Ca²⁺ and Ga³⁺ doped LaMnO₃ perovskite as a highly efficient and stable catalyst for two-step thermochemical water splitting

Abstract

High performance and stable catalysts for two-step thermochemical water splitting are key to synthesising direct fuels in the form of H₂ or liquid hydrocarbon fuels by the Fischer–Tropsch process. Herein, we designed and synthesised LaMnO₃ perovskite structured oxides doped on both the A and B sites for two-step thermochemical water splitting. First, Ca²⁺, Sr²⁺ and Ba²⁺ divalent cations were successfully doped on the A site of LaMnO₃ and the thermochemical water splitting performances were analysed. After that, Al³⁺ and Ga³⁺ ions were doped on the B site of the perovskites produced in the first step. Through this strategy, a novel perovskite composition (La_0.6Ca_0.4Mn_0.8Ga_0.2O₃) was found with remarkable water splitting performance, producing 401 μmol g⁻¹ of H₂ at low thermochemical cycle temperatures between 1300 and 900 °C. The as-prepared perovskite exhibits twelve times higher H₂ production than the benchmark CeO₂ catalyst under the same experimental conditions. This novel perovskite is also capable of maintaining steady-state redox activity during the water splitting cycles.

---

The thermochemical water splitting process represents an efficient way to convert solar energy into chemically storable fuels in the form of H₂, which is an important precursor for synthesising liquid hydrocarbon-based fuels.^1–3 Since the direct thermolysis of water takes place at extremely high temperatures^4–6 (>3000 °C), efficient catalysts are required to lower this processing temperature for practical applications. Typically, the catalysed thermochemical process involves two steps: a higher temperature endothermic reduction step and a subsequent lower temperature exothermic gas splitting step.^7–9 The processing temperature and solar fuel production performance of the two-step thermochemical method greatly depend on the redox properties of catalysts, typically metal oxides. Over the past several decades, hundreds of metal oxides have been tested but only a few (e.g. Fe₃O₄,^10–12 ZnO,^13–15 and CeO₂ ^16–18) were reasonably active for thermochemical solar fuel production. Currently, non-stoichiometric CeO₂ is regarded as the benchmark catalyst due to its rapid reaction kinetics and high stability.¹⁹ However, the high operation temperature (≥1500 °C) and relatively small oxygen exchange capacity of CeO₂ limit its large-scale utilisation. Being a versatile material, perovskite type oxides, ABO₃, have received widespread attention, particularly as a promising alternative catalyst for thermochemical water splitting. In principle, the perovskite-structured ABO₃ can release its lattice oxygen through a high-temperature thermal reduction process resulting in the highly reduced ABO_3−δ (eqn (1)). Subsequently, the ABO_3−δ can effectively reduce introduced H₂O to molecular H₂ (eqn (2)), while recovering back to its original ABO₃ state.

ABO₃ + thermal energy → ABO_3−δ + δ/2O₂ (1)

ABO_3−δ + δH₂O → ABO₃ + δH₂ (2)

To date, a number of perovskite oxides have been studied for two-step thermochemical gas splitting.^20–27 These include reports demonstrating the effectiveness of LaMnO₃ based perovskite catalysts, e.g. La_xSr_1−xMnO₃,^28–30 La_xCa_1−xMnO₃,³¹ and La_xSr_1−xMn_yAl_1−yO₃,³² in thermochemical processes. Based on the available scientific evidence, the catalytic performance of perovskite oxides is closely related to their inherent catalytic properties and corresponding chemical compositions.^33,34 The thermochemical gas splitting behaviour of LaMnO₃ perovskite oxide can be purposely tuned by doping different metal cations onto its A and B sites. Very recently, Demont et al. reported improved thermochemical CO₂ splitting performances via the successful incorporation of different chemical elements into the LaMnO₃ perovskite oxide.³⁵ To date, however, no work has investigated the thermochemical H₂O splitting performance of LaMnO₃ based perovskites with different elemental doping, especially the main group metal ions on the A and B sites.

