Evaluation of biofilm disruption on root dentin following conventional and nanoparticle-enhanced photodynamic therapies using scanning electron microscopy and FIJI-based image analysis

Abstract

Objective

Enterococcus faecalis biofilm is a major contributor to persistent root canal infections due to its high resistance to conventional irrigants. This study aimed to evaluate and compare different disinfection protocols, including sodium hypochlorite (NaOCl), photodynamic therapy (PDT) with methylene blue (MB), and MB functionalized with reduced graphene oxide (MB-rGO) nanoparticles with or without sonic activation, in disrupting E. faecalis biofilm on root canal dentin using SEM and quantitative FIJI-based image analysis.

Materials and methods

Eighty root dentin specimens were obtained from forty extracted single-rooted human teeth. Following canal preparation and instrumentation, thirty-five teeth were inoculated with Enterococcus faecalis and incubated for 21 days to develop mature biofilms, while five teeth served as negative controls. Each tooth was longitudinally sectioned to yield two standardized dentin specimens, which were randomly allocated to six experimental groups (n = 10 specimens per group) according to the disinfection protocol. Groups 1–2 received 5% sodium hypochlorite (NaOCl), Groups 3–4 underwent photodynamic therapy (PDT) using methylene blue (MB), and Groups 5–6 received PDT with methylene blue functionalized with reduced graphene oxide (MB-rGO). Sonic activation was applied in Groups 2, 4, and 6. Biofilm removal was assessed using scanning electron microscopy (SEM) and quantified with FIJI software employing the Trainable Weka Segmentation plugin. A 6-point visual scoring scale was additionally applied. Statistical analysis was performed using one-way ANOVA with Tukey’s post hoc test and the Wilcoxon signed-rank test (p < 0.05).

Results

The positive control showed extensive biofilm coverage (73.36 ± 6.36%). NaOCl-treated groups showed moderate reduction (Group 1: 35.57 ± 3.11%; Group 2: 31.67 ± 2.09%). PDT with MB exhibited greater reduction (Group 3: 21.68 ± 2.65%; Group 4: 19.11 ± 1.78%). The MB-rGO groups presented the highest efficacy (Group 5: 17.71 ± 2.66%; Group 6: 10.46 ± 1.92%), with Group 6 significantly outperforming all others (p < 0.001). Semi-quantitative scores reflected similar trends, with Group 6 showing the lowest score (1.20 ± 0.63, p < 0.001).

Conclusion

MB-rGO combined with sonic and laser activation demonstrated the highest efficacy against E. faecalis biofilm, indicating a promising approach for enhanced endodontic disinfection.

Introduction

Biofilms are self-produced, complex microbial structures in an extracellular polymeric substance (EPS), which includes polysaccharides, proteins, lipids, and extracellular DNA (eDNA). The EPS layer generally measures between 0.2 and 1.0 μm in thickness, while the entire biofilm typically ranges in size from 10 to 30 nm [1]. These biofilms complicate elimination of infections, often requiring antibiotic doses up to 1000 times higher than standard therapeutic levels [2].

Enterococcus faecalis (E. faecalis) is widely recognized as a leading cause of persistent endodontic infections and treatment failures due to its remarkable ability to survive harsh environmental conditions [3]. One of its key survival mechanisms is its resistance to elevated pH levels, allowing it to persist even in the presence of a common calcium hydroxide (Ca(OH)₂) intracanal medicament [4]. This resistance is largely attributed to the action of a proton pump within its plasma membrane, which helps maintain cytoplasmic homeostasis and enables the bacterium to withstand highly alkaline environments [5]. Consequently, the presence of E. faecalis in an infected root canals necessitate it completely eliminate to achieve treatment success. A major element contributing to the persistence of E. faecalis is its ability to form biofilms. Therefore, preventing and disrupting biofilms is essential to ensure the success of root canal therapy [6].

