*2.3. Freeze-Drying Process*

All experiments were carried out in a Martin Christ freeze-dryer, model Alpha 2-4 LSCplus (Martin Christ Gefriertrocknungsanlagen, Osterode, Germany), which can operate at a total vacuum pressure of up to 0.001 mbar, provided with an MKS Baratron 622 capacitance manometer (MKS Instruments) and a condenser that can operate at temperatures down to −85 ◦C (see Figure 2). It has three shelves of 0.021 m<sup>2</sup> each that can be temperature controlled with a wireless temperature monitoring system, which, in turn, allows for sample temperature monitoring through pt100 port sensors. The samples or blueberries were placed onto the three shelves, considering a total load of 60 units. In order to assess the product temperature, one blueberry fruit per shelf was inserted with a miniature pt100 sensor (PT 100 Mini-Epsilon LSC Plus, Martin Chris, ostero de am Harz, Germany). A standard freeze-drying processing condition was fixed at 0.13 mbar (13 Pa) with a condenser temperature of −85 ◦C and a shelf temperature of 20 ◦C, while the air temperature was also 20 ◦C. The freeze-drier system includes dedicated software (SCADA Software V1 LPC plus Martin Chris, ostero de am Harz, Germany) to set, control, and monitor the freeze-drying process at the selected processing condition. It samples

data every 5 s. A butterfly valve with an approximate closing time of 0.5 s, mounted in the cylindrical duct connecting the freeze-drying chamber and the condenser chamber, can be managed by the software to perform Pressure Increase Tests (PIT) (Figure 2). The SCADA Software LPC plus software allows graphic visualization and recording of several process variables: tray temperature, blueberry temperature, condenser temperature, chamber absolute pressure (capacitive sensor), and PIT data.

**Figure 2.** Freeze-drying system arrangemen<sup>t</sup> and its processing condition set up.

#### *2.4. Determination of Primary Drying Time*

After freezing, the FD process proceeds to primary-drying or sublimation, and finally to secondary drying or water desorption. The primary drying stage—conventionally, the most time-consuming part of the process—was investigated in the present study. In order to determine the end of the primary drying time or the transition from the primary drying stage to the final secondary-drying, which can be estimated with the automatic Pressure Increase Test (PIT), the freeze-drier software system was implemented. It works by temporarily closing the butterfly valve between the product chamber and the ice condenser (see Figure 2), allowing a pressure increase in the product chamber. If the pressure increase remains below a set limit, usually 10% whilst the valve is closed, the software program assumes that there is no further sublimation water left in the product, and the primary drying phase can be considered to be finished, and this time is recorded. Another way to estimate the end of the primary drying stage is to observe the difference between the shelf and assessed product temperature; when they equalize, no more sublimation heat is absorbed, meaning the sublimation is over.

#### *2.5. Visual Registry of the Blueberry FD Process*

A photo-sequence methodology was used to capture and visually analyze the dynamics of the busting process while blueberries are being freeze-dried. Images of the top-shelf blueberries were taken by a photographic camera (Flea®3 FL3-GE-03S2C-C Color GigE Camera, FLIR Systems, Wilson Ville, OR, USA) located at the upper part of the freeze drier system (Figure 2). A dataset of images taken at a rate of 12 photographs per min was then visually inspected to calculate the fraction of busted blueberries over time.

#### *2.6. Estimation of the Dry-Layer Mass Transfer Resistance (RT)*

The MTM method was used to evaluate the influence of product characteristics in the FD process. The needed chamber pressure-rise as a function of time can be experimentally monitored, and the acquired data can then be regressed through Equation (1), where unknown parameters such as Pi, X, AT (where AT = N × Aunitary), and RT can be estimated. The minimum conditions on the experimental procedure have been reported in order to obtain reliable results. The data collection time could not be longer than 30 s, since a longer time would allow a considerable product temperature increase because of the closed system, and within that time, there needed to be su fficient data collection to observe the complete development of the exponential part of Equation (1), which accounts for the pressure rise controlled by the product resistance. During pressure increase runs, the dynamic pressure increase was monitored with a cDAQ module/9215 data acquisition system and LabVIEW software, manufactured by National Instruments (11500 N Mopac Expwy, Austin, Texas, United States), allowing a sampling period of as low as 10 ms. To successfully apply the MTM method, the product of the geometric characteristics and the particular void volume of the chamber system needed to be adjusted. It has been demonstrated that the computed value of the exponential expression Q (without considering t) should be equal to or higher than 0.2 (1/h) to ensure complete depiction of the exponential phase of the pressure rise in the product chamber [28]:

$$\mathbf{Q} = \left(\frac{3.461 \text{NAT}\_s}{\text{VR}\_\Gamma}\right) \ge 0.2 \text{ (1/h)}\tag{2}$$

For a given value of RT, Q allows the minimum number of units (N, number of blueberry fruits) that must be loaded in a particular freeze-drier system (characterized by its void volume V) to be estimated to obtain a value equal to or higher than 0.2 (1/h). A total resistance value (RT) of approximately 3 (torr h cm<sup>2</sup>/g) has been reported to be acceptable to carry out this assessment.

#### *2.7. CO2 Laser Microperforations and Laser System Settings*

As mentioned in the introduction section, CO2 laser microperforation is a convenient pretreatment as it is a noncontact technology that significantly mitigates the chance of physical and microbiological contamination of materials that are typically associated with traditional cutting or contact devices or fluids. In this respect, the most-used pretreatment technology is to machine-cut the blueberry fruits into halves, which e ffectively avoids the busting process and reduces the primary drying time. Then, both pretreatment technologies were evaluated: CO2 laser perforation at four levels of perforation—1, 3, 6, and 9 perforations per blueberry fruit—and FD blueberry cut into halves. Finally, the results were compared with each other and with those of FD whole blueberries.

A 100 W CO2-laser system (Firestar t100, Synrad Inc., Mukilteo, Wash., U.S.A.) was used to carry out the perforations of blueberries. The system was equipped with a 125 mm focusing lens (FH series Flyer, Synrad Inc.) and a computer interface with laser marking software (WinMark Pro, Synrad Inc.).

The system was operated at a continuous wavelength of 10.6 μm and a frequency of 10 kHz. Perforations were made in a square grid pattern with a density of 2.0 × 2.0 mm.

The CO2 laser was set at 120 pulses, duration of 1 millisecond, linear speed of 100 cm per second, distance of 128 mm between the laser and the surface of the blueberry, and a percentage of power that was determined experimentally by observing in a microscope (Helmut Hund GmbH, H600/12) the depth of the perforation, until finding the configuration of the laser that allowed us to perforate until 1/3, varying the power of the laser (whose maximum was 100 watts).

A unit of frozen (−35 ◦C) blueberry was loaded onto an aluminum tray (20 cm × 20 cm) to perform microperforations; then, blueberry samples were refrozen at −35 ◦C and kept until FD experiments.

#### **3. Results and Discussion**