Herein, for the first time, we have investigated the two-step thermochemical H₂O splitting performances of LaMnO₃ based perovskites with Ca²⁺, Sr²⁺ and Ba²⁺ doped on the A site and Al³⁺ and Ga³⁺ doped on the B site. The doped LaMnO₃ exhibits significantly enhanced O₂ and H₂ production as well as re-oxidation yields compared to the parent LaMnO₃. More importantly, through this process, a novel La_0.6Ca_0.4Mn_0.8Ga_0.2O₃ perovskite oxide is developed which shows excellent H₂O splitting performance at a low thermal reduction temperature (1300 °C) with superb stability.

La_1−xA_xMn_1−yB_yO₃ (A = Ca²⁺, Sr²⁺, and Ba²⁺; x = 0 and 0.4; B = Al³⁺ and Ga³⁺; y = 0 and 0.2) perovskite oxides were prepared via a modified Pechini method³⁶ (refer to the ESI† for details). To facilitate effective testing and increase H₂O splitting activity, the freshly prepared perovskite powders (∼0.5 g) were transformed into a solid porous monolith structure by mixing with isopropanol (1.5 ml) and then heating at 1300 °C for 6 h in air.^29,37Fig. 1a shows the X-ray diffraction (XRD) pattern of La_0.6Ca_0.4Mn_0.8Ga_0.2O₃ (LCMGO) as a representative of all the samples and the main diffraction peaks are characteristic of the typical orthorhombic perovskite structure. However, when the Ga³⁺ doping content was further increased to 0.3, the impurity of Ga₂O₃ was detected in the XRD pattern (Fig. S1, ESI†). A similar observation was reported by Dey et al.³⁸ Compared to those of the parent structure of LaMnO₃ (LMO), the Bragg peaks of LMO perovskites doped on both the A and B sites were noticeably shifted due to the distinct ionic radii of the introduced metal ions (Fig. S2, ESI†). The partial substitution of La³⁺ (1.36 Å) with a smaller Ca²⁺ (1.34 Å) causes the decrease of lattice parameters, which is reflected by the shifted main Bragg peak (110) towards a higher diffraction angle. Similar upshift trends in the XRD results of LCMAO and LCMGO were also detected because of the introduced Al³⁺ (0.535 Å) and Ga³⁺ (0.62 Å) ions, which also have smaller ionic radii in comparison with Mn³⁺ (0.645 Å). However, the doping of Sr²⁺ and Ba²⁺ on the A site of LMO shows a downshift of XRD patterns due to the fact that the ionic radii of Sr²⁺ (1.44 Å) and Ba²⁺ (1.61 Å) are larger than that of La³⁺ (1.36 Å). Fig. 1b shows the scanning electron microscopy (SEM) image of LCMGO, revealing a porous structure with interconnected nano-sized particles (<100 nm, see the inset of Fig. 1b). Although the introduction of metal ions into the A and B sites of LMO leads to varied crystal lattice parameters, the surface morphologies of the perovskites remain similar, reflecting the retention of the parent structural morphology (Fig. S3, ESI†). The transmission electron microscopy (TEM) image (Fig. 1c) further confirms that the particle size of LCMGO is smaller than 100 nm. The high-resolution transmission electron microscopy (HRTEM) image (bottom inset of Fig. 1c) shows a lattice spacing of 0.27 nm corresponding to the (110) reflection plane of the crystal structure of LCMGO, which is consistent with the XRD result (Fig. 1a). Furthermore, the elemental distributions in the perovskites were mapped by the energy dispersive X-ray (EDX) spectroscopic technique. As shown in Fig. 1d, a homogeneous distribution of the corresponding Ca, Mn and Ga elements in the LCMGO sample was observed. Furthermore, the BET surface areas of LaMnO₃ based perovskites before and after thermochemical measurements were also investigated. As summarised in Table S1,† LCMGO showed the largest surface area among the tested perovskite samples, which could promote reactant transportation during the thermochemical process and lead to higher solar fuel production yield. Based on the XRD, SEM, HRTEM, EDX and BET analyses, it is evident that LCMGO perovskite was successfully synthesised through the manipulation of metal ion doping on both A and B sites of the parent LMO perovskite.

