Effect of printing orientation on mechanical properties of 3D-printed orthodontic aligners

Since the introduction of three-dimensional (3D) printing technologies in orthodontics, direct fabrication of orthodontic aligners has emerged as a promising alternative and has gathered the interest of clinicians and researchers alike. This highly efficient do-it-yourself method offers considerable advantages, such as same day availability of appliances within the orthodontic office at a lower cost by circumventing the dental laboratory, while giving direct control to the orthodontist, better geometrical accuracy, and improved efficacy [11, 29]. In contrast to the indirect technique, where aligners are thermoformed on models, directly 3D-printed aligners are formed layer by layer based on digital 3D models. The majority of available 3D printers utilize a light source capable of polymerizing resins in vertical, diagonal, or horizontal directions [19]. The process is quickest per aligner for the horizontal orientation, due to the smaller number of layers added by the 3D printer and the reduced lines to the aligner’s anterior teeth labial surfaces [18]. On the other hand, vertical printing favors a larger appliance volume production per single print job. Printing supports are also distributed differently according to the chosen orientation of the aligner during printing. It seems, however, that independent of the type of 3D printer used, 3D printing procedures using either horizontal or vertical orientation seem to be accurate and satisfy patient demands.

However, additive manufacturing techniques have the inherent disadvantage of potentially incorporating stresses between the material layers, which is attributable to shrinkage of the material after layering [4, 14, 31]. The direction of aligner printing may therefore influence the magnitude and orientation of these residual stresses. This phenomenon is a well-known side effect found in the 3D printing of metallic components [27] and is, at least to some degree, addressed by postthermal treatment [4]. Unfortunately, this treatment can be used on neither thermoformed aligners nor on resin-printed ones. The introduction of such internal stresses may lead to alterations of their major mechanical properties including hardness, elastic relaxation, and deformation. These changes are not to be taken lightly, since the clinical efficiency of the fabricated orthodontic appliances is largely dependent on them. In particular, the modulus of elasticity [9, 32, 33] and elastic relaxation [6, 10, 17] are crucial for the aligners to be able to exert light and continuous forces on the teeth, while hardness is related to the aligner’s resistance to abrasion [33]. Furthermore, the printing direction may have an adverse effect on surface roughness, which promotes appliance plaque accumulation and discoloration [13, 15] and might ultimately influence the, sometimes questionable [9, 21], clinical performance of aligners.

There is a lack of scientific evidence in the published literature regarding the mechanical properties of orthodontic aligners printed with different orientations. Thus, the current choice for 3D printing of such aligners is based mostly on efficiency aspects in terms of speed and mass production, and the actual efficacy of the end product in use is disregarded. Therefore, the aim of this study was to investigate the effect of printing orientation on the material mechanical properties of 3D manufactured orthodontic aligners. The null hypothesis was that there is no difference in the mechanical properties of orthodontic aligners directly printed with a 3D process in either a horizontal or a vertical orientation.

Representative stress–strain curves from both groups are presented in Fig. 2. In all instances, stresses increased steadily up to the yield point (point 1) and a local decrease in the specimen’s cross-sectional area started. After necking, the nominal stress started falling at a constant value as the neck extended along the full gauge length of the dumbbell and the polymer chains tended to align themselves parallel to the direction of tensile stress (point 2). Afterwards, the neck started to spread to the full gauge length (point 3) and, thanks to orientation of molecular chains, was stronger/stiffer and the stresses started to rise again up to final fracture (point 4). Several mechanical properties were measured from the tensile testing of the included dumbbells, including yield strength (mean 19.2 MPa; SD 1.7 MPa), breaking strength (mean 19.6 MPa; SD 1.2 MPa), and plastic strain (mean 0.77%; SD 0.11%), with no statistically significant differences between the horizontal and vertical group (P > 0.05 in all instances; Table 1).

Force-indentation depth curves for both groups are illustrated in Fig. 3a. Several mechanical properties were measured from IIT of the included aligners, including Martens hardness (median 89.0 N/mm²; IQR 84.5–90.0 N/mm²), indentation modulus (median 2670.5 MPa; IQR 2645.0–2726.0 MPa), elastic index (median 27.5%; IQR 25.9–28.1%), and indentation relaxation (mean 65.1%; SD 3.5%; Table 1), with no statistically significant differences between the horizontal and vertical group (P > 0.05 in all instances; Table 1). Finally, Fig. 3b presents the rectangular pulse applied with a period of standard indentation depth, while Fig. 3c depicts the degradation of the applied force over the experimental time.

The results of this in vitro study do not provide adequate justification to reject the null hypothesis of no difference between horizontal and vertical printing orientation for orthodontic aligners. This study employed both tensile testing and IIT to provide a full characterization of the wide spectrum of possible effect of printing orientation on total volume (tensile testing) and micro scale (IIT) of the material tested. Contrary to tensile testing, where bulky specimens are required, IIT can provide information on the mechanical properties of small and irregular specimens such as aligners [33].

