Effects of storage and toothbrush simulation on Martens hardness of CAD/CAM, hand-cast, thermoforming, and 3D-printed splint materials

Objectives To investigate Martens hardness parameters of splint materials after storage in liquids and toothbrush simulation. Materials and methods Ten specimens per material and group were fabricated (hand-cast CAST, thermoformed TF, CAD/CAM-milled CAM, 3D-printed PS, PL, PK, PV), stored in air, water, coffee, red wine, and cleaning tablets and investigated after fabrication, 24 h, 2- and 4-week storage or toothbrushing. Martens hardness (HM), indentation hardness (HIT), indentation modulus (EIT), the elastic part of indentation work (ηIT), and indentation creep (CIT) were calculated (ISO 14577-1). Statistics: ANOVA, Bonferroni post hoc test, between-subjects effects, Pearson correlation (α = 0.05). Results HM varied between 30.8 N/mm2 for PS (water 4 weeks) and 164.0 N/mm2 for CAM (toothbrush). HIT values between 34.9 N/mm2 for PS (water 4 weeks) and 238.9 N/mm2 for CAM (toothbrush) were found. EIT varied between 4.3 kN/mm2 for CAM (toothbrush) and 1.8 kN/mm2 for PK (water 2 weeks). ηIT was found to vary between 16.9% for PS (water 4 weeks) and 42.8% for PL (toothbrush). CIT varied between 2.5% for PL (toothbrush) and 11.4% for PS (water 4 weeks). The highest impact was identified for the material (p ≤ 0.001). Conclusions Storage and toothbrushing influenced Martens parameters. The properties of splints can be influenced by the choice of materials, based on different elastic and viscoelastic parameters. High HM and EIT and low CIT might be beneficial for splint applications. Clinical relevance Martens parameters HM, EIT, and CIT might help to evaluate clinically relevant splint properties such as hardness, elasticity, and creep.


Introduction
Temporomandibular disorders (TMD) can be effectively treated with splints [1].These appliances reduce symptoms including pain and functional limitations [2][3][4].Splints can be made on gypsum models either by applying methacrylate in the sprinkle-on technique or by vacuum thermoforming [5].The digitization of the clinical situation-either directly with an intraoral scanner or the extraoral scan of an impression/ plaster model-enables the use of computer-aided design/ computer-aided manufacturing (CAD/CAM) [6,7].Based on the digitalized clinical situation, the occlusal devices can be designed with a CAD software [8].CAM manufacturing can then be carried out using subtractive or additive processes.Subtractively machined splints are milled from a prefabricated resin-based blank using a computerized numerical control (CNC) machine [9].Additive processes include 3D printing with stereolithography (SLA) or digital light processing (DLP) technology [10,11], in which the splints are built up and cured layer by layer using a liquid photopolymer.Here, the mechanical properties and performance [12] are influenced by both the type of material and the processing.But cleaning and post-polymerization also play an important role in ensuring the material's properties [13,14].In addition to aesthetic [15][16][17], phonetic, and functional [18] requirements, splints should also meet mechanical requirements such as sufficient and long-term hardness.Stiffness, elastic relaxation, or hardness can influence treatment efficiency and may be associated with iatrogenic implications on the patient's health [19].Further on, energy dissipation capabilities and elastic and viscoelastic properties might be advantageous by the utilization of splints.
Hardness is generally defined as the resistance against plastic and permanent deformation measured by methods like Vickers or Brinell hardness testing [20].Instrumented indentation testing, so called Martens hardness (HM; ISO 14577-1), seems to be a suitable alternative for evaluating the surface hardness combined with information of the elastic and viscoelastic behavior of the splint materials.HM is derived from the applied force (F) divided by the indentation surface (As), which is a function of indentation depth (h).The constant measurement of force and indentation depth provides a force-indentation depth curve, which allows further interpretation.The indentation modulus (E IT ) is related to the modulus of elasticity [20].The elastic part of indentation work is expressed by η IT .The time-dependent response to the indentation of a viscoelastic material [21] can be expressed as the indentation creep (C IT ), showing the relative plastic character of a material, namely, the increase of strain under constant force application.C IT might therefore help to estimate the long-term dimensional and mechanical stability of a material [22][23][24].
The aim of the study was to assess how storage in various liquids (water, coffee, red wine, and denture cleaner solution) and toothbrush simulation affect Martens parameters of different splint materials.As a result of enduring contact with staining solutions and toothbrush simulation, Martens hardness (HM), indentation hardness (H IT ), indentation modulus (E IT ), the elastic part of indentation work (η IT ), and indentation creep (C IT ) are likely to change.The null hypothesis was that these changes would be dependent on the material, fabrication, type of storage, and the duration.
Materials were investigated using instrumented indentation testing according to ISO 14577-1.Testing was carried out in a universal hardness-testing machine (ZwickiLine Z2.5, ZwickRoell, Ulm, Germany).The indentation depth was constantly monitored at a loading speed of 0.5 mm/ min to a maximum force of Fmax = 10 N, using a Vickers indenter and a dwell-time of 10 s.Unloading was performed at 0.1 mm/min.The recorded force-indentation depth curves were used to calculate Martens hardness (HM), indentation hardness (H IT ), indentation modulus (E IT ), the elastic part of indentation work (η IT ), and indentation creep (C IT ) as defined in ISO 14577-1.Poisson's ratio of the diamond indenter was set to ν i = 0.07 and for the resin-based composite materials to νs = 0.3.Young's modulus of the indenter was E i = 1140 GPa.
Calculations and statistical analysis were performed using SPSS 26.0 for Windows (IBM, Armonk, NY, USA).