In clinical practice, irrigants such as sodium hypochlorite (NaOCl) and chlorhexidine (CHX) are routinely used to manage biofilm-associated infections. NaOCl is favored for its potent antibacterial properties and its ability to dissolve organic tissue; however, its cytotoxicity and risk of periapical tissue irritation necessitate careful clinical control [7]. In contrast, CHX offers broad-spectrum antimicrobial activity with reduced toxicity but lacks the ability to dissolve organic tissues [8, 9]. Despite their widespread use, both agents have inherent limitations that can affect the overall efficacy of root canal disinfection and long-term treatment outcomes.

In recent years, nanomaterials have been explored as adjunctive strategies in root canal disinfection due to their enhanced surface reactivity and antimicrobial potential. Graphene, a nanoparticle material consisting of a single atomic layer of carbon, exhibits exceptional properties such as high surface area and excellent thermal stability, which make it suitable for biomedical and dental applications [10, 11]. Despite its promising features, the large-scale synthesis of pristine graphene remains a challenge due to its complex production methods, high cost, and limited reproducibility. These limitations restrict its routine biomedical and dental applications. As a result, researchers have shifted focus toward more scalable and cost-effective graphene derivatives, such as graphene oxide (GO) and reduced graphene oxide (rGO) [10].

In a study by Martini et al. dentin discs coated with GO and subsequently incubated with E. faecalis for 72 h demonstrated a significant reduction in bacterial adhesion and biofilm formation [11]. Similarly, Kim et al. assessed the antibacterial performance of GO and RGO against E. faecalis biofilms on hydroxyapatite discs and observed effective inhibition by both materials. Notably, RGO exhibited lower cytotoxicity than GO, reflecting its superior biocompatibility and potential root canal disinfectant targeting at eliminating E. faecalis biofilms [10]. Additionally, photodynamic therapy (PDT) has gained attention as an adjunctive disinfection method in endodontics, owing to its antimicrobial efficacy and its unique ability to cross-link collagen with proteins, enhancing structural stability [12]. Among various photosensitizers, methylene blue (MB) has been extensively studied for its effectiveness, particularly against E. faecalis, a key pathogen associated with persistent root canal infections and treatment failures [13]. Despite its documented antimicrobial efficacy, conventional photodynamic therapy has certain limitations that may restrict its effectiveness in complex root canal systems. These include limited penetration of photosensitizers into mature biofilms and deep dentinal tubules, uneven photosensitizer distribution within the canal anatomy, and potential instability or rapid clearance of the photosensitizing agent. As a result, PDT alone may not always achieve complete biofilm disruption. To overcome these limitations, recent approaches have focused on the use of nanoparticle-based photosensitizer delivery systems and adjunctive activation methods to enhance penetration, retention, and overall photodynamic antimicrobial efficacy [13].

Ultrasonic agitation brings about the disruption of bacterial biofilms by breaking them apart and dispersing the bacteria into their planktonic form, which is more vulnerable to antimicrobial irrigants. On top of that, the cavitation effect generated during ultrasonic activation can temporarily compromise bacterial cell membranes, enhancing their permeability and allowing greater penetration of antimicrobial agents [14, 15]. Parolia et al. utilized ultrasonication to synthesize propolis nanoparticles and assessed their antibacterial efficacy against E. faecalis biofilms using SEM and confocal laser scanning microscopy (CLSM). Their findings revealed that propolis nanoparticles, when used as an endodontic irrigant, were as effective as 6% NaOCl and 2% CHX in reducing E. faecalis colony-forming units (CFUs) in both human tooth models and clinical isolates from patients with failed root canal treatments [16].

Persistent root canal infections, often caused by E. faecalis, are difficult to eradicate due to the bacterium’s ability to form a resilient biofilm that protects it from conventional antimicrobial treatments. While traditional irrigants like NaOCl and CHX have been widely used, they often fail to eliminate biofilms or penetrate deeply into dentinal tubules. Recent advances suggest that modified PDT techniques, such as functionalized photosensitizers or the incorporation of sonic or ultrasonic activation, may enhance biofilm disruption, but comparative data evaluating graphene-functionalized photosensitizer systems combined with adjunctive activation methods remain limited. This study evaluates reduced graphene oxide-functionalized methylene blue (MB-rGO) as a nanoparticle-enhanced photodynamic strategy for disrupting E. faecalis biofilms on root dentin using SEM and quantitative FIJI-based analysis.

