Slide 1: Laser Diffraction Method Development
HORIBA Scientific Particle Characterization Laser Diffraction Method Development Jeffrey Bodycomb, Ph.D. August 24, 2021 © 2021 HORIBA, Ltd. All rights reserved.
Slide 2: Method development overview
Method development overview • Consider Goal(s): Reproducible method that tracks product performance • Choose measurement approach (dry vs. suspension) • Choose Refractive Index (real and imaginary) • Vary measurement settings that can influence result – Dry: measurement duration, concentration, air
Slide 3: Goals
Goals • Reproducible method that tracks product performance • Accuracy: tricky subject, is it the “real” particle size? • Repeatability: liquid suspension re-circulating in sampler • Reproducibility: prepare, measure, empty, repeat • Resolution: optimize to find second populations • Match historic d
Slide 4: Accuracy vs. Precision
Accuracy vs. Precision HIGH ACCURACY HIGH PRECISION LOW ACCURACY HIGH PRECISION HIGH ACCURACY LOW ACCURACY LOW PRECISION LOW PRECISION (A) Low accuracy, low precision measurements form a diffuse, off-center cluster (B) Low accuracy, high precision measurements form a tight off-center cluster (C) Hig
Slide 5: Accuracy
Accuracy Is it the “real particle size”? • Comparison to referee technique • Microscope (image analysis) is referee technique for particle characterization • For submicron particles, also consider DLS and Nanoparticle Tracking analysis, though TEM images are more convincing. • Compare to sieving or
Slide 6: Image Analysis
Image Analysis Take a picture, analyze for size and shape © 2021 HORIBA, Ltd. All rights reserved. 6
Slide 7: Types of Precision
Types of Precision Repeatability Prepare sample, add to wet sampler, re-circulate, measure same multiple times (suspensions only) Provides limited information (mostly a test of analyzer performance) First test of a method Reproducibility Prepare sample, measure, drain, repeat (suspensions + dry) Dis
Slide 8: Reproducibility Samplings of Polystyrene Latex
Reproducibility 24 Samplings of Polystyrene Latex 30 Yes, we emptied instrument 25 20 between each analysis. Frequency % 15 10 5 0 15 4 7 8 9 5 3 5 4 8 9 8 6 4 2 9 5 11 19 25 33 44 58 76 00 31 72 26 97 90 12 71 81 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 1. 2. 2. 3. 5. 6. 8. -5 Size (microns) © 2021 HORIBA, Lt
Slide 9: Reproducibility
Reproducibility Reproducibility: prepare, measure, empty, repeat What would be good reproducibility? Look at accepted standards Measure 3 or more times, calculated COV at d10, d50, d90 COV (RSD) = st dev/mean * 100 ISO13320 Note: double all limits COV < 3% at median d50 when d10, d50, or d90 COV < 5
Slide 10: Repeatability
Repeatability Test repeatability first! If possible, check repeatability by leaving sample in the analyzer and running a multiple times. This lets you separate issues with reproducibility from repeatability. Your reproducibility is unlikely to be better than repeatability. © 2021 HORIBA, Ltd. All ri
Slide 11: Resolution
Resolution • Ability to measure small differences in particle size. • Small differences between successive samples (different production lots) are most important. This is tied to reproducibility. • Detection limit of small amount of material outside of main size distribution. • Best defined by user’
Slide 12: High Resolution?
High Resolution? • Resolve size difference between two materials of similar size • 552nm and 600nm PSL • Measured separately: high resolution • Measure mixed together: peaks would blend and you cannot discern two populations. • Directly connected to reproducibility. If your result is +/-5%, then you
Slide 13: Separate peaks
Separate peaks 83nm, 204 nm, 503nm PSL Resolution of multiple modes in a single sample Next peak at least 2x of previous size © 2021 HORIBA, Ltd. All rights reserved. 13
Slide 14: Separate Peaks
Separate Peaks 100µm, 200µm, 400µm glass beads Next peak 2x of previous size Resolution is independent of where you are on size scale © 2021 HORIBA, Ltd. All rights reserved. 14
Slide 15: Identifying Trace Impurities
Identifying Trace Impurities Ludox-50 Geltech Median Size : 0.031 μm Median Size : 1.65 μm Create Median (Peak 1) : 0.031 μm Median (Peak 2) : 1.75 μm method then refine by adding know amounts of “impurity” 0.05wt% Geltech 1.5 in LUDOX and testing. © 2021 HORIBA, Ltd. All rights reserved. 15
Slide 16: Dry vs wet?
