CuO-Based Oxygen-Carrier Particles for Chemical ... - ACS Publications

Article pubs.acs.org/IECR

CuO-Based Oxygen-Carrier Particles for Chemical-Looping with Oxygen Uncoupling − Experiments in Batch Reactor and in Continuous Operation Magnus Rydén,*,† Dazheng Jing,‡ Malin Kal̈ lén,† Henrik Leion,‡ Anders Lyngfelt,† and Tobias Mattisson† †

Department of Energy and Environment, and ‡Department of Environmental Inorganic Chemistry, Chalmers University of Technology, Göteborg 412 58, Sweden ABSTRACT: Chemical-looping with oxygen uncoupling (CLOU) is an innovative method to oxidize fuels with inherent CO2 sequestration, which utilizes a solid oxygen-carrier material to provide O2 for fuel combustion. In this study, a range of CuObased oxygen-carrier particles have been manufactured and examined. Out of 24 samples prepared, 10 were examined in a batch fluidized-bed reactor, of which three were selected for further examination by continuous operation in a small circulating fluidized-bed reactor system. Composite particles consisting of CuO as active phase and support material such as ZrO2, YSZ, CeO2, and MgAl2O4 were capable of providing full conversion of CH4 at 900 and 925 °C, and were also found to release gas phase O2 into inert atmosphere when fluidized with N2. Particles using semiactive support such as Fe2O3, Mn2O3, and Al2O3 formed combined spinel phases with CuO. Such materials were still capable of releasing gas phase O2 but at different concentrations as compared to particles with inert support. Materials with semiactive support had less good reactivity with CH4. No formation of unexpected phases could be detected by X-ray diffractometry, and all chemical reactions were completely reversible. The three materials that were examined in continuous operation were readily capable of providing more or less full conversion of natural gas under the chosen conditions. However, they also suffered from quick attrition and turned into a flourlike substance after a few hours of continuous operation with fuel. Crushing strength analysis showed that particles used in continuous operation were physically much weaker than fresh. In total, 23 h of continuous operation with fuel addition was recorded.

1. INTRODUCTION This Article describes a comprehensive experimental study examining the possibility to use CuO as oxygen-carrier for chemical-looping combustion (CLC) in general, or more specifically for chemical-looping with oxygen uncoupling (CLOU). The ultimate goal of developing these technologies is to provide efficient technology to capture the greenhouse gas CO2 during oxidation of different kinds of fuels, such as natural gas and coal. A thorough description of how so-called carbon capture and storage (CCS) could contribute to reduced emissions of CO2 to the atmosphere can be found in IPCC’s special report about the topic.1 A thorough description of the status of development of carbon capture and storage can be found in the recent article by Fennel et al.2

CnHm(g) + (2n + 1/2m)MeOx (s) → nCO2 (g) + (1/2m)H 2O(g) + (2n + 1/2m) MeOx − 1(s)

MeOx − 1(s) + 1/2O2 (g) → MeOx (s)

(2)

CnHm(g) + (n + 1/4m)O2 (g) → nCO2 (g) + (1/2m)H 2O(g)

(3)

The heat of reaction of the reactor system as a whole is the same as in ordinary combustion. The most commonly proposed design utilizes fluidized beds in a similar way as in a conventional circulating fluidized-bed boiler (CFB), with the difference that the inert bed material used in such facilities would be replaced with an active oxygen-carrier material. The operating temperature of each reactor is expected to be in the range of 800−1050 °C. A schematic description of chemicallooping combustion can be found in Figure 1. The basic ideas behind chemical-looping combustion can be traced back to the middle of the 20th century in work by Lewis and Gilliland et al.3,4 The purposes of these early process proposals were production of synthesis gas (CO + H2) or high-

2. BACKGROUND 2.1. Chemical-Looping Combustion. Chemical-looping combustion (CLC) is an innovative method to oxidize fuels in which the fuel is oxidized in two steps using two separate reactor vessels, one air reactor (AR) and one fuel reactor (FR). A solid oxygen-carrier, usually a transition metal oxide (MeOx), performs the task of transporting oxygen to the fuel and circulates continuously between the two reactors. In the fuel reactor, it is reduced by the fuel, which in turn is oxidized to CO2 and H2O according to reaction 1. In the air reactor, it is oxidized to its initial state with O2 from the air according to reaction 2. Combining reactions 1 and 2 yields reaction 3, which is complete combustion of the fuel with O2. © 2014 American Chemical Society

(1)

Received: Revised: Accepted: Published: 6255

November 26, 2013 February 3, 2014 March 20, 2014 March 20, 2014 dx.doi.org/10.1021/ie4039983 | Ind. Eng. Chem. Res. 2014, 53, 6255−6267

Industrial & Engineering Chemistry Research

Article

sufficient to obtain almost pure CO2, because separation of liquid water and gaseous CO2 is trivial. Thus, carbon capture via chemical-looping combustion makes it possible to avoid the high costs and high energy consumption for gas separation, which is associated with technologies such as amine absorption and oxyfuel combustion. Comprehensive reviews of chemicallooping combustion have recently been provided by Adanez et al.,14 Fan et al.,15 and Lyngfelt et al.16 2.2. CuO as Oxygen-Carrier Material. Using CuO as oxygen carrier was proposed already by Lewis and Gilliland in their patent from 1954.4 In chemical-looping applications, CuO can be reduced in two steps. The first step involves reduction to Cu2O: 2CuO(s) → Cu 2O(s) + 1/2O2 (g)(s) Figure 1. Schematic description of chemical-looping combustion.

The amount of O2 that can be released via reaction 5 corresponds to 10 wt % of fully oxidized CuO. The O2 concentration provided is a function of the equilibrium partial pressure of O2 over CuO/Cu2O, which is a function of temperature; see Table 1.

purity CO2. Other early work includes a study by Richter and Knoche,5,6 who suggested a fuel oxidation reaction scheme involving two intermediate reactions with a metal oxide as oxygen carrier to reduce irreversibility as compared to ordinary combustion. Ishida et al.7 coined the term chemical-looping combustion in 1987 while putting forward similar ideas. Chemical-looping combustion as a method to separate CO2 during fuel oxidation was proposed a few years later by Ishida and Jin,8 soon to be followed by a basic design for a circulating fluidized-bed reactor by Lyngfelt et al.9 in 2001. The process was demonstrated in continuous operation by Lyngfelt and Thunman in 2003,10 and since then a number of oxygen carriers have been used in continuous operation in pilot plants worldwide; see Lyngfelt et al.11 In reaction 1, it was assumed that the fuel is in the gas phase and reacts with the oxygen carrier in a gas−solid reaction. However, with some oxygen-carrier materials, gas-phase O2 can be released directly in the fuel reactor according to reaction 4. MeOx (s) ↔ MeO1 − x (s) + 1/2O2 (g)

(5)

Table 1. Equilibrium Concentration of O2 at p = 1 atm for Reaction 5 as a Function of Temperature, As Calculated with FactSage 6.1 and the FToxide Database

(4)

temp (°C)

equilibrium concn (%)