Fig. 1 (a) XRD pattern, (b) SEM images, (c) TEM, and (d) EDX elemental mapping of Ca, Mn and Ga distribution of LCMGO perovskite. The insets of (b) and (c) show the high magnification SEM and local HRTEM images.

The thermochemical H₂O splitting performances were evaluated in a laboratory-scale fixed-bed reactor (Fig. S4, ESI†) and the O₂ and H₂ produced during the thermochemical process were monitored using a mass spectrometer (MS). First, we compared the thermochemical performances of LMO with the A site doped La_0.6A_0.4MnO₃ (A = Ca²⁺, Sr²⁺, and Ba²⁺, labelled LCMO, LSMO and LBMO, respectively) perovskites. The extents of oxygen vacancies generated by the prepared samples in high-temperature thermal reduction processes were investigated by measuring O₂ production, which in turn will determine the maximum solar fuel production. As shown in Fig. 2a, LMO and the A site doped La_0.6A_0.4MnO₃ perovskites started to release O₂ at around 950 °C and the O₂ evolution process was completed after holding at 1300 °C for 1 h. The figure demonstrates that compared to the parent LMO, the A site doped La_0.6A_0.4MnO₃ perovskites exhibit significantly improved O₂ evolution profiles. In particular, LCMO required the shortest time to reach the peak rate due to its unique orthorhombic perovskite structure, which can significantly promote oxygen transportation during the thermal reduction process.³¹ As a result, LCMO (167 μmol g⁻¹) has the highest O₂ evolution performance among all the tested A site doped La_0.6A_0.4MnO₃ perovskites, which is 2.19 times higher than that produced by using the parent LMO (76 μmol g⁻¹). Thus we can conclude that partial substitution of La³⁺ with alkali earth metal ions (Ca²⁺, Sr²⁺ and Ba²⁺) on the A site of LMO can greatly promote O₂ evolution during the thermal reduction process and thereby achieve a larger extent of O₂ non-stoichiometry.

Fig. 2 (a) O₂ and (b) H₂ production curves of LaMnO₃ and A site doped La_0.6A_0.4MnO₃ (A = Ba²⁺, Sr²⁺ and Ca²⁺) perovskites.

In the second step, H₂O splitting tests were conducted at 900 °C and H₂ production was immediately detected after the aspiration of H₂O vapour into the system. Fig. 2b shows the H₂ production curves of the LMO and the A site doped La_0.6A_0.4MnO₃ perovskites. The H₂ production had not completely ceased after 60 min of reaction, leaving a long tail after 30 min, which is consistent with the results reported elsewhere.^29,38,39 The sluggish H₂O splitting reaction kinetics of perovskites may arise from several factors, such as limited chemical diffusion and decreased surface reaction constants.²⁹ Nevertheless, over 80% of the total H₂ production was achieved within 30 min. The H₂ production trend of the A site doped perovskite concurs with the trend obtained for O₂ production (Fig. 2a). As expected, LCMO shows significantly improved O₂ production (312 μmol g⁻¹), which is almost 6 times higher than that of LMO (48 μmol g⁻¹). From the final results summarised in Table 1, it is clear that both O₂ and H₂ production follow the same trend, i.e. LMO < LBMO < LSMO < LCMO. In addition, their re-oxidation yields also follow the same trend. The better thermochemical performance of LCMO may be attributed to the lower reduction enthalpy and lighter atomic weight of Ca²⁺ when compared to those of Sr²⁺ and Ba²⁺ at the same doping level.³⁵ To confirm the effectiveness of transforming perovskite powder into a porous monolith structure, we compared their thermochemical performances under the same conditions. The O₂ evolution results (Fig. S5, ESI†) obtained for the porous monolith and perovskite powder are very close. This confirms the identical material composition of both powder and porous monolith structured perovskites. However, a significant difference in the H₂ production performance was observed which is mainly due to the increased surface area of the monolith structure compared to that of the powder sample.

