Impact of post adhesion on stress distribution: an in silico study

Abstract

Objectives

This study aimed to evaluate the stress distribution in teeth restored with different post materials and bonding conditions using finite element analysis (FEA).

Methods

A two-dimensional FEA model of a maxillary central incisor restored with IPS-Empress-2 crown (Ivoclar Vivadent), composite resin core, and posts were created. The model simulated bonded and non-bonded conditions for both fiber-reinforced composite (FRC) and titanium (Ti) posts. Stress distribution was analyzed using ANSYS 14.0 software under a 100-N load applied at a 45° angle to the long axis of the tooth.

Results

The results revealed that stress concentration was significantly higher in non-bonded posts compared to bonded ones. FRC posts exhibited stress values closer to those of dentin, whereas Ti posts demonstrated higher stress concentration, particularly in non-bonded states, increasing the potential risk of damage to surrounding tissues.

Conclusions

FRC posts, with elastic properties similar to dentin and proper adhesion, minimize stress concentration and potential damage to surrounding tissues. Conversely, materials with higher elastic modulus like Ti, can cause unfavorable stress concentrations if not properly bonded, emphasizing the importance of post adhesion in tooth restoration.

INTRODUCTION

Endodontically treated teeth are generally considered mechanically compromised due to the loss of dental tissue resulting from caries, previous restorations, crown fractures, and the access cavity preparation for root canal treatment. The reduction in dentin moisture content and structural integrity further decreases their fracture resistance [1]. Consequently, such teeth require additional restorative interventions, ranging from direct composite resin restorations and onlays to full-coverage crowns, depending on the extent of tissue loss. In cases where substantial coronal destruction compromises retention and resistance, a post is inserted into the root canal to serve as a core substitute, subsequently supporting the final restoration [2].

Posts are available in various materials, shapes, and dimensions, broadly categorized into prefabricated and cast posts [2,3]. Prefabricated posts are advantageous in terms of time efficiency and cost-effectiveness; however, they may exhibit interfacial gaps or bubbles between the post and core material, and their adaptability to the canal morphology is limited. In contrast, cast posts, fabricated as a single unit with the core, offer superior adaptation but necessitate additional clinical visits and laboratory procedures.

The reinforcement effect of post placement in endodontically treated teeth remains a subject of debate [4–6]. Some studies have reported that posts reduce stress within dentin, while others have suggested that they increase the risk of root fractures [7,8]. Finite element analysis (FEA) and experimental studies have provided conflicting findings, with evidence indicating that posts do not significantly enhance the strength of root dentin under static or fatigue conditions [7,8].

Post materials vary from metals to fiber-reinforced composites (FRC), each exhibiting distinct mechanical properties. Materials with a high elastic modulus, such as titanium (Ti) and zirconia, efficiently withstand applied forces but may induce stress concentration within the root structure due to their rigidity [9]. Duret et al. [10] proposed that the ideal post should possess an elastic modulus similar to dentin, advocating for carbon fiber posts due to their comparable mechanical behavior (elastic modulus, 21 GPa vs. dentin, 18 GPa). Similarly, FRC posts have been shown to distribute stress more evenly, reducing the risk of catastrophic root fractures [3,11]. In contrast, metal posts, including Ti, stainless steel, and zirconia, exhibit significantly higher elastic moduli, leading to increased apical stress concentration and a greater likelihood of unfavorable root fractures [11–15].

Initially, post retention relied on mechanical retention, but due to the associated risk of root fractures, adhesive bonding techniques have become the preferred approach [2]. Zinc phosphate and polycarboxylate cements, which offer inadequate adhesion to both dentin and posts, have largely been replaced by resin cements. Resin cements demonstrate superior adhesion, particularly when used with FRC posts, and have been associated with improved long-term survival rates in clinical practice [16].

Post retention is influenced by multiple factors, including post material, shape, surface texture, thickness, length, and adhesion to the luting cement [17–19]. Among these, post length is particularly critical, with various studies recommending it be at least equal to the crown length, extend to at least two-thirds of the root length, or reach at least the midpoint between the alveolar crest and the root apex [2,14,20–22]. Unlike metal posts, FRC posts achieve retention through adhesion with resin cement. The effectiveness of this bond is determined by the resin cement type, post surface treatment, and the bonding protocol used [23–26].

