Chemicals and reagents
It is considered as all chemicals, solvents, media used in this work were of analytical grade. The chemicals such as 2,2-diphenyl-1-picrylhydrazyl (DPPH), 2,4,6-tripyridyl-s-triazine (TPTZ), butylated hydroxy toluene (BHT), quercetin, gallic acid and caffeic acid were obtained from Sigma-Aldrich, Germany. Along with these chemicals, ethanol, Folin and Ciocateu’s phenol reagent, aluminium chloride, HPLC grade water, acetonitrile and acetic acid were purchased from Merck, Mumbai, India. The crystal of sodium carbonate anhydrous and sodium nitrite were from LOBA CHEMIE, Mumbai, India. Especially milli-Q system with a 0.22 μm filter paper (Millipore, Bedford, USA) was used to purify the mobile phase solution.
Collection of microalgae sample
The microorganism Spirulina platensis var. lonor used in this study was collected from Antenna Green Trust, Madurai, Tamil Nadu, India. The samples were stored in screw capped amber colored plastic sealed container under dry and dark conditions.
Analysis of proximate composition of microalgae biomass sample
The proximate analysis is used for the quantitative assessment of macromolecules of foods. It is confirmed that the proximate composition was assessed to find moisture, protein, ash, lipid, crude fiber and carbohydrate content of the microalgae biomass sample . The moisture content was determined according to AOAC official method of 934.01by drying 5 g of sample in a hot air oven at 100–105 °C . Along with, the crude protein was estimated by Kjeldahl protein units. The total amount of protein was determined by nitrogen percentage (%) × 6.25  and total lipid content was measured by using hexane which described in AOAC official method of 920.39 . Apart from that total ash was estimated by incineration of microalgae sample (5 g) at 550 °C during 2–3 h . According to AOAC official method of 978.10, the crude fiber was estimated by subjecting 2 g of the de-fatted sample with 1.25% H2SO4 and 1.25% NaOH treatments as per AOAC, 2006 . The total carbohydrate content of microalgae biomass was analyzed by difference. All proximate determinations were done in triplicate.
Optimization of SC-CO2 conditions for extraction of caffeic acid from Spirulina
SC-CO2 extraction was accomplished by a Speed SFE System of Applied Separations, USA extraction unit according to the modified method of Dutta and Bhattacharjee, (2015) . The system includes of a pump fixed with a cooling bath refrigerator at − 4 °C to make colder the pump. The fifty gram (50 g) of freeze dried Spirulina powder was weighed accurately and filled into a 100 mL of stainless steel vessel. In order to avoid leakage from the extraction vessel the cotton wool was to be put at both ends of the vessel. The compressed CO2 was added at flow rate of 2 L/min from the bottom of the extraction vessel through micro-metering valves. As a co-solvent, 100% ethanol was added at 1 mL/min from the bottom end of the extraction vessel.
In the present context, a BBD was employed for the optimization of SC-CO2 extraction conditions with three parameters of extraction pressure (200, 350 and 500 bar), temperature (30, 40 and 50 °C) and extraction time (static + dynamic) (60, 90 and 120 min) was employed in the present study. The dynamic time of 30 min was kept constant since no extract was collected after this time. After the extraction process all extracts were collected in glass vials. The extracts were gravimetrically weighed and dissolved in 96% ethanol. Then the extracts were successively stored in amber colored screw capped glass vials at 4 °C by purging a stream of nitrogen gas until further analyses. For comparative evaluation with conventional extraction methodology, solvent extraction was carried out with 10 g lyophilized biomass (5 g) in 50 mL of ethanol and was shaken at 200 rpm in a rotary shaker during 5 h at 30 °C. After that, the solution was centrifuged (SIGMA Laborzentrifugen 2–16 PK refrigerated centrifuge) at 13,500 g for 15 min and concentrated on a rotary vacuum evaporator (Supervac-R/180; Superfit Continental Pvt. Ltd., Mumbai, India) at 50 °C–60 °C. The extract was stored under conditions described above.
Determination of caffeic acid content of Spirulina extracts by High performance liquid chromatography (HPLC)
The level of phenolic acid in extracts was quantified followed by the specific method with slight modifications . The separation of phenolic acid was done on an Agilent ZORBAX SB C-18 column (150 × 4.6 mm, 5 μm) at 280 nm and 330 nm. The obtained each sample was dissolved in mobile phase solution (B) and filtered through a 0.22 μm polytetrafluorethylene syringe filter prior to HPLC injection at an injection volume of 20 μL into an Agilent 1300 series HPLC (Agilent Technologies Inc., Alpharetta, GA, USA). The eluent phase of water–acetic acid (95:5, v/v) (A) and methanol–acetonitrile–acetic acid (95:5:1, v/v/v) (B), starting from 0 to 40% B in 10 min, 40%–100% B in 10 min, 100% B in 5 min, and 100%–5% B in 5 min was used. The flow rate of solution was programmed at 1 mL/min. Therefore, the standard caffeic acid was solubilised in mobile phase B solution and concentration prepared at 100, 80, 60, 40, 20 µg/mL . Every solution was injected three times using a needle. The mean peak areas were selected for the preparation of respective standard curve of caffeic acid and the results were expressed in µg/g of dry microalgae.
