Optimization of supercritical fluid extraction of phytosterol from roselle seeds with a central composite design model

Recovery of phytosterol from roselle (Hibiscus sabdariffa L.) seeds via supercritical carbon dioxide extraction modified with ethanol was investigated at pressures of 200–400 bar, temperatures from 40 to 80 °C and at supercritical fluid flow rates from 10 to 20 ml/min. It was found that an entrainer such as ethanol could enhance the solubility and extraction yield of roselle seed oil from the seed matrix, compared to values obtained using supercritical CO₂. After a typical run (holding period of 30 min, continuous flow extraction of 3 h), the results indicate that the oil recovery was optimal with a recovery of 108.74% and a phytosterol composition of 7262.80 mg kg^?1 at relatively low temperature of 40 °C, a high pressure of 400 bar and at a high supercritical fluid flow rate of 20 ml/min in the presence of 2 ml/min EtOH as entrainer. The solubility of roselle seed oil increased with temperature at the operating pressures of 200, 300 and 400 bar. Supercritical fluid extraction involved a short extraction time and the minimal usage of small amounts of entrainer in the CO₂.     

Supercritical fluid extraction of walnut kernel oil

The objective of this study was to investigate the effects of the main process parameters on supercritical fluid extraction of walnut (Juglans regia L.) kernel oil. The recovery of walnut kernel oil was performed in a green and high-tech separation process. CO₂ and CO₂ + ethanol mixtures were used as the supercritical solvent. The extraction was carried out at operating pressures of 30, 40 and 50 MPa, operating temperatures of 313, 323 and 333 K, mean particle sizes of 1.78×10⁻⁴, 3.03×10⁻⁴, 4.78×10⁻⁴, 7.00×10⁻⁴ and 9.00×10⁻⁴ m, supercritical CO₂ (SC CO₂) flow rates of 1.67×10⁻⁸, 3.33×10⁻⁸, 6.67×10⁻⁸ and 13.33×10⁻⁸ m³/s and entrainer (ethanol) concentrations of 2, 4, 8 and 12 vol-%. Maximum extraction yield and oil solubility in SC CO₂ obtained at 50 MPa, 333 K, 9.00×10⁻⁴ m, 3.33×10⁻⁴ m³/h were 0.65 kg oil/kg of dry sample and 37.16 g oil/kg CO₂, respectively. The results obtained in this study showed that the crossover pressure effect of walnut kernel oil was at 30 MPa. At 30 MPa and 313 K, the obtained extraction yields above 4 vol-% ethanol reached the organic solvent extraction yield of 68.5 kg oil/kg dry sample. Extraction time was decreased significantly because of the higher solubility of walnut kernel oil in SC CO₂ + ethanol mixtures.     

Lycopene solubility in mixtures of carbon dioxide and ethyl acetate

In order to improve the efficiency of processes using supercritical (sc) carbon dioxide (CO₂) to micronize the carotenoid “lycopene”, it is important to know the solubility of lycopene in mixtures of the organic solvent ethyl acetate (EA) and the antisolvent CO₂ at elevated pressures. The solubility of lycopene has been determined for different temperatures (313–333 K), pressures (12–16 MPa) and CO₂ molar fractions (0.58–1). The obtained data show that CO₂ acts as an antisolvent in the system lycopene/EA/CO₂ in the range of CO₂/EA ratios studied. The solubility of lycopene is rather small with lycopene molar fractions ranging from 0.1 × 10⁻⁶ to 46 × 10⁻⁶. The solubility of lycopene increases with temperature, pressure and EA concentration.     

