Mechanical Properties and Accuracy of Additively Manufactured Silicone Soft Tissue Materials

Abstract

The objective of this study was to measure and compare the mechanical properties of conventional and three additively manufactured soft tissue silicone materials, while evaluating the precision of additively manufactured (AMed) materials through different printing angles. Three additively manufactured soft tissue silicone materials were used, in addition to one conventional self-curing injectable silicone material as a control. AMed materials were divided into three groups with three build angles. Mechanical testing was conducted for tensile and compressive strength by a universal testing machine and Shore A hardness by a durometer. Accuracy analysis of additively manufactured materials (n = 20/group) was performed following superimposition and root mean square (RMS) calculation. Statistical differences between the groups were assessed with a one-way analysis of variance (ANOVA) and Tukey’s post hoc test at a significance level of p < 0.05. Scanning Electron Microscopy (SEM) analysis was performed for fracture surface analyses. The tensile strength of all additively manufactured silicone soft tissue materials was significantly lower (p < 0.0001) than that of the control material. All additively manufactured soft tissue material groups had significantly higher compressive strengths (p < 0.0001) and Shore A hardness values. Accuracy analysis showed no significant difference between the groups when compared at the same printing angle (0°, 45°, and 90°); however, within each material group, printing at 45° had higher RMS values than specimens printed at an angle of 0° and 90°. The conventional soft tissue material (control) had a significantly higher tensile strength than all the AMed soft tissue materials, whereas the opposite trend was found for flexural strength and shore hardness. When selecting an AMed material for soft tissue casts used during implant restoration fabrication, it is recommended to print the soft tissues at either 0° or 90°.

1. Introduction

Dental implants are a successful treatment modality for replacing missing teeth. The success of an implant restoration is marked by high esthetics and osseointegration [1,2,3]. The esthetics and biological harmony of an implant restoration are influenced by proper implant placement, quality of design, and fabrication [4]. A key gateway to esthetics and a good fit is through the preservation of gingival architecture in the mouth. This is achieved by accurate contouring of the region from the bone crest to the free gingival margin, also known as the emergence profile, as well as by proper communication with laboratory technicians about the gingival contour [5,6]. However, a risk exists when restorations are over- or under-contoured, without soft tissue as a point of reference. Errors of this type not only affect the restoration’s esthetics but also affect all other contours of the restoration [7]. There are several concerns regarding non-fitting contours. Over-contouring results in caries, gingival inflammation, and/or gingival hyperplasia and facilitates the accumulation of food and plaque in cervical regions, which are difficult to access and clean [8]. Moreover, an over-contoured restoration causes pressure on the surrounding gingiva, leading to blanching and discomfort in the short term and apical bone and gingiva migration in the long term [7]. In contrast, an under-contoured restoration can create problems with phonetics and esthetics as well as excessive interproximal spaces [8].

Dental implants differ from natural teeth in size, shape, and attachment to bone and soft tissue, which presents a significant challenge for quality production [9]. More specifically, at the gingival level, implants do not accurately replicate the cross-section of natural teeth in shape or size, typically having a significantly lower cervical width of the restoration. Overcoming this hurdle exists in the usage of a soft tissue cast, which gives a clear vision of the gingival margin and thickness of the gingiva and provides a medium for accurate contouring of the emergence profile [5,10], as shown in Figure 1. A trend in implant dentistry is the expanding usage of digital technology. The advancements in treatment planning software and computer-aided design (CAD) and computer-assisted manufacturing (CAM) technology have simplified and improved the workflow of digital implant treatment and fixed prosthodontics [11]. The fabrication of a soft tissue cast using the conventional method significantly differs from additive manufacturing, also known as the 3D-printing method. It is crucial that materials used for soft tissue casts have the following properties: dimensional stability for the duration of the prosthesis fabrication period; ease of fabrication as well as durability during use by having adequate flow, tensile tear strength, and resilience; compatibility and nonadherence to the impression material, and ability to be removed from the underlying stone; and to be shaped and modified if necessary [5].

