Heracles 2

Samples were analyzed using a gas chromatography electronic nose Heracles II (Alpha MOS, Toulouse, France). Weigh 3 g of gel and place into a sealed 20 mL headspace vial. The test unit consisted of two columns DB-5 and DB-1701 (capillary column, 10 m × 0.18 mm) with different polarity and two FID (dihydrogen flame) detectors. For headspace extraction, the temperature was set to 60 °C, while the inlet temperature was maintained at 200 °C. The temperature program consisted of an initial increase from 50 °C to 80 °C at a rate of 1 °C per second, followed by a further increase to 250 °C at a rate of 2 °C per second.

The aroma profile of samples was analyzed using an electronic nose (Heracles 2, Alpha MOS, France), following the method of Aunsa-Ard and Kerdcharoen (2022) (link) with an appropriate modification. The sample volume was 5 mL, in a 20 mL vial. The equilibration process was conducted at 60 °C for 25 min, with shaking at 500 rpm. A 100 μL aliquot was injected at a temperature of 250 °C, at a rate of 125 μL/s. The carrier gas flow rate was maintained at 10 mL/min. Analytes were collected over a temperature range of 50 °C to 240 °C, with a final collection temperature of 240 °C. The column temperature program involved an initial ramp of 1 °C/s to 80 °C, followed by a ramp of 2 °C/s to 250 °C, with a 60-s hold at the final temperature. Detection was performed using a flame ionization detector (FID) at 260 °C, with a detection time of 177 s. The FID amplification factor was set to 12. Calibration was conducted using nC2 – nC16 normal alkane standards, with retention indices calculated. Qualitative analysis of each compound was performed using the AroChemBase database.

Fingerprinting of volatile compounds was performed using an ultra-fast GC electronic nose (Heracles II, Alpha M.O.S., Toulouse, France), equipped with an autosampler (Odour Scanner HS 100, Alpha M.O.S., Toulouse, France), a cooled Tenax trap, two capillary columns of different polarities and two flame ionization detectors (FIDs). The two parallel columns were the non-polar MXT-5 (5% diphenyl) and a medium polar MXT-1701 (14% cyanopropylphenyl) column with dimensions of 10 m × 180 μm × 0.4 μm.
Prior to analysis, 2 g of cake was weighed in 20 mL vials, which were then tightly capped with PTFE/silicone seals to prevent the penetration of environmental odours. Each cake sample was analysed five times and blanks (empty vials) were added before, in between and after the measurements of different samples. To introduce the volatile compounds into the headspace, vials were incubated for 20 min at 60 °C with a constant agitation of 500 rpm. After incubation, the headspace was collected into a syringe of 70 °C at a speed of 500 µL/s. Then, 5 mL was injected into the GC system for 45 s at 125 µL/s at 200 °C and 10 kPa with a hydrogen N7.0 carrier gas (Parker Balston, Gateshead, Tyne & Wear, UK). Afterwards, a trap system with an initial condition of 40 °C for 50 s was heated in 35 s to 240 °C and sustained for 30 s. The vent rate of the injector was at 30 mL/min and the split of the trap at 10 mL/min. At a temperature of 250 °C, the valve sent the volatiles to the columns. The oven had an initial temperature of 50 °C for 2 s, whereafter it was raised to 80 °C at 1.0 °C/s and immediately ramped up to 250 °C at 3.0 °C/s and kept for 21 s. The temperature of the FIDs was 260 °C, and the gain and offset of both FIDs were 12 and 1000, respectively. The total acquisition time was 110 s with an acquisition period of 0.01 s [30 (link)].