Supercritical CO2 extraction of lutein esters from marigold (Tagetes erecta L.) enhanced by ultrasound

As a novel technique, supercritical CO₂ (SC-CO₂) extraction enhanced by ultrasound was applied to the extraction of lutein esters from marigold and the extraction curves were described by Sovová model. The mass transfer coefficient in the solid phase (k_s) increased from 3.1 × 10⁻⁹ to 4.3 × 10⁻⁹ m/s due to ultrasound. The effect of extraction parameters including particle size of matrix, temperature, pressure, flow rate of CO₂, and ultrasonic conditions consisting of power, frequency and irradiation time/interval on the yield of lutein esters were investigated with single factor experiments. The results showed that the yield of lutein esters increased significantly with the presence of ultrasound (p < 0.05). The maximal yield of lutein esters (690 mg/100 g) was obtained for a particle size fraction of 0.245–0.350 mm, extraction pressure of 32.5 MPa, temperature of 55 °C and CO₂ flow rate of 10 kg/h with ultrasonic power of 400 W, ultrasonic frequency of 25 kHz and ultrasonic irradiation time/interval of 6/9 s.     

Experimental measurement and correlation for solubility of piroxicam (a non-steroidal anti-inflammatory drugs (NSAIDs)) in supercritical carbon dioxide

Since the knowledge of pharmaceutical solubilities in the supercritical carbon dioxide is one of the first essential necessities for designing the supercritical carbon dioxide-based processes, solubility of piroxicam a non-steroidal anti-inflammatory drug was experimentally measured. In this regard, a static method coupled with gravimetric method was used to measure the solubility of piroxicam in the supercritical carbon dioxide in temperature and pressure range of 308.15–338.15 K and 16–40 MPa, respectively. The obtained solubility data were in the range of 1.17 × 10⁻⁵ and 5.12 × 10⁻⁴ based on the mole fraction (mole piroxicam/(mole piroxicam + mole CO₂)) then modeled using four different density based correlations namely Bartle et al., Mendez-Santiago-Teja, Chrastil and Kumar and Johnston models. The results of error analysis revealed that the used correlations were potential to correlate the solubility of piroxicam with minimum and maximum average absolute relative deviation percents (AARD%) of 14.4% and 15.2%, respectively.     

Extraction of sunflower oil with supercritical CO2: Experiments and modeling

Extraction of sunflower oil from sunflower seeds (Heliantus annuus L.) using supercritical CO₂ was studied. The shrinking core model was applied to the modeling of the packed-bed extraction process. The experimental data were obtained for extraction conducted at the pressures of 20, 30, 40, 50 and 60 MPa; the temperatures of 313, 333 and 353 K, the CO₂ flow rates of 1–4, and 6 cm³ CO₂ min⁻¹; the mean particle diameters of 0.23, 0.55, 1.09, 2.18 mm. The supercritical CO₂ extraction process was modeled by a quasi steady state model as a function of extraction time, pressure, temperature, CO₂ flow rate, and particle diameter. The supercritical CO₂ extraction process. The intraparticle diffusion coefficient (effective diffusivity) D_e was used as adjustable parameter. The model using the best fit of D_e was correlated the data satisfactorily.     

Response surface optimization of Smyrnium cordifolium Boiss (SCB) oil extraction via supercritical carbon dioxide

In this study, the essential oil of aerial parts of a species of a plant called Smyrnium cordifolium Boiss (SCB) was extracted by supercritical CO₂. The essence was analyzed by the method of GC/MS. Design of experiments was carried out with response surface methodology by Minitab 16 software to optimize four operating variables of supercritical carbon dioxide (SC-CO₂) extraction (pressure, temperature, CO₂ flow rate and extraction dynamic time). This is the first report announcing optimization of the operation of supercritical extraction of SCB in laboratorial conditions. Optimizing process was done to achieve maximum yield extraction. Independent variables were dynamic time (t_d), pressure (P), temperature (T) and flow rate of SC-CO₂ (Q) in the range of 30–150 min, 10–30 MPa, 40–60 °C and 0.5–1.7 ml/min, respectively. The experimental optimal recovery of essential oil (0.8431, w/w%) was obtained at 13.43 MPa, 40 °C, 150 min (dynamic) and 1.7 ml/min (CO₂ flow rate).     

