Voltage‐Controlled Nonstoichiometry in Oxide Thin Films: Pr_0.1Ce_0.9O_2−δ Case Study

Abstract

While the properties of functional oxide thin films often depend strongly on oxygen stoichiometry, there have been few means available for its control in a reliable and in situ fashion. This work describes the use of DC bias as a means of systematically controlling the stoichiometry of oxide thin films deposited onto yttria‐stabilized zirconia substrates. Impedance spectroscopy is performed on the electrochemical cell Pr0.1Ce0.9O2−δ (PCO)/YSZ/Ag for conditions: T = 550 to 700 °C, pO 2 = 10−4 to 1 atm, and ΔE = ‐100 to 100 mV. The DC bias ΔE is used to control the effective pO 2 or oxygen activity at the PCO/YSZ interface. The non‐stoichiometry (δ) of the PCO films is calculated from the measured chemical capacitance (Cchem ). These δ values, when plotted isothermally as a function of effective pO 2, established, either by the surrounding gas composition alone, or in combination with applied bias, agree well with each other and to predictions based on a previously determined defect model. These results confirm the suitability of using bias to precisely control δ of thin films in an in situ fashion and simultaneously monitor these changes by measurement of Cchem . Of further interest is the ability to reach effective pO 2s as high as 280 atm.

The Chemical Capacitance as a Fingerprint of Defect Chemistry in Mixed Conducting Oxides

“…According to Eqs. (23) and (24) the chemical capacitance can be determined from the fit parameter T W and the ionic and electronic resistances: (26) Assuming oxygen vacancies as minority charge carriers, we can use Eq. (26) and C chem = 4n v F 2 V/RT from Eq.…”

Dynamic Chemical Expansion of Thin Film Non-Stoichiometric Oxides at Extreme Temperatures

“…Figure 3c is derived according to the defect model available in Refs. 20,22 and the chemical expansion coefficient α c = 0.087 reported in Ref. 8 .…”

“…Since resistance to oxygen gas exchange at the PCO/Pt electrode is much higher than that of oxygen transport through the YSZ (>100 Ωcm 2 vs. 10 Ωcm 2 , respectively at 650 o C), as the PCO film adjusts its vacancy content to match the 'effective pO 2 ' caused by V WE , oxygen will be pumped primarily through the YSZ electrolyte. 19,20 The rate of adjusting oxygen content in the PCO film is then limited by diffusion through the YSZ, which will in turn determine the expansion rate. This interpretation is further validated by comparing the activation energy for τ/D 0 (1.05 ± 0.13 eV, see Figure 3a) with the activation energy for ionic conduction in YSZ (~1 eV).…”

“…Enforcing charge neutrality, mass, and site conservation for the composition of Pr 0.1 Ce 0.9 O 2-δ after Ref. 20 , (and using values of H r,Pr and k r 0 previously determined from the aforementioned literature), it is possible to determine vacancy concentration expected for this material under the range of temperatures and effective oxygen partial pressures pO 2,eff applied in this study.…”

“…The model film material of interest is Pr 0.1 Ce 0.9 O 2-δ (abbreviated as PCO, where δ 65 indicates non-stoichiometry), a functional oxide that exhibits large strain coupling to δ and has 66 been extensively characterized by a detailed defect model relating temperature and effective 67 pO 2 . 8,16,20 In PCO, chemical expansion arises largely from an increase in ionic radius of cations 68 during the following reduction reaction, written using Kröger-Vink notation: 16 respectively denote an oxide ion (neutral with respect to the lattice) and an oxide ion vacancy 72 (double positively charged with respect to the lattice) at an oxygen site. We have demonstrated 73 previously that in thin films supported on ionically conducting substrates, this equilibrium can be 74 controlled both by changing the surrounding atmosphere and by application of electrical bias.…”