shown in Figure 1 a, this process involves three simple steps: metal coating, diblock copolymer (BCP) lithography on the metal surface, and anisotropic ion beam milling (IBM) of the metal through the polymer etch mask. IBM has been previously used in patterning metal in conjunction with optical and electron-beam lithography. [ 18,19 ] Similarly, we combined IBM with the BCP lithography, since it is preferable, both economically and technically, to other lithographic techniques in patterning ultradense arrays on a macroscale area with large throughput. [ 20 ] Recently, BCP lithography has been nearly perfected with achieving a periodicity over a wafer scale, yet enabled by additional sparse patterning of lithographically-defi ned chemical [ 21,22 ] or physical templates. [23][24][25] BCPs can take different nanostructured morphologies of spherical, [ 20,26 ] cylindrical (columnar hexagonal) [ 20,22 ] or lamellar [ 25,27 ] blocks via microphase segregation. Compared with spherical ones, cylindrical BCPs have been more popular among researchers due to their high aspect ratio and negligible residual layers. The cylindrical BCPs has been applied on metal surfaces with an addition of an intermediate layer of random copolymer, [ 28 ] polydopamine [ 29 ] or a self-assembled monolayer of 3-(p-methoxyphenyl) propyltrichlorosilane [ 30 ] to match the surface energy. However, the use of such polymer intermediaries is not preferable as it abrogates the advantage of no residual layer. The challenge usually associated with large-scale formation of the self-assembled monolayer on metals also delists this method from the viable options for the macroscale BCP lithographic patterning.On the contrary, we show that it is possible to pattern spherical nanodomains of P(S-b-MMA) diblock copolymer directly on a metal thin fi lm at macroscales without use of any intermediary simply by lowering the roughness of the destination metal fi lm. Successive selective removal of the nanodomain (PMMA) yields a large-scale-patterned nanoporous matrix (PS) that we use as an etch mask to ion mill a nanoporous array (center-to-center distance: 50 nm) on the metal underlayer. In an attempt to assess the quality of our nanoporous metal fi lms, we defi ne a term, pore coverage, as a ratio between the areal number densities obtained experimentally and theoretically. This pore coverage measures 91.9% for our nanoporous metal fi lms over a 10-cm-diameter wafer with a great uniformity value of 99% (Figures S1 and S2). We characterize that the pore coverage becomes lower as the metal fi lm gets rougher, as empirically expected but yet to be clarifi ed quantitatively. In order to explain the relationship between pore coverage and roughness, we introduce the effect of rough localities of the metal surface on BCP self-assembly by discussing how they could impede the mobility of PS molecules thereby creating defect sites on Posing superior properties, nanoporous metals have been drawing attractions in a wide range of applications. For example, they possess a much ...