Controlling grain boundary segregation to tune the conductivity of ceramic proton conductors

22 July 2024, Version 1

Abstract

Acceptor-doped barium zirconates are of major interest as proton-conducting ceramics for electrochemical applications at intermediate operating temperatures. The proton transport through polycrystalline microstructures of yttrium doped barium zirconates is hindered by the presence of a positive space charge potential at grain boundaries. During high temperature sintering, the positive charge acts as a driving force for acceptor dopant segregation to the grain boundary. Acceptor segregation to grain boundaries has been observed in sintered ceramics, but the fundamental relationship between the segregation kinetics and the protonic conductivity is poorly understood. Here, we present a comprehensive study of the influence of acceptor dopant segregation on the electrochemical properties of grain boundaries in barium zirconate based protonic ceramics. To facilitate this study, we designed an out-of-equilibrium model material that is not in a state of thermodynamic equilibrium and displays no detectable Y segregation at its grain boundaries. This model material served as a starting point to measure the kinetics of segregation and the induced changes in grain boundary conductivity upon varying thermal histories. Furthermore, we correlated the electrochemical results from impedance spectroscopy to atomic resolution transmission electron microscopy and atom probe tomography. We discovered that acceptor dopant segregation drastically increases the proton conductivity in both our model system and several other application-relevant compositions. In all cases, high-temperature thermal treatments were necessary to equilibrate the space charge zones, allowing the segregation of cationic point defects to grain boundaries, compensating the core charge and resulting in high performance protonic ceramics.

Keywords

Defect segregation
Grain boundaries
Protonic ceramics
Space charge

Supplementary materials

Title
Description
Actions
Title
Supplementary Information file
Description
Supplementary figures and data
Actions

Comments

Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting Policy [opens in a new tab] - please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here [opens in a new tab] .
This site is protected by reCAPTCHA and the Google Privacy Policy [opens in a new tab] and Terms of Service [opens in a new tab] apply.
Comment number 1, Dmitry Medvedev: Sep 04, 2024, 06:12

Thank you very much for sharing your research. I have several comments regarding it: 1. You stated that ‘while the bulk conductivity remains relatively constant’, but Fig. 1e shows that the bulk conductivity also varies, by ~5 – 7 times. Perhaps it would be better to give possible reasons for these variations. 2. The mean grain size as a function of sintering temperature or annealing time would be welcome. 3. “size after the highest thermal treatment (+1600 °C) is still at 320 ± 172nm” what is an accuracy of the deviation presented? 320± 170 nm can be a better choice. 4. Fig. 3 g is amazing 5. We have been dealing with Y-doped Ba(Zr,Ce)O3 for a long time (http://dx.doi.org/10.1016/j.jpowsour.2014.09.116, http://dx.doi.org/10.1016/j.jpowsour.2014.12.024, http://dx.doi.org/10.1134/S1063783415020250, https://doi.org/10.1016/j.scriptamat.2015.07.012, https://doi.org/10.1016/j.jpowsour.2017.09.021). Under inappropriate sintering regimes for these materials, Y2O3 as a micro-scale phase was very often formed in triple junctions. However, when we completely replaced Y3+ with Yb3+, no Yb2O3 was observed in any of the experiments. This might indicate that Y and Yb have different segregation energies. Could you please give more details on this aspect (segregation energy differences of Y and Yb)? 6. If Y2O3 was formed in some experiments, this phase can be easily detected via EBSD. Have you tried to find it?

Response,
Moritz Kindelmann :
Oct 17, 2024, 15:05

Deat Dmitry, thank you for your feedback! Saw it a little late because I did not regularly check the preprint page. Just some replies from my side to the comments you made: 1. Its true that the bulk conductivity deviates a bit, which is not discussed in detail. We think this is probably caused by the tricky fitting procedure, especially on samples that have a very high GB resistivitiy. This adds a lot of error to the fit of the bulk contribution, possibily leading to the observed variations. 2. This could be some interesting additional data. however for 5h dwell the grain size does not change between 1000-1600C. Perhaps longer dwells at very high temp. might lead to some grain growth but we did not check this yet. 3.Thats true ;) 4. Thanks a lot. APT was done by my colleague Ivan Povstugar, and he did an amazing job! 5. To be honest, we did not find any Y2O3 secondary phases while investigating the BZCY721 samples that have different thermal histories, neither with SEM, EBSD or STEM. We are aware that this has been observed in literature, but we could not observe any secondary phases here. There are several reasons why this might be the case: We work with only 10%Y on the B-site, being well below the solubility limit (I myself observed secondary phases in BZY20 and also BZCY20 samples). Additionally, we were careful to prevent Ba evaporation during high temperature annealing by placing the samples in a powderbed. This might prevent a shift to A-site deficiency, which can also induce some Y2O3 secondary phase precipitation. I hope this could answeresome of your questions. Thanks a lot for the open and public feedback! It is highly appreciated. Best, Moritz