Materials and methods

This in-vitro investigation was conducted from June 2024 to March 2025. Ethical approval was obtained from the Research Ethics Committee, King Faisal University, Al-Ahsa, Saudi Arabia (KFU-REC-2024-MAR-ETHICS2072).

Sample preparation and biofilm development

Sample size estimation was guided by the expected effect size and variability reported in comparable in-vitro endodontic biofilm studies employing SEM-based quantitative analysis [16, 17]. Based on these studies, an anticipated difference of approximately 15–20% in biofilm coverage between treatment groups, with a standard deviation of approximately 3–5%, was considered biologically meaningful. At a significance level of 0.05 and 80% statistical power, a minimum of five specimens per group was deemed sufficient. To account for potential specimen loss during preparation and SEM processing and to strengthen analytical robustness, ten specimens per group were included. A total of eighty standardized root dentin specimens, obtained from forty freshly extracted single-rooted human teeth, were included in this study. The teeth were collected from orthodontic patients and stored in 0.1% thymol solution until use. Teeth exhibiting caries, restorations, cracks, fractures, or root curvature were excluded.

The crowns were sectioned to standardize the root lengths at 13 mm. Root canal instrumentation was performed using ProTaper rotary files up to size F3, stopping 1 mm short of the working length, and the apical ends were sealed with composite resin (Filtek™ Z250, 3M™, New Zealand), and all specimens were sterilized by autoclaving.

For biofilm development, thirty-five teeth were inoculated with Enterococcus faecalis (ATCC 29212) and cultured in brain heart infusion (BHI) broth, then adjusted to a turbidity of 0.5 McFarland. Five teeth served as negative controls and were inoculated with sterile BHI broth. All samples were incubated vertically in BHI broth at 37 °C for 21 days, with the medium replaced every 48 h. Following incubation, sterile paper points were used to collect the biofilm medium. Biofilm formation was verified after the 21-day incubation period by SEM observation of randomly selected positive-control specimens, which demonstrated dense and well-established E. faecalis biofilm coverage occluding dentinal tubules.

From the infected teeth, five teeth were randomly allocated as positive controls, while the remaining thirty teeth were assigned to six experimental treatment groups according to the disinfection protocol. The experimental groups were defined as follows:

• Group 1: 5% NaOCl with manual irrigation.

• Group 2: 5% NaOCl activated sonically (6000 cpm for 2 min).

• Group 3: Methylene blue (MB, 200 µg/mL) activated with a laser (650 ± 5 nm, 200 mW, 2 min).

• Group 4: MB with sonic activation followed by laser application.

• Group 5: MB functionalized with reduced graphene oxide (MB/rGO, 200 µg/mL) plus laser.

• Group 6: MB/rGO combined with sonic activation and laser treatment.

For Groups 5 and 6, methylene blue functionalized with reduced graphene oxide (MB-rGO) was employed as the photosensitizer. The synthesis, functionalization protocol, and comprehensive physicochemical characterization of MB-rGO, along with its antimicrobial efficacy, have been previously reported in the literature and were adopted without modification in the present investigation [18, 19].

Irrigation in Groups 1 and 2 was performed using 3 mL of NaOCl, agitated with a size 10K-file. While in Groups 2, 4, and 6 sonic activation was performed using an EndoActivator system (Dentsply Sirona, USA) operating at approximately 6000 cycles per minute. A 25/04 polymer tip was positioned 1 mm short of the working length and activated using gentle vertical in-and-out motion to facilitate irrigant and photosensitizer agitation within the canal. In Groups 3–6, photodynamic therapy (PDT) was performed using a diode laser (F3WW, Foshan, China) delivered via a 200 µm optical fiber positioned 1 mm short of the working length. The laser output power was verified using a calibrated optical power meter before each irradiation session. Laser irradiation was delivered in continuous wave mode at a wavelength of 650 ± 5 nm and an output power of 200 mW for 120 s, resulting in a total energy delivery of 24 J. These parameters were selected based on previously published endodontic PDT studies demonstrating effective bacterial reduction without inducing thermal damage to dentin, as well as optimal absorption characteristics of methylene blue for efficient reactive oxygen species generation [19].

