Several types of physical forces can induce pore formation.  These include sonoporation and electroporation. Ultrasound can be used to deliver compounds into viable cells for potential targeted drug delivery and non-viral gene transfection, revealing advantageous possibilities. The delivery is facilitated through sonoporation, the formation of temporary pores in the cell membrane induced by ultrasound. Pan et al. 2004 reviewed sonoporation mechanisms that can be used to achieve optimal delivery outcomes such as high delivery efficiency and minimal cell death. Using voltage clamp techniques, they obtained real-time measurements of sonoporation of single Xenopus oocytes in the presence of Optison, an agent consisting of albumin-shelled C3F8 gas bubbles. Ultrasound increased the transmembrane current as a direct result of decreased membrane resistance due to pore formation. The ability to real time monitor sonoporation of cells provides a novel tool to study the dynamic sonoporation process and obtain optimal delivery parameters. They confirmed the delivery of compounds into cells by using markers such as plasmid GFP. Sonoporation mechanisms have been reviewed (Bouakaz et al. 2016; Dasgupta et al. 2016; Castle and Feinstein 2016). 

Electroporation, the transient increase in the permeability of cell membranes when exposed to a high electric field, is an established in vitro technique and is used to introduce DNA or other molecules into cells. When the trans-membrane voltage induced by an external electric field exceeds a certain threshold (normally 0.2-1 V), a rearrangement of the molecular structure of the membrane occurs, leading to pore formation in the membrane and a considerable increase in the cell membrane permeability to ions, molecules and even macromolecules. This phenomenon is, potentially, the basis for many in vivo applications such as electrochemotherapy and gene therapy (Chen et al. 2006). Electrochemotherapy (ECT) can be used for the treatment for metastatic nodules of solid tumors on the skin or subcutaneous tissue. ECT is a combination of a physical effect, cell membrane poration, and cytotoxic drug administration (Giardino et al. 2006). Cell permeabilization by time-varying magnetic/electric fields have been modeled based on transmembrane potential and mechanical stress in-vitro (Chiaramello et al. 2021). Stress values evaluated under conditions in which transembrane potential values were too low to cause membrane permeabilization were comparable to those known to influence the pore opening mechanisms.

Single-cell electroporation (SCEP) has emerged for single-cell studies. When a large enough electric field is applied to a single cell, transient nano-pores form in the cell membrane allowing molecules to be transported into and out of the cell (Wang et al. 2010). Unlike bulk electroporation, in which a homogenous electric field is applied to a suspension of cells, in SCEP, an electric field is created locally near a single cell. Pore formation has been discussed from theoretical and experimental approaches. Current SCEP techniques using microelectrodes, micropipettes, electrolyte-filled capillaries, and microfabricated devices are all thoroughly discussed for adherent and suspended cells. SCEP has been applied in in-vivo and in-vitro studies for delivery of cell-impermeant molecules such as drugs, DNA, and siRNA, and for morphological observations (Wang et al. 2010). Lipid pores can be induced by external fields, stress, and peptides (Kirsch and Böckmann 2016). The electric field strength is positively related to the transmembrane voltage (TMV) and pore density. A minimal electric field strength is necessary to induce a critical TMV for the formation of pores. Pulse width also had to be long enough to charge the cell membrane, compared with the normal membrane charging time constant of about 1 mus (Zhou et al. 2021).

To understand the origin of transmembrane potentials, formation of transient pores, and the movement of anions and cations across lipid membranes, Kandasamy and Larson 2006 have performed systematic atomistic molecular dynamics simulations of palmitoyl-oleoyl-phosphatidylcholine (POPC) lipids. A double bilayer setup was employed, and different transmembrane potentials were generated by varying the anion (Cl^-) and cation (Na⁺) concentrations in the two water compartments. A transmembrane potential of approximately 350 mV was thereby generated per bilayer for a unit charge imbalance. For transmembrane potential differences of up to approximately 1.4 V, the bilayers were stable over the time scale of the simulations (10-50 ns). At larger imposed potential differences, one of the two bilayers breaks down through formation of a water pore, leading to both anion and cation translocations through the pore. The anions typically have a short residence time inside the pore, while the cations show a wider range of residence times depending on whether they bind to a lipid molecule or not. Over the time scale of the simulations, Kandasamy and Larson 2006 did not observe the discharge of the entire potential difference, nor did they observe pore closure, although the size of the pore decreased as more ions translocated. They also observed a rare lipid flip-flop, in which a lipid molecules translocated from one bilayer leaflet to the opposite leaflet via the water pore.

Pore-formation depends on the membrane structure and composition.  Nicolson and Ferreira de Mattos 2023 have updated recent views on biological membranes. The Fluid-Mosaic Membrane (FMM) model was originally proposed as a general, nanometer-scale representation of cell membranes (Nicolson 2014). The FMM model was based on some general principles, such as thermodynamic considerations, intercalation of globular proteins into a lipid bilayer, independent protein and lipid dynamics, cooperativity and other characteristics. The FMM model was consistent with membrane asymmetry, cis- and trans-membrane linkages, and associations of membrane components into multi-molecular complexes and domains. It has remained useful for explaining the basic organizational principles and properties of various biological membranes. New information has been added, such as membrane-associated cytoskeletal assemblies, extracellular matrix interactions, transmembrane controls, specialized lipid-protein domains that differ in compositions, rotational and lateral mobilities, lifetimes, functions, and other characteristics. The presence of dense, structured membrane domains has reduced significantly the extent of fluid-lipid membrane areas, and the FMM model is now considered to be more mosaic and dense than the original proposal (Nicolson and Ferreira de Mattos 2023).