Deacidification of soybean oil using supercritical fluid and membrane technology

Membrane processing has been used for oil purification, but low flux and membrane instability are major impediments. A technique that combines membrane processing and supercritical CO₂ was investigated. A specialized, high-pressure, dead-end membrane cell was designed, fabricated, and connected to two ISCO (Lincoln, NE) supercritical fluid extraction (SFE) systems. The cell has a base with a grooved bottom for permeate removal, plus a porous metal disc for membrane support; a cell body with threaded connections; and a cap with an inlet assembly. One SFE pump provided the appropriate pressure on the feed stream, the second maintained pressure on the permeate at a slightly lower pressure. The sample consisted of 50% TAG and 50% FFA. For example, for separations at 45°C and a transmembrane pressure of 7-atm, the β (selectivity factor) values (TAG, FFA) for the SE and BW membranes were 0.56, 3.63 and 0.60, 2.63, respectively, whereas the β values (TAG, FFA) for the DK and NF90 membranes were 0.58, 1.37 and 0.70, 1.28, respectively.     

Deacidifying rice bran oil by solvent extraction and membrane technology

Crude rice bran oil containing 16.5% free fatty acids (FFA) was deacidified by extracting with methanol. At the optimal ratio of 1.8:1 methanol/oil by weight, the concentration of FFA in the crude rice bran oil was reduced to 3.7%. A second extraction at 1:1 ratio reduced FFA in the oil to 0.33%. The FFA in the methanol extract was recovered by nanofiltration using commercial membranes. The DS-5 membrane from Osmonics/Desal and the BW-30 membrane from Dow/Film Tec gave average FFA rejection of 93–96% and an average flux of 41 L/m²·h (LMH) to concentrate the FFA from 4.69% to 20%. The permeate, containing 0.4–0.7% FFA, can be nanofiltered again to recover more FFA with flux of 67–75 LMH. Design estimates indicate a two-stage membrane system can recover 97.8% of the FFA and can result in a final retentate stream with 20% FFA or more and a permeate stream with negligible FFA (0.13%) that can be recycled for FFA extraction. The capital cost of the membrane plant would be about $48/kg oil processed/h and annual operating cost would be about $15/ton FFA recovered. The process has several advantages in that it does not require alkali for neutralization, no soapstock nor wastewater is produced, and effluent discharges are minimized.     

Optimization of Deacidification of Mixtures of Sunflower Oil and Oleic Acid in a Continuous Process

Mixtures of refined sunflower seed oil and oleic acid were deacidified using nitrogen as stripping gas in a pilot-plant scale continuous deodorizer. To optimize the process, two different levels were tested for the classical operating parameters of oil temperature, nitrogen flow rates, oil mass flow rates and initial free fatty acid contents. In addition, two heating procedures were tested, including one using separate electrical heating of the oil and gas distillates to maintain the same temperature in both parts of the deodorizer, and another in which only the oil was heated and controlled, resulting in a temperature difference between the oil and the gas distillates. The statistical technique of blocking with paired comparisons was used to analyze the final free fatty acid content, rate of free fatty acid loss with respect to the processed oil flow rate, free fatty acid content in the distillates recovered by condensation, and efficiency. The results showed that the oil temperature and maintaining the same temperature in the oil and the gas distillates produced the most relevant effects, having a positive effect on most of the responses.     

Deacidification of sulfur olive oil. II. Multi-stage liquid-liquid extraction of miscella with ethyl alcohol

In this study, laboratory-scale multi-stage cross- and counter-current extractions of sulfur olive oil miscella with 70 and 80% ethanol saturated with hexane were investigated. For cross-current extraction, the extraction factor for free fatty acids was constant in each extraction stage. Therefore, the extraction factors determined in single-stage extractions were used to calculate the extracted free fatty acid percentages for cross-current and counter-current multi-stage extractions and results were in close agreement with the experimental data. It was possible to determine the amount of solvent and the number of stages required for counter-current extraction to remove the desired amount of free fatty acids from a given sulfur olive oil with 70 or 80% ethanol. Comparison of the results for these two solvents showed that 80% ethanol was more suitable.     