800 825 850 875 900 925 950 975 1000

0.1 0.2 0.4 0.8 1.4 2.4 4.2 7.0 11.4

The implication of the data in Table 1 is that the fuel reactor should operate at as high temperature as practically possible, if a large driving force for O2 uncoupling is desired. The maximum temperature of operation of the air reactor is limited though, because reoxidation can only take place if the O2 partial pressure in the outlet of the air reactor is higher than the equilibrium concentration. Up to this point, most experimental work with CuO-based oxygen carriers has been conducted at 800−925 °C. Aside from reduction of CuO to Cu2O, further reduction of Cu2O to metallic Cu is a possibility, but only in the presence of reducing fuel gases such as CH4, CO, or H2. It is likely that formation of metallic Cu would like to be avoided because metallic copper has low melting temperature (1085 °C) and thus starts to sinter and defluidize already at moderate temperature levels, as has been reported by several research groups.17−19 CuO-based oxygen carriers have received a great deal of interest in the past few years, due to their O2 uncoupling properties, high reactivity with solid and gaseous fuels, good oxygen transport capacity by weight, and lack of thermodynamic limitation for complete fuel oxidation. Much work has focused on using CuO with cheap and available alumina (Al2O3) as support material.17,20−26 While this approach certainly is feasible, recent work has demonstrated that CuO and Al2O3 have a strong tendency to interact with each other during operation (or during calcination at high temperature)

O2 will be released until thermodynamic equilibrium for reaction 4 is obtained. If there is a fuel present in the fuel reactor, it will react directly with released O2 according to reaction 3, which in turn will facilitate further O2 release until all available fuel is consumed, or no more O2 is available in this solid. The reduced oxygen carrier can then be recirculated to the air reactor where it is reoxidized according to reaction 2. This second reaction scheme described above is referred to as chemical-looping with oxygen uncoupling (CLOU); see Mattisson et al.12 The sum of reactions is identical to those for chemical-looping combustion, but the mechanism by which the fuel is oxidized is different. In ordinary chemical-looping combustion, the oxidation of fuel takes place mainly via gas− solids reactions. So if the fuel is in solid phase such as coal, it has to be gasified first to be able to react with the solid oxygen carrier. By contrast, in chemical-looping with oxygen uncoupling, the oxidation of the fuel can proceed by direct combustion. Leion et al.13 have shown that oxidation of char can proceed much faster using this reaction scheme, as compared to a conventional chemical-looping combustion process, which relies on char gasification. As compared to conventional combustion, chemical-looping combustion and chemical-looping with oxygen uncoupling would provide some intriguing benefits. Most importantly, fuel is never mixed with N2 from the combustion air. Cooling and condensation of the steam produced in the fuel reactor is 6256

dx.doi.org/10.1021/ie4039983 | Ind. Eng. Chem. Res. 2014, 53, 6255−6267


Article

forming combined phases such as copper(II) aluminate (CuAl2O4) and copper(I) aluminate (CuAlO2).21−23,27 This may not necessarily be a problem for chemical-looping combustion of gaseous fuels because copper aluminates by themselves are active oxygen-carrier materials. However, if free CuO is replaced by CuAl2O4 as oxidized phase, the oxygen carrier loses its ability to release gas-phase O2 according to reaction 5. This may be a considerable drawback for oxidation of solid fuels where O2 release is crucial, and may also require improved mixing of gases and solids in the fuel reactor to achieve complete oxidation of gaseous fuels. Alternative support materials that have been proposed and examined include zirconia (ZrO2),14,15,28,29,32,33 magnesium aluminate (MgAl2O4),21,22,28,30,31 titania (TiO2),28,29 and silica (SiO2).28,29 These are materials that should be inert with respect to interactions with CuO at conditions suitable for chemical-looping applications. A different approach would be to use a support material, which forms combined oxides with CuO that, unlike copper aluminates, are capable of releasing gasphase O2 at relevant conditions. One such option would be CuO−Mn2O3 combined oxides,34−36 which have been shown to have such properties. 2.3. The Aim of This Study. The aim of this study is to contribute to the development of CuO-based oxygen carriers for chemical-looping with O2 uncoupling. Because CuO in combination with Al2O3 as support material seems destined to result in formation of CuAl2O4 and loss of O2 uncoupling properties, the focus has been to study CuO in combination with other inert or semiactive support materials.

Table 2. Summary of Oxygen-Carrier Materials Examined in This Studya designation

raw materials (wt %)

calcination temp (°C)

Cu40Z

40 CuO, 60 ZrO2 950, 1100b

Cu40ZLa5 Cu40(YSZ)

40 CuO, 55 ZrO2, 5 La2O3 40 CuO, 60 YSZ

Cu40Ce

40 CuO, 60 CeO2 950, 1100b

Cu40CeLa5

40 CuO, 55 CeO2, 5 La2O3 40 CuO, 60 MgAl2O4e 40 CuO, 60 MgAl2O4e 40 CuO, 60 MgAl2O4, 5 La2O3 60 CuO, 40 Fe2O3 40 CuO, 41 Fe2O3, 19 Mn3O4 40 CuO, 48 Fe2O3, 12 MgO 36 CuO, 40 Fe2O3, 24 Al2O3

Cu40(MgAl) Cu40(MgAlx) Cu40(MgAl) La5 Cu60F Cu40MF41 Cu40MgF48 Cu36FAl24

950, 1100b 1000,c 1100

950,c 1100b 950,c 1100 950,c 1100

bulk density (kg/m3)

crushing strength (N)

1950, 2270 1450, 1680 1480, 1700 2170, 2820 2130, 2780 1210, 1270 990, 1420

3.4, 4.5 1.4, 1.3 0.4, 0.6 1.0, 1.5 0.3, 1.3 0.4, 0.9 0.6, 1.4

950,c 1100

1200, 1180

0.4, 0.5

950, 1100d

2160d

1d

950,b 1100d

2100d

1.3d

950, 1100d

1810d

1d

950,c 1100

1090, 1870

0.4, 1.2

a

Cu = CuO, Z = ZrO2, La = La2O3, YSZ = Y2O3-stabilized ZrO2, Ce = CeO2, MgAl = MgAl2O4, F = Fe2O3, M = Mn3O4, Mg = MgO, with numbers indicating weight percentage during synthesis. bDefluidized during batch experiments. cVery soft particles excluded from testing. d Cake formation during calcination. eCu40(MgAl) from two MgAl2O4 qualities; finer material with higher purity is marked “x”.

3. EXPERIMENTAL SECTION 3.1. Manufacturing of Oxygen-Carrier Materials. All oxygen-carrier particles examined in this study were manufactured by VITO in Belgium by spray drying. The general procedure was as follows. Powder mixtures of the raw materials were dispersed in deionized water containing organic additives, organic binder, and dispersants. The water-based suspension was continuously stirred with a propeller blade mixer while being pumped to a 2-fluid nozzle, positioned in the lower cone part of the spray-drier. Particles obtained were sieved, and the fraction within the desired size range was separated from the rest of the spray-dried product. Sieved particles were then calcined in air at 950−1100 °C for 4 h. After calcination, the particles were sieved once more so that all particles used for experimental evaluation would be of well-defined size. A summary of manufactured materials and their basic physical properties can be found in Table 2. The bulk density was measured for particles in the size range of 125−180 μm using a graduated cylinder and a balance. The reported crushing strength is the force required to fracture a single particle, as measured with a digital force gauge on particles in the size range of 180−250 μm. The reported value is the average value for 30 particles. The materials can be divided into several categories, CuO using inert support materials based on ZrO2, CeO2, or MgAl2O4 and combined oxides of CuO and Fe2O3 with addition of other elements. Five wt % La2O3 was added to some samples. This was done because it has been suggested that La2O3 is beneficial for the physiochemical properties and thermal stability of copper-based oxygen carriers; see Cao et al.37 Several of the materials manufactured were found to be unsuitable for experiments in the batch fluidized-bed reactor.

