2.1.

Laser Parameters

An industrial marking laser, Impact 2500 from GSI Lumonics (Rugby, United Kingdom), operating at $9.3\mu \mathrm{m}$ , due to optical coatings on the resonator optics, which are capable of high repetition rates up to $500\mathrm{Hz}$ , was used for these experiments. The laser was modified for operation in a single-spatial mode optimized for the longer $5\text{-}\text{to}20\text{-}\mu \mathrm{s}$ laser pulses by LightMachinery (Ottawa, Canada) by installing an intercavity aperture and modified resonator optics. This laser also incorporates a solid-state switch for very stable and reliable high-voltage switching, solving the stability and heat dissipation problems encountered previously with thyratron-based systems. The laser gas handling system is sealed with an automatic gas exchange after $20\text{-}\min$ of operation. Fluence ranging from $0.2\text{to}160\mathrm{J}∕{\mathrm{cm}}^2$ and pulse energies of $0.1\text{to}45\mathrm{mJ}$ per pulse were used. The laser pulse temporal profiles shown in Fig. 2 were measured with a room temperature HgCdTe detector from Boston Electronics PD-10-6-3 (Boston, Massachusetts). The laser pulse was $18\mu \mathrm{s}$ in length for a 4% $\mathrm{C}{\mathrm{O}}_2$ gas mixture (20% ${\mathrm{N}}_2$ ) and $15\mu \mathrm{s}$ for the 2% $\mathrm{C}{\mathrm{O}}_2$ gas mixture (22% ${\mathrm{N}}_2$ ) with the use of a 75% reflectivity output coupler. Half the energy was contained in the initial $5\mu \mathrm{s}$ and 80% of the energy was in the first $10\mu \mathrm{s}$ . The laser pulse duration was approximately half the duration mentioned above, approximately $5\text{to}9\mu \mathrm{s}$ , with the same two gas mixtures if a higher reflectivity (85% reflectivity) output coupler was used. The laser energy was calibrated and measured using a laser energy and/or power meter, EPM 1000, Coherent-Molectron (Santa Clara, California) with a ED-200 Joulemeter from Gentec (Quebec, Canada). The laser was focused with a planoconvex ZnSe lens of $100\text{-}\mathrm{mm}$ focal length to a beam diameter of approximately $200\mu \mathrm{m}$ . The laser beam diameter $(1∕e^2)$ at the position of irradiation was determined by scanning with a razor blade across the beam. Two- and three-dimensional images of the laser spatial profile were acquired using a Spirocon Pyrocam III pyroelectric array (Logan, Utah). Both laser beam profile and spatial profile showed that the laser was operated in a single-spatial mode, Gaussian spatial beam (see Fig. 3 ). High spatial beam quality is a prerequisite for efficient coupling into hollow waveguide beam delivery systems.

Fig. 2

The temporal profiles of the TEA $\mathrm{C}{\mathrm{O}}_2$ laser pulses for the two laser gas mixtures with varying $\mathrm{C}{\mathrm{O}}_2$ and ${\mathrm{N}}_2$ concentration are shown: (a) 4.3% $\mathrm{C}{\mathrm{O}}_2$ and (b) 2% $\mathrm{C}{\mathrm{O}}_2$ . The pulse durations were 15.4 and $18\mu \mathrm{s}$ , respectively, for each gas mixture. The temporal profiles (left axis) were integrated (right axis-arrow) to show that the laser pulses produced using the 4.3% $\mathrm{C}{\mathrm{O}}_2$ gas mixture had a greater fraction of the pulse energy in the initial few microseconds than that measured during the same time interval for the laser pulse produced with the 2% $\mathrm{C}{\mathrm{O}}_2$ gas mixture, that is, 50% of intensity in first $3.9\mu \mathrm{s}$ for the 4.3% $\mathrm{C}{\mathrm{O}}_2$ gas mixture versus $5.2\mu \mathrm{s}$ for the 2% $\mathrm{C}{\mathrm{O}}_2$ gas mixture.