No statistically significant differences were identified for YS, BS, and ε (Table 1) between the horizontal and the vertical group, implying that the printing orientation has no effect on tensile properties of the printed aligners. All stress–strain curves (Fig. 2) presented the characteristic shape of a polymeric tensile curve with four characteristic points: (1) yield point and initiation of local decrease in cross-sectional area (neck); (2) after necking, the nominal stress falls at a constant value as the neck extends along the full gauge length of dumbbell. This process is known as cold drawing and the polymer chains tend to align themselves parallel to the direction of tensile stress, resulting in a strain-hardening mechanism; (3) neck has been spread to the full gauge length, becomes stronger/stiffer as the molecular chains are fully aligned, and stresses start to develop; and (4) rise again up to final fracture (4) [26].

The average modulus of elasticity for both aligner groups (median of 2670.5 MPa) was similar to previous studies, in which resin-made 3D-printed aligners were fabricated by different printers and under different conditions [3, 33]. The same was true when the results are compared to the polyurethane-made Invisalign® (Align Technology, Santa Clara, CA, USA) aligners [8, 20]. Conversely, the modulus of elasticity measured in the present study was higher than that reported from thermoformed aligners made of polyethylene terephthalate glycol (PET‑G; 1530 to 2370 MPa [1, 5, 28]). The increased resistance to elastic deformation of resins is associated with a higher magnitude of forces transmitted to the teeth upon aligner use. From a clinical point of view this means that 3D-printed aligners can be thinner, and therefore more patient-tolerated than their thermoformed counterparts, while being equally effective under the same strain [1, 12]. Concerning HM, the 3D-printed resin aligners displayed similar hardness (median of 89 N/mm²) to previously reported PET‑G aligners (92–101 N/mm²) [1] and resin aligners (94–108 N/mm²) [3, 33]. However, the material of Invisalign® aligners seems to be harder (118–122 N/mm²) [1, 8, 20] and could consequently be more resistant to intraoral wear. On the other hand, the small mean values of the 3D-printed resins’ elastic index (27.5%) emphasize their ductile nature [3], which is in accordance with the tensile results where a 0.75 (75%) plastic strain was found after fracture (Table 1). In contrast to the values of Invisalign® aligners (40%) [8, 20] and PET‑G (35%) aligners [1], 3D-printed aligners are less brittle and therefore less prone to fracture upon removal. In addition, 3D-printed resins showed a lower resistance to relaxation, given the high values of the relaxation index (65.1%) compared to the low values of Invisalign® (~ 4%) [20]. This is associated with a more dramatic force decrease under constant deformation for 3D printed than for Invisalign® aligners, agreeing with previous data [3], and which may be initially seen as problematic from an orthodontic treatment perspective. However, the high magnitude of initial forces applied to the teeth, due to the resin’s increased modulus of elasticity, needs to decrease rapidly to the ideal light and constant orthodontic forces, which are essential for efficient tooth movement [32]. Hence, an increased relaxation index might ultimately prove to be beneficial in this respect—but this requires further testing under in vivo conditions.

The results of both tensile and IIT testing agreed that printing orientation had no effect on the mechanical properties of 3D-printed aligners. This is in contrast with older studies reporting that printing orientation using stereolithography (SLA) had a significant effect on the mechanical properties of 3D-printed resin dumbbell specimens [16, 24, 25] and therefore appliances printed with SLA cannot be considered isotropic. On the other hand, data from a more recent study indicated that SLA printing with different orientations did not significantly influence the mechanical properties of printed resin dumbbell specimens [27]. It seems that the postcure process and the specific conditions during the aligner’s manufacturing may alleviate any residual stresses in the manufactured aligner and thereby affect the mechanical properties and clinical performance of the aligner. Unlike metallic 3D-printed components, which are characterized by high anisotropy in all axes [7] and where the printing orientation dictates the spatial orientation of the printed material, the resin material is not fully set anisotropic after printing. Thus, enough time is given for the polymeric chains to spatially rearrange, counteracting orientation differences, while simultaneously polymerization proceeds in multiple directions and more importantly in depth without the presence of oxygen inhibition phenomena—the latter augmented by the highly favorable nitrogen gas environment [22]. Therefore, it seems that any spatial orientation differences of the printed material are negated by the postcuring process, which was similar for the two groups. Nevertheless, future research that includes structural and chemical characterization of 3D-printed components under various conditions and configurations (including different postcuring processes) may shed further light on this.

In conclusion, the 3D-printed resin aligners tested showed similar mechanical properties to the material of Invisalign® aligners and thermoplastic materials and, thus, should be considered as a promising candidate for orthodontic tooth movements. However, there is still much room for additional investigations on biocompatibility issues and aging effects, which will surely pave the road for the optimization of these recently introduced materials.

The results of the present in vitro study indicate that printing orientation (horizontal or vertical) has no significant effect on the mechanical properties of three-dimensional (3D)-printed aligners from resin. The clinical implications of these results are that clinicians might consider 3D-printed aligners to be an isotropic material and, thus, a similar mechanical reaction might be anticipated intraorally under multidimensional activation.