1.

Introduction

A growing interest has been reported within the last decade concerning a variety of applications of photoacoustic (PA) or laser-ultrasound techniques for nondestructive examination (NDE) of complex materials. Some well-developed techniques that have been exploited in the industrial sector are now being further expanded for application in medical diagnosis and visualization.^{1, 2, 3, 4, 5, 6} In photoacoustic imaging of biological tissue, short light pulses (usually of nanoseconds of duration) are partially absorbed by a sample, leading to a local, small temperature rise (typically millikelvin). Heating of the restricted tissue volume results in its transient thermoelastic expansion. The stress-induced local deformation generates ultrasound that propagates through the sample. The initial ultrasonic pressure is (for short laser pulses) linearly proportional to the deposited optical energy. Because the amplitude and profile of a photoacoustic waveform depend strongly on tissue optical absorption coefficient ${\mu }_a$ , which is related to physiological (or pathological) status of the tissue, high contrast differentiation between tissue parts is possible. Different constituents of the tissue (i.e., different muscle structures, bones, blood veins, tumour formations, etc.) may be targeted just by appropriate selection of the excitation wavelength. Photoacoustic imaging in biological tissue overcomes the low-contrast limitation of pure ultrasound imaging (low contrast in acoustic properties between different tissue sections). Simultaneously, it also helps to overcome unsatisfactory spatial resolution of all-optical techniques in those cases when optical scattering processes are dominant, such as biological tissue.²

We demonstrate for the first time the existence of an unexpected decay phenomenon of the amplitude of photoacoustic signals that we have observed during moderate laser pulse irradiation of in-vitro dental enamel samples, and in-vitro chicken or turkey breast samples.

2.

Materials and Methods

2.2.

Experimental Arrangement

A multimode, Q-switched Nd:YAG laser (Ultra CFR, Big Sky Laser Technologies, Incorporated) was used to irradiate samples. The laser provided pulses of wavelength $\lambda =1064\mathrm{nm}$ with a duration of $8\mathrm{ns}$ ( $1∕\mathrm{e}$ fall). The laser was operated at a $1\text{-}\mathrm{Hz}$ repetition rate, delivering pulses of approximately $11\text{-}\mathrm{mJ}$ energy. Actual pulse energy was measured with a laser energy monitor (Nova, Ophir Optronics, Limited, Israel). Laser pulses were delivered via a $600\text{-}\mu \mathrm{m}$ -diam (core) glass fiber (PCS). The laser beam profile and its divergence were examined in air with a charge-coupled device (CCD) camera (BeamPro 2320, Photon, Incorporated). This beam divergence was next recalculated to take into account that actual experiments were conducted in distilled water. During all experiments, the optical fiber tip was kept about $7\mathrm{mm}$ above a sample surface, which led to beam size at the sample of approximately $2\mathrm{mm}$ in diameter $(1∕{\mathrm{e}}^2)$ . The corresponding optical energy density was about $300\mathrm{mJ}∕{\mathrm{cm}}^2$ at a sample surface using laser pulse energies of $11\mathrm{mJ}$ .

We exploited two different irradiation-detection systems, as seen in Fig. 1 . In the first arrangement, Fig. 1a, a needle hydrophone detector was used to measure photoacoustic signals. We used a $1.47\text{-}\mathrm{mm}$ (outer diameter) needle hydrophone ( $\mathrm{HPM}1∕1$ , Precision Acoustics, Limited, United Kingdom) with a submersible preamplifier (HP1, Precision Acoustics) which was aligned along the optical fiber at a small angle of about $2\mathrm{deg}$ . In the second arrangement, Fig. 1b, the hydrophone and optical fiber were replaced by a submersible home-made compact photoacoustic head. This construction was designed by our group in 1987⁸ and has been successfully developed since that time.⁹ The actual probe receiver had a $6\text{-}\mathrm{mm}$ outside diameter and utilized a $28\text{-}\mu \mathrm{m}$ -thick polyvinylidene difluoride (PVDF) piezoelectric film, with vacuum metallized nickel-copper electrodes. Both photoacoustic receivers had an upper frequency response of approximately $15\mathrm{MHz}$ and gave similar experimental results.