The present study aims to evaluate stress distribution variations between bonded and non-bonded conditions of FRC and Ti posts under identical loading conditions using FEA. In clinical practice, Ti posts are typically used in a non-bonded state, relying on mechanical retention. In contrast, FRC posts benefit from adhesive bonding, which enhances retention and stress distribution within the root canal. This fundamental difference between FRC and Ti posts allows for a direct comparison of their mechanical behavior under bonded and non-bonded conditions. By analyzing the impact of post adhesion, this study seeks to provide insights into the biomechanical behavior of different post systems. We hypothesize that bonded post conditions will result in lower stress concentrations compared to non-bonded conditions, regardless of post material.

To the best of our knowledge, no studies have separately evaluated the stress distribution within the post itself and in the surrounding dentin under both bonded and non-bonded conditions. Most FEA studies focus on overall stress patterns without distinguishing between stresses concentrated in the post and those affecting the dentin structure. However, in clinical practice, stress distribution in these two areas can differ significantly, influencing the long-term prognosis of post-retained restorations. Most previous FEA studies have assumed complete bonding between the post and dentin, which does not fully represent clinical reality. Metal posts, such as Ti, cannot achieve adhesion, whereas FRC posts rely on adhesive bonding, which serves as a significant advantage in clinical applications. However, the implications of this difference in bonding on stress distribution remain insufficiently explored. Therefore, this study addresses this gap by comparing bonded and non-bonded states for both Ti and FRC posts, providing clinically relevant insights into their stress behavior and implications for long-term success. Furthermore, by analyzing stress distribution separately in both the post and dentin, this study provides a more detailed understanding of how bonding conditions influence mechanical behavior. This differentiation is crucial, as non-bonded conditions can lead to localized stress accumulation in the post or dentin, potentially affecting clinical outcomes.

METHODS

Design and materials of the finite element model

Based on the human maxillary central incisors, a two-dimensional (2D) tooth model was created by sectioning buccolingually. After removing the coronal portion, the crown was reconstructed with a post, core, and a 2 mm thick IPS-Empress-2 ceramic (Ivoclar Vivadent, Schaan, Liechtenstein) crown (Figure 1A). The total length of the tooth was set to 23 mm, the diameter of the post to 1.4 mm, and the total length of the post to 13 mm, based on previous studies that utilized the same anatomical dimensions for FEA of maxillary central incisors in post-retained restorations [27,28]. The exposed post above the root was fixed at 1.6 mm, with 11.4 mm of the post embedded in the root, and 4 mm of gutta-percha. The embedded post was subjected to bonded or non-bonded conditions with core composite, dentin, and gutta-percha. The dimensions of the tooth model, including the 1.4 mm post diameter and 2 mm crown thickness, are shown in Figure 1B. The IPS-Empress-2 ceramic crown was modeled with a total thickness of 2 mm, comprising 0.8 mm inner thickness and 1.2 mm outer thickness. Both Ti posts and FRC posts were used, and the core was modeled with core composite resin. The FRC post used in this study was the D.T. Light-Post (BISCO Inc., Schaumburg, USA), and its material properties were obtained directly from the manufacturer. The Ti post used in this study was the ParaPost XH Titanium Post (Coltène/Whaledent Inc., Cuyahoga Falls, OH, USA), and its mechanical properties were referenced from a previous study [27]. Ti posts were included in this study as a positive control due to their significantly higher elastic modulus compared to dentin and their typical use in a non-bonded state [11,27,28]. This allows for a clearer evaluation of the effect of post adhesion by providing a distinct contrast with FRC posts. The material properties of the designed 2D model, including the elastic modulus (E) and Poisson’s ratio (υ), are shown in Table 1 [27–29].

Figure 1.

