壓縮包內(nèi)含有CAD圖紙和說明書,均可直接下載獲得文件，所見所得，電腦查看更方便。Q 197216396 或 11970985
Industrial Crops and Products
Volume 22, Issue 3, November 2005, Pages 207–222
Oil extraction of oleic sunflower seeds by twin screw extruder: influence of screw configuration and operating conditions
· I. Amalia Kartika,
· P.Y. Pontalier,
· L. Rigal
· Laboratoire de Chimie Agro-Industrielle, UMR 1010 INRA/INP-ENSIACET, 118 Route de Narbonne, 31077 Toulouse Cedex 4, France
· Received 8 October 2004. Accepted 6 January 2005. Available online 14 March 2005.
· http://dx.doi.org/10.1016/j.indcrop.2005.01.001, How to Cite or Link Using DOI
· Permissions & Reprints
Abstract
The objective of this study was to investigate the effects of screw configuration, position of screw elements and spacing between them allowing to realize oil extraction of oleic sunflower seeds on a twin-screw extruder. Experiments were conducted using a co-rotating twin-screw extruder (Model Clextral BC 45, France). Twelve screw profiles were examined to define the best performance (oil extraction yield, oil quality, mean residence time, and thermo-mechanical energy input) by studying the influence of operating conditions temperature pressing, screw rotation speed and seed input flow rate.
Generally, the position and spacing between two screw elements affected oil extraction yield. An increase of oil extraction yield was observed when the reversed screw elements were configured with increased spacing between elements or/and with smaller pitch screw. In addition, more oil extraction yield was produced as the temperature pressing, screw rotation speed and seed input flow rate were decreased. The higher oil extraction yield was obtained under operating conditions 80?°C, 60?rpm and 24?kg/h. Furthermore, the operating parameters influenced energy input and mean residence time of matter. Both energy input and mean residence time increased when the temperature pressing increased. However, increase of screw rotation speed and seed input flow rate decreased mean residence time. Effect of the operating parameters
on oil quality was unimportant. In all experiments tested, the oil quality was very good. The acid value was below 2?mg?KOH/g of oil and total phosphorus content was very poor, below 40?mg/kg.
Keywords
Twin-screw extruder;
Oleic sunflower;
Oil and extraction
1. Introduction
Industrial oil extraction from oleaginous seeds is commonly realized through mechanic pressing with a hydraulic or single expeller press, followed by solvent extraction. The hydraulic press is most effective but this process is discontinuous. Recently, the application of continuous oil extraction process using extrusion technology gets some attentions from few researchers ( [Vadke and Sosulski, 1988], [Isobe et al., 1992], [Clifford, 2000], [Wang and Johnson, 2001], [Crowe et al., 2001], [Singh et al., 2002]?and?[Zheng et al., 2003]). Extensive studies on extrusion processing of oilseeds using twin-screw extruder to generate oil ( [Guyomard, 1994], [Bouvier and Guyomard, 1997], [Dufaure et al., 1999a]?and?[Dufaure et al., 1999b]) and fatty acid ester (Lacaze et al., 1996) have been successfully carried out too.
The continuous oil extraction of oilseeds is widely carried out in a single-screw press. This type of machine consists of a single-screw of variable pitch and channel depth, slowly rotating in a cage type barrel (Isobe et al., 1992). Transport of material in a single-screw press depends mainly on friction between the material and the barrel's inner surface and screw surface during screw rotation. Thus, a solid core component is often necessary to produce the friction. This causes excess frictional heat, large energy consumption and oil deterioration. Furthermore, single-screw presses provide inadequate crushing and mixing if they are not configured with breaker bars or other special equipment. A twin-screw oil press can be expected to solve these problems because of the higher transportation force, similar to a gear pump, and better mixing and crushing at the twin-screw interface. In addition, energy consumption of the twin-screw press is more efficient ( [Isobe et al., 1992]?and?[Bouvier and Guyomard, 1997]).
The preparation of the raw material, such as size reduction, flaking, cooking and moisture preconditioning of the seeds are necessary to improve single-screw press performance, as well as the mechanical design of the worm and barrel assembly. Maximum pressure increased, and press throughput and residual oil (RO) in presscake decreased, with a reduction in choke opening and with lowering shaft speed of the single-screw press (Vadke and Sosulski, 1988). In addition, when whole seeds or flakes were preheated in the range 40–100?°C, the pressure and press throughput increased and RO decreased. Press throughput and oil output both achieved maximal at canola seed moisture content of 5%, while the RO showed a continuous increase with increasing seed moisture content. Oil recovery of crambe seed extraction on a single-screw press and sediment content increased, and residual oil and pressing rate decreased as seed moisture content decreased (Singh et al., 2002). In the case of flaxseed, oil recovery increased as whole seed moisture content increased (Zheng et al., 2003).
