nanoporous lithography methods, [18][19][20][21] etc. However, these techniques are still expensive and complicated for the fabrication of high quality SERS substrates over large areas, thus resulting in high prices for commercial SERS substrates. Furthermore, most commercial SERS substrates can only work for individual excitation wavelengths, i.e., one particular product works at one or two excitation wavelengths only. [22][23][24][25][26][27][28] When one wants to identify anonymous trace molecules or mixed samples, multiple excitation wavelengths will be required. [29][30][31] In this case, different substrates have to be used for different wavelength excitation, which consumes more biological/chemical materials, substrates, and measurement time. This is an obvious disadvantage for conventional SERS substrates. On the contrary, the SERS EF is proportional to the product of the fi eld intensity enhancements at both excitation and Raman scattering wavelengths. It was predicted that the maximum SERS enhancement can be achieved when localized surface plasmon resonance is located between the excitation and Raman scattering wavelengths. [ 32 ] To realize higher EF, double-resonance SERS substrates were proposed to realize strong enhancements for excitation and Raman scattered signals simultaneously using expensive e-beam lithography processes. [ 23 ] Due to the narrowband absorption spectra for both resonant bands, the enhanced SERS signal is still limited within narrow spectral regions. To address this problem, broadband resonant nanostructures are highly desired. For instance, a relatively broadband 1D metal-dielectric-metal metasurface (i.e., â70% optical absorption from 420 to 550 nm) was fabricated using e-beam lithography to realize uniform enhancement for SERS sensing. [ 24 ] However, the top-down lithography technique imposed a signifi cant fabrication cost barrier for large-scale practical applications. In addition, 1D grating structures are polarization dependent which can only work for given polarization states (usually transverse magnetic polarization). To overcome these limitations, here we report an ultrabroadband super absorbing metasurface substrate that can enhance the SERS signal for excitation wavelengths in a broad spectral region using lithography-free processes. [ 33 ] Most frequently used excitation wavelengths for SERS (e.g., from 450 to 1100 nm [23][24][25][26][27][28]34 ] are all covered due to the broadband light trapping and fi eld concentration within deep subwavelength Most reported surface-enhanced Raman spectroscopy (SERS) substrates can work for individual excitation wavelengths only. Therefore, different substrates have to be used for different excitation wavelengths, which consumes more biological/chemical materials, substrates, and measurement time. Here, an ultrabroadband super absorbing metasurface that can work as a universal substrate for low cost and high performance SERS sensing is reported. Due to broadband light trapping and localized fi eld enhancement, this structure can...