Suppression of cross-infection in actual dental office using viscoelastic polyacrylate solution as a handpiece coolant Free

Cross-infection usually refers to the transmission of harmful microorganisms, such as bacteria and viruses. The spread of infection can occur between people (e.g., patients, dentists, dental hygienists, dental assistants, and other healthcare workers), or between a person and dental equipment.^1,2 Thus, medical professionals should ensure equipment safety and a clean environment. Cross-infection can cause complications such as diarrhea, sepsis, pneumonia, meningitis, dehydration, and even death. Among cross-infectious diseases, some diseases, such as Severe Acute Respiratory Syndrome (SARS), influenza, measles, tuberculosis, hepatitis, and HIV/aids, can be transmitted through aerosols.² The COVID-19 pandemic prevalent worldwide is also an infectious disease that can be transmitted through aerosols.³ Recently, the prevalence of COVID-19 has greatly increased interest in cross-infection during dental procedures.

In dentistry, the probability of cross-infection is higher than in daily life for two reasons. First, dental treatment involves close contact between the patient and medical staff. During dental procedures, medical staff should be positioned close to a patient's oral cavity. This situation rises the possibility of cross-infection via bodily fluids such as saliva and sweat transferred directly from an infected patient to medical staff. Second, bodily fluid in a patient's mouth, including harmful microorganisms, can be scattered into surroundings owing to its aerosolization by dental equipment such as ultrasonic scalers, high-speed handpieces, and triple syringes. Since dental treatment is usually performed in a closed environment, as shown in Fig. 1(a), the aerosol can become airborne and remain in a confined space with a high risk of spreading infectious diseases. Among dental equipment, the high-speed handpiece is the most aerosol-generating machine.⁴ 

Wearing personal protective equipment (PPE) is one option to prevent aerosol transmission by physically blocking the path of aerosols [cf. Fig. 1(b)]. However, it is practically difficult because it is costly and hard to use in urgent situations, especially for a long time.⁵ In addition, complete protection from scattered aerosols and droplets is almost impossible. Aerosols can be airborne in air (at an altitude of 2 m) for several hours. Depending on environmental conditions, the floating time might be longer.⁶ In this context, preventive measures are required to prevent potential mass infection by microorganisms in dentistry. Recently, several studies on this topic have been conducted. Sergis et al. have achieved a 60-fold reduction in aerosols by controlling the rotational speed of the handpiece.⁷ Koch and Graetz have found that the use of high-vacuum suction devices can effectively reduce aerosol generation during handpiece operation.⁸ Izzetti et al. have demonstrated that rinsing a patient's mouth with povidone solution before dental treatment can significantly reduce the amount of bacteria.⁹ 

A completely different approach has recently been developed to suppress aerosol generation in dentistry. This can fundamentally prevent the generation of airborne aerosols by controlling the properties of a coolant. The addition of high molecular weight polymers to traditional coolant (water) can change rheological properties (e.g., viscosity and especially viscoelasticity) of a liquid. Plog et al. have used polyacrylic acid (PAA) to introduce viscoelastic properties of a coolant and confirmed that aerosol generation is suppressed in the vicinity of an ultrasonic scaler and a dental handpiece.⁶ In our previous study, we have confirmed the aerosol suppression effect according to the concentration of PAA at a near and far distances from a dental handpiece.¹⁰ In addition, it was confirmed that the addition of the polymer did not significantly affect cooling performance of the coolant.

