Lasers in Surgery and Medicine 28:435±444 (2001)
A Laser-Abrasive Method for the Cutting of Enamel
and Dentin
Gregory B. Altshuler, PhD,* Andrei V. Belikov, PhD, Yury A. Sinelnik
Institute of Fine Mechanics and Optics, Department of Quantum Electronics and Bio-Medical Optics, 14 Sablinskaya st.,
Saint Petersburg, 197101, Russia
Background and Objective: This paper introduced a
new method for the removal of hard dental tissue based
upon the use of particles accelerated by laser irradiation,
which the authors have called the laser-abrasive method.
Study Design/Materials and Methods: The particles
used were sapphire as powder or an aqueous suspension.
The effect of the products of enamel ablation was also
investigated. The particles were accelerated using sub-
millisecond pulses of Er:YAG and Nd:YAG lasers. A
strobing CCD camera was used to measure the speed of
the ejected particles. The additional contribution of these
particles to the ef®ciency of laser ablation of enamel and
dentin was also investigated.
Results: The results showed that the enamel particles
produced by the beam±tissue interaction were also acce-
lerated by this process of ablation and were effective in the
removal of enamel and dentin. The use of an aqueous sus-
pension of sapphire particles increased the ef®ciency of
enamel removal threefold when compared with the use of
an Er:YAG laser with water spray.
Conclusions: The laser-abrasive method allowed for the
removal of enamel and dentin at speeds approaching those
of the high-speed turbine. Lasers Surg. Med. 28:435±444,
2001. ß 2001 Wiley-Liss, Inc.
Key words: abrasive particles; enamel cutting ef®ciency;
dentin; laser acceleration
INTRODUCTION
Laser removal of hard dental tissue using Er and CO2
lasers has found a wide range of clinical applications in
recent years. The effects of laser removal of hard tissues
are distinguished from those of conventional technology by
a lower level of tissue necrosis and a signi®cant reduction
in the perception of pain. A fundamental limitation of laser
technology however, is the speed of cavity preparation [1],
which is considerably slower than that of a high-speed
turbine. Progress for hard tissue applications in dental
lasers depends on their ability to equal or exceed the speed
of hard tissue removal using high-speed turbines. It is
possible to approach this speed with an Er:YAG laser [2, 3]
but to do so requires optimisation of the characteristics of
the laser system resulting in a signi®cant increase in price
of the laser.
This paper presents a new method for laser ablation,
which addresses the problem of the speed of hard tissue
ß 2001 Wiley-Liss, Inc.
removal. The essence of the method lies in the use of hard
particles accelerated by laser radiation. It has been shown
that using sapphire particles of 30±50 mm in size and a
velocity of 50±100 m/sec, a rate of enamel removal higher
than that of high-speed turbines can be achieved. Both air
abrasion and laser-cavity preparation have been shown to
be relatively painless in clinical practice and resulted in
clinically acceptable cavity preparations [4,5,6]. The air-
abrasive method has a high particle divergence, which
requires exact positioning of the handpiece tip in relation
to the cavity surface and has no tactile feedback, a signi-
®cant disadvantage to the clinician. These factors do
not apply when using a conventional drill or an optical
laser tip. The air-abrasive method also requires a large
volume of sapphire particles, which quickly ®ll in a
patient's mouth demanding excellent suction. The use of
rubber dam (the application of which can be painful
without local anaesthetic) is also recommended. The laser-
abrasive method potentially eliminates these problems,
as the use of an optical tip provides for tactile feedback,
and the volume of abrasive particles is signi®cantly
diminished.
In this paper the authors discuss the basic physics of
laser abrasion and present the results of experiments
using sapphire and hydroxyapatite particles accelerated
by Er:YAG and Nd:YAG laser radiation, as well as data
comparing the ef®ciency of removal of enamel and dentin
using both hydroxyapatite and sapphire particles.
1. Mechanism of Destruction of Enamel and
Dentin by Accelerated Hard Particles
The concept and methodology of the destruction of hard
materials by air or liquid abrasion is well-understood
[4,5,6,7]. The basis of this methodology lies in the elastic
impact of a particle against a solid material. At the point of
impact there are two main effects on a material, plastic
deformation and brittle or viscous splitting. The ®rst
effect, i.e., plastic deformation, leads to a hardening of the
material surface and an alteration to its surface rough-
ness. The second effect, i.e., brittle or viscous splitting,
leads to its destruction and removal [7].
*Correspondence to: Dr. Gregory B. Altshuler, Palmolar
Medical Technologies, 82 Cambridge St., Burlington, MA 01803.
Accepted 5 January 2001
436 ALTSHULER ET AL.