Materials and finite element model design for tooth restoration analysis. (A) Materials involved in the investigated model. (B) Dimensions of the investigated model of a tooth restored with a cylindrical post and IPS-Empress-2 crown. (C) Finite element model for analysis. FRC, fiber-reinforced composite; Ti, titanium. IPS-Empress-2: Ivoclar Vivadent, Schaan, Liechtenstein.

Table 1.

Material properties used in Finite element models

Material	Young’s modulus (E) (MPa)	Poisson’s ratio (υ)	Note
Crown	100,000	0.25
IPS-Empress-2 layering ceramic [28]	65,000	0.19
IPS-Empress-2 layering ingot [28]	12,000	0.30
Core composite [27]	18,600	0.31
Dentin [29]
Post	112,000	0.33	Bonded or non-bonded
Titanium [27]	15,000	0.29
FRC	68.9	0.45
Periodontal ligament [29]	13,700	0.30
Cortical bone [29]	0.69	0.45
Gutta-percha [29]	1,370	0.30
Cancellous bone [29]

FRC, fiber-reinforced composite.

IPS-Empress-2: Ivoclar Vivadent, Schaan, Liechtenstein. FRC: D.T. Light-Post, BISCO Inc., Schaumburg, IL, USA (information given by the manufacturers).

The mesh generation, boundary conditions, and loading directions for the 2D FEA model of this study are shown in Figure 1C. The number of elements and nodes for each material in the entire model used for analysis is listed in Table 2. Each element was modeled as a 4-node quadrilateral element with incompatible nodes under plain strain conditions using Auto Mesh function of ANSYS Workbench ver. 14.0 (ANSYS Inc., Canonsburg, PA, USA) .

Table 2.

Number of elements on used meshes

Material	Number of elements	Number of nodes
Crown
Ceramic	638	2,125
Ingot	617	2,062
Core composite	505	1,650
Dentin	3,730	11,692
Post (FRC, titanium)	950	3,061
Ligament	603	2,214
Cortical bone	1,396	5,109
Gutta-percha	33	158
Cancellous bone	3,645	11,306
Total elements	12,117

FRC, fiber-reinforced composite.

Finite element analysis and boundary conditions

A finite element model was created based on measurements of an extracted maxillary central incisor fully restored with a ceramic crown, composite resin core, and cylindrical posts (either Ti or FRC). Figure 1C shows the mesh generation and boundary conditions of the 2D finite element model. For the boundary conditions of the analysis, the nodes along the x-axis at the bottom of the cancellous bone were fixed either in the y-direction or in both the x- and y-directions. The nodes in the same length portion as the upper part of the gutta-percha were fixed in both x- and y-axes, while the remaining nodes were fixed only in the y-axis direction. These boundary conditions were established following previous studies on FEA of post-retained teeth, which analyzed the effects of different post designs and loading directions on stress distribution in root and periodontal structures. Additionally, the general principles for defining boundary conditions in finite element modeling were applied based on established computational biomechanics guidelines [29,30]. A load of 100 N (≈10 kg) was applied at a 45° angle to the long axis of the tooth on the concave central lingual surface (right side of the model). The analysis was performed using the commercial software ANSYS 14.0. The analysis was conducted for two types of posts, Ti and FRC, each with a diameter of ϕ1.4 mm and a total length of 13 mm. In all cases, 1.6 mm of the post was embedded and fixed in the core, with the remaining portion of 11.4 mm implanted in the dentin part of the root. The core, dentin, and gutta-percha in contact with the posts were analyzed under bonded and non-bonded conditions for both Ti and FRC posts. However, Ti posts are generally used in a non-bonded state in practice, so the analysis was conducted under both fully bonded and non-bonded assumptions. For FRC posts, although they are initially bonded during implantation, continuous repeated loading during use can lead to partial or complete non-bonding. Therefore, analyses were performed under both fully bonded and non-bonded assumptions.

Based on the boundary conditions, the analysis was performed under single compressive loading conditions by applying loads. The stress distribution of maximum principal stress for each material was analyzed to understand stress concentration and characteristics. Finally, by examining the stress concentration in posts and dentin under bonded and non-bonded conditions of different post materials, the characteristics based on the material and bonding conditions were identified. The analysis was conducted using the FEA tool ANSYS 14.0 Workbench under plane strain conditions for a 2D analysis.

