2.1.

Tissue Irradiation

Two sealed TEA lasers designed and manufactured by Argus Photonics Group, Jupiter, FL were used to irradiate the tissue samples with fluences of 0.1–100 J/cm ² and energies from 0.1 to 80 mJ/pulse. The pulse durations were 2 and 8 μs in length (Figure 3) and the repetition rate was fixed at 1 Hz to avoid heat accumulation effects. The laser pulse temporal profiles shown in Figure 3 were measured at room temperature with a HgCdTe detector (Boston Electronics, Boston, MA). The temporal profiles of Figure 3 were integrated to indicate that 72 of the energy is partitioned in the first 600 ns, which is the full width of the gain-switched spike for the unmodified TEA laser pulse. In contrast, only 14 of the energy is contained in the first 600 ns for the stretched TEA laser pulse. The laser energy was measured and calibrated using laser calorimeters and the laser spot size was measured by both scanning a razor blade across the beam and by directly imaging the beam with a pyroelectric laser beam profilometer (Pyrocam I, Spirocon). The spot profile was single mode and fluences were defined using a Gaussian beam with a 1/e² beam diameter. The laser beam was focused to spot sizes between 200 and 1000 μm. Human coronal dentin and enamel thin sections of 200 μm thickness were prepared using a hard tissue microtome from caries free, unerupted molars, and premolars. The sections were mounted before an IR pyroelectric detector (perforation sensor) and the number of pulses required to drill through each section were recorded. After perforation, an Olympus microscope with a maximum magnification of 500 times, interfaced to a digital charge coupled device camera and image analysis software, was used to determine the crater volume. The diameter of the crater was measured at three positions: surface, halfway through, and on the backside of the respective sections by scanning the image plane of the calibrated microscope, and fitted to two conic sections to determine the volumes of the perforation craters for each set of irradiation conditions.

2.2.

Surface Dissolution Experiments and Fourier Transform Infrared (FTIR) Spectroscopy

The effectiveness of 9.6 and 10.6 μm laser radiation as a potential means for caries prevention, was assessed by irradiating 5×5 mm² blocks of polished bovine enamel. The surfaces of the enamel blocks were first serial polished to a 1 μm finish with 6, 3, and 1 μm diamond suspensions. The samples were irradiated using a 400 μm spot size that was scanned across the sample using a Newport MM-2000 motion control system interfaced to LABVIEW 4.0 software from National Instruments to uniformly treat the entire surface. The surface was scanned every 100 μm for sufficient overlap of the surface and five pulses were delivered to each spot at a repetition rate of 1 Hz.

After irradiation, an infrared microscope (XAD Plus, Laser Precision Analytical) with a 150×150 μm aperture attached to a Fourier transform spectrometer (Model RFX-30, Laser Precision Analytical) was used to acquire FTIR reflectance spectra of the surface of the irradiated bovine blocks and the control samples. The relative intensity of the carbonate bands located at 1450 and 1410 cm ⁻¹ was calculated by integrating the normalized spectra from 1200 to 1500 cm ⁻¹ . The carbonate loss was defined as one minus the ratio of the area of the irradiated enamel over that of the nonirradiated samples times a hundred. After spectra were taken the samples were mounted on high-density-polyethylene disks and placed in a reaction vessel containing a solution of 0.1 M acetic acid buffered to a pH of 4.5 Aliquots of the solution are collected for 20 min at 2 min intervals after immersion with continual stirring. The aliquots were subsequently analyzed for calcium and phosphate to determine the dissolution rate. Calcium was analyzed using atomic absorption and phosphate concentration photometrically using an ammonium molybdate colored complex that absorbs light at 820 nm. The technique(s) for evaluating surface dissolution and carbonate content via reflectance IR spectroscopy have been described previously.²²

3.1.

