|
1.IntroductionThe concept of selective targeting of strongly absorbing structures in weakly absorbing tissues was introduced by Anderson and Parrish.1 If heat is generated within an absorbing target more rapidly than heat can diffuse away, high temperatures can be obtained, leading to vaporization when applying appropriate energy. The resulting bubbles around laser-heated intracellular pigment particles cause spatial confined mechanical damage to the surrounding tissue.2 A prominent example of selective targeting is the retina. The aim of the selective retina treatment3 4 5 (SRT) is to selectively damage dysfunctional areas of the retinal pigment epithelium (RPE), which is a monocellular layer heavily loaded with pigment particles, while sparing the adjacent photoreceptors. Due to short pulsed laser exposure, bubbles form around the pigments in the RPE. However, large bubbles might also damage the neighboring photoreceptors causing scotoma.6 SRT is thought to share the therapeutic benefits from conventional photocoagulation, without affecting the function of the photoreceptors in the irradiated areas and avoiding loss of vision. The commonly used laser pulse duration in clinical SRT is 1.7 μs. Other biomedical applications, in which bubble formation around laser-heated microparticles is involved, are proposed for anticancer therapy. Originally unpigmented tumor cells are labeled selectively by bioconjugated light-absorbing exogenous microbeads. Selective destruction of cells, which contain these absorbers, can be triggered due to laser-induced microbubbles.7 8 9 10 Furthermore, bubbles forming around laser-heated micro- and nanoparticles can serve as contrast agents in optoacoustic tomography.11 Since high temperatures and high pressures can be obtained around laser-irradiated microparticles, chemical reactions might also be triggered.12 Therefore, knowledge of the particle’s surface temperature required for bubble nucleation is essential to estimate whether photothermal damage (thermal disintegration of the particle itself or thermal denaturation of proteins in the vicinity of the particle) or photomechanical effects (bubbles) cause cell damage.13 14 15 In this paper, we focus on the bubble nucleation around absorbing microparticles, especially melanosomes, which are the main pigments of the RPE, to achieve a detailed understanding of microbubble formation in SRT. A few models of bubble dynamics around laser-heated melanosomes have been developed in the past with regard to laser safety considerations.16 17 18 They improved understanding of laser-induced retinal damage, but there is still a lack of experimental data to validate these models. Therefore, the goal of this study is the experimental investigation of the nucleation temperature and threshold radiant exposures for bubble formation around melanosomes and other microbeads irradiated by nanosecond and microsecond laser pulses. The experimental bubble formation thresholds further enable the estimation of light absorption in RPE melanosomes. 2.Material and Methods2.1.MicroparticlesWe use porcine retinal pigment epithelial melanosomes. Enucleated eyes are obtained from a local slaughterhouse. The eyes are dissected, the vitreous is removed, and purified water, which has been degassed and demineralized, is added. The neural retina is peeled off after a few minutes. Subsequently, the superficial RPE is removed by a stiff brush and purified by filter paper (Schleicher & Schu¨ll 595 1/2, retention range 4 to 7 μm) to separate the melanosomes, which are released by expired cells, from cellular debris. As observed by light microscopy, porcine melanosomes are rotational ellipsoidal with mean diameters of d1=d2=0.8±0.1 μm and d3=2.3±0.6 μm. Additionally, we used polyethylenimine-coated spherical gold beads with a diameter of 0.5 μm (MG-2/1, Chemicell GmbH, Germany) and spherical magnetic silica beads (SiMAG-EP, Chemicell GmbH, Germany), which contain 60% magnetite crystals (<1 μm) embedded in a silica matrix. The surface of the SiMAG-EP beads is uncoated and contains hydrophilic silanol groups. The diameter of the SiMAG-EP particles is about 3 μm. SiMAG-EP and MG-2/1 beads were investigated to compare the results obtained for melanosomes to other particles. The 25 μL of the aqueous microparticle suspension, which contains 0.1% agarose for immobilization of the particles, is given on an object slide for the experiments. The fluid reservoir is prevented from drying by using Gene Frames (Abgene, United Kingdom), which seal the reservoir and keep a distance of about 250 μm between the object slide and cover slip. Temperature-dependent measurements are performed in a cuvette with a depth of 4 mm, which enables us to adequately place the thermocouple of a digital thermometer (Lutron TM-914C, Type-K thermocouple). The cuvette can be heated electrically by a resistor and cooled by a peltier device. 