In dentistry, plastic deformation causes hardening of

the surface, decrease of porosity and a rise in the chemical
and bacterial resistance of enamel and dentin. Further
research is needed to thoroughly examine these observa-
tions. In the present paper the authors address the aspects
of hard tissue removal, which are related to brittle or
viscous splitting.
At the point of impact of the speeding abrasive particle
against the enamel surface, the most obvious place for
crack development to occur is at the inter-prism spaces,
which have a lower mechanical hardness than the enamel
prisms. The minimum velocity of a particle, required to
develop a crack in a material, which has a hardness comp-
arable to or less than that of the particle, can be calculated
using [7]. Our calculations have demonstrated that for the
particle sizes of 30 . . . 50 mm, the minimum velocity of a
particle must attain the order of 100 . . . 30 m/sec.
In the following section, the authors examine the two
main mechanisms of laser acceleration of particles, asym-
metric ablation and microexplosion.

2. Mechanisms of Laser Acceleration

of Hard Particles
From previously reported works [8] it is known that
laser radiation could be used for acceleration of solid-state
or liquid particles. During laser ablation, non-isotropic
removal of material (from one direction) imparts in the
ejected particles a mechanical pulse (recoil pulse) directed
in a direction opposite to that of the total pulse of the
products of ablation which cause the acceleration of the
particles. The basic principle of particle acceleration dur-
ing ablation is similar to that of jet engine propulsion.
Assuming that the light source illuminates the surface
of the particle non-isotropically (i.e., from one particular
direction), then once the ablation threshold is exceeded,
material from the surface will be removed. If ablation
occurs in the form of vaporisation of material from the
target, then a layer of saturated vapour of the target's
particle atoms appears on the irradiated surface of the
particle.
Another mechanism of acceleration is associated with
the microexplosion of the target material due to the action
of the laser pulse. Using a composite material as an
example: it consists of a hard particle core surrounded by
an absorptive matrix (Fig. 1). Assuming the matrix has a
lower threshold of laser ablation than the core, then under
the action of a laser pulse, a rapid heating and evaporation
of the matrix accompanied by the formation of a high-
pressure vapour occurs. Gradients of this pressure result
in the ejection of the hard particles that accelerate to a
high velocity. In that case an asymmetric ablation will be
observed and the force acting on the particles will be
proportional to the saturated vapour pressure and the
pressure related to the direct out¯ow of atoms from the
particle's surface.
The composite material may be considered as a suspen-
sion of hard particles, or it may be compressed into the
form of capsules, in either case, each hard particle core is
surrounded by a layer of photo-absorbing, easily melted
Fig. 1. A: Explosion of a suspension of compressed capsules of
abrasive particles due to the action of laser irradiation. Laser
explosion of a suspension of abrasive particles. B: Explosion of
a suspension of compressed capsules of abrasive particles due
to the action of laser irradiation. Laser explosion of abrasive
powder.
material, e.g., a polymer. Hard material with absorbing
micro inclusions is another example. Similar properties
were already demonstrated for enamel and dentin [9].
Enamel has a small volume of water (2±3%) located in
pores and near-surface microcracks. The Er:YAG laser
wavelength (l ˆ 2.94 mm) is absorbed by water approxi-
mately ®ve times more than by hydroxyapatite. This leads
to a strongly overheated vapour with pressure in excess of
1000 bar, resulting initially in local cracking and subse-
quently in an avalanche of cracks, mainly in the inter-
prism areas (Fig. 2). As these microcracks join, the enamel
disintegrates into small fragments, which are ejected by
the build up of the pressure of water vapour as well as by
the energy released during the destruction of the enamel.
MATERIALS AND METHODS
Acceleration of Sapphire Particles by Er:YAG
or Nd:YAG Laser Radiation
Sapphire (aluminium oxide) is one of the most suitable
materials that can be used as particles in the laser-
 A LASER-ABRASIVE METHOD FOR THE CUTTING OF ENAMEL AND DENTIN
Fig. 2. Clusters of cracks in enamel.
abrasive method. It is practically transparent in the range
of 0.3±5.5 mm; therefore, for these particles to be acce-
lerated by laser radiation within this spectral range, they
need to be surrounded by a coating which will absorb this
wavelength, or be placed within a composite matrix.
The present paper investigated two types of materials,
which could be used for this purpose: dry sapphire part-
icles and an aqueous suspension of sapphire particles. The
dry sapphire particles had a rough surface structure,
which was formed during the manufacture of the powder
and were capable of absorbing laser radiation. Laser explo-
sion of the powder occurred as a result of the evaporation
on the surface of the particles as previously mentioned.
Five particle sizes of sapphire were investigated: 12 mm,
27 mm, 40 mm, 160 mm, and 320 mm (``Crystal Mark Inc., A
Swam Technologies'', USA; ``Danville Engineering'', USA).
