Kinetics of CaMn0.775Ti0.125Mg0.1O2.9-δ perovskite prepared at industrial scale and its implication on the performance of chemical looping combustion of methane

https://doi.org/10.1016/j.cej.2020.124863Get rights and content

Highlights

  • Reaction kinetics of a CaMn-perovskite with CH4, CO, H2 and O2 was determined.

  • Perovskite increased reactivity with redox cycles.

  • Kinetics was implemented in a reactor model for chemical looping combustion.

  • Design of the fuel reactor was explored for activated and non-activated material.

Abstract

The reaction kinetics of a CaMn0.775Ti0.125Mg0.1O2.9-δ perovskite, prepared at ton-scale to be used as an oxygen carrier in a CLC process with methane, was determined. The variation of the reaction rates, oxygen transport capacity, and oxygen uncoupling behaviour of the selected material with the number of cycles was evaluated in a thermogravimetric analyser.

The perovskite was activated for the reduction reaction with all the combustible gases, i.e., H2, CO and CH4. On the contrary, the oxidation rate barely changed throughout the cycles. The oxygen transport capacity underwent a slight decrease during the activation period.

Kinetics of reduction, oxidation and oxygen uncoupling reactions were determined for the fresh and activated material assuming a grain model controlled by chemical reaction at low values of solids conversion and diffusion through the product layer at high values of solids conversion. This model predicted correctly the solids conversion vs. time curves as a function of temperature and gas concentration for the oxygen carrier reactions evaluated in this kinetics study.

The kinetics data were introduced in a validated mathematical model of a fuel reactor to analyse the amount of solids required to fully combust CH4. A solids inventory of 350 kg/MWth was estimated in the fuel reactor at a temperature of 1243 K. However, this value could be reduced to 160 kg/MWth if the material operates under activation conditions within the CLC unit.

Introduction

According to the Intergovernmental Panel on Climate Change (IPCC), global warming is unequivocal and can be attributed to human action with a certainty of more than ninety percent [1], mainly due to emissions of CO2 to the atmosphere, e.g. by burning fossil fuels to produce energy. Avoiding this problem represents a colossal environmental challenge with effects on the global economy, health and social welfare. Most scenarios related to global energy consumption predict a substantial increase of CO2 emissions throughout the coming decades of the 21st century unless specific measures are taken to mitigate climate change [1], [2]. These scenarios also foresee that fossil fuels will dominate the primary energy supply at least until 2050. Combined with other technological options, carbon capture and storage (CCS) techniques will be required to fight against climate change and stabilize the CO2 concentration in the atmosphere.

Compared to other CO2 capture technologies, chemical looping combustion (CLC) allows the conversion of fuels with inherent CO2 separation. In CLC, an oxygen carrier (typically a metal oxide) supplies the oxygen for the fuel combustion. In a preferred configuration, the oxygen carrier is continuously circulating between two fluidized bed reactors: (1) the fuel reactor, where the fuel is burnt; and (2) the air reactor, where the oxygen carrier is regenerated by air [3]. Thus, the direct contact between fuel and air is avoided and pure CO2 can be obtained from the fuel reactor.

Having highly reactive oxygen carriers, commercially available in large quantities, low-priced, and environmentally acceptable is one key challenge for the CLC scale-up. The most reactive oxygen carriers, mainly based on NiO and CuO, are relatively expensive and may result in emissions of metals of concern to the environment and human health. On the contrary, the relatively slow fuel reaction rate for less expensive and environmentally friendly oxygen carriers, such as Fe-based ores, is the limiting step which reduces the fuel combustion efficiency resulting in overly large combustors.

In this regard, CaMnO3-δ materials with perovskite structure exhibit interesting characteristics to be utilized as oxygen carriers in CLC with gaseous fuels. The oxygen stoichiometry, 3-δ, depends on the temperature and oxygen partial pressure [4]. A reductive decomposition of CaMnO3-δ giving O2(g) is observed at low oxygen partial pressures and high temperatures [5]. This mechanism, often named oxygen uncoupling in chemical looping processes [6], can be exploited for fuel combustion. In addition, a greater reduction to CaMnO2 with cubic structure happens when a reducing gas such as CH4, H2 or CO is used [7]. The formation of non-perovskite phases such as Ca2MnO4-δ', CaMn2O4 or CaMnO2 in the reductive decomposition step hinders the complete regeneration of the perovskite structure in air, with the eventual loss of oxygen uncoupling capability [8].

