This chapter introduces the thermal microscopy techniques that have been developed and used to probe local temperature in plasmonic structures under illumination. Reliably imaging a temperature distribution with a sub-micrometric resolution is not trivial, which certainly explains why the first temperature measurement in plasmonics dates only from 2009.

To date, ten families of techniques have been developed for this purpose. Interestingly, most of them rely on far-field optical microscopy techniques, which explains why part of the optics community is now tackling problems of thermodynamics. Half of the techniques are based on fluorescence measurements of molecular probes and are gathered in the first section. Then, the other techniques are assigned to specific sections, namely nanodiamond spectroscopy, wavefront sensing, Raman scattering spectroscopy, X-ray absorption spectroscopy, and scanning thermal microscopy. Each section of this chapter begins with a rather detailed explanation of the underlying physics in play. For this reason, the interest of this chapter is also to enter the physics of concepts such as fluorescence emission, surface-enhanced Raman scattering, X-ray absorption or nanodiamond's photophysics.

Introduction

Probing or mapping temperature on the submicrometric scale is not an easy task, even in the twenty-first century. Standard IR thermal radiation measurements, which are usually done on the macro scale, no longer apply at such small dimensions since the involved wavelength of the radiations (more than 10 μm) would lead to a very poor spatial resolution [39]. Moreover, most optical components are not transparent in this wavelength range (lenses, glass coverslips, etc.). Unlike light (which is endowed with a propagative nature), heat just diffuses. This makes any temperature distribution arising from a nanosource of heat confined at its vicinity, and not propagating to the far field. For these reasons, first attempts to probe a temperature field at small scales were based on the use of local probes consisting of a small composite tip acting as a nanoscale thermocouple or bolometer. This is the socalled SThM (scanning thermal microscopy) technique, invented in 1986 [80]. It enables a spatial resolution of around 50 nm. This technique would therefore have been ideal for temperature measurements in nanoplasmonics … if only it were not so invasive

Magnetic resonance imaging (MRI) was one of the first thermal imaging techniques used in thermoplasmonics. The groups of West and Halas used MRI measurement to retrieve the temperature in living animals in the context of photothermal therapies [33].

This chapter is intended to present all the applications based on the use of metal nanoparticles as nanosources of heat, namely protein denaturation, photothermal cancer therapy, drug and gene delivery, heat-assisted magnetic recording, photoacoustic imaging, plasmonic-induced nanochemistry, photothermal imaging, solar steam generation generation and single living cell experiments. For each application, particular attention will be paid to (i) the pioneering works and how the thematics were born, (ii) the subsequent pivotal works that introduced the variants and new concepts and (iii) the current state of the art and remaining challenges. For each application, I shall also adopt a chronological, story-like description, and occasionally propose various reading templates (chronological, experimental, theoretical, …), even if it yields redundancy.

Protein Denaturation: The Very First Application of Thermoplasmonics (1999)

Although one usually presents photothermal cancer therapy (2003) and photothermal imaging (2002) as the two very first applications of thermoplasmonics, there exists a pioneer article published in 1999 by Hüttmann and Birngruber [87]. This work benefited from the photothermal effects of gold nanoparticles and I believe this work can be considered as the very first article reporting on an application of thermoplasmonics. This is the reason why I dedicate a full section to this article, although protein denaturation cannot be considered as an important application of thermoplasmonics today. The last section of this chapter, dedicated to the application in biology, will be the occasion to recall this work.

The idea of this article was to study the thermal-induced denaturation of proteins using a pulsed laser to heat gold nanoparticles. At that time, the authors already understood that the temporal and spatial confinement achieved when heating nanoparticles with a subnanosecond laser could help achieve temperature as high as 470 K without boiling. The authors evidenced the denaturation of chymotrypsin proteins within 300 ps at temperatures below 380 K. This work was carried out in the context of photothermal treatment of vessels or pigmented cells.