These particles are commonly used in industry for grind-
ing of hard materials.
The aqueous suspension contained sapphire particles of
12 or 27 mm carefully cleaned in a solution of phosphoric
acid. The experimental design for the study of the dyna-
mics of sapphire particle acceleration is shown in Figure 3.
Nd:YAG and Er:YAG lasers served as sources of laser
radiation. Both lasers operated in free-running mode. The
Nd:YAG laser radiation wavelength was 1.064 mm, laser
pulse duration measured at Full Width Half Maximum
(FWHM) was 300 microseconds. The Er:YAG laser wave-
length was 2.94 mm, and FWHM laser pulse duration was
300 microseconds. The laser radiation was focused by a
CaF2 lens onto the layer of powder or suspension which
437
Fig. 3. Experimental design for the measurement of the
acceleration of sapphire particles.
was 200 mm thick. The size of the laser beam in the focal
plane of CaF2 lens was 500  50 mm for the Nd:YAG laser
and 300  50 mm for the Er:YAG laser. At the point of
interaction between the laser radiation and the abrasive
particles, an intensive acceleration of particles in the
direction of the laser beam was observed. Images of the
plume of particles were recorded using a strobing CCD
camera. The strobing pulse was synchronised with the
laser pulse. A temporary shift of the strobing pulse relative
to the laser pulse could be continuously adjusted with an
accuracy of 1microsecond. In the majority of the experi-
ments the duration of the strobing pulse was 20 micro-
seconds.
The disintegrating particles were illuminated by plume
produced by the micro-explosion, which allowed the
determination of the direction, velocity, and degree of
divergence of the particles. The velocity of a particle was
de®ned as the length of the track divided on a duration of a
strobe pulse (illuminated part of particle's track in a shot
of the CCD camera). To evaluate the abrasive particle's
velocity at the front of the micro-explosion, the authors
used two photo-diodes, PD1 and PD2. Signals were recor-
ded from the plume in two areas: one in the centre of
the abrasive particle layer and the other at a distance of
L ˆ 9 mm. This length, divided by the temporal shift be-
tween scattering signals as seen on the oscilloscope, was
equal to the average velocity of the particles at the front of
the micro-explosion.
Both of the above methods for the measurement of par-
ticle velocity allowed only for the measurement of its
average value.
Use of Particles Accelerated by Laser Radiation
to Increase the Ef®ciency of Enamel Removal
As it was reported in the previous works, the ef®ciency
of enamel removal using submillisecond pulses of the
Er:YAG laser in a non-contact mode with water/air irriga-
tion of a processed surface reached a value of the order of
0.1 mm3/J [2, 3]. At a given ef®ciency, one laser pulse is
capable of removing a layer of enamel of up to 300 mm in
thickness.
The aim of these experiments was to determine whether
sapphire particles accelerated by Er:YAG laser radiation,
could increase the ef®ciency of enamel removal compared
with the conventional non-contact treatment with laser,
438 ALTSHULER ET AL.
Fig. 4. A drawing of an experimental design for the inves-
tigation of hard tissue removal using the laser-abrasive effect
with both aqueous suspension and powder.
using water irrigation. The experimental design is shown
in Figure 4.
In the experiment an Er:YAG laser beam with a pulse
energy density of 60 J/cm2, pulse duration of 300 micro-
second and a spot size of 300 mm was focused using a CaF2
lens onto the surface of the enamel of a freshly extracted
tooth. The teeth used (lower and upper ®rst and second
molars as well as central and lateral incisors), were extr-
acted for periodontal reasons and were free from other
defects. They were stored in a thymol solution until ready
for use, then rinsed under running water and cleaned with
a rubber cup and an aqueous solution of pumice using a
mechanical handpiece. Three effects of laser interaction
were examined. Firstly with a thin ®lm of water which was
sprayed onto the enamel surface prior to lasing (the cur-
rently used method). Secondly with a manually deposited
layer of an aqueous suspension of sapphire particles prior
to lasing and thirdly with a manually deposited layer of
powdered particles prior to lasing. The particle size for the
latter two methods was 12 mm. The thickness of water ®lm,
suspension ®lm, or powder was about 200 mm in all cases.
The aqueous suspension consisted of one part water to
nine parts abrasive particles.
Formation of High Velocity Hydroxyapatite
Particles During Enamel Ablation
by Er:YAG Laser Radiation
The aim of these experiments was to con®rm that abra-
sive particles could be created directly during the process
of laser ablation of enamel by submillisecond Er:YAG laser
pulses and that such particles had suf®cient velocity and
energy for the destruction of intact enamel and dentin.
The experimental design is shown in Figure 5. Laser
radiation was directed onto the surface of the enamel
through a dielectric (sapphire or fused silica) plate. Parti-
cles leaving the point of interaction ¯ew in a direction
opposite to the laser radiation and adhered to this plate
which was placed close to the tooth surface as shown in
Figure 5.
Fig. 5. Diagrammatic representation of the experimental
investigation of hydroxyapatite particles formed by laser des-