Table 1 Summarised thermochemical H₂O splitting performances of LaMnO₃ and LaMnO₃ based perovskites with A and B sites doped with Ba²⁺, Sr²⁺, Ca²⁺, Al³⁺ and Ga³⁺ metal ions, including O₂ and H₂ production amounts and re-oxidation yields

Redox materials	O2 (μmol g^−1)	H2 (μmol g^−1)	Re-oxidation yield (%)
LMO	76	48	31
LBMO	113	179	79
LSMO	126	215	85
LCMO	167	312	93
LCMAO	184	339	92
LCMGO	212	401	95

---

The B site doping of perovskites can also significantly affect their thermochemical performances. Trivalent metal ion doping on the B site of LMO has enhanced the extent of oxygen vacancies and redox behaviour for some catalytic reactions.^38,40,41 Therefore, to further improve the thermochemical performance of LCMO, two different trivalent metal ions, Al³⁺ and Ga³⁺, were partially doped on the B site (La_0.6Ca_0.4Mn_0.8Al_0.2O₃ and La_0.6Ca_0.4Mn_0.8Ga_0.2O₃, labelled LCMAO and LCMGO, respectively). As shown in Fig. 3, both LCMAO and LCMGO exhibited higher thermochemical performances than LCMO, confirming the positive effect of the B site doping strategy. More importantly, the doping of Ga³⁺ on the B site of LCMO showed better promotion of H₂ production performance than Al³⁺ doping. As summarised in Table 1, LCMGO exhibits a H₂ production of 401 μmol g⁻¹, which is 30% higher than that achieved by LCMO (312 μmol g⁻¹), 20% higher than that of LCMAO (339 μmol g⁻¹) and twelve times higher than that of the benchmark CeO₂ under the same experimental conditions (Fig. S6, ESI†).

Fig. 3 (a) O₂ and (b) H₂ production curves of LCMAO and LCMGO perovskites. The thermochemical process was carried out under the same conditions as those used for LCMO, operated between 1300 °C and 900 °C.

The noteworthy thermochemical performance of LCMGO may arise from its outstanding BET surface area, which greatly promoted the solid–gas reaction during the thermochemical catalytic process with remarkable H₂ production. Furthermore, as reported by Patrakeev et al., the doping of Ga³⁺ in similar La_1−xSr_xFeO₃ perovskite systems was proved to decrease the reduction enthalpy and resulted in enhanced oxygen vacancy formation.⁴¹ To the best of our knowledge, the H₂ production of LCMGO is at the top level of all the reported perovskite type catalysts (Table S2†). It is worth mentioning that the thermal reduction temperature (1300 °C) we applied in this study is lower than those used for most of the reported perovskites (≥1400 °C, refer to Table S2†). The lowered thermal reduction temperature is beneficial for the ease of solar reactor design and may result in higher energy conversion efficiency. The H₂ yield of LCMGO was further improved to 473 μmol g⁻¹ (Fig. S7, ESI†), which is the highest H₂ production amount reported to date when the thermal reduction temperature was increased to 1400 °C during the thermochemical H₂O splitting process. However, serious sintering and aggregation phenomena were detected in LCMGO when thermally reduced at 1400 °C (Fig. S8, ESI†) and the re-oxidation yield was found to drop from 95% (1300 °C) to 74% (1400 °C), which will significantly decrease the catalytic durability for long-term thermochemical application.