Twin-screw extruder played an important role in the food industry to transformer the material physically and chemically in a single step. The main application of twin-screw extruder is widely found in the production of various products such as snacks, cereals and pet food. In the present day, several studies have expanded the utility of twin-screw extruder as a reactor to conduct a thermo-mechano-chemical action plus a liquid/solid extraction, as in hemicellulose extraction ( [N’Diaye et al., 1996]?and?[N’Diaye and Rigal, 2000]), in a continuous mode.
The great capability of twin-screw extruder to conduct diverse functions and processes has a good correlation with advantages of their characteristics. According to Dziezak (1989), those advantages include (i) ability to provide better process control and versatility, especially in pumping efficiency, controlling residence time distribution and uniformity of processing, (ii) ability to process specialty formulation, in which the single-screw extruder can not handle it and (iii) flexibility to design machine, which permits self-cleaning mechanisms and rapid changeover of crew configuration without disassembling the extruder.
Twin-screw extruder is mainly built by elements, namely screw, including (i) forward pitch screw, principally conducts a conveying action, (ii) monolobe paddle (DM), primarily exerts a radial compression and shearing action, (iii) bilobe paddle (BB), exerts a significant mixing and shearing actions, a conveying and axial compression actions in combination with forward pitch screw, and (iv) reversed pitch screw, carries out intensive shearing and considerable mixing, and exerts a strong axial compression in combination with forward pitch screw (Rigal, 1996). The arrangement of different characteristics of screw elements (pitch, stagger angle, length) in different positions and spacing determine screw profile/configuration that is main factor influencing performance (product transformation, residence time distribution, mechanical energy input) during extrusion processing ( [Gogoi et al., 1996a], [Choudhury et al., 1998], [Gautam and Choudhury, 1999a]?and?[Gautam and Choudhury, 1999b]). Furthermore, by modularity of its configuration and screw profile, the twin-screw extruder enables a large number of basic operations, such as material transport, grinding/crushing, mixing, chemical reaction, liquid–solid extraction, liquid–solid separation and drying, to be carried out in a single step (Rigal, 1996) in which the conventional presses can not handle it.
The great amounts of researches concerning on study of screw configuration are found in the agro-industry field, particularly, for starch transformation. Screw configuration by placing longer reversed screw element (Barres et al., 1990) or nearer from the die (Colonna et al., 1983) increased starch breakdown. Furthermore, the systematic increases in starch breakdown ( [Gautam and Choudhury, 1999a]?and?[Gautam and Choudhury, 1999b]) and in mechanical energy input and water solubility index (Choudhury and Gautam, 1998) were observed as the mixing elements were moved farther away from the die, with longer elements, and with increased spacing between elements. The incorporations of reversed screw element (Gogoi et al., 1996b), kneading element (Choudhury et al., 1998) and mixing elements combination (Gogoi et al., 1996a) increased specific mechanical energy, expansion ratio and water solubility index.
In oil extraction case, a significant increase in oleic sunflower oil yield was observed as the length and the pitch of reversed screw elements were increased and reduced, respectively (Dufaure et al., 1999a). In addition, oil yield could be improved with adding the monolobe paddle screws (DM) in module 5 just above the filtration module and with increasing the stagger angle of bilobe paddle screws (BB). Furthermore, a investigation of continuous oil extraction method using extruder divided into two zones, (i) twin-screw zone, which was built from two co-rotating and co-penetrating screws and (ii) double single-screw zone, which was constructed from two co-rotating single-screw, increased oil extraction yield of whole sunflower seeds up to 90% with residual oil content in cake meal lower than 15% (Bouvier and Guyomard, 1997).
As well as screw configuration, the preparation of the raw material is also important to enhance oil extraction. The oil extraction yield from whole sunflower seeds in a contra-rotating twin-screw press was low (75%), but could be increased to 93.6% if raw materials was dehulled (Isobe et al., 1992). In the case of colza seeds, the oil extraction yield from dehulled seeds was always lower than whole seeds (Bouvier and Guyomard, 1997). The oil yield increased up to 80% with a high temperature pressing and natural moisture content of oleic sunflower seeds (Dufaure et al., 1999a).