truction of enamel.
The distance between the tooth surface and the plate
varied between 0.1 mm and 10 mm. The particles, which
remained on the surface of the plate, were then photo-
graphed. In order to ascertain whether the particles of
hydroxyapatite accelerated by laser radiation could des-
troy enamel or dentin the following experiment was
designed. The essence of the experiment was that the
hydroxyapatite particles emitted from the laser crater
were re¯ected back onto the enamel destroying its surface
as seen in Figure 6.
RESULTS
We shall ®rstly consider the results of the acceleration of
sapphire particles by Er:YAG or Nd:YAG laser radiation.
The experimental design is shown in Figure 3. Typical
strobe-photographs of disintegrating particles are shown
in Figure 7. The results showed that a particle's dispersion
was delayed relative to the beginning of the laser pulse
(Fig. 7a). Typical delays were of the order of 30±60 micro-
seconds. The density of particles in the plume reached its
maximum at the peak of the pulse, at which point the
particles began to descend (Fig. 7c).
Figure 7c shows that the central plume was strongly
overexposed and could not provide suf®cient resolution of
an individual particle's trajectory. Figure 8, which was
Fig. 6. Experimental set up designed to investigate the ability
of abrasive hydroxyapatite particles accelerated by laser
radiation to remove hard dental tissue.
 A LASER-ABRASIVE METHOD FOR THE CUTTING OF ENAMEL AND DENTIN
Fig. 7. A: Typical photographs of a plume of sapphire par-
ticles (ù ˆ 12 mm) in a ®eld of a 300-microsecond pulse of an
Er:YAG laser with an energy density of 100 J/cm2. The size of
the frame is 20  20mm. The exposition for all of the photo-
graphs is 20 microsecond the temporal shift from the
beginning of the laser pulse is 40 microsecond, B: Typical
photographs of a plume of sapphire particles (ù ˆ 12 mm) in a
®eld of a 300-microsecond pulse of an Er:YAG laser with an
energy density of 100 J/cm2. The size of the frame is 20  20
mm. The exposition for all of the photographs is 20 micro-
second, the temporal shift from the beginning of the laser
pulse is 80 microsecond, C: Typical photographs of a plume of
sapphire particles (ù ˆ 12 mm) in a ®eld of a 300 microsecond
pulse of an Er:YAG laser with an energy density of 100 J/cm2.
The size of the frame is 20  20 mm. The exposition for all of
the photographs is 20 microsecond, the temporal shift from
the beginning of the laser pulse is 180 microsecond, D. Typical
photographs of a plume of sapphire particles (ù ˆ 12 mm) in
a ®eld of a 300 microsecond pulse of an Er:YAG laser with
an energy density of 100 J/cm2. The size of the frame is
20  20 mm. The exposition for all of the photographs is 20
microseconds, the temporal shift from the beginning of the
laser pulse is 300 microsecond.
439
taken with normal exposition, clearly shows the trajectory
of the particles. For large particles (160 mm and 320 mm), a
rotation around their centre of gravity was observed, along
with a rectilinear movement. Axial rotation with frequen-
cies of up to 500 kHz contained a signi®cant fraction of the
total kinetic energy. In the strobe-photographs this rotat-
ion was seen as a helix-shaped trajectory (Figure 8). The
rectilinear nature of the bevelled trajectories in the strobe-
photographs con®rmed that braking of particles occurred
mostly due to air resistance, and that gravity did not play a
signi®cant role. The acceleration of particles occurred due
to a microexplosion of the powder or suspension.
The authors investigated in detail the threshold value
Eth of the microexplosion (acceleration), the maximal speed
and kinetic energy of the particles of various diameters
whilst irradiating both the powdered particles and the
aqueous suspension of particles independently with two
different lasers (Er:YAG and Nd:YAG). Maximum speed of
the particles decreased with an increase in the thickness of
the layer. The value of the layer thickness was close to
optimal for the investigation of different sized particles.
Increasing or decreasing the layer thickness decreased the
particle's average and maximal velocities. The results are
shown in Table 1.
The graph in Figure 9 shows the typical dependence of
particle's velocity on the energy density of the laser pulse.
From dependence given in Figure 9 it can be seen that
there is an Energy density threshold value Eth when the
particles motion starts. In Table 1, the typical threshold
values of the commencement of the motion of particles are
given. In the range of Eth±7Eth there is a steady growth in
velocity.
In order to evaluate the ef®ciency of the sapphire
particle's ability to remove enamel, the experiment whose
design is shown in Figure 4 was performed. SEM of craters
in enamel formed by Er:YAG laser radiation are shown in
Figure 10. All three craters were formed after action of
Er:YAG laser pulse with an energy density of 60 J/cm2
focused to a spot size of 300 mm on the enamel surface.
Crater 1 was formed by the action of a single pulse of an
Fig. 8. Photographs of sapphire particles with a temporal
delay of approximately 180 microseconds (corresponds to
Fig. 7c) in a ®eld of 300 microsecond Er:YAG laser pulse with
an energy density of 60 J/cm2. A helix-shaped trajectory of a
large particle (with a diameter of 160 mm) can be clearly seen.
The size of the frame is 20  20mm. The exposition is 20
microsecond.
440 ALTSHULER ET AL.