Deacidification of sulfur olive oil. I. Single-stage liquid-liquid extraction of miscella with ethyl alcohol

In this study, a systematic and detailed investigation on liquid-liquid extraction of sulfur olive oil miscella in hexane with aqueous ethanol solutions was performed. Optimal extraction conditions for recovery of free fatty acids (FFA) with a minor loss of neutral oil were determined in bench-scale single-stage extractions. It was concluded that, to ensure deacidification with a low triglycerides loss, it is appropriate to extract the miscella with 30% or more dilute ethanol solutions. It was also noted that under these circumstances the free fatty acid percentage extracted is not affected by increases in contents of FFA and partial glycerides of sulfur olive oil, and the solvent must be saturated with hexane before extraction. Changing the oil:hexane ratio in miscella from 1:2 to 2:1 by weight did not have any significant effect on extraction results.     

Deacidification of high-acid rice bran oil by reesterification with monoglyceride

Autocatalytic esterification of free fatty acids (FFA) in rice bran oil (RBO) containing high FFA (9.5 to 35.0% w/w) was examined at a high temperature (210°C) and under low pressure (10 mm Hg). The study was conducted to determine the effectiveness of monoglyceride in esterifying the FFA of RBO. The study showed that monoglycerides can reduce the FFA level of degummed, dewaxed, and bleached RBO to an acceptable level (0.5±0.10 to 3.5±0.19% w/w) depending on the FFA content of the crude oil. This allows RBO to be alkali refined, bleached, and deodorized or simply deodorized after monoglyceride treatment to obtain a good quality oil. The color of the refined oil is dependent upon the color of the crude oil used.     

膜过滤技术在盐水精制中的应用

介绍了戈尔膜过滤器、凯膜过滤器、呜泰"种植膜"过滤技术、颇尔膜过滤器及陶瓷膜等膜过滤技术在盐水精制中的应用情况。     

膜分离技术在盐化工中的应用

综述了膜分离技术的原理、特点及分类及在盐水精制和膜法脱硝中的应用。     

陶瓷膜过滤技术在盐水精制中的应用

主要介绍了陶瓷膜过滤技术在盐水精制中的应用情况。根据试验过程中出现的问题,总结出陶瓷膜过滤工艺应注意的问题,并与其他过滤工艺的运行费用进行了比较。     

陶瓷膜技术的应用

介绍了陶瓷膜过滤工艺,以及其工艺流程短和因采用PLC自动控制减小了操作人员的劳动强度的特点,从投资和运行两方面与聚四氟乙烯有机膜进行了对比,并总结了运行中出现的问题,提出了今后陶瓷膜过滤工艺的改进方向.     

Enrichment of carotene from palm oil by organic solvent nanofiltration

Palm oil is one of the richest sources of natural plant carotene with typical concentration of about 0.5–0.7 g/L. Unfortunately, during physical refining of palm oil, most of the carotenes were destroyed by high temperatures and this represents a loss of potential source of natural carotene. Various techniques have been developed to extract and recover carotenes from palm oil, however these processes often require high energy usage, and usually renders the oil useless for further consumption. Recently, organic solvent nanofiltration (OSN) has become an important method for molecular separation particularly for the separation of low molecular weight bioactive compounds. This work presents the application of OSN membranes for the separation of carotene from a crude palm oil/solvent system. Several commercial OSN membranes (DuraMem and PuraMem series) fabricated from polyimide were evaluated for their separation abilities. PuraMem 280 showed the best selectivity performance, with the concentration of carotene in permeate oil increased from 0.60 to 0.79 g/L when hexane was used as the solvent. Runs by using DuraMem 150, DuraMem 300 and DuraMem 500 showed low or no selectivity between carotene and triglyceride in all solvents. It was found that the rejection of carotene depends strongly on the type of solvents. A coupled solution diffusion and film theory was also utilized to model carotene transport through OSN membrane. It was demonstrated that OSN can serve as an alternative for the direct carotene recovery from palm oil and can be potentially applied for other minor compounds recovery from vegetable oils.     