CaMnO3-δ has been frequently doped with metals, e.g. La, Ti, Fe, Cu or Mg, in order to facilitate the regeneration to the perovskite structure [9], [10]. Doping with Mg showed improved performance related to stability, mechanical strength, fluidization behaviour and reactivity compared to others dopants [11]. In this regard, CaMn0.9Mg0.1O2.9‑δ has been successfully used in CLC units burning methane [12], [13]. However, Mg was not combined within the perovskite structure and appeared as a separated MgO phase [14]. Small amounts of CaMn2O4 were found after several redox cycles, which suggested that complete regeneration of the perovskite structure was not achieved. In fact, the perovskite reduction to CaMnO2 with cubic structure caused the loss of the oxygen uncoupling capability, although the reactivity towards CH4, H2 and CO was improved [15].

The addition of titanium also stabilized the perovskite structure [7], [8], [16]. Keeping these results in mind, a CaMnO3-based perovskite doped with Mg and Ti was developed and tested for CLC [17] showing an improved stability and reactivity compared to previous materials [18]. This kind of oxygen carriers was successfully tested in CLC units burning methane [19], [20]. CaMnO3-based perovskites doped with Mg and Ti were first prepared from high purity raw materials, but the use of low cost materials as a manganese source is an interesting option for their upscaling [21], [22], [23]. Therefore, they were replicated using low cost raw materials [24], [25].

Eventually, a material with a chemical formula CaMn0.775Ti0.125 Mg0.1O2.9-δ was selected in an EU project to be produced at ton-scale by spray drying method using low cost materials for Ca, Mn and Ti sources [26]. The calcination temperature of this material was between 1598 and 1623 K for 4 h, and it had to be carefully monitored in order to obtain particles with optimized properties in terms of crushing strength and reactivity [27]. This material may transfer oxygen to the fuel through the oxygen uncoupling mechanism, i.e. combining reactions (R1), (R2) [28].2/γ-δCaMn0.775Ti0.125Mg0.1O2.9-δ2/γ-δCaMn0.775Ti0.125Mg0.1O2.9-γ+O2CH4+2O2CO2+2H2O

In addition, a high reduction degree can be found by reaction of lattice oxygen with a gaseous fuel, e.g. CH4, H2 or CO; see reactions R3-R6.4/(γ-δ)CaMn0.775Ti0.125Mg0.1O2.9-d+CH44/(γ-δ)CaMn0.775Ti0.125Mg0.1O2.9-g+CO2+2H2O1/(γ-δ)CaMn0.775Ti0.125Mg0.1O2.9-d+CH41/(γ-δ)CaMn0.775Ti0.125Mg0.1O2.9-g+CO+2H21/(γ-δ)CaMn0.775Ti0.125Mg0.1O2.9-d+H21/(γ-δ)CaMn0.775Ti0.125Mg0.1O2.9-g+H2O1/(γ-δ)CaMn0.775Ti0.125Mg0.1O2.9-d+CO1/(γ-δ)CaMn0.775Ti0.125Mg0.1O2.9-g+CO2

The CaMn0.775Ti0.125 Mg0.1O2.9-δ material is the first and, in our best knowledge, the only oxygen carrier prepared at ton-scale, which exhibits promising features for the CLC scale-up [29] along with other materials based on Cu [30], Fe [31], [32] or Ni [33], [34], [35]. Its excellent performance has been demonstrated in different CLC units, namely 10 kWth [36], 120 kWth [37] and 1 MWth [38]. Complete combustion has been achieved at high oxygen carrier-to-fuel ratios (ϕ > 20). Under these conditions, the transference of oxygen by means of oxygen uncoupling was promoted; see reaction R(1) [28]. The oxygen carrier-to-fuel ratio, ϕ, is defined as the oxygen flow in the circulating oxygen carrier, assuming complete oxidation, divided by the stoichiometric oxygen demanded by the fuel; see Eq. (1) for CH4 combustion. In addition, the ϕ parameter is related to the variation of the oxygen carrier conversion between the fuel and air reactors, ΔXOC [39].ϕ=ṁOCROC4MCH4FCH4=ηcombΔXOC

The CaMn0.775Ti0.125 Mg0.1O2.9-δ oxygen carrier was usually tested at high ϕ values, which means that ΔXOC was low. In such condition, the oxygen carrier reactivity was found to be stable with the redox cycles [13]. However, the reactivity of lattice oxygen with combustible gases, e.g. CH4, CO and H2, might increase as the variation of the oxygen carrier conversion was increased, which was named as an activation process [18], [31]. In practice, this means to operate a CLC unit at low ϕ values. However, the rate for the oxygen uncoupling reaction could be negatively affected.

The objective of this work was to assess the behaviour of the CaMn0.775Ti0.125 Mg0.1O2.9-δ material prepared at ton-scale to burn methane in a CLC process. The variation of oxygen carrier reactivity with the number of redox cycles was analysed for reactions with lattice oxygen and oxygen uncoupling. The reaction kinetics was determined to be included in a validated mathematical model of the fuel reactor. Simulation results were used to discuss the potential of CaMn0.775Ti0.125 Mg0.1O2.9-δ as oxygen carrier under different operating conditions.

