as long as d, Engl in addition to Thursday February 26th 2009 Pitfalls In Ion Beam Analysis Chapter in the as long as thcoming updated IBA H in addition to book Chris Jeynes1, Nuno Barradas2 1 University of Surrey Ion Beam Centre, Guild as long as d, Engl in addition to 2Instituto Tecnológico e Nuclear, Sacavém, Lisbon J. A. Davies, W. N. Lennard in addition to I.V. Mitchell; responsible as long as the Pitfalls chapter in the 1994 IBA H in addition to book Context Elemental depth profiling : what is our competitive edge compared with other techniques (e.g. SIMS) IBA SIMS Quantitative Small footprint Good at interfaces High sensitivity IBA spectra can be treated with a numerical confidence of 0.2% Best available absolute experimental accuracy is 0.6% We should be aiming as long as absolute accuracy of 1% or better Barradas et al: NIMB262 (2007) 281-303, summary at NIMB266 (2008) 1338-1342

Anderson University US www.phwiki.com

This Particular University is Related to this Particular Journal

2009 H in addition to book Pitfalls Chapter Contents Note: here we will consider only spectral h in addition to ling (not data collection) pitfalls Lost Beam in addition to Events Fixed Parameter Calibration (gain, geometry, resolution) Algorithmic Issues (Matej) Accurate IBA (uncertainty estimation) Unwanted Beam-Target Interactions Other Effects Summary We want to interpret the spectra correctly How do we evaluate the uncertainties Which parameters do the spectra tell us Which parameters must we know precisely to interpret the spectra Chapter contents Lost Beam in addition to Events Fixed Parameter Calibration Energy Solid Angle Electronic Gain Calibration Scattering Angle Detector Resolution Algorithmic Issues Accurate IBA Unwanted Beam-Target Interactions Other Effects Summary Notes: Charge.solid-angle product is always uncertain – this is considered in “Ambiguity” talk Electronic Gain Why is gain D important AA = Q NA s´A (E, q) W not as long as number of counts! Y0,A = Q fA s´A W D / [ e0 ] AAB but to get Q W Jeynes C., Peng N., Barradas N.P., Gwilliam R.M. (2006), Quality assurance in an implantation laboratory by high accuracy RBS, Nucl. Instrum. Methods B249: 482-485