At that time, I do not think the authors realized that most of the problems they raised in this article were about to be the subject of a large number of forthcoming articles. The authors also made correct predictions, for instance when saying: “It is expected that only solid-state absorbing particles (e.g., metal spheres, melanin, graphite, or iron oxide particles) can be used as such an energy acceptor for thermal microeffects.

In the spring of 1999 I was working at my desk at C. P. M. O. H. (Centre de physique moléculaire optique et hertzienne) in Bordeaux when I received an unexpected phone call from Claude Boccara from E. S. P. C. I. (École supérieure de physique et de chimie industrielles de la ville de Paris) in Paris. Claude had been particularly interested in photoacoustic and photothermal spectroscopies, as he had used them to optimize the mirrors of the VIRGO gravitational wave detector. He asked me about work by a Japanese colleague – it turned out to be Tsuguo Sawada from Tokyo University – who used a thermal-lens microscope to detect very weak concentrations of molecules, potentially even down to single molecules. I did not know the work, but as we were thinking of detecting single molecules at room temperature through their absorption instead of their fluorescence, I was immediately thrilled to learn about this new possibility. Some weeks later, I attended a talk by David A. Schultz from San Diego about applications of gold nanoparticles in bioimaging. Putting these two pieces of information together, I proposed that my young colleagues Philippe Tamarat and Abdelhamid Maali should start with the photothermal detection of gold nanoparticles instead of trying the absorption of single molecules at room temperature. Indeed, these gold nanoparticles neither bleach nor blink, and they exist in different sizes. This makes it much easier to optimize a technique on large nanoparticles before attempting to detect smaller ones. Contrary to the work of Tsuguo Sawada and Takehiko Kitamori, who investigated fluid suspensions in which diffusing molecules could enter and leave the detection volume during the detection period, we decided to work on immobilized gold nanoparticles, with which we had previous experience. A further advantage of photothermal detection was the high-frequency modulation of the heating beam, leading to very efficient rejection of background scattering by the sample and of low-frequency noise, both appreciable features in the complex and heterogeneous environment of biological cells. This project eventually led to our group's first work on photothermal detection, published in 2002. Later work by Brahim Lounis’ group6 and by others has established the thermal-lens microscope as a unique tool for the detection of single absorbing objects down to individual molecules.

This introductory chapter deals with basic, general and important notions in nanoplasmonics that will be useful before entering the field of thermoplasmonics. The aim is to provide the reader with simple ideas and mathematical expressions that can be used to explain and understand the plasmonic response of metal nanoparticles.

The first section introduces the physics of the localized plasmon resonance for a dipolar spherical metal nanoparticle, for which closed-form expressions of the optical response exist. The section also describes what happens when enlarging the size of the nanoparticle or breaking its spherical symmetry. The second section explains why gold nanoparticles have been preferred in plasmonics compared with nanoparticles made of other materials. This second section also takes the opportunity to discuss a very recent branch of nanoplasmonics aiming to try and find alternative plasmonic materials. Finally, a third section introduces the field of thermoplasmonics by answering common experimental questions.

Localized Plasmon Resonance

Definitions

A localized plasmon (LP) is a normal mode of collective oscillation of the free electrons contained in a metal nanoparticle. A LP resonance can be excited using light when the electric field of the incoming light oscillates at a frequency close to the plasmon eigen frequency [49].

What I call a localized plasmon (LP) in this book is often coined localized surface plasmon (LSP) in the literature. I prefer to remove the word “surface” for the following reason. Apart from LP, there exist bulk plasmons (BP) and surface plasmons (SP). With a bulk plasmon, the excitation occurs in a metal extending over the three dimensions of space (3D). With a SP, the electronic oscillation occurs at the interface between a metal and a dielectric extending over two dimensions of space (2D). With a LP, the oscillation occurs in a space that is confined in all dimensions of space (0D). While “bulk” means 3D, “surface” means 2D, I find it appropriate to use the adjective “localized” to signify 0D, and not “localized surface.”

In this book, I distinguish between the fields of plasmonics and nanoplasmonics. While plasmonics encompasses the physics of LPs and SPs, nanoplasmonics rather focuses on LPs; hence the title of this chapter.