2.2.Experimental SetupThe experimental setup for microparticle irradiation is shown in Fig. 1. The suspension, which contains microparticles, is observed by means of a video microscope using a standard 40× microscope objective. The microparticles are irradiated by a frequency-doubled Q-switched Nd:YAG (12 ns at 532 nm) and a Nd:YLF laser (527 nm), alternatively. The pulse duration of the modified Nd:YLF laser (Quantronix, Model 527DP-H) can be adjusted in the range of 200 ns to 5 μs by an active feedback system controlling the Q-switch.19 The pulses are delivered by a fiber [Ceram Optec, Optron UV-A 105/125A/250, numerical aperture NA=0.22]. A top-hat beam profile at the specimen is achieved by imaging the fiber tip into the suspension. The fiber has a length of 200 m to reduce spatial intensity modulation due to speckles below 15%. The circular irradiation spot has a diameter of 39.5 μm. A photodiode (Centronics, AEPX 65) with a current-integrating circuit is used to determine the pulse energy. The photodiode is calibrated by an energy meter (Ophir Optronics, Laserstar). The pulse energy is adjusted by a variable attenuator. A probe HeNe laser (Laser Graphics LK 8623, 632.8 nm) is coupled by a dichroic mirror into the optical path of the irradiation beam. The HeNe laser is chopped synchronized to the irradiation laser pulses with a duty cycle of 10 μs to 125 ms to avoid additional heating of the particles. The Gaussian radius of the probe beam is about 1.5 μm at the specimen. The probe transmission is detected in an open aperture geometry by a photodiode (EG&G, FND-100) and a 100-MHz current amplifier (femto, HCA-100M-50K-C). Fast flash photographs are taken by a CCD (JAI, CV-A11) and a frame grabber (National Instruments, IMAQ 1409). A N 2 -pumped dye-laser (Laser Science, VSL-337-ND, DLM220) with a pulse duration of 3 ns serves as flash. Using the dye coumarin 102, the emission of the laser is tuned by an internal grating to 490 nm. The dye laser is triggered at an adjustable delay related to the irradiation pulse. The pulse energy of the dye laser at the particle suspension is below 1 μJ per 200 μm-diam spot, resulting in a radiant exposure of less than 3 mJ/cm2. Irradiation and illumination pulses are monitored by the same photodiode (Centronics, AEPX 65). Scattered light of the Nd:YLF/Nd:YAG laser is blocked by a short-pass filter in front of the CCD. The irradiation pulse shape; the illumination pulse shape; the irradiation pulse energy, which is taken from the calibrated photodiode; and the transmitted HeNe intensity are recorded by a digital oscilloscope (LeCroy, Waverunner LT374). 2.3.Thermal ModelingParticle heating due to the absorbed laser radiation is modeled with the aid of Mathematica 4 (Wolfram Research, Inc.). The absorption efficiency Q abs of a spherical particle is given by the ratio of the absorption cross section σ abs =E abs /H (E abs , absorbed energy; H, radiant exposure) to the geometric cross section of the particle where R is the particle radius. The absorption efficiency Q abs =1.05 of the spherical gold beads in water is calculated according to Mie theory20 using the refractive index of gold in water [m Au,H 2 O =0.59 to 1.67i at λ vac =525 nm (Ref. 20)] and a wavelength of λ H 2 O =λ vac /n H 2 O =395 nm. Since a variation in the particle size, magnetite content and magnetite distribution in the silica matrix of the SiMAG-EP beads leads to significant differences in the optical properties of the individual particle, no modeling of the absorption of the SiMAG-EP beads was carried out. The absorption of RPE melanosomes is not well known and can be determined from the experiments by thermal modeling (Secs. 3.2 and 3.3). The heat equation where T is temperature increase, κ is thermal diffusivity, ρ is density, cp is specific heat, A is input power density (see Table 1) for a spherical particle, whose thermal properties differ from the surrounding medium, has been solved by Goldenberg and Tranter.21 This solutionTable 1
