In the field of industrial crystallization, the control of crystal size distribution and purity is the most essential aspect, and has been investigated extensively. However, most of the previous studies were carried out on the final products, in other words, they were in a sense indirect results. It is therefore appropriate to observe the processes of crystallization so that how differently treated crystals grow can be determined under given solution conditions. It is the purpose of this study to investigate the crystallization process of NaCl on differently treated crystals by the in situ observation method under the optical microscope. From the observation results, it was found that the crystal surfaces were observed to have a large number of tiny rectangular columns originating from detached pieces which adhered to the crystal surface, with many inclusions and cracks, and these surface conditions always changed during growth under different solution conditions. Also we obtained the linear crystal growth rate and the crystal growth step velocity as follows, respectively, GL=7.1ΔC1.47, GS=3.4×102ΔC2.32.
Semibatch reactive crystallization of basic magnesium carbonate (BMC) and magnesium carbonate trihydrate (MCT) was conducted to clarify the mechanism of reactive crystallization of BMC and MCT produced by precipitation from magnesium hydroxide with carbon dioxide gas. The production regions of BMC and MCT were partitioned by the reaction temperature and the initial concentration of magnesium hydroxide. The reactive crystallization processes of BMC and MCT were confirmed by the experimental results of SEM photographs and X-ray diffraction analysis.
Seed crystals for which the size distribution is expressed by the Rosin-Rammler equation are assumed to grow by the ΔL law and changes in the uniformity number of size distribution are correlated with the growth increment of crystals from the uniformity number and size of seed crystals. The correlation is discussed based on the ratio of nuclei subsequently generated to that of the original crystals and test data on precipitation of lead sulfate were found to be satisfactory compared with the results estimated by the proposed correlation.
Precipitation of calcite has been studied by addition of calcium chloride solution into a mixed solution of sodium carbonate and bicarbonate. In this study, a calcium carbonate intermediate quickly appeared and was transformed to calcite. The number of calcite crystals produced in the solution was proportional to the amount of sodium bicarbonate, and the average growth rate of calcite crystals was 0.67μm/min.
One of four crystallizeres in our salt plant is the Oslo type producing large particle size and the others are the slurry circulation type producing middle or small particle size. In these particles, there are small cavities filled with mother liquor. The cavity ratio of product produced by the Oslo-type crystallizer is much higher than that of other products. As the first step in determining the reason for this, we studied the relation between the cavity ratio of products and operating conditions, which are epresented by growth rate and particle size. Using the Oslo-type crystallizer, test runs were done in two type of operations, stationary and nonstationary. The cavity ratio of the resulting products obviously showed particle size dependency, so we called it the “average cavity ratio” and gave the name of “cavity ratio” to the differential quotient of average cavity ratio by particle size, which corresponds to the cavity ratio of the thin surface layer of the particle. In the nonstationary mode, particle size distribution (PSD) was very sharp so that the growth rate of each size of particle (G(L)) could appear to be the same as the average growth rate (Gav). But in the stationary operation, PSD was wide and G(L) could not be determined from Gav. Therefore, as a temporary parameter in this case, Gav was substituted for G(L). In each mode, the cavity ratio data were expressed as a function of particle size and growth rate. For both types of operations trends were similar but not identical. Under the assumption that this difference in trends can be fully explained by the substitution of Gav for G(L), the relation between G(L) and Gavin the stationary mode was derived from the difference in trends of the cavity ratio. In both modes, each one cavity ratio corresponds to the smallest particle size sample was deviated from the trend of the other data. These deviations are considered to be affected too much by the carier of growth in the circulating slurry.
Particle size control procedures for the salt manufacturing process were established by utilizing the R-R chart and Toyokura's empirical equations. The results were obtained from an existing evaporating crystallizer designed for the production of small-sized of salt. It is evident that the suitable setting of the operating conditions such as the solid suspension concentration and evaporating rate give good results in manufacturing different sizes of salt using the same equipment. Further analysis of the data will yield general concepts for the design of an evaporating crystallizer to produce the optional salt-size.