Section snippets

Material

Natural ores were used as raw materials for the production of oxygen carrier particles by Euro Support Advanced Materials. Particles with good mechanical properties were produced by the spray drying method followed by a calcination process at 1608 K during 4 h [27]. The general formula of the perovskite is CaMn0.775Ti0.125 Mg0.1O2.9-δ. A particle size ranging from 125 to 200 μm was selected to determine the reaction kinetics. This material, prepared at industrial scale, is here named as

Results

Kinetics of the oxygen carrier for oxygen uncoupling in N2, reducing reactions with CH4, H2, CO, and oxidation with O2, as well as the oxygen transport capacity of the CaMnTi/Mg/Ind material, were determined from TGA tests. Besides, the evolution of reactivity and oxygen transport capacity with the number of redox cycles was analysed.

Reactivity

The determination of the rate index [49] or the minimum solids inventory [39] is often used to define the oxygen carrier reactivity and compare it with other materials. In both cases, the initial reaction rate at Xi = 0 with chemical reaction control is considered. The rate index is defined by Eq. (22) and calculated with a concentration of 15 vol% for the combustible gas and 10 vol% for O2. The minimum solids inventory is calculated using Eq. (23), and gives a more precise information because

Conclusions

The evolution with the redox cycles of the oxygen transport capacity, the redox reactivity and the oxygen uncoupling behaviour of CaMn0.775Ti0.125 Mg0.1O2.9-δ perovskite prepared at industrial scale was analysed in this work. The material increased its reactivity with the redox cycles for the reduction with CH4, H2 and CO, in a so-called activation process. The total oxygen transport capacity of the original material was ROC,t = 8.4–9.0 wt%, depending on the reaction temperature, but decreased

Acknowledgements

This work was partially supported by the project ENE2016-77982-R (AEI/FEDER, UE), ENE2017-89473-R (AEI/FEDER, UE) and the SUCCESS Project, funded by the European Commission under the Seventh Framework Programme (Grant agreement No. 608571).

References (51)

  • M. Jacobs et al.

    Synthesis and upscaling of perovskite Mn-based oxygen carrier by industrial spray drying route

    Int. J. Greenhouse Gas Control

    (2018)
  • A. Abad et al.

    Modelling Chemical-Looping assisted by Oxygen Uncoupling (CLaOU): Assessment of natural gas combustion with calcium manganite as oxygen carrier

    Proc. Combust. Inst.

    (2019)
  • J. Adánez et al.

    Chemical-looping combustion: Status and research needs

    Proc. Combust. Inst.

    (2019)
  • A. Cabello et al.

    Long-lasting Cu-based oxygen carrier material for industrial scale in Chemical Looping Combustion

    Int. J. Greenhouse Gas Control

    (2016)
  • L.F. de Diego et al.

    Performance of Cu- and Fe-based oxygen carriers in a 500 Wth CLC unit for sour gas combustion with high H2S content

    Int. J. Greenhouse Gas Control

    (2014)
  • C. Linderholm et al.

    Long-term integrity testing of spray-dried particles in a 10-kW chemical-looping combustor using natural gas as fuel

    Fuel

    (2009)
  • J.I. Baek et al.

    The effects of using structurally less-stable raw materials for the support of a spray-dried oxygen carrier with high NiO content

    Fuel

    (2012)
  • R.F. Pachler et al.

    Fate of sulfur in chemical looping combustion of gaseous fuels using a Perovskite oxygen carrier

    Fuel

    (2019)
  • P. Ohlemüller et al.

    Investigation of chemical looping combustion of natural gas at 1 MWth scale

    Proc. Combust. Inst.

    (2019)
  • A. Abad et al.

    Mapping of the range of operational conditions for Cu-, Fe-, and Ni-based oxygen carriers in chemical-looping combustion

    Chem. Eng. Sci.

    (2007)
  • A. Abad et al.

    Reaction kinetics of ilmenite for chemical-looping combustion

    Chem. Eng. Sci.

    (2011)
  • J. Szekely et al.

    A structural model for gas-solid reactions with a moving boundary

    Chem. Eng. Sci.

    (1970)
  • F. García-Labiano et al.

    Temperature variations in the oxygen carrier particles during their reduction and oxidation in a chemical-looping combustion system

    Chem. Eng. Sci.

    (2005)
  • A. Cabello et al.

    Kinetic determination of a highly reactive impregnated Fe2O3/Al2O3 oxygen carrier for use in gas-fueled Chemical Looping Combustion

    Chem. Eng. J.

    (2014)
  • C. Dueso et al.

    Reduction and oxidation kinetics of nickel-based oxygen-carriers for chemical-looping combustion and chemical-looping reforming

    Chem. Eng. J.

    (2012)
  • Cited by (26)

    View all citing articles on Scopus
    View full text