Polielektrolitlerin borik asit kristalizasyonuna etkisi

Abstract

Son 30 yıldır yapılan kristalizasyon çalışmalarının temel amaçlarından biri, kristalizasyona etki etki eden tüm değişkelerin belirlenmesi ve etki mekanizmalarının ortaya çıkarılmasıdır. Bu çalışmada, bugüne kadar yapılan kristalizasyon çalışmalarında göz ardı edilen, ancak, oldukça önemli bir değişken olan kristal yüzey şarjının, kristalizasyon mekanizmasına etkisi, borik asit için incelenmiştir. Çalışmada, üretilen veya aşı kristali olarak kullanılan borik asitin, yüzey şarjını değiştirmek amacı ile, yüksek molekül ağırlığına sahip anyonik, noniyonik ve katyonik polielektrolitler, katkı maddesi olarak seçilmiştir. Deneylerde, günümüz kristalizasyon çalışmalarında, kristal büyüme ve çözünme hızlarının belirlenmesinde, temel ölçüm teknikleri olarak kabul edilen akışkan yatak hücresi, MSMPR tipi kristalizör ve tek kristal hücresinden faydalanılmıştır. Polielektrolitlerin, borik asit kristalizasyonu üzerine olan gerçek etkisini belirlemek açısından, birer değişken olan, viskozite, pH, zeta potansiyeli değişimleri incelenmiştir. Yukarıda ifade edilen temel ölçüm sistemlerinden olan, akışkan yatak büyüme hücresinde yapılan deneylerde, kristal yüzey kalitesinin ve polielektrolitlerin, büyüme ve çözünme hızı üzerinde olan etkileri gösterilmiştir. Akışkan yatak sisteminde elde edilen sonuçlar doğrultusunda, MSMPR kristalizasyonunda sadece noniyonik ve katyonik polielektrolitler esas alınmıştır. Söz konusu bu deneylerde, ortalama tane boyutunun, nükleasyon ve görünür büyüme hızının kristal yüzey şarjına olan bağımlılığı belirlenmiştir. Ayrıca, polielektrolitlerin kristal ürünün şekline olan etkileri fotoğraflanarak değerlendirilmiştir. Öte yandan, akışkan yatak hücresinde belli tane boyutuna sahip, aşı kristallerinin büyüme ve çözünme davranımlarını, MSMPR sisteminde de nüuklei kademesinden gelişen büyüme adımlarını daha yakından takip edebilmek amacı ile tek kristal deneyleri yapılmıştır. Söz konusu deneyler, durgun çözelti ortamında kantitatif ve kalitatif olmak üzere iki şekilde yürütülmüştür. Çalışmanın sonucunda kristal yüzey şarjının, kristalizasyon mekanizmasını etkilediği, borik asit kristalizasyonu için ispatlanmıştır. Ayrıca, literatürde saf sizlik etkisi, büyümede saçılma, tercihli adsorbsiyon olarak geçen kavramların yüzey şarjı ile açıklanabileceği gösterilmiştir.

The most important parameters affecting the crystal size distribution in industrial crystallization are nucleation and growth rate mechanism. The production of large and homogeneous particles is very important in many industrial application. These two mechanisms were found to be affected by the presence of impurities. Therefore, it is possible to control particle size distribution by impurities. Boric acid produced in industrial crystallizers shows very broad range of crystal size distribution (CSD), very poor washing characteristic, effective dust problem in drier, caking density and high mother liquid content of centrifuged product and therefore high level of impurities than normally expected from CSD [66]. In order to solve these problems, behavior of crystal surface in impure medium sould be well known. In this study, the effect of the polyelectrolytes in different charge densities on the growth and dissolution kinetics of boric acid was investigated. The reason of choosing the polyelectrolytes as additive is the fact that boric acid crystals show strong surface charge which is important parameter in crystal growth. Polyelectrolytes used are selected ranging from strongly anionic to strongly cationic in order to understand the charge neutralization effect on crystal growth rates. Polyelectrolytes are expected to interact with the charged sites of surfaces, to modify the crystal habit and to change the crystal growth rate. In this study, commercial well-known polyelectrolytes shown in Table 6-1 are used. Polyelectrolytes are water-soluble organic compounds and they have high molecular weights. Therefore the viscosity naturally increases with increasing concentration of polyelectrolytes in solution. Since viscosity is one of the affecting parameters in diffusion phenomena, effects of the polyelectrolytes concentration on viscosity in boric acid solutions are determined. Results are presented in Fig.6-la and Fig.6-lb. IX Polyelectrolyte concentrations were determined by preliminary studies by measuring the zeta potentials of boric acid crystals in saturated solution. This value was found "-14 mV" and it was changed with the concentration of the polyelectrolytes. Changing of zeta potentials of boric acid crystal are plotted in Fig. 6-5 and Fig. 6-6. The experiments were carried out in the three different measurement techniques. These were fluidized bed, Mixed Suspension Mixed Produt Removal (MSMPR) type crystallizer and single crystal growth