The effect of equipment design and process scale-up on supercritical CO2 extraction: Case study for Silybum marianum seeds

https://doi.org/10.1016/j.supflu.2022.105676Get rights and content

Highlights

  • Kinetic of the SFE process significantly differs for four units having volumes from 0.28 to 40 L.

  • Three lab-scale units showed variation in extraction yield from 11.9 % to 18.0 %.

  • Sovova’s model can be used for description the presented SFE process phenomenology.

  • Scale-up study showed a good agreement between laboratory and semi-industrial scale unit.

  • Economic evaluation indicated optimal industrial process of 3 h and extract cost of 35.9 EUR/kg.

Abstract

The present study was focused on improving the supercritical fluid extraction process dedicated to the recovery of highly valuable extract from Silybum marianum seeds. At first, the process was tested using supercritical CO2 at temperatures of 40 and 80 °C and pressures ranging from 100 to 300 bar. Obtained values were described using two literature models with acceptable agreement of experimental and calculated data. Further, the effect of the equipment design and process scale-up was investigated in four high-pressure units having volumes in the range from 280 mL to 40 L resulting in extraction yields up to 26 %. Based on the values of the parameters determined at the laboratory scale, the extraction process was successfully transferred to a semi-industrial scale. Finally, estimated costs for supercritical extract production in Poland for the year 2021 using a 1000 L high-pressure unit amounted to the price of 35.9 EUR per kg of extract.

Introduction

The supercritical fluid extraction (SFE) process implies a separation of one compound (extract) from another (matrix) using a supercritical fluid as an extracting solvent. The SFE was shown to be an effective technique for the separation of commercially important extracts from raw plant materials that are rich in valuable bioactive compounds [1], [2], [3]. Besides being biologically active, supercritical extracts are pure and solvent-free, and therefore are eligible for application in food production, medicine, and pharmacy. Due to its high diffusivity, low viscosity, and near-zero surface tension, the most commonly used supercritical solvent is carbon dioxide (CO2) [2]. As a solvent, it has the GRAS (Generally Recognized as Safe) status, good solvation power, and can be easily and completely removed from the final product [2].

Although the SFE technique is widely used for the isolation of valuable plant extracts, a scale-up from laboratory to industry level still presents a challenging step. The Cambridge Dictionary defines scale-up as increasing something in size, amount, or production. If this definition is taken into account, one can find a certain number of studies available in the literature that report the scale-up of SFE processes [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14]. For example, Paula et al. [8] performed SFE scale-up for B. dracunculifolia leaves by variation in plant material mass from 12.5 g to 70 g and Fernández-Ponce et al. [9] performed SFE scale-up for mango leaves by an increase in extractor vessel volume from 0.1 L to 5.0 L. However, these studies presented scale-up on small-scale systems (due to the high costs of experimental tests at the industrial scale) while prediction models were used for a larger scale. Nevertheless, the resulting data were generally unreliable concerning the differences observed in processes conducted on significantly different equipment scales [3], [9]. The study of intermediate-scale experiments (semi-industrial scale) is a better strategy as it considers the restrictions that may occur on an industrial scale. Therefore, a scale-up of the SFE process proposed in this study was investigated in four high-pressure units, ranging from laboratory to semi-industrial scale, using supercritical CO2 (scCO2) as the working fluid. Silybum marianum (S. marianum) was selected as a model plant in this case study due to its high value as an industrial material. In 2018, S. marianum was placed among the eight top-selling herbal-based dietary supplements in the US natural retail outlets with sales reaching 9.0 million EUR and ranked 20th in the mainstream multi-outlet channel market in the US with total sales of 14.2 million EUR [15]. ScCO2 extracts from S. marianum contain valuable bioactive compounds such as fatty acids [16], [17], [18], α- and γ-tocopherol [1], [19], and silymarin [17]. Although there are several reports available in the literature that compare the effects of S. marianum seeds pretreatment [1], particle size [17], the SFE process pressure/temperature [17], [19], and CO2 flow rate [17] on the extraction yield and extract composition, the present study is the first report that describes the effects of high-pressure equipment design and the process scale-up from 0.28 to 40 L extractor volume. In addition, two literature models were compared for the description of the obtained SFE process kinetics. The first model is based on the adsorption-desorption mechanism that can be described in the following steps: (1) adsorption-desorption equilibrium of extractable component from solid tissue, (2) diffusion of extractable component dissolved in supercritical fluid to the surface, and (3) mass transfer through the external film into the bulk [19], [20], [21]. The second model introduced by Sovova [22] is to date the most used. The main assumption of this model is that, during the pretreatment of plant material, a fraction of oil-containing cells is destroyed, making the oil easily accessible to scCO2. The extraction process is thus divided into two periods. In the first period, easily accessible oil is extracted and an external mass transfer and/or oil solubility in scCO2 govern the process rate. In the second period, the oil that remained within the cells of the plant material is extracted, with internal diffusion governing the overall extraction rate.

