*3.1. Materials*

Kolliwax ® GMS II (glycerylmonostearate 40–55 type II, powder, Tm = 54–64 ◦C) was supplied by BASF GmbH (Ludwigshafen am Rhein, Germany). CO2 for the PGSS ® technique (purity 99.8%) and for the melting point determination (purity 99.998%) were purchased from Praxair (Madrid, Spain) and Air Liquide (Santiago de Compostela, Spain), respectively.

### *3.2. Determination of the Melting Point of GMS in the Presence of Compressed CO2 at Di*ff*erent Pressures*

The melting point of the GMS in the presence of compressed CO2 in a 0–200 bar pressure range was determined. A sample of GMS (approximately 3.5 mg) on a glass vial was placed inside a variable volume high-pressure cell, consisting of a horizontal stainless-steel cylinder with an internal diameter of 2 cm and a piston to adjust the volume from 7.9 to 29.5 cm3. The cell was equipped with a sapphire window (1.6 cm diameter) that allowed the detection of phase transitions through an endoscope (Olympus 5 series, Olympus, Tokyo, Japan) connected to a CCD-camera (Moticam 2000, Motic Asia, Hong Kong, China). In one sidewall of the cylinder, a second sapphire window (6 mm in diameter) made it possible to illuminate the interior of the cell through an optical fiber. A Pt100 probe with an uncertainty of 0.02 ◦C was used to measure the temperature in the cell wall. The pressure was measured with a Heise model DXD series digital pressure transducer, with an operating range 0–500 bar and an uncertainty of 0.02% of the full scale (FS).

For the experimental trials, the cell at its maximum volume was filled with CO2 at room temperature and supply pressure of 60–65 bar. Afterwards, the system was heated to the selected temperature (from 52 to 61 ◦C) and the pressure was gradually increased moving the piston (i.e., reducing the volume of the chamber) until the solid was completely molten to determine the melting point value. Thus, the melting pressure of the GMS at the selected temperature was determined. Subsequently, another temperature was selected and the procedure was repeated to obtain another value of the melting curve. Temperature measurements were carried out by triplicate. Results were expressed as the mean value ± standard deviation (SD). At a fixed temperature, this device shows repeatability for the pressure lower than 11.4%. The melting point temperature of the GMS at atmospheric pressure in the same equipment was also determined.

### *3.3. SLMPs Production by the PGSS Technique*

For the particle formation protocol, 6 g of GMS powder were placed into a 250-mL high-pressure autoclave (saturator) (Eurotechnica GmbH, Bargteheide, Germany). After heating the saturator to the desired temperature (T), CO2 entered the equipment at a constant flow of 7 g/min until the desired pressure (P) was reached. After 1 h of contact between the molten lipid and the compressed CO2 under stirring at 400 rpm, the system was depressurized by opening the valve placed at the bottom of the saturator. When the molten lipid leaves the saturator through a nozzle, rapid depressurization causes lipid microparticles precipitation within a 2.7 L borosilicate autoclave (precipitator).

Batches of GMS particles were produced following a D-optimal experimental design for three variables: nozzle diameter (2 levels), operating temperature (3 levels) and pressure (3 levels) (Table 3) carried out by DataForm ® v.3.1 software (Intelligensys Ltd., Stokesley, UK). GMS particles processed under di fferent pressure and temperature conditions were denoted as GMS-x-y-z, where x is the nozzle diameter in mm, y the processing temperature in degrees Celsius and z the processing pressure in bar.


**Table 3.** Nozzle diameters and processing temperatures (T) and pressures (P) tested for the preparation of SLMPs of GMS using the PGSS ® technique.

Microparticles were collected and weighed to determine the process yield according to Equation (1):

$$\% \text{ fine particles} = \frac{W\_f}{W\_0} \times 100\tag{1}$$

where W0 is the initial weight of GMS added to the saturator and Wf is the final weight of fine particles collected. Also, the amount of GMS remaining on the walls of the precipitator and the interior of the tubing was weighed to verify all the GMS had left the saturator, and what amount had not precipitated into SLMPs.

### *3.4. Morphological Analysis, Physicochemical Characterization and Particle Size Distribution (PSD)*

Four aliquots of each batch were characterized in terms of particle size distribution by optical microscopy using a camera (EP50, Olympus, Tokyo, Japan) provided with the software EP View (Olympus, Tokyo, Japan). The images were analyzed using the freeware ImageJ 1.49v. Calculated particle diameters correspond to the projected area equivalent diameter. The particle size distributions were fitted to a normal distribution, and mean particle size and standard deviations were obtained. The circularity of the particles was also evaluated by image analysis.

X-ray di ffraction (XRD) and attenuated total reflectance/fourier transform infrared spectroscopy were used to test possible physicochemical modifications in GMS caused by PGSS ® processing. XRD patterns were collected (PW-1710, Philips, Eindhoven, The Netherlands) in the 2–50◦ 2θ-range using a 0.02◦ step and CuK α1 radiation. ATR/FT-IR spectra (Gladi-ATR, Pike, Madison, WI, USA) were obtained in the 400–4000 cm<sup>−</sup><sup>1</sup> spectrum range from 32 scans and at a resolution of 2 cm<sup>−</sup>1.

Particles were also analyzed by scanning electron microscopy (SEM Zeiss EVO LS 15; Zeiss, Oberkochen, Germany) to evaluate their morphology and surface texture. Particles were previously sputtered-coated with a layer of 10 nm of iridium to improve the contrast (Q150 T S/E/ES, Quorum Technologies, Lewes, UK). Bulk density of the particles was determined by a volumetric method and the skeletal density was evaluated using helium pycnometry (MPY-2; Quantachrome, Delray Beach, FL, USA).