RESULTS

In a model sectioned parallel to the tooth’s long axis, the applied load of 100 N (≈10 kg) at a 45° angle to the long axis at the central part of the lingual surface was analyzed for stress distribution using FRC and Ti posts, both with a length of 11.4 mm implanted in the root and a total length of 13 mm. The maximum principal stress distribution patterns were analyzed to confirm stress concentration and distribution trends under bonded and non-bonded conditions with surrounding materials.

The maximum principal stress values for the entire model, dentin, posts, and core composite under load, applied at the central concave part of the lingual surface at a 45° angle to the x-axis, are shown in Table 3. In all four conditions (bonded and non-bonded states for FRC and Ti posts), the maximum stress values in the entire model were similar, with slightly higher values in non-bonded conditions. This trend was also observed in the dentin.

Table 3.

Maximum of von Mises stress in all body, dentin, post, and crown under load

Variable	Maximum principal stress (MPa)
	FRC		Titanium
	Bonded	Non-bonded	Bonded	Non-bonded
All body	996.90	1,043.60	986.74	1,019.70
Post	14.797	399.60	46.01	666.64
Dentin	429.91	438.89	426.06	431.03
Core composite	15.67	203.39	14.06	196.82

FRC, fiber-reinforced composite.

However, in posts, the maximum stress concentration values in non-bonded states were significantly higher than in bonded states for both post materials. The FRC posts showed about a 27-fold increase, while the Ti posts showed about a 14-fold increase, indicating a significant increase in stress concentration in FRC posts. This is because the strength of FRC posts is approximately one-sixth that of Ti posts. Both FRC and Ti posts in bonded conditions and FRC posts in non-bonded conditions exhibited high stress concentration values in the dentin. In contrast, Ti posts in non-bonded conditions showed higher stress concentration values in the posts.

The stress distribution patterns under different bonding conditions for FRC and Ti posts are presented in Figure 2. The maximum principal stress was concentrated at the upper outer dentin near the contact area with the ceramic. A force applied at a 45° angle to the long axis is similar to a bending force. The slightly higher maximum value of the FRC posts compared to the Ti posts is attributed to the stronger material properties of the Ti posts under bending loads.

Figure 2.

Maximum principal stress distribution across the tooth model under load conditions. (A) Maximum stress of tooth model with bonded fiber-reinforced composite (FRC) post under load. (B) Maximum stress of tooth model with non-bonded FRC post under load. (C) Maximum stress of tooth model with bonded titanium (Ti) post under load. (D) Maximum stress of tooth model with non-bonded Ti post under load.

The stress concentration in posts and dentin under bonded and non-bonded conditions is illustrated in Figure 3. Only the non-bonded Ti post showed maximum stress concentration at the upper buccal corner, while the other three conditions showed stress concentration at the upper outer dentin (Figure 3D). The maximum stress concentration values in dentin were similar under bonded and non-bonded conditions, with slightly higher values in non-bonded states. Both post materials showed increased stress concentration in posts and dentin under non-bonded conditions. Non-bonded FRC posts showed lower stress concentration values than dentin, while non-bonded Ti posts showed 1.6 times higher stress concentration values than dentin, with significant stress at the edges in contact with dentin. This indicates that non-bonded posts have reduced resistance to bending loads, leading to increased stress concentration in the posts and surrounding materials.

Figure 3.

Maximum stress distribution of maximum principal stress in post and dentin of tooth model under load: (A) bonded fiber-reinforced composite (FRC), (B) non-bonded FRC, (C) bonded titanium (Ti), and (D) non-bonded Ti.

Stress concentration was observed in the lingual cancellous bone and dentin area, with stress also concentrated at the lingual dentin and post areas. In non-bonded states, stress was concentrated at the upper and lower ends of the post, and at the upper lingual dentin edge in contact with the post. Stress concentration was mainly in the lingual dentin, not the posts. FRC posts showed maximum stress values in dentin 36 times higher than in posts, while Ti posts showed 13 times higher stress values in dentin, with a three-fold greater increase in FRC posts than Ti posts.

