Note
Effect of Initial Temperature and Slurry Density on Stable Crystallization Process of Xylitol Melt Containing Sorbitol
2020 Volume 26 Issue 2 Pages 235-238
Details
2020 Volume 26 Issue 2 Pages 235-238
There are many sugar products in foods, and it is desirable to satisfy both product quality and process operability. One example of a sugar product is fondant with crystalline material containing xylitol, and it is difficult to formulate products with desired qualities such as crystallinity. The production process of crystalline material can be regarded as melt crystallization. The objective of this study was to investigate the relationship between the condition of melt before depositing process and the crystallization phenomena for xylitol-sorbitol system. The results showed that the crystallization process after depositing was affected by the initial slurry density. The Avrami equation was applied to this system to analyze the mechanism of melt crystallization. It was confirmed that the crystallization process can be analyzed and crystallinity can be predicted by the Avrami equation. Melt with high slurry density and high temperature is preferable for formulation of desired crystal qualities.
There are many crystalline food products, for example, chocolates, ice cream and foods with sugar such as fondant. It is desirable to satisfy both product quality and operability in the production of these crystalline foods.
From the viewpoint of crystallization, the production process of these foods can be regarded as a melt crystallization process. One example of a crystalline food is fondant containing xylitol crystal. The production method of this fondant is simple. First, the raw material is heated until molten. Then, the melt is poured into molds and cooled. After standing, it hardens into the desired product. When the raw material is pure xylitol, crystallization is certain. However, an abrupt increase in viscosity can occur simultaneously, which causes undesirable operability. In other words, when pure xylitol is used, the crystallinity of the product quality is satisfied but the viscosity as an operability quality is not satisfied. In order to overcome undesirable operability (the abrupt increase in melt viscosity), sorbitol is mixed with xylitol at a certain composition. However, in this case, crystallization of xylitol is less likely to occur and the crystallinity quality of the final product is not satisfied or stable.
There have been several studies investigating crystallization of xylitol from alcohol solution since the year 2000 (e.g. Hao, 2006; Sampaio, 2006; Martínez, 2007, 2008; Canilha, 2008). These previous studies have reported nucleation rates or growth rates for xylitol crystal. However, these studies focused on solution crystallization as a separation technique for xylitol from raw materials (Sampaio, 2006; Martínez, 2007; Canilha, 2008) or purification from solution (Martínez, 2008; Hao, 2006). Conditions such as viscosity and composition of the raw material are very different between solution crystallization and melt crystallization for xylitol. For melt crystallization of xylitol, there is one report on crystal growth rate from melt containing methanol (Seppälä, 2010). The situation of xylitol crystallization from melt containing methanol differs from that of melt containing sorbitol because methanol is liquid at room temperature and complete solidification cannot be achieved. Therefore, the crystallization condition is very different from the crystallization from melt containing sorbitol. For successful crystallization of xylitol from melt containing sorbitol, it is necessary to understand the kinetics of the crystallization process. In this study, the authors focused on the relationship between the melt state before depositing process and the crystallization process phenomena after depositing. Factors such as slurry density, temperature, and melt composition can be considered as the melt state before depositing. The objective of this study was to investigate the relationship between the melt state before depositing and the crystallization process phenomena of xylitol after depositing for the xylitol-sorbitol system.
Materials Xylitol and sorbitol were purchased from Wako Pure Chemical Industries, Ltd.
Design of experimental conditions In order to investigate the crystallization process (phenomena) of xylitol after depositing, cooling crystallization was carried out for five kinds of melt with different states before depositing in terms of slurry density, overall composition and temperature (Runs 1 to 5). Slurry density was adjusted by using seed crystals of xylitol. The conditions are summarized in Table 1. The design of the initial slurry density MT0, and the amount of excess xylitol in the slurry to equilibrium melt was based on solid-liquid equilibrium data from a previous study by Perkkalainen et al. (1995). The amount of excess xylitol in slurry to equilibrium melt was defined as the difference of overall xylitol composition and equilibrium xylitol composition of slurry at T0.
| Run No. |
Initial temperature T0 [K] |
Initial slurry density MT0 [mol-crystal/mol-slurry] |
Overall composition of slurry [mole fraction] |
Excess xylitol in slurry [mole fraction] |
|
|---|---|---|---|---|---|
| Xylitol | Sorbitol | ||||
| 1 | 361 | 0 (No seed) | 0.80 | 0.20 | 0.02 |
| 2 | 361 | 0.26 | 0.84 | 0.16 | 0.05 |
| 3 | 358 | 0.26 | 0.80 | 0.20 | 0.07 |
| 4 | 355 | 0.26 | 0.74 | 0.26 | 0.09 |
| 5 | 355 | 0.44 | 0.80 | 0.20 | 0.15 |
Experimental procedure Sample melts were prepared by melting a solid mixture of xylitol and sorbitol at 120 °C (393 K). Then, the temperature was lowered to the predetermined value (Equilibrium temperature, refer to Table 1), and a fine powder of xylitol was added to the melt as seed crystals. After seeding, the temperature was kept at the predetermined value until the viscosity of the melt became constant. Viscosity during preparation of the sample melt was measured with a viscometer (Toki Sangyo Co., Ltd, TVB-15). Temperature control of the melt was carried out by using a thermostat bath (Tokyo Rikakikai Co., Ltd., NTB-221).