Since Goldenberg’s solution assumes spatial and temporal homogenous heat deposition within the particle, the model was extended to arbitrary laser pulse shapes. Green’s function, which is the temperature response for an infinitesimal short laser pulse, can be approximated at a given radius r by using a small time step Δt. A convolution of Green’s function with the time-dependent power density of the absorbed light where I(t) is laser intensity, which is derived from the laser pulse shape, results in the temporal temperature course at a given radius r: The condition of spatial homogeneous heat deposition within the particle is fulfilled, if the particle is optical thin (1/μ≫R, where μ is the absorption coefficient of the particle) or if the time of thermal relaxation is much smaller than the laser pulse duration (R2/4κ1≪τ laser ). Thermal parameters and the time of thermal relaxation of the particles can be found in Table 1.3.Results3.1.Threshold for Bubble FormationSurface heating of the particle due to the probe HeNe laser is calculated to be below 10 °C for all particles in the first 5 μs after the HeNe is switched on. The bubble is usually probed in this time interval. The particles are irradiated with pulse durations of 12 ns, 240 ns, and 1.8 μs. Above a certain threshold of radiant exposure the probe transmission signals contain dips due to scattering of the HeNe light at the bubble. Coincidence from transmission measurements with the occurrence of bubbles is obtained by simultaneous fast flash photography (Fig. 2). The photographs reveal, that at threshold radiant exposure semispherical microbubbles form only on a part of the particle surface (Fig. 3). Photographs during consecutive irradiation taken at the same delay demonstrate, that the sites of nucleation on an individual particle are reproducible (Fig. 4). No hysteresis of the thresholds for bubble formation is found during repeated irradiation of the same particle near bubble threshold [Fig. 5(a)]. No morphological alterations of the particles have been observed by light microscopy after laser irradiation near threshold. The threshold for an individual particle at a fixed position towards the irradiation beam is very steep [Fig. 5(b)]. Fitting a probit distribution, typical slopes of Hp=0.84/Hp=0.50<1.02 are obtained for irradiation of the same particle, but there is a larger variation in the threshold radiant exposure between different particles of the same type [Fig. 5(c)]. 3.2.Temperature-Dependent Thresholds for Bubble FormationThe nucleation temperature is obtained from temperature dependent threshold measurements for bubble formation. If the temperature of the particle suspension T amb is increased, less absorbed radiant exposure H thr is needed to reach the nucleation temperature T nuc on the particle surface and cause bubble formation. Due to the linearity of the heat equation in H, the temperature increase on the particle surface per radiant exposure ΔT surf (H,Q abs )/H is a constant, which depends for a given pulse shape and particle radius only on the absorption efficiency of the particle Q abs . The linear relationship can be extrapolated to the nucleation temperature, where no additional laser energy is needed to initiate vaporization.27 We use a heating interval of 20 to 60 °C for T amb in the experiments. Particle motion due to thermal fluctuations prevent threshold measurements at higher temperatures. Figure 6 shows exemplary measurements of the nucleation temperature for individual particles. No hysteresis of the thresholds was observed during heating and cooling cycles. The mean nucleation temperatures are obtained from nucleation temperature measurements of n=10 individual particles for each particle type and each pulse duration (see Table 2). No reliable determination of the nucleation temperature was possible for the gold beads due to strong thermal motion of the particles at elevated temperatures and a small temperature dependence of the thresholds for bubble formation.Table 2
The nucleation temperature depends strongly on the particle species. The heating rates, which can be approximated by (T nuc −T amb )/τ laser , are in the range of 108 K/s for microsecond pulses and 1010 K/s for nanosecond pulses. No significant dependency of the nucleation temperature from the pulse duration is found. Therefore, the mean nucleation temperature T mean is averaged from the nucleation temperatures of the 12-ns and the 1.8-μs irradiation. Modeling the melanosome as a sphere with a radius of R=0.5 μm, the efficiency factor of absorption Q abs can be determined from the slope of the temperature dependent bubble thresholds using the heat equation [Eq. (6)] (see Table 3). Table 3