Degumming of crude soybean oil by ultrafiltration using polymeric membranes

In this study, the ability of two ultrafiltration polymeric membranes to perform the degumming of a crude soybean oil/hexane mixture is tested. The performance of both membranes is defined in terms of their permeation flux, permeate color, and rejection of phospholipids. One of the membranes was synthesized in our laboratories from polyvinylidenefluoride (PVDF); the other one is a commercially available membrane made of polyimide. The degumming experiments were done in a stirred dead-end ultrafiltration cell pressurized with N₂. Results show that tested membranes are suitable for removing phospholipids from the crude oil/hexane miscella in the range of temperature and transmembrane pressure utilized in this work. Both membranes have high selectivity regarding phospholipids and produce a moderate reduction in permeate color. The PVDF membrane gives permeate fluxes up to threefold larger than those obtained with polyimide membrane at the same operational conditions, making the former more suitable for use at industrial scale.     

Enzymatic epoxidation of soybean oil methyl esters in the presence of free fatty acids

Epoxides of soybean oil methyl esters (SMEs) are biodegradable, non‐toxic, and renewable epoxy plasticizers. The objective of the present work was to investigate the effects of free fatty acids on the enzymatic epoxidation of SMEs. The results showed that the epoxidation of SMEs depended on the type of the added free fatty acid. For saturated (≤C_18:0) and monounsaturated free fatty acids, the epoxy oxygen group content (EOC) of SMEs increased with increasing carbon chain length of free fatty acids; for branched‐chain unsaturated free fatty acids, the EOC of SMEs decreased in the presence of hydroxyl group (OH) and hydroperoxide (OOH) of free fatty acids; the EOC of SMEs decreased with increasing number of double bonds of free fatty acids. The maximum EOC and the initial epoxidization rate (V₀) linearly decreased with increasing peroxide value of SMEs. The highest EOC (6.87 ± 0.3%) of SMEs was obtained using behenic acid as reaction material, which was similar with that of stearic acid (EOC 6.75 ± 0.2%).     

Analysis of soybean oil from Ohio

Five soybean cultivars-Pella 86, Ripley, Sherman, Williams 82, and Zane—were analyzed to determine the total fatty acid composition and triglyceride fatty acid composition. Palmitic, stearic, oleic, linoleic, and linolenic were the major fatty acids in these cultivars. Zane was significantly higher in saturated fatty acid content and lower in linolenic acid content than the other cultivars. Resolution by argentation thin-layer chromatography decreased with increased triglyceride unsaturation.     

Fatty acid‐based comonomers as styrene replacements in soybean and castor oil‐based thermosetting polymers

In this study, a fatty acid‐based comonomer is employed as a styrene replacement for the production of triglyceride‐based thermosetting resins. Styrene is a hazardous pollutant and a volatile organic compound. Given their low volatility, fatty acid monomers, such as methacrylated lauric acid (MLA), are attractive alternatives in reducing or eliminating styrene usage. Different triglyceride‐derived cross‐linkers resins were produced for this purpose: acrylated epoxidized soybean oil (AESO), maleinated AESO (MAESO), maleinated soybean oil monoglyceride (SOMG/MA) and maleinated castor oil monoglyceride (COMG/MA). The mechanical properties of the bio‐based polymers and the viscosities of bio‐based resins were analyzed. The viscosities of the resins using MLA were higher than that of resins with styrene. Decreasing the content of MLA increased the glass transition temperature (T_g). In fact, the T_g of bio‐based resin/MLA polymers were on the order of 60°C, which was significantly lower than the bio‐based resin/styrene polymers. Ternary blends of SOMG/MA and COMG/MA with MLA and styrene improved the mechanical properties and reduced the resin viscosity to acceptable values. Lastly, butyrated kraft lignin was incorporated into the bio‐based resins, ultimately leading to improved mechanical properties of this thermoset but with unacceptable increases in viscosity. © 2010 Wiley Periodicals, Inc. J Appl Polym Sci, 2011     

Densities of vegetable oils and fatty acids

Complete data for density as a function of temperature have been measured for a number of vegetable oils (crambe, rapeseed, corn, soybean, milkweed, coconut, lesquerella), as well as eight fatty acids in the range C₉ to C₂₂ at temperatures from above their melting points to 110°C (230°F). The specific gravity and density measurements were performed according to American Society for Testing and Materials (ASTM) standard test methods D 368, D 891 and D 1298 for hydrometers and a modified ASTM D 369 and D 891 for pycnometers. Correlation constants, based on the experimental data, are presented for calculating the density of fatty acids and vegetable oils in the range of temperature from 24°C (75°F) or the melting point of the substance, to 110°C (230°F). The constants are valuable for designing or evaluating such chemical process equipment as heat exchangers, reactors, process piping and storage tanks. Estimated density of fatty acids by a modified Rackett equation is also presented.     