In relation to these results, the studies more systematic should be realized to improve oil extraction yield and to reduce residual oil content in cake meal. Moreover, it has to optimize operating conditions and characterize oil extraction quality, residence time distribution and mechanical energy input.
This study purposed to evaluate the effects of screw configuration and operating parameters such as temperature pressing, screw rotation speed and seed input flow rate on oil extraction of oleic sunflower seeds using twin-screw extruder. The characterization of extraction performance was observed by the determinations of extraction yield, oil quality, mean residence time and thermo-mechanical energy input.
2. Materials and methods
2.1. Materials
All trials were carried out with whole and uncleaned sunflower seeds (3–6% of impurities content), which were supplied by La Toulousaine de Cereales (France). These seeds were from oleic type with the average acidity of 0.95%. All solvent and chemicals were analytical grades that were obtained from Sigma–Aldrich, Fluka, Prolabo and ICS, France.
The oil content of seed used in the first set of tests (screw configuration study), expressed in relation to the dry matter content of uncleaned seed, was 44.74% (NF V03-908). The seed moisture content at storage was 8.27% (NF V03-903).
In the second set of tests (operating conditions study), the oil content of uncleaned oleic sunflower seed was 42.49% in relation to the dry matter. The seed moisture content at storage was 7.13%.
The seeds were neither dehulled nor flaked prior to entering twin-screw extruder. Dehulling of oilseeds is adapted to the quality of the hulls and how easily they can be removed. Generally, European factories do not hull sunflower seeds. The hull by itself constitutes nearly 25% of the seed ( [Isobe et al., 1992]?and?[Karleskind, 1996]). Flaking of seeds is extremely important as a solid-liquid extraction by solvent is conceived after mechanic pressing. In this study, the oil extraction was only carried out with mechanic pressing, without solvent extraction.
2.2. Twin-screw extruder
Experiments were conducted with a co-rotating twin-screw extruder (Model Clextral BC 45, France). The extruder was built with seven modular barrels, each 200?mm in length, and different twin-screws which had segmental screw element each 50 and 100?mm in length. Four modules were heated by thermal induction and cooled by water circulation. Material was fed into the extruder inlet port by a volumic screw feeder (type 40, Clextral, France). A filter section consisting of six hemispherical dishes with perforation of 1?mm in diameter was outfitted on module 5 to separate extracted oil. Furthermore, screw rotation speed, seed input flow rate and barrel temperatures were monitored from a control panel. Fig. 1 shows the schematic modular barrel of twin-screw extruder.
Fig. 1.?Schematic modular barrel and global screw configuration of twin-screw extruder BC 45.
View thumbnail images
2.3. Experimental
2.3.1. Screw configuration study
Thirteen screw configurations were evaluated in this experiment (Fig. 2). First configuration was the best configuration obtained in previous work (Dufaure et al., 1999a). Seven screw configurations (profiles 1–7) were built by placing monolobe paddles (DM) and bilobe paddles (BB) elements at different position. Bilobe paddles (BB) were located at 50 and 150?mm from the right side module 5 (filter module). Monolobe paddles (DM) were positioned by spacing 50 or 100?mm from BB. Furthermore, position and interval effects of two reversed screw (CF) elements were studied by placing a CF element at 50 and 100?mm from the left side module 5 and by spacing second CF element at 0, 50, 100 or 150?mm from first CF element. Another five configurations (profiles 8–12) were developed from profile 5 by modifying the position of DM elements and/or by reducing the pitch of reversed screw elements. DM elements were positioned at module 2 before BB elements and the pitch of CF elements were decreased from 25 to 15.
Fig. 2.?Screw configurations for oil extraction of oleic sunflower seeds.
View thumbnail images
For all profiles, barrel temperature, screw rotation speed and seed input flow rate were fixed at 80?°C, 60?rpm and 24?kg/h, respectively. In the case of profiles 8, 11 and 12, screw rotation speed was fixed at 100 and 70?rpm. To ensure a stable flow rate and temperature, the extruder was operated for approximately 20–25?min before processing the actual samples. Upon achieving steady operation, filtrate (oil containing the foot) and cake meal samples were immediately collected over a period of 20?min. Sample collection time was determined with a stopwatch. The filtrate and cake meal were weighed. The filtrate was further centrifuged to separate the foot from the oil. The moisture and residual oil contents of the cake meal were measured according to standards NF V03-903 and NF V03-908. For each tests, sample collection was carried out just the once. The calculation of oil extraction yield was determined by relationships following:
(1)
(2)
where Qs is the inlet flow rate of seed (kg/h), QF and Qc are respectively the outlet flow rate of the filtrate (kg/h) and the cake meal (kg/h). Ts, TF and Tc are the oil content of the seed (%), the filtrate (%) and the cake meal (%), respectively.
