Pyro-catalysis for tooth whitening via oral temperature fluctuation (2024)

Concept and feasibility of pyro-catalytic tooth whitening

The concept of pyro-catalysis for tooth whitening is generally to harvest ubiquitous oral motion-induced temperature fluctuation without additional equipment or power sources. As shown in Fig.1b, oral temperature will change with typical daily activity. For example, when hot food (such as hot water) is ingested, oral temperature can rise from ∼36 to ∼48 °C. Conversely oral temperature will drop to ∼21 °C when consuming cold food (such as ice cream). As is demonstrated in this work, this readily available physical stimulus can be used to excite the properties of pyroelectric materials and enables the application of pyro-catalysts for tooth whitening.

The screening charge on the surface of the pyroelectric material is released when a temperature fluctuation occurs. Exploitation of these temperature changed-induced surface charges as catalysts is called the pyro-catalytic effect or pyro-catalysis³². The typical pyro-catalysis process illustrated in Fig.1c–e. When the pyroelectric material has been poled, the internal ferroelectric domains are arranged in an ordinal state, developing a heterogeneous charge (screening charge) on the surface of the material^33,34. It has been reported that the intensity of the spontaneous polarization of pyroelectric materials decreases as temperature increases³⁵. When the pyroelectric material is heated, the original orderly arrangement of ferroelectric domains will be somewhat randomized, resulting in a decrease in polarization. The change in internal polarization will decrease the required balancing surface charge and will result in the release of excess shielding charges on the surface. The released charge will combine with water molecules to form radicals (•OH or •O₂^-) with strong redox properties³⁶. Similarly, when a pyroelectric material is cooled, the decrease in temperature causes an increase in spontaneous polarization. As the thermodynamic equilibrium is broken once again, the surface of the material absorbs charge from the environment to make itself electrically neutral. When carried out in an aqueous environment, the remaining charge reacts with water molecules to form radicals³⁷.

From the perspective of a practical application, we have designed a dental brace that incorporates a pyroelectric material (Fig.1f) that uses the pyro-catalytic to whiten teeth. These retainers can achieve pyro-catalysis through temperature fluctuations in the mouth induced by daily oral activities (e.g., drinking, breathing, talking, exercising, etc.), without using any other assistant equipment. The generated constant stream of active radicals will attack and degrade the stains on the tooth surface. Degradation of the colored macromolecular groups into small, colorless molecules results in a tooth whitening effect (Fig.1g–i). As these dental braces require no active energy supply (photo stimulation, ultrasonic agitation, etc.), the act of whitening the teeth is essentially passive. Furthermore, typical temperature fluctuations in the mouth are well below the Curie temperature of the pyroelectric materials, resulting in extremely long lifetime of use.

Synthesis and structural characterization of BaTiO₃ nanowires

As a demonstration, classical ferroelectric tetragonal BTO nanowires were synthesized by the hydrothermal method (see methods section) using H₂TiO₃ nanowires as the template crystals. Figure2a shows the X-ray diffraction (XRD) pattern of the as-prepared BTO nanowires. All diffraction peaks can be well indexed to the perovskite structure and no impurity phases are detected. As the spontaneous polarization in BTO is due to the asymmetric distortion of the cubic perovskite structure, and because the diffraction technique has difficulty distinguishing the tetragonal phase from cubic symmetry (due to the broadened peak profile for samples with small crystallite size), Raman spectroscopic analysis was performed to investigate local distortions of the lattice. As shown in Fig.2b, the distinct bands observed at 186, 249, 306, 515, and 715 cm⁻¹ in Raman scattering profiles can be assigned to the splitting of degenerated 3F_1u + F_2u modes of the polar crystal BTO (P4mm). In particular, the peak at 306 cm⁻¹ is usually associated with the asymmetric vibration of the [TiO₆] octahedra, thus the distinct phonon mode at 306 cm⁻¹ was regarded as the indicator of tetragonal distortion of the perovskite structure. The distortion from the ideal cubic perovskite structure allows the pyroelectric effect in the BTO nanowires.