Fig. 3

Two- and three-dimensional images of the laser spatial profile acquired using a Pyrocam III pyroelectric beam analyzer.

2.2.

Laser Ablation Rate Determination

Thin sections, approximately $200\text{-}\mu \mathrm{m}$ thick, of human enamel and dentin from unerupted human molars and premolars and adult porcine alveolar bone were prepared using a MicroLux diamond band saw with water cooling from Micro-Mark (Berkeley Heights, New Jersey) and a Isomet 2000 precision saw from Buehler (Lake Bluff, Illinois). Each section was mounted over the edge of a glass slide and a ED-200 Joulemeter was placed directly behind each section to function as a perforation sensor. The number of pulses required to perforate each section was recorded to determine the ablation rate. It is important to point out that the single pulse ablation rate depends on the method of measurement with the highest rates measured for a single incident pulse on the tissue. However, such measurements require accurate measurement of the depth ablated. Sample perforation measurements are easier to carry out, however the ablation rate determined using this method is highly dependent on the sample thickness with the rate decreasing with increasing sample thickness. We have found that a $200\text{-}\mu \mathrm{m}$ thickness works well for laser pulses that ablate $5\text{to}40\mu \mathrm{m}$ per pulse, requiring a minimum of several laser pulses to perforate the tissue samples.

2.3.

Laser Incisions in Human Enamel, Dentin and Porcine Bone

Plano-parallel enamel blocks, $3\times 3{\mathrm{mm}}^2$ , approximately $2\text{-}\mathrm{mm}$ thick, were prepared from human unerupted molars. Surfaces were polished with 1200 carbide grit, and smear layers produced from polishing on the sample surfaces were removed by sonication. Dentin and bone sections with a thickness of at least $2\mathrm{mm}$ were cut using diamond saws and samples were kept well hydrated before ablation and either cooled using a computer-controlled water spray or left dry during ablation. Bone samples were kept frozen before ablation. A low-volume and low-pressure air-actuated fluid spray delivery system consisting of a EFD 780S-SS spray valve, a Valvemate 7040 controller, and a fluid reservoir from EFD Inc. (East Providence, Rhode Island) controlled by a computer was used to provide a uniform spray of water on the tissue surface.

Lateral incisions $3\mathrm{mm}$ long were produced in our dentin and bone sections by scanning the laser in spots spaced at $50\text{-}\mu \mathrm{m}$ intervals, each spot receiving 10 pulses. A computer-controlled stage was used to create controlled movement of the samples for the incisions. Three incisions were made for each frequency (10, 50, 100, 200, 300, and $400\mathrm{Hz}$ ) for both wet and dry conditions in both dentin and bone. Figure 10 shows examples of the incisions in dentin and bone. After ablation, tissue sections and enamel blocks were embedded onto epoxy resin disks with LX-112 epoxy resin from Ladd Research (Williston, Vermont), that were subsequently serially polished with 1200 carbide grit, and 6, 3, 1, and 0.3 diamond suspensions, using Ecomet 2 and Fibromet polishers from Buehler.

Fig. 10

Laser ablation in human dentin (left) and porcine bone (right) with and without the water spray at $400\mathrm{Hz}$ and $80\mathrm{J}∕{\mathrm{cm}}^2$ .

2.4.

Residual Heat Measurements with Thermal Imaging Analysis

Enamel blocks, $4\times 4{\mathrm{mm}}^2$ , $\sim 1\text{-}\mathrm{mm}$ thick, were prepared from bovine incisors. Surfaces were polished with 1200 carbide grit and each block was glued onto the edge of a glass slide (Fig. 4 ). A thermal camera Omega, from Indigo System Corporation (Goleta, California), incorporating an uncooled microbolometer focal plane array with $14\text{-}\text{bit}$ digital output was placed behind the block to measure the thermal emission during laser ablation in the wavelength band from $7.5\text{to}13.5\mu \mathrm{m}$ .