Durability is another critical parameter for assessing the viability of redox materials for thermochemical solar fuel production. Thus, multiple cycles of thermochemical H₂O splitting tests were conducted on LCMGO between 1300 °C and 900 °C. As shown in Fig. 4a, stable O₂ and H₂ evolution profiles were observed during the cycling measurements. This trend can also be seen in the summarised O₂ and H₂ production amount per cycle (Fig. 4b). Moreover, the molar ratio of H₂ to O₂ is quite close to the theoretical value of 2, which means that almost all the O₂ released during the thermal reduction process is recovered during the re-oxidation process. Furthermore, the sample exhibited reliable stability and no crystal structure change (Fig. S9, ESI†) or severe sintering phenomena were observed after the cycling test (Fig. S10, ESI†).

Fig. 4 (a) Six consecutive O₂ and H₂ production profiles and (b) O₂ and H₂ yields and their corresponding ratio as a function of cycle number.

Conclusions

A systematic investigation of the effects of doping LaMnO₃ based perovskites on both the A and B sites on their thermochemical H₂O splitting performances was carried out at a low operating temperature of 1300 °C. Our results demonstrate that doping on either A (Ca²⁺, Sr²⁺, and Ba²⁺) or B (Al³⁺ and Ga³⁺) sites of LaMnO₃ perovskites has a positive impact on the two-step thermochemical H₂O splitting performances. More importantly, the partial doping of Ca²⁺ on the A site and Ga³⁺ on the B site produces a novel perovskite composition (La_0.6Ca_0.4Mn_0.8Ga_0.2O₃) which shows a much higher H₂ production (401 μmol g⁻¹) than the benchmark CeO₂. Furthermore, La_0.6Ca_0.4Mn_0.8Ga_0.2O₃ perovskite exhibited superb cycling stability during the thermochemical H₂O splitting process. We, therefore, believe that this doping strategy can help develop highly efficient and stable perovskites for other energy conversion and storage applications.

Acknowledgements

This work was financially supported by the Australian Research Council (ARC) and the National Natural Science Foundation of China (Grant no. 51372248 and 51432009).