TABLE 1. Results of Experiments for the Er:YAG and Nd:YAG Lasers

Threshold of micro Maximum kinetic

explosion Maximum velocity energy
Size of sapphire (J/cm2) (m/s) (mJ)
Type of particles
abrasive (mm) Er:YAG Nd:YAG Er:YAG Nd:YAG Er:YAG Nd:YAG

Aquesous 12 1.5 125 600 30 0.7 0.0016

suspension 27 14 63 430 45 0.4 0.04
12 3 50 450 45 0.4 0.0036
Powder 40 9 85 120 40 0.8 0.1
160 26 95 210 35 180 5

Er:YAG laser with water spray, crater 2-by the action of a

single pulse of an Er:YAG laser with abrasive powder and
particles with a diameter of 12 mm, crater 3-by the action of
a single pulse of an Er:YAG laser with an aqueous
suspension and particles with a diameter of 12 mm.
Measurement of their volume showed that the ef®ciency
of enamel removal using an aqueous suspension of parti-
cles was 2±3 times greater than when only the laser with
water spray was used, and 1.5±2 times greater than when
the particles were used without water.

Fig. 10. SEM of craters formed in enamel by an Er:YAG

laser., 1-single pulse of an Er:YAG laser with water spray; 2-
single pulse of an Er:YAG laser with abrasive powder and
particles with a diameter of 12 mm; 3-single pulse of an
Er:YAG laser with water suspension and particles with a
diameter of 12 mm.

One can expect that the character of change of ef®ciency

Fig. 9. A: Dependence of an abrasive particle's (aqueous sus-
pension) velocity on the Er:YAG laser energy density. Par-
ticles size: 27 mm. B: Dependence of an abrasive particle's
(aqueous suspension) velocity on the Er:YAG laser energy
density. Particles size: 160 mm.
removal of dentin when using the abrasive powder will be
similar to the character of change of ef®ciency removal of
enamel.
Figure 11 shows the SEM of the crater formed by laser
pulses and an aqueous suspension of 12-mm particles. The
structure of undamaged dentin tubules can be clearly
seen. When dry sapphire particles were used, adhesion of
these particles to the cavity walls was observed. The
quality of a cavity wall in dentin when abrasive powder
was used is practically identical to the quality of a cavity
wall formed by the action of a single pulse of an Er:YAG
laser with water spray. The quality of a cavity wall can be
evaluated in Figure 11b.
The following experiment demonstrated an effect of a
hard material's destruction by sapphire particles acceler-
ated by laser pulse. Instead of the tooth, a sapphire plate
was ®xed in the focus of an Er:YAG laser beam. In
comparison to enamel, the absorption of sapphire at the
wavelength, l ˆ 2.94 mm is very low. During the action of
 A LASER-ABRASIVE METHOD FOR THE CUTTING OF ENAMEL AND DENTIN 441

Fig. 12. Photograph of hydroxyapatite particles ®xed onto the

surface of the sapphire plate (the distance between the plate
and enamel is 1 mm).

order of 20 mm in diameter. The divergence of the beam of

Fig. 11. A: SEM of the walls of a crater formed in dentin by an
Er:YAG laser with 12-mm particles in an aqueous suspension.
The energy density was 60 J/cm2. General view. B: SEM of the
walls of a crater formed in dentin by an Er:YAG laser with 12-
mm particles in an aqueous suspension. The energy density
was 60 J/cm2. Base of the crater.
enamel particles was about 30 . The velocity reached by
the 20-mm enamel particles was 400 m/sec and their
energy was approximately 1 mJ.
The effect of the re¯ection of ablated enamel particles
from the sapphire plate as observed in the experiment
shown in Figure 6. Figure 13 shows two craters, both
formed by single laser pulse, one by laser alone and the
other by re¯ection of hydroxyapatite particles. Both
craters are of approximately the same size. In other words,
the kinetic energy of the products of enamel ablation was
suf®cient for enamel removal in volumes similar to that