3.3.Thresholds for Bubble Formation versus Pulse DurationThe mean threshold radiant exposures and corresponding standard deviations of n>50 different melanosomes and gold beads (Chemicell MG-2/1) are determined for each pulse duration (Fig. 7). Thresholds for the SiMAG-EP beads (magnetite/silica) are not determined due to a major variation in absorption (Sec. 2.3). Irradiation of the gold beads and the melanosomes with increasing pulse duration results in an increasing threshold radiant exposure for bubble formation due to heat conduction in the surrounding water during the laser pulse. We use the thermal model from Sec. 2.3 to fit the experimental thresholds for bubble formation. In the model, the temperature on the particle surface is calculated as a function of the applied radiant exposure. The thermal model accounts for the laser pulse shape at each pulse duration. The model contains the absorption efficiency Q abs of the particles and the nucleation temperature T nuc as free parameters. For the gold beads, the absorption efficiency can be calculated according to Mie theory. Therefore, the nucleation temperature is the sole remaining fitting variable. For the melanosomes, the nucleation temperature has been determined experimentally (Sec. 3.2), which enables the calculation of the absorption efficiency. The nucleation temperature for the gold beads, which is obtained from the experimental bubble thresholds by a least-squares fit using the well-known optical and thermal properties of gold (Fig. 7, Table 1), is T nuc =327±112 °C (for a 95% confidence interval). Using a nucleation temperature of T nuc =150 °C on the melanosome’s surface, which is determined in Sec. 3.2, and a radius of R=0.5 μm, the absorption efficiency of a melanosome Q abs is calculated by least square fitting. A value of Q abs =0.56±0.14 (for a 95% confidence interval) is obtained. 3.4.Onset of Bubble FormationThe temporal onset of bubble formation is extracted from the transmitted intensity of the probe laser providing a temporal resolution of about 10 ns. This technique is sufficient to determine the point of bubble nucleation during the 240-ns and the 1.8-μs irradiation relative to the laser pulse. A few (n>2) individual particles were irradiated with increasing radiant exposure for each pulse duration. Since particles of the same type differ in their threshold for bubble formation [Fig. 5(c)], the applied radiant exposure for each particle is normalized to its bubble threshold irradiance H thr (Fig. 8). The temporal resolution obtained by the probe laser technique is not sufficient to resolve the bubble incipience relative to the laser pulse during the 12-ns irradiation. Therefore, we use fast flash images providing a temporal resolution of about 3 ns, which is given by the duration of the illumination pulse (Fig. 9). Due to larger heating rates, increasing radiant exposure leads to an earlier bubble onset at a constant nucleation temperature. The excess radiant exposure (expressed as a ratio H/H thr ) required to nucleate a bubble at a time t can be modeled due to the linearity of the heat equation in H by where ΔT surf (t,H thr ) is the surface temperature increase at time t due to laser irradiation at threshold radiant exposure H thr . These calculated curves, which are depicted in Figs. 8 and 9, are in reasonable agreement with the experimental bubble incipience, which supports the thermal model. Particularly, the experimental bubble onset at threshold radiant exposure coincides in most cases with the point of time, where the maximum relative surface temperature is calculated. Small discrepancies between the experimental and the calculated bubble onset of the melanosomes might be explained by their ellipsoidal shape, which is unaccounted for in the thermal model.4.Discussion4.1.Bubble ThresholdIdentical thresholds for bubble formation are obtained by probe laser transmission and photography (Fig. 2). The bubbles, which are observed at threshold have a diameter of about 0.7 μm, which is near the resolving power of our microscope (Fig. 2). These bubbles observed at threshold are significantly smaller than in previously reported threshold measurements of isolated porcine RPE melanosomes, where a minimal bubble size of 2 μm has been resolved.19 The enhanced spatial resolution reveals, that in most cases microbubbles form at threshold only on parts of the particle surface rather than around the whole particle like previously hypothesized.19 Despite the enhanced resolving power of our experiments, our measured threshold radiant exposure values for porcine melanosomes are 40% higher than previously reported.19 Although the threshold for a single particle is very steep [Fig. 5(b)], the deviation of bubble thresholds between individual particles of the same type [Fig. 5(c)] might be caused by differences in particle size (Sec. 4.5), surface topology (Sec. 4.2), and speckles during irradiation (Sec. 2.2). The melanosomes may also have a different melanin content depending on the stage of its maturation.28 4.2.Nucleation TemperaturesSince the applied pulses are much longer than the acoustic transit time of the particles 2R/cs (cs is velocity of sound, see Table 1), only small thermoelastic pressure amplitudes are expected at threshold radiant exposure. Thus, the influence of tensile stress on the bubble nucleation is neglected in the further discussion. In most experiments on rapid transient heating, the nucleation temperature