The thermoelectric module (TEM) is a solid-state device that converts thermal energy into electrical energy. A key feature of the TEM is that it delivers maximum power at only one specific operating point, defined by a certain localized voltage and current, referred to as the Maximum Power Point (MPP). Recently, much research has focused on improving the efficiency of power extraction from TEMs. Among the various Maximum Power Point Tracking (MPPT) methods proposed for thermoelectric systems, the perturb and observe (P&O) technique stands out for its low cost and simplicity of implementation. However, a limitation of the P&O method is that, under thermal equilibrium, the operating point oscillates around the MPP. Several enhancements have been suggested to reduce these oscillations. In this paper, a passivity-based P&O algorithm is introduced to address the shortcomings of the traditional P&O approach. To improve stability at the MPP, the Euler-Lagrange passivity technique is integrated with the standard P&O MPPT in a novel way. This combined approach, referred to as the P&O/EL-PBC algorithm, accelerates the tracking of maximum power and minimizes voltage fluctuations, showing excellent performance under varying load conditions.