An oxygen pool from YBaCo₄O₇-based oxides for soot combustion

Abstract

Soot, often referred to as black carbon emitted from diesel engines, is not only a particulate matter pollutant but also a light-absorbing agent that may affect global climate, but can be effectively controlled using a catalytic diesel particulate filter (DPF). A new YBaCo₄O_7+δ-type oxygen storage material is reported as an effective catalyst for soot combustion. Isotopic isothermal reactions demonstrate the activation of gaseous oxygen and subsequent oxygen storage and reaction/desorption during an oxidation process. High activity and structural stability are achieved by the substitution of Co with Al and Ga to form YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ. The specific rates at 300 °C of YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ, normalized by surface areas, are an order of magnitude higher than those of CeO₂-based oxides. This kind of oxygen-storage material acts as an oxygen pool, which ensures that the accumulated soot on a DPF can be promptly combusted.

---

1. Introduction

Soot, often referred to as black carbon, is not only a particulate matter pollutant but also a light-absorbing agent that may affect global climate.¹ Diesel engines are amongst the most abundant emission sources of soot that can be effectively controlled by a catalytic diesel particulate filter (DPF).² Commercial catalysts are composed of noble metals (Pt and Pd) supported on ceria-based oxides,³ oxidizing NO into NO₂, which is transported through a gas phase to soot aggregates where it oxidizes carbon while being reduced to NO.⁴ A so-called continuously regenerated trap (CRT) overcomes the problem of poor contact between soot and a catalyst. However, the limited amounts of NO_x present and the need to control NO_x for Euro IV standards are not ideal situations to meet the requirements of the latest generation of diesel engines, leading to a drive to develop catalysts which can produce highly reactive oxygen species from gaseous O₂ molecules.^5,6

Among this kind of catalysts, the extensively studied soot oxidation catalysts are ceria-based oxides due to the redox properties of the Ce³⁺/Ce⁴⁺ couple and the capacity of ceria to exchange oxygen with a gas phase.^7–12 Unfortunately, the effective utilization of such active oxygen species is limited by the extent of contact between soot and a catalyst. This limitation could be overcome by producing a sufficiently high amount of active oxygen to ensure that, at least, a proportion reaches the soot particle surface.⁵

In comparison with a conventional oxygen storage material, CeO₂–ZrO₂, YBaCo₄O₇ (114 structure) shows a markedly larger oxygen storage capacity (OSC), particularly at low temperatures (200–400 °C), which is just within the range of temperatures reached at the exhaust of a diesel engine.^13–16 The extraordinary oxygen storage capability of the YBaCo₄O₇-based oxides is due to the variable valence of Co ions between Co²⁺ and Co³⁺. The limitation of YBaCo₄O_7+δ is that it decomposes just above 600 °C in an oxygen-containing atmosphere, which limits its applications in catalytic combustion at elevated temperatures.¹³ Recently, Karppinen and colleagues identified that an YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ phase, where Co is co-substituted by Al and Ga is stable up to high temperatures under oxidizing conditions,¹⁶ which creates the potential for YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ to become a more effective catalyst for catalytic combustion.

In this paper, YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ were studied from the perspective of catalysis for diesel soot combustion. Their oxygen storage performance allows them to create an oxygen pool supply of active oxygen which is created from a gaseous phase.¹⁷ Although YBaCo₄O_7+δ has been reported as a robust catalyst for H₂O₂ oxidation of cyclohexene in the liquid phase,¹⁸ the present result is the first report on high-temperature oxidation reactions for this kind of non-stoichiometric transition metal oxide materials.

2. Experimental section

The YBaCo₄O₇-type samples were synthesized from a precursor powder prepared by an EDTA (ethylenediaminetetraacetic acid) and CA (citric acid) complex gel method,¹³ using Y(NO₃)₃·6H₂O, Ba(NO₃)₂, Co(NO₃)₃·6H₂O, Ga(NO₃)₃·6H₂O and Al(NO₃)₃·9H₂O as the starting materials. The precursor powder was calcined in a muffle at 350 °C for 2 h, then pelletized and sintered in air at 1000 °C for 24 h, followed by rapid cooling to room temperature. Further thermal treatment at 300 °C under a pure O₂ flow for 2 h obtained YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ.