Results
Martens hardness (HM, Fig. 2): HM varied between 30.8 N/mm 2 for PS (water 4 weeks) and 164.0 N/mm 2 for CAM (toothbrush).The highest impact in HM was identified for the material (p ≤ 0.001, η 2 = 0.809).TF and CAM show stable values of the different storage conditions and time, whereas CAST and printed systems provided decrease in HM with prolonged storage.
For TF, no influence on HM of any aging was observed (p ≥ 0.102).HM of CAM was significantly reduced only after red wine storage (p = 0.004, other p ≥ 0.165).CAST showed no change of HM (p ≥ 0.27) but for the storage in mouse rinse (p = 0.038).PL changed HM significantly (p ≤ 0.027, other p ≥ 0.196) due to storage in coffee, water, and red wine.HM of PK was significantly (p ≤ 0.026, other p ≥ 0.127) reduced by storage in coffee and mouth rinse.Storage in coffee, mouth rinse, and red wine also influenced PV significantly (p ≤ 0.012, other p ≥ 0.281.PS changed due to the influence of coffee, mouth rinse, red wine, and water (p ≤ 0.043), but not for the toothbrushing (p = 0.981).
Indentation hardness (H IT , Fig. 3): H IT ranged between 34.9 N/mm 2 for PS (water 4 weeks) and 238.9 N/mm 2 for CAM (toothbrush).The highest impact in H IT was identified for the material (p ≤ 0.001, η 2 = 0.825).TF and CAM showed stable values for the different storage conditions and Indentation modulus (E IT , Fig. 4): E IT varied between 4.3 kN/mm 2 for CAM (toothbrush) and 1.8 kN/mm 2 for PK (water 2 weeks).The highest impact in E IT was identified for the material (p ≤ 0.001, η 2 = 0.584).The materials showed only small changes with different storage conditions and time.
For TF, no influence on E IT of any aging was observed (p ≥ 0.080) but for toothbrushing (p = 0.008).E IT of CAM was not significantly reduced (p ≥ 0.054).CAST showed no changes of E IT (p ≤ 0.039) due to coffee and mouth rinse storage.E IT of PL was not influenced by aging (p ≥ 0.053) nor was E IT of PK (p ≥ 0.308).E IT of PV was significantly (p = 0.013) influenced by storage in red wine.PS changed due to the influence of coffee and water (p ≤ 0.011).
Elastic part of indentation work (η IT Fig. 5): η IT ranged between 16.9% for PS (water 4 weeks) and 42.8% for PL (toothbrush).The highest impact in E IT was detected for the material (p ≤ 0.001, η 2 = 0.659).The printed systems showed a decrease in η IT with prolonged aging.
For TF, no influence on η IT of aging was observed (p ≥ 0.123) but for mouth rinse (p = 0.010).η IT of CAM was not significantly reduced, but for toothbrushing (p = 0.022, other p ≥ 0.169).CAST showed no changes of η IT (p ≤ 0.089).η IT of PL was not influenced by coffee aging (p < 0.001, other p ≥ 0.069).PK showed no changes of η IT (p ≥ 0.160).η IT of PV was significantly (p ≤ 0.049) influenced by storage, but not by toothbrushing (p = 0459).PS changed due to the influence of coffee, mouth rinse, and water (p ≤ 0.004).
Indentation creep (C IT , Fig. 6): C IT ranged between 2.5% for PL (toothbrush) and 11.4% for PS (water 4 weeks).The highest impact in C IT was identified for the material (p ≤ 0.001, η 2 = 0.869).The values for TF, CAST, and CAM were stable over the storage period, while the printed systems showed an increase with increasing storage time.
For TF, no influence of any aging protocol on C IT was identified (p ≥ 0.074).H IT of CAM was significantly reduced only after coffee exposure (p < 0.001, other p ≥ 0.387).CAST showed no change of C IT (p ≥ 0.074) but for toothbrushing (p = 0.039).C IT of PL changed significantly (p ≤ 0.012, other p ≥ 0.128) due to storage in coffee, water, and red wine.C IT of PK was significantly (p < 0.001, other p ≥ 0.053) reduced by storage in coffee, water, and mouth rinse.PV and PS were significantly influenced by storage (p < 0.001) but not for toothbrushing p ≥ 0.530).

