Mid-infrared (MIR) spectroscopy is widely recognized as a powerful, non-destructive method for chemical analysis. However, its utility is constrained by a micrometre-scale spatial resolution imposed by the long-wavelength MIR diffraction limit. This limitation has been recently overcome by MIR photothermal imaging, which detects photothermal effects induced in the vicinity of MIR absorbers using a visible-light microscope. Despite its promise, the full potential of its spatial resolving power has not been realized. Here we present an optimal implementation of wide-field MIR photothermal imaging to achieve high spatial resolution. This was accomplished by employing single-objective synthetic-aperture quantitative phase imaging with synchronized subnanosecond MIR and visible light sources, effectively suppressing the resolution-degradation effect caused by photothermal heat diffusion. We demonstrated far-field MIR spectroscopic imaging with a spatial resolution limited by the visible diffraction, down to 120 or 175 nm in terms of the Nyquist–Shannon sampling theorem or full-width at half-maximum of the point spread function, respectively, in the MIR region of 3.12–3.85 μm (2,600–3,200 cm⁻¹). This technique—through the use of a shorter visible wavelength and/or a higher objective numerical aperture—holds the potential to achieve a spatial resolution of less than 100 nm, thus paving the way for MIR wide-field nanoscopy.

中红外 (MIR) 光谱被广泛认为是一种强大的、无损的化学分析方法。然而，其实用性受到长波长中红外衍射极限所施加的微米级空间分辨率的限制。最近，中红外光热成像克服了这一限制，它使用可见光显微镜检测中红外吸收体附近引起的光热效应。尽管有这样的前景，但其空间分辨能力的全部潜力尚未实现。在这里，我们提出了宽视场中红外光热成像的最佳实现，以实现高空间分辨率。这是通过采用同步亚纳秒中红外和可见光源的单物镜合成孔径定量相位成像来实现的，有效抑制了光热扩散引起的分辨率下降效应。我们展示了远场中红外光谱成像，其空间分辨率受可见衍射限制，根据奈奎斯特-香农采样定理或点扩散函数半峰全宽分别低至 120 或 175 nm，中红外区域为 3.12–3.85 μm (2,600–3,200 cm ^-1 )。该技术通过使用更短的可见光波长和/或更高的物镜数值孔径，有可能实现小于 100 nm 的空间分辨率，从而为 MIR 宽视场纳米显微镜铺平道路。