Three levels were assessed for each factor: power levels from 2 to 10 (dimensionless), processing times from 5 to 20 seconds and sample volumes from 10% to 65% of the total container volume (100 mL Schott beaker with external diameter of 50 mm and 70 mm height). The concentration of the sample expressed as degrees brix (45°Brix) and the position of the container within the cavity (center) were held constant. Using SAS software, a statistical design of the response surface (central composite) with three replicates at the center point was prepared, as shown in Table 1.

Table 1

Source: The authors.

Additionally, an optimization process was performed to determine which operating variables resulted in the greatest degree of biological destruction, i.e., a lower value of CFU/mL. During all the trials, the response variable was the colony forming units per volume, which was determined according to the standard NTC 4519 [12] and thermal kinetics of the process (extract temperature vs. process time). Quantification of the colony-forming units was performed in duplicate.

Once obtained the operating conditions at which the maximum biological destruction was achieved, the effects of the concentration of soluble solids (15 and 45 °Brix) and the position of the vessel in the cavity were evaluated: center, 6 cm from the center and 12 cm from the center (end of the rotary plate).

For the physicochemical and sensory testing of the sterilized product, it was assessed whether the drink was different from the conventional product, using the triangular sensory test according to the standards NTC 2681 and 4883 [13,14]. In addition, the physicochemical properties such as pH and acidity were determined according to standard NTC 4675 [15].

Table 2 presents the microbial destruction results using the conditions defined in the experimental design and an initial concentration of bacterial inoculum of 2x10⁵ CFU/mL for all the samples analyzed. The minimum value was obtained when the process conditions were: power level of 6, time of 19 seconds and 11 mL volume. In terms of biological destruction, low volumes and high processing times are the most suitable. Analysis of variance demonstrated the significant effect of the operating conditions (P <0.05) on the lethality. Table 3 lists the statistical results of experimental design.

Table 2

Source: The authors.

Table 3

Source: The authors.

This model is presented in Fig. 3. The figure illustrates that there is a minimum CFU/mL when the volume is low and the power level is between 5 and 10.

Figure 3 Source: The authors.

Table 4

Source: The authors

Figure 4 Source: The authors

Table 4 shows the optimization of the model. These values match the values reported in Table 2 and Fig. 3, where the lowest value of CFU/mL was obtained when the process was performed under these conditions. According to the values observed, the model makes an over-prediction of the CFU/mL value by approximately 6%.

Fig. 4 shows the variation of temperature of the sample using the optical fiber sensor with respect to processing time under the following conditions: power level of 6, time of 19 seconds and extract volume of 11 mL.

The results indicated that heating does not occur instantaneously, but a marked increase in temperature occurs for processing times higher than 50%. However, between 5-15 seconds, the temperature rapidly increases, coinciding with the results observed by Vadivambal and Jayas [16]. Another study conducted by Arimi and coworkers [17], reported similar behavior to that observed in this study. During the first 5 seconds, there is no change in the temperature of the sample, which can occur by the response time of the optical fiber

To analyze the thermal kinetics of the process, the values of the initial (Ti) and final temperatures (Tf) obtained for each sample are reported (Table 5). The conditions for which the minimum count of CFU/mL was obtained were a volume of 11 mL, a power level of 6 and a time of 19 seconds, for which the final temperature was 84.85°C, where this value is greater than the inactivation temperature (80°C) for bacterial vegetative cells [18].

Table 5

Source: The authors

Table 6

Source: The authors

According to Table 5, the optimum conditions for the microwave inactivation of Bacillus licheniformis in liquid coffee extract were a power level of 6, a processing time of 19 seconds and 11 mL volume. This result agrees with the conditions reported by Chandrasekaran and coworkers [19], who observed that microwave-processed foods must have a low volume because the microwave technique in high moisture product has a low microwave penetration depth.

On the other hand, the treatments applied reduced 2 logarithm cycles in 19 seconds, which corresponds to D value of 12 seconds, which is due to gram-positive bacteria such as Bacillus licheniformis are more resistant to microwaves in compared to gram negative [20], this is because the treated bacteria were not completely lysed. Previous studies report process times longer than 19 seconds to inactivate bacteria in liquid products such as juices [6].