It is clear that basic research, involving a literature review of the yields, composition, and bioactivity, shows potential for industrial application of the extract from S. marianum seeds. On the other hand, the analysis taking into account factors such as raw material prices, equipment, and labor costs, as well as prices on the market for a specific product, etc., can provide an approximate evaluation of the process economics and potential for investment. In that sense, this study provides, for the first time, information on the estimated costs for S. marianum extract production based on data obtained using a semi-industrial unit for the SFE process.

Section snippets

Materials

The S. marianum seeds were obtained from the Institute Dr. Josif Pancic (Pancevo, Serbia) and the Prowana (Radzymin, Poland). The seeds were stored in a dark place at room temperature (20 °C) prior to the extraction. Commercial CO2 (purity 99.9 %) purchased from the Messer-Tehnogas (Belgrade, Serbia) was used for the extraction process in laboratory scale units (units I and II, described in detail below) and CO2 (purity 99.9 %) produced by the Grupa Azoty (Zaklady Azotowe “Pulawy” S.A., Pulawy,

Plant material analysis and pretreatment

The moisture content in S. marianum seeds was determined using a laboratory moisture analyzer (MAC 50/1/WH, RADWAG®, Radom, Poland). Before the extraction, seeds were milled using a basic mill (Ika® A11, Warszawa, Poland) and sieved. The average particle size of the plant material used for SFE processes was 0.4 mm. The plant material density was 1192 ± 114 kg/m3. The bed density and porosity were 376 ± 6 kg/m3 and 68.2 ± 2.5 %, respectively.

Supercritical fluid extraction processes

Four high-pressure units were tested for the SFE

The SFE process in the lab-scale unit

The influence of pressure and temperature in the selected range was investigated using S. marianum seeds grown in Serbia and Unit I. The results obtained at 40 °C at various pressures are presented in Fig. 2a. It can be seen that pressure significantly affected the extraction yield. The highest yields obtained at 40 °C and 100, 200, and 300 bar were 1.4 %, 8.9 %, and 11.9 %, respectively. An increase in extraction yield with increasing pressure from 160 to 220 bar at 40 °C was also reported by

Conclusion

This study presented for the first time the scale-up of the SFE process from S. marianum seeds by in situ study (extractor vessels ranging from 0.28 to 40 L) as well as by theoretical model (for two 500 L extractor vessels). First, it was determined at the laboratory scale that the extraction yield increases significantly with increasing pressure from 100 to 300 bar for the SFE process performed at 40 °C and 80 °C. Further, it was shown that the construction of a high-pressure unit on a

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

Author S. Milovanovic acknowledges a scholarship from the Polish National Agency for Academic Exchange (NAWA), Warsaw, Poland (the agreement number PPN/ULM/2020/1/00023/U/00001). This work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia (Contract No. 451-03-68/2022-14/200135). Work was carried out in the frame of the COST-Action "Green Chemical Engineering Network towards upscaling sustainable processes" (GREENERING ref. CA18224)

References (49)

  • R.M. Hall et al.

    Simulating large scale SFE applied to recover bioactive compounds from papaya seeds

    J. Supercrit. Fluids

    (2018)
  • A. López-Padilla et al.

