By Dr. Graham R. Rideal, Managing Director, Whitehouse Scientific Ltd., Whitehouse Scientific
SubscribeCo-authors: Dr J Mitchell, Prof. J.A. Dodds, Dr M-N Pons, Prof. K Leschonski, Dr P.J. Lloyd and Mr H.G. Mercus Introduction - The European Community Bureau of Reference (BCR) brought out a range of crushed quartz reference standards in 1980 for particle sizing instrument calibration1.
These had maximum to minimum size ratios of approximately 10 to 1 and covered the size range 0.3 to 650 microns. With the exception of the largest size, which was measured by sieve analysis, all the other references were certified using the Andreasen Pipette method1.
The intervening years have seen a proliferation in Laser based particle sizing techniques, however, when compared to the more traditional methods, Laser analyses of the BCR standards have shown excessively large deviations of up to +/- 70%2,3. Such variations were attributed to a combination of factors including:
(a) Calibration problems from complex scattering of the angular quartz particles
(b) Poor sampling of the 10g bottles
(c) Dispersion problems (both surfactant and ultrasonic bath choice)
(d) Unspecified statistical modelling
(e) Non-prescriptive analytical procedure
To overcome these problems the BCR planned to bring out a new set of spherical standards that were to be analysed by as many primary particle-sizing methods as possible. (A primary method is defined as one where the dimensions of length and weight are directly traceable to International Standards and do not depend on second order effects such as diffraction patterns, turbidity, Brownian Motion or computer modelling.) Furthermore, prescriptive methodologies were to be developed to reduce the possibility of operator error.
The five primary methods selected were:
- Gravitational Sedimentation - guidelines developed by Prof. K Leschonski (Clausthal TU)
- Centrifugal Sedimentation - guidelines developed by Dr G Rideal (Whitehouse Scientific)
- Electroformed Sieve Analysis - guidelines developed by H G Mercus (Delft TU)
- Optical Microscopy - guidelines developed by Prof. J A Dodds and Dr M-N Pons (CNRS - France)
- Electrical Sensing Zone (Coulter) - guidelines by Dr P J Lloyd (Loughborough University)
As nearly 40 European Laboratories expressed an interest in certifying the new BCR references, Whitehouse Scientific was asked to produce a duplicate range of spherical standards in soda-lime glass both to short-list the laboratories and to provide a large quantity of single-shot bottles to back up the official range of standards.
This paper describes the subdivision of the Whitehouse bulk samples, reviews the development of the primary analysis guidelines and compares the results from the best laboratories for five of the polydisperse standards from 0.1 - 1mm to 150 - 650mm.
Results
1. Sample Subdivision
Custom built spinning rifflers developed Clausthal University and Whitehouse Scientific were used to subdivide the duplicate reference standards. Particle sizing data from the rifflers were independently evaluated at Loughborough University to ensure accurate subdivision. In the case of the Whitehouse riffler, 25 x 1g samples of the 10-100mm standard were taken from a 100 stage spinning riffler and analysed by the Electrical Sensing Zone method (Elzone - 280PC). The maximum deviation was only +/- 2.5% with the mean standard deviation across all the percentiles of only +/-1.07% confirming that all the samples sent to the participating laboratories were virtually identical.
100 stage spinning riffler & Particle size distribution variations from 25 positions around the spinning riffler (number data)
2. Particle Size Analysis Results
(a) Centrifugal Sedimentation
The small volume (180ml) Pipette Centrifuge as described by Allen 4 was used for this work. It is essentially a centrifugal analogue of the Andreasen Pipette and so utilises a similar methodology. The centrifuge was used only for the 0.1-1mm standard and run at 1500rpm. Seven extractions corresponded to a size range from 2.06mm down to 0.15mm. 5 analyses were performed from which a mean particle size distribution was calculated. A polynomial curve fitting routine was used to smooth the data points from each laboratory.
0.1 - 1mm Polydisperse Standard – summary of International results from the Pipette Centrifuge
(b) Gravitational Sedimentation
The Andreasen Pipette1 was used for the gravitational sedimentation of the 1-10mm and 3-30mm standards (the samples were dispersed in a 0.1% aqueous solution of sodium hexametaphosphate for 5 minutes in an ultrasonic bath). For each participating laboratory, the cumulative percent undersize points were plotted for 5 repeat analyses, from which mean sizes at the selected percentiles were calculated. The results were then compared to the other analytical techniques for the grade.