As a result, previous studies demonstrated that a relatively dilute aqueous polymer solution used as a coolant can fundamentally preclude aerosol generation from a handpiece. However, this finding is supported by experiments only conducted in a fully controlled environment in a laboratory. In addition, since the experiments have not been conducted with microorganisms, it is necessary to assume that microorganisms can be contained in aerosols and spread. The entrainment and transport in airborne droplets were demonstrated only using model aluminum nanoparticles of the COVID-19-virus size (∼100 nm) in Ref. 11. Therefore, it is necessary to quantitatively confirm how many microorganisms contained in the coolant could be spread in an actual dental environment and how much this spread can be suppressed by adding a polymer. In addition, there are issues such as safety and sour taste since PAA is an acidic substance. In this study, an experiment was conducted to identify aerosol generation near a dental handpiece using a high-speed camera with two polymer materials added: poly (sodium acrylate) denoted as NaPAA and polyacrylic acid denoted as PAA [Fig. 1(c)]. NaPAA is a type of polyacrylic acid that has been neutralized with sodium hydroxide. Furthermore, NaPAA is known as an edible substance approved by the FDA (U.S. Food and Drug Administration), which is advantageous for commercialization in terms of human body compatibility. On the other hand, although PAA is not on the FDA's list of edible substances, it is well-known as a highly viscoelastic material that is effective in suppressing aerosols.¹² As a result, as shown in Fig. 1(c), viscous and viscoelastic PAA was more effective in suppressing aerosol than only viscous NaPAA. Finally, water and PAA aqueous solution containing bacteria were supplied to the handpiece in an actual dental environment and the extent to which bacteria spread within the dental office was quantitatively determined.

Polymers [i.e., polyacrylic acid (PAA, M_w = 450 kDa) and sodium polyacrylate (NaPAA, M_w = 8 kDa, 45 wt. %)] were purchased from Sigma Aldrich (USA). Each polymer was mixed with a conventional coolant (de-ionized water) and stirred at 200 rpm for 24 h. The aqueous polymer solutions were maintained at 27 °C for all experiments.

In this experiment, a Staphylococcus epidermidis strain was chosen to investigate bacteria spread during handpiece operation. The bacterial strain (KCTC 13171) was purchased from Korea Collection for Type Cultures (Republic of Korea). Bacterial suspension for the observation of bacteria spread (Sec. III C) was prepared as follows. First, S. epidermidis strains were pre-cultured in LB-Broth media (BD 244620, Bectron, Dickinson and Company, France) and then incubated in an incubator (HG-TSI100, Hangil Science, Republic of Korea) at 37 °C with shaking (150 rpm) for 24 h. Overnight cultured bacteria suspensions were diluted 1:200 with fresh LB-Broth media and incubated at 37 °C with shaking (150 rpm) for 24 h. Subcultured S. epidermidis suspensions were diluted to OD₆₀₀ of 1 with LB-broth media. All OD₆₀₀ measurements were conducted with a UV-VIS spectrophotometer (OPTIZEN POP, KLAB, Republic of Korea). Diluted suspensions were finally inoculated to OD₆₀₀ of 0.01 with (i) water for the control group and (ii) polymer solutions (i.e., NaPAA 20 wt. % and PAA 2 wt. %) for the experimental groups.

A high-speed handpiece (KT 15-SQD) was provided by Dunamis Dental (Republic of Korea). A syringe pump (Legato 100) was purchased from KD Scientific (USA), wherein a 50 ml syringe (A50, Henke-Sass Wolf, Germany) was installed to supply the coolant at a fixed flow rate of 20 ml min⁻¹. A portable dental unit (HL-2070) was purchased from Hallim Dentech (Republic of Korea), which allowed high-pressure air of 4 bar to be supplied to the handpiece. A diamond bur (8845KR314016, TD2032A, Komet, Germany) was also prepared and mounted on the handpiece. Photographs of the experiments were obtained using a high-speed camera [cf. Fig. 1(c)].

