Characterization of Candies by DSC and Microscopy - METTLER TOLEDO

Characterization of Candies by DSC and Microscopy

Introduction

People who eat candies (sweets) are usually interested in their taste and consistency rather than in their chemical-physical properties. The consistency of a candy is however strongly dependent on structural aspects.

This article illustrates how certain properties of candies can be investigated using DSC and microscopy, in particular softening, melting and crystallization behavior. Detailed knowledge of these properties is important for defining proper production and storage conditions.

The term candy includes a large number of very different products (bonbons, milk and crème caramels, gelatinized sugar products such as jelly beans, etc.). In this article, we have investigated a commercially available three-layer candy.

 

Experimental Details

The candy consisted of three distinct layers: The two outer layers were transparent whereas the middle layer was white. This indicated that the two outer layers were probably largely amorphous and that the middle layer comprised mainly crystalline material.

All three layers were investigated; they will be referred to as “top”, “middle” and “bottom”. The samples were prepared by finely grinding them in a mortar. During the sample preparation, it was noticed that the two outer layers absorbed water relatively rapidly. The samples were therefore hermetically sealed in 40-µL aluminum crucibles immediately after grinding.

The measurements were performed using a DSC822e equipped with an FRS5 sensor and an IntraCooler. The samples were heated from –50 °C to 160 °C at 10 K/min, cooled to –50 °C at 5 K/min, and finally heated a second time to 160 °C at 10 K/min.

 

Results and Discussion

Figure 1 shows the heating curves recorded for samples of the top and bottom layers. In the first heating run, the curves exhibit glass transitions at approximately 15 °C (top) and 18 °C (bottom). At about 60 °C, the curve shows an endothermic peak that is accompanied by a stepwise shift of the baseline. The stepwise shift is typical indication of another glass transition; the peak can be interpreted as being due to enthalpy relaxation or melting. The second heating runs of both layers show only one glass transition in each curve. The step height of this glass transition corresponds roughly to the sum of the step heights of the two glass transitions recorded in the first heating run. This indicates that, prior to the first heating run, both layers consisted of two phase-separated components. After the first heating run the two components form a mixture so that only one glass transition is observed. The glass transitions of the upper and lower layers differ by only a few Kelvin.

Top and Bottom Layers of Candy by DSC 

The cooling curves (not shown) gave no indication of a crystallization process. From the measurements shown in Figure 1, it is therefore not possible to decide whether the endothermic peak at about 60 °C is a melting process or due to enthalpy relaxation. In this case, we were able to clarify the situation by performing a measurement in which the melting process was directly observed. The METTLER TOLEDO FP82 hot-stage microscopy system is ideally suited for this purpose. This instrument allows a sample to be heated under control and at the same time continuously observed by means of a polarization microscope. The (birefringent) crystals are visible, but non-crystalline regions remain dark. The images obtained are shown in Figure 2. The images show that crystals are indeed present at 50 °C. During the heating run the crystals do not however begin to melt until about 55 °C. The maximum melting rate is reached at about 70 °C and melting has finished by about 85 °C. Since the endothermic peak on the DSC curves already occurs at lower temperatures, the peak must be at least partly due to enthalpy relaxation.

The result can also be verified by DSC measurements using a simple trick. A sample is heated up to a temperature at which the glass transition has already taken place, in this case about 52 °C. This eliminates the thermal history of the material. When the sample is heated again an enthalpy relaxation peak should no longer occur. The corresponding measurements are shown in Figure 3. As expected, the second heating measurement no longer exhibits an enthalpy relaxation peak. A small relatively broad melting peak is however visible. If the sample is reheated a third and a fourth time it can be seen that the sample becomes more and more mixed each time. 

Conclusions

The melting and softening behavior of materials can be relatively easily investigated using DSC and optical measurements (e.g. using the FP82 hot-stage microscopy system). In the example, a candy consisting of three layers was analyzed using these methods. The result was that considerable differences between the two outer layers and the middle layer were found. The two outer layers were practically completely amorphous, whereas the middle layer was crystalline. Measurements with the FP82 hot-stage microscopy system indicated that the “amorphous” layer also contained crystals. The FP82 measurements also showed that the apparently complex melting process on the DSC curve was a result of recrystallization behavior of the material (Ostwald ripening).

Characterization of Candies by DSC and Microscopy | Thermal Analysis Application No. UC 276 | Application published in METTLER TOLEDO Thermal Analysis UserCom 27