Broadband infrared spectroscopic near-field microscope

Achieving a truly broadband mid-infrared microscope that can resolve 10 nm detail will be an enormous advance for the analysis of nanosystems, from material science to biology. Chemical recognition is offered through the use of “fingerprint” molecular resonances that have already been shown to yield strong contrast, in the scattering-type near-field microscope (s-SNOM), from very small material volumes. Use of line-tunable gas lasers requires repeated imaging, plagued with drift, to approach spectroscopic imaging1). Use of broadband IR combs from DFG of Ti:S oscillator beams is now being explored, but faint signals and long integration times will lead to a proof-of-principle, not an operational instrument.

We will build a femtosecond laser emitting at a central wavelength close to 2 μm. The laser consist of a high-energy Yb-based oscillator and optical parametrical amplifier operating at the full repetition rate of the oscillator. The light source will emit 10–100-nJ few-cycle (25 fs) phase-stabilised pulses at a repetition rate of 3–10 MHz. This source will overcome the present S/N problem of infrared s-SNOM, because DFG performs much more efficiently since 2-photon processes are much less a limiting factor at this longer wavelength. We expect at least 1-mW average power in the mid-infrared (20% wide bands, tunable from about 6 to 18 μm).

A s-SNOM will be set up and tested at MPB. Its operation with the 2 μm laser-pumped broadband mid-infrared beam will be tested at LMU. The operational microscope will be used there with nanostructured samples for studies of semiconductor physics, such as oxidation of Si nanoparticles, protein adhesion and conformational interaction (secondary-structure changes), and polymer nanostructures, in various cooperations. Exposed biological samples from C.3.3 will be examined for radiation damage ex situ, possibly in N₂ atmosphere to avoid oxidation, at up to, in principle, spatial resolution of 10 nm.