Er:YAG laser pulses onto the sapphire plate, no destruc-

tion of the sapphire was observed up to energy densities of
4 kJ/cm2. Addition of water spray forming a water ®lm on
the surface did not show any destruction. Addition of 27-
mm sapphire particles to the water spray however, caused
the destruction of the sapphire plate at energy densities of
0.4 kJ/cm2.
The formation of high-velocity hydroxyapatite particles
during enamel ablation by Er:YAG laser radiation obser-
ved in experiment shown in Figure 5. When using the
fused silica plate facing enamel, signi®cant damage caused
by the products of ablation was observed after only several
laser pulses. In the case of the sapphire plate, noticeable
damage was seen only after prolonged exposure. Figure 12
shows the base of sapphire plate located at a distance of 1
mm from the enamel surface during ablation.
An analysis of the image shown in Figure 12 allowed an
evaluation of the parameters of the beam of enamel
particles emitted from the laser crater, as well as their Fig. 13. SEM of enamel surface with two craters, one (1)
individual size. The size of the largest particles was 200 mm formed by laser alone, and one (2) formed by re¯ected hy-
in diameter, although they were most commonly of the droxyapatite particles.
442 ALTSHULER ET AL.
removed by laser alone. Crater 2 is of an elliptical shape
due to the direction of the trajectory of the particles.
DISCUSSION
The results showed that under the action of submilli-
second laser pulses it is possible to accelerate both
sapphire particles and the hydroxyapatite products of
laser ablation of enamel and dentin to speeds of hundreds
meters per second. The mechanical impact of these
particles can be successfully used for enamel and dentin
destruction. The results showed two methods of achieving
successful abrasion. The ®rst method utilised dry abrasive
particles or a suspension of these particles and the second
utilised the abrasive particles of re¯ected hyroxyapatite.
The abrasive effects of sapphire particles accelerated by
high pressure of air, were investigated as these have a
history of successful use in hard tissue removal with
commercially available air abrasion units [6]. Sapphire
has an extremely high hardness, which exceeds the
hardness of most other materials, and is also biologically
inert. In the range of l ˆ 0.2±5.5 mm sapphire is almost
transparent consequently, ablation of pure sapphire is
minimal for the majority of commercial lasers. Small
sapphire particles however, may have areas of defects,
which can affect absorption of laser irradiation. On the
particle's surface, in microcracks, water molecules can be
stored.
The results showed that under the action of radiation of
both a Nd:YAG and an Er:YAG laser on the powders used,
a microexplosion and acceleration of sapphire particles
was observed. The ablation threshold for the Er:YAG laser
was 10X less than that for the Nd:YAG laser. This
discrepancy can be explained by the fact that the water
absorption coef®cient [10] for Er:YAG laser is 105-times
greater than that for the Nd:YAG laser. Therefore, even
minor concentrations of water on the surface of particles
can produce signi®cant differences as shown in the obser-
ved thresholds.
The maximum speeds of particles differ as well, by
approximately 10 times at comparable energy densities.
The same relationship is observed during ablation of an
aqueous suspension of sapphire particles. Hence, the
Er:YAG laser is considerably more ef®cient for the acce-
leration of sapphire particles.
If highly absorbing particles (for example: graphite), or a
highly absorbing liquid (for example: an aqueous dye
solution) were used, the ef®ciency of sapphire particle
acceleration with the Nd:YAG laser or with any other laser
could be comparable to that of the Er:YAG laser.
The results of the experiments conducted showed that
the speed of the particles was almost directly proportional
to a difference between the laser's energy density and
threshold of microexplosion, up to 7X the microexplosion
threshold. The microexplosion threshold for the powder
increased as the size of sapphire particles increased.
Under the action of Er:YAG laser radiation the micro-
explosion threshold of the aqueous suspension for 12-mm
sapphire particles differed approximately 10X from that of
the aqueous suspension for 27-mm sapphire particles. The
microexplosion threshold for the particles of smaller sizes
was even lower with a value of only 1.4 J/cm2. For the
Nd:YAG laser an opposite relationship was observed.
The highest particle speeds (of about 600 m/sec) were
observed when using suspensions of 12-mm sapphire
particles and Er:YAG laser pulses. As particle size incre-
ased, the velocity decreased, but the kinetic energy incre-
ased (Table 1). The maximum kinetic energy was recorded
at 180 mJ for 160-mm particles using an Er:YAG laser. For
enamel destruction, the energy of a particle should exceed
0.17±0,4 mJ. Table 1 shows that this was achieved using
sapphire particles accelerated by Er:YAG laser pulses.
SEMs in Figure 10 showed the effect of a thin ®lm of
water, of dry powder, or aqueous suspension of sapphire
particles deposited on enamel surface. Each crater was
formed under action of ®ve pulses of the Er:YAG laser with
identical energy. Measurements of these craters showed
that the use of dry sapphire particles on enamel increased
enamel-removal ef®ciency by a factor of 1.5±2 compared to
water ®lm. The use of an aqueous suspension of particles
lead to an increase of enamel ef®ciency by 2±3 times when
compared with the Er:YAG laser using water spray only.
Using an aqueous suspension of sapphire particles for the
Er:YAG laser with a pulse duration of 300 microseconds
and an energy density of 100J/cm2 the enamel removal
ef®ciency was 0.3 mm3/J.
The cooling effect of the aqueous suspension should not
signi®cantly differ from that of the water-spray method
