Basic Research—Technology
In Vitro Particle Image Velocity Measurements in a Model Root Canal: Flow around a Polymer Rotary Finishing File Jon D. Koch, PhD,* Nicholas A. Smith, BSME,* Daniel Garces, BSME,* Luyang Gao, BSME,* and F. Kris Olsen, DDS, MS† Abstract Introduction: Root canal irrigation is vital to thorough debridement and disinfection, but the mechanisms that contribute to its effectiveness are complex and uncertain. Traditionally, studies in this area have relied on before-and-after static comparisons to assess effectiveness, but new in situ tools are being developed to provide real-time assessments of irrigation. The aim in this work was to measure a cross section of the velocity field in the fluid flow around a polymer rotary finishing file in a model root canal. Methods: Fluorescent microparticles were seeded into an optically accessible acrylic root canal model. A polymer rotary finishing file was activated in a static position. After laser excitation, fluorescence from the microparticles was imaged onto a frame-transfer camera. Two consecutive images were cross-correlated to provide a measurement of a projected, 2-dimensional velocity field. Results: The method reveals that fluid velocities can be much higher than the velocity of the file because of the shape of the file. Furthermore, these high velocities are in the axial direction of the canal rather than only in the direct of motion of the file. Conclusions: Particle image velocimetry indicates that fluid velocities induced by the rotating file can be much larger than the speed of the file. Particle image velocimetry can provide qualitative insight and quantitative measurements that may be useful for validating computational fluid dynamic models and connecting clinical observations to physical explanations in dental research. (J Endod 2014;40:412–416)
Key Words Flow visualization, PIV, polymer rotary finishing file, velocimetry
From the *Marquette University College of Engineering and †Marquette University School of Dentistry, Milwaukee, Wisconsin. Address requests for reprints to Dr Jon Koch, Department of Mechanical Engineering, 1515 West Wisconsin Avenue, Milwaukee, WI 53233. E-mail address: [email protected] 0099-2399/$ - see front matter Copyright ª 2014 American Association of Endodontists. http://dx.doi.org/10.1016/j.joen.2013.10.038
B
ecause apical leakage of bacteria or bacteria toxins from a root canal system is one of the main causes of periradicular disease, eliminating as much canal wall debris as possible, including tissue, bacteria, and biofilm, is important in nonsurgical endodontic treatment success (1). A study by Ricucci et al (2) reported that intraradicular infections were the primary cause of endodontic treatment failure. After instrumentation, many advocate activation of the final irrigant, commonly by sonic or ultrasonics, to achieve better canal debridement (3–5). Effective irrigation is thought to come from a combination of fluid dynamic or mechanical forces and chemistry. The fluid motion not only cleans the canal walls via hydrodynamic shear stress, but it also provides continuous bulk transport—transport of fresh chemical reagent to the wall and transport of reaction products and dislodged debris from the canal. Several studies have investigated specific factors that are thought to govern the effectiveness of irrigation such as the geometry of the canal, the volume and type of irrigant, and the size, shape, and type of activation instrument (4,6–10). In most cases, the results are explained in terms of how the instrument induces flow and transport of the fluid and how the flow interacts with the canal wall. Because such explanations nearly always revolve around a discussion of the velocity field within the fluid, the velocity field itself has become a key point of focus for endodontic research. The study of the velocity field has evolved from early work that focused on qualitative flow visualization (11–14) to more recent work involving quantitative velocimetry (15–17). The velocity field is also an important benchmark that may be directly compared with computational fluid dynamics (CFD) codes that are increasingly applied to the study of root canal irrigation methods (17–21). The F-file (PlasticEndo, LLC, Lincolnshire, IL) is a single-use rotary finishing file made of nontoxic plastic with imbedded diamond abrasives. The file removes dentinal wall debris and transports a chemical reagent such as sodium hypochlorite, EDTA, or chlorhexidine into areas of the canal that prior instrumentation did not reach, without further enlarging the canal. An in vitro study by Townsend and Maki (22) demonstrated that there was no significant difference (P > .05) in the effectiveness of the F-file to remove biofilm-forming bacteria from plastic simulated canals as compared with the EndoActivator, ultrasonic and sonic activation. All of these were significantly more effective (P < .05) in mechanically eliminating the established bacteria from the canal walls than the EndoVac or needle irrigation only. An in vitro study by Klyn et al (23) found that there were no significant differences in canal or isthmus cleanliness when the F-file, EndoActivator, and ultrasonic irrigation were compared. These clinical studies have indicated that the F-file performance is on par with other methods, but to date, flow patterns around the F-file have not been reported in the literature. Thus, the purposes of this study were to investigate the flow around the F-file and more generally to demonstrate quantitative fluid velocity measurements via a fluorescent particle image velocimetry (PIV) technique in an in vitro endodontic application.
