With the introduction of the Mastersizer 3000 comes the requirement to transfer methods from previous instruments. Laser diffraction is commonly used in quality control or regulated environments where specifications, and associated tolerances, are set on particle size results. Due to the high reproducibility of laser diffraction measurements these tolerances are often relatively narrow. In this environment it is a significant advantage to be able to transfer specifications (along with the method) to the new instrument.
|
To transfer methods between the Mastersizer 2000 and 3000 we need to consider the measurement parameters, analysis parameters and the state of dispersion of the samples (discussed in detail in a separate application note [1]). And when these factors have been considered, excellent agreement between the Mastersizer 2000 and Mastersizer 3000 can be achieved. However there are some cases where, even after these factors have been considered, differences between the results from the two systems are still observed. These cases include:
In these cases, the Emulated Mastersizer 2000 models (General Purpose, Single Narrow Mode, and Multiple Narrow Modes) are available to improve the agreement between the two systems. The Emulated Mastersizer 2000 models apply the dynamic range and analysis from the Mastersizer 2000 to measurements made on the Mastersizer 3000. These analysis models can be used as an aid to method transfer and, if required, in the transfer of specifications to from the Mastersizer 2000 to the Mastersizer 3000.The following sections discuss why differences are observed between the Mastersizer 2000 and Mastersizer 3000 for the sample types described above. In all of the examples below the general purpose models have been used on both systems.
Improvements in the dynamic range from the Mastersizer 2000 to the Mastersizer 3000 have been achieved through changes in the optical design of the system. Advances in detector electronics allow smaller detectors to be added at narrower angles, increasing the capability of the system to measure larger particles. A more powerful blue light source, from which data is collected over a wider angular range, has improved the submicron resolution of the system. This gives the Mastersizer 3000 the ability to measure from 10nm to 3500μm, compared to the Mastersizer 2000 which has a measurement range from 20nm to 2000μm. Therefore, if you measure samples containing particles at the extremes of the measurement range of either system then differences are expected in the measured particle size distributions.
An example of differences in results at the fine end of the dynamic range is shown in Figure 2. This shows the particle size distributions of an emulsion sample measured on the Mastersizer 2000 and Mastersizer 3000, as well as the Emulated Mastersizer 2000 result. The results on the Mastersizer 3000 show significantly more material below 100nm due to the increased sensitivity of the system in this region. Using the Emulated Mastersizer 2000 analysis applies the dynamic range and analysis of the Mastersizer 2000 to the Mastersizer 3000 data and brings the results into much better agreement with the Mastersizer 2000.
|
| Dv 10 (μm) | Dv 50 (μm) | Dv 90 (μm) | |
|---|---|---|---|
| Emulsion MS3000 | 0.03 | 0.11 | 0.30 |
| Emulsion Emulated MS2000 | 0.07 | 0.15 | 0.31 |
| Emulsion MS2000 | 0.07 | 0.14 | 0.28 |
Coffee is an application which may show differences at the top end of the measurement range as some grades contain particles larger than 2mm. Figure 3 shows the particle size distributions measured for a coffee sample where the Mastersizer 3000 shows material larger than 2mm. As this is beyond the measurement range of the Mastersizer 2000 the distribution stops at 2mm, but applying the Emulated Mastersizer 2000 analysis model to the Mastersizer 3000 data produces comparable results.
|
| Dv 10 (μm) | Dv 50 (μm) | Dv 90 (μm) | |
|---|---|---|---|
| Coffee MS3000 | 401 | 951 | 1910 |
| Coffee emulated MS2000 | 385 | 892 | 1510 |
| Coffee MS2000 | 398 | 878 | 1490 |
Using the Emulated Mastersizer 2000 analysis model in both of these examples aids the method transfer process by showing that the differences in the results is due to the change in the sensitivity of the system rather than any change in the sample.
The Mastersizer 3000 analysis uses a different mathematical inversion process to convert the measured scattering data into a particle size distribution. This algorithm has been changed to increase the sensitivity of the analysis. However, it does mean that the size distributions obtained using each algorithm on scattering data from the same sample may have slightly different shapes, due to the way the solution is reached. An example of which is shown for a battery material in Figure 4.