In this work, the effect of the concentration of sample measured as degree brix ((Brix) in the inactivation of the bacteria of interest was also assessed. The study to 15(Brix and 45(Brix was conducted under process conditions where the minimum recount CFU/mL was initially observed i.e., a power level of 6, a processing time of 19 seconds and an extract volume of 11 mL. The initial concentration of Bacillus lichenformis was 2x10⁴ CFU/mL.

Table 6 lists the values of CFU/mL of Bacillus licheniformis obtained by microwaves processing at various levels of concentration. .

The analysis of variance (ANOVA) indicates that the concentration of the coffee extract has no effect on the biological destruction process (P>0.05). This finding may be due to the electrical conductivity of the sample not changing significantly with the concentration of soluble solids, as indicated by Campañone and coworkers [21]. According to Simpson [22], if the heat generated by microwave heating is given by

Table 7

Source: The authors.

Figure 5 Source: The authors

(3)

where E is the electric field and σ is the electrical conductivity of the sample, the heat generated under the evaluated conditions should not exhibit significant differences; thus, no effect of the coffee extract concentration on the microorganism inactivation is observed.

The values of temperature and absorbed power for each treatment as a function of the solids concentration are listed in Table 7. In addition, Fig. 5 illustrates the heating degree reached as a function of the processing time.

The results in Table 7 indicate that for the liquid coffee extract, the concentration of the samples expressed as ºBrix had no significant effect on the final product temperature and the level of energy absorbed by the sample.

This result occurs because the composition and thermal properties (specific heat) do not vary noticeably with ºBrix (less than 8%). According to Datta and Rakesh [23], these properties are critical in heating via microwave and could generate a substantial change in the process.

According to Chandrasekaran and coworkers [5], one of the critical factors in the microwave process is the position of the container containing the sample in the cavity. Therefore, three positions were evaluated: center, 6 cm from the center and 12 cm from the center (end of the rotary plate). The initial concentration of bacteria was 2 x 10⁴ CFU/mL in the liquid coffee extract at 45°Brix. Table 8 lists the biological destruction achieved as a function of position within the oven cavity.

Table 8

Source: The authors.

Table 9

Source: The authors.

Figure 6 Source: The authors.

Using a completely randomized unifactorial design, the positions were evaluated, and a statistically significant effect of the position on the final degree of biological destruction was observed (P <0.05).

Table 9 lists the power absorbed by the sample using the temperatures measured at the beginning and end of the process. The final temperature reached during the process does not change significantly with the position inside the microwave. However, Fig. 6 illustrates how the temperature profile is highly variable. The results in Fig. 6 demonstrate that the thermal inertia of the process is higher at positions further from the center. Additionally, the temperature increases are higher for the positions close to the center.

Table 10

Source: The authors.

Evaluating the position of the container within the microwave cavity revealed that the temperature increase occurs faster in the center than at the ends because the field is uniform to the extent that the product rotates [23]. Additionally, the increases in temperatures are higher at positions close to the center, which is due to the waves in these positions being reflected in the plate and losing energy that could be used to increase the temperature in the extract [24].

With respect to the power absorbed by the sample, Cañumir and coworkes [25], observed that the inactivation of Escherichia coli required a power between 900 and 720 W for processing times of 60 to 90 seconds in apple juice, while the results obtained from this study indicate that less time and a lower power for the inactivation of a more microwave-resistant bacteria are required. The possible differences may be due to variations in the dielectric properties of the products and to the distribution of the electromagnetic field within the oven cavity.

Table 10 lists the physicochemical properties of the liquid coffee extract. No statistically significant differences (P>0.05) were observed between the control and treatment in terms of the values of pH, acidity and ºBrix.

In the sensory analysis, it was assessed whether there are significant differences between the coffee extract that has not been subjected to microwave treatment (control) and the treated coffee extract. No significant differences between the control and treatment (P>0.05) were observed; thus, the microwave treatment is considered a good alternative compared with conventional thermal processes as the microwave pasteurization preserves more flavor, color, nutritional value and quality [5,8].