For these materials, it was not possible to extract meaningful data. These instances have been pointed out in Table 2. Experienced problems included: • Defluidization during batch experiments. This phenomenon appears to have been connected with particle density, as samples with high bulk density were much more likely to defluidize during investigation. Similar correlation has previously been reported for NiO-based particles in a study by Jerndal et al.38 • Poor particle strength. Many samples, in particular those supported on MgAl2O4 calcined at 950 °C, had very low crushing strength (below 0.6 N). These were not examined if the sample calcined at higher temperature appeared more promising in this respect. Because of the poor strength of the first batch Cu40(MgAl), a second batch Cu40(MgAl)x was made, using MgAl2O4 of finer grain size and higher purity. The difference in particle strength as compared to the first batch was small though. In general, we believe that a crushing strength of at least 1 N or preferably 2 N would be required for practical applications; see Rydén et al.39 • Agglomeration or melting during calcination. All combined CuO−Fe2O3 materials except Cu36FAl24 formed a hard cake rather than particles during calcination at 1100 °C. Despite these difficulties, batch experiments were successfully conducted for all material compositions except Cu40CeLa5 and Cu40MF41. 6257



Article

Figure 2. Schematic description of the experimental setup used for batch experiments.

Table 3. Examination Scheme for Batch Experimentsa

3.2. Experimental Setup for Batch Experiments. Batch experiments were carried out in an 820 mm long quartz reactor with an inner diameter of 22 mm. A porous quartz plate, on which the sample of oxygen carrier was applied, was located 370 mm above the bottom. During operation, the sample was fluidized by adding gases to the bottom of the reactor, and the porous plate acted as gas distributor. The reactor temperature was measured with thermocouples located 5 mm below and 25 mm above the porous plate. The pressure drop over the particle bed was measured with a pressure transducer operating with a frequency of 20 Hz. The pressure drop over a quartz plate is approximately constant for constant flows, so by measuring fluctuations in the pressure drop it was possible to determine if the particle bed was fluidized or not. At the top of the reactor was a plug of quartz wool, in which elutriated solid fines were captured. The product gas was then subject to cooling and condensation of water. After this step, the composition of the dry gas was measured with a gas analyzer. CO2, CO, and CH4 were measured with infrared sensors, while O2 was measured with a paramagnetic sensor. A schematic description of the experimental setup is shown in Figure 2. The experimental procedure was as follows. Fifteen grams of particles in the size range 125−180 μm was placed on the porous plate. The reactor was then assembled and placed inside an electrically heated furnace. During heating to 900 °C, the sample was fluidized with an oxidizing gas mixture to ensure full oxidation of the carrier prior to actual experiments. Experiments were then conducted by switching between different fluidization gases, as is described in Table 3. The following gases and gas flows were used: • Oxidation: 900 mL/min with a mixture of 5% O2 and 95% N2 was used to simulate the expected conditions at the top of the air reactor in a future chemical-looping combustion facility. • Inert: 600 mL/min with 100% N2 was used to examine O2 release in inert atmosphere via reaction 5, and also to flush the reactor of reactive gases for 60 s between each fuel and oxidizing period.

cycle

temp (°C)

reducing gas

reduction time (s)

1−2 3−5 6−7 8−10 11−12 13b 14b 15b 16b 17b

900 900 925 925 925 925 925 925 925 925

inert (N2) fuel (CH4) inert (N2) fuel (CH4) inert (N2) fuel (CH4) fuel (CH4) fuel (CH4) fuel (CH4) fuel (CH4)

360 20 360 20 360 30 40 50 60 70

a

Each cycle constitutes reduction for a specified time period, either passively with N2 or actively with CH4, followed by oxidation until the sample is fully reoxidized. bNot done for Cu60F and Cu40F48Mg.

• Reduction: 450 mL/min with 100% CH4 was used to examine reactivity with natural gas and examine behavior during deep reduction of each sample. The gas flows in the reactor inlet correspond to 4−20 times the minimum fluidization velocity of the particles, depending on fluidization gas used and particle density. In this Article, the reactivity of each oxygen carrier toward CH4 is quantified in terms of gas yield, γCO2, defined as the fraction of CO2 in dry flue gas divided by the sum of the fractions of carbon containing gases; see eq 6: γCO2 =

yCO

2

yCO + yCH + yCO 2

4

(6)

in which yi is the dry gas concentration (vol %) of gas component i. The mass-based conversion of the oxygen carrier, ω, is defined as the mass of the sample, m, divided by the mass of the fully oxidized sample, mox, as is defined in eq 7: ω=

m mox

(7)

The mass-based conversion of the oxygen carrier for a specific time period can be calculated by integration over a time interval, as is described in eq 8, in which MO is the molar mass of oxygen atom and ṅout is the molar flow out of the reactor. 6258



Article

these particles falls into the downcomer, entering a J-type loopseal. From the loop-seal, particles overflow into the fuel reactor via the return orifice. The fuel reactor is a bubbling bed. At the bottom particles return to the air reactor through a U-type loop-seal, and thus a continuous circulation of oxygen-carrier particles is obtained. The two loop-seals are fluidized with small amounts of inert gas such as argon, which is added via thin pipes perforated by holes. To make it possible to reach and sustain a suitable temperature, the whole reactor is placed inside an electrically heated furnace. The temperature in each reactor section is measured with thermocouples located inside the particle beds, a few centimeters above each bottom plate. The reactor is operated at approximately atmospheric pressure. However, a water-seal is located downstream the fuel reactor, which makes it possible to apply an overpressure of ∼250 Pa to the fuel reactor, to inhibit leakage of air into the fuel reactor. Along the reactor sections, there are 13 separate pressure measuring taps. By measuring differential pressures between these spots, it is possible to estimate where particles are located in the system, and to detect abnormalities in the fluidization. For gas analysis, roughly 0.50 Ln/min gas is extracted downstream of the air reactor and the fuel reactor, respectively. The two extracted flows pass through separate particle filters, coolers, and water traps. Hence, all measurements are made on dry gas. CO2, CO, and CH4 are measured using infrared analyzers, while O2 was measured with paramagnetic sensors. The gas from the fuel reactor was also examined with a gas chromatograph, which measured H2, N2, and light hydrocarbons up to C3, in addition to the gases mentioned above. The gas chromatograph provides one measuring point every 3 min. Excess gas that was not needed for analysis passed through a textile filter to catch elutriated fines and particles, prior to release in a chimney.

Equation 8 simply describes a species balance over the reactor, and is valid only when CH4 is used as fuel. ωi = ωi − 1 −

∫t

t1 0

nout ̇ MO (4yCO + 3yCO + 2yO − yH ) dt 2 2 2 mox (8)

3.3. Experimental Setup for Continuous Operation. The continuous experiments were carried out in a small-scale laboratory reactor made of 253MA steel, which is a temperature and deformation resistant stainless steel with the composition 67.9% Fe, 21% Cr, 11% Ni, and 0.1% C. The same reactor has previously been used for different kinds of chemical-looping related experiments involving liquid fuels33,40 and hydrogen generation.41 A schematic description of the reactor is shown in Figure 3.

