Introduction

Fast and sensitive methodologies of food analysis, especially for industrial purposes, are useful tools to determine the quality of final commercial products. Moreover, the possibility to use always easier tools to analyze the authenticity of the certificated food composition is a daily challenge. Thermal analysis is recognized as instrumental method of food analysis able to give unique informations regarding the nature of the sample or the modifications induced by industrial processing. Books and reviews report the applications of thermoanalytical techniques to the food science [1–5].

Chocolate is a complex suspension of around 70 % of fine solid particles (from sugar and cocoa), in a continuous fat phase. At ambient temperature (around 25 °C), it is solid and it melts at oral temperature (37 °C) generating a smooth suspension of solid particles in cocoa butter. There are different types of chocolate (dark, milk, and white), according to their composition in terms of cocoa solids, milk fat, and cocoa butter, hence the final products have different compositions. “Dark” chocolate is a complex food product in which sugar crystals and cocoa particles are surrounded by a continuous phase of crystalline and liquid cocoa butter. Due to the hydrophilic nature of the sugar crystals, a small portion of emulsifier, e.g., soy lecithin or polyglycerol polyricinoleate (PGPR), is often added to improve compatibility with the hydrophobic cocoa butter [6, 7]. Thermal analysis has been extensively applied to characterize the thermal, structural, and rheological properties of the dark chocolate [8–10].

The aim of this work, on the basis of our group experience in thermal analysis [11–23] and in food characterization [24, 25], was to propose the thermogravimetry (TG) and the derivative thermogravimetry (DTG) as fast, simple, cost-effective, and sensitive tools to determine the composition of dark chocolate and consequently to control the dark chocolate quality and to certify the cocoa percent declared by the producer.

Experimental

Material

All the analyzed materials were either purchased in specialized stores (brands are omitted) or kindly gifted by chocolate producing companies: five chocolate samples declared as 40 % cocoa (40 %-C chocolate), two 47 % cocoa (47 %-C chocolate), five 70 % cocoa (70 %-C chocolate), 85 % cocoa (85 %-C chocolate), 99 % cocoa (99 %-C chocolate), three different sugars (S), three different cocoa liquor (CL), and four different cocoa butter (CB).

Instrumental

The thermoanalytical curves were obtained by using a Perkin Elmer TG7 thermobalance (range 20–1,000 °C); the atmosphere was either pure nitrogen or air, at a flow rate of 100 mL min−1; the heating rate was varied between 5 and 40 °C min−1, with the best resolution achieved at a scanning rate of 10 °C min−1.

Results and discussion

Although the U.S. has no official definition for dark chocolate, European rules specify a minimum of 35 % cocoa solids [26]. In addition to very specific and sensitive analytical approaches already published [27–29], a fast, cost-effective, and simple tool to determine the percent composition of the main ingredients could be very useful to control the effectiveness of the declared quality. To this end, thermogravimetric analysis (TG) and first DTG were performed on purchased dark chocolate samples. The reproducibility of the TG experiments was always good, with a fluctuation below 2 %. At least three experiments were performed for each sample, but the number was up to five or six experiments for those samples with fluctuations higher than 0.5 %.

The typical thermogravimetric behavior of a dark chocolate sample is shown in Fig. 1 by a 70 % cocoa declared TG profile.

Fig. 1
figure 1

TG and DTG curves of a representative 70 % cocoa declared chocolate. Heating rate: 10 °C min−1. Nitrogen flow rate at 100 mL min−1

Full size image

Under nitrogen purging flow, the TG curve shows three main steps in the temperature range 20–210, 210–340, and 340–750 °C, the second step being the convolution of at least three processes.

By the analysis of the starting materials (sugar, cocoa liquor, and cocoa butter, Fig. 2) it is very easy to assign each TG step in the dark chocolate curve: the first one is related to the sugar contribution, the second step is the release of the cocoa liquor and the final loss is related to the cocoa butter. Figure 3 clearly shows the correspondence.

Fig. 2
figure 2

TG and DTG curves of representative starting materials: sugar, cocoa liquor, and cocoa butter. Heating rate: 10 °C min−1. Nitrogen and air flow rate at 100 mL min−1

Full size image
Fig. 3
figure 3

Comparison among representative starting materials and one chocolate sample. Heating rate: 10 °C min−1. Nitrogen flow rate at 100 mL min−1

Full size image

For the starting materials the reproducibility was high, with results never different of more than 0.3 %.

The nitrogen purging flow gives a clear qualitative profile, but the quantitative interpretation of the analysis is not allowed since inert flows depress the complete decomposition, without a final constant mass value.

By changing the atmosphere to air, a final plateau is obtained, thus allowing quantitative calculations. The TG and DTG curves of Fig. 4 show that the presence of oxygen has no effect on the first release, while shifts to lower temperature, the second mass loss step mainly influences the third one.

Fig. 4
figure 4

TG and DTG curves of a representative chocolate sample under different purging flow. Heating rate: 10 °C min−1. Nitrogen and air flow rate at 100 mL min−1

Full size image

Figure 4 clearly shows that the quantitative results have to be calculated from the curves recorded under air flow. By this method, in Table 1 the calculated and declared cocoa percent for each analyzed dark chocolate sample have been reported. The cocoa is the sum of cocoa liquor and cocoa butter. The higher fluctuation in the lower percent cocoa samples has been attributed to the higher amount of excipients that influences the shape of the curve and consequently the calculated cocoa percent. At least five TG experiments have been run for the 40 %-C samples.

Table 1 Calculated and declared cocoa percent for the analyzed samples
Full size table

It is interesting to note that there is a clear sugar/cocoa dependence when considering different chocolate amount in products of the same brand. This trend is clearly enhanced in Fig. 5, where several representative TG curves show a decrease of the sugar peak when the cocoa percent increases.

Fig. 5
figure 5

TG and DTG curves of representative chocolate samples with increasing cocoa percent declared. Heating rate: 10 °C min−1. Air flow rate at 100 mL min−1

Full size image

A final comment is due for samples with cocoa <70 %: in the final TG step, the derivative curve often shows an additional overlapped peak that is due to excipients (i.e., vanillin) and must be subtracted to avoid wrong evaluations. This process lowers intensity or disappears in samples with higher cocoa percent. In Table 1 all the results have been subtracted by this contribution.

Conclusions

TG and DTG revealed to be useful tools for the determination of the dark chocolate composition. The results show the possibility to give fast, cost-effective, and accurate informations when the purging flow is air and to propose TG and DTG for the control and the certification of the dark chocolate quality.