Fig. 1

Experimental arrangements (out of scale): (a) needle hydrophone detection system and (b) scheme of a compact integrated photoacoustic head used as an alternative probe system.

Electrical signals coming from the ultrasonic receiver were captured by a $200\text{-}\mathrm{MHz}$ digital oscilloscope (TDS 640A, Tektronix, Incorporated) and transferred to a personal computer via a GPIB/IEEE488 interface. Due to the nature of our investigation, no signal averaging procedures could be applied. Transistor-transistor logic pulses from the laser’s Q-switch were utilized to trigger the scope, and special software, developed under a LabView® (National Instruments) platform, was used to acquire and monitor photoacoustic waveforms.

A sample under investigation was placed on a flipped Petri dish immersed in distilled water, in a 1 litre glass beaker. Small dental samples were clamped between two microscope slides to secure their steady-state position during examination.

3.

Results and Discussion

Typical photoacoustic waveforms delivered during the first few laser pulses by a human dental enamel section are presented in Fig. 2 . Similar (yet different in shape) bipolar transients were generated also by bovine enamel and chicken/turkey breast samples (rare or cooked). We focused on investigating the variation in peak-to-peak amplitude of these bipolar signals when consecutive laser pulses irradiated the sample.

Fig. 2

Typical bipolar photoacoustic waveforms delivered by a human dental enamel section sample under distilled water. Data were taken using the arrangement shown in Fig. 1a, with additional electronic amplification.

Figure 3 shows a graph of the peak amplitude of photoacoustic waveforms versus the consecutive laser pulse count when a sample was irradiated. The samples considered were human dental enamel, and rare and cooked chicken breast tissue. The figure shows a rapid decay of the photoacoustic signal amplitude to be clearly visible, confirming the phenomena shown in Fig. 2. An analogous decay phenomenon was also observed during experiments with bovine and turkey samples. Figure 3 reveals one more important detail: cooked chicken tissue produced initial photoacoustic signal amplitudes that were much higher than ones corresponding to an uncooked sample. This fact was consistent during all our tests, and was also confirmed in the case of turkey breast samples. Thus, the decay phenomenon itself could not be attributed to “laser cooking” or to thermal damage of the samples investigated.

Fig. 3

Decay of the photoacoustic signal amplitude during sample irradiation: optical energy density $300\mathrm{mJ}∕{\mathrm{cm}}^2$ . Data were taken using the arrangement shown in Fig. 1a, with additional electronic amplification.

These observations have been supported by theoretical calculations we performed of the temperature rise at a sample surface during laser irradiation. As a first approximation, we employed a simple 1-D model (not involving scattering) to estimate the surface temperature increase resulting from a single laser pulse^{1, 10}

The parameter values used for human dental enamel were: ${\mu }_a=1{\mathrm{cm}}^{-1}$ ,¹¹ $\rho =2.9\mathrm{g}∕{\mathrm{cm}}^3$ , and $C=0.75\mathrm{J}∕(\mathrm{g}\bullet \mathrm{K})$ .¹² Corresponding values assumed for rare chicken breast tissue were ${\mu }_a=2{\mathrm{cm}}^{-1}$ , $\rho =1.1\mathrm{g}∕{\mathrm{cm}}^3$ , and $C=3.5\mathrm{J}∕(\mathrm{g}\bullet \mathrm{K})$ , which are typical for muscle tissue.^{13, 14} In both cases, an average laser energy density $\Phi =300\mathrm{mJ}∕{\mathrm{cm}}^2$ was used in calculations. The optical absorption coefficient of water, ${\mu }_a$ , at $\lambda =1064\mathrm{nm}$ was taken to be $0.12{\mathrm{cm}}^{-1}$ .¹⁵ The resulting analytical predictions are presented in Fig. 4 , which shows that the expected surface temperature rise above ambient temperature for each laser pulse is predicted to be well below $0.25°\mathrm{C}$ . In later calculations, we have considered optical scattering processes in a 3-D model. Introduction of these processes do not lead to more than a doubling of predicted temperatures at the surface of the sample.

