*2.4. Experimental Data Analysis*

The specific capacitance *cBLM* was determined for at least 25 planar lipid bilayers of each lipid composition and for each measurement method. Mean value and standard deviation were calculated for each experimental group. Differences in *cBLM* means between two measurement methods for each lipid composition were determined by t-test. For each lipid composition, at least three measurements of *Ubr* were made for each slope of linearly rising voltage and current. To compare *Ubr* measured for a given lipid composition at different slopes of the voltage signal, the one-way ANOVA test was used. When comparing

the mean breakdown voltages at the same slopes *k<sup>u</sup>* or *k<sup>i</sup>* between lipid bilayers of one component, the unpaired t-test was used. Since the variances of the mean breakdown voltages at slopes *k<sup>u</sup>* = 7.8 kV/s and *k<sup>u</sup>* = 11.5 kV/s were statistically different, the comparisons were made using Mann–Whitney's test. The difference was considered statistically significant at the values of *p* < 0.05.

Using nonlinear regression, a two-parameter strength–duration curve

$$
\mathcal{U} = \sqrt[4]{a + \frac{b}{t}} \tag{2}
$$

based on the viscoelastic model of Dimitrov [30] and proposed by Sabotin et al. [31] was fitted to experimentally obtained points (*tbr*, *Ubr*) measured at different slopes of the linearly rising voltage. The parameter √<sup>4</sup> *a* is an asymptote of the strength–duration curve corresponding to minimal breakdown voltage *Ubrmin* for planar lipid bilayers of a given lipid composition at fast transmembrane voltage build-up. The parameter *b* determines the inclination of the strength–duration curve. As in the case of strength–duration curve for excitable cell membranes of nerves and muscles, we can define for each lipid composition a kind of a chronaxie time constant *t<sup>c</sup>* = *<sup>b</sup>* 15·*a* as the minimum time required for a stimulus of twice *Ubrmin* to cause rupture of the planar lipid bilayer.