cell. The laboratory scale fluidized bed crystallizer shown in Fig. 5-1 is used. Solution was pumped by magnetic centrifugal pump was set as closly as possible to the settling velocity of the seed crystal and this was controled by electronic rotameter. When the solution temperature become steady, chambers were changed as short time as possible. The crystals were allowed to grow or dissolve for about twelve minutes. At the end of each run the grown or dissolved crystals were removed, filtered and dried. After drying, crystals are weighed and the growth rates were calculated by known method which was described in section 5.1.1. On the other hand in order to explain real effects of the polyelectrolytes, viscosity has been considered. In the experimental study two different type boric acid crysalts were used. One of these was perfect, and the other was normal crystal. Perfect crystals were prepared in fluidization condition in the fluidized bed by dissolving the >710 /xm particles to -500+400jum. Boric acid grows in pure solution dendritically at all supersaturation levels. This behavior results from the charge distribution and from the dislocation points seen on normal crystals. Therefore, in order to see the effect of dislocation points, crystal growth rates of perfect crystals are also measured. Fig. 6-9 shows that growth rates of normal and perfect boric acid crystals as a function of supersaturation. As it can be seen from this figure, perfect crystals have higher growth rates than normal crystals. This can be explained by the breaking of dendritic part of growing crystals during the operation. Effects of polyelectrolytes on the growth and the dissolution rates of boric acid crystals are given in Appendix B. As it can be seen from these figures, each polyelectrolyte has a different effect on the growth and the dissolution rate of the boric acid. Results are given as follows: - In the presence of anionic polyelectrolytes growth rate depressed and dissolution rate increased by increasing the concentration of polyelectroytes. Crystals grow in stongly dendritic shape and it causes formation of secondary nuclei. x - In the presence of nonionic polyectrolytes growth rate depressed and dissolution rate partially increased by increasing the concentration of polyelectrolytes. Dentritic growth was less when nonionic polyelectrolytes were used. In the presence of cationic polyelectrolytes different behaviors were observed. Cationic polyelectrolytes which have weak charge density prevent the dendritic growth. On the other hand, solution viscosity has definite effect on growth and dissolution rates. But increasing of dissolution rate in the presence of polyelectrolytes are not explainable by viscosity effect. Therefore it is very probable that the surface charge is the most important in the growth rate of boric acid and it can be controlled by polyelectrolytes which can neutralize the surface charge. Fluidized bed experiments show that polyelectrolytes have an important effect on the growth and dissolution kinetics of boric acid. The relationship between the zeta potential and the growth rate in a constant super- saturation are plotted in Fig. 6-32. As it can be seen very clearly from this figure, the growth rate of boric acid was strongly dependent on the zeta potential. It means, growth rate can be control by using polyelectrolytes. On the other hand, continuous crystallization experiments give the possibility to see the effect of the polyelectrolytes on the particle size distribution. For this purpose, MSMPR type crystallizer were selected. Because this distribution can be best characterized by population balance theory which is proved for Mixed Suspension Mixed Product Removal (MSMPR) type crystallizer. Experimental set up shown in Fig. 5-6 was used. The MSMPR type crystallizer was made of cylindrical plexiglas and contained three flow breakers in it. In order to keep constant the inside temperature of crystallizer, a spiral (10 cm in a diameter, 10 cm in height) made from 316 stainless steel was provided at the center of crystal lizer. This spiral acts either as a heat exchanger or as a draft tube. Crystallizer temperature was controlled by a contact thermometer which was calibrated with ASTM thermometer and by control units. Feed solution was pumped from 25 1 closed tank which was jacketed with steam and made from 316 stainless steel. The temperature was controled by a contact thermometer and by a magnetic valve linked to the steam line. In spite of the continuous feeding, the suspension withdrawal is made discontinuously as usual. This ensures isokinetic withdrawal and prevent the particle classification in the outlet. For this reason 10 % increase in volume of crystallizer is permitted and this excess is sucked by using vacuum. XI Boric acid solution used in experiments is prepared