Estimation of phytochemical properties of obtained Spirulina extracts
The total phenol content of the microalgae extracts was estimated by Folin and Ciocalteu reagent method  with slight modification. The total phenol contents were thus calculated as milligram gallic acid equivalent (GAE)/g dw. The reducing power of the extract was determined following the method of Ghosh et al. . The reducing power was expressed as mg BHT equivalent/g dw from respective standard curve. Total flavonoids content of the extract was estimated by the AlCl3 method as described by Srivastava et al. (2012) . The total flavonoid content was expressed as mg QE/g dw.
Determination of antioxidant activities of extracts
The antioxidant activity of all extracts was determined by measuring the radical scavenging activity of DPPH  and expressed as IC50 values. The reducing ability was determined by Ferric reducing antioxidant power (FRAP) assay based on the reduction of a ferric-tripyridyltriazine complex to its ferrous colored form in the presence of antioxidants  and expressed as mM FeSO4 equivalent/g dw (from the standard curve prepared).
Gas chromatography (GC–MS) analysis of the best SC-CO2 extract
The extract having the best combination of caffeic acid content and phytochemical properties was analyzed by GC–MS. GC–MS analysis of SC-CO2 extract were set according to the following method of Pantami et al. (2020)  with some modification. An Elmer GC Clarus Perkin 680 chromatograph (M/S, Perkin Elmer, MA, and USA) was used for the analyses. In particular, nitrogen gas was used at 1 mL/min as carrier. The column temperature was set firstly at 70 °C for 2 min, then increased to 150 °C at 25 °C/min, 200 °C at 3 °C/min, 260 °C at 8 °C/min and finally, settled at 260 °C for 1 min. After that, the best microalgae extract was diluted and automatically injected (1 µL) with 1:2 in split mode. As of now, injector and detector temperatures were put at 250 °C and 260 °C, respectively. The septum purge flow and gas saver mode were programmed at 3 mL/min and 20 mL/min. The MS source and MS Quad temperature were set at 230 °C and 150 °C with solvent delay by 12 min. The obtained peaks representing mass to charge ratio were compared with NIST mass spectrum library.
Application of SC-CO2 extract of Spirulina platensis in litchi beverage
Ripe litchi was collected from Kharagpur local market. After washing with distilled water litchi juice were extracted by proper grinding and filtrating by muslin cloth. 2% sugar was added to prepared juice and heated on low flame. Then caffeic acid rich SC-CO2 algal extract (1 µg/mL) was added as preservative in litchi beverage. The treated juice was filled in glass bottle under sterilized condition. The glass bottles were shield immediately using a hand shieling machine for preventing contamination. Finally total two sets of studies were conducted by varying storage temperature viz. refrigerated (4 ± 2 °C) and ambient storage (room temperature) at specific time interval (0, 7, 15, 30, 45, 60 days).
Sensory evaluation of litchi beverage
The litchi beverage preparations were sensorial evaluated by a panel of 10 trained judges comprising of research scholars of IIT Kharagpur. The sensory evaluation was selected by 9 point Hedonic Scale for different sensory attributes like colour, flavour, appearance and overall acceptability .
Physical and chemical analysis
The litchi juice samples were taken from bottle at specific time interval and were analysed for total bacterial counts (TBC), total fungal count, total soluble/suspended solid, pH, acidity. TBC were estimated on nutrient agar media (Hi-media). After incubation at 37 °C for 24 h colony forming units (CFU) were counted. Fungal counts were estimated on potato dextrose agar (Hi media) after serial dilution of sample. After proper incubation at 28 °C for 72 h, CFU were counted. Total soluble solids (TSS) was determined using hand held refractometer (ERMA, Japan; 0–32) at room temperature. The pH of the litchi juice was determined with the help of digital pH meter (Systronics, Model 361). According to the method of Sadler and Murphy, (2010)  the titratable acidity was estimated.
One-way analysis of variance (ANOVA) with Duncan's multiple range tests have been carried out to study the activities of extraction parameters such as temperature (°C) and pressure (bar) on the total yield of extract, total flavonoids, total phenolic content, antioxidant activity and reducing power of dried SC-CO2 extracts of microalgae. All statistical experiments were conducted using IBM SPSS Statistics 22 software (Statsoft, OK, USA) to verify the significance level at the p value of 0.05.