Materials and Methods The polymer rotary finishing file used in these experiments was a size 20 F-file (Plastic Endo, Buffalo Grove, IL; 25 mm long, 0.20 mm in diameter at the tip, with a 0.04 taper). The file was driven by an AEU-20T Endodontic System (Dentsply, Tulsa, OK) with an Aseptico (Woodinville, WA) motor and a 1:8 ratio Anthogyr (Sallanches,
412
Koch et al.
JOE — Volume 40, Number 3, March 2014
Basic Research—Technology France) E-type handpiece at a setting of 600 rpm, but measurements based on acquired images indicated that the actual tool speed was 683 rpm. An acrylic resin block (Sybron Endo, Orange, CA) with a 30 simulated root canal was instrumented to a size 30/0.06 by using ProFile rotary files (Dentsply) in a crown-down technique. The canal was irrigated with distilled water by using a 30-gauge needle (ProRinse; Dentsply, Johnson City, TN) attached to a monoject syringe and then dried with paper points. The prepared block was clamped to a translation stage to control its location. After needle injection of the water/particle mixture as described below, the F-file was inserted into the canal to a depth of 22 mm and activated in the static position. The optical configuration is shown in Figure 1. The light source consisted of a double-pulsed, frequency-doubled YAG laser (New Wave Research, Sunnyvale, CA; Solo PIV) with 10 mJ/pulse output at 532 nm. This green laser light was expanded through a lens, steered by a dichroic reflector (Semrock RazorEdge, Lake Forest, IL; 532 nm), and focused through a microscope objective onto an acrylic root canal model that was filled with a mixture of dye-coated microparticles (Fluka 74097, 6-mm melamine resin particles coated with rhodamine B) and distilled water. The melamine particles are somewhat more dense than the surrounding fluid (specific gravity = 1.5), but their small size ensured that the particles would faithfully follow the flow field (Stokes number of 0.05 is based on a characteristic velocity of 150 mm/s). The particles absorbed the green laser light from the YAG and subsequently fluoresced in the orange region of the spectrum. Fluorescence was captured by using the objective lens and a Navitar (Rochester, NY) 12 zoom lens as an ocular and focused onto a frame-transfer unintensified CCD camera (LaVision Imager Pro X 2M; LaVision, Goettingen, Germany) at a magnification of 5.7. Scattered green light from the surfaces of the canal model and the file was partially rejected via the dichroic filter, but complete rejection was accomplished with the use of an orange bandpass filter (Semrock FF01 583-nm bandpass, 22-nm width), thus ensuring low background interference and an ability to focus only on the motion of the particle-laden fluid in the canal. After calibrating the spatial scales of the image by using a reticle, sets of 2 images were collected by using the double-pulsed YAG and the dualframe capability of the camera. The frames were separated in time by 600 microseconds, and pairs of images could be acquired at 10.5 Hz (once every 1.083 revolutions). Sets of image pairs were acquired and cross-correlated by using DaVis 7.1 (LaVision) to provide velocity field measurements. By using one-fourth of the initial interrogation
window size as a limit, the maximum measurable velocity component was 141 mm/s. The minimum measurable velocity component based on a 1-pixel displacement created a minimum discernible velocity of about 2 mm/s. Hence, regions of the flow with components slower than 2 mm/s should be regarded as essentially motionless in the projected view. The depth of field was 42 mm, limited by the collection optics. By using this depth of field as a measure of the possible out-of-plane displacement that could appear to be displacement within the measured plane (the perspective effect), the possible bias was found to be 1 pixel. Because this potential bias was the same as the minimum measurable velocity, it did not introduce significant bias to the measurement of in-plane velocities. The focal plane was adjusted by using the translation stage to be close to the edge of the file but between the file and the wall of the canal in the direction of the objective lens to maintain a field of view that was completely within the fluid. Because of imperfect index matching between the water (refractive index, n = 1.33) and the acrylic (n = 1.49) model, slight curvature of the object plane is expected but was not quantified in detail for this work. The single-camera PIV technique used here provides a measure of 2 velocity components; those are the orthogonal displacements measurable within the field of view, but in actuality, the flow is 3dimensional. On the basis of the rotational motion of the file, one might expect the predominant flow direction to be in the direction perpendicular to the image plane. As explained below, however, this was not the case. Nonetheless, the out-of-plane motion that was not measured should be kept in mind when interpreting the results. Threedimensional information can be obtained with the use of a second camera to observe the imaged volume from a different angle, thus enabling stereo-PIV, but that was beyond the scope of this investigation.
