Accurate, repeatable, and efficient fiber-length measurements

31 October 2017
Sebastian Goris, Tim A. Osswald, Teresa Back, Dietmar Drummer, Angel Yanev, and Dave Brands
Conventional fiber-length measurement techniques for discontinuous fiber-reinforced composites are evaluated and a novel characterization methodology is proposed.

Discontinuous fiber-reinforced polymers have become important materials in the transportation industry because of their favorable material properties, low manufacturing costs, and lightweight characteristics.1, 2 The configuration of the reinforcing fibers, however, changes significantly over the course of the entire polymer production process and these changes are reflected in numerous ways (e.g., favored fiber orientation, fiber jamming, and separation of fibers from the matrix).3–6 Indeed, one of the major challenges in processing discontinuous fiber-reinforced thermoplastics is the process-induced reduction of fiber length, often referred to as fiber attrition or fiber breakage. For example, during injection molding, the embedded fibers are subjected to extensive stresses from the plasticating phase and injection stages, which inevitably reduces their length.7, 8 It is therefore important to quantify the residual fiber-length distribution so that the fiber breakage phenomenon can be studied experimentally. The characterization of fiber-length distribution is a cumbersome task, however, because even small samples contain millions of fibers. Moreover, no standardized measurement protocol has yet been established.9

The majority of the currently applied measurement procedures (see Table 1) for assessing the fiber-length distribution of fiber-reinforced thermoplastics have four main steps in common. In the first step—matrix removal—the matrix is removed via pyrolysis (after the sample has been extracted from the molded part). Next—down-sampling and fiber dispersion—a subset of the fibers is selected and prepared for scanning. In the third step—digital imaging—either a microscope or an optical document scanner is used to create a digital image. Finally—during fiber detection—image processing software is used to analyze the digital image and determine the length of the individual fibers. During this step, the fiber detection is achieved either manually (by clicking on the endpoints of the fibers) or with the support of (semi-) automated image-processing algorithms. Despite the similarities between the many measurement techniques, there are also large variations in important aspects of the procedures (see Table 1).

Overview of fiber-length measurement techniques used in recent studies. PP: Polypropylene. GF: Glass fiber. PA: Polyamide. SFT: Short-fiber-reinforced thermoplastic. NF: Natural fibers. CF: Carbon fibers.

MaterialCorrection forFiber dispersionFiber detectionFibers per sampleStudy
PP/GFNoManual dispersionManual100Bajracharya et al.10
PP&PA12/GFNoManual dispersionManual300Bumm et al.11
PP/GFNoManual dispersionManual500Priebe et al.1
PP/GFNoManual dispersionManual500Wang et al.12
PP/GFNoManual dispersionManual500Thomason13
PP/GFNoManual dispersionManual500Jin et al.14
PP/GF&CFYesManual dispersionManual2,000Nguyen et al.15
PP/GF&CFYesManual dispersionManual2,000Kunc et al.16
PA6/GFNoDiluted suspensionManual400Inceoglu et al.17
PA6/GFNoDiluted suspensionManual400Yilmazer et al.18
PA6/GFNoDiluted suspensionManual1,000Lafrance et al.19
PP/GFNoDiluted suspensionManual1,000Inoue et al.20
GFNoManual dispersionSemi-automatic1,000Huq et al.21
PP/GFNoDiluted suspensionSemi-automatic800Zhuang et al.22
PP/GFNoDiluted suspensionSemi-automatic2,000Teixeria et al.23
PP/GFNoDiluted suspensionSemi-automatic2,000Ren et al.24
PP/GFNot neededDiluted suspensionSemi-automatic2,000Rohde et al.8
PP/GFNot neededDiluted suspensionSemi-automatic3,000Hartwich et al.25
PP/GFNoManual dispersionAutomatic1,000Giusti et al.26
SFTNoDiluted suspensionManual360±60ISO27
GF/CF/NFYesTurbulent airAutomatic>15,000Goris et al.9

To test the repeatability of conventional manual fiber-length measurement procedures, we analyzed9 10 identical injection-molded samples of long-glass fiber-reinforced polypropylene, with a glass concentration of 30 or 40wt% (PPGF30 and PPGF40, respectively). Following the published measurement protocols, we extracted samples from the molded part, removed the matrix via pyrolysis, and collected a random set of fibers for the fiber-length analyses. The results from a conventional measurement protocol are illustrated in Figure 1. These results show that there is substantial variation in the measured fiber-length distributions, i.e., varying from 1.03 to 3.96mm (0.95mm standard deviation). In addition, the number of analyzed fibers can vary by more than an order of magnitude, and it is unclear whether a set of 1000 (or fewer) fibers can be statistically representative. Our results thus emphasize the difficulty of characterizing (in terms of comparability, repeatability, and accuracy) fiber length samples, and highlight the need for a standard measurement protocol. To that end, we have developed a novel technique that can be used to provide highly reproducible measurements in a timely manner.