4. RESULTS 4.1. Experiments in Batch Reactor. As was pointed out in Table 2, there were some practical problems with a number of the samples, prohibiting complete testing in accordance with Table 3. Yet for the rest of the particles for which experiments could be successfully carried out, the results were quite intriguing. 4.1.1. O2 Release during Fluidization with N2. All samples were found to release O2 during fluidization with N2, as can be seen in Figure 4. In Figure 4, it can be seen that all samples with CuO on inert support material (ZrO2, YSZ, CeO2, MgAl2O4) showed very similar behavior. When the oxidizing gas was switched to N2, the O2 concentration dropped rapidly and stabilized at about 2% for the whole period at 925 °C, and to 1.0% at 900 °C. These numbers are slightly below the equilibrium values reported in Table 1. One interpretation could be that the temperature in the reaction zone may be slightly lower as compared to the value measured with the thermocouple used to control the temperature of the oven, which is located 25 mm above the porous plate. This assumption seems reasonable because reaction 5 is strongly endothermic. The total amount of O2 released during 360 s fluidization with N2 in Figures 4 and 5 amounts to approximately 0.6 wt % at 925 °C, and 0.3 wt % at 900 °C. This is far from the amount of O2 that theoretically could be released via reaction 5, which is 10 wt %. It is easily realized that the reason is that the duration of the N2

Figure 3. Schematic description of the two-compartment continuous reactor system.

The reactor is 300 mm high. The fuel reactor measures 25 mm × 25 mm. The base of the air reactor is 25 mm × 42 mm, while the upper narrow part is 25 mm × 25 mm. Fuel and air enter the system through separate wind boxes, located in the bottom of the reactor. Porous quartz plates located between the wind boxes and the reactor sections act as gas distributors. For the experiments presented in this Article, 300−400 g of particles in the size range 90−212 μm was added to the reactor. This corresponds to a bed height in the air and fuel reactor of roughly 10 cm, taking into consideration that a considerable share of the particles was located in the downcomer during operation. In the air reactor, the gas velocity is sufficiently high for oxygen-carrier particles to be thrown upward. Above the reactor (not shown in Figure 3), there is a particle separation box in which the cross-section area is increased and gas velocity reduced so that particles fall back into the reactor. A fraction of 6259



Article

Figure 4. O2 concentration profile of the second inert cycle for examined oxygen-carrier materials at 900 and 925 °C.

period is insufficient to release all available O2. At 900 °C, an N2 period of about 30 h would have been required. The samples intended to form combined oxides did also release O2, but at slightly different levels as compared to unmodified CuO. CuO−Fe2O3 (Cu60F) released significantly higher concentrations of O2 as compared to other materials examined. The likely reaction pathway for such materials is reduction of CuO and CuxFe2−xO4 spinel solid solution to delafossite CuFeO2; see reaction 9: CuO + CuFe2O4 → 2CuFeO2 + 1/2O2 (g)

Particles with iron and magnesium (Cu40MgF48) released O2 at the same level as particles with inert support material. The likely reason is that iron and magnesium formed magnesium ferrite MgFe2O4 during calcination; see section 4.3 for details. This apparently left free CuO in the sample, which could react according to reaction 5 in similar fashion as the samples with inert support. Particles with iron and alumina (Cu36FAl24) released much less O2 as compared to the other samples. This was not surprising considering that in those particles all CuO was present in a combined spinel phase Cu0.95Fe1.05AlO4 rather than as free CuO; see section 4.3 for details. 4.1.2. Reactivity with CH4. All examined samples showed high reactivity toward CH4. Figure 5 shows the dry flue gas concentration profile during oxidation of CH4 for Cu40Z at 925 °C. The time delay in the reactor system is approximately 20 s, depending on gas flow. At t = 20 s, the oxidizing gas was switched off and inert N2 was injected for 60 s. As can be seen, the O2 concentration in the reactor drops to values similar to what was reported in Figure 4, that is, 2%. At t = 80 s, the N2

(9)

While precise thermodynamic data for this reaction are unavailable, it is clear that the oxygen potential in the relevant temperature interval is higher for reaction 9 than for reaction 5; see Jacob et al.42 According to a correlation proposed by Jacob et al.,42 the equilibrium O2 partial pressure of reaction 9 should be in the order of 0.05 atm at 925 °C. So again the experimental results presented in this Article suggest slightly lower temperature in the reaction zone as compared to what was measured with the thermocouple. 6260



Article

flow was turned off, and CH4 was added for 20 s. Added CH4 was converted to CO2 via reactions 3 and 4, giving rise to a sharp CO2 peak. At the same time, the O2 concentration measured on dry basis increases by roughly a factor of 3. This is an artifact caused by the steam generated in the reaction between oxygen carrier and CH4, providing two H2O for each CO2 in the outlet gas. Because O2 release is dictated by equilibrium partial pressure of O2 in the reaction zone, that is, in the presence of steam, this means that O2 measured on dry basis appears 3 times as high as compared to what could be expected. Another factor influencing the results is that the reaction between CH4 and CuO is exothermic. Hence, the bed temperature increases about 25 K during reduction with CH4. At t = 100 s, flushing with N2 is initiated. As can be seen, the concentrations of detected gaseous species decrease rapidly toward zero. The exception is O2, which returns to a somewhat higher value as compared to before CH4 was added to the reactor. The slight increase is because of a lingering temperature increase, due to the exothermic nature of the preceding reactions. In Figure 5, it is apparent from the lack of CO during fuel oxidation that the conversion of CH4 to CO2 is more or less

Figure 5. Dry gas concentration profile for reduction of Cu40Z with CH4 at 925 °C.

Figure 6. Gas yield (γCO2) as a function of the degree of reduction of the oxygen carrier (ω) for examined materials at 900 and 925 °C. 6261



Article

Table 4. Summary of Batch Experimentsa

a

designation


O2 900 °C (%)

O2 925 °C (%)

γ 900 °C

γ 925 °C

Cu40Z Cu40ZLa5 Cu40(YSZ) Cu40Ce Cu40(MgAl) Cu40(MgAlx) Cu40(MgAl)La5 Cu60F Cu40MgF48 Cu36FAl24

950 950 1100 950 1100 1100 1100 950 950 1100

1.0 1.0 1.2 1.0 1.1 1.0 1.1 1.7 1.0 0

1.8 2.0 2.1 1.7 1.7 1.9 2.0 2.7 1.9 0.3

1.00 1.00 0.98 1.00 0.98 1.00 0.96 0.90 0.80 0.67

1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.90 0.86 0.83

Reported O2 release during O2 uncoupling experiments and γCO2 for experiments with CH4 are average values for the period ω 1.00→0.98.

