Optimum Condition of Hydration of Quicklime for Efficient Cleaning of Flue Gas from Municipal Waste Incineration

Transcription

1 Optimum Condition of Hydration of Quicklime for Efficient Cleaning of Flue Gas from Municipal Waste Incineration Tetsushi Iwashita 1, Kyaw Choh 2, Yoshimi Gotoh 2 Mitsuhiro Kubota 1 and Hitoki Matsuda 1 1. Department of Energy Engineering and Science, Nagoya University, Nagoya, Japan 2. Yabashi Industries Co., Ltd., Ogaki, Japan Abstract: To develop highly efficient hydrated lime for cleaning of flue gas from municipal waste incineration is most important for conserving energy and limestone resources as well as reducing solid wastes discharged. In the present study, an attempt was made to prepare highly efficient hydrated lime sorbent for acid gas cleaning by slaking quicklime with various amount of water without chemical additives. In our experiment, highly reactive fine quicklime powder was slaked with 0.65 to 10 times wt/wt of water at 313K, dried and ground. The obtained hydrated lime was evaluated by analyzing the pore volume by N 2 adsorption/desorption, and dry SO 2 adsorption method. The crystallinity of the produced Ca(OH) 2 was evaluated by XRD. As a result, it was found that the hydrated lime prepared could be classified into three groups according to the weight ratio of slaking water to quicklime, namely SWR (Slaking Water Ratio to quicklime). The hydrated lime from i) SWR=0.65 to developed pore volume in the range to 0.11cm 3 /g and SO 2 adsorptivity of to 22.15%, ii) SWR=0.85 to 1.0 developed the largest range pore volume, 3 to 0.133cm 3 /g, and showed the highest SO 2 adsorptivity of to 24.00%, and iii) SWR=1.4 to 10 developed pore volume to 0.11 cm 3 /g and SO 2 adsorptivity of to 21.84%. The XRD intensity of the hydrated lime was increased with SWR, indicating that Ca(OH) 2 crystal growth was more liable to be achieved with higher amount of slaking water. From these findings, it was concluded that there was an optimum slaking water ratio to quicklime controlling the pore structure characteristics of calcium hydroxide with high SO 2 adsorptivity. Keywords: Calcium Hydroxide, Hydrated Lime, Slaking Water, SO 2 Adsorptivity, Pore Volume 1. INTRODUCTION Hydrated lime has been one of the important agents for application of acidic medium treatment processes. One significant application where hydrated lime has played a major role is in the area of flue gas treatment. Hydrated lime is usually used on a large scale at municipal waste incineration where large amount of flue gas is produced. The production figure of hydrated lime was 1.13MT and the total usage of lime for flue gas and environmental treatment was 0.84MT in Japan for 2004 [1,2]. In order to preserve global environment for using the lime, which is one of the natural resources available in Japan [3], it is highly required to put into effective use of the lime related products. It has been reported that the reactivity of the lime varies with the source of the limestone and generally increases with its surface area as well as pore volume [4,5]. Moreover, it has been reported that the reactivity of lime is closely associated with the pore structure characteristics [6]. Industrial production of hydrated lime is generally based on reacting quicklime, a CaO rich lime product obtained by calcining limestone, with water. CaO + H 2 O = Ca(OH) 2 ΔH = -65.3kJ/mol (1) The reaction occurs readily and can reach completion within a few minutes. Theoretically, the exothermic heat can heat up the reaction mass to around 973K; the temperature region which is high enough to burn up and impair the reactivity of the produced hydrated lime. In order to avoid this unfavorable event, production of dry hydrated lime powder normally introduces an excess amount of water, around twice the stoichiometric amount, i.e kg of water per kg of quicklime, which will remove the reaction heat by boiling off as steam and keep the temperature of the reaction mass around 373K. The process of reacting quicklime with water is called slaking as well as hydration and the water used in the process is termed slaking water. Most of the hydrated lime is produced according to the method described above which is known as dry hydration/slaking method. A method called semi-dry slaking method had been proposed by using almost equal in weight of slaking water to that of quicklime used for producing hydrated lime much superior in reactivity than that of the