Characterization of electroactive polymer films and their application in practical devices (electronic, chemical, sensing, energy storage, optical, actuator) tends to focus on the transfer of charged species (electrons and counter ions), since these are the entities most directly associated with the response to the electrochemical control function. However, transport rates of ions (“dopant”, in the context of conducting polymer systems) are dictated by solvent content of the film, which governs the local viscosity of the medium through which mobile species move. Further, the presence of solvent (and other small molecules) creates free volume that facilitates the dynamics of polymer motion, macroscopically seen as a plasticising effect on film viscoelasticity.

In the light of the above, it is of value to determine the solvent content of electroactive polymer films; this is the focus of the present work. Experimentally, the challenge is that the solvent is (by choice) electrochemically “silent” and optically transparent, and its selective observation within a film exposed to a vast reservoir of bulk solution is not trivial. Nanogravimetric measurements made using the EQCM have provided innumerable insights into changes in film solvent population, e.g. in response to redox state changes, but the technique cannot provide absolute solvent population in the film. Spectroscopic techniques generally have difficulty in distinguishing solvent in a film from that in bulk solution, but reflectivity techniques do possess surface selectivity. Of these, ellipsometry has the advantage of rapid data acquisition, i.e. good time resolution, but modelling the data can present ambiguities.

Here we use neutron reflectivity (NR) which, through the penetrating nature of neutrons, allows access to "buried" interfaces. Traditionally, the negative consequence of this weak interaction with matter is long data acquisition times; addressing this is the central feature of this presentation. Since neutrons interact with the nuclei in the system (cf. photons interacting with the electrons in ellipsometry), NR has the huge advantage of isotopic selectivity. This is commonly referred to as “contrast variation” and has the critical characteristic of allowing one to vary the visibility of a selected chemical component without (to a first approximation) altering the (electro)chemistry of the system. Most frequently, this is exploited by deuteration, since the neutron scattering lengths of H and D are very different. Correlation of the h- and d-system NR data sets removes ambiguities in modelling the data.  

We have previously used NR with solvent contrast variation to determine solvent volume fraction depth profiles, Φ_S(z), in redox polymer films at equilibrium (fixed potential) and on long timescales during slow cycling. We now present data under dynamic conditions more relevant to fabrication or operation of a working device. This necessitates an order of magnitude improvement in time resolution. Instrumentally, the challenge can be met using more sophisticated detection systems, including event mode data capture, in which every incident/reflected neutron is detected individually. The dramatic increase in the size of data sets presents challenges in modelling the data, but the facility to (re-)select signal averaging time post experiment permits optimisation of competing aspirations of time resolution and signal-to-noise. This is key for single shot experiments, such as film deposition or evolution of film behaviour.

Here we present neutron reflectivity data for aniline-, pyrrole- and thiophene-based polymer films. For these systems, we have the opportunity to exploit h- and d- variants of both solvent and monomer. Four significant insights are revealed. First, the data reveal that film deposition protocol (i.e. potentiostatic, potentiodynamic or galvanostatic control function) influences film solvation state and spatial distribution; while anecdotally suspected, this provides unambiguous quantitative evidence. Second, there are substantial spatial variations in film solvent content; this is a feature commonly not accommodated in models of modified electrodes. Third, for some systems, film solvent population and spatial disribution are "frozen" into the system during fabrication, so deposition protocol determines film characteristics in the longer term. Fourth, in selected cases the structure evolves with electrochemical manipulation, such that differently prepared films subsequently exposed to the same environment and history evolve to a common solvation profile. The implications of these phenomena for film design and handling are discussed for particular types of application.