Results Figure 2A shows how the file fit within the curved canal and provides a visual context for obtaining PIV measurements in the model canal. Spatial resolution is critical to PIV, and the aspect ratio of the entire canal is too large to enable accurate velocity measurements of its entirety in a single image. As a result, only a portion of the canal, a small section roughly 2 mm in height, in the lower mid-root section is shown to provide example PIV results. This section is just above the point where the file makes contact with the bend in the canal wall as indicated in Figure 2A. In Figure 2B, the bright spots are the fluorescent
Figure 1. Schematic of fluorescent PIV measurements in an optically accessible root canal model.
JOE — Volume 40, Number 3, March 2014
PIV of Flow around a Polymer Rotary File
413
Basic Research—Technology
Figure 2. (A) Image of entire canal stitched from several photographs. (B) A single image of the fluorescent particles in the lower mid-root region, with the boundaries of the canal wall and file outlined. (C) Two-dimensional velocity field from cross-correlation of an image pair.
particle signals and are visible throughout the canal. The edges of the file have been outlined to highlight its location behind the focal plane. It is important to note that the F-file is significantly closer to the right wall than the left within the canal. Two images similar to Figure 2B were correlated to determine local displacements of the fluorescent particles. Because of the known time between the images, the velocity field is deduced as shown in Figure 2C. In this image, the flow near the center of the image moves with the rotating tool (from right to left) at a speed that is close to the nominal file surface speed of 16 mm/s. The peak velocity in this image, about 50 mm/s, is actually parallel to the central axis of the canal. The flow was highly unsteady, so to illustrate the types of velocity fields observed, measured fields are also shown in Figure 3A and B. These serve to highlight the observation that sometimes the flow in the image plane was in the opposite direction of file rotation (Fig. 3A), and frequently, the flow was found to have a strong axial component (Fig. 3B). The average speed from 200 images is shown in Figure 3C, and the maximum measured speed at each location is shown in Figure 3D.
Discussion For many images, the projected velocities are consistent with flow that is rotating circumferentially with the F-file. Flow in the measured plane is slow along the edges of the canal and somewhat faster in the middle of the image, as would be expected because of the similarity to the well-studied canonical Taylor-Couette flow (24). However, much of the fluid motion consists of oscillating flow axially along the canal. In fact, the largest velocities were generally not in the direction of file rotation but in the orthogonal direction, and at times, the bulk flow was observed to move against the flow of the file, as exhibited in Figure 3A. The strong axial flow was likely caused by 2 forwardfacing steps that acted as flutes on opposite sides of the F-file. The interaction between the forward-facing steps and the wall of the canal created a helical gap that opened and closed twice per revolution, 414
Koch et al.
and the fluid in the gap responded by moving axially along the canal wall. At the location measured in Figure 3, the tool velocity is approximately 16 mm/s at the outer edge of the F-file as calculated from the diameter and rotational speed. It is clear from the average and maximum images that a significant portion of the fluid moves much faster than the surface of the file, at times in excess of 100 mm/s, as indicated by Figure 3D. It is also clear from this experiment that with the F-file in static position, there are regions of consistently low velocity, as can be observed on the lower left-hand portion of the canal in Figure 3C and D. Such regions may exhibit less chemical and debris transport and may correlate to regions of remaining smear layers and remaining debris, but this question is left for a more comprehensive study. CFD codes have recently been applied to gain insights into the fluid dynamics and transport in endodontic treatments (16–20). The results of such codes can be strongly dependent on boundary conditions and submodels used within the code. For example, the velocity field during needle injection is strongly dependent on the volumetric injection rate. The flow around a moving file depends on the degree of deformation and displacement of the file. The PIV techniques demonstrated here may provide a directly observable benchmark for such simulations, thereby ensuring realistic boundary conditions and enhancing understanding of the flow field and our ability to model its complexities. Such experiments could also enable direct quantitative comparison of different irrigation systems.
Conclusions We have demonstrated a fluorescence PIV technique for quantitatively measuring the velocity field during in vitro experiments involving activation of root canal irrigation. A polymer rotary finishing file was observed to induce flows that were much faster than the speed of the file. Furthermore, these large speeds were predominantly detected in the axial direction, orthogonal to the motion of the file. PIV is an experimental tool that may be valuable to researchers in root canal irrigation. It can provide qualitative insight and quantitative JOE — Volume 40, Number 3, March 2014
Basic Research—Technology
Figure 3. Example images of instantaneous velocity fields that demonstrate (A) flow against the direction of motion of the file (the file rotates right to left; the flow is clearly left to right) and (B) strong axial flow upwards. The average speed from 200 PIV measurements is shown in (C), and the maximum speed from the same group of measurements is shown in (D). The speeds tend to be much higher on the side where the gap between the file and the wall is narrow.
measurements that may be useful for understanding the complex fluid dynamics and transport processes in root canal irrigation and for validating CFD models in dental research.
Acknowledgments The authors deny any conflicts of interest related to this study.