Scanning electron microscopy (SEM) was used to characterize the morphology of the material (Fig.2c). SEM results reveal a large quantity of nanowire materials with uniform geometrical morphology. The nanowires are ∼100 nm in diameter, with 95% having lengths of ∼5 μm, as shown in the TEM image of a typical BTO nanowire (Fig. 2d). The resulting aspect ratio (as calculated by length divided by diameter) of the nanowires is more than 50 for the majority of our samples.

The crystallographic structure of the BTO nanowires is further confirmed by high-resolution transmission electron microscopy (TEM). The high-resolution TEM image in Fig.2e indicates a single crystalline structure with lattice fringes of (001). The interplanar spacing of 3.983 Å obtained by the selected area electron diffraction pattern (Fig.2f) is consistent with the interplanar distance of (001) planes of BTO. The lattice parameter measurement confirms the tetragonal symmetry of the perovskite structure. Moreover, the orientation of (001) lattice fringes clearly shows the extension of BTO nanowires along the polar axis [001]. Furthermore, the STEM and energy dispersive X-ray spectroscopy elemental mappings (Fig.2g–j) show, all the involved elements (Ba, Ti, and O) are hom*ogeneously distributed and well-matched with each other, resembling the original STEM morphology.

The ferroelectricity of BTO nanowires was characterized by piezoelectric force microscopy (PFM). The diameter of the nanowires from Fig.2k is on the order of ∼90 nm. The piezoresponse magnitude can be estimated from the amplitude image of Fig. 2l. The phase image is shown in Fig.2m, where bright and dark regions in the nanowires are corresponding to the domains oriented upwards and backwards directions, indicating the robust ferroelectricity of BTO nanowires. The local piezoelectric hysteresis loop of a BTO nanowire is shown in Fig.2n. The phase angle shows a full 180° change under a 5 V DC bias field. The phase switching, coupled with the butterfly shape of the amplitude loop implies the existence of well-defined polarizations along the vertical direction of the nanowires. And the finite element simulations of pyroelectric potential for different morphological nanomaterials showed that BTO nanowires exhibited outstanding pyroelectric potential compared to other nanostructures due to the spontaneous polarization along the length orientation. (Supplementary Fig.1).

Indigo Carmine degradation based on pyro-catalysis

To characterize the pyro-catalytic properties of BTO nanowires and to evaluate the potential use as a tooth whitening agent, the common food additive Indigo Carmine was selected as the target contaminant for degradation experiments. Briefly, the degradation of Indigo Carmine solution was investigated using BTO nanowire turbid liquid with a concentration of 1 mg mL⁻¹ under different temperature fluctuations. To simulate the temperature change in the human oral cavity more realistically, we chose 36 °C as the initial temperature and the pyro-catalysis experiment was carried out at different temperature ranges (ΔT = −10, −5, +5, +10, +15, +20 °C)³⁸ (Supplementary Fig.2). The UV-Vis absorption spectra of an Indigo Carmine solution exposed to the BTO turbid liquid at various temperature fluctuations is presented in Fig.3a–f. The maximum absorption peak of Indigo Carmine around 611 nm shows a notable decrease with increased number of thermal cycles, and the degradation rate can be higher than 98%. Impressively, at temperature fluctuation of ΔT = 20 °C, only three thermal cycles are required to degrade the Indigo Carmine (Supplementary Movie1). In contrast, the degradation of Indigo Carmine was minimal when BTO nanowires were used as catalyst or when BTO nanowires were not added (Supplementary Fig.3). The concentration of H₂O₂ monitored by liquid chromatography shows no change during the pyro-catalysis process (Supplementary Fig.4), while the DPD-POD experimental results indicate the generation of 300 μM of H₂O₂ in the reaction system after 30 hot and cold cycles at a temperature difference of +20 °C (Supplementary Fig.5). Anomalous to photo-catalysis, it can be inferred that H₂O₂ is as an intermediate product and remains an equilibrium state between consumption and generation during the pyro-catalysis process. The charges released from the pyro-catalytic BTO nanowires was first combined with the addition of H₂O₂, and subsequently react further into reactive radicals in a Fenton-like reaction^39,40. The addition of small amount of H₂O₂ can accelerate the pyro-catalytic process and improve the efficiency. These comparative experiments unambiguously verify that the degradation of Indigo Carmine due to the catalysis effect is strongly associated with the pyroelectricity of the nanowires.

