![Sample our Physical Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110819/TOC-Subject-Sample-2021-Physical-Sciences.jpg"})
Abstract
The structure of our material world is characterized by a large hierarchy of length scales that determines material properties and functions. Increasing spatial resolution in optical imaging and spectroscopy has been a long standing desire, to provide access, in particular, to mesoscopic phenomena associated with phase separation, order, and intrinsic and extrinsic structural inhomogeneities. A general concept for the combination of optical spectroscopy with scanning probe microscopy emerged recently, extending the spatial resolution of optical imaging far beyond the diffraction limit. The optical antenna properties of a scanning probe tip and the local near-field coupling between its apex and a sample provide few-nanometer optical spatial resolution. With imaging mechanisms largely independent of wavelength, this concept is compatible with essentially any form of optical spectroscopy, including nonlinear and ultrafast techniques, over a wide frequency range from the terahertz to the extreme ultraviolet. The past 10 years have seen a rapid development of this nano-optical imaging technique, known as tip-enhanced or scattering-scanning near-field optical microscopy (s-SNOM). Its applicability has been demonstrated for the nano-scale investigation of a wide range of materials including biomolecular, polymer, plasmonic, semiconductor, and dielectric systems.
We provide a general review of the development, fundamental imaging mechanisms, and different implementations of s-SNOM, and discuss its potential for providing nanoscale spectroscopic including femtosecond spatio-temporal information. We discuss possible near-field spectroscopic implementations, with contrast based on the metallic infrared Drude response, nano-scale impedance, infrared and Raman vibrational spectroscopy, phonon Raman nano-crystallography, and nonlinear optics to identify nanoscale phase separation (PS), strain, and ferroic order. With regard to applications, we focus on correlated and low-dimensional materials as examples that benefit, in particular, from the unique applicability of s-SNOM under variable and cryogenic temperatures, nearly arbitrary atmospheric conditions, controlled sample strain, and large electric and magnetic fields and currents. For example, in transition metal oxides, topological insulators, and graphene, unusual electronic, optical, magnetic, or mechanical properties emerge, such as colossal magneto-resistance (CMR), metal–insulator transitions (MITs), high-T C superconductivity, multiferroicity, and plasmon and phonon polaritons, with associated rich phase diagrams that are typically very sensitive to the above conditions. The interaction of charge, spin, orbital, and lattice degrees of freedom in correlated electron materials leads to frustration and degenerate ground states, with spatial PS over many orders of length scale. We discuss how the optical near-field response in s-SNOM allows for the systematic real space probing of multiple order parameters simultaneously under a wide range of internal and external stimuli (strain, magnetic field, photo-doping, etc.) by coupling directly to electronic, spin, phonon, optical, and polariton resonances in materials. In conclusion, we provide a perspective on the future extension of s-SNOM for multi-modal imaging with simultaneous nanometer spatial and femtosecond temporal resolution.
Acknowledgements
We would like to thank our collaborators on correlated matter, in particular, Manfred Fiebig, Stanislaus Wong, David Cobden, Weida Wu, Dan Dessau, Sang-Wook Cheong, and Junqiao Wu. Several present and former group members were instrumental in the success of many of the experiments, foremost Catalin Neacsu, Honghua Yang, Erik Josberger, Erik Hebestreit, Emily Chavez, Robert Olmon, Molly May, Ben Pollard, and Greg Andreev. We are indebted to many colleagues over the years, too numerous to list individually, but hopefully cited as completely as possible, who have helped us along the way with inspiring discussions and invaluable advice. This work was supported by funding from the Department of Energy Division of Materials Sciences and Engineering (Grant No. DE-SC0002197), and the National Science Foundation (NSF CAREER Grant CHE 0748226).
Notes
A plethora of acronyms developed in the early phases of tip-based scanning probe optical microscopies. Different adjectives were chosen in reference to what seemed the most relevant attribute of a certain application. “Tip-enhanced” (e.g, tip-enhanced Raman scattering (TERS)) has been used where the local field enhancement of the tip seemed most relevant. “Apertureless”, (e.g., apertureless (a-SNOM) or apertureless a-NSOM) were used in reference to the distinction from the conventional use of optical fiber tips with illumination or collection of light via the aperture at the fiber terminal. “Scattering” (e.g., in scattering (s-SNOM)), in reference to the function of the tip apex to scatter the evanescent tip-localized near-field into propagating and detectable far-field radiation, is also used interchangeably with “apertureless”. However, despite differences in terminology, the underlying physical mechanisms between all these techniques is very similar, if not identical. Most current descriptions of the imaging mechanism, in fact, overemphasize field-enhancement or near-field scattering, yet neglect the at least equally important function of the tip as an optical antenna defined by its capture cross section, antenna resonant properties, and its efficiency for the far- to near- to far-field transformation in and out of the localized sample excitation (parameters which are highly variable and as of yet poorly characterized or understood). We choose, unless otherwise specified, the term s-SNOM as a unified term.
Following initial demonstrations of NSOM, a debate ensued on the extent to which the observed optical contrast and resolution arose from a true optical near-field response, rather than topographic or related imaging artifacts. For further discussion see, for example, Ref. Citation210.
Much of the difficulty in estimating SERS enhancement factors arises from uncertainty in the volume of the “hot spot” and the number of molecules probed. Planar sample geometries and known surface coverages in TERS make the precise determination of the enhancement factor more readily possible.