Fig. 4

(Inset) The experimental setup for residual heat deposition measurements on bovine enamel block samples. The thermal camera was placed behind the sample to record thermal radiation from the tissue sample throughout the laser ablation. Simultaneous thermal measurements using the thermal camera (+)-left-axis and a microthermocouple (o)-right axis for 3000 laser pulses delivered with a fluence of $0.28\mathrm{J}∕{\mathrm{cm}}^2$ and a repetition rate of $300\mathrm{Hz}$ .

The magnitude of the residual energy $\left(E_R\right)$ remaining in the tooth as heat during enamel ablation was directly calculated by taking the ratio of the rise in signal due to thermal radiation on the opposite side of laser irradiated block for ablative versus nonablative laser pulses. Every image consisted of $164\times 129\text{pixels}$ , and a region of interest, where the pixels correlate with the block surface, was selected before each series of measurements. The average value of all the $14\text{-}\text{bit}$ pixels within the region of interest was computed and recorded continuously throughout the laser treatment. The accuracy of this approach was validated and calibrated by comparison of the thermal emission from the block with simultaneous micro-thermocouple measurements. Figure 4 shows the temperature rise that was simultaneously measured using the thermal camera and a thermocouple cemented to the enamel surface. Nonablative laser pulses were delivered at a fluence of $0.28\mathrm{J}∕{\mathrm{cm}}^2$ and a repetition rate of $300\mathrm{Hz}$ . The average residual heat deposition values were $71.2\pm 5.0\%$ for the thermocouple readings and $69.9\pm 4.8\%$ for the thermal camera readings.

In order to measure the residual energy remaining at each respective fluence of interest, each block was irradiated with two laser pulse trains: nonablative and ablative. The first train of pulses irradiated the surface of the block with 3000 low-energy nonablative laser pulses delivered at a repetition of $300\mathrm{Hz}$ (Fig. 5 ). The measurements were repeated after the block had equilibrated (returned to room temperature), then a second train of ablative pulses were delivered to the surface of the block at the fluence of interest. In order to vary the fluence while keeping the total energy delivered and total delivery time, of $10\mathrm{s}$ , constant, the repetition rate, total number of laser pulses, energy per pulse, and the spot size were varied. These measurements are similar to those described in Ref. 30.

Fig. 5

Residual heat on the backside of a block of bovine enamel was monitored during irradiation with two series of ablative and nonablative $\mathrm{C}{\mathrm{O}}_2$ laser pulses. The repetition rate, fluence, and number of laser pulses were varied, but the total delivery time and cumulative energy was constant. The 3000 nonablative laser pulses (top crosses) were delivered at a repetition rate of $300\mathrm{Hz}$ and an incident fluence of $0.28\mathrm{J}∕{\mathrm{cm}}^2$ . The 90 ablative laser pulses (bottom trace filled) were delivered at a rate of $9\mathrm{Hz}$ and an incident intensity of $9.5\mathrm{J}∕{\mathrm{cm}}^2$ . The residual energy $E_R$ was given by the ratio of the two peak intensities.

2.5.

Polarized Light Microscopy (PLM)

Images of tissue surfaces were taken before and after laser irradiation with a Secolux BF/DF/DIC microscope manufactured by Leitz-Wetzlar (Germany). A low-volume and/or low-pressure air-actuated fluid spray delivery system consisting of a 780S spray valve, a Valvemate 7040 controller, and a fluid reservoir from EFD, Inc. was used to provide a uniform spray of fine water mist onto the enamel surfaces at $16\mathrm{mL}∕\min$ . Each incision, approximately $6\text{-}\mathrm{mm}$ long and $700\text{-}\mu \mathrm{m}$ wide, was prepared by scanning the laser beam, operating at $300\mathrm{Hz}$ , across the sample at $6\mathrm{mm}∕\mathrm{s}$ , with a $50\text{-}\mu \mathrm{m}$ separation between each longitudinal scan.