Notes and references

1. S. Z. Baykara, Int. J. Hydrogen Energy, 2004, 29, 1451.
2. W. C. Chueh, C. Falter, M. Abbott, D. Scipio, P. Furler, S. M. Haile and A. Steinfeld, Science, 2010, 330, 1797.
3. T. Kodama and N. Gokon, Chem. Rev., 2007, 107, 4048.
4. M. P. Pandey and C. S. Kim, Chem. Eng. Technol., 2011, 34, 29.
5. M. Romero and A. Steinfeld, Energy Environ. Sci., 2012, 5, 9234.
6. J. E. Miller, A. H. McDaniel and M. D. Allendorf, Adv. Energy Mater., 2014, 4, 1300469.
7. S. Abanades and G. Flamant, Sol. Energy, 2006, 80, 1611.
8. S. Abanades, P. Charvin, G. Flamant and P. Neveu, Energy, 2006, 31, 2805.
9. P. Furler, J. R. Scheffe and A. Steinfeld, Energy Environ. Sci., 2012, 5, 6098.
10. T. Kodama, Y. Nakamuro and T. Mizuno, J. Sol. Energy Eng., 2004, 128, 3.
11. P. Charvin, S. Abanades, G. Flamant and F. Lemort, Energy, 2007, 32, 1124.
12. M. E. Gálvez, P. G. Loutzenhiser, I. Hischier and A. Steinfeld, Energy Fuels, 2008, 22, 3544.
13. K. Wegner, H. C. Ly, R. J. Weiss, S. E. Pratsinis and A. Steinfeld, Int. J. Hydrogen Energy, 2006, 31, 55.
14. P. G. Loutzenhiser, A. Meier and A. Steinfeld, Materials, 2010, 3, 4922.
15. A. Stamatiou, P. G. Loutzenhiser and A. Steinfeld, Chem. Mater., 2010, 22, 851.
16. W. C. Chueh, C. Falter, M. Abbott, D. Scipio, P. Furler, S. M. Haile and A. Steinfeld, Science, 2010, 330, 1797.
17. P. Furler, J. Scheffe, M. Gorbar, L. Moes, U. Vogt and A. Steinfeld, Energy Fuels, 2012, 26, 7051.
18. W. C. Chueh and S. M. Haile, ChemSusChem, 2009, 2, 735.
19. W. C. Chueh and S. M. Haile, Philos. Trans. R. Soc., A, 2010, 368, 3269.
20. Q. Q. Jiang, J. H. Tong, G. L. Zhou, Z. X. Jiang, Z. Li and C. Li, Sol. Energy, 2014, 103, 425.
21. A. Demont, S. Abanades and E. Beche, J. Phys. Chem. C, 2014, 118, 12682.
22. Z. Zhang, L. Andre and S. Abanades, Sol. Energy, 2016, 134, 494.
23. C. L. Muhich, B. D. Ehrhart, I. Al-Shankiti, B. J. Ward, C. B. Musgrave and A. W. Weimer, Wiley Interdiscip. Rev.: Energy Environ., 2016, 5, 261.
24. M. M. Nair and S. Abanades, ChemistrySelect, 2016, 1, 4449.
25. M. E. Galvez, R. Jacot, J. Scheffe, T. Cooper, G. Patzke and A. Steinfeld, Phys. Chem. Chem. Phys., 2015, 17, 6629.
26. T. Cooper, J. R. Scheffe, M. E. Galvez, R. Jacot, G. Patzke and A. Steinfeld, Energy Technol., 2015, 3, 1130.
27. A. M. Deml, V. Stevanović, A. M. Holder, M. Sanders, R. O'Hayre and C. B. Musgrave, Chem. Mater., 2014, 26, 6595.
28. J. R. Scheffe, D. Weibel and A. Steinfed, Energy Fuels, 2013, 27, 4250.
29. C. K. Yang, Y. Yamazaki, A. Aydin and S. M. Haile, J. Mater. Chem. A, 2014, 2, 13612.
30. A. Demont and S. Abanades, RSC Adv., 2014, 4, 54885.
31. S. Dey, B. S. Naidu, A. Govindaraj and C. N. R. Rao, Phys. Chem. Chem. Phys., 2015, 17, 122.
32. A. H. McDaniel, E. C. Miller, D. Arifin, A. Ambrosini, E. N. Coker, R. O'Hayre, W. C. Chueh and J. H. Tong, Energy Environ. Sci., 2013, 6, 2424.
33. M. A. Peña and J. L. G. Fierro, Chem. Rev., 2001, 101, 1981.
34. L. G. Tejuca, J. L. G. Fierro and J. M. D. Tascon, Adv. Catal., 1989, 36, 237.
35. A. Demont and S. Abanades, J. Mater. Chem. A, 2015, 3, 3536.
36. Y.-S. Han and H.-G. Kim, J. Power Sources, 2000, 88, 161.
37. Y. Hao, C. K. Yang and S. M. Haile, Chem. Mater., 2014, 26, 6073.
38. S. Dey, B. S. Naidu and C. N. R. Rao, Dalton Trans., 2016, 45, 2430.
39. S. Dey, B. S. Naidu and C. N. R. Rao, Chem.–Eur. J., 2015, 21, 7077.
40. S. Cimino, L. Lisi, S. De Rossi, M. Faticanti and P. Porta, Appl. Catal., B, 2003, 43, 397.
41. M. V. Patrakeev, E. B. Mitberg, A. A. Lakhtin, I. A. Leonidov, V. L. Kozhevnikov, V. V. Kharton, M. Avdeev and F. M. B. Marques, J. Solid State Chem., 2002, 167, 203.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6se00097e

---