Instrumental Validation of a Scanning Electron Microscope with Energy Dispersive X-ray - PowerPoint Presentation

1. Instrumental Validation of a Scanning Electron Microscope with Energy Dispersive X-ray Spectroscopy (SEM/EDS)BOSTON Police Department Crime LaboratoryAmy Reynolds, Erin Corcoran, Emily runt1

2. SEM/EDS AcquisitionSept. of 2014 our previous Amray SEM with PGT EDS stopped workingAfter much research and soliciting for fundsJEOL IT300LV SEM with an Oxford X-MaxN 50 EDSPurchased in 2016AZtec softwareTrace Evidence analysisPaintTapeGlassGeneral UnknownsGunshot primer residue analysis2

3. SEM/EDS ValidationANAB 17025 Accreditation5.4.5, Supplemental and our own labs validation requirementsSpoke to several Forensic Scientists in the field about their validationsJohn Druggan, MA State Police Crime LaboratoryMatney Wyatt, US Army Criminal Investigation LaboratorySent their validation plan for validating an SEM/EDS for GSRDavid Edwards, JEOLRichard McLaughlin, Oxford InstrumentsStarted out with validation of the instrument for trace evidence samples other than Gunshot primer residues3

4. Magnification and Measurement ToolsCopper Grid (200 Mesh, Pelco XCS-6 Standard)3 locations on gridOne edge of the mesh to the adjacent edge 2 magnifications (350x and 550x)JEOL and Oxford image Average measurements from the 3 locations at each magnification, compared to expected 127µm grid spacingJEOL- 350x was 126.8µmJEOL- 550x was 127.6µmOxford- 350x was 126.9µmOxford- 550x was 126.6µmUnder 1µm difference observed was acceptable4

5. Verify Factory Resolution –127kV for MnManganese (Pelco XCS-6 Standard)FWHMProcess Time 4 and 6, 20kVDetermine height of K peak, left side of peak at half height subtracted from right side of peak at half heightNumber of channels counted at half width, multiplied by the amount of eV’s per channelAverage of 3 spectra for each method determinedProcess time 4: approx. 133eVProcess time 6: approx. 127eVProtocol-Process time 6 is used, any eV under 127eV is acceptable for QC, or +/- 10eV of 127eV 5HeightFWHM

6. Verify X-ray RangeFull Scale VerificationMolybdenum (Mo) aperture3 spectra at 30kVMo peaks observed at approx. 2.3keV (2.266keV) & 17.5keV (17.446keV)Low eV Limit VerificationBoron (Pelco XCS-6 Standard)3 spectra at 20kVB peak observed at 0.179keV (0.179keV)Beam Energy RangeManganese (Pelco XCS-6 standard)1 spectra each at 20kV & 25kVBremsstrahlung Curve observed to drop off at approx. 20kV (20kV run) and 25kV (25kV run)6

7. Verify Consistency of Probe CurrentManufacturer-Probe current over the course of a run <1% difference per hourCurrent meter was purchasedrbd 9103 Autoranging PicoammeterOld aperture placed over the top of a hole in stub holder (Faraday cup) Focused upon the aperture hole and magnified until the aperture hole completely covered field of viewFilament stabilized for ½ hourProbe current data collected for 9.69 hoursDifference in probe current data over 9 hours was approx. 2.6%<1% per hour, 0.28% per hour7

8. Old vs. NewDo we get the same or better results with the new SEM/EDS vs. the old SEM/EDS?Used old proficiencies to answer this questionTape- 2010 Items 1 and 2 (electrical), 2012 Items 1-3 (duct)Paint- 2012 Items 1-3 (automotive), 2013 Items 1 and 2 (architectural)Glass- 2011 Items 1 and 2, 2012 Items 1 and 4Determine the best parametersUsed different Pa in Low Vacuum (LV) mode (30, 50 and 70)Observed charging at low PaObserved contamination from other paint layers or the environment at high Pa due to beam spreadUsed different kV (10, 15, and 20)Unable to obtain all of the elements found in previous spectra with lower kVObserved contamination from the other paint layers or the environment with higher kVDetermined that the samples needed to be sputter coated with carbon8