Dry vs wet? Wet: if your sample is already in liquid dispersion smaller particles (<10 microns>) that need liquid to keep agglomerates apart need to use surfactants Dry: sample is already a dry powder larger particles © 2021 HORIBA, Ltd. All rights reserved. 16
Slide 17: Wet Method Development
Wet Method Development © 2019 2021 HORIBA, Ltd. All rights reserved. 17
Slide 18: Workflow
Workflow First determine RI Choose dispersant (water, surfactants, hexane, etc.) Sampler selection: sample volume Pump & stirrer settings Concentration Measurement duration Does the sample need ultrasound? Document size-time plot Disperse sample, but don’t break particles Check for reproducibility ©
Slide 19: Refractive Index
Refractive Index Refractive index affects accuracy, but not repeatability Real component via literature or web search, Becke line, etc. Measure sample, vary imaginary component to see if/how results change, choose value that minimizes R parameter error calculation More in the refractive index webina
Slide 20: LA-960 Method Expert
LA-960 Method Expert © 2021 HORIBA, Ltd. All rights reserved. 20
Slide 21: Dispersant Choice
Dispersant Choice Which liquid? Set up vials or beakers Water is easy to handle and with each candidate inexpensive so it is quite popular. dispersant. Gently put Choose organic solvents a little bit of fine when dispersion in water is particle on the top of difficult or if particles are water each.
Slide 22: Poor wetting vs good wetting
Good wetting: particles separate Poor wetting: particles don’t and stream through surface break surface tension © 2021 HORIBA, Ltd. All rights reserved. 22
Slide 23: Sample Handling
Sample Handling Agitation and sampling (sample volume) affect repeatability. Larger, broad distributions require larger sample volume LA-960 Dispersing Volume (mL) Sample Handlers Lower volume samplers for precious materials or solvents Aqua/SolvoFlow 180 - 330 MiniFlow 35 - 50 Fraction Cell 15 Smal
Slide 24: Pump and Stir Start
Pump and Stir Start • Must be high enough to Measure 2 or 3 times in a suspend & circulate heavy row particles-if too low speed, larger particles won’t be measured and apparent size decreases with time. Size going Increase stir / • Add sample to 85-90% down? pump speed transmission. • Do this to get
Slide 25: Concentration
Concentration Laser diffraction works best in limit of zero concentration High enough for good S/N ratio Low enough to avoid multiple scattering Typically 95 – 80 %T Rough out pump/stir Add sample to ~95% T (low concentration) Measure size Add another drop to go to ~90% T Measure Add another drop t
Slide 26: Duration
Duration • Long enough for reproducibility • Typically 5 sec (almost always…) • Longer time for large, broad distributions • Can be automated in software • Could be used for robustness testing during method validation Check repeatability. If size is varying at random, try increasing duration or conc
Slide 27: Ultrasonic Dispersion
Ultrasonic Dispersion Adding energy to break up agglomerates – disperse to primary particles, without breaking particles Similar to changing air pressure on dry powder feeder Typically set to 100% energy, vary time (sec) on Investigate tails of distribution High end to see if agglomerates removed Sm
Slide 28: Ultrasound
Ultrasound On board ultrasound can break up loose agglomerates. Get sample at middle of T% range and rough pump and stir. Measure Ultrasound 10 sec measure 10 more seconds measure Etc. Review results. No significant change means way too little or you don’t need it. If changing, go to where size s
Slide 29: Plot summary data
Plot summary data 180 160 140 D90 120 D50 D10 Size (mm) 100 80 60 40 20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Ultrasound (sec) © 2021 HORIBA, Ltd. All rights reserved. 29
Slide 30: Reproducibility
Reproducibility Need the same answer after drain and refill. Driven by • Concentration repeatability/sensitivity T% • Hard to hit the same T%, predilute before adding • Sampling • Ensure the suspension is well mixed to entrain all particles, even large ones that settle. • Suspension stability (e.g.
