Fibrous particle analysis has consistently presented a challenge. In systems that are volumetrically based like laser diffraction, dynamic light scattering and Coulter Counters Equivalent Spherical Diameter (ESD) is the only measurement supplied, which is obviously not appropriate for fibers (example shown in Figure 1).
Due to this hindrance, manual microscopy has been the principal technique for the classification of fibers when both width and length are being analyzed. This process can be accelerated through the use of Flow Imaging Microscopy, where computer algorithms are employed rather than a human observer. The overview results of a FlowCam® characterization of industrial fibers are shown in Figure 2.
Figure 1. How a volumetric-based system calculates ESD for a fiber: a 290 µm x 11 µm fiber is characterized as a sphere of diameter = 64 µm. Image Credit: Yokogawa Fluid Imaging Technologies, Inc.
Figure 2. Screenshot of FlowCam analysis of industrial fibers. Image Credit: Yokogawa Fluid Imaging Technologies, Inc.
This is relatively easy to achieve in straight fibers but becomes considerably more challenging in curled fibers. In several applications, measuring the extent of curl in the fibers can be crucial to the performance of the final product. With its VisualSpreadsheet® software, the FlowCam offers high-speed measurement of fiber width and length, along with the appropriate classification of fiber curl and straightness.
Using the FlowCam, a sample of cellulose fibers suspended in acetone was analyzed. The bulk of imaging particle analysis systems use a technique for width and length measurement according to Feret’s diameter. Feret’s diameter is also known as a caliper diameter. It is determined by the distance from two tangents produced by the particle’s outline, much like when employing a caliper (illustrated in Figure 3).
Figure 3. Feret’s Diameter used to calculate length and width. Image Credit: Yokogawa Fluid Imaging Technologies, Inc.
VisualSpreadsheet and the FlowCam exceed the capacities of fundamental Feret measurements and also offer Geodesic Thickness and Geodesic Length as quantified measurements for each particle. For the same fiber particle image depicted in Figure 3, Figure 4 exhibits the results for these two measurements (particle #523 from the initial run as presented in Figure 2).
Figure 4. Geodesic Length and Thickness calculated by VisualSpreadsheet. Note the differences versus Figure 3. Image Credit: Yokogawa Fluid Imaging Technologies, Inc.
Additionally, the image of fiber particle #523 from the first run in Figure 2 is exhibited in Figure 5, alongside the exact measurements for both the Feret-based calculations and Geodesic-based calculations as computed by VisualSpreadsheet. As can be seen in Figure 5, VisualSpreadsheet includes two additional calculations to depict the fibers: Fiber Curl and Fiber Straightness.
Fiber Straightness is identified by separating the ratio of Length (Feret-based) by the Geodesic Length. A completely straight fiber would be determined by a straightness value of one, and as the complexity of the fiber increases (i.e. non-straightness), this value moves closer to zero.
Fiber Curl is established by dividing the ratio of Geodesic Length by Length (Feret-based) minus one. A totally straight fiber would be determined by a value of zero, and increasingly greater values depict an increasing curl.
As would be predicted, by comparing the measurements of the curled particle in Figure 5 with the straight fiber particle in Figure 6, the curled particle possesses a lower ‘Fiber Straightness’ measurement and a higher “Fiber Curl” measurement. It should be noted that in the straight particle, the lengths and widths determined by the two different techniques come very close.
Figure 5. Curved fiber particle image with length and width measured using the Feret method and Geodesic method. Image Credit: Yokogawa Fluid Imaging Technologies, Inc.
Figure 6. Straight fiber particle image with length and width measured using the Feret method and Geodesic method. Image Credit: Yokogawa Fluid Imaging Technologies, Inc.
Results and Conclusions
In applications where fiber straightness is critical in the conclusive use of the fibrous materials, filters can be pre-installed in VisualSpreadsheet to report Fiber Straightness or Fiber Curl automatically for the entire sample when the run has been finalized.
For the particular material used in this instance, a sample was deemed to pass if over half of the fibers possessed a Fiber Straightness above 0.75. As can be seen in the summary statistics in Figure 7, this sample would only just pass, including fibers with a Straightness of >0.75 to a count percentage of 53.47.
Figure 7. Results of automated filters for fiber straightness on the overall fiber data from Figure 2. Image Credit: Yokogawa Fluid Imaging Technologies, Inc.
Finally, by double-clicking on the filter, the images of those fibers that meet the specification, as seen in Figure 8, can be presented for review by opening the View screen in VisualSpreadsheet.
Figure 8. Screenshot of FlowCam analysis of fibers, with images on right representing those with Fiber Straightness >0.75. Image Credit: Yokogawa Fluid Imaging Technologies, Inc.
FlowCam with VisualSpreadsheet is a pioneering system for the effective classification of fibrous particles relating to their shape. As outlined in this application, over 10,000 fibers were characterized automatically in just 23 seconds, producing data with considerably greater statistical confidence than what was previously deemed possible.
This information has been sourced, reviewed and adapted from materials provided by Yokogawa Fluid Imaging Technologies, Inc.
For more information on this source, please visit Yokogawa Fluid Imaging Technologies, Inc.