hydrated lime produced from dry slaking method [7]. Hydrated lime putty and slurry are produced by using slaking water of three to four times in weight of the quicklime used and the method is known as wet slaking method. In the wet slaking method where slaking water of around ten times in weight of the quicklime used, the produced hydrated lime slurry is termed milk of lime. The produced lime slurry and milk of lime are usually kept under moderate temperature and agitation to mature for improved settling, plasticity and other values. When the slaking medium is steam or air (moisture), the process is termed steam or air slaking respectively. Instead of the stoichiometric term, the amount of slaking water used will be evaluated in terms of the ratio of Corresponding author: H. Matsuda, matsuda@nuce.nagoya-u.ac.jp 825

2 the weight of the slaking water used to the weight of quicklime being slaked and will be termed as SWR (Slaking Water Ratio to quicklime) hereafter in the present paper. In this work, the effect of excess slaking water on the pore structure and acidic gas reactivity of the produced hydrated lime was studied, particularly in the range of 0.65 to 10 of SWR. For lime sorbent, it is known that the HCl adsorptivity is strongly dependent on the surface area and the SO 2 adsorptivity is highly influenced by the pore volume [8,9]. For this reason, pore volume improvement of the prepared hydrated lime was evaluated also by means of dry SO 2 adsorption in an isothermal fixed bed reactor. 2. EXPERIMENT 2.1. CaO material A JIS industrial grade quicklime powder (CaO>95%, 45μm>95%wt, surface area 2.28m 2 /g, pore volume 8cm 3 /g and the reactivity of quicklime with water (t 60 ) 26s) obtained by calcining limestone from Kinsho mountain, Ogaki, was used as a CaO material. A specific time of the reactants to reach 333K, t 60 designates one of chemical reactivities for a quicklime with water. A more small value by second of t 60 indicates a more reactive quicklime. The measurement of t 60 is carried out by adding 150g of quicklime (all particles smaller than 250μm) into 600g of water (initially at 293K and 300rpm agitation) in a 83.5mmφ x 225mmH Dewar flask in accordance with ASTM C A and recording the changes in temperature with time Slaking method The slaking of quicklime with water was carried out in a mixing/ agitating machine, Model C138A, Maruto Manufacturer. First, a predetermined amount of slaking water was added as quickly as possible to 1kg of quicklime, which was placed in the bowl and under agitation. After 15min, the hydrated lime mass was removed and dried in a vacuum dryer, Model Vacuum Oven ADP-21, Yamato, at 393K and -0.1MPa for 12h. The dry hydrated lime of 20g, was ground in an airtight pulverizer, IKA Labortechnik Type M 20 S 8, Janke & Kunkei GmbH & Co. (100V, 50/60Hz, 500W, 20000rpm), for 30s. For SWR of 10, the obtained hydrated lime in a slurry form was allowed to mature at 313K, 100rpm for 0, 48 and 96h and then filtered with a filter paper of 1μm opening to obtain a wet hydrated lime cake which was subsequently put through the same drying and grinding preparations Pore characterization Pore characteristics of the prepared hydrated lime sorbents were measured by means of N 2 adsorption and desorption at 78K using a gas sorption analyzer, NOVA 3000, Quantachrome corporation Dry SO 2 adsorptivity A schematic diagram of the dry SO 2 adsorptivity measurement apparatus is shown in Fig. 1. First, 0.6g of hydrated lime sorbent is evenly placed on the glass filter 33.5mmφ x 100μm opening. The glass filter unit is then set up at the bottom of the reactor, 400mmφ x 500mmH glass tube which has been maintained at 443K by a temperature controlling/ heating arrangement. After allowing 10min for the sample unit to establish the thermal equilibrium with the reactor unit, simulated municipal waste incineration flue gas is introduced into the reactor for 50min. The simulated gas composes of 1%SO %N 2 gas mixture 9100cm 3 and steam 3900cm 3 at STP thereby the total amount of SO 2 in the simulated gas is one half of the stoichiometric amount of Ca(OH) 2 in the sample. And the flow rate of the simulated gas at the sample unit is 8mm.s -1. These parameters are typical operating conditions kept at municipal waste incineration with dry SO 2 adsorption. Then the reactor is purged with 2000cm 3 of N 2 gas at STP for 10min. The gas from the reactor outlet is scrubbed with 400cm 3 of 3.5%H 2 O 2 solution