WO2007004894A2

WO2007004894A2 - Method for producing tetragonal lithium zirconate

Info

Publication number: WO2007004894A2
Application number: PCT/NO2006/000254
Authority: WO
Inventors: Dag Øistein ERIKSEN; Kwang Bok Yi
Original assignee: Institutt For Energiteknikk
Priority date: 2005-06-30
Filing date: 2006-06-30
Publication date: 2007-01-11
Also published as: WO2007004894A3; NO20053197D0; NO323419B1; NO20053197L

Abstract

This invention relates to a method for producing substantially tetragonal lithium zirconate with small crystallite sizes by preparation of an aqueous solution of zirconium oxynitrate, ZrO(NO3)2 and a water solvable lithium salt in a molar ratio of Li:Zr of 2, precipitate lithium zirconate, Li2ZrO3 by adding aqueous ammonia, NH3 to the solution, filtering the solution and recover the filter cake, drying the filter cake, and calcinate the filter cake until a dry powder of lithium zirconate is obtained.

Description

METHOD FOR PRODUCING LITHIUM ZIRCONATE

Field of invention

This invention relates to a method for producing substantially tetragonal lithium zirconate with small crystallite sizes. Background

During the last decades the scientific community has reached an understanding that the earth is presently going through a rapid climate change that, if allowed to continue at the same pace in the coming decades, probably will result in a global warming with serious ecological and climatic consequences on a global scale. The human releases of greenhouse gases are identified as one of the main forcings behind this climate change.

The single most important human made greenhouse gas is carbon dioxide (CO₂) resulting from combustion of fossil fuels. To date, the human made emissions of CO₂ have increased the atmospheric concentration from a pre-industrial level of 270 ppm to about 385 ppm, and this increase of CO₂ in the atmosphere is estimated to have shifted the earth's energy radiant balance by about 1.4 W/m². Obviously, human made emissions of CO₂ from the use of fossil fuels should be significantly reduced in order to halt the accumulation of CO₂ in the atmosphere.

The global reserves of fossil fuels are estimated to allow continuing the present day consumption for decades, and possibly for centuries in the case of coal. Therefore, even though there are presently many attempts to develop competitive renewable energy carriers, fossil fuels will most probably be the major energy carrier for decades ahead.

This has resulted in an increased activity in developing various approaches for CO₂ reduction that may be implemented in present and future technology for combustion of fossil fuels. Capture of CO₂ from existing processes producing substantial amounts of CO₂, such as coal-burning power plants and steam-fuel reforming by a sorption-desorption cycle is one of them.

Prior art In both power plants and fuel reforming processes flue gas or product gas containing CO₂ is present at temperatures above 400°C, and there are known a broad range of concepts for utilizing sorption-desorption cycles to separate CO₂ from a flue gas (post combustion capture) or fuel gas (pre combustion capture). These include commercially available amines process and physical sorbents. In order to obtain proper economic evaluation of processes, sorbent properties should be obtained accounting the unit cost of the sorbent and high temperature sorbent selection is very important for both processes [I]. Thus the sorbent must be heat tolerant, possess high CO₂ sorption capacity, and exhibit fast kinetics. Satisfactory heat tolerance and high capacity in multi-cycle operation are seldom provided with currently available sorbents [2]. It is known that lithium zirconate (Li₂ZrO₃) can react with CO₂ over the temperature range from 45O⁰C to 550°C based on the following reaction [3]:

Li₂ZrO₃ + CO₂ = Li₂CO₃ + ZrO₂ (A)

The stoichiometric capacity is 28.7% CO₂ by weight, and this number will be considered as the maximum target value in this application. Currently, around 20% CO₂ capture by weight has been achieved with Li₂ZrO₃ in other studies, a value that is higher compared to most other known sorbents.

