CAMSIZER Method Development

941 views
829 views

Published on

Published in: Technology
0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
941
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
29
Comments
0
Likes
2
Embeds 0
No embeds

No notes for slide

CAMSIZER Method Development

  1. 1. © 2013 HORIBA, Ltd. All rights reserved.CAMSIZER Method DevelopmentKiwan Parkkiwan.park@horiba.com
  2. 2. © 2013 HORIBA, Ltd. All rights reserved.ContentsOverview of CAMSIZERSampling TechniquesCAMSIZER HardwareSize and Shape Parameter ChoicesSoftware Set-ups
  3. 3. © 2013 HORIBA, Ltd. All rights reserved.Quick Overview of CAMSIZER
  4. 4. © 2013 HORIBA, Ltd. All rights reserved.Two-Camera-SystemBasic-CameraMeasuring PrincipleZoom-Camera
  5. 5. © 2013 HORIBA, Ltd. All rights reserved.Advantages• precise frame taking of particle projections (60frames/second)• large dynamic measuring range: 30µm to 30mmMeasuring Principle
  6. 6. © 2013 HORIBA, Ltd. All rights reserved.CCD - Basic CCD - ZoomMeasuring PrincipleDetection of particlesOne pixel is element ofa projection when atleast half of the pixelis covered.Basic Camera Range:240 um – 30000 umZoom Camera Range:30 um – 1250 umOverlapping Range:625 um – 1250 umCamera Resolution
  7. 7. © 2013 HORIBA, Ltd. All rights reserved.Range of useMeasuring Principle30 µm to 30 mmShape Size
  8. 8. © 2013 HORIBA, Ltd. All rights reserved.CAMSIZER Samples Must Be…1. In dry state2. Free flowing3. Lager than 30 um4. Preferably not cohesive and/orstatically chargeda) Flow-agent can be added (Fumed Silica)b) Ionizer can be usedc) Ultrasonic probe can be used
  9. 9. © 2013 HORIBA, Ltd. All rights reserved.ContentsOverview of CAMSIZERSampling TechniquesCAMSIZER HardwareSize and Shape Parameter ChoicesSoftware Set-ups
  10. 10. © 2013 HORIBA, Ltd. All rights reserved.Good Sampling PracticesLearn to Love Your Riffler!
  11. 11. © 2013 HORIBA, Ltd. All rights reserved.From: NIST Recommended Practice GuideSpecial Publication 960-1Particle Size CharacterizationAjit JillavenkatesaStanley J. DapkunasLin-Sien H. LumMaterials Science andEngineering LaboratoryJanuary 2001Particle Sizing Workflow
  12. 12. © 2013 HORIBA, Ltd. All rights reserved. SMALL PARTICLES POTENTIALLY SMALLEXTRACTION ERRORS (A) POTENTIALLY LARGESAMPLE PREP ERRORS (C) LARGE PARTICLES POTENTIALLY LARGEEXTRACTION ERRORS (B) POTENTIALLY SMALLSAMPLE PREP ERRORS (D) INSTRUMENT ERROR IS SMALL ANDRELATIVELY CONSTANTPARTICLE SIZESAMPLE PREPARATION SAMPLE EXTRACTIONINSTRUMENT ERROR%ERROR ABCDMeasurement Error Sources
  13. 13. © 2013 HORIBA, Ltd. All rights reserved.Sampling from bulkSub-sample for measurementSometimes second sub-samplingPerform measurementSample Extraction Overview
  14. 14. © 2013 HORIBA, Ltd. All rights reserved.PLACE SPATULA INTO POWDEREXTRACT SMALL AMOUNT FORANALYSISACCEPTABLE FOR NARROWDISTRIBUTIONSSEGREGATE LARGE AND SMALL WHENPOLYDISPERSE LARGE PARTICLES PERCOLATE UPWARD SMALL PARTICLES GRAVITATE DOWNWARDEASY METHOD BUT NOT THE BEST METHODMOST USED METHODTechnique: Grab Sampling
  15. 15. © 2013 HORIBA, Ltd. All rights reserved. When a powder is stored in acontainer, it can be mixed by rollingand tumbling the container. Thecontainer should not be more thanhalf to two-thirds full. It is importantto perform this action before“grabbing” a sample with a spatula. Then pull sample with a spatula…..This image cannot currently be displayed.Grab Sampling from Bottle
  16. 16. © 2013 HORIBA, Ltd. All rights reserved.Pile of powder is divided into 4sections.Two diagonal sections arediscarded, and two are retainedand mixed together.Mixture is again divided into 4sections, and two diagonalsections are mixed.Process is repeated untilremaining sample is correctamount for analysis.Can be carried out with verysmall sample amount or very largesamples.Technique: Coning & Quartering
  17. 17. © 2013 HORIBA, Ltd. All rights reserved.Chute splitting allows sample tovibrate down a chute to vaneswhich separate the mass into twoportions. Each portion movesfurther where they each aredivided into two parts, now givingfour parts. This may be continueduntil usually 8 or 16 portions areobtained.Technique: Chute Riffling