Surface Modification

The irradiation fluence thresholds that produced microscopically observable surface changes on enamel were determined for CO ₂ laser radiation at 9.6 and 10.6 μm for the conventional TEA laser with a 2 μs pulse duration. There were laser induced surface changes on enamel at incident fluences above 0.5–1 J/cm ² for 9.6 μm and approximately 2–3 J/cm ² for 10.6 μm. With the appropriate adjustments due to reflectance losses, 49 at 9.6 μm and 13.5 at 10.6 μm, the difference in absorbed energy between the two wavelengths is significantly greater. The absorbed fluences are calculated to be 0.25–0.5 J/cm ² for 9.6 μm and 2–3 J/cm ² for 10.6 μm, demonstrating a factor of 6–8 times the difference in the amount of absorbed energy required for the surface modification of enamel using 2 μs CO ₂ laser pulses between the two wavelengths. With regards to the pulse duration, the required energy for 2 μs pulses is markedly lower than the 3–4 J/cm ² required for surface modification with 9.6 μm laser pulses of 100 μs duration that were previously employed in surface modifications studies.^{23 24 25}

Morphological changes in enamel during short (<1 μs) pulsed CO ₂ laser irradiation was first observed by Nelson et al.;²⁶ They observed that very distinct morphological differences were induced in the mineral phase for each of the CO ₂ wavelengths. Similar wavelength dependent morphological differences were not noted between 10.6 and 9.6 μm after irradiation after 2 μs TEA laser pulses. This discrepancy may be due to the fact that the previously reported study²⁶ reflected differences in absorbed fluence and not in wavelength, since all the samples were irradiated at the same incident fluence level for each wavelength. The ablation thresholds vary markedly with wavelength due to the 40 difference in reflectance and the order of magnitude variation in the absorption coefficient between 9.6 and 10.6 μm. Moreover, 400 laser pulses were used in those previous studies and the periodic surface structures produced may have been unique to that particular laser mode structure due to extensive surface bombardment.²⁶ During short pulsed laser irradiation without the addition of water as a coolant, dentin exhibited charring at 10.6 μm but not at the more highly absorbed 9.6 μm wavelength.²⁷ The differences in the fluence threshold for ablation and the distinct differences noted in surface morphology between 9.6 and 10.6 μm are entirely consistent with the markedly different absorption depths and thermal relaxation times for those two wavelengths. It is interesting to note that with longer laser pulses, namely 100 μs, this difference does not manifest itself²⁸ because the thermal diffusion length at 9.6 μm for a 100 μs laser pulse is about 10 μm, which is roughly equivalent to the absorption depth of 10–12 μm at λ=10.6 μm.

3.2.

Infrared Spectroscopy and Caries Prevention Studies

FTIR was used to determine chemical changes in dental enamel as a result of laser irradiation. In previous studies, we discovered that FTIR spectroscopy in spectral reflectance mode was well suited for resolving chemical changes on the surface of enamel.^{22 29 30} The principal advantage of this technique is that the tissue reflectance is only influenced by a surface layer of a thickness on the order of the wavelength of the light. Thus, surface changes localized to the outer 10 μm of tissue are probed. The carbonate bands centered at 6–7 μm disappeared completely as a result of irradiation at fluences of approximately 0.5–1 J/cm ² (Figure 4) with the (2 μs) TEA 9.6 μm CO ₂ laser pulses. This is a four-to-eight fold reduction in the required energy as compared to the longer 100 μs pulses used in previous studies which required 4 J/cm ² . While for the 10.6 μm laser pulses the reduction in the required fluence was less dramatic, the carbonate band disappeared at fluence levels of 3 J/cm ² as compared to approximately 5–6 J/cm ² for the longer 100 μs laser pulses a factor of 1.5–2 difference in fluence.²⁹ Integration of the area of carbonate bands in the spectra enabled a relative carbonate loss to be calculated based on normalization to the carbonate band of the nonirradiated samples. In previous studies, we directly correlated the changes in the carbonate bands with the transformation of the carbonated hydroxyapatite mineral to the more acid resistant purer phase hydroxyapatite.²² The incident fluence that was sufficient to induce loss of carbonate in the FTIR spectra coincided with the optimum laser fluence that inhibited both surface and subsurface acid dissolution. Therefore, the FTIR is extremely useful as a rapid feedback nondestructive probe to elucidate the optimum laser parameters to inhibit acid dissolution.