9. Carbon Sputter CoatingDenton Vacuum Desk II Sputter/Etch Unit and Carbon Evaporation accessoryResurrected and sent off for maintenanceDetermine the best height of the carbon chuck for proper carbon coating Heights- (5.5, 7.5 and 9.5 centimeters above the samples)Higher the carbon chuck the lighter the coatingLower heights could burn the sample7.5 cm height chosenSlight charging still observed on the electrical tape and backing of duct tapeCarbon coat twice or use low Pa in LV mode 9

10. Proficiency ComparisonsGlass ProficiencyCTS 11-548 and CTS 12-548 old system vs. new systemAll the same major peaks observedPaint ProficiencyCTS 12-546 old system vs. new systemSame major peaks observed plus more peaks with new systemBetter Quality DetectorCTS 13-546 old system vs. new systemSame major peaks observedTape ProficiencyFTS 10-Tape and FTS 12-Tape old system vs. new systemSame expected peaks observed plus more peaks with the old systemContamination10

11. Protocol for SEM/EDSJEOL SEM with Oxford EDS was comparable with the reliability, reproducibility and sensitivity to the old Amray SEM with PGT EDSProtocol adjustmentsDaily QCMn is run at 10mm, 20kV, Aperture 3, Process Time 6FWHM at 5.9eV should be below 127eV, but at least within +/-10eV from 127eVCharging samplesMay use LV mode below 30Pa, starting at 15kVHighly recommended to carbon coat the samplesIf after carbon coating the sample charging is still observed:Increase the working distance (possible loss of elemental information)Decrease the accelerating potential (possible loss of elemental information)Recoat the sample with another layer of carbon (best option)Parameters to start:10mm, 20kV, Aperture 3, Probe current at a level to observe 40-50% deadtime, Secondary detector for good image quality, Backscattered detector-Compositional for good elemental analysis11

12. Gunshot Primer Residue Validation AZtec softwareSeems like all other labs with Oxford are using INCA softwareENFSI/PLANO standard Determine the best parameters to be used 90% of the >1µm particles10mm, 20kV, Process time 4, Aperture 2, these parameters stayed the sameGray level threshold changes first8000, 8800, 9000, 10000 at 3µs first passFirst pass µs changed second5, 4, 3, 2, 1 µs at 8800 gray level thresholdMagnification changed last190x, 250x 12

13. Gunshot Primer Residue Validation Each different parameter run was done in at least triplicateData from the runs were exported to ExcelX and Y coordinates were plotted to create an overlay on the PLANO standard plotDetermine the % of the particles >1µm After countless runs with the different parameters using the PLANO standardSeveral discussions and visits with both Dave Edwards (JEOL) and Richard McLaughlin (Oxford Instruments)We got the best parameters for a Gunshot primer residue run nailed down10mm, 20kV, Aperture 2, Process time 4Image Scan size: 2048, Dwell time: 2µs, Gray level threshold: 8800Second pass imaging: 20µs, Acquisition mode: Live Time of 0.5s, Second Pass EDS filter: Lead containing features with an additional time of 1.5s250x magnification13

14. Negative Control Study5 air blank samples using the optimum parameters5 lab spaces, SEM stubs were left open for 3 working daysSerology lab, pre-SEM room, SEM room, Trace lab, Evidence Receiving areaResultsZero characteristic GSR particles found on any of the stubsSome consistent GSR particles found Found on the actual aluminum stub surface not the adhesive tabNegative controlAir blank from the SEM room or pre-SEM room (sample prep area)14

15. Known GSR stubStub collected directly after shooting occurredTest the optimum parametersStopped after 100 characteristic particles detected Test the reacquisition portion of the softwareChange to GSR Reacquisition Program 15 features ranging from 20µm to 1µm were reacquired-AutomaticallyConfirmed 3 features >9µm in sizeDetermined that if feature is <8-9µm in size it will not relocate properly to those featuresPoint and ID as an end aroundUnacceptableVersion 3.4 now has a Manual Reacquisition15