To observe the spread of aerosols due to dental handpiece procedure in an actual dental environment [cf. Fig. 1(a)], a dental handpiece with a set of diamond burs, a syringe pump with a 50 ml syringe, and a portable dental unit were used as described in Sec. II C. The flow rate and pressure of the coolant supplied to the handpiece were 20 ml min⁻¹ and 4 bar, respectively. Artificial teeth (Single tooth model, Dongguan Liyue Model Technology Co., Ltd., China) were inserted into a phantom model [Oral training model, Dongguan Liyue Model Technology Co., Ltd., China; cf. Fig. 1(b)] and used in this experiment. This experiment was performed in an actual dental operating room (340 cm × 340 cm × 260 cm) in Anam Hospital of Korea University after obtaining permission from the College of Dentistry of Korea University. The phantom model was fixed to a dental chair. A professional dentist (Prof. Song) performed a dental procedure on the teeth (mandibular right first premolars) attached to the phantom model [Fig. 1(b)]. The dentist reproduced the situation of crown preparation on the mandibular first molar using a high-speed handpiece with a maximum 380 000 rpm air turbine. A tapered chamfer bur (FG 6856 014, Komet, Germany) was attached to the high-speed handpiece and applied to the buccal, lingual, and occlusal surfaces of the teeth for 4 min. Water and a viscoelastic liquid (e.g., 2 wt. % PAA solution) were used as coolants. Bacteria (S. epidermdis, KCTC 13171) suspensions were added into each coolant at a concentration as described in Sec. II B. To check the spread of bacteria entrained in the coolant, agar plates (BD 244520, Becton, Dickinson and Company, France) were installed at distances of 45 cm on the floor of the operating room. Media were installed on each wall at a height of 45 cm from the ground and at intervals of 45 cm. Media were also attached to foreheads and chests of the dentist and the assistant and LED (Light Emitting Diode) light. Dental procedures were performed three times for each case. At the end of each experiment, the operating room was left for 30 min for bacteria to sufficiently spread and all the agar plates were collected. The room was then sterilized using a nebulizer containing 70% ethanol. The collected plates were incubated at 37 °C for 24 h, and the number of colonies grown on the solid medium was counted to determine the extent of bacterial transmission.

Physical properties affecting fluid aerosolization include density ρ, surface tension σ, shear viscosity μ, and viscoelasticity characterized by the relaxation time θ.^6,10,13 Figure 2 presents the measured physical properties of polymer solutions (NaPAA and PAA) to be used in the present work as handpiece coolants. The detailed method of the polymer characterization is presented in Sec. S1 in the supplementary material.

Figure 2(a) presents density variation with increasing polymer concentration. Regardless of the type of the polymer, the density was linearly proportional to the concentration. As the concentration of NaPAA increased from 10 to 20 wt. %, the solution density changed from 1062 to 1137 kg m⁻³, i.e., increased by 7.5 kg m⁻³ for every 1 wt. % of NaPAA added. As the concentration of PAA increased from 1 to 5 wt. %, the solution density increased from 1046 to 1072 kg m⁻³, i.e., revealed an increase in 6.5 kg m⁻³ for every 1 wt. % of PAA added. Although the increase in density with concentration is higher for NaPAA than PAA, the PAA solution possesses a higher density than NaPAA even at a lower mass concentration which seems to be influenced by the molecular weights of the polymers¹⁴ (compare the molecular weights in Sec. II A).

Figure 2(b) shows the variation of surface tension of NaPAA and PAA solutions when the polymer concentration is increased. In the case of NaPAA, the surface tension hardly changed with the concentration, with σ change within 5.5% when the concentration of NaPAA was increased from 0 to 30 wt. %. On the other hand, in the case of PAA, the solution surface tension decreased significantly when the polymer concentration was increased. In particular, when the PAA concentration was increased from 0 to 5 wt. %, the surface tension decreased by 43%. However, in the concentration range of the PAA solution used in this experiment (i.e., at 1–2 wt. %), there was only a 14% difference in σ.

