Allyson Pulsipher and Dr. Scott Thomson, Department of Mechanical Engineering

Over the past 25 years, research has been done to analyze the response of vocal fold tissue to tensile loading. Although the research provided better understanding of vocal fold behavior, there has been significant variability in the results of each research endeavor. Many factors may have contributed to the differences, such as dissimilar experimental set-ups, difficulty of obtaining and analyzing vocal fold samples, etc. However, the most significant factor of variance is most likely related to the structure of the tissue itself.

Cartilage, bones, and some soft tissues in the body are composed largely of small protein fibers called collagen. The vocal folds have high densities of these fibers. The presence of collagen fibers in cartilage causes a non-linear response to tensile loading. In its relaxed form, collagen has a wave-like shape. As these fibers are stretched, they exert minimal tensile resistive force as they straighten and become more parallel. However, once they are straight, they cause a significant increase in the tension until they finally deform and fail.

The aim of this project was to model the non-linear behavior of collagenous vocal fold tissue through experimentation with various fiber materials and densities. In this project, fibers of various types and materials were added to liquid silicone compound and allowed to cure in a polymer mold, producing synthetic models of the vocal folds with non-linear properties. Three general types of fibers were considered: cotton, foam, and an acrylic/polyester mix. Each material was cut to a specific length (roughly 3 cm), then immersed in the liquid silicone prior to curing. The resulting models were then tested in a tensile testing apparatus to measure and compare the response of each to tensile loading. The elongation, force, stress, strain, and modulus of elasticity of each sample were calculated from the measurements.

Each material yielded different results, with foam and cotton being significantly stiffer than the acrylic/polyester fibers. From all the materials tested, the acrylic/polyester fibrous models most closely approximated the properties of vocal folds reported by R.F. Chan. Figure 1 compares his published results with our experimental results using acrylic/polyester fibers. The red and blue lines show the behavior of the cover (outside layer) and the ligament (inner layer) of the vocal fold, respectively. These two layers differ slightly in stiffness, but have similar general trends. Our results are represented by the three remaining lines on the graph. In each case, the acrylic/polyester mixed fibers were used. Each were tested with a modulus of 2.2 kPa, but the amount of fibers per sample was varied.

The samples with 9mg and 31mg of fibers represent low-strain and high-strain extremes. The best approximation of the collagenous vocal fold tissue, however, is the third sample. In this case, fiber amounts are estimated to be approximately 15-20mg. The graph resulting from stress and strain data of this material is almost identical in shape to that reported for the male ligament. It is also very similar in range to that of the male cover.
Our results illustrate two major conclusions. First, it is feasible to approximate the behavior of collagenous vocal fold tissue using acrylic/polyester mixed fibers. Second, behavior is most closely approximated when the fiber amount is within the range of 15-20mg per sample (where each sample is roughly 3cm in length).

Although our results contribute significantly to the development of synthetic vocal folds, they require further experimentation to enhance their accuracy and repeatability. Research should be continued in two areas. First, the issue of density uniformity should be addressed. The acrylic/polyester fibers being used currently are difficult to distribute with uniformity because of their large size and raveled texture. Smaller fibers that could be more methodically placed would provide a more uniform density, enhancing the accuracy and repeatability of our results.

A second issue to address for improved results is the issue of isotropy. Our samples were created by running fibers longitudinally through the silicone models. Each sample was stretched longitudinally to test the material properties in that direction. Although properties in the longitudinal direction are most significant physiologically, future research should consider the material properties in several directions to improve the accuracy of this material.

I would like to express appreciation for your generosity in granting me ORCA funding. Your financial contributions have made possible many learning activities directly related to engineering, including but not limited to this research. I would like to especially thank Dr. Scott Thomson for his mentorship and teaching. He has not seen the effect of his influence, but I have seen it as I have continued my education and research endeavors. His training helped me learn to ask questions and follow up with the results. Because of the contributions of ORCA and Dr. Thomson, I have been able to learn how to contribute in the field of engineering.

References

• Alipour-Haghighi, Fariborz, & Titze, Ingo R. (1991). “Elastic Models of Vocal Fold Tissues.” J. Acoustical Society of America., Vol. 90, No. 3, September 1991.
• Nordin, Margareta, & Frankel, Victor H. (Eds.). (1989). Basic Mechanics of the Musculoskeletal System (2nd ed.). Malvern: Lea & Fribiger.
• Chan, R. F., M; Young, L; Tirunagari, N (2007). “Relative contributions of collagen and elastin to elasticity of the vocal fold under tension.” Annals of Biomedical Engineering 35(8): 1471-83.