Discussion
The null hypothesis that the Martens parameters are dependent on the material, fabrication, type of storage, and the duration could be partly confirmed.
The examined materials showed clearly different hardness levels ranging between 30 and 165 N/mm 2 .Since Martens hardness is calculated from the course of the indentation depth during loading, it provides information not only about the plastic but also about the elastic material properties.However, e.g., Martens hardness and flexural strength of composite resins are not correlated [25].To relate the indentation hardness to conventional values, it can also be converted to Vickers hardness, but HV and HM do not necessarily correlate [21].The HM values identified in the current study were in the order of magnitude that has been measured earlier for other resin-based materials [15,26,27].Varying values have been reported, ranging between ~100 N/mm 2 [19,28] for aligner and ~500 N/mm 2 [27] for resin-based composites, strongly dependent on the type and composition of the materials.General variations might be attributed to different test setups [29].Differences between composites and resin based-splint materials might be attributed to a different filler content of the materials, which varies between 0% and 87%.The CAD/CAM materials, which can be filled to a higher extent due to the manufacturing technology, showed the highest HM values [30].Since the printed systems must have a lower viscosity for processing-and are therefore generally filled to a lower extent-their HM hardness is also 20-30% lower [31].Since the hardness of PL is higher, a higher filler content can be assumed for this material.An additional indication for the higher filler content might be the higher resistance against toothbrush abrasion, because the abrasion is also associated with a loss of resin matrix and filler particles (Valente et al. 2013; Lai et al. 2018).
It is not only the resin composition, but also the conversion, and therefore fabrication and post-processing that might influence hardness.Incomplete polymerization [32,33] and chemical reactivity could be reasons for the decrease of individual properties [34,35].Photo-polymerization influences the structure of dental resin matrices [36], and therefore, the degree of polymer polymerization may be a key to the decrease in HM.It is known that a combined heat-and light-post-curing unit can improve the degree of conversion of 3D-printed occlusal splints [37] and, e.g., the in vitro performance [13,35].Light-curing occlusal splint resins have comparable hardness as auto-polymerizing systems (Danesh et al. 2006;Więckiewicz et al. 2014), but the hardness of 3D-printed occlusal splint materials is also influenced by the print angle (Grymak et al. 2021).Against expectations [19,26,31], differences between printed materials do not seem to be not or only marginally influenced by manufacturing parameters such as cleaning or post-polymerization.However, the combination of material, processing, and finishing will affect the results.Moreover, the HM results could be influenced by the changed superficial roughness resulting from storage [17].
A decrease of HM was observed during all storage conditions.It was noticeable that only CAD and TF were not affected by storage.The decrease in HM is evident with longer storage time.Similar effects of long-term laydown could also be identified for other properties such as roughness, color, or gloss [17].This phenomenon will certainly have an impact on the long-term clinical application of the splints.Since the drop in hardness already occurs during water storage and is somewhat equal for all storage media, it can be expected that the materials absorb water.It can be assumed that water absorption reduces the hardness [38,39] and Martens hardness [26].A general influence of an individual stirring agent could not be confirmed.Only CAST showed a somewhat indifferent behavior, as with storage in red wine hardness even slightly increased.However, large variations were also observed for this system, which should be attributed to the manual production process and the resulting inhomogeneous structure of the specimens.The results showed that the highest HM stability can be achieved with both milled and printed materials.