  18. 18. © 2013 HORIBA, Ltd. All rights reserved. The best method of representativesplitting of powders is the ROTARYRIFFLER. The complete sample to besplit is directed down a chute into opencontainers. Each container will receive asample which is representative of theoriginal bulk material because thedistribution of material is averaged overtime. The complete amount of theoriginal bulk sample must be consumed. These splitters are commercially available from companies that marketvarious types of sample splitters. See: www.retsch.com www.quantachrome.com www.microscal.comTechnique: Rotary Riffling
  19. 19. © 2013 HORIBA, Ltd. All rights reserved. Standard Deviation () in % Sugar-Sand Mixture SCOOP SAMPLING 6.31 TABLE SAMPLING 2.11 CHUTE RIFFLER 1.10 SPINNING RIFFLER 0.27 Density of sand and sugar respectively 2.65 and 1.64 g/ml Reference: Allen, T. and Khan, A.A. (1934), Chem Eng, 238, CE 108-112Sampling Technique Error Levels
  20. 20. © 2013 HORIBA, Ltd. All rights reserved.ContentsOverview of CAMSIZERSampling TechniquesCAMSIZER HardwareSize and Shape Parameter ChoicesSoftware Set-ups
  21. 21. © 2013 HORIBA, Ltd. All rights reserved.CAMSIZER Hardware
  22. 22. © 2013 HORIBA, Ltd. All rights reserved.Chute Choice 60 mm Chute works well with particles larger than 240 um. 15 mm Chute is recommended if you need to be accurate forparticles below 240 um (Note: Particles below 625 um will bemainly measured by the Zoom camera)
  23. 23. © 2013 HORIBA, Ltd. All rights reserved.Guidance Sheet & Vacuum Each scale is equal to 1 mm (Total 10 scale) Unless sample are larger than 10 mm, use the guidance sheetat all time (the gap needs to be larger than the largest particle)
  24. 24. © 2013 HORIBA, Ltd. All rights reserved.Maintenance-Free byVenturi-FlushingFeaturesAlso use Vacuum tokeep the windowsclean, especially forparticles that arecohesive.The guidance sheetwill help particles fallon the same planefield and preventparticles fromtouching windows
  25. 25. © 2013 HORIBA, Ltd. All rights reserved.ContentsOverview of CAMSIZERSampling TechniquesCAMSIZER HardwareSize and Shape Parameter ChoicesSoftware Set-ups
  26. 26. © 2013 HORIBA, Ltd. All rights reserved.Size and Shape Parameter Choices
  27. 27. © 2013 HORIBA, Ltd. All rights reserved.Choosing Size ParametersHow do I determine which sizeparameter to use??
  28. 28. © 2013 HORIBA, Ltd. All rights reserved.What is thesize of thisparticle?Measurement ResultsParticle Size
  29. 29. © 2013 HORIBA, Ltd. All rights reserved. Martins’s Diameter: The distance between oppositesides of a particle measured on a line bisecting theprojected area. To ensure statistical significance allmeasurements are made in the same directionregardless of particle orientation. Feret’s Diameter: The distance between paralleltangents on opposite sides of the particle profile.Again to insure statistical significance, allmeasurements are made in the same directionregardless of particle orientation. Note: Both Martin’s and Feret’s diameters aregenerally used for particle size analysis by optical andelectron microscopy. Equivalent Circle Diameter: The diameter of a circlehaving an area equal to the projected area of theparticle in random orientation. This diameter is usuallydetermined subjectively and measured by oracularmicrometers called graticules. Equivalent Spherical Diameter: The diameter of asphere that has the same volume as the irregularparticle being examined.Martin’sDiameterFeret’s DiameterEquivalent SphericalDiameterSize Definitions
  30. 30. © 2013 HORIBA, Ltd. All rights reserved. Martin’s andFeret’s Diameter’swill vary asparticles areviewed in differentorientations. Theresult will be aDISTRIBUTIONfrom smallest tolargest.Martin’sDiameterFeret’s DiameterEquivalent SphericalDiameterMartin’sDiameterFeret’s DiameterEquivalent SphericalDiameterParticle Orientation
  31. 31. © 2013 HORIBA, Ltd. All rights reserved.Particle SizeMeasurement Resultsxc minxc min“width”AA‘ = Axarea“diameter overprojection surface”xarea“length”xFemaxxFe max
  32. 32. © 2013 HORIBA, Ltd. All rights reserved.Comparison of Size Definitionsdifferentsize definitionsdifferentresultsxc min  xArea  xFe maxMeasurement Resultsx [mm]0.2 0.4 0.6 101020304050607080Q3 [%]Sample A_Basic_0.2%_xc_min_001.rdfSample A_Basic_0.2%_x_area_001.rdfSample A_Basic_0.2%_xFemax_001.rdf2 x[mm]