Different variables may impact additively manufactured products, for example, printer settings, material composition, build orientation, and post-processing. These variables influence the material’s mechanical properties [12]. Recently, multiple additive manufacturing liquid resins for soft tissue have been made available. However, there is a lack of information available on the properties of these additive manufacturing resins used for soft tissue materials. Research on this can provide useful selection and manufacturing guidelines for dental practitioners to select the most suitable and accurate additively manufactured soft tissue gum materials for fixed implant restorations. Therefore, the objective of this research was to measure and compare the mechanical properties (tensile strength, compressive strength, and shore hardness) of conventional and three different additively manufactured soft tissue gum materials. Additionally, this study aimed to compare the accuracy of additively manufactured soft tissue gum materials manufactured under different printing parameters (print angle and presence of supports). The null hypotheses H0 were as follows:

• There is no statistically significant difference in the mechanical properties of conventional and different additively manufactured soft tissue materials in terms of their compressive strength, tensile strength, and Shore A hardness.
• There are no statistically significant differences in the accuracy of different additively manufactured soft tissue materials manufactured under different printing parameters.

2. Materials and Methods

2.1. Specimen Fabrication

The materials used in this study are listed in

Table 1. Silicone materials included in the study.

Material	Name	Manufacturer	Composition
Conventional injection silicone material	Gingifast	Zhermack	≥80% to <90% vinylpolysiloxane
Additively manufactured silicone material	DentaGUM	Asiga	7,7,9 (or 7,9,9)-trimethyly-4,13-dioxo-3,14-dioxa-5,12-diazahexadecane-1,16-diyl bi,methacrylate: 10–25% Tetrahydrofurfuryl methacrykate: 10–20% Diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide: 10–20%
Additively manufactured silicone material	KeyMask	KeyMask Industries	Urethane acrylate oligmomer: ≥25 to ≤50% Alphatic Urethane Acrylate Oligomer: ≤10% Proprietary ingredient #1: >1% Proprietary ingredient #2: ≤1%
Additively manufactured silicone material	Dima Print Gingiva Mask	Kulzer	Poly(oxy-1,4-butanediyl), alpha.-hydro-omega.-hydroxy-, polymer with 5-isocyanto-1-(isocyanatomethyl)-1,3,3-trimethylcyclohexane, 2-hydroxyethyl acrylate-blocked: ≥25 to ≤50% Phenoxy polyethylene glycol acrylate: ≥25 to ≤50% Ethoxylated o-phenylphenol acrylate: ≥25 to ≤50% Diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide: ≥1 to ≤2.5%

. This study included four test groups: three currently available additively manufactured resins for soft tissue materials printed using three types of build angles (0°; flat to the build plate, 45°, and 90°; perpendicular to the build plate) and one conventional self-curing injectable soft tissue material was included in the study as a control for tensile, compression, Shore A hardness tests and accuracy. The three print build angles were selected in the study as they are the most used in dentistry [13]. The additively manufactured soft tissue material specimens for tensile, compression, Shore A hardness testing, and accuracy analysis were designed via AutoCAD (Autodesk, San Francisco, CA, USA) to generate STL files that were then printed using an Asiga 3D printer (Asiga, Alexandria, NSW, Australia). The printing and post-curing processes were conducted as per the manufacturer’s instructions. The printing layer thickness for all materials and groups was set at 50 µm, as this was the minimum thickness available for the Dima Print Gingiva Mask group.

Figure 2 shows the measurements of each specimen fabricated in this study. The geometry of the tensile specimens is Type 1a according to EN ISO 527-1:2019. Compression specimens are illustrated in Figure 2b. Shore A hardness specimens were designed to the dimensions shown in Figure 2c,d. The Shore A hardness specimens followed recommendations by Hoto Instruments that the minimum specimen dimensions must be 44 mm long, 19 mm wide, and 6 mm thick. We chose a square shape for our accuracy specimens, with the hole in the middle (Figure 2e) being of similar dimensions to that of an average implant. The overall size of the specimen was similar to the typical size of a soft tissue cast required for a molar. A sample size calculation for mechanical testing was performed based on a previously conducted experiment of similar nature and outcomes using the software G*power v3.0.10 (Heinrich-Heine Universität, Düsseldorf, Germany). The effect size (dz = 4.9706) and the required sample size were calculated for α = 0.05 and a power of 0.95 (1-β err prob), assuming a normal distribution. The calculation shows that a minimum of n = 20 per material group for tensile strength testing, n = 20 per material group for compressive strength testing, and n = 10 each per material group for Shore A hardness testing and accuracy analysis were required.