The indentation hardness H IT is determined using the maximum force and applying tangents to the unloading curve and represents a measure of resistance to permanent and plastic deformation.Since H IT shows a similar behavior and ranking as HM, a high proportion of plastic deformation is to be assumed for all materials.The elastic proportion is to be classified as low, especially for parts of the printed materials [40].The highest stability under clinical bruxism loadings might be expected for CAM and astonishingly one printed system PL.Again, a higher filler content in comparison to the other print systems might be the reason for this behavior [30].But also storage conditions, such as pH in solution, appear to be related to the hydrophilicity of the matrix and the chemical composition of the filler, which in turn affect sorption and solubility (Örtengren et al. 2001).
The indentation modulus E IT is calculated from the indentation relief curve.The increase in the E IT values for PS and TF therefore indicates an embrittlement of these materials due to storage.The cast material showed an indifferent picture for E IT , probably again due to the influence of manual processing.On the other hand, it confirms the same quality of the other production processes: all other materials exhibited a more homogeneous behavior with smaller variations.The combination of material and storage seems to be decisive for the results in E IT .The elasticity, which is essential for the insertion and removal of the splint during application, therefore seems only minimal.The elasticity of splint specimens is expected to depend on the type of material, their cleaning and post-polymerization.Therefore, E IT is also expected to have an influence on the improvement of the dynamic load capacity of splints [12,13,27].
A further aspect for the flexibility of the splints is η IT .The elastic fraction of indentation work (η IT ) is calculated from the areas under loaded and unloaded parts of the load-relief curve.The plastic fraction Welast/Wtotal 100 % = η IT is calculated, which means a high η IT is associated with elastic properties.Surprisingly, the examined materials show a comparable η IT level.Only two of the printed materials have a lower η IT level, i.e., the elastic content is lower.Since the differences between the materials seem to become smaller with η IT , E IT seems to be more meaningful to evaluate elastic properties.Due to the aging process, the elastic portion is reduced in three of the four printed materials, i.e., the materials become visibly brittle.This phenomenon may reduce the retention force of the splint and increase the fracture risk during clinical application [41].More brittle materials may although provide sharp-edged fractures, which bear the risk for cutting [12].
The H IT results correlate with the C IT results, where, surprisingly, the investigated materials showed a comparable C IT level.C IT behaves in the opposite direction to η IT .Only for two printed materials a higher value could be identified.Since the creep behavior C IT describes the further deformation of the material under constant force, the indentation depth for these materials increases under load [40].Under clinical conditions with continuous load, such as bruxism, the deformations for these materials would therefore be higher.Occlusal deformation might correlate with a loss of contact situation and function.To determine C IT , the indenter is pressed into the specimen with a constant force over a longer period of time.Polymers with a tendency to creep continuously yield and the penetration depth increases.For three out of four printed materials examined, the value increases due to storage.An influence of the print parameters on the surface quality has been reported [42,43], and thus, specimens with a lower (e.g., 25 μm) or varying layer thickness (inside 50 μm and outside 25 μm) might show different Martens parameters.All other materials showed a relatively good creep behavior.
Clinically desirable would certainly be a splint material with a high hardness and high resilience, i.e., low creep.
Despite the importance of mechanical properties, there is no evidence that the significant differences in in vitro mechanical properties have implications for clinical therapy.For the comprehensive evaluation of the splint materials, the consideration of different Martens parameters seems important, since not all parameters correlate with each other.

Conclusion
Martens hardness parameters such as hardness (HM), indentation hardness (H IT ), indentation modulus (E IT ), elastic part of indentation work (η IT ), and indentation creep (C IT ) varied significantly between different splint materials.

Clinical consequence
The clinical behavior of dental splints might be influenced by the selection of materials that feature different elastic and viscoelastic parameters.Materials with high HM, E IT , and low C IT might be beneficial for clinical splint applications.

Table 1
Materials, fabrication, and composition