depends only weakly on the heating rate.29 30 Such a weak dependence could not be observed in our measurements. Therefore, no significant difference has been found between the 12-ns and the 1.8-μs pulses in Sec. 3.2. The temperature range, in which nucleation of water on heated surfaces can occur at atmospheric pressure, is restricted by the boiling point in thermodynamic equilibrium (100 °C) and the spinodal (∼315 °C), where water becomes mechanically unstable.31 Reviewing classical nucleation theory, there are different scenarios that lead to bubble nucleation on heated surfaces.31 4.2.1.Heterogeneous nucleationIf no bubble nuclei preexist on the heated surface, bubbles can form only due to random statistical energy fluctuations on a molecular scale. If a bubble nucleus, which is produced by these thermodynamic fluctuations, grows beyond its critical radius, where the vapor pressure overcomes the ambient pressure and the surface tension, an observable bubble nucleates. For reasonable contact angles Θ between water and the solid (Fig. 10), the rate for heterogeneous nucleation on the particle surface is at the same temperature some orders of magnitude larger than the homogeneous nucleation rate in the surrounding fluid, which usually results in nucleation on the particle surface rather than in the surrounding fluid.31 Since the fluid is heated by the particle, the highest fluid temperatures are obtained at the particle surface, which supports surface nucleation. Thus, we neglect the homogeneous nucleation rate in the following calculations. An approximation of the nucleation rate J het per surface area for heterogeneous nucleation is given in Ref. 31 by where Tl is the liquid temperature, ρ(T) is the liquid density,22 pl=1013 hPa is the liquid pressure, p sat (T) is the saturated vapor pressure,22 and σ(T) is the surface tension.32 Avogadro constant N A =6×1026/kmol, molecular mass M H 2 O =18 kg/kmol, specific gas constant R H 2 O =462 J/kg K, and Boltzmann’s constant kB=1.38×10−23 J/K. Assuming a temperature course that follows the laser pulse, one obtains for the bubble threshold, which is defined by a probability of 50% for bubble formation The contact angle Θ as a function of the nucleation temperature T nuc according to Eq. (9) is plotted in Fig. 10. Pulse shape and duration as well as the different surface area of the particle types are negligible for this calculation. The nucleation temperature decreases with increasing contact angle. Nucleation temperatures near the spinodal, which are consistent with the nucleation temperature determined for gold beads, are obtained31 33 for typical contact angles Θ in the range of 0 deg<Θ<Θ teflon =108 deg. The contact angles, which correspond to the nucleation temperatures determined in Sec. 3.2 (Fig. 10: Θ mel (147 °C)=175 deg, Θ SiMAG-EP (244 °C)=153 deg) suggest a quite nonwettable particle surface, which is extremely unlikely, for the surface of the magnetic silica particles (SiMAG-EP) contains hydrophilic silanol groups. Moreover, the photographs of transient bubbles on the particle surface (e.g., Fig. 4) are not consistent with contact angles Θ≫100 deg. It is more likely that other mechanisms lead to bubble nucleation on SiMAG-EP beads and melanosomes.4.2.2.Inhomogeneous nucleationGas or steam can be entrapped in crevices on the particle surface.34 Due to heating, the vapor pressure in the crevice increases and can overcome surface tension, which leads to bubble growth. The critical radius of a steam bubble is given by31 Bubbles smaller than the r crit will collapse due to insufficient vapor pressure, while bubbles larger than r crit will grow spontaneously. Thus, the nucleation temperature depends mainly on the size of the preexisting nucleus. Assuming pure steam in the crevice, the size of active nucleation cavities can be estimated by Hsu’s criterion for nucleate boiling on heated surfaces in nonuniform temperature fields.35 According to Hsu, inhomogeneous nucleation takes place if a preexisting steam bubble radius with the radius r becomes critical within the thermal boundary layer of the heated surface. Thus, for the bubble the condition must be satisfied to overcome ambient pressure and surface tension. T crit (r) is the temperature of a bubble, which is critical at a radius of r, [inverse of Eq. (10); for T2(r,t), see Eq. (2)]. The plots of the activation curve and the temperature in the surrounding fluid show, that nucleation on SiMAG-EP particles by this mechanism is very likely (Fig. 11). Nucleation by steam or gas cavities often corresponds to reproducible nucleation sites, which were observed during repeated irradiation above threshold for bubble formation (Fig. 4). Also the nucleation on gold particles is consistent with Hsu’s criterion. [The