References 1. Lin LM, Skribner JE, Gaengler P. Factors associated with endodontic treatment failures. J Endod 1992;18:625–7. 2. Ricucci D, Siqueira JF Jr, Bate AL, Pitt Ford TR. Histologic investigation of root canaltreated teeth with apical periodontitis: a retrospective study from twenty-four patients. J Endod 2009;35:493–502.
JOE — Volume 40, Number 3, March 2014
3. Averbach RE, Kleier DJ. Clinical update on root canal disinfection. Compend Contin Educ Dent 2006;27. 284, 286–9. 4. Cheung GS, Stock CJ. In vitro cleaning ability of root canal irrigants with and without endosonics. Int Endod J 1993;26:334–43. 5. Haapasalo M, Shen Y, Qian W, Gao Y. Irrigation in endodontics. Dent Clin North Am 2010;54:291–312. 6. Stamos DE, Sadeghi EM, Haasch GC, Gerstein H. An in vitro comparison study to quantitate the debridement ability of hand, sonic, and ultrasonic instrumentation. J Endod 1987;13:434. 7. Munley PJ, Goodell GG. Comparison of passive ultrasonic debridement between fluted and nonfluted instruments in root canals. J Endod 2007;33:578–80. 8. Curtis TO, Sedgley CM. Comparison of a continuous ultrasonic irrigation device and conventional needle irrigation in the removal of root canal debris. J Endod 2012;38:1261. 9. De Gregorio C, Estevez R, Cisneros R, et al. Efficacy of different irrigation and activation systems on the penetration of sodium hypochlorite into simulated lateral canals and up to working length: an in vitro study. J Endod 2010;36:1216.
PIV of Flow around a Polymer Rotary File
415
Basic Research—Technology 10. Jiang LM, Verhaagen B, Versluis M, van der Sluis LW. Evaluation of a sonic device designed to activate irrigant in the root canal. J Endod 2010;36:143–6. 11. Lumley PJ, Walmsley AD, Laird WR. Streaming patterns produced around endosonic files. Int Endod J 1991;24:290–7. 12. Ahmad M, Roy RA, Kamarudin AG. Observations of acoustic streaming fields around an oscillating ultrasonic file. Endod Dent Traumatol 1992;8:189–94. 13. Ahmad M, Roy RA, Kamarudin AG, Safar M. The vibratory pattern of ultrasonic files driven piezoelectrically. Int Endod J 1993;26:120–4. 14. Roy RA, Ahmad M, Crum LA. Physical mechanisms governing the hydrodynamic response of an oscillating ultrasonic file. Int Endod J 1994;27:197–207. 15. Jiang LM, Verhaagen B, Versluis M, van der Sluis LW. Influence of the oscillation direction of an ultrasonic file on the cleaning efficacy of passive ultrasonic irrigation. J Endod 2010;36:1372–6. 16. Boutsioukis C, Verhaagen B, Versluis M, et al. Irrigant flow in the root canal: experimental validation of an unsteady computational fluid dynamics model using highspeed imaging. Int Endod J 2010;43:393–403. 17. Verhaagen B, Boutsioukis C, Heijnen GL, et al. Role of the confinement of a root canal on jet impingement during endodontic irrigation. Exp Fluids 2012;53: 1841–53.
416
Koch et al.
18. Boutsioukis C, Lambrianidis T, Verhaagen B, et al. The effect of needle-insertion depth on the irrigant flow in the root canal: evaluation using an unsteady computational fluid dynamics model. J Endod 2010;36:1664. 19. Boutsioukis C, Verhaagen B, Versluis M, et al. Evaluation of irrigant flow in the root canal using different needle types by an unsteady computational fluid dynamics model. J Endod 2010;36:875. 20. Boutsioukis C, Lambrianidis T, Kastrinakis E. Irrigant flow within a prepared root canal using various flow rates: a computational fluid dynamics study. Int Endod J 2009;42:144–55. 21. Gao Y, Haapasalo M, Shen Y, et al. Development and validation of a three-dimensional computational fluid dynamics model of root canal irrigation. J Endod 2009;35:1282. 22. Townsend C, Maki J. An in vitro comparison of new irrigation and agitation techniques to ultrasonic agitation in removing bacteria from a simulated root canal. J Endod 2009;35:1040–3. 23. Klyn SL, Kirkpatrick TC, Rutledge RE. In vitro comparisons of debris removal of the EndoActivator system, the F file, ultrasonic irrigation, and NaOCl irrigation alone after hand-rotary instrumentation in human mandibular molars. J Endod 2010;36:1367–71. 24. Bird RB, Stewart WE, Lightfoot EN. Transport Phenomena. Revised, 2nd ed. New York: J. Wiley; 2007.