To further verify the pyro-catalysis effect itself as a tooth whitening process, a duplicate experiment was carried out using [001]-oriented Pb(Mg_1/3Nb_2/3)-PbTiO₃ (PMN-PT) single crystal, a material with well-documented pyroelectric properties⁴¹. Bulk plate PMN-PT specimens were broken into pieces by ball-milling. After thermal cycling over the temperature ranges of interest, more than 95% of the Indigo Carmine was degraded when PMN-PT was used as the pyro-catalysis agent. Catalytic rate constant calculations show that the catalytic efficiency of a single heating/cooling cycle was improved as the temperature fluctuation range increases, with the same results observed when BTO nanowires were used as the catalyst (Supplementary Fig.6).

It is worth noting that the degradation rate of Indigo Carmine solutions with the same concentration of pyro-catalyst is dependent on the absolute temperature, with the rate significantly faster at higher absolute temperature. For example, 13 thermal cycles were required for a 95% degradation of Indigo Carmine at 26–36 °C (ΔT = −10 °C), while only 6 thermal cycles were required for the same degradation of Indigo Carmine at 36–46 °C (ΔT = 10 °C) (Fig.2a, d).

The pyroelectric effect relates the polarization (P), the temperature change (ΔT), and the change in the surface charge of a pyroelectric material. The surface charge, ΔQ can be characterized by Eq.(1) when the temperature change (ΔT) is known.

Where p is the pyroelectric coefficient, and A is the area of the surface from which the charge is released. From this equation we can obtain that for the same pyroelectric material, the charges released from the surface should be equal when experiencing the same temperature change. From this perspective, the number of thermal cycles required to degrade the Indigo Carmine solution at 26–36 °C and 36–46 °C should be the same which is different from the observed phenomenon. In fact, in the pyro-catalysis experiments, the entire process was carried out in aqueous solution, and instead of simply raising and lowering the temperature of the pyroelectric material, the entire reaction system underwent thermal cycling together. Temperature changes in a reaction system can drastically affect reaction kinetics, resulting in accelerated degradation when the absolute system temperature is increased.

Although reaction kinetics are proportional to system absolute thermal energy, a comparison of experiments at 26–36 °C (ΔT = −10 °C) with those at 31–36 °C (ΔT = −5 °C) shows that the degradation rate is faster at 26–36 °C due to the greater temperature fluctuation experienced despite an overall decrease in average temperature. And it can be clearly seen from Fig.3g that the number of cycles required for the degradation of Indigo Carmine decreases with increasing temperature fluctuation in the same temperature trend (heating or cooling). Therefore, we can conclude that the phenomenon observed in this experiment is still consistent with the theory, and the increased temperature variation range can induce more reactive radicals at the same temperature variation trend.

As natural oral temperature fluctuation rates vary (Supplementary Movie2), the pyro-catalysis degradation experiment was carried out using different heating rates under the same cooling rate of 1 °C m⁻¹ (Supplementary Fig.2b). The slower heating rate leads to a more efficient pyro-catalysis (i.e., k_Slow = 0.467, k_Mid = 0.303 and k_Fast = 0.199, Supplementary Fig.7), which can be attributed to a trade-off between positive and negative charge recombination and the ROS generation. The total charges induced by the pyroelectricity will be identical under the same temperature fluctuation (Eq.1), while they may recombine to each other (Eq. 2), rather than release to the liquid to produce ROS (Eqs.3 and 4) at a rapid heating rate. In order to further discuss the charge recombination and release efficiency on cooling or heating rate, we have additionally performed pyro-catalysis experiment using different catalyst concentrations under the same temperature fluctuation (i.e., ΔT = +10 °C) and temperature rise rate (Supplementary Fig.8). The pyro-catalytic efficiency increases and then decreases with the increase of catalyst concentration, indicating that charge release efficiency higher than a center degree will result in the recombination of charges with each other instead of reacting with water molecules to form radicals^42,43.