Figure 4

The carbonate loss (solid symbols—left axis) determined using IR spectroscopy and the surface dissolution rate reduction (hollow symbols—right axis) relative to nonirradiated control samples for bovine enamel samples is plotted vs incident fluence after 9.6 μm (triangular symbols) and 10.6 μm (square symbols) (2 μs) TEA CO ₂ laser irradiation.

The results of the surface dissolution studies are shown graphically in Figure 4. The surface dissolution rate of the laser irradiated 5×5 mm ² bovine blocks relative to nonirradiated control samples was determined for fluence levels up to 4.0 J/cm ² for 10.6 and 9.6 μm laser irradiation. After irradiation with five 9.6 μm laser pulses of 2 μs duration in each spot at an incident fluence of 0.6 J/cm ² /pulse (absorbed fluence of 0.3 J/cm ² ), the rate of acid dissolution of enamel decreased by 40–50. The reduction in the relative surface dissolution rate closely paralleled the reduction in the carbonate band intensities, as was previously observed with longer 100 μs laser pulses.²² Significantly higher irradiation intensities were required to reduce the rate of acid dissolution by a similar amount after 10.6 μm laser irradiation, namely an incident fluence of 3.0 J/cm ² and an absorbed fluence of 2.6 J/cm ² . Moreover, the required irradiation intensities using the 2 μs TEA laser pulses are from five to ten times lower than those required using 9.6 μm laser pulses of a 100 μs duration, namely an incident fluence of 4–5 J/cm ² .²²

As indicated in the earlier section of this paper dealing with the problem of heat accumulation and thermal damage, the greatest potential danger of laser irradiation is the risk of excessive heat accumulation. Dental pulp, the soft tissue of the center of the tooth, is rigidly encased in a confined space enclosed by hard tissue, therefore it is extremely susceptible to permanent damage due to thermally induced inflammation.³¹ The 1965 Zach and Cohen study of the effect of heat on the pulp of Rhesus monkeys indicated that a temperature rise of 5.5°C in the pulp caused irreversible puplitis in 15 of the pulps.³² The accumulation of heat after use of the dental drill for cavity preparations can raise the pulpal temperatures to dangerous levels if air and/or water cooling are not used.³³ Similarly, the use of multiple pulse irradiation without air/water cooling may also result in the accumulation of heat to levels dangerous to the pulp. Subsurface thermocouple measurements, simulations of heat conduction, and histological examinations during laser irradiation show that the extent of pulpal heating is determined by the rate of deposition of the laser energy in the tooth, the distance from the laser spot to the pulp, and the rate of energy loss from the tooth.^{34 35 36 37} By using the shorter laser pulse we are modifying a 1 mm spot size on the enamel surface with a total absorbed energy delivered of only 20–100 mJ for 5–25 laser pulses (0.5 J/cm ² ) versus 100–2500 mJ in our previous studies utilizing 100 μs 9.6 μm laser pulses (2.5 J/cm ² ). Note that other groups report the use of continuous wave CO ₂ and argon ion lasers for caries prevention and they utilize absorbed fluences in the 100–200 J/cm ² range, factors of 200–400 times higher than what we report in this study. The high fluences required for effective use of those systems may lead to excessive heat accumulation in the tooth.^{38 39 40} This study indicates that by using shorter laser pulses the energy needed to modify the surface of enamel for caries prevention can be reduced by a factor of 5–10 using short (<10 μs) 9.6 μm laser pulses.

3.3.