complete. Such conditions were achieved for 7 out of 10 examined materials at 925 °C, and for 5 out of 10 materials at 900 °C. Figure 6 shows the gas yield (γCO2) as a function of the oxygen-carrier conversion (ω) for examined materials. In Figure 6, it can be seen that the oxygen carriers with inert support materials outperformed those with support materials, which reacted with CuO such as Fe2O3, at least with respect to CH4 conversion. This was perhaps to be expected because CuO is an excellent oxygen carrier on its own. Somewhat surprisingly, Cu40MgF48 performed worse than other materials with inert support materials, even though it should contain as much free CuO as the samples with inert support. The results from the batch experiments are summarized in Table 4. 4.1.3. Effect of Prolonged Reduction. Most samples were examined also in longer reduction than 20 s (cycles 13−17 in Table 3). This was done to determine the effect of reducing the particles beyond Cu2O, that is, down to metallic Cu. As has been explained above, metallic Cu has a low melting point, so formation of metallic Cu could be assumed to increase the propensity to sintering and defluidization. For 15 g of a sample containing 40 wt % CuO being reduced by 450 mL/min CH4, complete reduction of Cu to Cu2O takes 30 s. After that, Cu2O will start to become reduced to metallic Cu. All materials worked well during the 30 s reduction. Beyond that point, defluidization did occur for most materials. Only Cu40(MgAl) and Cu40(MgAlx) were unaffected and maintained fluidization throughout the scheme shown in Table 3. The other materials defluidized either during the oxidation period that followed the 40 s reduction (Cu40Ce) or during the oxidation period following the 50 s reduction (all others). 4.2. Continuous Operation. Three materials were selected for further examination in the continuously operating reactor, Cu40Z calcined at 950 °C, Cu40Ce calcined at 950 °C, and C40(YSZ) calcined at 1100 °C. The materials were selected on the basis of their perceived performance during batch experiments, and were produced in larger batches using the same procedure as was outlined in section 3.1. A summary of experiments conducted can be found in Table 5. The results obtained with each material will be discussed below. 4.2.1. CuO Supported on Unmodified ZrO2. Particles supported on ZrO2 (Cu40Z), being highly reactive with high crushing strength, were an obvious choice for further examination. Two experimental campaigns were undertaken, starting with Cu40Z calcined at 950 °C. During heatup and preliminary experiments at 800−850 °C fluidizing all reactor sections with inert gases such as Ar and N2,

Table 5. Summary of Oxygen-Carrier Materials Examined in Continuous Operation designation Cu40Z Cu40Z

a

Cu40Ce Cu40(YSZ) a

raw materials (wt %) 40 CuO, ZrO2 40 CuO, ZrO2 40 CuO, CeO2 40 CuO, YSZ


bed mass (g)

60

950

350

60

a

300

60

950

400

60

1100

330

1030

operation with fuel 2 h CLC natural gas 5 h CLC diluted natural gas 5 h CLC diluted natural gas 11 h CLC natural gas

4 h 950 °C + 4 h 1030 °C.

the particles were found to release gas-phase O2 in predictable fashion, that is, at concentrations close to those in Table 1. After about 2 h of operation at hot condition, the inert gases added to the fuel and air reactor were switched to 0.45 LN/min natural gas and 7.0 LN/min air. Already at the low temperature applied (820 °C), there was close to complete conversion (γCO2 > 99.5%) of CH4 into CO2 and H2O. After 15 min, the fuel flow was reduced to 0.30 LN/min to be able to observe excess O2 from the fuel reactor and ensure that reduction to metallic Cu did not take place. At these conditions, about 0.3% excess O2 in the fuel reactor gas was observed; see Figure 7.

Figure 7. Measured dry gas concentration for continuous operation with CU40Z. Ffuel = 0.30 LN/min, Fair = 7.0 LN/min, T = 820 °C. Balance is argon used to fluidize the slot and downcomer. 6262



Article

Opening the reactor revealed that the particles had eroded and the remaining oxygen carrier resembled flour. Microscope investigation revealed that a large fraction of the solids had been turned to dust; see Figure 9. On the basis of these experiments, it could be concluded that Cu40Z particles had remarkably good reactivity with fuel already at 820 °C, but were subject to severe attrition. Unfortunately, the particles were destroyed after only about 4 h of operation at hot conditions, of which only 2 h involved addition of fuel. Because of this, a second experimental campaign was conducted with the same material, this time using particles that had been recalcined for another 4 h at 1030 °C to increase their hardness further. This batch of particles was then examined at 850−880 °C, this time using natural gas (0.10− 0.20 LN/min) diluted with N2 (0.50 LN/min) as reducing gas, and air (7.0 LN/min) as oxidizing gas. The results were similar to those of the first campaign, but this time the solid circulation ceased after about 8 h of operation at hot conditions, of which 5 h involved addition of fuel. The actual fuel flows were lower in this case though due to dilution with N2. The apparent reason for the loss of circulation again was severe particle attrition. 4.2.2. CuO Supported on CeO2. Particles supported on CeO2 (Cu40Ce) were examined in a fashion similar to that described above. Following 4 h of operation at hot condition examining O2 release, inert gases were switched to natural gas (0.15−0.30 LN/min) and air (7.0 LN/min). Again, the fuel was diluted with N2 (0.30 LN/min), and this time great care was taken to always have sufficiently high fuel conversion so that excess O2 could be observed in the fuel reactor gas. Excess O2 was considered as an indicator to ensure that further reduction of Cu2O to metallic Cu in the fuel reactor could not occur. Fuel conversion γCO2 > 99.5% with excess O2 was achieved at 830− 840 °C, with the somewhat higher temperature required for higher fuel flows. Cu40Ce was much softer and had lower density as compared to Cu40Z. Despite the different physical properties between the two particles, the results were similar. The solids circulation collapsed after 9 h of operation at hot conditions, of which 5 h involved fuel. The reason for collapsed circulation again appears to have been severe attrition, with particles eroding to flour-like substance.

Problematically, the pressure drops at key locations such as fuel reactor and downcomer were found to decrease during operation; see Figure 8. Already after 60 min of operation there

Figure 8. Measured pressure drops in the downcomer (bottom to top on fuel reactor side = 82 mm) and fuel reactor for continuous operation with CU40Z.

were irregularities with respect to measured gas concentrations, after 80 min there were irregularities in the circulation of solids, and after 2 h of operation the circulation of solids between the reactor sections ceased and could not be re-established; see Figures 7 and 8. The irregularities in gas concentrations seen in Figure 7 from 60 min onward are most likely associated with fluidization gas for the loop seals and possibly also air choosing different flow patterns in the system. Sudden peaks and shifts in concentrations at 80 and 110 min are an effect of mild physical stimulation of the reactor (i.e., hitting it with a hammer). The peaks at 105 min are an effect of blowing the system with N2. Of the pressure curves in Figure 8, the downcomer is measured over a particle pillar with a fixed height of 82 mm and thus should be constant. Reduced pressure drop here suggests particle swelling, which also was verified by measuring the bulk density of fresh material (1950 kg/m3) and used (1590 kg/m3).

Figure 9. Fresh particles of Cu40Z-950 (left) as compared to used material (right), which has been subject to continuous operation. The black squares measure 1 mm. 6263



Article

resembling flour; that is, it did not flow or pour satisfactorily. It was readily apparent that solids of this quality could not work in a reactor such as the one used in this study. All used samples were found to have slightly lower bulk density as compared to fresh samples of the same material. It seems that the structural integrity of the particles was weakened, resulting in less rigid solids. The flour-like samples were carefully sieved to be able to examine crushing strength of remaining material in the size range 180−250 um. Unsurprisingly, such particles were found to be much softer as compared to fresh material; see Table 6.