Figure 2(c) illustrates the variation of the shear viscosity with concentrations of polyacrylate (NaPAA or PAA) in aqueous solutions. The NaPAA solutions revealed the shear viscosities practically independent of the shear rate, which is a first (but not an ultimate!) hint of Newtonian behavior. On the other hand, both PAA solutions reveal shear-thinning behavior. The 1 wt. % PAA solution revealed a higher value of μ than the 10 wt. % NaPAA solution, although its concentration was much lower. The high molecular weight (M_w) of the linear polymers is proportional to the length of their macromolecules. Lengthy PAA coils are more easily entangled with each other in semi-dilute solutions than those of NaPAA, resulting in higher shear viscosity values.¹⁵  Figure 2(c) illustrates that the higher the concentration, the higher is the shear viscosity of NaPAA. In the PAA case, when the shear rate was below 50 s⁻¹, the shear viscosity of the 2 wt. % PAA aqueous solution was higher than that of the 1 wt. % PAA solution. However, when the shear rate was above 50 s⁻¹, the shear viscosity of the 2 wt. % PAA solution became lower than that of the 1 wt. % PAA solution. On the other hand, in the case of NaPAA, there was practically no change in the shear viscosity with the shear rate within the measured concentration range (<20 wt. %). As the shear rate increased from 100 to 400 s⁻¹, the shear viscosity of NaPAA changed within 5 wt. % compared to the value at 100 s⁻¹, meaning that it behaved as a practically Newtonian fluid.¹⁶ It should be emphasized that the lower shear viscosity of the 2 wt. % PAA solution compared to that of the 20 wt. % NaPAA, makes the former much more preferable as a coolant pumped through the pipe in a dental chair from a bottle to the handpiece with a rotating bur.

Figure 2(d) illustrates the results of the observations of self-thinning capillary threads of the NaPAA and PAA aqueous solutions. In particular, the thread of the 20 wt. % NaPAA solution is never uniform throughout the entire self-thinning process. Also, the corresponding results for NaPAA solutions in Fig. 2(e) reveal a practically linear decay of the cross-sectional diameter in the middle of the thread D_thread, which is expected for Newtonian fluids.¹⁷ Accordingly, both simple shear and uniaxial elongational flows of Figs. 2(d) and 2(e), respectively, demonstrate that solutions of NaPAA are Newtonian fluids irrespective of the polymer concentration, which stems from its low molecular weight.

Equation (1) involves time t, the initial diameter of the thread at t = 0 denoted as D₀, and the relaxation time θ which characterizes fluid viscoelasticity.²⁰ The values of θ for the 1 and 2 wt. % PAA solutions were θ = 0.25 and 0.28 ms, respectively. The higher value of the relaxation time θ corresponds to stronger elastic effects.

Additionally, the toxicity of polymer solutions (e.g., the 20 wt. % NaPAA and the 2 wt. % PAA) against bacteria (Staphylococcus epidermidis) was assessed. The detailed experimental method is presented in Sec. S2 in the supplementary material. In the case of the 2 wt. % PAA solution, OD₆₀₀ of the suspension did not change from the initial collected state even after 24 h incubation. In other words, S. epidermidis growth was 97.9% inhibited by the PAA solution. Because PAA contains a large number of acid groups, it has been reported that PAA is active against bacteria, including Staphylococcus strains.²¹ On the other hand, the NaPAA group revealed bacterial growth similar to the control group (2.5% difference). The absence of an acidic group in the molecular structure of NaPAA might have caused a lower anti-bacterial activity than PAA.

First, droplets were called as “aerosol” if their size was less than 50 μm, as “droplet” if the size was greater than 50 μm but less than 100 μm, and as “splatter” encompassing all the sizes (>100 μm) similarly to the definition used in the previous studies.^22,23 Among all these liquid bodies, aerosols generated during dental treatment pose a risk for transmission of infectious diseases. They can contain bacteria or viruses, and being airborne for a long time could transfer them to others, such as a dentist and other patients.

Figure 3 presents snapshots of a rotating bur and its surroundings taken with a high-speed camera with water or polymer solutions used as coolants for a dental handpiece. A detailed description of the experimental setup is given in Sec. II C and Sec. S3 in the supplementary material. Figure 3(a) reveals countless small aerosols, which were generated around the bur when water was used as the coolant for the dental handpiece. Table I contains data related to the number and size- distribution for water disintegration by the rotating bur. More than 1000 aerosols with a diameter of 0–50 μm were generated by water, accounting for 93% of total liquid bodies generated.