Results from manual fiber-length measurements of 10 identical 40wt% glass, injection-molded long-glass fiber-reinforced polypropylene (PPGF40) samples. The weight-average fiber length (red bars) and number of measured fibers (white squares) are given for each measurement.

In the first stage of our measurement technique (illustrated in Figure 2), we conduct pyrolysis to remove the sample matrix. We then perform a representative downsampling step, adapted from previous work.15, 16 All fibers in the defined region are collected by injecting a UV-curable epoxy resin through the thickness of the fiber bed, by using a fine hypodermic needle. This downsampling step allows the collection of a controlled and representative subset of fibers that statistically resemble the fiber-length distribution of the initial sample. After the epoxy plug has cured, we use a brush to carefully remove all the surrounding fibers and we conduct a second pyrolysis step to recover all the fibers that are embedded within the epoxy. In the next step, we use bursts of slightly compressed air to disperse the fibers within an enclosed chamber system.9 In this way, the air turbulence causes the fibers to disentangle and disperse uniformly. Finally, after we have obtained a digital image of the dispersed fibers, we use an image processing algorithm (developed in-house) to automatically detect all the fibers (including bent and intersecting fibers) in the image. To ensure statistically representative results, we analyze at least 15,000 fibers for each sample and process the data set statistically (i.e., we calculate the number-average and weight-average fiber length).

Illustration of the steps in the proposed fiber-length measurement technique.

As part of our study, we also compared our novel measurement technique with a manual measurement, the commercially available (from Karg Industrietechnik, Germany) ‘FASEP’ measurement system,25 and a full analysis (provided by SABIC, the Netherlands). During the full analysis, the entire fiber population in the sample was measured, i.e., there was no downsampling step and up to 750,000 fibers were assessed in a single measurement. The results of our comparison, for the PPGF40 and PPGF30 samples are summarized in Figure 3. In particular, from our manual measurement we obtained a value that was 27% higher than our weight-average fiber length measurement, and which had a substantial standard deviation. In general, however, we observe a strong agreement between the results of our proposed method, FASEP, and the full analysis procedures.

Comparison of the number-average and weight-average fiber-length measurements obtained from the new method proposed in this work, the commercial ‘FASEP’ approach, the full analysis procedure (from SABIC), and a manual technique for the (a) PPGF40 and (b) PPGF30 (i.e., long-glass fiber-reinforced polypropylene with 30wt% glass concentration) samples. Error bars indicate standard deviations.

In summary, we have highlighted the important need for a standard fiber-length measurement technique for discontinuous fiber-reinforced composites. We have thus developed a new technique with which we can obtain accurate and highly reproducible fiber-length measurements in an efficient manner. In the next steps of our work, we will apply the developed measurement technique as part of a fundamental study on fiber attrition during long fiber thermoplastic processing. Our aim is to thus obtain a fundamental understanding of the underlying physics of fiber breakage.


Sebastian Goris
University of Wisconsin–Madison

Sebastian Goris is a PhD candidate in mechanical engineering and the chief engineer of the PEC.

Tim A. Osswald
University of Wisconsin–Madison

Tim Osswald is a professor of mechanical engineering and director of the PEC.

Teresa Back
University of Erlangen-Nuremberg

Teresa Back was a visiting scholar at the PEC and is now a design engineer at Harburg-Freudenberger Maschinenbau GmbH (Germany).

Dietmar Drummer
University of Erlangen-Nuremberg

Dietmar Drummer is a professor and head of the LKT.

Angel Yanev

Angel Yanev is a molding and processing specialist.

Dave Brands

Dave Brands is an application development engineer.


  1. M. Priebe and R. Schledjewski, Processing and properties of glass/polypropylene in long fibre compounding extrusion, Plastics Rubber Compos. 40, pp. 374-379, 2011.

  2. H. E. Friedrich, Leichtbau in der Fahrzeugtechnik, Springer, 2013.

  3. J. H. Phelps and C. L. Tucker, An anisotropic rotary diffusion model for fiber orientation in short- and long-fiber thermoplastics, J. Non-Newton. Fluid Mechan. 156, pp. 165-176, 2009.

  4. J. H. Phelps, A. I. A. El-Rahman, V. Kunc and C. L. Tucker, A model for fiber length attrition in injection-molded long-fiber composites, Compos. Part A: Appl. Sci. Manufac. 51, pp. 11-21, 2013.

  5. T. A. Osswald and G. Menges, Materials Science of Polymers for Engineers, Hanser, 2012.

  6. S. Goris, J. Puentes and T. A. Osswald, Polymer-composites manufacturing processes, Manufacturing Engineering Handbook, McGraw Hill, 2015.

  7. A. Durin, P. De Micheli, J. Ville, F. Inceoglu, R. Valette and B. Vergnes, A matricial approach of fibre breakage in twin-screw extrusion of glass fibres reinforced thermoplastics, Compos. Part A: Appl. Sci. Manufac. 48, pp. 47-56, 2013.

