Macroscopic indicators of fault diagnosis and ageing in electrochemical double layer capacitors

Abstract

The ageing behaviour of electric double layer (EDLC) supercapacitors during over-temperature and overvoltage stresses is described. Specific experimental tests were carried out in order to reproduce, under laboratory controlled conditions, the above-mentioned stresses. The obtained results are discussed and interpreted to define indicators that distinguish between over-temperature and over-voltage faults occurring in these devices. For each stress type post-mortem examinations of the EDLCs were performed to observe any specific physical damage. These examinations have been used to define the fundamental assumptions of the proposed fault indicators. Three fault indicators were identified and related to the evolution of the following SC parameters: (i) the equivalent series resistance, (ii) capacitance and (iii) decay voltage during the relaxation phase. Additionally, new accelerated ageing and recovering phases were detected during combined high-pulsed power cycling and life endurance stresses.

Modelling of current and temperature effects on supercapacitors ageing. Part II: State-of-Health assessment

“…The main difference between LE and PC stress is that the latter amplifies the degradation of the electrolyte and the production of impurities since it is associated to the current extraction (the so-called 2 If the boiling point of the electrolyte (78 C) and/or its decomposition voltage (3.5 V) are reached, other degradations, that are out of scope of the work here presented, take place and involve more important C SC fading [15]. 3 We remind that this fitting refers to a LE stress obtained at constant average T SC .…”

“…During a normal usage of a SC, in the transition from LE and PC, and vice versa, the phenomena described in [14,15] will take place. The effects of these phenomena are a temporary fast C SC fading during the transition PC !…”

“…Before illustrating the structure of the proposed SOH model, we summarize the main findings of [15] since we are interested to model the RP and AP.…”

“…The function DC SC ðtÞ, obtained from the results presented in [15], are described by Eqs. (4) DC SC ðtÞ ¼ 0:65…”

“…1) it requires a limited number of sensing devices since it just relies on the measurement of the cell voltage, current and temperature; 2) it relies on a set of measurements that can be off-line performed on a sample SC cell; 3) it takes into account the effect of stresses associated to highpulsed delivered current and, also, high temperatures; 4) it takes into account temporary performance changes due to the SC recovering phases (RPs) and accelerated ageing phases (APs) [14,15].…”