Hyperspectral chemical imaging via sum-frequency generation microscopic holography

Project Summary: This proposal outlines a new investigator grant (NIBIB Trailblazer R21) for the PI to develop a new holographic platform that enables high-speed, noninvasive 3D chemical imaging with broad biomedical applications, including analysis of tumor microenvironment and real-time imaging of live cell cultures/animal

models/engineered tissues. While nonlinear optical microscopy is an attractive approach because it is label-free, nondestructive, and offers higher resolution and richer information than linear microscopy, it usually involves 3D scanning that limits the imaging refresh rate. To obviate this bottleneck, recent research has sought to combine

nonlinear optical microscopy with digital holography to vastly improve imaging speed. However, current implementations either lack chemical selectivity or require tuning the laser wavelength in order to measure vibrational spectra. To achieve a deeper understanding of biomedical processes at the molecular level, there is

clearly a significant need to further improve imaging speed and obtain rich spectral information. The long-term goal is to develop a new nonlinear digital microscopic holography approach capable of high-speed 3D imaging with spectroscopic vibrational contrast: i.e., 5D imaging in spatial, temporal, and spectral dimensions, in live cell

cultures and animal tissues. This transformational tool will enable discoveries of disease mechanisms and new treatment paradigms. This application’s objective is to demonstrate the feasibility of a new approach to achieving time-domain hyperspectral microscopic holography through vibrationally resonant (VR) sum-frequency

generation (SFG) and third-order sum-frequency generation (TSFG). Using mid-infrared photons to resonantly excite vibrational modes will enable chemical mapping of different functional groups in specimens, while the nonlinear processes offer submicron resolution. Combining SFG and TSFG in one instrument will enable

multimodal probing of non-centrosymmetric sample components as well as other components. Hyperspectral holography will enable 3D imaging and simultaneous recovery of the signal field’s amplitude, phase, and spectral

frequency in a single time scan. Three specific aims will be pursued: Aim 1 is to demonstrate hyperspectral VR-SFG microscopic holography and validate its performance. Based on comparison to the phase-sensitive multiplex VR-SFG microscopy previously demonstrated in the applicant’s hands, a single holographic

interferogram can be measured with a 4-μs exposure time and a signal-to-noise ratio of 10, with further improvement in signal-to-noise ratio expected. Aim 2 is to demonstrate multimodal VR- SFG/TSFG and to accelerate the acquisition rate by ~10x via compressive sensing. Aim 3 is to expand the spectral range to the

fingerprint region through building a new mid-infrared light source. The approach is innovative because it integrates concepts in a new unproven format for imaging with unprecedented speed and vibrational spectroscopic contrast. The proposed research is significant because it is expected to vertically advance

nonlinear digital microscopic holography for chemical mapping with submicron/subcellular resolutions.