JOE — Volume 40, Number 3, March 2014
In Vitro Particle Image Velocity Measurements in a Model Root Canal: Flow around a Polymer Rotary Finishing File Jon D. Koch, PhD,* Nicholas A. Smith, BSME,* Daniel Garces, BSME,* Luyang Gao, BSME,* and F. Kris Olsen, DDS, MS† Abstract Introduction: Root canal irrigation is vital to thorough debridement and disinfection, but the mechanisms that contribute to its effectiveness are complex and uncertain. Traditionally, studies in this area have relied on before-and-after static comparisons to assess effectiveness, but new in situ tools are being developed to provide real-time assessments of irrigation. The aim in this work was to measure a cross section of the velocity field in the fluid flow around a polymer rotary finishing file in a model root canal. Methods: Fluorescent microparticles were seeded into an optically accessible acrylic root canal model. A polymer rotary finishing file was activated in a static position. After laser excitation, fluorescence from the microparticles was imaged onto a frame-transfer camera. Two consecutive images were cross-correlated to provide a measurement of a projected, 2-dimensional velocity field. Results: The method reveals that fluid velocities can be much higher than the velocity of the file because of the shape of the file. Furthermore, these high velocities are in the axial direction of the canal rather than only in the direct of motion of the file. Conclusions: Particle image velocimetry indicates that fluid velocities induced by the rotating file can be much larger than the speed of the file. Particle image velocimetry can provide qualitative insight and quantitative measurements that may be useful for validating computational fluid dynamic models and connecting clinical observations to physical explanations in dental research. (J Endod 2014;40:412–416)
Key Words Flow visualization, PIV, polymer rotary finishing file, velocimetry
From the *Marquette University College of Engineering and †Marquette University School of Dentistry, Milwaukee, Wisconsin. Address requests for reprints to Dr Jon Koch, Department of Mechanical Engineering, 1515 West Wisconsin Avenue, Milwaukee, WI 53233. E-mail address: [email protected] 0099-2399/$ - see front matter Copyright ª 2014 American Association of Endodontists. http://dx.doi.org/10.1016/j.joen.2013.10.038
B
ecause apical leakage of bacteria or bacteria toxins from a root canal system is one of the main causes of periradicular disease, eliminating as much canal wall debris as possible, including tissue, bacteria, and biofilm, is important in nonsurgical endodontic treatment success (1). A study by Ricucci et al (2) reported that intraradicular infections were the primary cause of endodontic treatment failure. After instrumentation, many advocate activation of the final irrigant, commonly by sonic or ultrasonics, to achieve better canal debridement (3–5). Effective irrigation is thought to come from a combination of fluid dynamic or mechanical forces and chemistry. The fluid motion not only cleans the canal walls via hydrodynamic shear stress, but it also provides continuous bulk transport—transport of fresh chemical reagent to the wall and transport of reaction products and dislodged debris from the canal. Several studies have investigated specific factors that are thought to govern the effectiveness of irrigation such as the geometry of the canal, the volume and type of irrigant, and the size, shape, and type of activation instrument (4,6–10). In most cases, the results are explained in terms of how the instrument induces flow and transport of the fluid and how the flow interacts with the canal wall. Because such explanations nearly always revolve around a discussion of the velocity field within the fluid, the velocity field itself has become a key point of focus for endodontic research. The study of the velocity field has evolved from early work that focused on qualitative flow visualization (11–14) to more recent work involving quantitative velocimetry (15–17). The velocity field is also an important benchmark that may be directly compared with computational fluid dynamics (CFD) codes that are increasingly applied to the study of root canal irrigation methods (17–21). The F-file (PlasticEndo, LLC, Lincolnshire, IL) is a single-use rotary finishing file made of nontoxic plastic with imbedded diamond abrasives. The file removes dentinal wall debris and transports a chemical reagent such as sodium hypochlorite, EDTA, or chlorhexidine into areas of the canal that prior instrumentation did not reach, without further enlarging the canal. An in vitro study by Townsend and Maki (22) demonstrated that there was no significant difference (P > .05) in the effectiveness of the F-file to remove biofilm-forming bacteria from plastic simulated canals as compared with the EndoActivator, ultrasonic and sonic activation. All of these were significantly more effective (P < .05) in mechanically eliminating the established bacteria from the canal walls than the EndoVac or needle irrigation only. An in vitro study by Klyn et al (23) found that there were no significant differences in canal or isthmus cleanliness when the F-file, EndoActivator, and ultrasonic irrigation were compared. These clinical studies have indicated that the F-file performance is on par with other methods, but to date, flow patterns around the F-file have not been reported in the literature. Thus, the purposes of this study were to investigate the flow around the F-file and more generally to demonstrate quantitative fluid velocity measurements via a fluorescent particle image velocimetry (PIV) technique in an in vitro endodontic application.