alone; therefore, laser radiation with a pulse energy of
0.35 J and a repetition rate of 10 Hz (mean power: 3.5W)
should be no more detrimental to the pulpal tissue than
that of using water spray alone (as is used in current
technology). At these parameters the speed of enamel
removal constitutes a value of 0.3 mm3/J  3.5 W 1 mm3/
sec, which approximates the typical speed of enamel
removal by a conventional high-speed turbine.
The results also showed that apart from an increase in
the speed of enamel removal, the damage to the crater
walls was no different to that produced by the Er:YAG
laser with water spray alone, showing no cracking or
melting of crater walls as seen in Figure 11. This crater
was formed with the Er:YAG laser and a spray in the form
of an aqueous suspension of sapphire particles. Figure 11
shows that the dentinal tubules remained undamaged.
From the above results it can be concluded that the
application of a spray of an aqueous suspension of sapphire
particles (aluminium oxide) allowed an increase in the
speed of enamel removal by a factor of 2±3 without a
change in laser power. The spray performed two functions,
tooth cooling and delivery of abrasive particles to the
target tissue. Both laser abrasion and laser ablation with
the Er:YAG laser preserve natural tooth structure. It may
be possible that both laser abrasion and laser ablation may
negate the need for acid etching. In contrast to air
abrasion, in the laser abrasion process, the abrasive parti-
cles are contained in an aqueous solution, therefore, a
signi®cant amount of the unpleasant and potentially
unhealthy cloud of abrasive particles is eliminated or
more easily dealt with via normal suction procedures. The
 A LASER-ABRASIVE METHOD FOR THE CUTTING OF ENAMEL AND DENTIN
number of particles is also signi®cantly reduced in the
laser abrasive process.
Enamel ablation using the Er:YAG laser occurs mainly
due to microexplosions in areas of high water concentra-
tions leading to the formation of microcracks [9,11] as seen
in Figure 2. This is due to the fact that water is absorbed by
the Er:YAG laser wavelength 5±8 times more than is
hydroxyapatite therefore, the formation of high pressure
within these areas due to overheated steam occurs at
considerably smaller energies, than hydroxyapatite melt-
ing and evaporation [12]. Since hydroxyapatite does have
relatively high absorption of the Er:YAG wavelength, the
heating of the solid phase of enamel takes place along
with that of water especially at large energy densities.
The dominant mechanism however is due to water heating
and microcrack formation which is shown in the results of
experiments represented by the design seen in Figure 13.
In these experiments enamel ablation products were
collected and analyzed. Figure 12 shows that these pro-
ducts are the particles of hydroxyapatite up to 200 mm in
size with the main size being 20 mm. Melting was observed
on the surface of certain particles which indicated that
hydroxyapatite heating and melting also occurred. How-
ever, this mechanism is not the dominant one. A small
quantity of water (about 2±7%) is always located in enamel
mainly in the interprism spaces. Water, located in
microcracks on the enamel surface also plays a signi®cant
role as a site for microexplosions since, water irrigation
between laser pulses ®lls the microcracks formed after
ablation. By way of comparison, in the case of a CO2-pulsed
laser an opposite effect should occur, as the absorption
coef®cient of water in range of 9.3±10.6 mm is much lower
than that of hydroxyapatite. The products of enamel
ablation for the CO2-pulsed laser should be the atoms
and ions or liquid drops of hydroxyapatite. For this reason
the CO2-pulsed laser can not be used for the acceleration of
hydroxyapatite particles for the purpose of enamel des-
truction.
As shown in the results, the velocity of hydroxyapatite
particles at ablation can reach several hundreds of meters
per second. For 20-mm particles the maximum velocity
observed was 400 m/sec which corresponds to energy of
1 mJ (energy suf®cient for enamel destruction should such
a particle collide with the enamel surface). An important
observation was that the main stream of particles leaving
a laser crater during ablation was concentrated within a
relatively narrow range of approximately 30 . This was
shown in the ®nal experiment the design of which can be
seen in Figure 6.
The formation of the second crater shown in Figure 13
can be explained only by the action of particles accelerated
by the laser pulse. The light nature of this crater excludes
it as being due to as enamel re¯ection coef®cient for the
Er:YAG laser is very low. After re¯ection from the sap-
phire plate (re¯ection coef®cient is approximately 0.3), the
energy density is decreased more than 300 times and lies
below the ablation threshold. Thus, the formation of the
second crater can only be due to the particles of ablated
enamel re¯ected from the plate's surface. The microhard-
443
ness of hydroxyapatite particles is equal to or exceeds the
microhardness of enamel, which besides hydroxyapatite
contains a protein component. Therefore at suf®cient
speeds and with suf®cient kinetic energy, the hydroxya-
patite particles should be able to destroy enamel as well as
the sapphire particles accelerated by compressed air. This
result showed that the velocity of particles leaving the
laser crater was suf®cient for enamel destruction following
re¯ection from the sapphire plate. What remains unclear
however, is what happens to these particles when they
collide with the surface of the sapphire plate. Under what
conditions and to what degree is this impact elastic? It is
clear that under certain conditions these particles have a
completely inelastic impact and adhere to the surface of
sapphire plate. This may cause the destruction of the
sapphire plate surface by the next pulse of the Er:YAG
