Photoacoustics

Photoacoustics is the generation of acoustic waves by the absorption of light. A short laser pulse, typically 5–10 ns long, is absorbed by a chromophore such as hemoglobin; the absorbed energy heats the region by roughly 10 mK for blood under typical imaging fluence, the region expands thermoelastically, and the resulting pressure transient propagates outward as broadband ultrasound in the 1–100 MHz range. An ultrasound detector records the wave, and its arrival time and amplitude give the position and strength of the absorber. Bell first reported the effect in 1880; it now underlies photoacoustic imaging and a family of trace-gas sensors.

Stress and thermal confinement

Efficient pressure generation requires that the energy be deposited before it can leave the heated region. Two conditions apply. Stress confinement requires the pulse to be shorter than the acoustic transit time across the region of size $d$:

$$\tau_s = \frac{d}{v_s}$$

For a 150 µm heated region in soft tissue, with a speed of sound $v_s$ of about 1500 m/s, $\tau_s = 100$ ns, so a 10 ns pulse is well inside the limit. Thermal confinement requires the pulse to be shorter than the heat-diffusion time, $\tau_{th} = d^2/\alpha_{th}$; with a tissue thermal diffusivity of about $1.4 \times 10^{-7}$ m²/s, the same region gives $\tau_{th} \approx 0.16$ s. Stress confinement is the stricter condition, which is why Q-switched nanosecond lasers are used.

Initial pressure

When both conditions hold, the local initial pressure is

$$p_0 = \Gamma\,\mu_a F$$

where $\mu_a$ is the absorption coefficient, $F$ is the local fluence, and $\Gamma = \beta v_s^2/C_p$ is the dimensionless Grüneisen parameter, with $\beta$ the volume thermal expansion coefficient and $C_p$ the specific heat. For water at 20 °C, $\Gamma \approx 0.11$; for soft tissue at body temperature it is often taken as roughly 0.2, though it varies with tissue type and temperature.

Worked example: whole blood near 800 nm has $\mu_a \approx 4$ cm⁻¹ (400 m⁻¹). With $F = 10$ mJ/cm² (100 J/m²) and $\Gamma = 0.2$,

$$p_0 = 0.2 \times 400 \times 100 = 8000\ \text{Pa}$$

or about 8 kPa. A background tissue with $\mu_a = 0.1$ cm⁻¹ under the same pulse gives 0.2 kPa, so a vessel stands out by the ratio of absorption coefficients.

Microscopy and tomography

Photoacoustic imaging comes in two broad forms. In optical-resolution photoacoustic microscopy, the laser is focused tightly and the lateral resolution is set by the optics, about $0.51\lambda/\text{NA}$: 2.7 µm at 532 nm with an NA of 0.1. This works only within roughly the first millimeter of tissue, where unscattered light still forms a focus. In acoustic-resolution microscopy and in photoacoustic tomography, the light is diffuse and the resolution comes from the ultrasound detection, which allows imaging at depths of a few centimeters.

The acoustic resolution scales with the ultrasound wavelength, $\lambda_a = v_s/f$. At 1500 m/s, 10 MHz corresponds to 150 µm and 50 MHz to 30 µm. Higher frequencies resolve finer detail but attenuate faster: at about 0.5 dB/(cm·MHz) in soft tissue, 1 cm of travel costs about 5 dB at 10 MHz and 25 dB at 50 MHz. Every photoacoustic system therefore trades depth against resolution through its transducer frequency.

Optical coherence tomography images scattering in the first one or two millimeters and multiphoton microscopy images fluorescence at cellular resolution; photoacoustics images absorption and reaches deeper at coarser resolution.

Lasers and contrast

The most common source is a flashlamp-pumped Nd:YAG laser, Q-switched to 5–10 ns pulses at 10–20 Hz, either frequency-doubled to 532 nm or used to pump an optical parametric oscillator tunable over roughly 680–950 nm.

The main endogenous absorbers are oxy- and deoxyhemoglobin, whose spectra cross at an isosbestic point near 800 nm, and melanin, whose absorption falls steadily with wavelength. Imaging at several wavelengths and unmixing the spectra yields maps of oxygen saturation and total hemoglobin. Dyes and nanoparticles add exogenous contrast.

Photoacoustic gas spectroscopy

In gas sensing, a modulated laser tuned to an absorption line heats a gas sample periodically, and a microphone in a resonant cell, or a quartz tuning fork that itself acts as the resonator, detects the resulting sound. The signal scales with absorbed power and needs no long optical path, unlike transmission methods such as TDLAS.

Pitfalls

Fluence at depth is rarely known accurately, so absolute $\mu_a$ values from images carry a large uncertainty unless fluence is modeled or calibrated. Limited detector views produce reconstruction artifacts and lose structures oriented along the detection direction.

Common questions

What is the difference between photoacoustic and ultrasound imaging?

Both detect ultrasound, but conventional ultrasound sends sound in and images acoustic reflections, while photoacoustic imaging sends light in and images optical absorption.

How deep can photoacoustic imaging see?

Optical-resolution microscopy is limited to about 1 mm. Tomography with near-infrared light, where tissue absorption is lowest, reaches a few centimeters, limited by the fluence that survives scattering.

Why use near-infrared wavelengths?

Hemoglobin and water absorb less between about 650 and 950 nm, so light penetrates farther, and the spectral differences between oxy- and deoxyhemoglobin in that window allow oxygenation to be measured.

References: L. V. Wang and H.-I. Wu, Biomedical Optics: Principles and Imaging (Wiley, 2007); L. V. Wang and S. Hu, "Photoacoustic tomography: in vivo imaging from organelles to organs," Science 335, 1458 (2012); A. G. Bell, "On the production and reproduction of sound by light," American Journal of Science 20, 305 (1880).