In some cases, a dead stop procedure allowed us to collect material at different locations (mainly on modules 1–4) in screw channel for particle size distribution analysis. A number of materials (±50?g) taken from screw channel were successively filtered on 3, 2, 1 and 0.5?mm opening sieves to separate all particles. All fractions were weighed for measurement of the particle size distribution.
2.3.2. Operating conditions study
Experiments were done in three steps and conducted with screw profile 5. First, experiment was conducted at various temperatures (80–120?°C) with a fixed screw rotation speed of 60?rpm and a seed input flow rate of 24?kg/h. The choice of these temperature limits were based on information reported in the literature (Dufaure et al., 1999a). Furthermore, a number of feed input flow rates (17–49?kg/h) and screw rotation speeds (60–200?rpm) were tested to determine the optimal operating condition. In this case, the temperature along the barrel was fixed at 80?°C. Sample collections and analysis were determined according to procedure in previous study.
2.4. Oil quality analysis
The quality parameters of a crude oil included (i) the acid value, expressed in mg of KOH/g of oil (standard NF T 60-204), indicates the free fatty acid content of the oil; (ii) the iodine value, expressed in terms of the number of centigrams of iodine absorbed per gram of oil (standard AOCS-Cd 1d-92), is a measure of the unsaturation of oils. The higher the iodine value the greater the unsaturation of oils; (iii) the saponification value, expressed in mg of KOH/g of oil (standard ISO 3657), is the amount of alkali necessary to saponify a definite quantity of the oil; (iv) the phosphorus content, expressed in mg of phosphorus per kg of oil (standard AOCS Ca 12-55), determines phosphorus or the equivalent phosphatide content by ashing the oil in the presence of zinc oxide followed by the spectrophotometric measurement of phosphorus as a blue phosphomolybdic acid complex. Total phospholipids content was determined by multiplying phosphorus content by 30. Moreover, the fatty acids and tocopherols compositions of crude oil were determined with gas chromatography (FAME method) and HPLC (IUPAC 2, 432 COFRAC CM 40), respectively.
2.5. Specific mechanical energy
The specific mechanical energy (SME) was calculated by the following equation:
(3)
(4)
where P is the motor power, I and Ss are correspondingly electrical intensity and screw rotation speed.
2.6. Residence time distribution
The residence time distribution (RTD) was determined by introducing directly a certain amount of seeds (±5?g) colored with erythrosine into the entrance of the extruder. Samples (filtrate and cake meal) were collected every 10?s. The cake meal samples were dried (105?°C, 24?h) and ground in a micro-grinder. Furthermore, the quantity of colorant in samples was determined by CIE L*a*b* method using a spectrocolorimeter (Minolta Seri CM-500i, Japan). The color values measured are presented as L*, a* and corresponding to lightness, the green-red and the blue-yellow components, respectively. Those results are the average of five consecutive measurements.
3. Results and discussion
3.1. Effect of screw configuration
3.1.1. Oil extraction yield
The position of BB, DM and CF elements and the spacing between two elements affected generally the oil extraction yield R and Ro (Fig. 3). High oil extraction yield based on residual oil content of cake meal (R) was observed when first reversed screw element was moved farther from second reversed screw, as observed on profiles 3, 5, 8 and 11, compared to profiles 0, 7 and 9 where no spacing between reversed screw elements. For certain interval of two CF elements, oil extraction yield increased with decreasing interval between BB and DM elements, as observed on profiles 2 and 5. In another cases, the reduction of the interval of BB and DM elements decreased the oil yield (profiles 1 and 4, 3 and 6). The modifications of the configuration of BB and DM elements or/and the pitch of screw elements did not influence the oil yield (profiles 8 and 11), in contrary the oil extraction yield decreased (profiles 5 and 10).
Fig. 3.?Variation of oil extraction yield on different screw configurations.
View thumbnail images
In the case of Ro, high oil yield separated from filtrate by centrifugation was mainly observed when interval between BB and DM