16. Known GSR stubVersion 3.4 was installed in Nov 2017Manual selection in the GSR Reacquisition programRelocate to a featureMagnify and focus on this featureEnter this magnification into the programStart reacquisitionNeed to find and refocus on the particle at the previously determined magnificationReacquires a better image and spectrum10mm WD, 20kV, Aperture 2, Process Time 4Image Scan size: 512, Dwell time: 5µs, Gray level threshold 5000Acquisition mode: Live Time of 10sConfirm the reacquired featuresCreate a report16

17. Weapon and Time of Collection StudyCurrently running analysis on a weapon and time of collection study5 different weapons, 5 different shootersGlock 9mm pistol with PMC bronze FMJ 9mm LugerSmith & Wesson 38 revolver with PMC bronze FMJ 38 specialHi Point 9mm pistol with Federal American Eagle FMJ 9mm LugerWalther 380 auto pistol with Winchester FMJ 380 AutoRuger 357 Magnum revolver with Blazer Total Metal Jacket 38 Special + P4 different times since collection0 hrs, 2 hrs, 4 hrs and 6 hrs after shooting17

18. Preliminary Results-Weapon and Time of Collection StudyShooter #1- Glock 9mm pistol 18TimeHand# of Characteristic Features# of Confirmed FeaturesApprox. size of Confirmed Features0Right603528µm-3.5µm0Left499517µm-3µm2hrsRight22 (Non-spherical)14µm-5µm2hrsLeft82 4µm-3.5µm4hrsRight0--------4hrsLeft21 2µm6hrsRight21 5µm6hrsLeft0--------

19. Preliminary Results-Weapon and Time of Collection StudyShooter #2- Smith & Wesson 38 Revolver 19TimeHand# of Characteristic Features# of Confirmed FeaturesApprox. size of Confirmed Features0Right1050520µm-1.5µm0Left291312µm-4µm2hrsRight4333µm-1.3µm2hrsLeft83 (Non-spherical)5µm-2µm4hrsRight11 (Non-spherical)5.5µm4hrsLeft0--------6hrsRight0--------6hrsLeft111.3µm

20. Contamination StudyPrep area with the Denton Vacuum was used as our GSR Distance Determination area previouslyAdjacent to the SEM roomSince been moved to another location in the labStubs were collected from the lab bench, fume hood and examination tableBefore cleaningAfter cleaning Determine if any contamination is observed from these locations for future caseworkClean again?Retest areas with further stubs?20

21. Training ProgramTraining CoursesJEOL JSM-IT300LV operator’s training courseOxford AZtec operator’s training course with AZtec GSRGunshot Residue Identification-Wayne Niemeyer, Hooke CollegeReadingsASTM Standard Guide for Gunshot Residue Analysis by Scanning Electron Microscopy/Energy Dispersive X-Ray Spectrometry (newest version)SWGGSR Guide for Primer Gunshot Residue Analysis by Scanning Electron Microscopy/Energy Dispersive X-Ray SpectrometrySummary of the FBI Laboratory’s Gunshot Residue Symposium, 2005Related journal articlesWorkshopsStudies using brake pad stubs and fireworks stubsRun 2-3 of the Time of Collection setsOld proficiency test(s)CompetencyPractical- Old proficiency or a Mock case created for our analystsTheoretical- Competency questions and Mock trial21

22. Thank youMatney WyattUS Army Criminal Investigation LaboratoryJohn DrugganMassachusetts State Police Crime LaboratoryRusty WhiteTexas DPS, Austin Crime LaboratoryMike MartinezBexar County Crime LaboratoryDave EdwardsJEOLRichard McLaughlinOxford InstrumentsEveryone else in the SEM/EDS and GSR community22

23. 23Questions?