from GR product of Merck company. Crystallization experiments are conducted at a constant residence time and concentration. Characteristic samples were withdrawn after eight residence times and fed to a measuring flask. After measurement of volume of suspension, chemical and sieve analysis were carried out. MSMPR experiments are made by adding the polyelectrolytes to feeding solution at three different concentrations to see the individual effects of polyelectrolytes on boric acid crystal growth rate, nucleation rate, average particle size and the crystal shape. The results are given as follows: - There is a strong relation between the zeta potential and the growth rate, nucleation rate, average particle size (Fig. 6-45, 6-46, 6-47) - The average particle size of boric acid increases the vicinity of zero point of charge. In the range of 0+5 mV, crystals are strongly agglomerated. In the range of +5+15 mV crystals grow without excessive agglomeration. - In the range of +15+25 mV the average size of crystals are decreased and growth defects are observed. In the presence of N100, F04115 and CatFlocDL polyelectrolytes agglomerates are obsorved. This agglomeration is seen especially below 250jum particle size. - EM949 emulsion type cationic polielectrolyte have a specific effect on the boric acid crystal shape. It is found that strong dentrite formation are provided when the polyelectrolyte concentration increase from 10 ppm to 100 ppm. In this study, single crystal measurement technique was also conducted. For this aim, aparatus shown in Fig. 5-7 was used. In order to explain the real effect of poly electrolytes in fluidized bed and MSMPR experiments qualitative and quantitative experiments were made in single crystal growth cell. For this purpose, N100, F04115, WT2479 and CatFlocDL polyelectrolytes that give different behaviours in continuous crystallization were selected. In the qualitative experiments seed crystal having 450jum particle size was used. XII It is found that in the presence of N100 and WT2479 growth rate is increased but in the presence of F04115 and CatFlocDL growth rate is decreased. These results do not show the real effect of the polyelectrolytes. Because the growth behaviour of crystal in the presence of poly electrolytes is found different. In order to measure the crystal growth rate, microscobic projective area of crystal is used. Because of measurement technique dendritic grown crystals are show higher growth rate whereas compact grown crystals show lower growth rate. Therefore results are not comparable with each other. Qualitative experiments were carried out in the single crystal cell. For this purpose, saturated boric acid solution was filtered from 0.2/xm membrane filter and was added to the growth cell. By controlled cooling of solution small nuclei are produced and growth behaviour was observed. The results are: In the pure solution starting nuclei had hexagonal shape but during the growth this shape changed and directed to strongly dentritic crystal formation. In the presence of F04115 polyelectrolyte starting nuclei are in hexagonal and rod shape. During the growth new growth layers are grown parallel to first layer but in the slightly sliding shape. When the nuclei become to three dimensional crystals, faces of crystals grow in different velocity and habit of crystal changed. At the end, crytals from nuclei grown very compactly and have a prismatic shape. Dendritic growth are not observed. In the presence of N100 polyelectrolyte, starting nuclei are in the shapes of hexagonal and rod. In the beginning of growth, nuclei grow as in the presence of F04115. But after a period lots of crystal defects are produced on the crystal. At the end of the experiment, crytals from nuclei grow having a defected prismatic shape. Also, agglomeration occurs. In the presence of WT2479 polyelectrolytes, starting nuclei are in the shapes of hexagonal and rod. During the growth, highly crystal defects are produced. As a result of these, crystal lost its prismatic shape. There are obviously a number of different parameters effecting the crystallization mechanism. Understanding the mechanism will help us to control the crystallization operations. As a conclusion, in this study the effect of surface charge as a parameter on the crystallization kinetics is shown and also the presence of this parameter is confirmed by using three different growth rate measurement techniques. xm The suggestion reached by this work explaines the impurity effect, size dependent growth and growth dispersion. It is also found that every particle has its own surface charge and there is a charge distribution in the crystal state. Although the surface charge of particles of different size can not be measured, it is belived that the surface charge distribution should be valid in all size ranges.