Table S1 in the supplementary material lists velocity distribution according to the size of water bodies shed by the rotating bur. In the case of water, the aerosols flew straight at a speed of about 0.2 m s⁻¹ away from the bur (cf. Movie S1 in the supplementary material). The length of the region observed around the bur was ∼5 mm from the bur. Accordingly, the aerosol was observed during the time interval Δt = 5 mm·(0.2 m s⁻¹)⁻¹ = 25 ms. Therefore, the furthest liquid bodies seen in one snapshot in Fig. 3(a) were in flight for 25 ms.

On the other hand, when a polymer solution was used as a dental coolant, the aerosol generation was greatly reduced (Movie S2 in the supplementary material). Figures 3(b) and 3(c) display liquid bodies in flight around the bur when NaPAA solution was used as a coolant. In particular, when the 10 wt. % NaPAA and 20 wt. % NaPAA solutions were used as handpiece coolants, the numbers of aerosols (0–50 μm in size) were 378 and 41, respectively, which corresponds to a decrease by 69% and 97% compared to the number in the case of water. The ratios of the number of aerosols to the total number of liquid bodies resulting from disintegration on the rotating bur were 87% and 37% for the 10 wt. % NaPAA and 20 wt. % NaPAA solutions, respectively. That means that as the polymer concentration increased, the aerosol ratio decreased significantly for these viscous Newtonian solutions. The numbers of droplets (50–100 μm size) were 52 and 36 for the 10 wt. % NaPAA and 20 wt. % NaPAA solutions, respectively, which constitutes a decrease of 44% and 61% compared to the corresponding number in the case of water. The maximum size was 182 μm for the 10 wt. % NaPAA solution and 283 μm for the 20 wt. % NaPAA solution.

Figures 3(d) and 3(e) illustrate the cases where PAA solutions were used as dental handpiece coolants. For the 1 and 2 wt. % of PAA solutions, the numbers of aerosols were 145 and 5, respectively. In other words, 88% and 99.6% of aerosols were suppressed compared to those in the case of water. It should be emphasized that almost all aerosols were suppressed with 2 wt. % PAA solution. A number of generated droplets were 55 and 18 for the 1 and 2 wt. % PAA solutions, respectively, which corresponds to a decrease of 40% and 80%, respectively, compared to the case of water. Compared to the NaPAA solutions, the PAA solutions inhibited more aerosols and droplets even at 10-fold lower concentrations, which is a direct consequence of their viscoelasticity responsible for the suppression of aerosol or drop detachment.⁶ Moreover, the 2 wt. % PAA solution forms long liquid threads attached to the rotating bur and sheds directly underneath. It should also be noted that PAA inhibited aerosols more economically than NaPAA. From Sigma-Aldrich, PAA (M_w ∼ 450 kDa) was $1.81/g and NaPAA (M_w ∼ 5 kDa) was $0.43/g, meaning that the polymer cost of the 20 wt. % NaPAA solution was 2.4 times higher than that of the 2 wt. % PAA solution.

In Table S1 in the supplementary material, the velocity of the aerosol flying away from the bur was about 0.2 m s⁻¹ in all cases. Because the aerosol size is very small, the flow of the surrounding air driven by the bur rotating at a high speed plays a more important role than the physical properties of different liquids affecting the aerosol detachment. Therefore, as shown in Movie S3 in the supplementary material, the fine particles (which appear to be aerosols) drifted spirally near the dental handpiece in the rotational direction of the bur. As the size of the liquid bodies increased, the influence of the airflow around the bur diminished and the liquid velocity had the tendency to increase. This stems from the fact that smaller liquid bodies are more affected by the air drag force than the bigger ones.²⁴ However, big liquid bodies easily change their shape from spherical to irregular in flight, which inevitably increases the drag coefficient and force,²⁵ and could diminish their velocity of motion through air. Accordingly, as shown in Table S1 in the supplementary material, the splatter velocity seemingly decreased above a certain size, which is beneficial for reducing the travel distance of such splatter.