4.2.3. CuO Supported on Y-Stabilized ZrO2. Because Cu40Z was perceived as physically hard but still experienced severe attrition, it was assumed that some mechanism other than mechanical abrasion could be responsible for its poor behavior. One such possibility is the well-established fact that ZrO2 experiences changes in crystal structure at certain temperature levels (such as monoclinic to tetragonal at 1170 °C). These phase transitions are known to induce stress and cracks in ZrO2, because each crystal structure of ZrO2 has different unit cell volume and density. This often leads to material failure during cool down; see for example Chevalier et al.43 The temperatures for calcination and experiments used in this study were well below the expected temperature for this kind of phase transition to take place. However, because oxygen-carrier particles in operation experience highly exothermic redox reactions, it was deemed a possibility that such reactions could be an explanation for the experienced problems anyway. Therefore, further experiments were conducted with CuO supported on fully Y2O3-stabilized ZrO2 (8YSZ from Tosoh), Cu40(YSZ), which should be much less prone to undergo this kind of phase transition. A similar experimental procedure was used as for Cu40Z. First, O2 release was examined for 4 h at temperatures ranging from 680 to 930 °C, by fluidization with inert gases. Next, the reactor was operated for a total of 11 h using natural gas (0.35− 0.50 LN/min) as fuel. The air flow was 6.0−7.0 LN/min, and the temperature was set to 920 °C. At these conditions, there was essentially full conversion of natural gas (γCO2 > 99.9%) with almost 3% excess O2 in the flue gas on dry basis. While these particles lasted longer than the other samples, their ultimate fate was the same. The pressure drops measured at key locations in the reactor such as the downcomer slowly decreased as a function of time, and the solids circulations eventually ceased and could not be restarted. Yet again used particles resembled flour rather than sand. 4.2.4. Summary of Experiments with Continuous Operation. The aim of doing experiments in a continuously operating reactor was to examine a few selected samples at a range of temperatures (700−950 °C) using different flows of fuel and air to examine the effect of such parameters on the performance of the process. Because of the problems with rapid particle attrition explained above, the stated aims could not be achieved with any of the materials. So it turned out that each material happened to be examined in slightly different ways, an effect of changing parameters between the experimental campaigns to find benign conditions. Therefore, it is hard to compare the performances of different materials with one another. It is possible to draw some conclusions though. All materials examined in continuous operation were found to release gasphase O2, in fashion similar to that during experiments in batch reactor. All materials also had very high reactivity with fuel, readily providing almost full conversion of CH4 to CO2 and H2O. There was a very small CH4 slip for all materials, while the measured concentrations of CO and H2 were practically zero. There was also excess O2 present in flue gas once γCO2 > 99.5% was achieved. The reason for experimental failure was similar in all four campaigns. The particles were subject to severe attrition, which could be observed as a buildup of fine material in filters downstream the reactor. When the reactor was disassembled and emptied, it was found that used material had a quality

Table 6. Summary of Oxygen-Carrier Materials Examined in Continuous Operation designation

bulk density fresh (kg/m3)

bulk density used (kg/m3)

crushing strength fresh (N)

crushing strength used (N)

Cu40Z-950 Cu40Z-1030 Cu40Ce Cu40(YSZ)

1950 2050 2170 1700

1590 1550 1900 1410

3.4 not available 1.0 0.6

0.4 0.5 0.3 0.3

It can be concluded that all samples examined in continuous operation had low structural integrity. The reason for this is not obvious. Microscope investigation also revealed that many individual particles were defective. As can be seen in Figure 9, many were doughnut shaped, for example. It can be speculated that this may have contributed to the poor performance with respect to attrition. Yet fresh Cu40Z was very hard and did appear to be robust. Still, it did not fare better in operation than the softer Cu40Ce and Cu40(YSZ). Because of this, it seems reasonable to believe that the effect may be due to continuous reduction and oxidation, rather than due to mechanical attrition. 4.3. X-ray Diffractometry Analysis. All fresh samples were examined by X-ray powder diffraction using a Siemens D5000 diffractometer utilizing copper Kα1 radiation. Also, particles from all successfully conducted experiments, both in batch reactor and in continuous operation, were examined. In this case, the samples were obtained in oxidized form. The results are summarized in Table 7: On the basis of the results presented in Table 7, it is possible to draw a few conclusions. First, it can be seen that all samples achieved the desired phase composition. Support materials supposed to be inert (ZrO2, YSZ, CeO2, MgAl2O4) did in no instance react with CuO. La2O3 on the other hand reacted with the support materials forming separate phases of La2Zr2O7, Ce0.9La0.1O2−δ, and LaAlO3. Addition of Fe2O3, Mn3O4, MgO, and Al2O3 resulted in combined spinel structures with composition corresponding to the molar ratio of the ingredients, as could be expected. Nonstabilized ZrO2 was of monoclinic symmetry, while YSZ was of cubic symmetry, with possible traces of monoclinic ZrO2. The only unclear XRD result was for Cu40MgF48, which had spectra possibly indicating Mg1.5Fe1.5O3.75 rather than MgFe2O4. The latter seems more likely though and thus is the phase reported in the table. Second, it was found that different calcination temperatures of fresh particles did not influence the phase composition of the samples. This may seem unexpected because calcination in air at 1100 °C should have resulted in decomposition of CuO to Cu2O according to reaction 5. It seems reasonable to assume 6264



Article

Table 7. Phase Composition of Oxygen-Carrier Materials Examined in This Study sample Cu40Z Cu40ZLa5 Cu40(YSZ) Cu40Ce Cu40CeLa5 Cu40(MgAl) Cu40(MgAl*) Cu40(MgAl)La5 Cu60F Cu40MF41 Cu40MgF48 Cu36FAl24 a

raw materials (wt %) 40 40 40 40 40 40 40 40 60 40 40 36

CuO, CuO, CuO, CuO, CuO, CuO, CuO, CuO, CuO, CuO, CuO, CuO,

60 55 60 60 55 60 60 60 40 41 48 40

ZrO2 ZrO2, 5 La2O3 YSZ CeO2 CeO2, 5 La2O3 MgAl2O4 MgAl2O4 MgAl2O4, 5 La2O3 Fe2O3 Fe2O3, 19 Mn3O4 Fe2O3, 12 MgO Fe2O3, 24 Al2O3

phases in fresh sample (calcined 950−1100 °C)

phases in used sample (from batch reactor)

CuO, ZrO2a CuO, ZrO2,a La2Zr2O7 CuO, ZrxY1−xO2b CuO, CeO2 CuO, CeO2, Ce0.9La0.1O2−δ CuO, MgAl2O4 CuO, MgAl2O4 CuO, MgAl2O4, LaAlO3 CuO, CuxFe3−xO4 CuO, MnxFe3−xO4 CuO, MgFe2O4 Cu0.95Fe1.05AlO4

CuO, ZrO2a CuO, ZrO2,a La2Zr2O7 CuO, ZrxY1−xO2b CuO, CeO2 CuO, CeO2, Ce0.9La0.1O2−δ CuO, MgAl2O4 CuO, MgAl2O4 CuO, MgAl2O4, LaAlO3 CuO, CuxFe3−xO4 CuO, MnxFe3−xO4 CuO, MgFe2O4 Cu0.95Fe1.05AlO4