laser as the deposited particles of hydroxyapatite have
considerable absorption and these particles of destroyed
plate thus formed may also add to the destruction of the
enamel surface. We have examined the energy balance
during enamel ablation with Er:YAG laser pulses. At
energy densities of 75±100 J/cm2 the energy remaining in
a tooth (as a heat) constitutes a value of about 40% [13].
The remaining 60% is shared among the kinetic energy of
the ablation products, the thermal energy of the ablation
products and the decomposition energy used in destroying
the missing volume of enamel.
If we assume that the particles are heated to melting
point DT ˆ 103K and that the enamel breaks into cubic
fragments with an average size of 20 mm per side of the
cube, then the speci®c decomposition energy can be deter-
mined as 200 J/m3. At ablation ef®ciency 10±10 m3/J, it
is easy to demonstrate that the share of energy spent for
enamel decomposition is very small and constitutes
2  10 ÿ 6 % of deposited laser energy. In consideration
that 20% of deposited laser energy were spent on heating
the products of enamel ablation and the energy of enamel
decomposition is extremely low, one can calculate that the
kinetic energy of fragments of enamel particles leaving the
destruction zone is about 40% of deposited laser energy. In
this case, the velocity of some particles is enough for
making an extra destruction at secondary impact. This
argument serves as an additional explanation for the for-
mation of the second crater seen in Figure 13, which shows
that the volume of the second crater is approximately the
same as that created by the laser itself. We can, therefore,
predict that using the kinetic energy supplied by the
products of ablation, the ef®ciency of enamel removal can
be at least doubled. Re¯ection and recycling of the
particles of ablation would require a special tip attached
to the handpiece, the discussion of which is outside the
scope of this paper.
CONCLUSIONS
This paper introduces a new method for the destruction
of hard dental tissue named by the authors as the laser-
abrasive method. The method can be utilised in two ways:
®rstly, using hard particles delivered from an external
source to the target area and secondly, using particles
444 ALTSHULER ET AL.
formed as products of the ablation process itself. In both
instances, the actual process of laser ablation accelerates
the particles. The method described utilise the generated
kinetic energy of the particles for hard tissue destruction
in addition to the process of laser ablation as it is generally
understood which allows for a 2±3 times increase in the
speed of destruction of enamel and dentin for the Er: YAG
laser. The resulting cavity walls are comparable to the
conventional laser method showing no cracking or melting
in areas adjacent to the laser crater. The authors also
argue that using this method the intrapulpal temperature
changes should be compatible to that of the conventional
laser method although this work still needs to be done.
ACKNOWLEDGMENTS
The authors are grateful to Ms O.B. Chistyakova (Laser
Center, St. Petersburg) and Dr. I. Cernavin, BDS (Adel),
MDSc (Melb) Prosthodontist, Senior Fellow School of
Dental Science, Faculty of Medicine Dentistry and Health
Sciences University of Melbourne, Melbourne, Australia,
for their help to prepare this article.
REFERENCES
1. Jelinkova H, Dostalova T, Dolezalova L, Krejsa O, Kubelka J,
Prochazka S. Comparison of preparation speed of Er: YAG
laser and conventional drilling machine, Proceedings of
SPIE: Lasers in Dentistry III, Febr. 08±14, 1997, San Jose,
CA, V. 2973. 2±10.
2. Altshuler GB, Belikov AV, Erofeev AV, Sam RC. Optimum
regimes of laser destruction of human tooth enamel and
dentin, Proceedings of SPIE: Lasers in Orthopedic, Dental
and Veterinary Medicine, Jan 16±18, 1993, Los-Angeles,
California, V. 1880. p. 101±107.
3. Altshuler GB, Belikov AV, Erofeev AV. Laser treatment of
enamel and dentin by different Er-lasers, Proceedings
of SPIE: Laser Surgery: Advanced Characterization, Ther-
apeutics and Systems IV, Jan 22±25, 1994, Los-Angeles,
California, V. 2128. p. 273±281.
4. Goldstein RE, Parkins FE. Using air abrasive technology to
diagnose and restore pit and ®ssure caries. J Am Dent Assoc
1994;126(7):761±765.
5. Christensen G. Cavity preparation: cutting or air abrasion? J
Am Dent Assoc 1996;127(11):1651±1654.
6. Goldstein RE, Parkins FE. Air abrasive technology: Its new
role in restorative dentistry. J Am Dent Assoc 1994;125(5):
551±557.
7. Smanev VA, Zulepov OP, Mesheryakov OV. Jet-hydro-
abrasive processing of GTD mechanisms. Moscow: Machinos-
troenie; 1995.21±22.
8. Ascarian GA, Moros EM. Recoil pulse during laser ablation.
Sov J Theor Exp Phys V. 1962;43:2787±2795.
9. Altshuler GB, Belikov AV, Erofeev AV, Egorov VI. Simula-
tion of laser destruction of hard tooth tissues. Proceedings of
SPIE: Dental Applications of Lasers, Aug 29±Sept 03, 1993,
Budapest, Hungary, V. 2080. 10±19.
10. Zolotaryov VM, Morozov VN, Smirnova EV. 1984. Optical
constants of natural and technical environments, Leningrad,
Chimiya.
11. Belikov AV, Erofeev AV, Shumilin VV, Tkachuk AM. Comp-
arative study of the 3 mkm laser action on different hard
tooth tissue samples using free running pulsed Er-doped
YAG, YSGG, YAP and YLF lasers, Proceedings of SPIE:
Dental Applications of Lasers, Aug 29±Sept 03, 1993,
Budapest, Hungary, V. 2080. p. 60±67.
12. Altshuler GB, Belikov AV, Erofeev AV. Comparative study
of-contact and noncontact operation mode of hard tooth tissues
Er-laser processing, 5th Congress of the International
society for Lasers in Dentistry, May 5±9, 1996, Jerusalem,
Israel, Monduzzi Editore, International Proceedings Division,
p. 21±26.
13. Altshuler GB, Erofeev AV. Modern Optics and Dentistry, In:
Leo Misserendino, Robert. M. Pick, editors. Lasers in
Dentistry. Quintessence Publishing; 1995. p. 283±299.