At present, 3-5mm particle size salt is produced by an evaporator crystal classifying crystallizer (OSLO type) or a conical crystal classifying crystallizer. Only a few technical reports or documents are available on salt production systems yielding particle sizes larger than 3-5 mm, such as an old test report by the HOUFU Salt Manufacturing Laboratory. The reasons for the lack of research and development by companies in this field are: 1) technical difficulties; 2) production costs; and 3) low demand. Furthermore, problems exist in product quality (shape, purity), which is the most important aim of the crystallization process generally. Satisfactory results on the test production of super large particle size salt were obtained, and 15mm diameter salt was obtained which was larger than the 10 mm size targeted initially.
In industrial sodium chloride crystallizers, the oscillation phenomena of the crystal size distribution is known even under constant operational conditions. Crystal hardness is also considered to oscillate, accompanied by crystal size distribution oscillation. In order to study the correlation between crystal hardness and operational factors, the hardness of crystals produced under the constant operational condition was measured using an industrial crystallizer with an external heat exchanger. Crystal hardness was found to change with the change in crystal size distribution. Also, crystal hardness has a strong correlation with the change in slurry concentration, and crystal became harder when the slurry concentration was low and the average particle size of salt nears its maximum. Furthermore, crystal hardness is also connected with the nucleation rate and average values of crystal growth rate, which are calculated by Toyokura's model. The smaller the nucleation rate, and the higher the crystal growth rate, the harder the crystal. We examined crystal hardness from the aspect of crystal internal structure, and observed that crystal hardness was related to the crystal void fraction. The void fraction of crystals increased when the particle size of salt increased, and the fraction began to increase more steeply when the particle size reached 800-900 μm. This may be explained by a change in crystal surface conditions, as observed using electron microscopic photographs.
Generally speaking, it is not easy to produce hard, cubical, large-sized salt crystals in the mother liquor concentration process. We studied the characteristics of salt crystallization in this process using Toyokura's equation for the crystal nucleation rate and the crystal growth rate, taking microscopic photographs, and measuring crystal compressive strength. Under the conditions within the range of 0.005-0.020 m3/m3/h of production rate and of circa 33°Bé of liquid density in the evaporating chamber, grown crystals converged to the maximum size (e. g., 500μm) within 6 hours in that chamber. It was confirmed that maximum crystal size became smaller with higher liquid density. This limitation of crystal growth is considered to be caused by impurities. The salt obtained in this process was found to be weaker in compressive strength and to contain more coagulated or irregularly shaped crystals compared to common evaporated salt.
Using a well-mixed type and cone-shaped continuous classified bed-type crystallizer, we studied the effect of classification through a crystallizer and a salt leg without operation, and the influence of the crystallization rate, suspension density, growth rate, and nucleation rate on the crystal size distribution with a common crystallizer. Based on these results, we verified the results of narrowing the crystal size distribution experimentally. To calculate the crystal size distribution of products, we used the cumulative amount obtained using the Rosin-Rammler method for practical use, and investigated the operational conditions for narrowing the crystal size distribution, which was characterized by enlarged n. We also calculated the growth rate and the nucleation rate using Toyokura's design chart.
For the purpose of antiblocking of salt in crude salt, we conducted slurry classification tests using a classification-type evaporator. In the classification test, the mother liquid flowed from the bottom to the top of the apparatus. It follows that: 1) microcrystals of salt products are reduced in size by about half; and 2) the average diameter of crystals tends to become larger by adding seed crystals, and by holding salt crystals longer in the apparatus.
Sodium chloride seed crystal was prepared in a stationary solution to yield asmooth surface, and was placed in a stream of supercooled solution having imaginary NaCl nuclei. In each test, seed crystal barely grew during the period correlated with supersaturation, but its surface was observed to become rough. After it began to grow, the average growth rate was obtained from increment of crystal seed size for the operational period. The observed crystal growth rate increased with increases in the populations of nuclei supposed to be suspended in supercooled solution. Crystal growth rate is discussed considering two processes: formation of surface roughness and becoming smooth with the increase in crystal size.
Fine crystals appearing on the surface of fixed sodium chloride crystals growing in suspension revealed two kinds of behavior. Some of the fine crystals bury into the seed crystals, while the others develop on the surface of seed crystals.