It has been suggested that the reaction mechanism involves a double shell formation around the solid Li₂ZrO₃ according to the following reactions:

Li₂ZrO₃ = ZrO₂ + 2Li⁺ + O₂ ^" (B) CO₂ + 2Li⁺ + O₂ ^' = Li₂CO₃ (C)

Li₂CO₃ forms the outer shell leaving the layer of ZrO₂ behind. As a consequence, the inner core of Li₂ZrO₃ shrinks as ZrO₂ and Li₂CO₃ layers grow. Lithium and oxygen ions diffuse through ZrO₂ and Li₂CO₃ layers for further reaction. Because reaction (C) is very fast, the rate-limiting step is the diffusion of lithium and oxygen ions through the ZrO₂ and Li₂CO₃ layers, which is a slow process.

It has been reported that the reaction rate is accelerated when lithium zirconate is doped with sodium carbonate and/or potassium carbonate. The doping produces a eutectic molten carbonate that reduces CO₂ diffusion resistance. Others have compared the performance of lithium zirconate prepared using tetragonal and monoclinic zirconia, and reported that tetragonal lithium zirconate reacted with CO₂ significantly faster [4]. It is also reported that particle size and starting material size influences the CO₂ sorption rate.

Despite several attractive characteristics of lithium zirconate, its sorption rate is too slow. Usually approximately 250 min are required to reach maximum sorption capacity. Most lithium zirconate to date were prepared by high temperature (850 to 100O⁰C) solid-state reaction. These materials required a carbonate additive to achieve a reasonable reaction rate. Even with the additive the reaction rate strongly depends on particle size. The preparation method of solid-state reaction is energy intensive, requires long reaction times, and it is difficult to control the product particle size. In addition, the introduction of carbonate additives will reduce CO₂ capacity of the sorbent. Objective of the invention

The main objective of the invention is to provide a low-temperature method for producing substantially tetragonal lithium zirconate with small crystallite sizes.

Summary of the invention The objective of the invention may be obtained by a method that shows the features as described in the following description of the invention and/or in the appended claims.

The method according to the invention for producing tetragonal lithium zirconate with small crystallite sizes is characterised in that it comprises: - preparation of an aqueous solution of zirconium oxynitrate, ZrO(NO₃)₂-2H₂O and a solvable lithium salt, preferably lithium nitrate LiNO₃-XH₂O, in a molar ratio of Li:Zr of 2,

- precipitation of lithium zirconate, Li₂ZrO₃ by adding aqueous ammonia, NH₃ to the solution, - filtration of the solution and recovering of the filter cake,

- drying the filter cake, and ^■

- calcinations of the filter cake, followed by a slight grinding to obtain a dry powder of lithium zirconate.

List of figures Figure 1 shows XRD spectra of a) conventionally produced lithium zirconate and b) lithium zirconate produced according to the invention.

Figure 2 shows SEM images of a) lithium zirconate produced according to the invention and b) conventionally produced lithium zirconate.

Figure 3 shows a SEM image of the lithium zirconate shown in Figure 2 a) at a larger magnification.

Figure 4 shows a comparison of the sorption rate of CO₂ of conventionally produced lithium zirconate and lithium zirconate produced according to the invention.

Figure 5 shows the effect of temperature on the sorption rate of CO₂ of the lithium zirconate produced according to the invention. Figure 6 shows a multi cycle test for the sorption rate of CO₂ of the lithium zirconate produced according to the invention.

Figure 7 shows the effect of temperature on the release rate of CO₂ of the lithium zirconate produced according to the invention.

Figure 8 shows the effect of partial pressure of CO₂ on the sorption rate of lithium zirconate produced according to the invention. Figure 9 shows the effect of the partial pressure steam in the flue gas on the sorption rate of CO₂ of the lithium zirconate produced according to the invention. The partial pressure Of CO₂ was fixed at 0.1 bar.

Detailed description of the invention The invention will be described in more detail under reference to a preferred embodiment and verification tests performed on the preferred embodiment as compared to conventionally produced lithium zirconate.