  8. M. Rohde, A. Ebel, F. Wolff-Fabris and V. Alstädt, Influence of processing parameters on the fiber length and impact properties of injection molded long glass fiber reinforced polypropylene, Int'l Polym. Process. 26, pp. 292-303, 2011.

  9. S. Goris, T. Back, A. Yanev, D. Brands, D. Drummer and T. A. Osswald, A novel fiber length measurement technique for discontinuous fiber-reinforced composites: a comparative study with existing methods, Polym. Compos., 2017.

  10. R. M. Bajracharya, A. C. Manalo, W. Karunasena and K.-T. Lau, Experimental and theoretical studies on the properties of injection moulded glass fibre reinforced mixed plastics composites, Compos. Part A: Appl. Sci. Manufac. 84, pp. 393-405, 2016.

  11. S. H. Bumm, J. L. White and A. I. Isayev, Glass fiber breakup in corotating twin screw extruder: simulation and experiment, Polym. Compos. 33, pp. 2147-2158, 2012.

  12. J. Wang, C. Geng, F. Luo, Y. Liu, K. Wang, Q. Fu and B. He, Shear induced fiber orientation, fiber breakage, and matrix molecular orientation in long glass fiber reinforced polypropylene composites, Mater. Sci. Eng. A 528, pp. 3169-3176, 2011.

  13. J. L. Thomason, The influence of fibre length and concentration on the properties of glass fibre reinfored polypropylene. 6. The properties of injection moulded long fibre PP at high fibre content, Compos. Part A: Appl. Sci. Manufac. 36, pp. 995-1003, 2005.

  14. G. Jin, X. Lin, G. Tian, S. Zhang, M. Wang and X. Wang, Entrance flow of long glass fiber reinforced polypropylene through contraction die, J. Reinforced Plastics Compos. 35, pp. 111-123, 2016.

  15. B. N. Nguyen, S. K. Bapanapalli, J. D. Holbery, M. T. Smith, V. Kunc, B. J. Frame, J. H. Phelps and C. L. Tucker, Fiber length and orientation in long-fiber injection-molded thermoplastics---part I: modeling of microstructure and elastic properties, J. Compos. Mater. 42, pp. 1003-1029, 2008.

  16. V. Kunc, B. Frame, B. N. Nguyen, C. L. Tucker and G. Velez-Garcia, Fiber length distribution measurement for long glass and carbon fiber reinforced injection molded thermoplastics, Proc. 7th Annu. Automotive Compos. Conf. Exhibit. 1, pp. 866, 2007.

  17. F. Inceoglu, J. Ville, N. Ghamri, J. L. Pradel, A. Durin, R. Valette and B. Vergnes, Correlation between processing conditions and fiber breakage during compounding of glass fiber-reinforced polyamide, Polym. Compos. 32, pp. 1842-1850, 2011.

  18. U. Yilmazer and M. Cansever, Effects of processing conditions on the fiber length distribution and mechanical properties of glass fiber reinforced nylon-6, Polym. Compos. 23, pp. 61-71, 2002.

  19. E. Lafranche, P. Krawczak, J.-P. Ciolczyk and J. Maugey, Injection moulding of long glass fiber reinforced polyamid 66: processing conditions/microstructure/flexural properties relationship, Adv. Polym. Technol. 24, pp. 114-131, 2005.

  20. A. Inoue, K. Morita, T. Tanaka, Y. Arao and Y. Sawada, Effect of screw design on fiber breakage and dispersion in injection-molded long glass-fiber-reinforced polypropylene, J. Compos. Mater. 49, pp. 75-84, 2015.

  21. A. M. A. Huq and J. Azaiez, Effects of length distribution on the steady shear viscosity of semiconcentrated polymer-fiber suspensions, Polym. Eng. Sci. 45, pp. 1357-1368, 2005.

  22. H. Zhuang, R. Pu, Y. Zong and G. C. Dai, Relationship between fiber degradation and residence time distribution in the processing of long fiber reinforced thermoplastics, eXPRESS Polym. Lett. 2, pp. 560-568, 2008.

  23. D. Teixeira, M. Giovanela, L. B. Gonella and J. S. Crespo, Influence of injection molding on the flexural strength and surface quality of long glass fiber-reinforced polyamide 6.6 composites, Mater. Design 85, pp. 695-706, 2015.

  24. P. Ren and G. Dai, Fiber dispersion and breakage in deep screw channel during processing of long fiber-reinforced polypropylene, Fiber Polym. 15, pp. 1507-1516, 2014.

  25. M. R. Hartwich, N. Höhn, H. Mayr, K. Sandau and R. Stengler, FASEP ultra-automated analysis of fibre length distribution in glass-fibre-reinforced products, Proc. SPIE 7389, pp. 738921, 2009.

  26. R. Giusti, G. Lucchetta and I. Dubrovich, Effect of screw geometry on long glass fiber breakage during injection molding, SPE ANTEC, 2015.

  27. Plastics---glass-fibre-reinforced products---determination of fibre length Int'l Org. Standardization ISO 22314, 2006.

DOI:  10.2417/spepro.006973