Typically, the pyro-catalyzed degradation of organic dyes is described as a series of chemical reactions (Eqs.2–5). As illustrated in Fig.1c, when a pyroelectric material is poled by an applied electric field, screening charges accumulate on its surface to balance the bound charges of electric polarization. The electric polarization will decrease with temperature increase, leading to a change in the bound and screening charge, and the excess screening charge will release into the aqueous solution to combine with water molecules to form radicals. Thermodynamically, the potential for generating •OH and •O₂^- needs to be at least ∼1.7 V and 1.9 V, respectively^44,45. The pyroelectric potential (∅_pyro) induced by temperature fluctuations (ΔT) can be governed by Eq.(6).

where p is the pyroelectric coefficient, l is the length of nanowires, ε₀ is the permittivity of vacuum, ε_r is the permittivity of BTO. The reported p and ε_r of BTO nanowires are ∼210 μC m⁻² K⁻¹ and ∼100^19,46. It can be calculated that the required nanowire length is at least 1.6 μm when the temperature fluctuation ΔT = 5 °C. The fabricated BTO nanowires have a length of ∼5 μm, which are capable to realize the pyro-driven ROS generation for degradation of organic dyes.

Total organic carbon (TOC) measurement was performed to monitor the total amount of organic carbon in the Indigo Carmine degradation experiment. As the reduction of TOC reflects the extent of degradation or mineralization of an organic species, the TOC value in the pyro-catalysis experiment was studied as a function of thermal cycles (Supplementary Fig.9). The initial TOC value of the Indigo Carmine is 39.85 mg L⁻¹. After 9 thermal cycles with a temperature fluctuation of +5 °C, the TOC value decreased to 8.36 mg L⁻¹ (i.e., ∼20% of the initial TOC value). The reduction of TOC confirms that color change was due to degradation of Indigo Carmine macromolecules (Eq. 5), rather than decolorization or decomposition.

Electron paramagnetic resonance (EPR) of the BTO nanowires pyro-catalyst using 5,5-dimethyl-1-pyrroline N-oxide (DMPO) as a spin trapper was used to detect •OH and •O₂^- radicals that play a major role in the degradation of Indigo Carmine. After three thermal cycles at ΔT = 5 °C and ΔT = 10 °C, the signals of both DMPO- •OH (Fig.3h) and DMPO- •O₂^- (Fig.3i) were observed. The additional peaks in Fig.3i are from the intermediate product DMSO- •CH₃ formed by the reaction of •OH and DMSO^47,48. The signal of radicals was enhanced with the increase of temperature change, which implies an elevated number of radicals. Furthermore, level of radicals also increased with increasing number of thermal cycles (Supplementary Fig.10). The radicals created by PMN-PT via pyro-catalysis were also tested by the same process, the signals of EPR for both •OH and •O₂^- demonstrated a strong dependence on the pyroelectric effect (Supplementary Fig.11). The correlation of the radical signal with the temperature fluctuation and the number of thermal cycles confirms that the radical was derived from the pyroelectric effect of BTO nanowires.

Due to the complexity of the human oral environment resulting in saliva with many other metal ions as well as enzymes, artificial saliva was employed as a solvent for pyro-catalytic indigo carmine degradation in order to exclude the effect of saliva environmental complexity. The BTO nanowires were subjected to three cycles at a temperature fluctuation of +5 °C. It can be seen that the BTO nanowires exhibit excellent cycling stability in the artificial saliva environment (Supplementary Fig.12), and their pyro-catalytic performance in this artificial saliva environment is almost the same as that in deionized water (Fig.3c). In addition, the phase structure and morphology of the BTO nanowires themselves remains stable (Supplementary Fig.13). This result provides strong support for the application of pyro-catalysis for tooth whitening.