One sample of tetragonal lithium zirconate with small crystallite sizes produced according to the inventive method and on sample of lithium zirconate produced by the conventional high temperature solid state reaction were made in order to verify the effect of the lithium zirconate produced according to the invention in a series comparison tests. The sample produced according to the invention are referred to as LiZr-LL (low temperature liquid reaction) and the conventionally produced comparison sample is referred to LiZr-HS (high temperature solid state reaction) in the following.

Preferred embodiment

The preferred embodiment of the invention was prepared as follows:

Zirconium oxynitrate (ZrO(NO₃)₂-2H₂O, from Merck) and lithium nitrate (LiNO₃-XH₂O, approx. 78% LiNO₃, from Merck) at Li/Zr molar ratio of 2 were dissolved in double distilled deionised water at 70°C using separate beakers. 500 ml of 12.5 mM LiNO₃ and 25 mM ZrO(NO₃)₂ were used. LiNO₃-XH₂O dissolved instantly while ZrO(NO₃)₂-2H₂O needed more time to dissolve and produced a cloudy solution. It is possible to perform the reaction at temperatures ranging from 25 to 70 ⁰C₅ but 70 ⁰C has shown to function best and is preferred. The solutions were then mixed and the final mixture was stirred using a magnetic bar for 30 minutes (minimum 10 minutes). The mixed oxide was then precipitated by the drop wise addition of aqueous ammonia solution (NH₃, 25% solution, Merck). The addition of ammonia solution was continued until no more precipitation occurred. Stirring was stopped and the precipitate was filtered using ash-free paper filter (Schleicher & Schull) using a hydraulic vacuumed funnel. The filter cake was dried overnight in air at 150°C and then calcined at 500°C for 2 h. The drying and calcinations may be performed in air temperatures from 100 to 200 ⁰C and from 400 to 700 °C, respectively. The preferred temperatures are 150 ⁰C and 500 °C, respectively. The drying of the powder should be performed until a dry powder is formed (usually around 12 hours). The calcination should be continued for 1 to 4 hours, but preferably around 2 hours. Comparison sample

The conventionally produced comparison sample was prepared using a high temperature solid-state reaction:

Lithium carbonate (Li₂CO₃, 99%, Fluka), zirconia (ZrO₂, 99%, Aldrich), and potassium carbonate (K₂CO₃, 99%, Fluka). The molar ratio of the starting materials was adjusted to Li₂CO₃IZrO₂IK₂CO₃ = 1.1 :1.0:0.1. Each material was weighed, mixed, and ground in an agate mortar. Then the mixture was heat treated at 1000°C for 1O h. The calcined material was quenched in air and again ground in an agate mortar before testing. Verification of the invention

In order to verify the invention, the LiZr-LL and LiZr-HS were characterised and subject to a series of tests.

Characterisation

The formed lithium zirconates, the sample according to the invention (LiZr-LL) and the conventional comparison sample (LiZr-HS) were characterised by their XRD spectra using an Inel XRG 3000 diffractometer with CuKa radiation. XRD test results were retrieved and stored using commercial software (Inel Acquisition). Crystal sizes were calculated based on resolved peaks using the Scherrer equationl2. Sample surface morphology was investigated by scanning electron microscopy (SEM, Hithachi S-4800 Field Emission). ,

The CO₂ sorption kinetics and capacity were studied using a thermogravimetric analyzer (TGA₉ CI electronics). The TGA test was started in a N₂ atmosphere with temperature increased to the desired reaction temperature at 10°C/min. CO₂ was introduced after 20 minutes at the desired temperature. The N₂/CO₂ ratio was controlled by mass flow controllers (Bronkhorst, EL-FLOW Digital series). Steam flow was controlled using a liquid flow controller (Bronkhorst, Liqui-flow) with a controlled evaporation mixing system (Bronkhorst, CEM). The reverse reaction was studied at each temperature by turning off the CO₂ flow after carbonation was complete. The effects of CO₂ and steam partial pressures on the carbonation of Li₂ZrO₃ were also investigated.