X-ray powder diffraction (XRD) patterns were recorded on a Rigaku D/max-RC diffractometer employing Cu Kα radiation. Surface areas and pore size distributions were determined by N₂ adsorption–desorption at 77 K using a Micromeritics ASAP 2020 instrument after outgassing at 300 °C for 5 h prior to analysis.

Temperature programmed desorption of O₂ (O₂–TPD) experiments were conducted in a fixed-bed flow reactor. A 150 mg sample was heated under a flow of high purity O₂ (30 ml min⁻¹) at 300 °C for 1 h. After cooling to room temperature, high purity He was introduced. Desorption was started at a heating rate of 2 °C min⁻¹ in He (30 ml min⁻¹). The desorbed O₂ was monitored by a quadruple mass spectrometer (MS, OminiStar 200, Balzers).

Temperature programmed reduction with H₂ (H₂–TPR) experiments were performed in a quartz reactor with a thermal conductivity detector (TCD) to monitor H₂ consumption. A 50 mg sample was pretreated in situ at 300 °C for 1 h under a flow of O₂ and cooled to room temperature in the presence of O₂. After purging with N₂, TPR was conducted at 10 °C min⁻¹ up to 700 °C under a 30 mL min⁻¹ flow of 5 vol% H₂ in N₂. To quantify the total amount of H₂ consumption, CuO was used as a calibration reference.

“Dynamic” OSC (DOSC) measurements with CO–O₂ pulses were carried out at 200–500 °C. CO (4% CO/1% Ar/He at 300 mL min⁻¹ for 10 s) and O₂ (2% O₂/1% Ar/He at 300 mL min⁻¹ for 10 s) streams were pulsed alternately with at a frequency of 0.05 Hz. A DOSC value was obtained by integrating the CO₂ formed during one CO–O₂ cycle and was expressed as μmol of O per gram of catalyst (μmol [O] g⁻¹). The concentration of CO₂ was determined using a MS.

Temperature programmed oxidation (TPO) reactions were conducted in a fixed bed micro reactor consisting of a quartz tube (6 mm i. d.). Printex-U from Degussa was used as the model soot. The soot was mixed with the catalyst in a weight ratio of 1 : 9 in an agate mortar for 30 min, which resulted in a tight contact between the soot and the catalyst. A 50 mg sample of the soot/catalyst mixture was pre-treated under a flow of He (50 mL min⁻¹) at 200 °C for 30 min to remove adsorbed species. After cooling to room temperature, a gas flow with 5 vol% oxygen in He was introduced and then TPO was initiated at a heating rate of 5 °C min⁻¹ until 880 °C. For pure soot combustion (non-catalytic), the catalyst was substituted by silica. CO and CO₂ concentrations in the effluent gas were monitored using an online gas chromatograph (GC) (SP-6890, Shandong Lunan Ruihong Chemical Instrument Corporation, China) fitted with a methanator. The ignition temperature for soot combustion was evaluated by the value of T₁₀, which is defined as the temperature at which 10% of the soot is converted. The selectivity to CO₂ is defined as the percentage CO₂ outlet concentration divided by the sum of the CO₂ and CO outlet concentrations.

Isothermal reactions at 300 °C, at which a stable and low soot conversion (<15%) was achieved, were conducted within the kinetic regime. The reaction rate for soot combustion was obtained from the slope of the conversion lines with time. Specific rates normalized by BET surface areas and turnover frequency (TOF)¹⁹ were used to characterize the activity for soot combustion.

An isotopic isothermal reaction was performed by switching the flowing gas from 1% ¹⁶O₂ to 1% ¹⁸O₂ diluted in Ar at 350 °C. 50 mg of a mixture of the soot and catalyst in a tight contact mode was employed. The effluent gas from the reactor was continuously monitored by a MS for all of the isotopic molecules of CO₂ (at m/z = 44, 46 and 48).

3. Results and discussion

The fresh YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ were confirmed to be composed of a single phase (Fig. S1†), similar to that of hexagonal LuBaAlZn₃O₇ (JCPDS 40-4099).¹³ As the shifts in the XRD peaks for the separate substitution of Co by Al (r = 0.039 nm) and Ga (r = 0.047 nm) are opposed for each sample (Fig. S2†), the simultaneous substitution has a slight effect on the cell volume parameters. Furthermore, both ionic radii of Al and Ga are smaller than the high spin Co²⁺ ionic radius (r = 0.058 nm), thus Maignan et al. suggested that Al and Ga are substituted for Co³⁺.²⁰ The BET surface areas are fairly low, in accordance with the high-temperature sintering preparation (Table S1†) and the highly crystalline nature of the samples.