nucleation temperature is often estimated by Eq. (10) using the particle radius as critical radius r crit . This implies, that the whole particle acts as nucleation center, which is not supported by Figs. 2, 3, and 4. The assumption, that the whole particle is the nucleation center, is equivalent to an unrealistic contact angle of 180 deg and gives, therefore, only the lower limit of the nucleation temperature.]A comparison to nucleation experiments on film heaters,30 which are used in bubble jet technology, or thin wires,29 show that an electrical heating duration in the microsecond range results on metallic surfaces in nucleation temperatures in the range of 270 to 300 °C. On silica surfaces a nucleation temperature of 250 °C was found during pulsed laser heating.36 These temperatures are similar to the nucleation temperatures we have measured for gold and magnetic silica beads. The nucleation temperature of melanosomes is significantly lower and cannot solely be explained by steam crevices, because the thermal boundary layer is too small to activate preexisting bubble nuclei on the surface (Fig. 11). In contrast to the chemically inert materials of the inorganic microbeads, a chemical reaction of melanin, which produces gas, or vaporization of hydratation water in the melanosome, cannot be excluded by the experiments and might support bubble formation at lower temperatures. Glickman et al.37 found morphological damage on the surface of bovine RPE melanosomes, which induces oxidative stress, by scanning electron microscopy (SEM) after multiple irradiation by a 10-ns laser pulse at 214 mJ/cm2 (y=λ532 nm). Glickman et al. hypothesized that vaporization of the melanosome’s hydratation water, which is released during continuous heating in a temperature range of 100 to 130 °C (Ref. 38), causes this kind of damage. Furthermore, Piattelli and Nicolaus39 have observed thermal decarboxylation of melanin, which results in a release of CO 2 during heating of melanin to 140 to 150 °C. On the other hand, Hayes and Wolbarsht have found no morphologic changes of dog RPE melanosomes by SEM after heating them in standard atmosphere to 350 °C. Since we do not observe any threshold hysteresis or residual bubbles after transient bubble formation [Fig. 5(a)], only a small amount of noncondensable gas can be produced within the melanosome, whereas irradiation a few times above threshold leads to stable bubbles of noncondensable gas, which is usually explained by thermal disintegration of the melanosome.40 4.3.Onset of Bubble FormationA comparison of the experimental and the calculated bubble onset shows a good agreement, which supports our thermal model of the heated particles (Figs. 8 and 9). The bubble nucleates during the laser pulse even for nanosecond pulses (Fig. 9). In some models on bubble formation around laser heated melanosomes, nucleation takes place after the laser pulse.18 As no evidence for a delayed phase transition was found, our experiments do not support this hypothesis. 4.4.Absorption Coefficient of RPE MelanosomesThe mean absorption efficiency Q abs obtained by temperature dependent (Q abs,T nuc ) and by pulse duration dependent (Q abs,τ) bubble thresholds differ. In principle, both techniques should lead to the same absorption efficiency values. A reason for the differences between Q abs,T nuc and Q abs,τ was not found. However, the error bars obtained for both absorption efficiencies overlap. To compare the measured absorption efficiency of RPE melanosomes to absorption coefficients reported by other authors, we neglect the effects of Mie scattering and calculate the internal absorption coefficient of a RPE melanosome according to Gerstman et al.16 by geometric optics and Beer’s law. Using the first-order series approximation in μR, which corresponds to optical thin particles (homogeneous deposition of the absorbed energy) one obtains The resulting homogeneous absorption coefficients are listed in Table 4. These calculated absorption coefficients are in reasonable agreement with absorption coefficients of RPE melanosomes, which have been found in previous studies (see Table 5).Table 4