Tooth whitening demonstration based on pyro-catalysis

After a successful demonstration of pyro-catalytic generation of reactive species and the subsequent degradation of Indigo Carmine solution by these radicals, tooth whitening experiments were performed using BTO nanowires. Human teeth were soaked in a mixture of black tea, red wine, and blueberry juice for one week to simulate the staining of teeth caused by habitual food intake. The stained teeth were then placed in a BTO nanowire suspension at an initial temperature of 36 °C (to simulate the oral temperature), and pyro-catalysis experiments were performed at different temperature fluctuations. The photographs of the teeth treated under various thermal cycles at ΔT = −10, +10, +25 °C were obtained using the same tooth and a standard greyscale card was used as a reference when the photographs were taken (Fig.4a–c). It is obvious that the tooth enamel is significantly whitened after 2000 thermal cycles of pyro-catalysis (compare left-most and right-most photo in each sequence), which is consistent with the previous experimental results that the higher the temperature and the greater the temperature fluctuation the more effective the teeth whitening effect is. It should be noted that at ΔT = +25 °C, even the roots, which are the most deeply stained part of the teeth and the hardest part to deal with, were completely whitened (Fig.4c). In contrast, at ΔT = +25 °C, the tooth whitening effect of the control (aqueous solution without BTO nanowires) was almost negligible (Fig.4d).

Typically, sugars and amino acids in food will remain on the surface of tooth and interact with the components in saliva to form colored chromogens through complex chemical changes, so the process of tooth whitening is to degrade the chromogen into colorless small molecules. Under pyro-catalysis, continuous temperature cycles excite the pyroelectric property of BTO nanowires, continuously releasing screening charges to form active radicals. These radical species attack the stains on the tooth surface by oxidizing the multiple conjugated double bonds of the large organic molecules that generated the stains⁴⁹. The Commission International De L’Eclairage (CIELab) system was also used to quantitatively characterize the degree of tooth whitening. The CIELab system characterizes the change in tooth shade by three elements: L (bright-dark), a (red-green), and b (blue-yellow)⁵⁰. The comparison of tooth chromaticity before and after pyro-catalysis was measured for different treatment environments, and the values of L, a, b are given in Fig.4e–g, respectively. When the pyro-catalyst was present, the brightness L of the teeth was significantly enhanced, while the value of a decreased significantly, and the value of b changed slightly. Conversely, changes in the values of L, a, and b for the teeth without BTO nanowires as the whitening agent are negligible. The chromaticity changes for the pyro-catalysis effect are more obvious with increased absolute temperature and increased temperature change. The ΔE calculated from Eq.(7) was used to further characterize the effect of whitening teeth⁵¹.

Figure4h shows the calculated ΔE for teeth treated in different environments. The ΔE for teeth treated in the environment containing the pyro-catalyst was about four times higher than for deionized water when ΔT = 25 °C. Moreover, similar to the experimental results for the degradation of organic dyes there is an obvious correlation between the value of ΔE and both system temperature and the magnitude of temperature fluctuation. The tooth whitening procedure was also verified using PMN-PT single crystal powder as catalyst, with the PMN-PT samples showing a better tooth whitening effect after 2000 thermal than BTO nanowires (Supplementary Fig.14).

Pyro-catalysis performance of BTO-Gel

After successful demonstration of tooth whitening by pyro-catalysis, we developed a practical delivery method. A photo-cured hydrogel was used as a carrier for the pyroelectric powders in preparation of composite gels to serve in the fabrication of medical braces. To verify the retention of pyro-catalytic performance of the BTO nanowires in the composite gel form, degradation experiments of Indigo Carmine solution were repeated for the composite. As shown in the UV-VIS absorption curves of Fig.5a, the gel with the composite pyroelectric material exhibited pyroelectric catalytic performance, which was comparable to that of the pyroelectric nanowires alone, confirming that the presence of the medical gel did not interfere with the pyroelectric catalytic performance. The pyro-catalytic performance stability of the composite gel was then tested by using the same gel for five sequential degradation experiments with Indigo Carmine solution. The composite gel showed no changes between successive cycles (Fig.5b).

The ability of the pyroelectric composite gel to generate radicals was also characterized. The composite gel was placed in water with DMPO as the trapping agent and subjected to three thermal cycles. The DMPO- •OH signal detected at ΔT = 10 °C was approximately twice as high as that at ΔT = 5 °C (Fig.5c). This result was consistent with the EPR test results for BTO nanowires (Fig.3h–i). Furthermore, the signal of DMPO- •OH was detected at ΔT = 5 °C for multiple, successive thermal cycles (Fig.5d), and the intensity of the DMPO- •OH signal increased with the increase of the number of thermal cycles, which proves that the radicals are continuously generated as the thermal cycles proceed. The detection of the DMPO- •O₂^- signal performed in DMSO solution exhibited a similar trend (Fig.5f–g), which further confirms that the signals of the radicals generated in the composite gel come from the pyroelectric material contained therein.