Each tooth was subsequently longitudinally sectioned into two halves using a low-speed diamond saw (IsoMet™ 1000; Buehler Ltd., Illinois, USA) operating at 200 rpm under continuous water irrigation to prevent heat generation, yielding two dentin specimens per tooth and a total of ten dentin specimens per group (n = 10) for SEM analysis.

Fixation and SEM analysis of E. faecalis biofilm on root dentin

The prepared root dentin specimens were fixed in 2.5% glutaraldehyde prepared in 0.1 M phosphate-buffered saline (PBS; pH 7.4) at 4 °C for 24 h, followed by three rinses in PBS. Dehydration was performed using a graded ethanol series (30–100%), with each step lasting 10 min; the 100% ethanol step was repeated twice. The specimens were air-dried overnight, mounted on aluminum stubs using carbon adhesive tape, and sputter-coated with a 10–20 nm gold–palladium layer.

SEM analysis was conducted at an accelerating voltage of 20 kV and a magnification of 3000× to evaluate biofilm morphology and the interaction between E. faecalis biofilm and root dentin surfaces. SEM evaluation was standardized by imaging specimens from the middle third of the root canal. For each specimen, three representative non-overlapping fields were captured at standardized magnification to ensure consistency and minimize selection bias.

Biofilm quantification using SEM images and FIJI software

Quantitative analysis of biofilm from SEM images was conducted using FIJI (Fiji Is Just ImageJ; version 2.16.0/1.54p, Java 21.0.7, National Institutes of Health, Bethesda, MD, USA), an open-source platform for image analysis. The Trainable Weka Segmentation plugin (version 4.0.0) was employed to develop a pixel-based classifier. Initially, representative SEM images were manually labelled to identify three distinct classes: biofilm, dentin, and dentinal tubules. The classifier was programmed to distinguish and highlight biofilm regions as white in the binary output, while dentin and tubules remained black, reducing the likelihood of misidentifying tubules as biofilm. Formal classifier performance metrics (e.g., accuracy, precision, recall) were not generated; therefore, the potential for minor misclassification cannot be excluded and is acknowledged as a methodological limitation.

Once programmed, the classifier was applied uniformly across all treatment groups to generate consistent binary images. These outputs were processed using FIJI’s thresholding and measurement functions to determine the percentage of surface covered by biofilm (Fig. 1). The following formula was used:

Fig. 1.

Illustration of the sequential process for segmenting and quantifying E. faecalis biofilm on dentin surfaces using the Trainable Weka Segmentation plugin in FIJI (ImageJ). A Manual labeling of biofilm, dentin, and tubules during classifier training. B Output produced by the trained classifier, displaying clearly defined overlay classes with adjusted thresholds. C Final binary mask highlighting biofilm regions in white, which were utilized for calculating surface area coverage

In this context, the biofilm area corresponds to the white pixels in the binary image, and the total number of pixels represents the entire image field. In the negative control group, images appeared almost entirely black; therefore, its binary image was captured without Weka training.

In addition to this quantitative approach, a semi-quantitative visual scoring system was employed to assess biofilm coverage. This scoring scale was adapted and refined from the 4-point system developed by Rusu and Stratul [20]. The semi-quantitative scoring system used to assess biofilm coverage was based on a 6-point scale (ranging from 0 to 5) after validation with experts in the field. A score of 0 indicated a completely clean surface with no visible biofilm. Score 1 represented trace amounts of biofilm, covering less than 10% of the surface. Score 2 denoted light biofilm coverage, ranging between 10 and 30%. Score 3 corresponded to moderate biofilm presence, covering 30–60% of the surface. A score of 4 indicated heavy biofilm accumulation, with 60–90% of the area affected. Finally, score 5 was assigned to samples with extensive biofilm, where more than 90% of the surface was covered. SEM image analysis and scoring were performed independently by two calibrated evaluators who were blinded to group allocation. Any discrepancies were resolved by consensus. Inter-examiner agreement was assessed before analysis to ensure scoring consistency.