2.2. Tensile Strength Testing

Flat dumbbell-shaped specimens (n = 20 per group) were loaded in tension along their longitudinal axis in a universal testing machine (Instron 3369; Instron, Norwood, MA, USA), using a 50 N (±2) load cell at a constant speed of 1 mm/min with an extensometer (W-6280, Instron, Norwood, MA, USA) to record the strain until failure occurs. The specimen dimensions and testing were conducted following the previous testing protocol [14]. The maximum force (N), tensile stress (MPa), and strain (mm) were recorded (Figure 3A). Each specimen was marked 13 mm from each end indicating where the bottom of the clamp would sit. This ensured that specimens were clamped in at the same spot every time to allow for standardization.

2.3. Scanning Electron Microscopy (SEM) Analysis

One fractured specimen from each group (one control and one specimen from additively manufactured material with a build angle of 0°) was examined by scanning electron microscopy (SEM) at ×45, ×150, and ×2000 magnifications for analysis of the specimens. Samples were mounted on aluminum stubs with double-sided carbon tape and carbon paste and coated with approximately 15 nm of gold-palladium (Quorum Q150V Plus modular coating system, Quorum Technologies Limited, Laughton, UK). Samples were viewed and imaged in a Zeiss Sigma 300 VP FESEM (Carl Zeiss Inc., Oberkochen, BW, Germany) at an accelerating voltage of 5 KV.

2.4. Compressive Strength Testing

Compression testing to maximum compression of 4 mm displacement for each specimen was performed in an Instron universal testing machine (Instron 3369; Norwood, MA, USA), using a 500 N (±2) load cell at a crosshead speed of 1 mm/min (Figure 3B).

2.5. Shore A Hardness Testing

The hardness measurement of each specimen was calculated using a handheld digital Shore A durometer (Instrument Choice, Wingfield, SA, Australia), as shown in Figure 3C. Each specimen was marked 13 mm from the edge of the specimen and from any neighboring marks to indicate where the indenter would be used for testing. Ten hardness measurements (HA) were taken from each specimen as Shore units and the average values were calculated as the final Shore A hardness values.

2.6. Accuracy Analysis

To perform superimposition and obtain STL files, specimens were individually digitized by a single operator using a laboratory scanner (3Shape E4, 3Shape, Copenhagen, Denmark) while coated with a 3D laser scanning spray (AEsub Blue 3D-Scanning Spray, Magdeburg, ST, Germany). The STL files were then edited on Autodesk MeshMixer (Autodesk, San Francisco, CA, USA) to remove any debris present on a scan. The STL files of the scanned soft tissue gum material were then superimposed on the master STL file (the original file) that was used to print off the specimens using CloudCompare (Grenoble, France) to calculate root mean square (RMS) values (Figure 3D) using the following formula: $RMS=\sqrt{{\displaystyle \frac{1}{n}}\sum _i{x}_i^2}$ where n is the number of points that were measured in each specimen and $x_i$ is the value. Any deviations between the printed soft tissue gum material file and the control file were computed.

2.7. Statistical Analyses

The data sets were analyzed with statistical software (Prism 9, GraphPad, Boston, MA, USA). Descriptive statistics with mean and standard deviation for all tests and groups were computed. The statistical differences between the groups were assessed with a one-way analysis of variance (ANOVA) and Tukey’s test. All tests were performed at a significance level of p < 0.05.

3. Results

The summary of mean tensile strength for all the soft tissue materials studied is presented in

Table 2. Mean tensile strength (MPa ± SD), compressive strength (MPa ± SD), hardness (HA ± SD), accuracy of conventionally made material, and three different additively manufactured materials. (Note: N/A = Not applicable).

