skip to content
 

Fuel Cells


Background

There has been great interest in studying properties of ionic conductors so as to develop new materials for use in solid state devices such as solid oxide fuel cells (SOFCs), sensors and gas separation membranes. SOFCs have the potential to be a clean, low emission, quiet, reliable, fuel adaptable, transportable and highly efficient method for the generation of electricity. The basic principle of SOFC operation involves the transport of ions (oxide, O2- or proton, H+) through a solid electrolyte interposed between a cathode and anode, enabling the electrochemical reaction of oxygen with fuel to produce an electric current (Figure 1).

Figure 1: A general schematic of a solid oxide fuel cell (reproduced from here).

In general, electrolyte materials are required to have ionic conductivity and no electronic conductivity. Y-stabilized ZrO2 (YSZ) is one of the traditional oxygen-conducting solid electrolytes, with high ionic conductivity on the order of 10−3 Scm-1 with no electronic conductivity at 500°C. However, much higher ionic conductivity is observed in contemporary materials such as La0.8Sr0.2Ga0.8Mg0.2O3 (LSGM) with σi = 10−1.5 Scm-1 at 500°C.

Hydrated BaZrO3, on the other hand, is a traditional proton-conducting electrolyte with ionic conductivity of 10−1.5 Scm-1. Yet BaZrO3 requires acceptor doping (such as Y) to initiate O vacancy, and consequently H2O uptake, by the material (Blanc et al. [1,2], Middlemiss et al. [3], Buannic et al. [4]).

SOFC electrodes, in particular the cathode, are required to have mixed ionic and electronic conductivity (MIEC). The most common anode material for SOFCs is the Ni-YSZ cermet: nano-particle sized Ni metal supported by a high surface area YSZ-based ceramic. By contrast, the most common cathode material is the single-phase perovskite La1-xSrxMnO3 (LSM) which has a very high electronic as well as acceptable oxygen ion conductivity. Improvements in the O2- conductivity of MIEC cathodes are critically important in lowering the operational temperature of SOFCs.

We use multinuclear (17O, 31P, 1H, 2D; 89Y, 119Sn, 69/71Ga, 25Mg, 27Al) magic angle spinning NMR to study oxide and proton conduction as well as local ordering or "trapping" of species near dopants. First-principles DFT calculations in the GIPAW framework are also commonly used to assign experimental spectra to specific structural configurations. By identifying individual crystallographic or interstitial sites in often highly disordered materials, we can determine the local environments conducive to ionic conduction and obtain a much deeper understanding of how these materials function as "superionic" conductors.

Recent work

As mentioned, doped lanthanum gallate perovskites (LSGM, La1-xSrxGa1-yMgyO3) are second-generation SOFC electrolyte materials with improved oxide-ion conductivity at intermediate temperatures. 17O NMR spectra (Figure 2), as well as 17O 3Q-MAS experiments, reveal distinct resonances assigned by GIPAW NMR calculations to anions occupying equatorial and axial positions with respect to the GaV–VO axis. Similar studies using 25Mg and 69/71Ga MAS NMR confirm that all O vacancies preferentially locate in the first coordination sphere of Ga. [2,3] Variable temperature measurements also characterize the nature of the orthorhombic-to-rhombohedral phase transition in LSGM and its dependence on doping. [5]

Figure 2: 17O MAS NMR spectrum of 17O enriched La0.8Sr0.2Ga0.8Mg0.2O2.8 obtained at 19.6 T with assignments from GIPAW NMR calculations. [2]

Hydrated BaSn1–xYxO3–x/2 is a protonic conductor with unusually good conductivity at high dopant levels (0.10 ≤ x ≤ 0.50). As seen through 119Sn and 89Y NMR, increasing Y substitution on the Sn sublattice leads to strict dopant ordering at x = 0.50 (Figure 3). The avoidance of more basic Y–O–Y linkages as a consequence of this ordered configuration lowers proton trapping and directly enables the high protonic mobility seen experimentally. [6]

In contrast to the stannate, hydrated BaZr1–xYxO3–x/2 shows clear evidence for proton trapping, as revealed by two distinct resonances in its 1H spectra. Moreover, 1H to 89Y cross polarization (CP) experiments reveal that the "trapped" proton sites preferentially reside near yttrium atoms, in agreement with the lowest-energy configuration found by DFT. [7,8] Much of the experimental work on this system has incorporated sophisticated dynamic nuclear polarization (DNP) techniques for enhancing spectra where the NMR-active nuclei suffer from low receptivity (17O and 89Y). [8,9]

Figure 3: 119Sn NMR spectra of BaSn1–xYxO3–x/2 for 0.10 ≤ x ≤ 0.50 (left) show the transition to structural configurations favoring dopant ordering at x = 0.50 (middle) that contribute to high protonic conductivity in the hydrated state (right). [6]