Statistical analysis

Data analysis was carried out using IBM SPSS Statistics (Version 23.0, IBM Corp., Armonk, NY, USA). Each root dentin specimen was considered an independent unit of analysis, as all specimens were prepared, treated, and analyzed individually. Quantitative measurements obtained from FIJI software were reported as mean ± standard deviation with corresponding 95% confidence intervals. The Shapiro–Wilk test was used to evaluate the normality of the data. Since the data satisfied the assumptions of normal distribution and equal variances, one-way ANOVA was employed to compare the mean percentages of biofilm coverage between groups. Homogeneity of variances was assessed using Levene’s test before one-way ANOVA. Tukey’s honest significant difference (HSD) test was used for post hoc comparisons to identify differences between specific groups. Semi-quantitative biofilm scores, obtained using an ordinal scoring system (0–5), were compared between the positive control and each experimental group using the Mann–Whitney U test. SEM image selection, FIJI-based biofilm quantification, and semi-quantitative scoring were performed by evaluators blinded to group allocation. A p value less than 0.05 was considered statistically significant.

Results

All quantitative and semi-quantitative results are reported per root dentin specimen, which served as the unit of analysis. The percentage of biofilm coverage measured from SEM images showed that the negative control group exhibited minimal biofilm presence, with a mean coverage of 0.65 ± 0.32%, whereas the positive control group showed extensive biofilm formation (73.36 ± 6.36%). Among the treatment groups, methylene blue functionalized with reduced graphene oxide (MB-rGO) demonstrated the most substantial biofilm reduction, particularly Group 6 (MB-rGO + Sonic + Laser), which exhibited the lowest coverage at 10.46 ± 1.92%, followed by Group 5 (MB-rGO + Laser) at 17.71 ± 2.66%, as presented in Fig. 2.

Fig. 2.

Quantitative analysis of E. faecalis biofilm coverage across treatment groups using FIJI software

Pairwise comparisons revealed that both NaOCl-treated groups (Groups 1 and 2) showed significantly higher biofilm coverage than the photodynamic therapy-based groups (Groups 3–6) (all p < 0.001). While no statistically significant difference was observed between Group 3 and Group 4 (p = 0.585), Group 6 demonstrated significantly lower biofilm coverage compared with Groups 3, 4, and 5 (p < 0.001), as summarized in Table 1 and illustrated in Fig. 3.

Table 1.

Comparative analysis of biofilm coverage across groups

Comparison of groups	Biofilm coverage (%)
	Mean ± SD	p value
Positive controls vs. Group 1	73.36 ± 6.36 vs. 35.57 ± 3.11	< 0.001
Positive controls vs. Group 2	73.36 ± 6.36 vs. 31.67 ± 2.09	< 0.001
Positive controls vs. Group 3	73.36 ± 6.36 vs. 21.68 ± 2.65	< 0.001
Positive controls vs. Group 4	73.36 ± 6.36 vs. 19.11 ± 1.78	< 0.001
Positive controls vs. Group 5	73.36 ± 6.36 vs. 17.71 ± 2.66	< 0.001
Positive controls vs. Group 6	73.36 ± 6.36 vs. 10.46 ± 1.92	< 0.001
Group 1 vs. Group 2	35.57 ± 3.11vs. 31.67 ± 2.09	0.128
Group 1 vs. Group 3	35.57 ± 3.11 vs. 21.68 ± 2.65	< 0.001
Group 1 vs. Group 4	35.57 ± 3.11 vs. 19.11 ± 1.78	< 0.001
Group 1 vs. Group 5	35.57 ± 3.11 vs. 17.71 ± 2.66	< 0.001
Group 1 vs. Group 6	35.57 ± 3.11 vs. 10.46 ± 1.92	< 0.001
Group 2 vs. Group 3	31.67 ± 2.09 vs. 21.68 ± 2.65	< 0.001
Group 2 vs. Group 4	31.67 ± 2.09 vs. 19.11 ± 1.78	< 0.001
Group 2 vs. Group 5	31.67 ± 2.09 vs. 17.71 ± 2.66	< 0.001
Group 2 vs. Group 6	31.67 ± 2.09 vs. 10.46 ± 1.92	< 0.001
Group 3 vs. Group 4	21.68 ± 2.65 vs. 19.11 ± 1.78	0.585
Group 3 vs. Group 5	21.68 ± 2.65 vs. 17.71 ± 2.66	0.115
Group 3 vs. Group 6	21.68 ± 2.65 vs. 10.46 ± 1.92	< 0.001
Group 4 vs. Group 5	19.11 ± 1.78 vs. 17.71 ± 2.66	0.789
Group 4 vs. Group 6	19.11 ± 1.78 vs. 10.46 ± 1.92	< 0.001
Group 5 vs. Group 6	17.71 ± 2.66 vs. 10.46 ± 1.92	< 0.001