Polarized light microscopy was used to determine the degree of thermal damage produced while creating our lateral incisions. Thermally damaged dentin and bone appears dark brown to black due to carbonization of protein (collagen). Under examination with a polarized microscope with crossed polarizers, thermal damage to the mineral and protein phases result in loss of birefringence. Thin sections were taken from each sample incision sectioning our samples using the Isomet 2000 diamond saw. Sections were made perpendicular to the direction of the lateral incisions to view them in cross section. After cutting, sections were imbibed with water and viewed in the optical microscope using crossed polarizers and a Red I wave plate. Thin sections were examined visually using a Westover Scientific Series 7 optical microscope (Westover Scientific, Mill Creek, Washington) with an integrated digital camera, Canon EOS Digital Rebel XT (Canon Inc., Tokyo, Japan) and IMAGE PRO PLUS software (Media Cybernetics, Silver Spring, Maryland). A moire ruling with 100 lines per millimeter was used to calibrate the length measurements on the images. Figure 11 shows images at $100\times$ magnification of wet and dry dentin and bone.

Fig. 11

PLM images under $100\times$ magnification show dentin (a) irradiated at $200\mathrm{Hz}$ without the water spray and with the water spray at $100\mathrm{Hz}$ (b). Bone irradiated at $50\mathrm{Hz}$ without the water spray (c) shows bone irradiated at $50\mathrm{Hz}$ with the water spray in (d). Note thermal damage zones in (a) and (c).

The wave plate introduces a phase retardation of $530\mathrm{nm}$ so that addition and subtraction colors introduced by the tissue birefringence induce resolvable changes in color. In the case of thermal damage, increased attenuation of the illumination light causes altered tissue to appear black, which is easier to discern under polarized light. Subtle changes in birefringence should introduce resolvable color changes. However, we were unable to resolve such changes and there was a sharp demarcation between normal tissue and the thermal damage zone.^{12, 13} A complete loss in birefringence of dentin without any other changes in optical properties would have resulted in a subtle change in color around the rim of the crater. In cross section, the char layer appears as loosely attached debris along the crater walls, and some is probably lost during the sectioning procedure. The remaining layer is of uniform thickness and it most likely represents increased scattering and absorption due to denaturation and partial thermal decomposition of the collagen matrix.

2.6.

Synchrotron Radiation—Fourier Transform Infrared Spectromicroscopy (SR-FTIR)

FTIR spectroscopy used in the specular reflectance mode is well suited for resolving thermally induced changes in dental hard tissue as a result of laser irradiation. Chemical changes in the mineral phase and loss of water and protein can be clearly seen as a result of thermal damage during laser irradiation. High spatial resolution $\left(5\mu \mathrm{m}\right)$ is achievable with a high brightness synchrotron radiation source such as the advanced light source (ALS) at Lawrence Berkeley National Laboratory. IR spectra of laser treated dentin and bone in both wet and dry conditions were obtained at ALS. All samples had been irradiated at $400\mathrm{Hz}$ . A Nicolet Magna 760 FTIR interfaced to a Nic-Plan IR microscope, equipped with a motorized sample stage connected to beam line 1.4.3 of ALS at Lawrence Berkeley National Laboratory was used to acquire spectra of the incisions produced with the lasers.

Plano-parallel sections approximately $1\text{-}\mathrm{mm}$ thick were cut at right angles to the laser incision. Each section was embedded onto $2\mathrm{in.}$ diameter LX-112 epoxy resin disks, which were subsequently serially polished with 1200 grit carbide and 3 and $1\text{-}\mu \mathrm{m}$ diamond suspensions. Specular reflectance spectra were acquired with a spatial resolution of $5\mu \mathrm{m}$ by scanning the $5\text{-}\mu \mathrm{m}$ spot imaged by the FTIR microscope across the laser intensity profile from the normal dentin to the resin outside the incision.