The proton conductor CsH2PO4 exhibits a unique superprotonic phase transition at 230°C wherein the proton conductivity increases by up to four orders of magnitude (to 0.01 S cm-1). 1H and 17O NMR spectra at temperatures near the phase transition show distinctive lineshapes that depend sensitively on the dynamics of proton motion. [10] As seen in Figure 4, the experimental 17O spectra are consistent with an elegant model of proton mobility wherein the phosphate ion rapidly reorientates along two distinct axes, transferring protons from one tetrahedral unit to another. [11]

Figure 4: Variable-temperature 17O NMR spectra of CsH2PO4 below 230°C (left) can be fit to simulated spectra (middle) obtained from a model of two independent rotations of the phosphate anion that together facilitate proton motion (right). [11]

We recently re-investigated the defect perovskite Brownmillerite phase Ba2In2O5 and found the sharp resonance in the 17O NMR spectra at 220 ppm instead originates from surface-bound H2O, in agreement with the ease of hydration of this material. [12]

Another perovskite-derived structure, La2NiO4+δ, is a prospective mixed ionic (O2-) and electronic conductor for SOFC cathodes. The oxygen hyperstoichiometry (0 ≤ δ < 0.3) is incorporated as interstitial oxide anions. Again by way of 17O NMR measurements at operational SOFC temperatures, we observe two distinct mechanisms of oxide ion motion by vacancies and interstitials (Figure 5). Calculation and assignment of the extensively broadened paramagnetic 17O spectra remains a major challenge; our previous theoretical formalism for describing NMR shifts in paramagnetic materials shows some limited success here.

Figure 5: 17O MAS VT-NMR spectra of La2NiO4+δ at operational SOFC temperatures (600-800°C) provide evidence for two different mechanisms of oxygen exchange. [13]

References

  1. F. Blanc, D. S. Middlemiss, L. Buannic, J. L. Palumbo, I. Farnan and C. P. Grey, Solid State Nucl. Magn. Reson., 2012, 42, 87-97 (DOI: 10.1016/j.ssnmr.2012.01.003).
  2. F. Blanc, D. S. Middlemiss, Z. Gan and C. P. Grey, J. Am. Chem. Soc., 2011, 133, 17662-17672 (DOI: 10.1021/ja2053557).
  3. D. S. Middlemiss, F. Blanc, C. J. Pickard and C. P. Grey, J. Magn. Reson., 2010, 204, 1-10 (DOI: 10.1016/j.jmr.2010.01.004).
  4. L. Buannic, F. Blanc, I. Hung, Z. Gan and C. P. Grey, J. Mater. Chem., 2010, 20, 6322-6332 (DOI: 10.1039/c0jm00155d).
  5. Blanc, F., Middlemiss, D. S., Buannic, L., Palumbo, J. L., Farnan, I., Grey, C. P. Solid State Nucl. Magn. Reson., 2012, 42, 87–97 (DOI: 10.1016/j.ssnmr.2012.01.003).
  6. Buannic, L., Blanc, F., Middlemiss, D. S., Grey, C. P. J. Am. Chem. Soc. 2012, 134, 14483–14498 (DOI: 10.1021/ja304712v).
  7. Yamazaki, Y., Blanc, F., Okuyama, Y., Buannic, L., Lucio-Vega, J. C.; Grey, C. P.; Haile, S. M. Nat. Mater., 2013, 12, 647–651 (DOI: 10.1038/nmat3638).
  8. Blanc, F., Sperrin, L., Lee, D., Dervişoğlu, R., Yamazaki, Y., Haile, S. M.; De Paëpe, G.; Grey, C. P. J. Phys. Chem. Lett., 2014, 5, 2431–2436 (DOI: 10.1021/jz5007669).
  9. Blanc, F., Sperrin, L., Jefferson, D. A., Pawsey, S., Rosay, M., Grey, C. P. J. Am. Chem. Soc., 2013, 135, 2975–2978 (DOI: 10.1021/ja4004377).
  10. Kim, G., Blanc, F., Hu, Y.-Y., Grey, C. P. J. Phys. Chem. C, 2013, 117, 6504–6515 (DOI: 10.1021/jp312410t).
  11. Kim, G., Griffin, J. M., Blanc, F., Halle, S. M, Grey, C. P., J. Am. Chem. Soc., 2015, 137, 3867-3876 (DOI: 1010.1021/jacs.5b00280).
  12. Dervişoğlu, R., Middlemiss, D. S., Blanc, F., Holmes, L. A., Lee, Y.-L., Morgan, D., Grey, C. P. Phys. Chem. Chem. Phys., 2014, 16, 2597–2606 (DOI: 10.1039/c3cp53642d).
  13. Halat, D. M., Dervişoğlu, R., Kim, G., Dunstan, M. T., Blanc, F., Middlemiss, D. S., Grey, C. P. J. Am. Chem. Soc., 2016, 138, 11958 11969 (DOI: 10.1021/jacs.6b07348).