    Supercritical carbon dioxide extraction of Calendula officinalis: Kinetic modeling and scaling up study

    J. Supercrit. Fluids

    (2017)
  • H.T. Çelik et al.

    Extraction of oil and silybin compounds from milk thistle seeds using supercritical carbon dioxide

    J. Supercrit. Fluids

    (2015)
  • I. Lukic et al.

    Separation of high-value extracts from Silybum marianum seeds: Influence of extraction technique and storage on composition and bioactivity

    Lwt

    (2022)
  • M. Hadolin et al.

    High pressure extraction of vitamin E-rich oil from Silybum marianum

    Food Chem.

    (2001)
  • H. Sovova

    H. sovova

    Chem. Eng. Sci.

    (1994)
  • I. Zizovic et al.

    Supercritical carbon dioxide extraction of essential oils from plants with secretory ducts: Mathematical modelling on the micro-scale

    J. Supercrit. Fluids

    (2007)
  • Z.S. Zhang et al.

    Constituents and thermal properties of milk thistle seed oils extracted with three methods

    Lwt

    (2020)
  • P.T.V. Rosa et al.

    Rapid estimation of the manufacturing cost of extracts obtained by supercritical fluid extraction

    J. Food Eng.

    (2005)
  • Z. Huang et al.

    Benign recovery of carotenoids from Physalis alkekengi L.var. francheti through supercritical CO2 extraction: Yield, antioxidant activity and economic evaluation

    J. CO2 Util.

    (2020)
  • M.M.R. De Melo et al.

    Supercritical fluid extraction of spent coffee grounds: measurement of extraction curves, oil characterization and economic analysis

    J. Supercrit. Fluids

    (2014)
  • G.A. Núñez et al.

    Supercritical CO2 oilseed extraction in multi-vessel plants. 2. effect of number and geometry of extractors on production cost

    J. Supercrit. Fluids

    (2014)
  • P.C. Veggi et al.

    Production of phenolic-rich extracts from Brazilian plants using supercritical and subcritical fluid extraction: Experimental data and economic evaluation

    J. Food Eng.

    (2014)
  • K. Tomita et al.

    Extraction and solubility evaluation of functional seed oil in supercritical carbon dioxide

    J. Supercrit. Fluids

    (2013)
  • Cited by (9)

    • Supercritical fluid extraction of bioactive compounds from quinilla (Manilkara bidentata) seed

      2023, Journal of Supercritical Fluids
      Citation Excerpt :

      As the Bi number is the ratio of internal diffusion resistance to the convection resistance, a Bi close to unity in this table demonstrate the reliability of the assumption of significant internal and external mass transfer in the mathematical model. Although Bi numbers for SFE from most of biological substrates are very high, Bi numbers smaller than 10 have been recorded for SFE from Silybum marianum seeds [49], Pepper, Korean green tea, and neem seeds, among others [50]. Fig. 6 compares the SFE curve adjusted by the model to the experimental data.

    • Dandelion seeds as a new and valuable source of bioactive extracts obtained using the supercritical fluid extraction technique

      2022, Sustainable Chemistry and Pharmacy
      Citation Excerpt :

      The milled material that had diameters ranging from 0.2 to 0.6 mm was used for the SFE process. The SFE process was performed in a high-pressure unit (SITEC, Zurich, Switzerland) described in detail elsewhere (Milovanovic et al., 2022b). The schematic representation is shown in Fig. 1.

    • Green manufacturing of high-value extracts from milk thistle seeds: Parameters that affect the supercritical CO<inf>2</inf>extraction process

      2022, Journal of CO2 Utilization
      Citation Excerpt :

      The material collected between 0.2 and 0.6 mm sieves was further used. The effect of parameters on the SFE process kinetics was investigated using three high-pressure units: Autoclave Engineers Screening System (AESS, Autoclave Engineers Group, USA), High-Pressure Extraction-Adsorption Unit (HPEA, Eurotechnica GmbH, Germany), and Micronisation Unit (MCR, Sitec, Switzerland) previously described in detail [19]. The AESS unit was used for studying the influence of seeds storage (1 year vs. 8 years) and cultivation year (2011 vs. 2019).

    View all citing articles on Scopus
    View full text