Search results

The presence of polysaccharides in cane and beet raw juices causes several negative effects during the sugar manufacture. These are usually mitigated by enzymatic decomposition of dextrans. Such effects not only depend on the content, but also on the molecular mass distribution. This means that the different dextran fractions specifically affect the process. An accurate process control hence requires the most precise knowledge about the existing content and the molecular mass distribution present. A detailed understanding of the specific processing problems and also a targeted enzyme application hence requires the determination of a total dextran content and also its characterization including the differentiation between the different dextran fractions. An accurate analytical tool which equally satisfies industrial applicability is still lacking. To improve on this situation, two new approaches for the determination of dextran were developed and benchmarked against the commonly used and established Haze Method, which is rather inaccurate and also sensitive to molecular mass variation. The two new approaches are both based on polarimetry. These two methods indicate to be superior over the Haze Method with respect two molecular mass variation and hence enable the determination of a broader molecular size range including also low molecular mass dextrans.

This manuscript covers two aspects of the subject area described in the title. In order to improve the understanding and control of sucrose solubility in molasses it is believed that it is necessary to firstly gather experimentally detailed and consistent solubility and composition data. Secondly, the mathematical description of sucrose solubilities in molasses should be reconsidered. In the contribution data on 49 different molasses are described. The data show significant variation in both the sucrose solubility and the composition of the nonsucrose components. Current models to describe the solubility are discussed in light of thermodynamical considerations and their success in representing the data gathered.

The effects of high and low molecular mass dextran (T2000 and T40) on the size and shape of particles precipitated during carbonatation and their correlation with filtration performances were key to this study. Varying contents of T2000 and T40 dextran in sugar solutions corresponding to DS contents of thin juice were investigated. For particle size and shape analysis, static image analysis and laser particle size analysis were used. Both methods, static image analysis and laser diffraction, revealed that the presence of T2000 and T40 dextran leads to a higher amount of large-sized particles at the expense of small-sized particles, indicating pronounced agglomeration. The additional evaluation of shape parameters (circularity, roundness, solidity) obtained from static image analysis indicates that the agglomeration is oriented in the absence and in the presence of lower T40 dextran levels. Besides, non-oriented agglomeration, resulting in more round agglomerates with smoother surfaces, was found for samples loaded with T2000 dextran and high T40 dextran levels. Only the latter samples have shown to negatively affect the filtration performance. Thus, in the presence of T2000 dextran and high T40 dextran levels, the filtration was hampered. This appears to be mainly caused by a tighter packing of more round calcium carbonate agglomerates in the porous structure of the filter cake.

In the production of sugar from mixed syrups containing feedstock from beet and cane origin the non-sugar fraction plays an equally important role as in processing of material from a single source. However, the consequences of the resulting mixed non-sugar fraction on processes such as impurity, mainly colorants, incorporation and chemical reactions have not been studied yet. In this contribution the results of a study elucidating the kinetics of colorants formation in mixed syrups on thermal stress are given. It is found that the relative increase of syrup color value with respect to the starting value behaves quite systematic for wide range of mixed syrups. In all syrups the color values increase linearly with exposure time. The rate of increase is the highest for beet syrups. The generation rate increases with temperature in line with Arrhenius' law. Different to earlier work no promoting effect of increased dry matter contents could be found. Furthermore, a model, assuming a first order reaction, to predict the color generation in mixed syrups based on the information of the pure syrups was formulated. This model is in essence generating data that match the experimental data within the range of the accuracy of the color value determination method. In conclusion the data suggest that there is no excess color generation on thermal stress due mixing of the non-sugar materials from different sources.

Following the changes in legislation the situation for sugar production in Europe is going to change in the near future. Due to market liberalization the option to co-process raw materials from beet and cane origin becomes more attractive. To this end the crystallization of sugar from blended syrups was studied with particular emphasis on the color inclusion. Evaporation crystallization was performed for various blended syrups and the feed syrups and resulting sugar were characterized in detail. The crystallization is performed at practically constant supersaturation followed by centrifugation and drying. It is found that the colorants from the different origin mutually suppress their inclusion or adhesion. For the industrial raw materials based system it is indicated that color values of less than 30 IU can be achieved with levels of up to 40% of cane material. Based on the experimental finding a simple relation to predict the color of sugars crystallized from blended syrups is formulated. The input data to the model are the color values of both the two pure feed syrups mixed and their respective sugars.

The industrial production of sugar is structured such that the development of process innovations is difficult. The evaluation of ideas directly at full scale is risky and costly. Testing basic feasibility at small scale is hence desirable. To this end a pilot scale crystallizer and adequate processing developed is described here. The evaluation covers next to basic quality parameters also the comparison of computed online data with laboratory analysis on samples drawn. The designed evaporating crystallization appeared to be well controlled. At constant supersaturation approximately 50g/100 g of crystals were produced reliably. It turned out that next to the actual well controlled crystallization process the separation and drying processes are very important to produce relevant data. Centrifugation at high centrifugal forces (G=1450) for 120 s with dosage of plain water (2.5g/100g massecuite) as washing fluid yielded best results. Drying in a fluidized bed at temperatures of 343K for 10min was superior to drying at open air or in climate cabinets, this with respect to formation both of agglomerates and fines. It is shown that the designed process allows to crystallize sucrose in small scale with industrially relevant characteristics of process and product.

The effect of dextran's molecular mass distribution on the sucrose crystal shape was key to this study. Therefore, sucrose crystals were produced by evaporating crystallization experiments using synthetic thick juices in the form of pure sugar syrups containing high (T2000) and low (T40) molecular mass dextran fractions as well as enzymatically decomposed dextran. The combined analysis of molecular mass distributions by size exclusion chromatography and sucrose crystal shapes by static image analysis were used to identify the least harmful reaction products resulting from the enzymatic decomposition of dextran. The combined evaluation of two shape parameters, circularity and width-to-length ratio, has shown that three different shape modifications can be related to the presence of dextran, namely cube-shaped crystals, elongated needle-shaped crystals and agglomerates. In the main, the data indicated that high T2000 contents and generally all T40 dextran contents led to an increased occurrence of agglomerated and occasionally elongated crystals. The latter was especially found for high T2000 dextran contents. In contrast, low T2000 dextran contents predominantly increased the amount of cube-like crystals. The enzymatic decomposition of dextran resulted in a gradual reduction of the molecular mass. It was shown that an insufficient decomposition to broadly distributed low molecular mass dextran fragments, which are realistic to assume for technical cane and beet juices, still dramatically affected the sucrose crystal shape. Once dextran was decomposed to molecules with molecular masses of less than 5 kDa, no dextran-related effects on the sucrose crystal shape were found.