Manufacturing Process for Non-Adhesive Super-Soft Vocal Fold Models

Summary

This study demonstrates the manufacturing of non-sticky and super-soft vocal fold models by introducing a specific way to create the vocal fold layers, providing a detailed description of the manufacturing procedure, and characterizing the properties of the models.

Abstract

This study aims to develop super-soft, non-sticky vocal fold models for voice research. The conventional manufacturing process of silicone-based vocal fold models results in models with undesirable properties, such as stickiness and reproducibility issues. Those vocal fold models are prone to rapid aging, leading to poor comparability across different measurements. In this study, we propose a modification to the manufacturing process by changing the order of layering the silicone material, which leads to the production of non-sticky and highly consistent vocal fold models. We also compare a model produced using this method with a conventionally manufactured vocal fold model that is adversely affected by its sticky surface. We detail the manufacturing process and characterize the properties of the models for potential applications. The outcomes of the study demonstrate the efficacy of the modified fabrication method, highlighting the superior qualities of our non-sticky vocal fold models. The findings contribute to the development of realistic and reliable vocal fold models for research and clinical applications.

Introduction

Vocal fold models are used to simulate and investigate human voice production under normal and pathological conditions^1,2. One of the biggest challenges in creating vocal fold models is to achieve a realistic softness and flexibility that closely approximates those of humans. To achieve these properties, silicone elastomers are often used, which are diluted with high amounts of silicone oil to achieve the corresponding elasticity moduli^3,4. Another crucial factor in creating realistic vocal fold models is layering, as vocal folds consist of multiple layers of varying softness, which determine the pattern of flow-induced vibration and the frequency at which vibration is possible.

In this study, we created a typical vocal fold model. We used the common geometry provided by Scherer⁵, which represents typical dimensions for male vocal folds with 17 mm length according to Zhang⁶ and consists of three layers: one layer for the vocalis muscle (body layer), one for the entire mucosal layer (cover layer), and one for the epithelium. This structure can be seen in the coronal cross-section view in Figure 1.

Figure 1: Coronal cross-section of the larynx modules. Coronal cross-section of the larynx modules illustrating the widest width of the vocal folds (8.5 mm). Each vocal fold comprises a body layer, a cover layer, and an epithelium layer. This figure has been modified from¹³. Reproduced from Häsner, P., Prescher, A., Birkholz, P. Effect of wavy trachea walls on the oscillation onset pressure of silicone vocal folds. J Acoust Soc Am.149 (1), 466-475 (2021) with the permission of the Acoustical Society of America. Please click here to view a larger version of this figure.

Other publications use partially only one layer⁷, two layers without epithelium layer² or model the mucosa with multiple layers³. Usually, the layers are cast from the inside out, i.e., starting with the deepest layer. The epithelium, which is very thin with 30 µm thickness, is cast at the end over the entire body to envelop it with a sturdy skin⁸.

The cover layer in the model is the softest part, with Young's modulus of about 1.1 kPa⁹. For the body layer, the approximate Young's modulus in the transverse direction using in vitro measurements¹⁰ is 2 kPa. In vivo, the Young's modulus of the thyroarytenoid muscle may be higher due to the presence of fibers in the longitudinal direction as well as the possible tensing of the muscle. To achieve this extremely low Young's modulus, it is necessary to add a high amount of silicone oil to the silicone mixture (approximately 72%). However, the manufacturer strongly advises against using an oil proportion greater than 5%. In general, the addition of silicone oil to the elastomer is intended to increase flow and drip time, as well as reduce the shrinkage of the cured silicone polymer. It helps the silicone to cure more uniformly, thereby reducing stress in the material. Its purpose is to optimize the moldability and properties of the cured material, rather than to increase its softness, although this is also a consequence. This is because silicone oil is chemically inert, meaning that it cannot polymerize itself and is not integrated into the network of the silicone polymer¹¹. Instead, it remains as a liquid phase in the polymer matrix, weakening the polymer structure at higher levels and potentially causing it to dissolve out of the cured material and adhere to the surface. As a result, other negative properties such as curing disorders, uneven vulcanization, chemical shrinkage, and brittleness are possible. Vocal fold models with high silicone oil contents were investigated with regard to aging and reproducibility, and it was found that there is a high variability in the properties of different models and a change in their properties over time¹¹.

When producing vocal fold models in the conventional way^7,12, the stickiness of the epithelial layer can be a problem as it can affect the homogeneity of vibration and lead to rupture of the epithelium. Although the silicone used to make the epithelium is undiluted, it can be assumed that the oil that leaks from the neighboring mucosa layer has similar effects on the silicone as if it had been diluted. The problem of stickiness was addressed by adding various powders such as talcum or carbon powder as an intermediate layer between the mucosa and the epithelial layer¹². This approach may have been successful because the oil was partially absorbed by the powder and, as a result, the stickiness of the epithelial surface could be reduced.

In this publication, we show that the problem of stickiness can be circumvented by a slight modification of the process of vocal fold manufacturing. By changing the order of layering and starting with the undiluted epithelial silicone (so-called closed silicone), non-sticky super-soft vocal fold models can be produced. This change involves unusual types of molds and methods that are best presented and explained in the form of a video. In this paper, we describe our manufacturing process in detail and demonstrate how the properties of the vocal fold models can be characterized in an application.

Protocol

1. Design of the vocal fold models and 3D printing of parts Create a multi-layered representation of the common M5 geometry of silicone vocal folds using various soft silicone materials. Design the individual parts using computer-aided design (CAD) software. Check the Supplementary Coding File 1, Supplementary Coding File 2, Supplementary Coding File 3, Supplementary Coding File 4, Supplementary Coding File 5, Supplementary Coding File 6, Supplementary Coding File 7, Supplementary Cod…

Representative Results

The fabricated vocal fold model was integrated into the measurement setup depicted in Supplementary Figure 3 at the vocal folds position. The setup, extensively detailed in a previous publication13, comprises a multi-stage controllable airflow source that stimulates the vocal fold models into oscillation, along with an array of measuring instruments that record data such as sound pressure, static pressure at specific positions, and volume velocity. For the measurements, the airflo…

Discussion

The manufacturing process presented here involves critical steps that significantly impact its success. Firstly, it should be noted that the presented manufacturing process does not solve the problem of oil saturation in the vocal fold body material but rather circumvents certain negative side effects. The outgassing and the associated shrinkage and surface waviness still persist, albeit to a lesser extent. A solution to these problems would involve the use of an ultra-soft silicone or alternative material that combines …

Materials

3D Printer	ULTIMAKER		Type S5
3D Printing software	ULTIMAKER CURA		Version 5.2.2
CAD Software	Autodesk Inventor		Version 2023
High Speed Camera	XIMEA GmbH		MQ013CG-ON
PLA+ 3D Printer Material	eSun	none	white
Primary silicone	KauPo Plankenhorn	09301-005-000041	EcoFlex 00-30
Release Agent	KauPo Plankenhorn	09291-006-000001	UTS Universal
Secondary silicone	KauPo Plankenhorn	09301-005-000181	DragonSkin NV10
Silicone Thinner	KauPo Plankenhorn	09301-010-000002
Tougth PLA 3D Printer Material	BASF		black