To further characterize the stability of the pyro-catalysis of the composite gels, pyroelectric radical generation was tested after subsequent degradation cycles. Each test was performed at ΔT = 10 °C for three thermal cycles, and the results (Fig.5e, h) showed that after five dye degradation experiments, the composite gels still exhibit excellent radical production of both •OH and •O₂^-. The fluidity of the BTO gel is demonstrated in Fig.5i. The excellent fluidity before curing allows the gel to be molded into any shape, making it possible to prepare intricate dental braces, but also allows for customization for a pinpoint treatment of a single tooth. A stained tooth was placed in a mold and BTO gel was injected into the mold to completely encapsulate the enamel of the tooth. The gel was then cured for 15 min in UV light. Afterwards, the tooth was placed in water and subjected to 2000 thermal cycles at a temperature fluctuation of 25 °C with different heating rate (Fig.5j, k). It was evident that the gel-coated parts of the teeth were significantly whitened compared to the initial state, while the non-gel-coated parts showed no significant color change. In order to simulate more realistic oral temperature environment, the stained tooth with BTO gel was immersed into hot and cold side each for 5 s (Supplementary Movie3). The tooth whitening effect is also obviously occurred, while in contrast, there was no perceptible change in the color of the one with 5 min, which agrees well with the pyro-catalytic dye degradation (Supplementary Fig.7).

Tooth structure characterization

Dental enamel, as the hardest tissue in the body, acts as a protective covering of teeth and can withstand a wide range of functional and non-functional loads⁵². Classical tooth whitening methods using peroxide are effective, but can cause adverse such as increased surface roughness, cracking, and enamel changes⁵³. To evaluate the safety of the BTO composite gels, the microscopic morphology of tooth enamel before and after whitening with different whitening agents was examined. As shown in Fig.6a, the same area of the same tooth was observed and recorded using scanning electron microscopy (SEM). The enamel surface of the tooth was rough and stained in the initial state, and after 2000 thermal cycles with BTO gel as the whitening agent, stains are removed from the tooth surface, and no damage was caused to the enamel owing to the gentle and continuous release of ROS (Fig.6b). In contrast, enamel whitened with commercial tooth whitening gels showed significant and irreversible damage to the enamel due to the violent nature of the response caused by the dramatic release of ROS from high peroxide concentrations (Fig.6c). To further verify that the pyroelectric-catalyzed tooth whitening technique does not cause damage to teeth microhardness of the tooth enamel was measured at five locations before and after whitening (Fig.6d, e). In the as-received condition, the hardness at the five different positions is similar, with an average value of ∼350 HV, indicative of a complete and healthy tooth. There is little to no change in the hardness across the tooth’s surface between the as-received, the stained, and the pyro-catalytically whitened condition. This suggests that the pyroelectric gel does not cause mechanical damage to the enamel while whitening the tooth, much less cause any functional defects to the tooth.

To evaluate the biocompatibility of pyro-catalysis gel, fibroblast cells (L-929) were used to co-cultured with BTO nanowires and evaluated using the live/dead fluorescence staining and typical cell-counting kit 8 (CCK-8) assays. Fluorescence microscope images of tissue cultures exposed to different BTO nanowires doses across a three-day period are presented in Fig.7a-d. Since no significant cell viability decrease can be detected, even at the BTO nanowires dose of as high as 0.3 g mL⁻¹, the BTO nanowires can be confirmed as biocompatible. Figure7e-g presents the results of the CCK-8 assay. The cell survival rate of all experimental groups was above 70%, showing that the BTO nanowires have no cytotoxicity to the fibroblast cells. As a further precaution, Ba²⁺ leakage during the whitening process was also examined. The results showed no detectable Ba²⁺ creation after 2000 thermal cycles (Supplementary Fig.15).