Measurement and correlation solubility of cefixime trihydrate and oxymetholone in supercritical carbon dioxide (CO2)

The equilibrium solubilities of cefixime trihydrate and oxymetholone in supercritical carbon dioxide (CO₂) were measured using a “static method”. Cefixime trihydrate is a cephalosporin antibiotic drug and Oxymetholone is a 17alpha-alkylated anabolic-androgenic steroid drug. The experimental measurements for cefixime trihydrate were performed at temperatures of 308, 318 and 328 K as well as pressure range from 183 to 335 bar. The solubility of oxymetholone was determined at 308, 318 and 328 K and pressure range from 121 to 305 bar. The experimental solubility data (mole fraction) for cefixime trihydrate and oxymetholone was greater than 1.6 × 10⁻⁷ and 1.6 × 10⁻⁵ and less than 3.02 × 10⁻⁷ and 1.49 × 10⁻⁴, respectively. The solubilities for two drugs in CO₂ were correlated by using four semi-empirical models such as Bartle, Kumar and Johnstone (K–J), Mendez-Santiago and Teja (M–T) and Chrastil models. The results obtained from the semi-empirical models show that there is good agreement between the experimental data and the results of semi-empirical models. By using the correlation results, the heat of drug–CO₂ solvation and heat of drug vaporization for cefixime–CO₂ and oxymetholone–CO₂ systems may be approximately estimated. Also, the Peng–Robinson (PR) cubic equation of state (CEOS) along with the van der Waals combining rule was applied to correlate the drugs solubilities in supercritical CO₂. The average absolute deviation between the experimental data and the results of PR equation for cefixime trihydrate and oxymetholone was 11.92% and 11.74%, respectively.     

A shrinking core model and empirical kinetic approaches in supercritical CO2 extraction of safflower seed oil

Utilization of supercritical CO₂ in safflower seed extraction was performed using a semi-batch extractor. Different extraction parameters, such as 40–60 MPa pressure, 323–347 K temperature, 20–76 min time, and 1–3 mL/min CO₂ flow rate were applied. A two-stage experimental design application was performed in order to maximize the oil yield. First of all, a 3² factorial design was applied to estimate the effect of the main factors and their interactions. The second part of the experimental design was improved and accelerated using the steepest ascent method. Optimum extraction parameters were determined to be 50 MPa pressure, 347 K temperature and 76 min time at a constant CO₂ flow rate (3 mL/min) according to the 2² design. Under these conditions, the oil yield obtained was 39.42%, comparable with Soxhlet extraction (40%) for 8 h. Shrinking core and empirical kinetic models were applied in order to generalize the extraction process. The predicted data was compatible with the experimental data.     

Measurement of entrainer effects of water and ethanol on solubility of caffeine in supercritical carbon dioxide by FT-IR spectroscopy

The solubilities of caffeine in supercritical CO₂, supercritical CO₂ + water, supercritical CO₂ + ethanol, and supercritical CO₂ + water + ethanol were measured with a circulation-type apparatus combined with an on-line Fourier transform infrared (FT-IR) spectrometer at 313.2 K and 15.0 MPa. The solubilities of caffeine were determined with the peak absorbances of caffeine at 1190 cm⁻¹. The solubilities of caffeine increase until water is saturated in supercritical CO₂. The maximum increase rate is 22%. In CO₂ + caffeine + ethanol system, the solubilities of caffeine increase with increasing the concentration of ethanol. The solubility of caffeine becomes five times when 1000 mol m⁻³ of ethanol is added. In CO₂ + caffeine + water + ethanol system, the solubilities of caffeine are smaller than those with single entrainer of water or ethanol. The shape of the peaks of two CO stretching bands for caffeine were changed by the addition of ethanol. It was confirmed that the interaction species of caffeine interacting with ethanol are produced by deconvolution of the CO stretching bands. The enhancement of solubility for caffeine in supercritical CO₂ by the addition of ethanol is due to the hydrogen bonding between caffeine and ethanol.     