In order to confirm the effectiveness of the viscous and viscoelastic coolants as the inhibitors to the spreading of infectious diseases, an experiment was conducted using bacteria as a tracer in an actual dental environment. Bacteria (S. epidermidis, KCTC 13171) suspensions were added to the coolants before dental procedures. They were identified from agar plates distributed throughout the operating room [Figs. 4(a)–4(d)]. The fact that bacteria released from the handpiece could be identified in an agar plate in the operating room after a dental procedure would imply that contaminants originating from the patient or dental equipment had been transmitted to the location of the plate during the dental procedure. Detailed experiments on bacterial spread are described in Sec. II D.

Figure 4(a) is a snapshot of the dental operating room in Anam Hospital of Korea University. The operating room was 340 cm × 340 cm in size. Ventilation was blocked to determine aerosol propagation in an isolated environment. Note that aerosol propagation might worsen depending on the ventilation environment.²⁶ Circular agar plates were installed on the floor and walls of the operating room at the distances of 45 cm in-between to determine the degree of bacteria spread by aerosols, droplets, and splatters. Movable LED (Light Emitting Diode) lights were installed on the dental chair and positioned ∼50 cm above the dentist's head during dental simulation. Figure 4(b) presents the results related to bacteria distribution after a dental procedure with bacteria originally suspended in coolants: (left) water and (right) the 2 wt. % PAA solution. When water was used as a coolant for the dental handpiece, it was found that the floor was contaminated in the direction above the patient's head. However, the floor in the other direction to the equipment, the dentist, or the assistant was not contaminated because transmission of aerosol, droplets, and splatter was obstructed by obstacles (e.g., the equipment, the dentist, and the assistant). The office wall was more contaminated than the floor, providing evidence of aerosol propagation.²⁷ In the case of droplets and splatters propagation, the closer the distance to the patient, the higher the concentration of bacteria would be. Since the treatment was performed for the right second premolar of the mandible, it was confirmed that bacteria spread more to the patient's right-hand side wall. When the 2 wt. % PAA solution was used, there was little or no transmission of bacteria around the office [the right-hand side figure in Fig. 4(b)]. Because of the bio-toxicity of PAA to S. epidermidis, bacteria might tend to be under-reported. However, the higher level of contamination on the floor than that on the wall indicated that the aerosol was sufficiently suppressed, whereas long filaments of the coolant formed settle down underneath the handpiece. Accordingly, the addition of viscoelastic polymers to the cooling water to remove harmful aerosols originated from dental surgery and preclude the infection transmission holds great potential for dental applications. This result is consistent with the previous high-speed camera results (Fig. 3) in which the generation of aerosol was significantly suppressed by the viscoelastic coolant in comparison with the viscous coolant. In any case, large-sized splatters (if any) and filaments generated by the viscous and viscoelastic coolants could only settle on the floor near the handpiece.¹⁰ 

Figures 4(e) and 4(f) present the distribution of the number of bacterial colonies N by distance from the handpiece in the case of water used as a coolant. Figure 4(e) presents the number of bacteria colonies to the right-side from the phantom patient. A total of 227 bacterial colonies were identified at the dentist's forehead where it was closest to the handpiece during the dental procedure (Table II). Excluding the case denoted as Wall in Fig. 4(e), the value of N decreased as the distance from the phantom increased (cf. N_Dentist > N_Equipment > N_Floor). On the floor at the distance of 225 cm, the number of bacteria colonies N was reduced by 94% compared to that of the dentist. On the wall, N was higher than that on the floor because of the airborne aerosol floatation, which caused entrapped bacteria to spread over long distances in the case of water coolant. The smaller the aerosol size, the longer it remains airborne and the further the settling distance.²⁸ At a distance of 260 cm, N was reduced only by 60% compared to that of the dentist. Figure 4(f) describes the distribution of bacterial colonies on the left-hand side of the phantom patient. A total of 91 bacterial colonies were identified on the assistant's forehead, which was decreased by 60% compared to that on the right-hand side of the dentist’s forehead (Table II). This was due to the fact that the aerosol was sprayed more to the right-hand side as the right-hand side tooth was treated. In the opposite direction (to the left), where the bur of the handpiece was not oriented, the value of N decreased with increasing the distance. Thus, the effect of aerosol transmission did not seem to be significant in that direction.