Electronic Gain How precisely can it be obtained Seah et al, 1988: <0.5% An intercomparison of absolute measurements of the oxygen in addition to tantalum thickness of tantalum pentoxide reference materials, BCR 261, by 6 laboratories, Nucl. Instr. Methods B30, 140-151 Munnik et al, 1996: 0.16% F. Munnik, A.J.M. Plompen, J. Räisänen, U. Wätjen, Stopping powers of 200-3000 keV 4He in addition to 550-1750 keV 1H ions in Vinyl, Nucl. Instr. Methods B119, 445-151 Bianconi et al, 2000: 0.2% reported in Barradas et al: Nucl. Instr. Methods B262 (2007) 281-303 Gurbich & Jeynes, 2007: <0.1% (PHD corrected) Evaluation of non-Ruther as long as d proton elastic scattering cross section as long as magnesium, Nucl. Instr. Methods B265 447-452 Pulse Height Defect Detector non linearity (gain varies across energy range). Arises from: Energy loss in detector dead layer / entrance window Nuclear (non-ionising) energy loss (NIL) of projectile in the detector Radiation damage in the detector (recombination sites) NIL can be accounted as long as using Lennards’ correction (W.N. Lennard, S. Y. Tong, G. R. Massoumi, L. Wong, Nucl. Instr. & Methods B45 (1990) 281-284) Pulse Height Defect (Left) Spectra of bulk magnesium with 68.1015 C/cm2 in addition to 800.1015 MgO/cm2 on the surface (Above) SigmaCalc scattering cross-sections as long as natural Mg (the isotopes behave differently) at two different scattering angles with sharp resonance at 1483keV (FWHM 400eV) Gain uncertainty is <0.1% if PHD is accounted as long as Gain changes by 5% over dataset without PHD Important if spectra from different ions or energies are to be compared Gurbich & Jeynes, NIM B265 447-452 Bulk Mg target Electronic Gain Do we care about the offset No, but we need to determine both o & D (correlated variables) as long as Gurbich & Jeynes 2007, o was determined at <1keV as long as both detectors independently Isn’t this an experimental problem No, it’s a data analysis problem E = DC + o energy E of channel C depends on gain D in addition to offset o I have never seen non-linearity, if PHD correction is used Electronic Gain Peaks can be fitted at very high precision Edges can be fitted at high precision PROVIDED the detector resolution is known Multi-elemental sample is GOOD (doesn’t matter if it is dirty) NB: gain(beam energy) Correction as long as energy loss in metal layers is easy C. Jeynes, N. P. Barradas, M. J. Blewett, R. P. Webb, Nucl. Instr. in addition to Meth. B 136-138 (1998) 1229 Au/Ni/SiO2/Si C. Jeynes, A.C.Kimber, High accuracy data from RBS, J.Phys.D, 18 (1985) L93-L97 Detector Resolution 4% resolution change 1.5% thickness change 4% resolution change 0.25% gain uncertainty (need to control contributions to uncertainty at ¼%) 25 keV 4457 eV/ch 28.55 TFU 24 keV 4472 eV/ch 28.07 TFU 26 keV 4460 eV/ch 28.96 TFU Au/Ni/SiO2/Si Electronic Gain Is it right Why are the two signals different heights Check! As signal of As implanted Si (previous example) two detectors Chapter contents Lost Beam in addition to Events Fixed Parameter Calibration Algorithmic Issues Accurate IBA Uncertainty Estimation Spectral Ambiguity Model-free Analysis in addition to Occam’s Razor Common Pitfalls in Data Analysis Unwanted Beam-Target Interactions Other Effects Summary Uncertainty “Error” implies that you (or someone) has made a mistake “Uncertainty” expresses our ignorance of reality Only God knows as long as sure! “Type A” uncertainty: obtained as a variance from ensemble of data “Type B” uncertainty: from a qualitative discussion "GUM:1995" BIPM / IEC / IFCC / ISO / IUPAC / IUPAP / OIML “Guide to the expression of uncertainty in measurement ”. Identical to EN 13005:1999. See also the valuable websites www.gum.dk, www.npl.co.uk/scientific-software/research/uncertainties. Sjöl in addition to KA, Munnik F, Wätjen U, Uncertainty budget as long as ion beam analysis, Nucl. Instrum. Methods B161 (2000) 275-280 Uncertainty Type A & Type B Think of measuring a length with a ruler Type A: multiple measurements (perhaps by different people) Type B: consideration of the spacing of the marks on the ruler in addition to how well they could possibly be read (precision), checking whether the ruler is marked true (!), etc Type A is good, but never sufficient! We should remember always the famous case of the Astronomer Royal, Nicholas Maskeleyne, [in post 1765-1811] who dismissed his assistant Kinnebrook as long as persistently recording the passage of stars more than half a second later than he, his superior. Maskeleyne did not realize that an equally watchful observer may register systematically different times by the method employed by him; it was only Bessel's realization of this possibility which 20 years later resolved the discrepancy in addition to belatedly justified Kinnebrook. Michael Polanyi: Personal Knowledge, 1958 C.Jeynes, N.Peng, N.P.Barradas, R.M.Gwilliam (2006), Quality assurance in an implantation laboratory by high accuracy RBS, NIM B249: 482-485 Revised in the light of the IAEA Software Intercomparison Final Observation The first attempt by the leading labs at a Round Robin as long as RBS showed charge measurement errors of 20%! Davies, Lennard in addition to Mitchell, Chapter 12, Black H in addition to book (1995) It is very easy to make mistakes We should beware! Acknowledgements as long as 2009 IBA H in addition to book Pitfalls Chapter J. A. Davies, W. N. Lennard in addition to I.V. Mitchell; responsible as long as the Pitfalls chapter in the 1994 IBA H in addition to book (the “Black Bible”), some of which is copied into this chapter Max Döbeli (Zurich) as long as the electronic noise in addition to background loops section Alex in addition to er Gurbich as long as detailed comments Harris, Michael Kinetic Strategies Founder,President, Principal Analyst www.phwiki.com

Harris, Michael Founder,President, Principal Analyst

Harris, Michael is from United States and they belong to Kinetic Strategies and they are from  Phoenix, United States got related to this Particular Journal. and Harris, Michael deal with the subjects like Broadband Technologies; Cable/Satellite Industry

Journal Ratings by Anderson University

This Particular Journal got reviewed and rated by Anderson University and short form of this particular Institution is US and gave this Journal an Excellent Rating.