Results and discussion

XRD results of LiZr-HS and LiZr-LL are shown in Figure 1. The major component of LiZr-HS is monoclinic Li₂ZrO₃ while that of LiZr-LL is tetragonal Li₂CO₃. The diffraction peaks of LiZr-LL are much lower and wider than those of LiZr-HS, indicating that the crystallite size of LiZr-LL is much smaller than LiZr-HS.

Crystallite sizes of LiZr-HS and LiZr-LL, calculated using the Scherrer equation, were 40 and 5nm, respectively. The crystallites formed by the method according to show a variation in size ranging from about 4 to 10 nm. Figure 2 shows scanning electron micrographs of LiZr-LL and LiZr-HS at the same magnification. These two materials have completely different particle morphology. It is clear that LiZr-LL particles are extremely small and have relatively uniform particle sizes. However, LiZr-HS particles have a smooth, dense surface with no indication of cracks or agglomeration of small particles. This suggests that it may be difficult for CO₂ penetrate to inside of individual particles, and, as a consequence, the reaction rate with CO₂ is expected to be slower than that of LiZr- LL. The average particle size of LiZr-HS is around 2.5 μm. This particle size is different from crystallite size described in the previous section. A Crystallite is domain of solid-state matter that has the same structure as a single crystal. Particle is a result of agglomeration of crystallites.

At larger magnification the particle size of LiZr-LL is determined to be around 40 nm as shown in Figure 3. Its particle size is around 40nm. This particle size is almost same size of crystallite of reported Li₂ZrO₃. Figure 4 shows TGA test results of LiZr-HS and LiZr-LL. The tests were carried out in pure CO₂ at 500°C. The CO₂ uptake rate of LiZr-LL was much faster than that of LiZr-HS. Less than 6 min were required to reach 20% weight change. The rate then decreased rapidly after reaching 23% weight change and flattened out. The ultimate weight gain of 26% was about 90% of the stoichiometric capacity (the dash line in Figure 4) Of Li₂ZrO₃. The rate of CO₂ sorption was calculated using Excel spread sheet regression function in the range of 0 to 20% weight gain and was about 38 mg/(g-min), which is at least 3 to 10 times faster than reported rates for other LiZrO₃ sorbents. The small particle size of LiZr-LL seems to be main factor contributing to the increase in CO₂ uptake rate. During carbonation, the layers of Li₂Cθ₃ and ZrO₂ formed outside of the Li₂ZrO₃ core are expected to be extremely thin. Before those layers reach sufficient thickness to limit diffusion of CO₂, most of the Li₂ZrO₃ would be already reacted.

Figure 5 shows the effect of temperature on CO₂ uptake rate. Tests were carried out in a pure CO₂ atmosphere at temperatures ranging from 450°C to 600°C. As temperature increased from 450⁰C to 500°C the CO₂ uptake rate increased significantly. However, the rate decreased at 600°C and was similar to that at 500°C. Explanation for this behaviour has been based on the effect of temperature on CO₂ equilibrium partial pressure [5]. Increasing temperature will result in an increase in both the reaction rate constant and the equilibrium partial pressure of CO₂. This is a very important concept to be considered when effective CO₂ removal is desired. In pure CO₂, the reaction driving force, which is characterized by the difference between experimental and equilibrium CO₂ pressure, does not change by a large amount because the experimental partial pressure of CO₂ is significantly higher than the equilibrium CO₂ partial pressure. However, this partial pressure driving force will be more dependent on temperature when lower experimental 00254

concentrations of CO₂ are tested. Lower temperature would be required to achieve lower exit CO₂ concentration. However, this positive effect of lower temperature on equilibrium CO₂ partial pressure will be cancelled by slow kinetics at some point.