Comparison of steam and nitrogen in the physical deacidification of soybean oil

Deacidification in physical refining is one of the most sensitive steps in refining edible vegetable oils because of its large impact on the quality of the oil. The removal of volatile compounds such as FFA is accomplished at elevated temperatures and a high vacuum with a stripping gas, usually steam. The aim of this work was to verify, at the laboratory level, the advantages of using an alternative stripping gas, nitrogen, instead of steam. An ideal vapor-liquid equilibrium model (IVLE) was used to compare the stripping capacities of steam and nitrogen and to analyze the effects of various operational parameters (temperature, pressure, amount of stripping gas) on the residual acidity of the oil. There was no clear evidence that nitrogen showed a higher capacity to strip FFA than steam. The IVLE model seemed suitable to describe FFA laboratory distillation by using steam or nitrogen, provided the final residual content of FFA was not too low.     

Extraction of free fatty acids from soybean oil using ionic liquids or poly(ethyleneglycol)s

The alternative solvents poly(ethyleneglycol)s (PEGs) with various molar masses and room temperature ionic liquids (RTILs) AMMOENG100 and 1‐butyl‐3‐methylimidazolium dicyanamide (bmimDCA) were studied for the purpose of extraction of linoleic acid from soybean oil. Liquid‐liquid phase equilibrium was measured for binary (PEG + soybean oil), (RTIL + soybean oil) as well as ternary (PEG + soybean oil + linoleic acid), (RTIL + soybean oil + linoleic acid) mixtures, as a function of temperature and composition. The influence on distribution coefficients and separation factor of temperature, initial acid content of the oil and solvent to oil ratio were studied. The effect of molar mass in the case of the PEGs was also examined. The experimental results were modeled using the Peng‐Robinson cubic equation of state coupled with the Mathias‐Klotz‐Prausnitz mixing rule, with good results. © 2010 American Institute of Chemical Engineers AIChE J, 2011     

Deacidification of black cumin seed oil by selective supercritical carbon dioxide extraction

The deacidification of high-acidity oils from Black cumin seeds (Nigella sativa) was investigated with supercritical carbon dioxide at two temperatures (40 and 60°C), pressures (15 and 20 MPa) and polarities (pure CO₂ and CO₂/10% MeOH). For pure CO₂ at a relatively low pressure (15 MPa) and relatively high temperature (60°C), the deacidification of a highacidity (37.7 wt% free fatty acid) oil to a low-acidity (7.8 wt% free fatty acid) oil was achieved. The free fatty acids were quantitatively (90 wt%) extracted from the oil and left the majority (77 wt%) of the valuable neutral oils in the seed to be recovered at a later stage by using a higher extraction pressure. By reducing the extraction temperature to 40°C, increasing the extraction pressure to 20 MPa, or increasing the polarity of the supercritical fluid via the addition of a methanol modifier, the selectivity of the extraction was significantly reduced; the amount of neutral oil that co-extracted with the free fatty acids was increased from 23 to 94 wt%.     

Changes of the tocopherol and fatty acid contents in rapeseed oil during refining

Among the most important metabolic compounds there are some which are not synthesized by human and animal organisms and have to be supplied in appropriate quantities in due time. Vitamin E and the essential unsaturated fatty acids have crucial physiological significance, and their greatest quantities occur in plant oils. During refining, apart from unnecessary substances, nutritionally advantageous compounds are also being eliminated. In the present paper changes of tocochromanols taking place during refining of rapeseed oil obtained from seeds of two subsequent crops were investigated. It was observed that losses of tocopherols exceeded 30%, two thirds of which resulted from distilling off during deodorization. The ratio of vitamin E to essential unsaturated fatty acids expressed as the Harris coefficient decreased in the refined oil obtained from seeds of two subsequent crops by about 28%.