Materials and Methods The polymer rotary finishing file used in these experiments was a size 20 F-file (Plastic Endo, Buffalo Grove, IL; 25 mm long, 0.20 mm in diameter at the tip, with a 0.04 taper). The file was driven by an AEU-20T Endodontic System (Dentsply, Tulsa, OK) with an Aseptico (Woodinville, WA) motor and a 1:8 ratio Anthogyr (Sallanches,
412
Koch et al.
JOE — Volume 40, Number 3, March 2014
Basic Research—Technology France) E-type handpiece at a setting of 600 rpm, but measurements based on acquired images indicated that the actual tool speed was 683 rpm. An acrylic resin block (Sybron Endo, Orange, CA) with a 30 simulated root canal was instrumented to a size 30/0.06 by using ProFile rotary files (Dentsply) in a crown-down technique. The canal was irrigated with distilled water by using a 30-gauge needle (ProRinse; Dentsply, Johnson City, TN) attached to a monoject syringe and then dried with paper points. The prepared block was clamped to a translation stage to control its location. After needle injection of the water/particle mixture as described below, the F-file was inserted into the canal to a depth of 22 mm and activated in the static position. The optical configuration is shown in Figure 1. The light source consisted of a double-pulsed, frequency-doubled YAG laser (New Wave Research, Sunnyvale, CA; Solo PIV) with 10 mJ/pulse output at 532 nm. This green laser light was expanded through a lens, steered by a dichroic reflector (Semrock RazorEdge, Lake Forest, IL; 532 nm), and focused through a microscope objective onto an acrylic root canal model that was filled with a mixture of dye-coated microparticles (Fluka 74097, 6-mm melamine resin particles coated with rhodamine B) and distilled water. The melamine particles are somewhat more dense than the surrounding fluid (specific gravity = 1.5), but their small size ensured that the particles would faithfully follow the flow field (Stokes number of 0.05 is based on a characteristic velocity of 150 mm/s). The particles absorbed the green laser light from the YAG and subsequently fluoresced in the orange region of the spectrum. Fluorescence was captured by using the objective lens and a Navitar (Rochester, NY) 12 zoom lens as an ocular and focused onto a frame-transfer unintensified CCD camera (LaVision Imager Pro X 2M; LaVision, Goettingen, Germany) at a magnification of 5.7. Scattered green light from the surfaces of the canal model and the file was partially rejected via the dichroic filter, but complete rejection was accomplished with the use of an orange bandpass filter (Semrock FF01 583-nm bandpass, 22-nm width), thus ensuring low background interference and an ability to focus only on the motion of the particle-laden fluid in the canal. After calibrating the spatial scales of the image by using a reticle, sets of 2 images were collected by using the double-pulsed YAG and the dualframe capability of the camera. The frames were separated in time by 600 microseconds, and pairs of images could be acquired at 10.5 Hz (once every 1.083 revolutions). Sets of image pairs were acquired and cross-correlated by using DaVis 7.1 (LaVision) to provide velocity field measurements. By using one-fourth of the initial interrogation
window size as a limit, the maximum measurable velocity component was 141 mm/s. The minimum measurable velocity component based on a 1-pixel displacement created a minimum discernible velocity of about 2 mm/s. Hence, regions of the flow with components slower than 2 mm/s should be regarded as essentially motionless in the projected view. The depth of field was 42 mm, limited by the collection optics. By using this depth of field as a measure of the possible out-of-plane displacement that could appear to be displacement within the measured plane (the perspective effect), the possible bias was found to be 1 pixel. Because this potential bias was the same as the minimum measurable velocity, it did not introduce significant bias to the measurement of in-plane velocities. The focal plane was adjusted by using the translation stage to be close to the edge of the file but between the file and the wall of the canal in the direction of the objective lens to maintain a field of view that was completely within the fluid. Because of imperfect index matching between the water (refractive index, n = 1.33) and the acrylic (n = 1.49) model, slight curvature of the object plane is expected but was not quantified in detail for this work. The single-camera PIV technique used here provides a measure of 2 velocity components; those are the orthogonal displacements measurable within the field of view, but in actuality, the flow is 3dimensional. On the basis of the rotational motion of the file, one might expect the predominant flow direction to be in the direction perpendicular to the image plane. As explained below, however, this was not the case. Nonetheless, the out-of-plane motion that was not measured should be kept in mind when interpreting the results. Threedimensional information can be obtained with the use of a second camera to observe the imaged volume from a different angle, thus enabling stereo-PIV, but that was beyond the scope of this investigation.