Enrichment of Nigella damascena extract with volatile compounds using supercritical fluid extraction

β-Elemene, germacrene A and damascenine were extracted from lady-in-a-mist (Nigella damascena L.) seeds with supercritical carbon dioxide at 10–30 MPa and 40–60 °C. The influence of supercritical fluid extraction (SFE) conditions on the yield and concentration of volatiles in the extract and the extraction kinetics were studied. The extraction yields and the apparent solubility of volatile compounds increased with increasing density of CO₂. The highest total yield was obtained at 30 MPa and 40 °C but the selectivity for volatiles was low under these conditions. With respect to both yield of volatiles and their concentration in extract, the best results were at 12 MPa and 40 °C, either with one separator or with additional separator maintained at 5 MPa and 25 °C. The yields of β-elemene, germacrene A and damascenine reached 0.72, 3.31 and 3.65 mg g⁻¹ and their concentration in the extract was 2.62, 12.04 and 13.28 wt.%, respectively. Though the yields of germacrene A and damascenine were by about 20% higher using Soxhlet extraction with hexane than using SFE, their concentration in the extract where fatty oil prevailed was only 1.19 and 1.20 wt.%, respectively. Under the conditions of hydrodistillation, partial conversion of germacrene A to β-elemene occurred and its yield was higher than using the other methods but the composition of volatiles in the SFE extracts better corresponds to the original raw material.     

Extraction of curcuminoids from deflavored turmeric (Curcuma longa L.) using pressurized liquids: Process integration and economic evaluation

Pressurized liquid extraction (PLE) of curcuminoids from deflavored turmeric rhizomes was optimized. The rhizomes were initially deflavored by extraction with supercritical CO₂. Immediately after SFE, PLE process was performed using ethanol as the solvent and a static extraction time of 20 min, and the independent variables were the temperature (333–353 K) and pressure (10–35 MPa). The results indicate that the optimum extraction temperature and pressure were 333 K and 10 MPa, respectively. PLE required three and six times less extraction time than low-pressure solvent extraction and Soxhlet extraction, respectively, to produce similar extraction yields. The cost of manufacturing (COM) decreased from US$ 94.92 kg⁻¹ to US$ 88.26 kg⁻¹ when the capacity of the two-extractor system increased from 0.05 m³ to 0.5 m³ and from US$ 94.92 kg⁻¹ to US$ 17.86 kg⁻¹ when the cost of the raw materials decreased from US$ 7.91 kg⁻¹ to US$ 0.85 kg⁻¹ for a two 0.05 m³ extractor system.     

Optimization of essential oil supercritical extraction from Algerian Myrtus communis L. leaves using response surface methodology

The present work deals with the application of the supercritical fluid extraction process to extract essential oils from the leaves of an Algerian myrtle plant (Myrtus communis L.). Using the surface response methodology, an optimization of the extraction recovery was carried out, varying the pressure in the range of [10–30 MPa], the temperature within [308–323 K], a solvent flow rate fixed at 0.42 kg h⁻¹ and a mean particle diameter equal to 0.5 mm or less than 0.315 mm. The maximum value of essential oil recovery relative to the initial mass of leaf powder was 4.89 wt%, and was obtained when the SC–CO₂ extraction was carried out under 313 K, 30 MPa and with a particle diameter less than 0.315 mm. A second-order polynomial expression was used to express the oil recovery. The calculated mass of recovered oil using the response surface methodology was very close to the experimental value, confirming the reliability of this technique.     