Monoclinic. bCubic.

concentration as compared to particles with inert support. Materials with semiactive support also had less good reactivity with CH4. No formation of unwanted or unexpected phases could be detected by X-ray diffractometry, and all chemical reactions appear to have been completely reversible. Three material compositions (CuO−ZrO2, CuO−CeO2, CuO−YSZ) were chosen for operation in a continuous fluidized-bed reactor system. All three materials were capable of providing more or less full conversion of natural gas during operation. However, they also suffered from fast attrition, and all turned into a flour-like substance after only a few hours of operation. Crushing strength analysis showed that particles remaining after having been used in continuous operation were physically much weaker than fresh. In total, 23 h of continuous operation with fuel addition was recorded. Among the materials examined, CuO supported on YSZ was the most durable. It can be concluded that CuO as active phase in chemicallooping combustion with O2 uncoupling has potential to provide very high rates of O2 release and very high reactivity with fuel at 900−925 °C. CuO therefore remains a prime candidate for this application. The experiments in continuously operating reactor show severe attrition taking place during repeated reduction and oxidation though. It is clear that this phenomenon must be carefully studied and better understood.

that the particles were oxidized back to CuO during the cool down period following calcination though, because calcination took place in air. Third, in no case could it be seen that the phase composition of oxidized samples changed over the course of the experiments. Not even prolonged reduction in batch reactor or continuous operation affected the phases identified in the samples. This suggests that all redox reactions involved are completely reversible.

5. DISCUSSION The results of the current study are both intriguing and disappointing. It is clear that CuO-based oxygen-carrier particles in many respects have superior properties as compared to those of materials such as NiO, Mn3O4, and Fe2O3. The rapid release of O2 should be a huge advantage when oxidizing solid fuels, and the reactivity with CH4 is also very good. Further, CuO could be expected to be resistant to impurities such as sulfur, and all chemical reactions involved appear to be completely reversible. Despite all of those advantages, the experiments performed in the continuously operating reactor reveal that attrition may be a considerable problem for CuO-based oxygen carriers. Unfortunately, the attrition mechanisms involved are not well understood. It seems that attrition phenomena during continuous reduction and oxidation of CuO and related support materials will need to be further studied, before certain conclusions about the feasibility of CuO as oxygen carriermaterial can be drawn.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

6. CONCLUSIONS Twelve different CuO-based oxygen-carrier particles for chemical-looping with oxygen uncoupling have been manufactured. Each material was calcined at two different temperatures between 950 and 1100 °C. Some particles became very soft while others showed melting behavior during calcination, or defluidization during experiments. Out of the 24 prepared samples, 10 were successfully examined in a batch fluidized-bed reactor. Composite particles consisting of CuO and inert support material such as ZrO2, YSZ, CeO2, and MgAl2O4 provided full conversion of CH4 at 900−925 °C, and were also found to release gas-phase O2 into inert atmosphere at these temperatures when fluidized with N2. Particles using semiactive support such as Fe2O3, Mn2O3, and Al2O3 formed combined spinel structures with CuO. Such materials still released gas-phase O 2 , but at different

ACKNOWLEDGMENTS This research has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) via grant agreement no. 241401. Ali Hedayati and Cristina Dueso are acknowledged for having participated in some of the experimental work.

■

REFERENCES

(1) Intergovernmental Panel on Climate Change. Carbon Dioxide Capture and Storage; Cambridge University Press: UK, 2005. (2) Boot-Handford, M. E.; Abanades, J. C.; Anthony, E. J.; Blunt, M. J.; Brandani, S.; Dowell, N. M.; Fernandez, J. R.; Ferrari, M. C.; Gross, R.; Hallett, J. P.; Haszeldine, R. S.; Heptonstall, P.; Lyngfelt, A.; Makuch, Z.; Mangano, E.; Porter, R. T. J.; Pourkashanian, M.; Rochelle, G. T.; Shah, N.; Yaoa, J. G.; Fennell, P. S. Carbon capture

6265



Article

and storage update. Energy Environ. Sci., in press, DOI: 10.1039/ c3ee42350f. (3) Lewis, K. W.; Gilliland, E. R.; Sweeney, M. P. Gasification of carbon: Metal oxides in a fluidized powder bed. Chem. Eng. Prog. 1951, 47, 251−256. (4) Lewis, K. W.; Gilliland, E. R. U.S. Patent 2,665,972, 1954. (5) Richter, H. J.; Knoche, K. Reversibility of combustion processes. ACS Symp. Ser. 1983, 235, 71−85. (6) Knoche, K. F.; Richter, H. Verbesserung der reversibilitaet von verbrennungsprozessen. Brennst.-Waerme-Kraft 1968, 20, 20−210. (7) Ishida, M.; Zheng, D.; Akehata, T. Evaluation of a chemicallooping combustion power-generation system by graphic exergy analysis. Energy 1987, 12, 147−154. (8) Ishida, M.; Jin, H. A new advanced power-generation system using chemical-looping combustion. Energy 1994, 19, 415−422. (9) Lyngfelt, A.; Leckner, B.; Mattisson, T. A fluidized-bed combustion process with inherent CO2 separation; application of chemical-looping combustion. Chem. Eng. Sci. 2001, 56, 3101−3113. (10) Lyngfelt, A.; Thunman. H. Construction and 100 h of operational experience of a 10-kW chemical looping combustor. Carbon Dioxide Capture for Storage in Deep Geologic Formations Results from the CO2 Capture Project; Elsevier Science: London, 2005; Vol. 1, pp 625−646. (11) Lyngfelt, A. Oxygen carriers for chemical-looping combustion 4000 h of operational experience. Oil Gas Sci. Technol. 2011, 66, 161− 172. (12) Mattisson, T.; Lyngfelt, A.; Leion, H. Chemical-looping with oxygen uncoupling for combustion of solid fuels. Int. J. Greenhouse Gas Control 2009, 3, 11−19. (13) Leion, H.; Mattisson, T.; Lyngfelt, A. Using chemical-looping with oxygen uncoupling (CLOU) for combustion six different of solid fuels. Energy Procedia 2009, 1, 447−453. (14) Adanez, J.; Abad, A.; Garcia-Labiano, F.; Gayan, P.; de Diego, L. F. Progress in chemical-looping combustion and reforming technologies. Prog. Energy Combust. Sci. 2012, 38, 215−282. (15) Fan, L. S. Chemical Looping Systems for Fossil Energy Conversion; American Institute of Chemical Engineers and John Wiley & Sons: New York, 2010. (16) Lyngfelt, A. Chemical looping combustion. In Fluidized-Bed Technologies for Near-Zero Emission Combustion and Gasification; Scala, F., Ed.; Woodhead Publishing Limited: UK, 2013; Chapter 20, pp 895−930. (17) de Diego, L. F.; Gayán, P.; García-Labiano, F.; Celaya, J.; Abad, A.; Adánez, J. Impregnated CuO/Al2O3 oxygen carriers for chemicallooping combustion: Avoiding fluidized bed agglomeration. Energy Fuels 2005, 19, 1850−1856. (18) Comparison of iron-, nickel-, copper- and manganese-based oxygen carriers for chemical-looping combustion Cho, P.; Mattisson, T.; Lyngfelt, A. Fuel 2004, 83, 1215−1225. (19) Copeland, R.; Alptekin, G.; Cesario, M.; Gershanovich, Y. Sorbent energy transfer system (SETS) for CO2 separation with high efficiency. Proceedings of the 27th International Technology Conference on Coal Utilization & Fuel Systems, Clearwater, U.S, 2002. (20) Chuang, S. Y.; Dennis, J. S.; Hayhurst, A. N.; Scott, S. A. Development and performance of Cu-based oxygen carriers for chemical-looping combustion. Combust. Flame 2008, 154, 109−121. (21) Arjmand, M.; Azad, A. M.; Leion, H.; Lyngfelt, A.; Mattisson, T. Prospects of Al2O3 and MgAl2O4-supported CuO oxygen carriers in chemical-looping combustion (CLC) and chemical-looping with oxygen uncoupling (CLOU). Energy Fuels 2011, 25, 5493−5502. (22) Gayán, P.; Forero, C. R.; Abad, A.; de Diego, L. F.; GarcíaLabiano, F.; Adánez, J. Effect of support on the behavior of Cu-based oxygen carriers during long-term CLC operation at temperatures above 1073 K. Energy Fuels 2011, 25, 1316−1326. (23) Forero, C. R.; Gayán, P.; García-Labiano, F.; de Diego, L. F.; Abad, A.; Adánez, J. High temperature behaviour of a CuO/γAl2O3 oxygen carrier for chemical-looping combustion. Int. J. Greenhouse Gas Control 2011, 5, 659−667.