Method	Materials	Build Angle	Tensile Strength (MPa ± SD)	Compressive Strength (MPa ± SD)	Hardness (HA ± SD)	Accuracy
Conventional	Gingifast	N/A	1.30 ± 0.19	5.35 ± 0.44	42.00 ± 0.41	N/A
Additively manufactured	DentaGum	0	0.57 ± 0.17	3.56 ± 0.37	47.30 ± 0.63	0.13 ± 0.20
		45	0.39 ± 0.13	5.29 ± 0.55	61.50 ± 0.81	0.08 ± 0.01
		90	0.46 ± 0.16	7.47 ± 1.33	71.97 ± 2.14	0.06 ± 0.01
	KeyMask	0	0.48 ± 0.18	8.53 ± 2.64	61.60 ± 0.46	0.06 ± 0.02
		45	0.31 ± 0.05	8.00 ± 2.46	57.75 ± 1.27	0.07 ± 0.01
		90	0.36 ± 0.08	10.86 ± 1.90	60.25 ± 0.92	0.06 ± 0.01
	Dima Gingiva	0	0.63 ± 0.11	8.54 ± 1.14	57.55 ± 0.60	0.06 ± 0.02
		45	0.30 ± 0.09	6.68 ± 0.37	59.85 ± 1.11	0.08 ± 0.02
		90	0.41 ± 0.1	9.12 ± 0.33	61.8 ± 0.82	0.06 ± 0.02

and Figure 4. The tensile strength of all additively manufactured soft tissue materials regardless of the build angle was significantly lower (p < 0.0001) than that of the control material (Gingifast) (1.30 ± 0.19 MPa). Between the three additively manufactured materials printed at 0°, it was found that KeyMask had a statistically significantly lower tensile strength than Dima Print Gingiva Mask (p < 0.0134) with no statistically significant difference between DentaGUM and KeyMask, nor between DentaGUM and Dima Print Gingiva Mask. Materials that were printed with build angles of 45° and 90° had no statistically significant difference between them (p = 1.000). Within its groups, it was found that DentaGUM had a statistically significant higher tensile strength when printed at a 0° build angle (p < 0.001) (0.57 ± 0.17 MPa) compared to when printed at a 45° build angle (0.39 ± 0.13 MPa) and had a statistically insignificant (p = 1.000) lower tensile strength values at a 90° build angle. A similar trend was observed in Dima Gingiva (p < 0.001) with tensile strength values (0.63 ± 0.11 MPa) and (0.30 ± 0.09 MPa) for 0° and 45°, respectively, and again a statistically insignificant (p = 1.000) lower value when printed at a 90° angle. When KeyMask material was printed at 0°, it produced the highest tensile strength (0.48 ± 0.18 MPa) compared to the other printing angles (p < 0.0001). SEM analysis was performed only on the control specimen and additively manufactured group materials with a build angle of 0° as a reference (Figure 5). SEM images of the fracture surfaces showed cracks initiating that were formed from porosity on the outer surface of the control material (Figure 5a). The fracture surface surrounding the large crack was not smooth showing many micro-cracks, suggesting the material’s higher resistance to fracture, and explaining the control’s significantly higher tensile strength compared to the additively manufactured materials. Three-dimension-printed materials (Figure 5b–d) showed larger cracks propagating through the fracture surface by having a smoother appearance surrounding the large cracks compared to the control group, suggesting that there was less resistance to fractures and explaining the lower tensile strength.

Between the three additively manufactured soft tissue materials at ×2000 magnification, individual printing layers were visible in both DentaGum and KeyMask materials, whereas this was not present in the Dima Gingiva group and could explain its higher tensile strength. SEM images of the outer surface of the tensile specimens revealed that only the control specimen had visible surface porosity. Control and DentaGum groups had cracks on their surface, which could be seen at ×2000 magnification. Polymerized polymer chains were observed on Dima Gingiva and KeyMask outer surfaces.