2.7.

Hollow Waveguide Transmission Measurements

Quartz hollow waveguides, from Polymicro Technologies, LLC (Phoenix, Arizona), $2\text{-}\mathrm{m}$ long and hollow-core diameters of 300, 500, 750, and $1000\mu \mathrm{m}$ were coupled to the laser using $f=50\text{to}125\mathrm{mm}$ ZnSe lenses and a laser energy and/or power meter was used to measure and record the transmitted energy. The measured transmitted energy reflects losses due to both coupling in and out of the waveguide in addition to the attenuation of the $2\text{-}\mathrm{m}$ length of the waveguide.

3.1.

Laser Ablation Rate Determination

Ablation studies were performed at $\lambda =9.3\mu \mathrm{m}$ with TEA laser pulses of $18\text{-}\mu \mathrm{s}$ duration (4% $\mathrm{C}{\mathrm{O}}_2$ gas mixture with 75% output coupler) for enamel, dentin, and alveolar bone. See Refs. 31, 32 for measurements for the laser pulses of $15\text{-}\mu \mathrm{s}$ duration (2% $\mathrm{C}{\mathrm{O}}_2$ gas mixture). The gain switched spike present in the beginning of the $18\text{-}\mu \mathrm{s}$ pulse initiated the onset of a plasma, which reduced the ablation effectiveness at high fluence. Ablation rates as a function of the incident laser fluence are shown in Fig. 6 for all three tissues: enamel, dentin, and alveolar bone. These rates are for the perforation of $200\text{-}\mu \mathrm{m}$ thick sections with no application of a water spray. The spot size was varied from $200\text{to}1080\mu \mathrm{m}$ by moving the sample away from the focal spot in order to achieve the targeted fluence, and the repetition rate was $1\mathrm{Hz}$ . The ablation rate of enamel saturated at a maximum rate of approximately $19\mu \mathrm{m}$ /pulse for irradiation intensities above $40\mathrm{J}∕{\mathrm{cm}}^2$ for the 4% gas mixture. The ablation rate for dentin and bone was approximately twice as high, $30\text{to}50\mu \mathrm{m}$ per pulse.

Fig. 6

Ablation rate versus fluence curves for porcine jaw bone (blue circle), human dentin (red cross), and enamel (gray triangle) carried out with the 4% $\mathrm{C}{\mathrm{O}}_2$ gas mixture with the 75% reflectivity output coupler. Each data point is the mean of three $\text{measurements}\pm \text{standard}$ deviation. Curves are drawn as a visual aid to represent the expected increase in the ablation rate with increasing fluence, followed by saturation of the ablation rate above the plasma initiation threshold (color online only).

3.2.

Residual Heat Deposition Measurements

Figure 5 shows an example of the residual heat deposition measurements performed at a nonablative fluence of $0.28\mathrm{J}∕{\mathrm{cm}}^2$ and an ablative fluence of $9.53\mathrm{J}∕{\mathrm{cm}}^2$ with the 4% of $\mathrm{C}{\mathrm{O}}_2$ gas mixture. Ablative laser pulses have a lower residual heat deposition than nonablative pulses due to energy transformed into kinetic and internal energy of the ablated material.