(24) de Diego, L. F.; García-Labiano, F.; Gayán, P.; Celaya, J.; Palacios, J. M.; Adánez, J. Operation of a 10 kWth chemical-looping combustor during 200 h with a CuO-Al2O3 oxygen carrier. Fuel 2007, 86, 1036−1045. (25) Wang, S.; Luo, M.; Wang, G.; Wang, L.; Lv, M. Analysis of reactivity of a CuO-based oxygen carrier for chemical looping combustion of coal. Energy Fuels 2012, 26, 3275−3283. (26) Wang, B.; Zhao, H.; Zheng, Y.; Liu, Z.; Yan, R.; Zheng, C. Chemical looping combustion of a Chinese anthracite with Fe2O3based and CuO-based oxygen carriers. Fuel Process. Technol. 2012, 96, 104−115. (27) Arjmand, M.; Azad, A. M.; Leion, H.; Lyngfelt, A.; Mattisson, T. Evaluation of CuAl2O4 as an oxygen carrier in chemical-looping combustion. Ind. Eng. Chem. Res. 2012, 51, 13924−13934. (28) Gayán, P.; Adánez-Rubio, I.; Abad, A.; de Diego, L. F.; GarcíaLabiano, F.; Adánez, J. Development of Cu-based oxygen carriers forchemical-looping with oxygen uncoupling (CLOU) process. Fuel 2012, 96, 226−238. (29) de Diego, L. F.; GarcIa-Labiano, F.; Adánez, J.; Gayán, P.; Abad, A.; Corbella, B. M.; MarIa Palacios, J. Development of Cu-based oxygen carriers for chemical-looping combustion. Fuel 2004, 83, 1749−1757. (30) Adánez-Rubio, I.; Gayán, P.; Abad, A.; de Diego, L. F.; GarcíaLabiano, F.; Adánez, J. Evaluation of a spray-dried CuO/MgAl2O4 oxygen carrier for the chemical looping with oxygen uncoupling process. Energy Fuels 2012, 26, 3069−3081. (31) Abad, A.; Adánez-Rubio, I.; Gayán, P.; García-Labiano, F.; de Diego, L. F.; Adánez, J. Demonstration of chemical-looping with oxygen uncoupling (CLOU) process in a 1.5kWth continuously operating unit using a Cu-based oxygen-carrier. Int. J. Greenhouse Gas Control 2012, 6, 189−200. (32) Mattisson, T.; Leion, H.; Lyngfelt, A. Chemical-looping with oxygen uncoupling using CuO/ZrO2 with petroleum coke. Fuel 2009, 88, 683−690. (33) Moldenhauer, P.; Rydén, M.; Mattisson, T.; Lyngfelt, A. Chemical-looping combustion and chemical-looping with oxygen uncoupling of kerosene with Mn- and Cu-based oxygen carriers in a circulating fluidized-bed 300 W laboratory reactor. Fuel Process. Technol. 2012, 104, 378−389. (34) Abdul-Majeed, A.; Hedayati, A.; Rydén, M.; Leion, H.; Mattisson, T. Examining the Cu-Mn-O spinel system as an oxygen carrier in chemical looping combustion. Energy Technol. 2013, 1, 59− 69. (35) Mohammad Pour, N.; Leion, H.; Rydén, M.; Mattisson, T. Combined Cu/Mn oxides as oxygen carrier in chemical-looping with oxygen uncoupling (CLOU). Energy Fuels 2013, 27, 6031−6039. (36) Rydén, M.; Leion, H.; Lyngfelt, A.; Mattisson, T. Combined oxides as oxygen carrier material for chemical-looping combustion with oxygen uncoupling. Appl. Energy 2014, 113, 1924−1932. (37) Cao, Y.; Zhao, H. Y.; Song, P. S.; Wei-Ping, P. Lanthanumpromoted copper-based oxygen carriers for chemical looping combustion process. Proceedings of the 41st Annual Conference of the North American Thermal Analysis Society (NATAS), Bowling Green, OH, August 4−7, 2013. (38) Jerndal, E.; Mattisson, T.; Thijs, I.; Snijkers, F.; Lyngfelt, A. Investigation of NiO/NiAl2O4 oxygen carriers for chemical-looping combustion produced by spray-drying. Int. J. Greenhouse Gas Control 2010, 4, 23−35. (39) Rydén, M.; Moldenhauer, P.; Lindqvist, S.; Mattisson, T.; Lyngfelt, A. Measuring attrition resistance of oxygen carrier particles for chemical looping combustion with a customized jet cup. Powder Technol. 2014, in press. (40) Moldenhauer, P.; Rydén, M.; Younes, M.; Mattisson, T.; Lyngfelt, A. The use of ilmenite as oxygen carrier with kerosene in a 300W CLC laboratory reactor with continuous circulation. Appl. Energy 2014, 113, 1846−1854. (41) Rydén, M.; Arjmand, M. Continuous hydrogen production via the steam-iron reaction by chemical looping in a circulating fluidizedbed reactor. Int. J. Hydrogen Energy 2012, 37, 4843−4854. 6266



Article

(42) Jacob, K. T.; Fitzner, K.; Alcock, C. B. Activities in the spinel solid solution, phase equilibria and thermodynamic properties of ternary phases in the system Cu-Fe-O. Metall. Trans. B 1977, 8, 451− 460. (43) Chevalier, J.; Gremillard, L.; Virkar, A. V.; Clarke, D. R. The tetragonal-monoclinic transformation in zirconia: Lessons learned and future trends. J. Am. Ceram. Soc. 2009, 92, 1901−1920.

6267


CuO-Based Oxygen-Carrier Particles for Chemical ... - ACS Publications

Recommend Documents