Fig. 4

Predicted temperature rise distribution within samples above ambient temperature due to a single laser pulse of $11\text{-}\mathrm{mJ}$ energy. The incident optical energy density was $300\mathrm{mJ}∕{\mathrm{cm}}^2$ .

All photoacoustic signal amplitudes varied on a shot-to-shot basis due to the fluctuations in laser pulse energy. As a consequence, an error of about $\pm 10\%$ is associated with the experimental data points shown in Fig. 3. Additionally, it is worth noting that similar decay of photoacoustic signal amplitudes was also observed during laser irradiation of dental enamel samples immersed in high performance liquid chromatography-grade methanol. Repeat measurements of signal amplitude also showed that the samples did not “recover” within several hours after laser irradiation.

Laser light can result in various processes such as tissue charging, coagulation, dehydration, ablation, or even trigger a synthesis of lethal complexes. Fried ¹⁶ for example, mentioned a decay phenomenon of photoacoustic signals delivered by $\mathrm{C}{\mathrm{O}}_2$ laser $(\lambda =9.6\mu \mathrm{m})$ irradiated dentin samples. These authors, however, used fluences exceeding $1\mathrm{J}∕{\mathrm{cm}}^2$ which, together with a very high optical absorption coefficient of dentin at $\lambda =9.6\mu \mathrm{m}$ , ${\mu }_a=6500{\mathrm{cm}}^{-1}$ ,¹⁶ led to sample surface temperature rises exceeding $1000°\mathrm{C}$ and, as a consequence, to ablation processes. Thus, it is difficult to say whether there is any direct correlation between their findings and ours, since under our experimental conditions there is no ablation. On the other hand, Sasaki ¹⁷ have provided evidence based on Fourier transform infrared spectroscopy (FTIR) that laser irradiation of dental samples may result in critical changes of their chemical structure. They investigated the influence of Er:YAG $(\lambda =2.94\mu \mathrm{m})$ and $\mathrm{C}{\mathrm{O}}_2$ $(\lambda =10.6\mu \mathrm{m})$ laser irradiation of extracted human teeth sections. Their data showed that laser radiation results in the formation of two additional bands in FTIR spectra of irradiated samples, at $2010{\mathrm{cm}}^{-1}$ $\left(4.98\mu \mathrm{m}\right)$ and also at $2200{\mathrm{cm}}^{-1}$ $\left(4.55\mu \mathrm{m}\right)$ . The first peak was recognized as a cyanamide $\left(\mathrm{N}\mathrm{C}{\mathrm{N}}^{2-}\right)$ band, while the second one was attributed to cyanate $\left(\mathrm{N}\mathrm{C}{\mathrm{O}}^{-}\right)$ ion formation. (Both compounds may be toxic.)

Neither of these interaction mechanisms at laser wavelengths extending into the infrared is likely in our present experiments. Our results at a near-infrared laser wavelength imply that the samples exhibit a decrease in the optical absorption coefficient from consecutive laser pulses. As a consequence, the samples undergo some optical transparency. Further studies are therefore required with infrared spectroscopy, and with detailed surface examinations of the samples using scanning electron microscopy. These studies may help to reveal the nature of the laser-matter interaction that results in a reduction of samples’ optical absorption coefficient, leading to a photoacoustic signal decay as shown in Figs. 2 and 3.

4.

Conclusions

We report on a new decay phenomenon in photoacoustic signal amplitudes from successive near-infrared laser pulse irradiation of biological tissue. The phenomenon is observed and investigated using moderate laser energy density $(300\mathrm{mJ}∕{\mathrm{cm}}^2)$ when irradiating biological materials at a laser wavelength of $1064\mathrm{nm}$ . Laser pulse durations were of the order of $8\mathrm{ns}$ . We recognize that this phenomenon arises from a direct consequence of changes to the sample’s optical properties when the sample is irradiated with such laser pulses. Future studies using analytical techniques will be required to gain further insight into these processes. A detailed understanding of the photoacoustic decay phenomena is necessary, and may be crucial for safe in-vivo biomedical applications of laser techniques for diagnostic and therapeutic purposes.