The summary of mean compressive strength for all the soft tissue materials studied is presented in Table 2 and Figure 6. The compressive strength of the control group was 5.35 ± 0.44 MPa, and all the additively manufactured materials had significantly higher compressive strengths regardless of the build angle (p < 0.0001). When comparing within its groups, DentaGum had a statistically significant difference (p < 0.001) between all three build angles with the highest compressive strength of 7.47 ± 1.33 MPa when printed at a build angle of 90°. KeyMask material had no statistical significance between build angles 0° and 45°, but it did have a statistically significant result (p < 0.0001) at a build angle of 90° with the highest value of 10.86 ± 1.90 MPa for this group. Similarly, Dima Gingiva printed at 90° exhibited the highest results (9.12 ± 0.33 MPa) and was statistically significant to a build angle of 90° and 45° (p < 0.0001) but not to 0° (p = 1.000). When compression strength is compared between additively manufactured materials within same build angle, it was found that DentaGUM had a statistically significant lower compressive strength (p < 0.0001) to both KeyMask and Dima Print Gingiva Mask, while KeyMask and Dima Print Gingiva Mask had no statistical differences between each other (p > 0.99) when printed at a build angle 0° and 45°, but they did have a statistically significant difference when printed with a build angle of 90° with KeyMask being significantly higher (p < 0.05).

The summary of mean Shore A hardness for all the soft tissue materials studied is presented in Table 2 and Figure 7. The Shore A hardness of the control specimens was the lowest with a mean of 42.00 ± 0.41. The Shore A hardness of all the additively manufactured soft tissue materials was significantly higher than the control group (p < 0.0001). Between the three additively manufactured soft tissue materials, it was found that a statistically significant difference was present within groups and between materials (p < 0.0001), except for one KeyMask group where no statistical significance was found between specimens that were printed with build angles of 0° and 90°. DentaGum material when printed at 90° had the highest Shore A hardness measured (71.97 ± 2.14) among all groups.

The summary of mean RMS for all the soft tissue materials studied is presented in Table 2 and Figure 8. When comparing the RMS values of the additively manufactured soft tissue materials at build angles of 0°, 45°, and 90°, no significant difference was found between the three groups. Within the DentaGum group, specimens printed at a build angle of 45° were found to have a significantly higher mean RMS than those printed at 0° (p < 0.0342) and 90° (p < 0.0008). In the KeyMask group, specimens printed at 45° were shown to have a slightly higher mean RMS than those printed at 0° and 90°, but no statistically significant difference was observed. In the Dima Gingiva group, specimens printed at 45° were shown to have a higher mean RMS than those printed at both 0° and 90°. However, it was only significantly (p < 0.0287) higher than those printed at 0°.

4. Discussion

The aim of this study was to measure and compare the mechanical properties (tensile strength, compressive strength, shore A hardness, and printing accuracy) of conventional and different additively manufactured soft tissue materials manufactured under different print angles. The null hypotheses were rejected since a statistically significant difference between the control and the different additively manufactured soft tissue materials was found. There was also a statistically significant difference in the printing accuracy of the additively manufactured soft tissue materials at different print angles in both the DentaGum and Dima Gingiva groups.

The additively manufactured soft tissue materials, regardless of their build angle in comparison to the conventionally manufactured (control) material, were found to be statistically significantly lower in tensile strength. In a previous study, the tensile strength of Thiel-embalmed human gingival tissues were 3.81 ± 0.94 MPa, and buccal mucosa 1.54 ± 0.52 MPa [14]. The current findings suggest that the tensile strength of additively manufactured materials was not as high as that of the Thiel-embalmed specimens and was slightly lower than that of buccal mucosa. Nonetheless, it should be noted that previous research has only been conducted on two Thiel-embalmed cadavers, and that gingival tissue can exhibit variations in tensile strength due to factors such as age and overall health status. Therefore, observing additively manufactured soft tissue material with a tensile strength most like the conventional material, it is recommended that they use Dima Gingiva using a build angle of 0° as it had the highest result among all additively manufactured materials.