Table 54.5.Calculated Threshold Dependence on the Particle SizeUsing the experimental homogeneous absorption coefficient of μ=0.7 μm−1 for melanosomes, the surface temperature as a function of the melanosome size is calculated and compared to the calculated surface temperature of gold beads (Fig. 12). The absorption of the melanosomes is approximated by Eq. (12), whereas the absorption of gold particles is determined according to Mie theory. In Fig. 12, we choose radiant exposure values for the calculations, which lead for melanosomes at R=0.5 μm to the nucleation temperature of T nuc =150 °C and for gold beads at R=0.25 μm to T nuc =300 °C, respectively. Since melanosomes are optically thin, the energy is absorbed homogeneously within the volume resulting in a temperature increase, which does not depend on the particle size (12-ns laser pulse duration). But if the particle size becomes smaller, the condition of thermal confinement τ laser ≪R2/(4κ H 2 O ) (see Table 1) is not fulfilled for long pulses (240 ns and 1.8 μs) and appreciable heat diffusion during the pulse leads to less effective melanosome heating. Growing particle size of the gold beads, which are optically thick (Q abs ≈1, if the surface reflectivity is neglected) leads to a decreasing density of the absorbed energy which can be observed in Fig. 12 for the 12-ns pulse duration, because in the thermal confinement the temperature increase is proportional to the absorbed energy density at constant radiant exposure. Again, the particle heating becomes less effective for long pulses (240 ns and 1.8 μs) and small particle sizes due to appreciable heat conduction in the surrounding water during the laser pulse.5.ConclusionThe nucleation of bubbles on laser-heated microparticles was investigated. It was demonstrated by flash photography, that the whole particle does not act as the nucleation center, but the bubbles nucleate heterogeneously on the particle’s surface. The thresholds for bubble nucleation on melanosomes were determined for pulse durations in the nanosecond and microsecond range. A mean absorption efficiency of Q abs =0.46 was calculated by thermal modeling of bubble formation thresholds as a function of temperature and pulse duration. The nucleation temperature of bubbles on a melanosome, which is extrapolated from temperature dependent bubble thresholds, is T nuc =147±20 °C. For inorganic silica magnetite beads and gold beads, nucleation temperatures of T nuc =244±30 °C and T nuc =327±112 °C are found, which are significantly higher than for melanosomes. The bubble incipience relative to the irradiation pulse was studied and was found to be in good agreement with thermal modeling. It was shown that the bubbles can nucleate for all applied pulse durations, including for 12 ns pulses, during the pulse. AcknowledgmentsWe are grateful to Dr. Gereon Hu¨ttmann for making his self-written Mathematica library on heat diffusion and Mie scattering available to us. We also thank Chemicell GmbH, Berlin, Germany, for providing us with microbeads. One of the authors (J.N.) is supported by a fellowship of the FAZIT-Stiftung, Frankfurt am Main, Germany. REFERENCES
R. R. Anderson
and
J. A. Parrish
,
“Selective photothermolysis: precise microsurgery by selective absorption of pulsed radiation,”
Science , 220 524
–527
(1983). Google Scholar
B. S. Gerstman
,
“Nanosecond laser pulses: from retinal damage to refined surgical treatment of congenital nevi and melanoma,”
Proc. SPIE , 2975 180
–191
(1997). Google Scholar
J. Roider
,
R. Brinkmann
,
C. Wirbelauer
,
H. Laqua
, and
R. Birngruber
,
“Retinal sparing by selective retinal pigment epithelial photocoagulation,”
Arch. Ophthalmol. (Chicago) , 117 1028
–1034
(1999). Google Scholar
J. Roider
,
R. Brinkmann
,
C. Wirbelauer
,
H. Laqua
, and
R. Birngruber
,
“Subthreshold (retinal pigment epithelium) photocoagulation in macular diseases: a pilot study,”
Br. J. Ophthamol. , 84 40
–47
(2000). Google Scholar
J. Roider
,
F. Hillenkamp
,
T. Flotte
, and
R. Birngruber
,
“Microphotocoagulation: selective effects of repetitive short laser pulses,”
Proc. Natl. Acad. Sci. U.S.A. , 90 8643
–8647
(1993). Google Scholar
J. Roider
,
E. El Hifnawi
, and
R. Birngruber
,
“Bubble formation as primary interaction mechanism in retinal laser exposure with 200 ns laser pulses,”
Lasers Surg. Med. , 22 240
–248
(1998). Google Scholar
C. P. Lin
,
M. W. Kelly
,
S. A. B. Sibayan
,
M. A. Latina
, and
R. R. Anderson
,
“Selective cell killing by microparticle absorption of pulsed laser irradiation,”
IEEE J. Sel. Top. Quantum Electron. , 5 963
–968
(1999). Google Scholar
D. Leszczynski
,
C. M. Pitsillides
,
R. K. Pastila
,
R. R. Anderson
, and
C. P. Lin
,
“Laser-beam-triggered microcavitation: a novel method for selective cell destruction,”
Radiat. Res. , 156 399
–407
(2001). Google Scholar
C. M. Pitsillides
,
E. K. Joe
,
X. Wei
,
R. R. Anderson
, and
C. P. Lin
,
“Selective cell targeting with light-absorbing microparticles and nanoparticles,”
Biophys. J. , 84 4023
–4032
(2003). Google Scholar
V. P. Zharov
,
V. Galitovsky
, and
M. Viegas
,
“Photothermal detection of local thermal effects during selective nanophotothermolysis,”
Appl. Phys. Lett. , 83 4897
–4899
(2003). Google Scholar
A. A. Karabutov
,
E. V. Savateeva
, and
A. A. Oraevsky
,
“Optoacoustic supercontrast for early cancer detection,”
Proc. SPIE , 4256 179
–187
(2001). Google Scholar
T. E. McGrath