Results Figure 2A shows how the file fit within the curved canal and provides a visual context for obtaining PIV measurements in the model canal. Spatial resolution is critical to PIV, and the aspect ratio of the entire canal is too large to enable accurate velocity measurements of its entirety in a single image. As a result, only a portion of the canal, a small section roughly 2 mm in height, in the lower mid-root section is shown to provide example PIV results. This section is just above the point where the file makes contact with the bend in the canal wall as indicated in Figure 2A. In Figure 2B, the bright spots are the fluorescent
Figure 1. Schematic of fluorescent PIV measurements in an optically accessible root canal model.
JOE — Volume 40, Number 3, March 2014
PIV of Flow around a Polymer Rotary File
413
Basic Research—Technology
Figure 2. (A) Image of entire canal stitched from several photographs. (B) A single image of the fluorescent particles in the lower mid-root region, with the boundaries of the canal wall and file outlined. (C) Two-dimensional velocity field from cross-correlation of an image pair.
particle signals and are visible throughout the canal. The edges of the file have been outlined to highlight its location behind the focal plane. It is important to note that the F-file is significantly closer to the right wall than the left within the canal. Two images similar to Figure 2B were correlated to determine local displacements of the fluorescent particles. Because of the known time between the images, the velocity field is deduced as shown in Figure 2C. In this image, the flow near the center of the image moves with the rotating tool (from right to left) at a speed that is close to the nominal file surface speed of 16 mm/s. The peak velocity in this image, about 50 mm/s, is actually parallel to the central axis of the canal. The flow was highly unsteady, so to illustrate the types of velocity fields observed, measured fields are also shown in Figure 3A and B. These serve to highlight the observation that sometimes the flow in the image plane was in the opposite direction of file rotation (Fig. 3A), and frequently, the flow was found to have a strong axial component (Fig. 3B). The average speed from 200 images is shown in Figure 3C, and the maximum measured speed at each location is shown in Figure 3D.
Discussion For many images, the projected velocities are consistent with flow that is rotating circumferentially with the F-file. Flow in the measured plane is slow along the edges of the canal and somewhat faster in the middle of the image, as would be expected because of the similarity to the well-studied canonical Taylor-Couette flow (24). However, much of the fluid motion consists of oscillating flow axially along the canal. In fact, the largest velocities were generally not in the direction of file rotation but in the orthogonal direction, and at times, the bulk flow was observed to move against the flow of the file, as exhibited in Figure 3A. The strong axial flow was likely caused by 2 forwardfacing steps that acted as flutes on opposite sides of the F-file. The interaction between the forward-facing steps and the wall of the canal created a helical gap that opened and closed twice per revolution, 414
Koch et al.
and the fluid in the gap responded by moving axially along the canal wall. At the location measured in Figure 3, the tool velocity is approximately 16 mm/s at the outer edge of the F-file as calculated from the diameter and rotational speed. It is clear from the average and maximum images that a significant portion of the fluid moves much faster than the surface of the file, at times in excess of 100 mm/s, as indicated by Figure 3D. It is also clear from this experiment that with the F-file in static position, there are regions of consistently low velocity, as can be observed on the lower left-hand portion of the canal in Figure 3C and D. Such regions may exhibit less chemical and debris transport and may correlate to regions of remaining smear layers and remaining debris, but this question is left for a more comprehensive study. CFD codes have recently been applied to gain insights into the fluid dynamics and transport in endodontic treatments (16–20). The results of such codes can be strongly dependent on boundary conditions and submodels used within the code. For example, the velocity field during needle injection is strongly dependent on the volumetric injection rate. The flow around a moving file depends on the degree of deformation and displacement of the file. The PIV techniques demonstrated here may provide a directly observable benchmark for such simulations, thereby ensuring realistic boundary conditions and enhancing understanding of the flow field and our ability to model its complexities. Such experiments could also enable direct quantitative comparison of different irrigation systems.
Conclusions We have demonstrated a fluorescence PIV technique for quantitatively measuring the velocity field during in vitro experiments involving activation of root canal irrigation. A polymer rotary finishing file was observed to induce flows that were much faster than the speed of the file. Furthermore, these large speeds were predominantly detected in the axial direction, orthogonal to the motion of the file. PIV is an experimental tool that may be valuable to researchers in root canal irrigation. It can provide qualitative insight and quantitative JOE — Volume 40, Number 3, March 2014
Basic Research—Technology
Figure 3. Example images of instantaneous velocity fields that demonstrate (A) flow against the direction of motion of the file (the file rotates right to left; the flow is clearly left to right) and (B) strong axial flow upwards. The average speed from 200 PIV measurements is shown in (C), and the maximum speed from the same group of measurements is shown in (D). The speeds tend to be much higher on the side where the gap between the file and the wall is narrow.
measurements that may be useful for understanding the complex fluid dynamics and transport processes in root canal irrigation and for validating CFD models in dental research.
Acknowledgments The authors deny any conflicts of interest related to this study.