Figure 4 shows the detailed stress distribution in the core composite, post, and dentin under bonded and non-bonded conditions. Bonded posts had lower stress concentration values than dentin, while non-bonded posts showed significantly higher stress concentration than bonded states (Figure 4B). Bonded FRC posts showed maximum stress concentration at the right edge in contact with dentin and core, while non-bonded FRC posts showed maximum stress concentration at the upper buccal edge and high stress at the boundary with dentin. Ti posts showed stress concentration at the lower part and upper contact edges, with maximum stress concentration at the lower buccal edge in bonded states and upper buccal edge in non-bonded states. Under load, non-bonded posts showed several-fold increases in maximum stress concentration values, with more complex stress distribution. Posts may transition from bonded to non-bonded states under repeated loads, leading to increased stress concentration and potential damage to surrounding tissues, emphasizing the importance of post adhesion in tooth restoration.

Figure 4.

Maximum stress distribution of maximum principal stress in tooth model under load: (A) core composite, (B) post, and (C) dentin. FRC, fiber-reinforced composite; Ti, titanium.

Figure 4C shows stress distribution in dentin under bonded and non-bonded conditions, with similar patterns and slightly higher stress concentration values in non-bonded states, concentrated at the upper buccal corner. Non-bonded conditions also showed stress concentration at the dentin in contact with the post lower edges, indicating the potential for internal fractures due to stress concentration.

Figure 4A shows the maximum principal stress distribution in the core composite. Bonded FRC posts showed stress concentration away from post contact areas, while non-bonded conditions in both posts showed stress concentration at the upper post and lower core edges, significantly increasing stress concentration values by 13-fold in non-bonded states. This highlights the importance of effective load distribution in bonded states to prevent stress concentration and potential tooth damage.

DISCUSSION

This study employed a 2D FEA method that provides easily interpretable results compared to three-dimensional (3D) FEA and excluded the cement layer between the root dentin and the post. While 3D models offer a more comprehensive simulation of complex anatomical structures and occlusal forces, they require significantly higher computational resources and more complex meshing processes. In contrast, 2D FEA allows for efficient analysis with well-controlled boundary conditions and has been widely used in previous studies for evaluating stress distribution in post-retained restorations [27–30]. Although the 2D model has limitations in fully replicating the 3D biomechanical behavior of teeth and surrounding tissues, it is a widely accepted approach for comparative stress analysis and remains a valuable tool for preliminary evaluations in dental biomechanics research. This is because the elastic modulus of resin cement is not significantly higher than that of dentin, and there is limited information on the thickness of the cement layer encountered in clinical practice.

Posts used clinically are made from various materials, including gold alloys, Ti, stainless steel, ceramics, zirconia, and FRC. FRC posts are manufactured by embedding carbon fibers, quartz, or glass fibers in epoxy resin or methacrylate resin [3,9]. The fibers are aligned parallel to the post’s long axis, with diameters of 6 to 15 µm. The number of fibers in the post’s cross-section varies by type, ranging from 25 to 30 fibers, occupying 30% to 50% of the area. Adhesion between the fibers and resin matrix is enhanced by silanization before embedding. Strong interfacial adhesion between fibers and the resin matrix allows stress transfer from the matrix to the fibers, essential for effective reinforcement properties [9,31].

Previous FEA studies related to stress distribution in posts did not consider different interface conditions, implying that both metal and FRC posts were bonded to the dentin. However, metal posts are not actually bonded to the cement, and FRC posts can become non-bonded over time due to increased fatigue stress. In this study, models of both bonded and non-bonded states of Ti and FRC posts were created and analyzed, considering actual usage conditions. Unlike previous studies, which assumed complete bonding of posts to the root, dentin, and composite resin core, this study accounted for the realistic conditions of usage [11]. Ti posts generally remain in a non-bonded state upon implantation, whereas FRC posts maintain a bonded state due to silanization. However, some may mistakenly believe that metal posts are bonded to surrounding tissues due to their excellent mechanical properties, expecting superior performance in use [9]. From an engineering perspective, repeated loading on a non-bonded contact state leads to fretting, which can damage surrounding tissues due to the post material’s excellent mechanical properties [11]. FRC posts, despite being bonded initially, may also become non-bonded due to repeated masticatory loads and moisture absorption by the cement [24]. Therefore, it is necessary to predict potential non-bonding scenarios. Consequently, the analysis included not only the bonded and non-bonded states of the materials but also the opposite conditions.