Extraction of jojoba oil with liquid CO2 + propane solvent mixtures

In this work liquid CO₂/propane mixtures were used to extract jojoba oil from oilseeds. First, experiments at 313 K and pressures of 70 bar and 200 bars were carried out on jojoba oil deposited on glass spheres, using different solvent concentrations (30 wt%, 50 wt% and 70 wt% CO₂), to assess the influence of the solvent composition and phase behavior on the extraction rate. Then, jojoba oil was extracted from the milled seeds under homogeneous liquid conditions, using solvent mixtures containing 30 wt% and 50 wt% CO₂ at 70 bar and 313 K. A solvent mixture with 30 wt% CO₂ exhibited good solvent power. Oil extraction yields of 98% were obtained using a minimum solvent to oilseed mass ratio of 5 g solvent/g oilseed and operating the extractor at 313 K and 70 bar.     

Supercritical carbon dioxide extraction of Gac oil

Supercritical carbon dioxide extraction of Gac (Momordica cochinchinensis Spreng) aril was performed at pressures ranging from 200 to 400 bar, temperatures from 313 to 343 K and specific flow rates from 50 to 90 kg h⁻¹ CO₂ kg⁻¹ Gac aril. Total oil recovery and carotenes concentration were investigated in the course of extraction. Mathematical modelling of oil solubility data was also performed. The results showed that at specific flow rate of 70 kg h⁻¹ kg⁻¹, pressure of 400 bar and temperature of 343 K, Gac oil recovery exceeded 95% after 120 min of extraction. Gac oil loading of supercritical carbon dioxide was successfully described by Chrastil's model. Carotenes concentration of extracted Gac oil was found at level of thousands of ppm.     

Supercritical CO2 and highly selective aromatase inhibitors: Experimental solubility and empirical data correlation

The solubility of highly selective and potent third-generation aromatase inhibitors includes the non-steroidal agents letrozole and anastrozole and the steroid exemestane in supercritical carbon dioxide (SC-CO₂) has been investigated. The experiments were carried out using the simple and static method at pressures in the range of 12.1–35.5 MPa and temperatures ranging from 308 to 348 K. The mole fraction solubilities ranged from 0.22 × 10⁻⁵ to 1.88 × 10⁻⁴. Solubility data were correlated using six empirical models (Chrastil model, dV–A model, K–J model, Bartle model, Yu model and Gordillo model). The results showed that these models can be applied to satisfactory solubility predictions at different pressures and temperatures. A comparison among the six models revealed that the K–J, and Gordillo models gave much better correlations of the solubility data with an average absolute relative deviation (AARD%) ranging from 0.2 to 2.3 and from 1.6 to 2.5%, respectively. Using the correlation results, the heat of drug–CO₂ solvation and that of drug vaporization was separately approximated in the range of −17.3 to −17.5 and 93.0–112.1 kJ mol⁻¹.     

Static mixers as heat exchangers in supercritical fluid extraction processes

The performance of a Kenics static mixer as a heat-transfer device for supercritical carbon dioxide (CO₂) flow is studied and compared with conventional tube-in-tube heat exchangers. Measurements were carried out at pressures ranging from 8 to 21 MPa, temperatures from 283 to 323 K, and mass flowrates from 2 to 15 kg/h. The corresponding Reynolds and Prandtl numbers, at bulk conditions, ranged between 10³ and 2 × 10⁴ and between 2 and 7, respectively. The temperature increase experienced by the supercritical CO₂ stream varied between 10 and 35 K. The heat fluxes obtained with the static mixer are one order of magnitude higher than the ones observed with a tube-in-tube heat exchanger for the same set of operating conditions. The heat-transfer enhancement is caused by the cross-sectional mixing of the fluid and to a lesser extent by conduction across the metallic mixing elements. Heat-transfer is also affected by temperature-induced variation of physical properties, especially in the pseudocritical region of the fluid. From the experimental data, a correlation was developed for convective heat-transfer to supercritical CO₂ in terms of the Nusselt number.     