In agreement with conclusions in the literature,^13,16 the O₂–TPD profiles (Fig. 1) show that both YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ desorb a large amount of oxygen below 400 °C, corresponding to δ = 0.37 and δ = 0.34, respectively. The peak maximum of YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ is 50 °C lower than that of YBaCo₄O_7+δ, which suggests a promotion effect on O₂ desorption by doping, albeit with a slight decrease of the overall oxygen storage capability.¹⁶ Above 700 °C, further desorption of O₂ is observed, corresponding to the possible decomposition of the 114 structure.¹³ However, YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ shows less pronounced O₂ desorption than YBaCo₄O_7+δ. After O₂–TPD at about 900 °C, the Y₂O₃ and CoO phases are segregated from YBaCo₄O_7+δ, which is opposite to the stable structure of YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ (Fig. S3†).

Fig. 1 O₂–TPD spectra of YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ from room temperature to 900 °C.

A similar situation is observed in H₂–TPR (Fig. S1†). The 114 structure of YBaCo₄O_7+δ is completely destroyed by H₂ in TPR to 700 °C (Fig. S1a and b†). In contrast, YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ was preserved with no formation of new oxide phases (Fig. S1c and d†). The structural stability of YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ is vital to high-temperature redox reactions. As shown in Fig. 2, two peaks were observed for both YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ. The low-temperature TPR peak can be assigned to the removal of non-stoichiometric excess O accommodated within the lattice, and the values of which are slightly larger than those consumed by δ (Table 1). The 114 structure of YBaCo₄O_7+δ is stable at this stage (Fig. S1a and b†). The second peak corresponds to the reduction of bulk and surface Co³⁺ of YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ, respectively. The substitution of Co³⁺ by Al and Ga protects the structure from decomposition under reducing atmospheres (H₂–TPR and O₂–TPD).²⁰ Furthermore, the lower TPR temperature of YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ in comparison with that of YBaCo₄O_7+δ coincides with the O₂–TPD results.

Fig. 2 H₂–TPR spectra of YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ from room temperature to 700 °C.

Table 1 H₂–TPR peak temperatures (T, °C), H₂ consumption (C, μmol g⁻¹), specific rates at 300 °C normalized by BET surface areas (mol s⁻¹ m⁻² × 10⁻⁸), and TOF (s⁻¹ × 10⁻³) at 290 °C

Sample	Peak 1		Peak 2		Specific rates^a	TOF
	T	C	T	C
a The specific rate at 300 °C normalized by BET surface areas (80.4 m^2 g^−1) for Ce0.43Zr0.57O2 is 0.21 mol s^−1 m^−2 × 10^−8.
YBaCo4O7+δ	275	74.9	475	298.5	2.11	1.34
YBa(Co0.85Al0.075Ga0.075)4O7+δ	255	57.8	390	10.0	2.17	1.57

---

Although H₂–TPR and O₂–TPD data may be useful in rapidly evaluating the potential OSC of the candidate materials, DOSC provides better simulation of instantaneous oscillations between lean (oxidizing) and rich (reducing) exhaust conditions during real operation and is therefore much more useful in the evaluation of the activity of OSC materials.²¹Fig. 3(a) shows the collected DOSC data and the corresponding transition curves at 320 °C of YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ as an example, with alternate dynamic pulses of 4% CO/1% Ar/He (10 s) and 2% O₂/1% Ar/He (10 s) under 0.05 Hz given in Fig. 3(b). In comparison with CeO₂–ZrO₂, the normalized DOSC values by the BET surface areas of YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ are more than thirty times larger (DOSC values of YBaCo₄O_7+δ, YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ and Ce_0.43Zr_0.57O₂ at 400 °C are 40.9, 41.6 and 1.3 μmol [O] m⁻², respectively),²² confirming the fast responses between reduction and oxidation environments. Furthermore, the higher DOSC of YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ in comparison with that of YBaCo₄O_7+δ is consistent with the O₂–TPD and H₂–TPR results.