The compressive strength of a soft tissue material is a crucial mechanical property, as the shape and contour of an implant restoration are highly dependent on the soft tissue cast [15]. If the material is too susceptible to compression, it may lead to failure under a smaller load than desired. During testing, the only material in this study that failed under the compressive load was KeyMask, but it also had a significantly higher compressive strength than the other three materials (p < 0.0001). An easily compressed soft tissue material may also result in an over-contoured restoration, which may apply too much pressure on the patient’s gingiva once placed in the mouth, potentially causing short-term blanching and discomfort, which may be followed by apical bone and gingiva migration in the long term [7].

Despite the importance of this property, there are only two follow-up studies that have directly investigated the compressive stress of tissue [16,17]. These studies were found to be quite limited due to the lack of standardization and inconsistencies in data reporting. This is largely because pressure levels can only be measured until the patient experiences pain, and this threshold varies among individuals. In a study conducted by Goktas et al. (2011), the peak compressive stress of the buccal-attached gingiva of pigs was 1.17 ± 0.17 MPa, and they found that this was similar to the peak compressive strength of the lingual-attached gingiva with no significant difference between the two [15]. The results from the current study showed that none of the additively manufactured materials had similar compressive strengths to pig gingiva, with the material having the closest compressive strength found to be DentaGum with a build angle of 0° (3.56 ± 0.37 MPa).

The Shore hardness of a material indicates its ability to resist deformation at the surface under compression [18]. This property is important in soft tissue casts as it is unfavorable for the gingival topography to deform under compression when an implant restoration is fitted onto the model. The gingival topography plays a large role in the esthetics and function of a restoration, therefore making a soft tissue material that deforms too easily during compression unfavorable as it may result in an over-contoured implant restoration [6]. All the additively manufactured soft tissue materials had significantly higher shore A hardness values than the control group. Indicating that they are better than the control group at resisting deformation at the surface under compression, making them more desirable material for a soft tissue cast.

The accuracy of the additively manufactured soft tissue materials was expressed in RMS values. RMS values show the mean and standard deviation of the distribution of distances calculated between two point clouds. While root mean square (RMS) values are commonly used to evaluate accuracy, they may not provide clinically significant information about the direction and amount of deviation [19], which is the limitation of this study that future studies could investigate. However, the overall accuracy of soft tissue material is still important to ensure that the most accurate gingival topography is used for the fabrication of the implant restoration. If a gingival mask is printed with large inaccuracies, it could affect how it sits on the additively manufactured or stone casts. Soft tissue casts that do not sit flush will be lifted slightly off the model, which will affect the contours and esthetics of the final restoration. In terms of the accuracy of the additively manufactured materials, no material was significantly more accurate than the others, indicating that if choosing a material based on accuracy, any of the three additively manufactured materials can be used. However, by printing at a 45° build angle, we found that each material group had a higher RMS with a significantly higher difference in the DentaGum and Dima Gingiva groups. Therefore, it is recommended that the additively manufactured soft tissues be printed at either 0° or 90° rather than at 45° build angles.

Future studies could focus on investigating the effect of specimen layout on the build plate during printing, as Tian et al. (2019) suggested that samples printed on the edge exhibited superior performance in both tension and compression [12]. Additionally, research could explore the storage of printed materials and determine an appropriate time range for their use. This could aid in identifying the optimal conditions for printing and storing soft tissue materials to ensure consistent and reliable results in clinical settings.

5. Conclusions

Within the limitations of this study, the following can be concluded:

• The conventional injected soft tissue material had a significantly higher tensile strength than all the additively manufactured soft tissue materials. All the additively manufactured soft tissue materials within all build angles studied showed a statistically significantly lower tensile strength.
• A statistically significantly higher compressive strength was found in all the additively manufactured soft tissue materials compared to the control group. Similar to their compressive strengths, the additively manufactured soft tissue materials all had significantly higher Shore A hardness values than the control group.
• Additively manufactured soft tissue materials printed at 45° were found to have significantly higher mean RMS values (less accuracy) than specimens printed at 0° and 90° for both DentaGum and Dima Gingiva.