Experimental evaluation of temperature increase and associated detection sensitivity in shot noise-limited photothermal microscopy

Abstract

Abstract Photothermal microscopy is useful for visualizing non-fluorescent chromophores in a light-scattering medium. However, it remains a major challenge to enhance sensitivity while reducing thermal damage to samples, particularly for biological tissues. In this study, a highly sensitive laser scanning photothermal microscope was constructed by implementing a low-noise balanced photodetector into a spatially segmented balanced detection which was proposed and demonstrated to be useful for high sensitivity detection in our previous studies. We confirmed that the noise level was only 1.05-fold greater than the shot noise. Temperature increase and the detection sensitivity for this parameter were evaluated by observing the motion of individual 20-nm-sized gold nanoparticles embedded in agarose gel in response to incident light at varied intensities. We found that the nanoparticles remained trapped in the gel structure at a low pump power and it exhibits random motion as increase in the pump power because local temperature of agarose gel around the nanoparticle reached the melting point of 338 K. The peak SNR is 290 at about the melting point with an integration time of 10 μ s per pixel. On an assumption that the SNR is proportional to the temperature increase, the estimated limit of detection of a temperature increase at the nanoparticle surface, which is defined as the ratio of temperature increase to the SNR, ranges from 0.1 to 0.2 K. Additionally, findings from single carbon nanotube imaging suggest that a larger or longer light absorber would improve the trade-off relationship between the signal-to-noise ratio and temperature increase.