Tukey’s HSD test

Fig. 3.

Scanning electron microscopy (SEM) images (× 3000 magnification) illustrating dentin surface morphology in the negative control (A), positive control (B), and six experimental groups (C–H). The negative control exhibits clean dentin with patent dentinal tubules, whereas the positive control demonstrates dense biofilm coverage occluding the tubules. The treatment groups show progressive biofilm disruption and varying degrees of dentinal tubule re-exposure, with the MB-rGO + sonic activation + laser group (H) demonstrating the most pronounced biofilm removal and tubule patency

Semi-quantitative scoring based on SEM image evaluation showed the highest mean biofilm score in the positive control group (3.80 ± 0.42). All treatment groups demonstrated significantly lower scores compared with the positive control (p < 0.001). Group 1 (NaOCl) showed a moderate reduction (2.40 ± 0.51), while Group 2 (NaOCl + sonic) exhibited a further decrease (1.80 ± 0.42). The PDT-based groups (Groups 3 and 4) demonstrated comparable reductions, with mean scores of 2.20 ± 0.42 and 2.10 ± 0.31, respectively. More pronounced reductions were observed in the MB-rGO groups, with Group 5 scoring 2.00 ± 0.00 and Group 6 exhibiting the lowest score (1.20 ± 0.63), indicating minimal residual biofilm (Table 2; Fig. 4).

Table 2.

Semi-quantitative scoring of E. faecalis biofilm on dentin surfaces based on SEM image evaluation

Comparison of groups	Scoring of E. faecalis biofilm
	Mean ± SD	p value
Control positive vs. Group 1—NaOCl	3.80 ± 0.42 vs. 2.40 ± 0.51	0.006
Control positive vs. Group 2—NaOCl + Sonic	3.80 ± 0.42 vs. 1.80 ± 0.42	0.004
Control positive vs. Group 3—MB + Laser (PDT)	3.80 ± 0.42 vs. 2.20 ± 0.42	0.004
Control positive vs. Group 4—MB + Sonic + Laser	3.80 ± 0.42 vs. 2.10 ± 0.31	0.004
Control positive vs. Group 5—MB-rGO + Laser	3.80 ± 0.42 vs. 2.00 ± 0.00	0.003
Control positive vs. Group 6—MB-rGO + Sonic + Laser	3.80 ± 0.42 vs. 1.20 ± 0.63	0.004

Mann–Whitney U test

Fig. 4.

Binary images of SEM micrographs analyzed using FIJI software for biofilm quantification. Binary representations of SEM images generated using FIJI software (Fiji Is Just ImageJ, v2.16.0) with Trainable WEKA Segmentation. White areas represent biofilm coverage, while black areas correspond to exposed dentin and dentinal tubules. The negative control shows no biofilm (black), whereas the positive control shows dense coverage (white). A progressive reduction in biofilm is evident from Group 1 through Group 6, with Group 6 showing the least white area, indicating the most effective disinfection

Discussion

This study demonstrated that photodynamic therapy (PDT) combined with functionalized photosensitizers and sonic activation is more effective in disrupting Enterococcus faecalis biofilms on root dentin compared to conventional irrigants. While sodium hypochlorite (NaOCl), with or without sonic activation, showed moderate efficacy, methylene blue (MB)-mediated PDT produced a further reduction in biofilm, underscoring the added value of light-activated antimicrobial mechanisms beyond chemical disinfection alone.