kept at K which is later analyzed for the amount of SO 2 dissolved in it. The adsorptivity of the sample is calculated as: SO 2 adsorptivity = 1/2 x (S S S H )/S S x 100% (2) where S S = amount of SO 2 in the simulated gas S H = amount of SO 2 dissolved in the H 2 O 2 solution N2 gas 1%SO2 99%N2 Gas mixture 443K superheated steam Thermocouple + Temperature Controller Reactor + Heater + Sample unit 3.5%H2O K Gas meter Fig.1 Schematic diagram of the dry SO 2 adsorptivity measuring apparatus 2.5. Crystallinity determination Crystallinity of the samples were studied by packing the sorbent powders in an Al holder using a powder diffractometer with a proportional detector and Cu-Kα radiation of an X-ray diffraction analyzer, XRD 6100, Shimadzu. 3. RESULTS AND DISCUSSION 3.1. Pore structure and SO 2 adsorptivity Surface area, pore volume and SO 2 adsorptivity data of the prepared hydrated lime sorbents in terms of SWR are listed in Table 1. Pore volume distribution curves of the prepared hydrated lime sorbents prepared by using SWR , and are plotted in Figs

3 And the curves for SWR 10 with 0, 48 and 96h maturing are plotted in Fig. 5. For comparison, pore distribution of quicklime is also shown in Fig. 2. It is quite evident from these figures that the maximum pore volumes for all hydrated lime sorbents are located around 18nm of pore radius. Also it can be seen from Table 1 that the pore volume of lime is expanded from 8cm 3 /g to more than cm 3 /g when quicklime is transformed to hydrated lime indicating that slaking with water plays a significantly important role in developing the pore volume of hydrated lime. Hydrated lime sorbents with pore volume of about 0.11cm 3 /g are obtained by using SWR in the range and with much larger pore volume of about 0.2cm 3 /g are obtained by using SWR in the Table 1. Summary of pore values and SO 2 adsorptivity of the hydrated lime sorbents in terms of SWR SWR Surface Area m 2 /g Pore Volume cm 3 /g SO 2 adsorptivity % 0.0(CaO) h h much higher range But pore volume of the hydrated lime sorbents prepared by using SWR higher than 1 are only about 0.1cm 3 /g. This result implies that there exists an optimal amount of slaking water for maximum total pore volume development of the hydrated lime sorbent. From Fig. 2, it can be seen that the pore volume distribution of the hydrated lime sorbents obtained from slaking with SWR of 0.65, and are almost the same; e.g. peak pore volumes Dv(log r) at 18nm pore radius are about 0.09 to 0.1cm 3 /g for all three curves. Fig. 3 shows the pore volume distribution of hydrated lime sorbents obtained by using SWR of 0.85, 0.9 and 1 and it can be seen that the peak pore volumes Dv(log r) at 18nm pore radius are 0.145cm 3 /g for SWR 0.85 and 0.18cm 3 /g for SWR 0.9 and 1. Fig. 4 shows the pore volume distribution of hydrated lime sorbents obtained by using SWR of 1.4, 2 and 10, and it can be seen that the peak pore volumes Dv(log r) at 18nm pore radius are about 0.09 to 0.11cm 3 /g. From Fig. 5, it can be concluded that pore volume slightly improves with maturing. These pore volume distribution figures indicate that hydrated lime sorbent with higher pore volume distribution can be obtained by using SWR in the range and that pore volume development occurs in the pore radius 2 to 50nm range with the highest pore expansion occurring at pore radius 18nm. Emphasizing on the similarity in pore characteristics of the hydrated lime sorbents, the authors classified them into three groups according to the SWR used; group I = SWR , group II = SWR and group III = SWR SWR=0.725 SWR=0.65 SWR=0.775 quicklime Fig. 2 Pore size distribution of hydrated lime sorbent obtained by slaking with SWR= and of quicklime SWR=0.85 SWR=0.9 SWR=1 Fig. 3 Pore size distribution of hydrated lime sorbent obtained by slaking with SWR= SWR=2 SWR=10 SWR= Pore Radius / nm Fig. 4 Pore size distribution of hydrated lime sorbent obtained by slaking with SWR=