  33. 33. © 2013 HORIBA, Ltd. All rights reserved.Size Parameter Selection Questions1. Do I know what size parameter(s) Iwant to measure? (Maybe you alreadyknow)2. Do I need to correlate the CAMSIZERdata to another data? (Sieve data,laser diffraction data, or any otherdata?)3. Which technique do I need tocorrelate?4. Which parameter will be most critical?
  34. 34. © 2013 HORIBA, Ltd. All rights reserved.Size Parameter Selection Based on TechniquesXc_min (width): To correlate sieve data(may need a fitting file on theCAMSIZER)Xarea (equivalent spherical diameter):To correlate laser diffraction dataXfe_max (length): No other techniquescan really measure this. If length needsto be measured, one may have to useimage analyzer.
  35. 35. © 2013 HORIBA, Ltd. All rights reserved.Size Parameter Selection Process1. Measure multiple size parameters (upto 5 different parameters can bemeasured with the CAMSIZER)2. Compare each size parameters andidentify the parameter that give thedesired data
  36. 36. © 2013 HORIBA, Ltd. All rights reserved.x [mm]0.2 0.4 0.6 101020304050607080Q3 [%]Sample A_BZ_0.2%_xc_min_001.rdfSample A_.refDigitale Imaging  Sieving
  37. 37. © 2013 HORIBA, Ltd. All rights reserved.Choosing Shape ParametersHow do I determine which shapeparameter to use??
  38. 38. © 2013 HORIBA, Ltd. All rights reserved.• width/length =aspect ratio• Roundness• Symmetry• ConvexityxFe maxxc minA Pr1r2SA convexA realParticle ShapeMeasurement ResultsmaxFemincxxPA2421rrmin121convexrealAA
  39. 39. © 2013 HORIBA, Ltd. All rights reserved.Shape Parameter Selection Questions1. Do I know what shape parameter(s) Iwant to measure? (Maybe you alreadyknow)2. Do I want to quantify an outlier?3. Do I want to differentiate mixtures?
  40. 40. © 2013 HORIBA, Ltd. All rights reserved.Shape Parameter Selection Process1. Measure multiple shape parameters (all theparameters that are available in theCAMSIZER software can be measured)2. Save images to compare particle to particle3. Compare shape parameters and identify theparameter that give the most discrepancy4. Set a parameter threshold to clearly identifyamount of each group
  41. 41. © 2013 HORIBA, Ltd. All rights reserved.Symm (symmetry) is the parameter with the most discrepancyImage Evaluation
  42. 42. © 2013 HORIBA, Ltd. All rights reserved.ContentsOverview of CAMSIZERSampling TechniquesCAMSIZER HardwareSize and Shape Parameter ChoicesSoftware Set-ups
  43. 43. © 2013 HORIBA, Ltd. All rights reserved.Method Design(Creating a Task File in the CAMSIZERSoftware)
  44. 44. © 2013 HORIBA, Ltd. All rights reserved.Goal for a Method Accuracy: Not really questioned (Images supportits accuracy) Repeatability: The CAMSIZER lets users recycleits samples and repeatability is usually good Reproducibility: A good sampling technique is thekey to achieve reproducibility Resolution: Extremely good since the CAMSIZERwill use every particle to generate data Match historic data (sieves, LD, and etc..):Possible to match any data as long as a correctparameter gets measured
  45. 45. © 2013 HORIBA, Ltd. All rights reserved.New Task File
  46. 46. © 2013 HORIBA, Ltd. All rights reserved.
  47. 47. © 2013 HORIBA, Ltd. All rights reserved.Feeder and Funnel Control
  48. 48. © 2013 HORIBA, Ltd. All rights reserved.Cameras
  49. 49. © 2013 HORIBA, Ltd. All rights reserved.Image Saving
  50. 50. © 2013 HORIBA, Ltd. All rights reserved.Saving Specific Images
  51. 51. © 2013 HORIBA, Ltd. All rights reserved.Saving Different Size and Shape Parameters
  52. 52. © 2013 HORIBA, Ltd. All rights reserved.Saving a Task File
  53. 53. © 2013 HORIBA, Ltd. All rights reserved.Method Development Summary1. Know how to practice a good samplingtechnique: Try to get the representativesample all the time2. Know your hardware: Choose properhardware to get more accurate and precisedata3. Know what you want to measure: Choosesize and shape parameters will achieveyour goal4. Know the software settings: Choose propersettings to get more robust data
  54. 54. © 2013 HORIBA, Ltd. All rights reserved.GraciasDankeБольшое спасибоGrazie‫ر‬ْ‫ك‬ُ‫ش‬‫ا‬Σας ευχαριστούμε감사합니다ObrigadoTacka dig謝謝ขอบคุณครับdhanyawadありがとうございました
  55. 55. © 2013 HORIBA, Ltd. All rights reserved.www.horiba.com/particleReceive news of updatesView application &technical notes (180+),webinars (70+), whitepapers.Talk to us, ask questionslabinfo@horiba.comKiwan ParkApplications Scientistkiwan.park@horiba.com

×