Understanding and controlling the interactions between lipid membranes and nanomaterials are important for drug delivery, toxicity studies, and sensing. In the literature, the perception is that cationic nanomaterials can damage lipid membranes, although some reports suggest the opposite. In this work, instead of using different materials for testing the effect of charge, we used the same material and adjusted the pH. A total of three types of liposomes including zwitterionic phosphocholine (PC) and negatively charged phosphoserine (PS) with saturated and unsaturated tails were tested with three types of metal oxide nanoparticles and two types of cationic polymers. A calcein leakage assay was used to probe membrane leakage. We found that cationic polymers had very little advantage for leaking PC liposomes. On the other hand, the PS liposomes were leaked by TiO2 nanoparticles regardless of their charge tuned by pH. ZnO with a high pKa value was studied in detail, and it only leaked the 1,2-dioleoyl-sn-glycero-3-phosphocholine liposomes at low pH when ZnO was positively charged, but leakage was inhibited by adding NaCl to weaken electrostatic attraction and by capping ZnO. In addition, dissolution of adsorbed ZnO also caused leakage, suggesting that adsorption and desorption induced reversible lipid phase transitions. Overall, the interaction strength was a key factor for leakage. Leakage does not necessarily mean membrane damage, and cryogenic transmission electron microscopy was used to study membrane integrity. Previously observed cationic polymer/nanoparticle-induced damages in supported membranes could be due to electrostatic attraction between the polymers and the underlying negatively charged supporting surface.