Ti posts, regardless of bonding, cannot avoid stress concentration due to their higher elastic modulus compared to dentin. However, if Ti posts could be implanted in a bonded state, they might demonstrate better characteristics despite the unavoidable risk of root fractures from stress concentration. Practically, bonding Ti posts during implantation is difficult. Thus, higher stress concentration in the post than dentin during non-bonded state loads may increase the risk of surrounding tissue damage. Conversely, FRC posts, while showing increased maximum stress concentration in non-bonded states, exhibit stress concentration values similar to those of dentin in both bonded and non-bonded states. Despite the limitations of this study, our findings suggest that FRC posts may have less impact on surrounding tissue damage from stress concentration compared to metal posts.

Compared to bonded posts, non-bonded posts showed an increase in stress concentration. Our findings underscore the critical role of adhesion in stress distribution. Non-bonded posts exhibited significantly higher stress concentrations compared to bonded posts, particularly for Ti posts due to their high elastic modulus. This aligns with clinical observations where non-bonded posts increase the likelihood of mechanical failures such as root fractures [11,16,28]. While FRC posts maintained more favorable stress distributions even in non-bonded states, their adhesion degradation over time cannot be overlooked [3]. In clinical practice, adhesive failure is a significant concern for FRC posts due to moisture absorption and fatigue stress [23,24]. This highlights the necessity of robust bonding techniques and periodic assessment of bonded restorations. Future studies should explore the long-term effects of cyclic loading on adhesion degradation in FRC posts and investigate various materials or surface treatments to improve the durability of bonded restorations. Additionally, clinical trials comparing the failure modes and survival rates of bonded and non-bonded posts in various patient populations are necessary to validate these findings in real-world scenarios. Importantly, previous FEA studies assumed ideal bonding conditions, which do not align with clinical realities, particularly for metal posts. Our study, by incorporating both bonded and non-bonded scenarios, bridges this research gap and provides a more accurate representation of post-restored teeth under functional loads [11,16,23,29]. Additionally, Ti posts in non-bonded states exhibited higher stress concentration in the post compared to dentin under loads, whereas FRC posts showed lower values compared to dentin under loads. The load caused significantly greater stress concentration and impact on posts and surrounding tissues, especially in non-bonded states. FRC posts bonded to root canal dentin distributed stress uniformly to the dentin. While non-bonded FRC posts showed slight stress concentration at the post-dentin interface and root tip, the stress magnitude was not significantly higher than in dentin. This supports the conclusion of other studies that the primary failure cause of restorations with FRC posts is not root fracture but post non-bonding [11,12,23].

Therefore, ensuring the adhesion of FRC posts with resin cement to both the post and dentin is crucial for maintaining restorations. This includes proper surface treatment of the post, adhesion process, appropriate use of resin cement, and effective removal of gutta-percha, sealers, and other materials previously applied in the root canal during post hole preparation.

CONCLUSIONS

Considering all the limitations of this study, we can conclude that bonded FRC posts demonstrate a more favorable stress distribution compared to Ti posts, which exhibited higher stress concentration, particularly in non-bonded conditions. The results suggest that post adhesion plays a crucial role in minimizing mechanical complications and reducing the risk of surrounding tissue damage.

Furthermore, our findings emphasize the importance of selecting appropriate post materials and ensuring proper bonding techniques to improve long-term clinical outcomes. Clinicians should carefully consider the mechanical behavior of different post materials and the effects of bonding conditions when restoring endodontically treated teeth. Future studies with 3D models and in vivo validation are needed to further explore the biomechanical performance of post-retained restorations.