Figures 4(g) and 4(h) display the distribution of bacteria according to the distance in the case of the 2 wt. % PAA solution used as a coolant. Unlike the case of water, almost no bacteria were observed except on the floor. In the PAA case, the heavy long filaments should be ejected at high speed from the handpiece (see Fig. 3 and Table S1 in the supplementary material). Because of their parabolic trajectories, the locations that were slightly lower than the patient's oral cavity (cf. dental chair and dental equipment) were not contaminated. Also, the locations higher than the patient's mouth (cf. LEDs and the forehead/chests of medical staff) were not contaminated. However, the ejected long filaments eventually settled down, contaminating the floor similar to the case of water.

Table II shows the values of N on the forehead/chest of the dentist and the assistant, and on the LED. In the case of water, the forehead is located farther from the droplet source (i.e., patient's mouth) than the chest, but the values of N on foreheads are larger than those on the chests by 52%–70%. That indicates that droplets move in the frontal direction of the patient. (The staff’s foreheads were in the frontal direction of the patient, and the chest was slightly deflected from the frontal direction.) The value of N decreases with the frontal distance increase resulting in a lower N on the LED than on the doctor's head by 46% (the LED was located 50 cm above the head). In the case of 2 wt. % PAA, the value of N was less than 2 (i.e., insignificant), which means that its application essentially prevented droplet formation and propagation.

The viscoelastic aqueous 2 wt. % PAA solution used as a coolant during the dental procedure with a rotating bur of a handpiece revealed the following significant benefits compared to the other coolants tested (water, which is practically, inviscid, and the 10 and 20 wt. % NaPAA solutions, which are viscous Newtonian fluids).

• The viscoelastic aqueous 2 wt. % PAA solution suppressed the formation of aerosols (<50 μm) and droplets (50–100 μm), as well as individual splatters (>100 μm). Instead, it formed intact filaments settling down to the floor underneath the bur.

• Accordingly, the viscoelastic aqueous 2 wt. % PAA solution contaminated by bacteria (Staphylococcus epidermidis) completely prevented transmission of the infection to large distances from the phantom patient to the walls of dental office.

• Also, the lower shear viscosity of the 2 wt. % PAA solution compared to that of the 20 wt. % NaPAA facilitates pumping through a pipe in the existing dental chairs from a bottle to the rotating bur.

• In addition, PAA macromolecules possesses a large number of acid groups, which make it an active anti-bacterial agent, whereas NaPAA demonstrated a lower anti-bacterial activity than PAA because of the absence of the acidic group in the NaPAA molecular structure.

See the supplementary material for characterization methods, behavior of aerosols around a bur, and velocity distribution of liquid bodies from the handpiece.

This research was supported by Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (Grant No. HI21C0049010021). This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government under Nos. NRF-2020R1A5A1018153, NRF-2021R1A2C2010530, and 2022M3J1A106422611 and also by a Korea University grant.

The authors have no conflicts to disclose.

Yong Il Kim, Gang Min Noh, and Jungwoo Huh contributed equally to this paper.

Yong Il Kim: Conceptualization (equal); Data curation (equal); Formal analysis (lead); Funding acquisition (equal); Investigation (equal); Methodology (equal); Validation (lead); Visualization (equal); Writing – original draft (lead). Gang Min Noh: Data curation (lead); Formal analysis (equal); Validation (equal); Writing – original draft (equal). Jungwoo Huh: Data curation (lead); Methodology (equal); Visualization (lead); Writing – original draft (equal). Seongpil An: Data curation (equal); Methodology (equal); Validation (equal); Writing – review & editing (equal). Seongdong Kim: Data curation (equal). In-Seok Song: Conceptualization (lead); Data curation (equal); Funding acquisition (lead); Project administration (equal); Resources (lead). Alexander L. Yarin: Conceptualization (equal); Supervision (equal); Validation (equal); Writing – review & editing (lead). Sam S. Yoon: Funding acquisition (lead); Project administration (lead); Resources (lead); Supervision (lead); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.