Before the depositing operation, the melt sample was agitated at 500 rpm with a six-blade turbine impeller. When the sample viscosity became constant, the sample was poured into a petri dish at 25 °C, and the dish was placed into an incubator at 25 °C.
X-ray diffraction (XRD) measurement was carried out every hour for 4 h after sampling. For the XRD measurement, MiniFlex (Rigaku, Co., Ltd., Cu Kα radiation) was used.
Analysis of crystallization process The peak intensity of XRD indicates the crystallinity of the sample. The change in crystallinity shows the change in the crystallization process. Therefore, in this study, the intensity of this peak was focused on to analyze the crystallization process. The Avrami equation (Eq. 1) was used to analyze the crystallization process.
 |
where “y” is the volume fraction of the transformed phase or crystallinity. “k” is the constant of crystallization rate and “n” is the Avrami exponential. In the present study, “y” was defined as the ratio of the diffraction intensity of the sample melt after depositing to the initial melt as the quantity corresponding to the transformed crystal phase from the amorphous phase, and the equation was modified to (Eq. 2)
 |
where “a” is the coefficient related to the final intensity ratio. The coefficient “a” shows the ratio of the peak intensity corresponding to the ratio of the total deposition amount to the initial slurry density including background intensity.
Effect of presence of solid phase For Run 1, the melt was almost clear although a thin film was observed on the air-melt interface after 4 h. On the other hand, for Run 2, there were many crystals in the sample melt after 4 h. XRD peak patterns of samples after depositing for Run 1 are shown in Fig. 1. The peak near 2θ = 19° can be regarded as the characteristic peak of xylitol crystal based on a previous study (Kim, 1969). Therefore, the diffraction intensity of the peak near 2θ = 19° was used to estimate the ratio of diffraction intensity in the present study.
XRD measurement results of sample melt for Run 1.
Figure 2 shows the plots of the ratio of the peak intensity to the initial peak intensity for the peak at 2θ = 19° against time, and fitting curves using (Eq. 2) for Runs 1 to 5. As a result, the change in crystallinity was well fitted by the Avrami equation (Eq. 2) in all conditions. Therefore, it became clear that the crystallization process of xylitol from melt containing sorbitol can be analyzed and predicted using the Avrami equation.
Plots of the ratio of the peak intensity to the initial peak intensity for the peak at 2θ = 19° against time, and fitting curves by Avrami equation (Eq. 2) for Runs 1 to 5.
Effect of initial slurry density and temperature on crystallization process Firstly, the effect of initial slurry density is discussed by comparing among Runs 1, 3 and 5. When a large amount of seed crystal was introduced, there was almost no change in the intensity ratio. Therefore, the melt crystallization process and the crystallinity of the melt were stable under the high seed load condition. When comparing among Runs 2, 3 and 4, the intensity ratio hardly changed with melt temperature condition. From these results, the melt crystallization process was stable under the high seed load condition and the high temperature condition. The changes in the intensity ratio such as crystallinity were well fitted by the Avrami equation in all Runs with different conditions, although the experiments were not repeated. The parameters of the Avrami equation are summarized in Table 2. The Avrami exponential n was in the range of 2 to 3 for all Runs in the present study. Considering the experimental condition of this study and the results of a previous study (Hinrichs, 1996), the mechanism of crystallization is predicted to be as follows: the nucleation of melt is heterogeneous, crystal growth occurs at the interface and the crystals are sphere-like in shape.
| Run No. |
Initial temperature T0 [K] |
Initial slurry density MT0 [mol-crystal/mol-slurry] |
a [–] |
k [h−n |
n [–] |
|---|---|---|---|---|---|
| 1 | 361 | 0 {No seed) | 32 | 0.02 | 2.8 |
| 2 | 361 | 0.26 | 1.2 | 2.3 | 3.0 |
| 3 | 358 | 0.26 | 10 | 0.08 | 2.1 |
| 4 | 355 | 0.26 | 2.2 | 0.02 | 2.7 |
| 5 | 355 | 0.44 | 0.8 | 0.69 | 2.5 |
In the case of Runs 1, 3 and 5, the overall composition of the slurry was the same, and the temperature and initial slurry density MT0 were different. The initial slurry density MT0 of Run 1 was the lowest, so the amount of crystallized xylitol crystal after depositing for Run 1 was the highest among them. According to the k value of the Avrami equation, the crystallization process of Run 1 was the slowest and did not become stable during the observation period, and Run 5 was the fastest. In the case of Run 5, the initial slurry density MT0 was high, so the crystal growth interface was large. From the results mentioned above, it was found that a high initial slurry density of melt before depositing is desirable for rapid crystallization after depositing.
In the case of Runs 2, 3 and 4, the initial slurry density MT0 was the same, and the temperature was different. From the comparison, the crystallization process of Run 2 was the fastest, and Run 4 was the slowest according to analysis using the Avrami equation. From these analysis results, it was found that a high temperature of melt is desirable for rapid crystallization after depositing when the initial slurry density MT0 of the melt is the same.
From these considerations, when a large amount of seed crystal is introduced, crystal growth proceeds and finishes faster. For melt containing a certain amount of seed crystals, the higher the temperature the faster the crystallization process finishes, and the presence of a solid phase is necessary for a stable and progressive crystallization process. Melt with a high slurry density and high temperature is preferable for formulation of desired crystal qualities.