Since LiZr-LL was synthesized at lower temperature than actual carbonation or calcinations temperature it is very important to check its thermal stability. The multi-cycle test result is shown in Figure 6. Temperature increased to 550⁰C at 10°C/min in N₂ atmosphere. After 10 minutes of stabilization, the atmosphere was switched to CO₂ (200ml/min, 25°C) and 20 minutes of reaction time was given. Then temperature increased 690°C along with switching atmosphere to N₂ for regeneration. After 10 minutes of dwelling at 69O⁰C, temperature decreased to

550⁰C at 3°C/min for another CO₂ uptake. The same manipulations were done to obtain second and third CO₂ uptakes. As shown in Figure 6, CO₂ capacity was not reduced after two cycles of carbonation and regeneration. CO₂ uptake rate of second cycle, however, was slightly slower than that of first cycle. The third CO₂ uptake rate remained as same as the second one. CO₂ uptake rates were calculated using Excel spread sheet regression function in the range of 0 to 20% CO₂ uptake. First CO₂ uptake rate was 4.98 mg/g min, second and third were 2.92 mg/g-min and 2.93 mg/g min respectively. Even though LiZr-LL was prepared at room temperature and calcined at 500⁰C for short period of time, test result identified fine thermal stability.

CO₂ release rates from carbonated sorbents are presented in Figure 7 at different temperatures from 500°C to 65O⁰C. 70 mg of LiZr-LL was reacted with 200 ml/min (25°C, 1 bar) of pure CO₂ for an hour. Then, the CO₂ atmosphere was switched to N₂. At the moment of switching atmospheres the weight of CO₂ absorbed is designated 100%. 80% of the CO₂ captured during carbonation was released after 60 min at 600⁰C. 90% of the CO₂ was released after two hours. The CO₂ release rate was significantly smaller at 550⁰C, and at 500⁰C, effectively no CO₂ was released during the 2 hr test period. This result is somewhat consistent with the argument described above that lower temperature is favoured in low concentration of CO₂ capture using Li₂ZrO₃.

The effect of CO₂ partial pressure on CO₂ sorption using LiZr-LL at a reaction temperature of 500⁰C is presented in Figure 8. The flow rate of the CO₂/N₂ mixture was 300 ml/min (25°C, 1 bar). CO₂ partial pressure was varied from 1 to 0.1 bar. As expected, the CO₂ uptake rate was extremely fast at P_COi =1 bar, and decreased slowly asP_COz decreased until 0.3 bar. However, the uptake rate became quite slow when P_CQ2 decreased to 0.1 bar. Only 10% weight gain was achieved in 10% CO₂ in N₂ in 2 hours. 6 000254

8

This result confirms that the major challenge of utilizing Li₂ZrO₃ in CO₂ capture is achieving low concentration of CO₂ in the exit gas. Reduced CO₂ capacity in low partial pressure of CO₂ can be explained by associating it with diffusion and reaction rate. It is assumed that at high P_c0 diffusion of CO₂ into the core of macro-particles and reaction to form Li₂CO3 occur at the same time. Therefore, the global reaction is under kinetic control. When P_∞ϊ is low, however, all CO₂ molecules are reacted on the surface of particle before they diffuse into the core. Since volumetric expansion occurs when Li₂ZrO₃ is converted to Li₂CO₃ and ZrO₂, the diffusion path becomes increasingly blocked, and the global rate becomes diffusion controlled. Considering that gas leaving the steam methane reformer (SMR) or coal-burning power plant contain 10 to 13 %15, 16, the challenge becomes evident.

However, the gases leaving both processes also contain steam. Typical steam contents are 30% in the SMR exit gas and 10% in coal fired power plant flue gas. Since steam usually affects the kinetics of solid-gas reactions, the effect of steam addition on CO₂ sorption of LiZr-LL was also investigated and is illustrated in Figure 9. P_COj was fixed at O.lbar, temperature at 500⁰C₅ and total flow rate of the

CO₂/H₂O/N₂ mixture was 300ml/min (25°C, 1 bar). The concentration of steam was varied from 0% to 30%. Steam clearly increased the CO₂ uptake rate, especially when the steam concentration was increased from 10% to 20%. 20% and 30% steam addition resulted in somewhat similar CO₂ uptake rates. This positive effect of steam on CO₂ uptake suggests that Li₂Zr03 sorbent might be more suitable to SMR than coal-burning power plants because of its higher steam content.