1 - 10mm Polydisperse Standard – summary of International results
3 - 30mm Polydisperse Standard – summary of International results
(c) Electroformed Sieve Analysis
Electroformed nickel sieves having square apertures accurate to +/-2 microns were specified for this method. The sieves were calibrated by microscope using the National Institute of Science and Technology (NIST)9 or the National Physical Laboratory (NPL) graticule5 as the traceable unit of length. Five analyses were performed from which the sizes at the prescribed percentiles were calculated.
10 - 100mm Polydisperse Standard – summary of International results
Electroformed sieve analysis was used to analyse the 10 - 100mm and the 150 - 650mm polydisperse standards. In the case of the 10 - 100mm standard, the analysis was performed either in suspension using an ultrasonic bath (40W, 40kHz) or as a dry powder using a sonic sifter e.g. Gilson. The method was exceptionally consistent and gave very repeatable results.
There was also good agreement with the other primary particle sizing methods.
In the case of the 150 - 650mm standard, where electromechanical shaking was employed, the method showed the best agreement of all the techniques and the results were almost identical to the microscopic analysis.
150 - 650mm Polydisperse Standard – summary of International results
(d) Microscopy and Image Analysis
Microscopy and Image Analysis were used for the 3-30mm, 10-100mm and 150-650mm powders. The minimum recommended particle count was 6000. The Miles-Lantuejoul6,7,8 correction was implemented for particles cut by the edge of the video frame.
As can be seen from the illustrations above, good correlation was seen with the other three methods.
(e) The Electrical Sensing Zone Method - (Coulter Principle)
The Electrical Sensing Zone (ESZ) method was employed for the 1-10mm, 3-30mm and 10-100mm samples. Great care was taken in preparing a representative standard from a well-dispersed suspension. All the instruments used were pre-calibrated using traceable, monodisperse reference standards.
3. Conclusion
The BCR round robin has shown that, by using spherical materials, smaller 1g (single-shot) samples and prescriptive analytical guidelines, it has been possible to obtain very good agreement between the four primary particle sizing methods. The maximum deviations were about 7% and typically below 5%. This compares very favourably with the results previously achieved by the single method of sedimentation used for the BCR quartz standards back in 1980.
4. The Complete Range of Single-Shot Standards
Since the completion of this work, the range of standards has been increased to seven as follows:
0.1 - 1, 1 - 10, 3 - 30, 10 - 100, 50 - 350, 150 - 650 and 500 -2000mm
The single-shot bottles contain from 10mg to 10g so can be used for any particle sizing technique without the need for further subdivision.
5. Participating Laboratories
The following table shows the final analytical laboratories selected from the initial team of forty candidate laboratories.
5. Acknowledgements
The authors wish to thank the sponsors of the work, the Community Bureau of Reference (BCR) Brussels, and the main co-ordinator, AEA Technology (UK).]
6. References
- Wilson, R., Leschonski, K., Alex, W., Allen, T Koglin, B., Certification Report on Reference Materials of Defined Particle Size, Commission of the European Communities Report EUR 6825 EN, 1980.
- Allen, T and Davies R., Evaluation of Instruments for Particle Size Analysis. p17-46 in Proc. 4th European Symposium on Particle Characterisation (PARTEC), Nuremberg, Germany 1989.
- Mercus, H.G., Bischof, O., Drescher, S., and Scarlett, B., Precision and Accuracy in Particle Sizing. Round-robin Results from Sedimentation, Laser Diffraction and Electrical Sensing Zone, using BCR 67 and BCR 69 in Proc.6th European Symposium on Particle Characterisation (PARTEC), Nuremberg, Germany 1995
- Allen, T., Particle Size Measurement, Chapman and Hall, 1981
- National Physical Laboratory, Teddington, Midddlesex, TW11 0LW, England
- Miles, R. E., ‘Stochastic Geometry’ p228 ed Kendall, D.G. and Harding, E, F., Wiley and Sons, N.Y (1974)
- Lantuejoule, C., Microscopica Acta, S4, p266, (1980)
- French Standard, AFNOR NF X11-696 (1989)
- National Institute of Science and Technology (NIST), Gaithersburg, MD 20899-0001
PDF Brochures