Fig. 3 DOSC of YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ (a) and enlarged transition curves of YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ (b) at 320 °C with dynamic pulses of 4% CO/1% Ar/He and 2% O₂/1% Ar/He under 0.05 Hz. CO₂(I) was produced by CO and surface oxygen; CO₂(II) was attributed to the reaction of absorbed CO and oxygen gas.²²

The catalytic activity for soot combustion was first checked by TPO (Fig. 4a). YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ decrease from a T₁₀ value of 530 °C for non-catalytic combustion to 387 and 379 °C, respectively, confirming the catalytic effect of the YBaCo₄O_7+δ-type material and the higher activity of the latter than the former. In terms of the selectivity towards CO₂ formation, the non-catalytic combustion is only 43.3%, while YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ yield nearly 100% CO₂. After the TPO reactions, no phase decomposition occurs even for YBaCo₄O_7+δ (Fig. S4†), probably due to the high heating rate in 5 vol% oxygen in He.²³ Furthermore, the XRD peaks of YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ after TPO shift to higher angles, suggesting a lattice shrinkage, which confirms the participation of bulk oxygen.

Fig. 4 TPO patterns of CO_x for soot combustion with O₂ over YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ (a); isothermal reactions for soot combustion at 300 °C within the kinetic regime (b) under tight contact conditions between the soot and catalysts.

The intrinsic activity was further demonstrated by kinetic rates at 300 °C, which can be obtained from the slope of the lines shown in Fig. 4b. As observed in Table 1, the specific rates of YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ, normalized by BET surface areas, are an order of magnitude larger than that of Ce_0.43Zr_0.57O₂.^19–22 This is significant because 300 °C is a relevant temperature for light diesel engines. This particularly high reaction rate can ensure that the accumulated soot on the DPF can be readily combusted, leading to a lower balance point temperature (BPT) at which the rate of soot oxidation is matched with the rate of soot accumulation.²⁴ Furthermore, both the specific rate and TOF (Table 1) of YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ are a little higher than that of YBaCo₄O_7+δ, confirming the higher intrinsic activity of YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ.

In order to explore the origin of the active oxygen, isotopic isothermal oxidation at 350 °C was performed (Fig. 5). Before switching from ¹⁶O₂ to ¹⁸O₂ (to the left of the dashed line), the main product was C¹⁶O₂, confirming that the soot oxidation occurs with the bulk oxygen species. The concentration of C¹⁶O₂ for YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ is much larger than that for YBaCo₄O_7+δ, which is again consistent with the O₂–TPD, H₂–TPR, DOSC, T₁₀ and specific rates. After switching from ¹⁶O₂ to ¹⁸O₂ (to the right of the dashed line), the sum of the products of C¹⁶O¹⁸O, C¹⁸O₂ and ¹⁶O¹⁸O still possesses higher concentrations of YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ than that of YBaCo₄O_7+δ. However, the concentration of C¹⁶O₂ decreased gradually to a very low value due to the depletion of ¹⁶O₂ in the gaseous phase. Comparatively, for both YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ, the production of C¹⁶O¹⁸O increases and then reaches a maximum, while the C¹⁸O₂ production monotonically increases. This indicates that only the gaseous oxygen which has been activated by YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ can be used to oxidize the soot. In addition, the desorption of ¹⁶O¹⁸O is detected, which suggests that the desorbed ¹⁶O¹⁸O species do not interact with the soot to produce the product containing carbon, demonstrating that the intimate contact between the soot and catalysts is essential. Since the high DOSC can make up for the missing oxygen, the intrinsic activity of YBaCo₄O₇-based oxides is much more active than that of CeO₂-based oxides.

Fig. 5 Isothermal reactions for soot combustion at 350 °C after 1% ¹⁶O₂ was switched to 1% ¹⁸O₂ in He on YBaCo₄O_7+δ (a) and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ (b).