Influence of pretreatment on supercritical CO2 extraction from Nannochloropsis oculata

This work demonstrates that supercritical carbon dioxide extraction is efficient for the complete recovery of neutral lipids from microalgae with a water content up to 20 wt%, allowing thus a further full characterization of this oil. This is a first useful step in the framework of lipid production from microalgae either for nutraceutical, food or for energy applications. This study is particularly focused on the influence of the pretreatments upon extraction kinetics and yields. This study proposes a complete study at laboratory scale (10 g per batch of dry biomass) of the influence of pretreatments (type of drying and grinding) and of water content on the extraction kinetics and yields as well as on the oil composition in terms of lipidic classes and profiles. Two drying pretreatments (drying under air flow and freeze-drying) applied on Nannochloropsis oculata were studied. Extraction experiments were carried out at 40 MPa, 333 K, with a carbon dioxide flow rate of 0.5 kg h⁻¹ and for different granulometries. Results showed that drying under air flow at 308 K is the most adequate pretreatment leading to the most rapid kinetics. Whatever the pretreatment used, the extracted oil contains more than 90 wt% of triglycerides and does not contain phospholipids. As expected, the smaller the particle size, the faster the extraction kinetics. Finally, an increase in the biomass water content up to 20 wt% increases the global extraction kinetics (extraction of both water and oil) but appears to have no influence on oil extraction yields. Moreover, the extraction of neutral lipids happens to be complete for a CO₂/charge mass ratio ranging from 30 to 130 depending on the operating conditions and on the characteristics of the treated biomass. Finally, pilot scale experiments were performed with batches up to 15 kg in order to evaluate the influence of pressure and particle size on the extraction kinetics and yields. Extracts obtained at 333 K with operating pressures of 50 MPa and 85 MPa have similar compositions and do not contain phospholipids.     

Supercritical CO2 extraction of rosehip seed oil: Fatty acids composition and process optimization

Rosehip seed oil has been extracted using supercritical CO₂ at various operating conditions to optimize extraction process. The effect of extraction conditions on the fatty acids composition in the oil was also observed. The extraction conditions were as follows: pressures (P) of 150, 300 and 450 bar, temperatures (T) of 40, 60 and 80 ^oC, and CO₂ flow rate (F) of 2, 3 and 4 mL/min. A full 3³ factorial design coupled with statistical and graphical analysis of the results, by using analysis of variance (ANOVA) was applied to optimize variables in the process of rosehip seed oil extraction with SC-CO₂. The experimental result showed that the seed oil extracted mainly contained linoleic acid (C18:2) as the most abundant followed by linolenic (C18:3), palmitic (C16:0) and stearic acid (C18:0); and the extraction conditions influenced the fatty acids composition. The analysis of experimental design for process optimization results demonstrates that temperature and pressure were to be the influential variables on the extraction yield of seed oil. Furthermore, the apparent solubility of oil in SC-CO₂ was also determined from the experimental data and correlated using empirical equations for further model developing.     

Near critical and supercritical impregnation and kinetic release of thymol in LLDPE films used for food packaging

The impregnation of organic compounds in polymeric materials using supercritical carbon dioxide (scCO₂) is a well-known technique, which is currently used in drug/polymer formulation. In this work, near critical and supercritical impregnation of thymol in linear low-density polyethylene (LLDPE) films was done in order to develop a new technique for preparation of active polymers to be used as food packages. The properties of thymol as a natural antimicrobial and antioxidant agent have motivated this study about the assessment of its migration from the polymer to different food simulant. Impregnation assays of thymol in LLDPE films were done in a high-pressure cell, where pure thymol was solubilized in supercritical carbon dioxide at 313 K and pressures varying from 7 to 12 MPa. This procedure allowed the preparation of plastic films with thymol concentrations ranged between 5100 and 13,200 ppm. Migration tests showed that the pressure applied during the impregnation procedure is a key parameter that affects the content of the active compound into the polymer, since thymol solubility in scCO₂ and absorption phenomena in the polymer increased with the pressure. The correlation between experimental data and a phenomenological transfer model allowed the estimation of the diffusion coefficient of thymol in LLDPE, which was ranged from 7.5 × 10⁻¹³ to 3.0 × 10⁻¹² m² s⁻¹.