PSI - Issue 78

Virginio Quaglini et al. / Procedia Structural Integrity 78 (2026) 105–112

106

Barone et al. (2019), Çavdar et al. (2023), Constantinou et al. (2007), Furinghetti et al. (2019), Gandelli et al. (2019), Mokha et al.(1990, 1991, 1998), Pavese et al. (2019); Quaglini et al. (2012, 2014, 2022, 2023) are among the relevant references on this subject. However, the behavior of CSS at low temperature has not been adequately deepened yet. Recently Çavdar et al. (2024, 2025) investigated the effect of water and ice contamination on the coefficient of friction of sliding surfaces, an aspect not considered in prior literature. Dolce et al. (2005) and Quaglini et al. (2014) conducted at small scale showed that an increase in the friction coefficient at low temperature in comparison to regular temperature (20 °C), due to stiffening of the thermoplastic sliding polymer. In these studies, specimens were cooled down and tested under controlled conditions, usually within a thermal chamber. However, laboratory conditions are very different from the real environment where sliding isolators are supposed to operate in outdoor applications. Freezing winds strike the bridge above and below and on both sides, so it loses heat from every side. The sliding plates of the CSS are made of steel that is a good heat conductor and freezes shortly after temperature in the atmosphere hits the freezing point, resulting in the formation of an ice layer on the sliding surface. Such layer behaves as a lubricant, with a contrary effect on the friction coefficient than the one promoted by the stiffening of the slider material. Dry friction indeed describes the solid-solid sliding contact between two surfaces without any intermediate layer, and dry friction coefficient is typically high. However, as argued by Çavdar et al. (2024), in presence of ice dry friction is extremely rare in practice because a thin liquid-like lubricating layer with a thickness of a few molecules is always present on the ice surface even at very low temperatures. Therefore, friction coefficient on ice is typically lower than in dry condition, and this mechanism is well known and exploited in ice sports such as skying and skating (Rosenberg 2005; Petrenko and Whitworth 1999). The effect of ice on the friction coefficient of CSSs has been generally disregarded up to now, with few exceptions such as Alvarado and Ryan (2023), and current codes indeed recommend considering an increase in the friction coefficient as temperature decreases (CEN, 2009). The present paper reports the results of an experimental investigation conducted on a full-scale isolator to deepen this aspect, which can have important practical implications for applications in cold and very cold regions, and provide an insight for the friction coefficients of CSSs in icing conditions. 2. Methodology 2.1. Full -scale isolator The tested isolator is a Double Curved Surface Slider (DCSS) comprising a slider between two concave plates with a sliding pad at either interface (Fig. 1). The concave plates are 700 mm in diameter and the radii of curvature of both top and bottom concave surfaces are identical and equal to 1800 mm. The maximum displacement capacity of the DCSS is 300 mm. The main geometrical properties of the specimen are shown in Fig. 2.

(a)

(b)

Fig. 1. (a) cross-section of a DCSS; (b) motion of top and bottom plates.

2.2. Test set - up The experimental campaign was conducted at the ESQUAKE Seismic Isolator Test Laboratory of Eskişehir Technical University. The test set-up shown in Fig. 3 is capable of applying force and displacement controlled