Conclusion Li₂ZrO₃ was prepared using liquid base precipitation method and solid-state reaction method in this study. Its CO₂ sorption kinetics and capacity were investigated in various conditions. CO₂ sorption rate of Li₂ZrO₃ prepared using precipitation method was several tens times faster than that prepared using solid- state reaction method. The Li₂ZrO₃ possessed average aggregate size of 40nm and absorb CO₂ amount to 20 wt% of sample weight less than 6 min at 500°C. CO₂- release tests from saturated sorbent were carried out and recommended CO₂ absorption temperature using Li₂ZrO₃ was 500⁰C Absorption-desorption cycle test using the Li₂ZrO₃ in 500°C and 69O⁰C was carried out up to three cycles and its CO₂ uptake rate and capacity were maintained as fresh one. CO₂ uptake rate decreased significantly as CO₂ partial pressure decreased less than 0.3 bar.

However, more than 20% of steam addition increased CO₂ uptake rate substantially. References:

1 Abanades, J.; Rubin, E.; Anthony, E. INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH 2004, 43, (13), 3462-3466.

2 Xiong, R.; Ida, J.; Lin, Y. CHEMICAL ENGINEERING SCIENCE 2003, 58, (19), 4377-4385.

3 Nakagawa, K.; Ohashi, T. JOURNAL OF THE ELECTROCHEMICAL SOCIETY 1998, 145, (4), 1344-1346.

4 Wang, Y.; Qi, L. CHINESE JOURNAL OF INORGANIC CHEMISTRY 2004, 20, (7), 770-774. 5 Ida, J. ; Xiong, R. ; Lin, Y. SEPARA TION AND P URIFICA TION TECHNOLOGY ^'2004, 36, (1), 41-51.

Claims

1. A method for producing substantially tetragonal lithium zirconate with small crystallite sizes, characterised in that it comprises: - preparation of an aqueous solution of zirconium oxynitrate, ZrO(NO₃)2 and a water solvable lithium salt in a molar ratio of Li:Zr of 2,

- precipitate lithium zirconate, Li₂ZrO₃ by adding aqueous ammonia, NH₃ to the solution,

- filtering the solution and recover the filter cake, - drying the filter cake, and

- calcinate the filter cake until a dry powder of lithium zirconate is obtained.

2. Method according to claim 1, characterised in that it is employed lithium nitrate, LiNO₃ as the water solvable lithium salt.

3. Method according to claim 1 , characterised in that the method produces at least 95 % tetragonal lithium zirconate.

4. Method according to claim 1, characterised in that the zirconium oxynitrate and lithium nitrate are dissolved in separate containers with double distilled deionised water with temperature in the range of 25 to 90°C and then mixing the solutions under stirring for at least 10 minutes.

5. Method according to claim 4, characterised in that the solution temperature is 70 ⁰C and that the stirring takes place in about 30 minutes.

6. Method according to claim 1, characterised in that the precipitation of solid lithium zirconate is obtained by drop wise adding of aqueous ammonia until no further precipitation is observed.

7. Method according to any of claim 1, characterised in that the filter cake is dried in an air stream at 100 to 200 ⁰C for about 12 hours, and then calcined for 1 to 4 hours at 400 to 700°C.

8. Method according to claim 7, characterised in that the filter cake is dried in air at 150 ⁰C in 12 hours and then calcined at 500 °C in 2 hours.

9. Tetragonal lithium zirconate,

Characterised in that it is produced by the method as specified in one of claims 1 to 4.

10. Tetragonal lithium zirconate according to claim 5, Characterised in that the crystallite sizes is in the range from about 4 nm to about 10 nm.