4 96h maturing 48h maturing 0h Fig. 5 Pore size distribution of hydrated lime sorbent obtained by slaking with SWR=10 and maturing for 0, 48 and 96h The relations between SWR and its corresponding hydrated lime sorbent s pore volume and SO 2 adsorptivity are plotted in Fig. 6. Both the SO 2 adsorptivity and pore volume show similar relation to SWR; both curves increase in value with the increase in SWR from 0.65 to 0.9, and then decrease in value with the increase in SWR from 0.9 to 1.4 and then stay almost constant in higher SWR regions. This proves that SO 2 adsorptivity and pore volume are almost directly related and are highest for SWR 0.85 to 1 (group II). Although pore volume values of group I and III are almost the same, group I shows a slightly better average SO 2 adsorptivity (refer to Table 1). To explain this event, we have to refer back to Figs. 2 and 4. In these two figures, it can be seen that even though the maximum pore volume distribution peaks at 18nm pore radius for sorbents in group III, SWR=1.4-10, are slightly higher than those for sorbents in group I, SWR= , pore volumes in the pore radius 2 to 10nm of group I are significantly higher than that of group III. This points out that the two groups possess the same amount of pore volume but different pore volume distribution and suggests that pores in the 2-10nm radius might be more effective than pores in the 18-50nm radius for SO 2 adsorption. On the whole, it can be concluded that the total pore volume/ pore size distribution of the hydrated lime sorbent is significantly influenced by the amount of slaking water used and that the hydrated lime sorbent possessing an optimum pore structure for dry SO 2 adsorption was obtained by slaking with SWR= under the present experimental conditions XRD analyses The XRD graphs of the hydrated lime sorbents prepared by using SWR of 0.65, 1 and 10 are shown in Fig. 7. It can be seen that the intensity of the peaks increases with the increase in SWR; e.g. the corresponding intensities at 18.1º and 34.09º (2θ) for SWR 0.65 are 1400 and 2000cps, for SWR=1 are 2250 and 3000cps, and finally for SWR=10 are 3600 and 4000cps respectively. For the hydrated lime sorbent prepared by slaking with SWR of 10, no significant change in intensities is observed with respect to maturing. It can be concluded that higher amount of slaking water promotes the crystalline development of the hydrated lime Slaking Water Ratio Fig. 6 The relationship between SWR and its corresponding hydrated lime sorbent s pore volume and SO 2 adsorptivity Intensity / Cps(Count per second) -1 3 g Pore Vol / cm I II III SWR=10 SWR=1 SWR=0.65 SO 2 adsorptivity Pore Volume ~~ ~ ~~ ~ ~~ ~ ~~ ~ θ (Cu Kα) Fig. 7 XRD patterns of the hydrated lime sorbents prepared by slaking with SWR 0.65, 1 and CONCLUSION Pore structure and SO 2 adsorptivity of the hydrated lime sorbents were significantly influenced by the amount of slaking water used. The hydrated lime sorbents could be classified roughly into three groups according to SWR; the ratio of the amount of slaking water used to the amount of quicklime being slaked: I. Low SWR, from 0.65 to II. Medium SWR, from 0.85 to 1 III. High SWR, from 1.4 to 10 The hydrated lime sorbents of group II, obtained by slaking with medium SWR possessed the highest pore volumes and exhibited the highest SO 2 adsorptivity. The sorbents of group III, obtained by slaking with higher SO2 adsorptivity / % 828

5 SWR showed that the pores in 2 to 10nm radius range were comparatively small and that the SO 2 adsorption performances were in the lowest range. XRD analysis showed that the peak intensity of the sorbent increases with the increase in slaking water used, implying that the growth of the hydrated lime is largely affected by water and is more advanced with larger amount of water. It was found that slaking water promotes the development of pore volume/ structure of the hydrated lime and the crystalline development of the calcium hydroxide, and that there exists an optimal amount of slaking water, with which the produced hydrated lime sorbent would possess high pore volume with pore structure appropriate for maximum SO 2 adsorptivity. REFERENCES 1. Japan Lime Association, JLA production statistics (2004). 2. Japan Lime Association, The Environmental Measure Report of Japan Lime Association Action Plan for Global Warming (2003), Lime, 541 (2004), pp M. M. Miller, U. S. Geological Survey, Mineral Commodity Summaries, January (2005), pp R. S. Boynton, Chemistry and Technology of Lime and Limestone, John Wiley & Sons, (1980), p J. A. H. Oates, Lime and Limestone: Chemistry and Technology, Production and Uses, Wiley-VCH (1998), p D. Beruto, L. Barco, A. W. Searcy and G. Spinolo, J. A. Cerm. Soc., 63 (1980), pp Japanese Patent, P (2004). 8. E. Sasaoka, Md. A. Uddin and S. Nojima, Ind. Eng. Chem. Res., 36 (1997), pp N. Mihara, S. Okumura, S. Ozawa, Y. Kojima, H. Matsuda, K. Kyaw, T. Iwashita and Y. Goto, J. Soc. Inorg. Mater. Japan, 11 (2004), pp