References 1. Lin LM, Skribner JE, Gaengler P. Factors associated with endodontic treatment failures. J Endod 1992;18:625–7. 2. Ricucci D, Siqueira JF Jr, Bate AL, Pitt Ford TR. Histologic investigation of root canaltreated teeth with apical periodontitis: a retrospective study from twenty-four patients. J Endod 2009;35:493–502.
JOE — Volume 40, Number 3, March 2014
3. Averbach RE, Kleier DJ. Clinical update on root canal disinfection. Compend Contin Educ Dent 2006;27. 284, 286–9. 4. Cheung GS, Stock CJ. In vitro cleaning ability of root canal irrigants with and without endosonics. Int Endod J 1993;26:334–43. 5. Haapasalo M, Shen Y, Qian W, Gao Y. Irrigation in endodontics. Dent Clin North Am 2010;54:291–312. 6. Stamos DE, Sadeghi EM, Haasch GC, Gerstein H. An in vitro comparison study to quantitate the debridement ability of hand, sonic, and ultrasonic instrumentation. J Endod 1987;13:434. 7. Munley PJ, Goodell GG. Comparison of passive ultrasonic debridement between fluted and nonfluted instruments in root canals. J Endod 2007;33:578–80. 8. Curtis TO, Sedgley CM. Comparison of a continuous ultrasonic irrigation device and conventional needle irrigation in the removal of root canal debris. J Endod 2012;38:1261. 9. De Gregorio C, Estevez R, Cisneros R, et al. Efficacy of different irrigation and activation systems on the penetration of sodium hypochlorite into simulated lateral canals and up to working length: an in vitro study. J Endod 2010;36:1216.
PIV of Flow around a Polymer Rotary File
415
Basic Research—Technology 10. Jiang LM, Verhaagen B, Versluis M, van der Sluis LW. Evaluation of a sonic device designed to activate irrigant in the root canal. J Endod 2010;36:143–6. 11. Lumley PJ, Walmsley AD, Laird WR. Streaming patterns produced around endosonic files. Int Endod J 1991;24:290–7. 12. Ahmad M, Roy RA, Kamarudin AG. Observations of acoustic streaming fields around an oscillating ultrasonic file. Endod Dent Traumatol 1992;8:189–94. 13. Ahmad M, Roy RA, Kamarudin AG, Safar M. The vibratory pattern of ultrasonic files driven piezoelectrically. Int Endod J 1993;26:120–4. 14. Roy RA, Ahmad M, Crum LA. Physical mechanisms governing the hydrodynamic response of an oscillating ultrasonic file. Int Endod J 1994;27:197–207. 15. Jiang LM, Verhaagen B, Versluis M, van der Sluis LW. Influence of the oscillation direction of an ultrasonic file on the cleaning efficacy of passive ultrasonic irrigation. J Endod 2010;36:1372–6. 16. Boutsioukis C, Verhaagen B, Versluis M, et al. Irrigant flow in the root canal: experimental validation of an unsteady computational fluid dynamics model using highspeed imaging. Int Endod J 2010;43:393–403. 17. Verhaagen B, Boutsioukis C, Heijnen GL, et al. Role of the confinement of a root canal on jet impingement during endodontic irrigation. Exp Fluids 2012;53: 1841–53.
416
Koch et al.
18. Boutsioukis C, Lambrianidis T, Verhaagen B, et al. The effect of needle-insertion depth on the irrigant flow in the root canal: evaluation using an unsteady computational fluid dynamics model. J Endod 2010;36:1664. 19. Boutsioukis C, Verhaagen B, Versluis M, et al. Evaluation of irrigant flow in the root canal using different needle types by an unsteady computational fluid dynamics model. J Endod 2010;36:875. 20. Boutsioukis C, Lambrianidis T, Kastrinakis E. Irrigant flow within a prepared root canal using various flow rates: a computational fluid dynamics study. Int Endod J 2009;42:144–55. 21. Gao Y, Haapasalo M, Shen Y, et al. Development and validation of a three-dimensional computational fluid dynamics model of root canal irrigation. J Endod 2009;35:1282. 22. Townsend C, Maki J. An in vitro comparison of new irrigation and agitation techniques to ultrasonic agitation in removing bacteria from a simulated root canal. J Endod 2009;35:1040–3. 23. Klyn SL, Kirkpatrick TC, Rutledge RE. In vitro comparisons of debris removal of the EndoActivator system, the F file, ultrasonic irrigation, and NaOCl irrigation alone after hand-rotary instrumentation in human mandibular molars. J Endod 2010;36:1367–71. 24. Bird RB, Stewart WE, Lightfoot EN. Transport Phenomena. Revised, 2nd ed. New York: J. Wiley; 2007.
JOE — Volume 40, Number 3, March 2014