The most substantial biofilm reduction was observed with methylene blue functionalized with reduced graphene oxide (MB-rGO), particularly when combined with sonic activation and laser irradiation (Group 6). This finding suggests a potential synergistic interaction between functionalized photosensitizers and physical activation methods. The enhanced performance of MB-rGO-mediated PDT may be attributed to the unique physicochemical properties of rGO, including improved photosensitizer loading and enhanced light absorption, which could contribute to increased reactive oxygen species (ROS) generation and bacterial membrane disruption, as suggested in previous studies [10, 11]. From a nanomaterial perspective, rGO may further enhance photodynamic efficacy through its conjugated π–electron structure, which can facilitate electron transfer processes during light activation and potentially amplify ROS generation. In addition, rGO has been reported to exhibit localized photothermal effects and improved light absorption, which may contribute to enhanced bacterial membrane destabilization when combined with laser irradiation [10, 21]. These mechanisms are proposed based on existing literature and were not directly measured in the present study. These effects may be further amplified by sonic activation, which facilitates deeper penetration of the photosensitizer into the biofilm matrix.

Quantitative analysis using SEM and FIJI-based image analysis confirmed the robustness of the experimental model, with minimal biofilm presence in the negative control compared to extensive biofilm formation in the positive control. These marked contrasts support the validity of the observed treatment effects. However, the present findings differ from those reported by Teves-Córdova et al., who evaluated a high-power 810 nm diode laser combined with NaOCl and passive ultrasonic irrigation (PUI) on dual-species biofilms and smear layer removal [22]. This discrepancy may reflect fundamental differences in disinfection mechanisms, as PDT primarily relies on ROS-mediated microbial damage rather than thermal or hydrodynamic effects alone. The present study focused primarily on structural biofilm disruption and surface coverage reduction rather than direct bacterial viability. SEM combined with FIJI-based quantitative analysis provides high-resolution morphological assessment but does not distinguish between viable and non-viable bacterial cells. The antimicrobial efficacy of MB-rGO-mediated photodynamic therapy against E. faecalis has been previously reported [18, 19]. Future studies incorporating CFU quantification or additional viability assays would further strengthen translational validation.

Similarly, Liu et al. introduced a hydrodynamic cavitation system (Odne™Clean) using saline and demonstrated biofilm removal comparable to conventional NaOCl protocols [23], highlighting the role of physical fluid dynamics rather than chemical enhancement. Sehitoglu et al. evaluated graphene oxide and its nanocomposites with EDTA and chitosan and reported that graphene oxide alone did not enhance smear layer removal [24]. These findings suggest that graphene derivatives do not inherently improve debridement outcomes unless appropriately functionalized and activated. In contrast, the present study demonstrates that functionalization of graphene derivatives with photosensitizers, combined with laser and sonic activation, is critical for accentuating antibiofilm effects. The superior performance observed appears to result from the combined chemical, photodynamic, and mechanical actions rather than the presence of graphene derivatives alone. Veeraiyan et al. similarly reported that EDTA and chitosan were among the most effective agents for smear layer removal [25].

With respect to conventional irrigants, NaOCl-treated groups demonstrated moderate biofilm reduction. The absence of a statistically significant difference between NaOCl with and without sonic activation suggests that, under the present conditions, sonic agitation alone may not substantially enhance NaOCl efficacy. This contrasts with observations by de Gregorio et al., who reported enhanced biofilm disruption with ultrasonic activation of NaOCl [26]. Although NaOCl remains a gold-standard irrigant due to its tissue-dissolving and antimicrobial properties, its limited penetration into mature biofilms has been consistently reported [7].

PDT-based groups demonstrated superior biofilm disruption compared to NaOCl-based protocols, supporting previous evidence that PDT effectively targets E. faecalis biofilms through ROS-mediated mechanisms [21]. More pronounced effects were observed in MB-rGO-treated groups, particularly when combined with dual activation. These findings are consistent with earlier studies reporting the antibiofilm potential of graphene derivatives, where rGO has been suggested to enhance antimicrobial efficacy through improved ROS generation and bacterial membrane interactions [10, 11].