The mean and individual values of the residual heat $E_R$ that were measured for increasing irradiation intensities are plotted in Fig. 7 for enamel and dentin. For laser pulses near the threshold for ablation, the residual heat remaining in the tooth is high. Previous measurements have shown that ablation rapidly stalls after only a few laser pulses at irradiation intensities just above the ablation threshold.^{33, 34, 35} The initial few laser pulses ablate some material before ablation stalls and subsequent laser pulses contribute significantly to heat accumulation. As the incident fluence increases, ablation proceeds without stalling and the $E_R$ values converge to a minimum value due to plasma or debris shielding. The minimum $E_R$ value represents the most efficient ablation attainable for this wavelength and pulse duration in bovine enamel. This value for enamel is approximately 34%, which indicates that over 66% of the deposited laser energy is carried away by the ablative laser pulses delivered at $33\mathrm{J}∕{\mathrm{cm}}^2$ . The magnitude of the fluence at which the residual heat reaches a minimum value, 34%, is closely correlated with the maximum ablation rate for enamel, which occurs near $40\text{to}60\mathrm{J}∕{\mathrm{cm}}^2$ (Fig. 6). The ablation rate saturates because additional laser energy is absorbed or reflected by the highly opaque plasma, preventing further removal. The minimum $E_R$ value for dentin is significantly lower than for enamel, below 20% most probably due to the more volatile water and collagen content.

Fig. 7

The mean and individual values of the residual energy, $E_R\pm$ standard deviation $(n=3)$ for the ablation enamel (top) and dentin (bottom) measured using $\mathrm{C}{\mathrm{O}}_2$ laser pulses delivered in varying irradiation intensity.

3.3.

Peripheral Thermal Damage in Enamel, Dentin, and Bone (PLM and SR-FTIR)

Figure 8 shows reflected light images of incisions produced in enamel at a repetition rate of $300\text{-}\mathrm{Hz}$ scanning at $6\mathrm{mm}∕\mathrm{s}$ with and without the water spray at a fluence of $31\mathrm{J}∕{\mathrm{cm}}^2$ . The base of each incision appears to be uniformly melted. SR-FTIR spectra acquired in the area of the enamel in the melted region show that highly crystalline hydroxyapatite has been formed in that region even on the incisions produced with the water spray (Fig. 9 ). The loss of carbonate and narrowing of the phosphate peak near $1050{\mathrm{cm}}^{-1}$ are indicative of conversion to the more highly crystalline, pure phase hydroxyapatite that manifests a greater resistance to acid dissolution. Incisions produced at lower scanning speeds without water manifested greater thermal damage with the formation of chalky-white asperities that are most likely nonapatitic CaP phases formed due to disproportionation of hydroxyapatite at high temperature. SR-FTIR spectra of such asperities did not resemble spectra of hydroxyapatite.^{36, 37}

Fig. 8

Images of incisions in enamel produced without (a) and with (b) the water spray treatment ( $200\times \text{magnification}$ ). $\mathrm{C}{\mathrm{O}}_2$ laser pulses of $18\text{-}\mu \mathrm{s}$ duration at a fluence of $31\mathrm{J}∕{\mathrm{cm}}^2$ , scanning at $6\mathrm{mm}∕\mathrm{s}$ .

Fig. 9

Comparison of SR-FTIR spectrum of sound enamel (thick gray trace) versus spectrum of the incision area on the bovine enamel block produced scanned at $6\mathrm{mm}∕\mathrm{s}$ using a fluence of $31\mathrm{J}∕{\mathrm{cm}}^2$ with the water spray (thin black trace).

A comparison of incision depth in dentin for varying repetition rates showed no significant variation up to $400\mathrm{Hz}$ , the maximum repetition rate investigated. This demonstrates that high repetition rates can be employed without interference from ejected material in between pulses.