Slide 31: Systematic reporting
Systematic reporting • Exp # Agitation Circulation Dmean (nm) D10 (nm) D90 (nm) Systematically measure at several settings & 1 1 1 187.03 137.5 245.7 select optimum (repeatability) 2 1 3 184.23 135.9 242.1 3 3 1 187.28 137.8 245.8 • Pump +/-1 4 3 3 184.61 136.1 242.5 • Stir +/-1 5 6 1 1 1 3 185.32 1
Slide 32: Reproducibility methods
Reproducibility 58 methods Not acceptable per USP<429> Laser diffraction for PSD If RSD for d50 < 20%, then acceptable for QC environment Acceptable Note: RSD increases with decreasing size *Barber, Keuter, and Kravig, A Logical Stepwise Approach to Laser Diffraction Particle Size Distribution Analy
Slide 33: Sampler Selection
Sampler Selection Remove points from not acceptable region using Fraction Cell Not acceptable per USP<429> Not acceptable Acceptable Acceptable Systematic analysis gives good hint about fraction cell use and importance of pumping for larger particles. © 2021 HORIBA, Ltd. All rights reserved. 33
Slide 34: Automation
Automation LA-960 Method Expert will systematically vary measurement parameters: Refractive Index Pump Ultrasound Generate result graphs © 2021 HORIBA, Ltd. All rights reserved. 34
Slide 35: Dry Method Development
Dry Method Development © 2019 2021 HORIBA, Ltd. All rights reserved. 35
Slide 36: Workflow
Workflow First get sampling right & determine RI Measure at 3 different pressures (low, medium, high) Determine optimum pressure based on good dispersion while not breaking particles Can also compare dry vs. wet measurements Adjust other settings to optimize sample concentration (T%) duration (amoun
Slide 37: Dispersion vs Breakage
Dispersion vs Breakage Theoretical Actual Stability Size Size Increasing energy Increasing energy Higher air pressure or longer ultrasound duration © 2021 HORIBA, Ltd. All rights reserved. 37
Slide 38: Dispersion vs Breakage
Dispersion vs Breakage • Dispersion and milling can be parallel rather than sequential processes Theoretical Actual © 2021 HORIBA, Ltd. All rights reserved. 38
Slide 39: Pressure Titration
Pressure Titration © 2021 HORIBA, Ltd. All rights reserved. 39
Slide 40: Mg Stearate
Mg Stearate High = 3 bar Mid = 2 bar Low = 1 bar © 2021 HORIBA, Ltd. All rights reserved. 40
Slide 41: Mg Stearate
Mg Stearate 18 16 14 12 10 D90 D50 8 D10 6 4 2 0 1 2 3 D90 15.258 14.394 12.822 D50 8.626 8.149 7.502 D10 4.862 4.564 4.234 © 2021 HORIBA, Ltd. All rights reserved. 41
Slide 42: Reproducibility at 3 Bar
Reproducibility at 3 Bar © 2021 HORIBA, Ltd. All rights reserved. 42
Slide 43: Reproducibility at 2 Bar
Reproducibility at 2 Bar © 2021 HORIBA, Ltd. All rights reserved. 43
Slide 44: Reproducibility at 1 Bar
Reproducibility at 1 Bar © 2021 HORIBA, Ltd. All rights reserved. 44
Slide 45: Summarize Reproducibility
Summarize Reproducibility Pressure D10 D50 COV D90 COV COV 1 Bar 0.659 0.612 1.007 2 Bar 0.242 0.288 0.896 3 Bar 0.956 1.456 2.209 This is reproducibility; Sampling is important! Must have representative sample © 2021 HORIBA, Ltd. All rights reserved. 45
Slide 46: T%, quantity
T%, quantity T% is done much like you did with wet. Vary T% setting and monitor results Sample Quantity. Vary quantity and monitor results. © 2021 HORIBA, Ltd. All rights reserved. 46
Slide 47: Summary
Summary Design for maximum precision. Follow guidelines in standards. Liquid Suspensions: wet, disperse Investigate system settings: concentration, agitation, ultrasound. Dry Powders: select air pressure Must have representative sample. Systematically evaluate the method. Prepare a report in case th
Slide 48: © 2019 HORIBA, Ltd. All rights reserved.
© 2019 HORIBA, Ltd. All rights reserved. 48
Slide 49: Thank you
Thank you Cảm ơn 감사합니다 ありがとうございました Dziękuję Grazie धन्यवाद நன்ற Merci 谢谢 ขอบคุณคร ับ Obrigado Σας ευχαριστούμε Tack ska ni ha Большое спасибо Danke Gracias © 2019 HORIBA, Ltd. All rights reserved. 49