The use of SEM provided a detailed visualization of biofilm morphology and dentinal tubule involvement, corroborating the quantitative findings. Progressive biofilm disruption and re-exposure of dentinal tubules were evident across treatment groups. Semi-quantitative scoring further supported these observations, with PDT-based and MB-rGO-treated groups exhibiting lower residual biofilm scores compared to conventional irrigants. Comparable nanoparticle-assisted PDT strategies have been reported, such as polydopamine nanoparticles functionalized with Curcuma longa, which demonstrated effective E. faecalis biofilm disruption comparable to NaOCl [27]. Conversely, Özdemir et al. reported significant biofilm reduction using EDTA and NaOCl combinations [28], while other studies have shown that silver nanoparticles exhibit variable antibacterial and antibiofilm efficacy compared to NaOCl [29].

Overall, these findings highlight the potential of integrating functionalized nanomaterials with PDT and adjunctive activation techniques to overcome the limitations of conventional irrigants. From a clinical perspective, nanoparticle-enhanced PDT combined with sonic activation may represent a feasible adjunct to existing irrigation protocols; however, factors such as delivery feasibility, cost, chairside time, and potential cytotoxicity require careful consideration before clinical translation.

Limitations

This study has several limitations. The in vitro design does not fully replicate the complexity of clinical root canal infections. A single-species biofilm model (E. faecalis) was used, whereas clinical infections are typically polymicrobial. Only a limited number of disinfection protocols and outcome measures were evaluated, which may not reflect all clinically relevant scenarios. Although SEM provided high-resolution structural visualization of biofilm coverage, the semi-quantitative scoring approach remains inherently subjective despite evaluator calibration and blinding. Furthermore, SEM-based analysis does not distinguish between viable bacteria, non-viable remnants, or metabolically inactive cells. In addition, intratubular bacterial penetration was not assessed. The occurrence of zero standard deviation in certain scoring outcomes likely reflects uniform biofilm responses rather than observer bias and represents a recognized limitation of ordinal SEM-based assessment methods. These factors should be considered when interpreting the findings.

Conclusion

This study demonstrates that all evaluated disinfection protocols were effective in reducing Enterococcus faecalis biofilm on root dentin compared with the untreated control. Conventional sodium hypochlorite (NaOCl), with or without sonic activation, showed moderate antibiofilm efficacy. Photodynamic therapy (PDT) using methylene blue further enhanced biofilm disruption, with additional improvement observed when combined with sonic activation. The most pronounced biofilm reduction was achieved using methylene blue functionalized with reduced graphene oxide (MB-rGO) in combination with sonic activation and laser irradiation, suggesting that nanoparticle functionalization and adjunctive activation strategies may enhance the antibiofilm potential of PDT. Further in vivo and clinical studies are required to confirm safety, feasibility, and translational applicability of this approach in routine endodontic practice.

Author contributions

Conceptualization, R.J. and M.S.H.; methodology, R.J. and M.S.H.; software, R.J. and M.S.H.; validation, R.J., M.S.H., M.A.A. and S.A.Q.; formal analysis, R.J. and M.S.H.; investigation, R.J..; resources, R.J. and M.S.H.; data curation R.J., M.S.H. and M.A.A..; writing-original draft preparation, R.J.; writing-review and editing, M.S.H., M.A.A. and S.A.Q.; visualization, M.S.H., M.A.A. and S.A.Q.; supervision, M.S.H. and M.A.A; project administration, M.S.H, and R.J.; funding acquisition, R.J. All authors reviewed the manuscript.

Funding

This work was supported by the Deanship of Scientific Research, Vice Presidency for Graduate Studies and Scientific Research, King Faisal University, Saudi Arabia [Grant No. KFU 261478].

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval was obtained from the Research Ethics Committee, King Faisal University, Al-Ahsa, Saudi Arabia (KFU-REC-2024-MAR-ETHICS2072). All experiments were performed following relevant guidelines and regulations regarding the experiments on the use of human tissue samples. Informed consent to participate was obtained from all of the participants (owners of the extracted teeth) in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.