If a method transfer process is followed then the agreement between the two systems is extremely good, see Figure 4 for a battery material (Lithium Iron Phosphate, LiFePO4). Without the emulated Mastersizer 2000 analysis the results on the two systems are within the ISO limits for the repeatability for a laser diffraction measurement [2]. However there is a difference in the shape of the distribution and by using the Emulated Mastersizer 2000 model the results are almost identical.
|
| Dv 10 (μm) | Dv 50 (μm) | Dv 90 (μm) | |
|---|---|---|---|
| LiFePO4 Mastersizer 3000 | 0.77 | 1.73 | 3.96 |
| LiFePO4 Emulated Mastersizer 2000 | 0.75 | 1.72 | 3.87 |
| LiFePO4 Mastersizer 2000 | 0.75 | 1.71 | 3.84 |
In this case the Emulated Mastersizer 2000 analysis model can be used to understand what has caused the slight differences in the results between the two systems.
Due to the increased resolution of the analysis used in the Mastersizer 3000, small modes of agglomerates or coarse particles can be measured on the Mastersizer 3000 which may not have been reported on the Mastersizer 2000. An example of which is shown in Figure 5. This shows the particle size distributions of a cement sample measured using the Mastersizer 3000, which shows a small mode of large particles (or agglomerates) at approximately 300μm. The presence of these large particles in the sample can be verified using image analysis, which also allows these particles to be identified as large primary particles or undispersed agglomerates.
These large particles are not observed in Mastersizer 2000 measurements, and by applying the Emulated Mastersizer 2000 analysis model to the Mastersizer 3000 data this mode is not reported. In this case the Mastersizer 2000 emulation model could be used as part of method transfer to determine whether the differences in results are caused by changes in the dispersion mechanisms or the analysis.
|
| Dv 10 (μm) | Dv 50 (μm) | Dv 90 (μm) | |
|---|---|---|---|
| Cement MS3000 | 1.77 | 9.27 | 26.7 |
| Cement Emulated MS2000 | 1.74 | 9.08 | 25.6 |
The Mastersizer 3000 collects scattering data at wider angles than the Mastersizer 2000, which improves the sensitivity to fine particles even without the use of the secondary blue light source. Therefore, in dry measurements the increased angular range, combined with improvements in the dry measurement cell, increases the sensitivity to submicron particles. Therefore, differences may be observed in the measurement of dry samples containing significant amounts of material smaller than 1μm.
Figure 6 shows the particle size distributions of a calcium carbonate sample measured in dry dispersion using the Mastersizer 2000 and Mastersizer 3000. As the Mastersizer 3000 can capture data at higher angles the measured particle size distribution shows material smaller than 200nm which is not reported by the Mastersizer 2000. The Emulated Mastersizer 2000 analysis model can then be used to produce equivalent results to the Mastersizer 2000 by restricting the data that is used, and limiting the dynamic range.
|
| Dv 10 (μm) | Dv 50 (μm) | Dv 90 (μm) | |
|---|---|---|---|
| Fine dry sample MS3000 | 0.53 | 4.72 | 28.2 |
| Fine dry sample Emulated MS2000 | 0.68 | 4.93 | 28.9 |
| Fine dry sample MS2000 | 0.81 | 4.96 | 29.0 |
In this example the Emulated Mastersizer 2000 analysis model can aid in method transfer by determining whether the differences in the results are due to differences in the state of dispersion or the sensitivity of the system.
The Mastersizer 2000 emulation model has been designed to aid the process of method transfer from the Mastersizer 2000 to Mastersizer 3000. Improvements in the optical design and analysis implemented in the Mastersizer 3000 have improved the sensitivity and resolution of the system. However, the advantages in terms of increased dynamic range can also cause problems when methods and specifications are transferred from one system to another. Hence, the Mastersizer 2000 emulation model can be used to apply the analysis and dynamic range of the Mastersizer 2000 to measurements made on the Mastersizer 3000.
In general, when the particle size range of the material is within the dynamic range of both systems, excellent comparability between the Mastersizer 2000 and Mastersizer 3000 is achieved and the emulation model is not required. However there are some cases where better agreement between the two systems can be achieved using the Emulated Mastersizer 2000 analysis model. These cases include: samples at the extremes of the dynamic range, polydisperse samples containing non-spherical particles, dry measurements of samples with significant submicron populations and samples containing small mode of agglomerates or coarse particles. This technical note has provided examples of materials where the Emulated Mastersizer 2000 model has improved the comparability of results obtained on the Mastersizer 2000 and Mastersizer 3000 systems.