Minimal thermal damage to dentin and bone was observed in both wet and dry conditions for high scanning rates $(6\mathrm{mm}∕\mathrm{s})$ at the highest repetition rate $\left(400\mathrm{Hz}\right)$ . Figure 10 shows reflected light images of incisions produced in both dentin and bone at $400\mathrm{Hz}$ at a fluence of $80\mathrm{J}∕{\mathrm{cm}}^2$ . Very clean uniform incisions without discoloration were produced with use of the water spray. Charring occurred without the water spray. Thin sections were cut across all the incisions in dentin and bone for examination with PLM (Fig. 11 ). PLM showed thermal damage zones of less than $22\mu \mathrm{m}$ for dry dentin and less than $16\mu \mathrm{m}$ for wet dentin for all repetition rates investigated up to $400\mathrm{Hz}$ . Thermal damage zones for bone were slightly higher: $27\mu \mathrm{m}$ for dry bone and $18\mu \mathrm{m}$ for wet bone. Thicker sections were embedded and examined with SR-FTIR with a spatial resolution of $5\mu \mathrm{m}$ . Incisions were produced at a fluence of $80\mathrm{J}∕{\mathrm{cm}}^2$ . Results were also compared for lower irradiation intensities of 40 and $20\mathrm{J}∕{\mathrm{cm}}^2$ . The thermal damage zones at 40 and $20\mathrm{J}∕{\mathrm{cm}}^2$ were both smaller than those found at $80\mathrm{J}∕{\mathrm{cm}}^2$ at $400\mathrm{Hz}$ . Graphs of the thermal damage in dentin and bone for irradiation intensities of 80, 40, and $20\mathrm{J}∕{\mathrm{cm}}^2$ all at $400\mathrm{Hz}$ were reported in our SPIE proceedings paper.³²

Figure 12 shows reflected light images of incisions produced in dentin at $40\mathrm{J}∕{\mathrm{cm}}^2$ at a repetition rate of $400\mathrm{Hz}$ with $15.4\mu \mathrm{s}$ with and without the water spray. There is charring visible for the incisions produced without the water spray. The incisions produced with the water spray are much shallower than the dry incision pulses; however, they are clean without discoloration due to thermal damage.

Fig. 12

Reflected light images of cross sections of human dentin produced with and without the water spray at $400\mathrm{Hz}$ and $80\mathrm{J}∕{\mathrm{cm}}^2$ .

SR-FTIR spectra were acquired across the edge of the incisions shown in Fig. 12. Spectra taken at $5\text{-}\mu \mathrm{m}$ intervals across the incision wall as shown in Figs. 13 and 14 manifest few chemical changes due to peripheral thermal damage. These results show that within a resolution of $5\mu \mathrm{m}$ , minimal changes in the amide structure of the dentin could be detected. Therefore, FTIR showed no spectral changes within $5\mu \mathrm{m}$ around the periphery of the incisions of our samples in wet and dry conditions at $400\mathrm{Hz}$ .

Fig. 13

Sequential SR-FTIR spectra taken at $5\text{-}\mu \mathrm{m}$ intervals across the embedded dentin samples with incisions produced with the water spray, $400\mathrm{Hz}$ and $80\mathrm{J}∕{\mathrm{cm}}^2$ . The reflected light image contains a cross showing the position of each spectrum. The spectra are displayed progressing from the unmodified dentin (top spectrum) toward the center of the lesion. The peak near $1700{\mathrm{cm}}^{-1}$ marked by the gray line is due to the embedding resin and its rise indicates that the position is across the wall of the incision.

Fig. 14

Sequential SR-FTIR spectra taken at $5\text{-}\mu \mathrm{m}$ intervals across the embedded dentin samples with incisions produced without the water spray, $400\mathrm{Hz}$ and $80\mathrm{J}∕{\mathrm{cm}}^2$ .

3.4.

Hollow Waveguide Transmission Measurements

The transmission rates through $2\text{-}\mathrm{m}$ long 300-, 500-, 750-, and $1000\text{-}\mu \mathrm{m}$ quartz hollow waveguides (Table 1 ) were measured. The bending diameter for the hollow waveguides was approximately $45\mathrm{cm}$ . The highest transmission was achieved for the $750\text{-}\mu \mathrm{m}$ waveguide with a bending diameter of $45.5\mathrm{cm}$ .

Table 1

The transmission properties of quartz hollow waveguides of 300-, 500-, 750-, and 1000-μm diameter, a length of 2m , and a bending diameter of 45.5cm for λ=9.3-μm laser pulses of 18-μs duration. The focal length of the lens used for coupling into the waveguide is indicated in the left column.