Caries Ablation

Ablation studies were performed at 9.6 and 10.6 μm with unmodified TEA laser pulses of 2 μs duration with the gain switched spike present in the first 100–200 ns of the pulse. The onset of a plasma initiated by the initial gain switched spike, prevented ablation at rates sufficient for clinical application. The rates were restricted to 2–3 μm/pulse for enamel and 6–8 μm/pulse for dentin for both 9.6 and 10.6 μm laser pulses.²⁷ The pulse duration was subsequently lengthened to 8 μs, resulting in almost an order of magnitude reduction in the magnitude of the gain switched spike. This change in the temporal profile of the laser pulse raised the threshold for plasma formation (see Figure 3) and markedly increased the maximum achievable ablation rate. Enamel and dentin ablation rates and efficiencies as a function of the incident laser fluence for 8 μs, 9.6 μm CO ₂ laser pulses are shown in Figures 5 and 6. These values are for the perforation of 200 μm thick sections without the application of a water spray. The spot size was 280 μm and the repetition rate was 2 Hz. The ablation rate saturates at a maximum rate of approximately 25 μm/pulse at fluences above 10–15 J/cm ² . The peak enamel ablation efficiency at 9.6 μm with the 8 μs pulse approaches 0.1 mm ³ /J and is comparable to the highest values reported in the literature for either the longer pulsed CO ₂ laser (20–300 μs) or the erbium laser systems. Maximum dentin ablation rates approached 50 μm/pulse at fluences of only 10–20 J/cm ² and ablation efficiencies exceeded 0.1 mm ³ /J .

Figure 5

The ablation efficiency in (mm ³ /J) (circles) and the single pulse ablation rate (μm/pulse) (triangles) for perforation of 200 μm thick sections of human enamel. Each data point is the mean of three measurements ±standard deviation (s.d.). A dotted line is empirically drawn through the ablation rate data as a visual aid. It represents an increase in the ablation rate with increasing influence followed by saturation of the ablation rate above the plasma-shielding threshold.

Figure 6

The ablation efficiency (in mm ³ /J ) (circles) and the single pulse ablation rate (in μm/pulse) (triangles) for perforation of 200 μm thick sections of human dentin. Each data point is the mean of three measurements ± s.d. A dotted line is empirically drawn through the ablation rate data as a visual aid. It represents an increase in the ablation rate with increasing fluence followed by saturation of the ablation rate above the plasma-shielding threshold.

The CO ₂ laser couples more strongly to the mineral component of dental hard tissue than the Er:YAG laser and therefore it may not be as dependent on the requirement of the addition of a water spray to alleviate stalling problems such as those encountered with the erbium laser systems. We have been able to perforate 2 mm of enamel and dentin with relatively low fluences. The enamel ablation rates of 10–25 μm/pulse are lower than the longer pulsed CO ₂ and Er:YAG laser systems which can ablate at rates exceeding 100 μm/pulse. However, high repetition rates are readily achievable for transverse excited CO ₂ lasers and very high enamel and dentin removal rates are feasible.

Studies have shown that pulsed lasers can generate strong compressive and tensile stress waves due to rapid laser heating.⁴¹ These shock waves propagate through the tissue radiating outward from the site of absorption. Generally, biological tissue has a relatively high compressive strength, however the tensile strength is much weaker and tensile forces can cavitate soft tissue and generate cracks in hard tissue. We have previously reported extensive mechanical damage peripheral to the respective ablation sites with Q-switched Er:YAG ablation⁴² and such cracks have been reported for high single pulse energy Er:YAG and Er:YSGG laser pulses.⁴³ Similar cracks due to thermally induced stresses have also been produced in teeth using continuous wave CO ₂ lasers.⁴⁴ We did not observe any obvious stress related peripheral damage for either the conventional (2 μs pulse) or the long pulse (8 μs pulse) TEA laser systems. Ablation craters produced via perforation of human enamel and dentin are shown in Figure 7. The craters are extremely clean without any laser associated peripheral damage.

Figure 7

Holes drilled through 200 μm thick sections of enamel and dentin with λ=9.6 μm laser pulses of 8 μs pulse duration show no evidence of cracking and extended thermal damage. The image on the left was taken at 50 times magnification and it demonstrates the large number of holes that can be drilled through a thin tooth section without inducing any fractures. The image on the right shows the surface morphology of one of the holes produced in enamel at a higher magnification. The enamel is melted and there is a ring of fused enamel of 10 μm thickness.