Fig. 4. Facile clot detachment.

a Illustration of fibrin fibers formation from blood on a superhydrophobic CNF surface. Due to its non-wetting property and the presence of micro-air pockets (Cassie–Baxter state^28,49), the CNF surface would only partially attach onto the clot. b During clot maturation, the fibrin mesh would squeeze out the serum and shrink the clot into a smaller size;³⁹ contractile stress during clot contraction and solidification would remove CNFs connected with micro-fibrin fibers, causing the clot to free itself from the superhydrophobic CNF surface. c SEM image of the micro-fibrin fibers adhered on the CNFs-coated cotton fiber after clot shrinkage (Supplementary Fig. 12a, b). d SEM images show CNFs transferred onto clot after clot detachment, resulting in a smooth cotton fiber, and a hairy clot surface (Supplementary Fig. 12c). e Clot self-peel from a stiff CNF Ti mesh surface (Supplementary Fig. 11d–f). f Facile clot detachment; clot formed on the CNF gauze can be easily picked off by a foam tape and the corresponding clot-peeling tension is about 54 times smaller than that for peeling a clot from the normal gauze (n = 4, the setup in Supplementary Fig. 13); briefly, the gauze with a clot was mounted onto a force sensor, which was used to record the clot-peeling force for the calculation of peeling tension; data in f are shown as mean ± SD, the error bar represents SD, and individual data points are represented by black stars. Scale bars are 25 μm in c and 50 μm in d. Source data for f are provided as a Source Data file.