,
G. J. Diebold
,
D. M. Bartels
, and
R. A. Crowell
,
“Laser-initiated chemical reactions in carbon suspensions,”
J. Phys. Chem. A , 106 10072
–10078
(2002). Google Scholar
G. Hu¨ttmann
and
R. Birngruber
,
“On the possibility of high-precision photothermal microeffects and the measurement of fast thermal denaturation of proteins,”
IEEE J. Sel. Top. Quantum Electron. , 5 954
–962
(1999). Google Scholar
G. Hu¨ttmann
,
J. Serbin
,
B. Radt
,
B. I. Lange
, and
R. Birngruber
,
“Model system for investigating laser-induced subcellular microeffects,”
Proc. SPIE , 4257 398
–409
(2001). Google Scholar
G. Hu¨ttmann
,
B. Radt
,
J. Serbin
, and
R. Birngruber
,
“Inactivation of proteins by irradiation of gold nanoparticles with nano- and picosecond laser pulses,”
Proc. SPIE , 5142 88
–95
(2003). Google Scholar
B. S. Gerstman
,
C. R. Thompson
,
S. L. Jacques
, and
M. E. Rogers
,
“Laser induced bubble formation in the retina,”
Lasers Surg. Med. , 18 10
–21
(1996). Google Scholar
J. M. Sun
,
B. S. Gerstman
, and
B. Li
,
“Bubble dynamics and shock waves generated by laser absorption of a photoacoustic sphere,”
J. Appl. Phys. , 88 2352
–2362
(2000). Google Scholar
V. K. Pustovalov
,
“Thermal processes under the action of laser radiation pulse on absorbing granules in heterogeneous biotissues,”
Int. J. Heat Mass Transfer , 36 391
–399
(1993). Google Scholar
R. Brinkmann
,
G. Hu¨ttmann
,
J. Ro¨gener
,
J. Roider
,
R. Birngruber
, and
C. P. Lin
,
“Origin of retinal pigment epithelium cell damage by pulsed laser irradiance in the nanosecond to microsecond time regimen,”
Lasers Surg. Med. , 27 451
–464
(2000). Google Scholar
H. Goldenberg
and
C. J. Tranter
,
“Heat flow in an infinite medium heated by a sphere,”
Br. J. Appl. Phys. , 3 296
–298
(1952). Google Scholar
O. L. Anderson
,
D. G. Isaak
, and
S. Yamamoto
,
“Anharmonicity and the equation of state for gold,”
J. Appl. Phys. , 65 1534
–1543
(1989). Google Scholar
S. L. Jacques
and
D. J. McAuliffe
,
“The melanosome: threshold temperature for explosive vaporization and internal absorption coefficient during pulsed laser irradiation,”
Photochem. Photobiol. , 53 769
–775
(1991). Google Scholar
S. Glod
,
D. Poulikakos
,
Z. Zhao
, and
G. Yadigaroglu
,
“An investigation of microscale explosive vaporization of water on an ultrathin pt wire,”
Int. J. Heat Mass Transfer , 45 367
–379
(2002). Google Scholar
Y. Iida
,
K. Okuyama
, and
K. Sakurai
,
“Boiling nucleation on a very small film heater subjected to extremely rapid heating,”
Int. J. Heat Mass Transfer , 37 2771
–2780
(1994). Google Scholar
M. Holmberg
,
A. Ku¨hle
,
J. Garnaes
,
K. A. Morch
, and
A. Boisen
,
“Nanobubble trouble on gold surfaces,”
Langmuir , 19 10510
–10513
(2003). Google Scholar
Y. Y. Hsu
,
“On the size range of active nucleation cavities on a heating surface,”
J. Heat Transfer , 84 207
–216
(1962). Google Scholar
T. Brendel
and
R. Brinkmann
,
“Mid-IR laser induced superheating of water and its quantification by an optical temperature probe,”
Appl. Opt. , 43 1856
–1862
(2004). Google Scholar
R. D. Glickman
,
S. L. Jacques
,
J. A. Schwartz
,
T. Rodriguez
,
K.-W. Lam
, and
G. Buhr
,
“Photodisruption increases free radical reactivity of melanosomes isolated from retinal pigment epithelium,”
Proc. SPIE , 2681 460
–467
(1996). Google Scholar
P. Baraldi
,
R. Capelletti
,
P. R. Crippa
, and
N. Romeo
,
“Electrical characteristics and electret behavior of melanin,”
J. Electrochem. Soc. , 126 1207
–1212
(1979). Google Scholar
M. Piattelli
and
R. A. Nicolaus
,
“The structure of melanins and melanogenesis I,”
Tetrahedron , 15 66
–75
(1961). Google Scholar
J. Neumann
and
R. Brinkmann
,
“Microbubble dynamics around melanosomes irradiated with microsecond pulses,”
Proc. SPIE , 4617 180
–186
(2002). Google Scholar
R. D. Glickman
,
S. L. Jacques
,
R. T. Hall
, and
N. Kumar
,
“Revisiting the internal absorption coefficient of the retinal pigment epithelium melanosome,”
Proc. SPIE , 4257 134
–141
(2001). Google Scholar
M. A. Williams
,
L. H. Pinto
, and
J. Gherson
,
“The retinal pigment epithelium of wild type (C57BL/6J +/+) and pearl mutant (C57BL/6J pe/pe) mice,”
Invest. Ophthalmol. Visual Sci. , 26 657
–669
(1985). Google Scholar
M. Strauss
,
P. A. Amendt
,
R. A. London
,
D. J. Maitland
,
M. E. Glinsky
,
C. P. Lin
, and
M. W. Kelly
,
“Computational modeling of stress transients and bubble evolution in short pulse laser irradiated melanosome particles,”
Proc. SPIE , 2975 261
–270
(1997). Google Scholar
J. R. Hayes
and
M. L. Wolbarsht
,
“Thermal model for retinal damage induced by pulsed lasers,”
Aerosp. Med. , 39 474
–480
(1968). Google Scholar
I. A. Vitkin
,
J. Woolsley
,
B. C. Wilson
, and
R. R. Anderson
,
“Optical and thermal characterization of natural melanin,”
Photochem. Photobiol. , 59 455
–462
(1994). Google Scholar
|