Conclusions

In conclusion, a new YBaCo₄O_7+δ-type oxygen storage material was used to catalyze soot combustion. Isotopic isothermal reactions demonstrate the activation of gaseous oxygen and subsequent oxygen storage and reaction/desorption during the oxidation process. Higher activity and structural stability are achieved by the substitution of Co with Al and Ga to form YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ. The specific rates at 300 °C of YBaCo₄O_7+δ and YBa(Co_0.85Al_0.075Ga_0.075)₄O_7+δ, normalized by BET surface areas, are an order of magnitude larger than that of CeO₂-based oxides. This type of oxygen-storage material provides an oxygen pool, which ensures that the accumulated soot on a DPF can be readily combusted.

Acknowledgements

This work is financially supported by the National Natural Science Foundation of China (no. 21477046, 21277060 and 21547007).

Notes and references

1. B. Booth and N. Bellouin, Nature, 2015, 509, 167–168.
2. Barry A. A. L. van Setten, M. Makkee and J. A. Moulijn, Catal. Rev.: Sci. Eng., 2001, 43, 489–564.
3. S. Sumiya, L. Wang and G. Zhang, SAE [Tech. Pap.], 2012, 2012-28-0009.
4. K. Leistner, A. Nicolle and P. Da Costa, J. Phys. Chem. C, 2012, 116, 4642–4654.
5. I. Atribak, F. E. López-Suárez, A. Bueno-López and A. García-García, Catal. Today, 2011, 176, 404–408.
6. E. Obeid, L. Lizarraga, M. N. Tsampas, A. Cordier, A. Boréave, M. C. Steil, G. Blanchard, K. Pajot and P. Vernoux, J. Catal., 2014, 309, 87–96.
7. Q. Liang, X. Wu, D. Weng and H. Xu, Catal. Today, 2008, 139, 113–118.
8. J. Liu, Z. Zhao, J. Wang, C. Xu, A. Duan, G. Jiang and Q. Yang, Appl. Catal., B, 2008, 84, 185–195.
9. E. Aneggi, C. de Leitenburg and A. Trovarelli, Catal. Today, 2012, 181, 108–115.
10. A. Buenolopez, K. Krishna, M. Makkee and J. Moulijn, J. Catal., 2005, 230, 237–248.
11. K. Harada, T. Oishi, S. Hamamoto and T. Ishihara, J. Phys. Chem. C, 2014, 118, 559–568.
12. A. Bueno-López, Appl. Catal., B, 2014, 146, 1–11.
13. M. Karppinen, H. Yamauchi, S. Otani, T. Fujita, T. Motohashi, Y. H. Huang, M. Valkeapää and H. Fjellvåg, Chem. Mater., 2006, 18, 490–494.
14. S. Kadota, M. Karppinen, T. Motohashi and H. Yamauchi, Chem. Mater., 2008, 20, 6378–6381.
15. Y. Jia, H. Jiang, M. Valkeapää, H. Yamauchiä, M. Karppinen and E. I. Kauppinen, J. Am. Chem. Soc., 2009, 131, 4880–4883.
16. O. Parkkima, H. Yamauchi and M. Karppinen, Chem. Mater., 2013, 25, 599–604.
17. K. Sato, M. Yamaguchi, S. Fujita, K. Suzuki and T. Mori, Catal. Commun., 2006, 7, 132–135.
18. O. Parkkima, A. Silvestre-Albero, J. Silvestre-Albero and M. Karppinen, Catal. Lett., 2014, 145, 576–582.
19. Z. Zhang, D. Han, S. Wei and Y. Zhang, J. Catal., 2010, 276, 16–23.
20. A. Maignan, V. Caignaert, V. Pralong, D. Pelloquin and S. Hébert, J. Solid State Chem., 2008, 181, 1220–1226.
21. M. Boaro, F. Giordano, S. Recchia, V. D. Santo, M. Giona and A. Trovarelli, Appl. Catal., B, 2004, 52, 225–237.
22. Z. Zhang, Y. Fan, Y. Xin, Q. Li, R. Li, J. A. Anderson and Z. Zhang, Environ. Sci. Technol., 2015, 49, 7989–7995.
23. T. Komiyama, T. Motohashi, Y. Masubuchi and S. Kikkawa, Mater. Res. Bull., 2010, 45, 1527–1532.
24. J. Oi-Uchisawa, A. Obuchi, S. Wang, T. Nanba and A. Ohi, Appl. Catal., B, 2003, 43, 117–129.

---

Footnote

† Electronic supplementary information (ESI) available: Textural properties and XRD patterns of the samples. See DOI: 10.1039/c5cy02181b

---