Comparative study of methods of determining oil content of sunflower seed

An extraction-gravimetric method (AOCS Official Method Ai 3-75) was compared with 2 instrumental techniques, near-infrared reflec-tance (NIR) spectroscopy and wide-line nuclear magnetic resonance (NMR), for the determination of the oil content of oilseed-type hybrid sunflower seed. Eight sunflower seed samples of varying oil contents, replicated 5 times, were analyzed by the 3 procedures. The overall mean oil contents and standard deviations for the 8 samples were: AOCS method, 44.5% ± 0.33%; NMR, 44.8% ± 0.27%; and NIR, 44.2% ± 0.81%. Analysis of variance of the means of the 3 methods of analysis indicated no difference (p>0.05) in oil content due to the method. However, there was a difference (p>0.001) in total oil content due to replicated analyses of the same sample with the NIR method. With the AOCS and NMR methods, no effect (p>0.05) of replicated analyses of the same sample was found. The NMR method was more precise and repro-ducible than the other 2 methods. Although the NIR mean oil contents were not significantly different from the means of the other 2 methods, the coefficient of variations for all samples were consistently higher for the NIR analyses than for the AOCS and NMR analyses.     

Analysis of oil content of sunflower seed by wide-line NMR

The wide-line nuclear magentic resonance (NMR) analyzer is routinely used to determine the oil content of sunflower seed by plant breeders. This technique is now under consideration as the official method for the domestic trading of sunflower seed. A study of the effect of depth (volume) of sunflower seed in the NMR 130 ml sample tube showed that between a depth of 30–75 mm (23.5–62.5 g seed) the NMR response was uniform, but beyond 75 mm, the response rapidly decreased. Oil analysis of 10 sunflower seed samples showed that coefficient of variation (C.V.) was lower with a 130 ml sample tube (C.V. 0.4%) than with a 34 ml tube (C.V. 0.8%). As the temperature of the sample was increased 1C, the instrument response decreased by 0.4%. Analysis of sunflower seed with 31–71% linoleic acid contents analyzed 0.1% higher for each 1% decrease in linoleic acid. Data show that linoleic acid content of NMR sunflower seed standard is important in NMR total oil analysis. Results of this study showed that the sample of sunflower seed for total oil analysis by NMR should be contained at least within the bottom 70 mm of the 130 ml sample tube, and NMR response of the standard calibration seed and sample being analyzed should be read at the same temperature, and their fatty acid compositions should be similar.     

Measurement of oil in whole flaxseed by near-infrared reflectance spectroscopy

A near-infrared reflectance (NIR) Infralyzer 500 was calibrated for determination of oil with samples of ground and whole flaxseed grown over three years. Wavelength selection by the computer software interfaced with the Infralyzer, analytical and regression statistic data, such as standard deviation of laboratory analysis (SD_x), correlation coefficient, standard error of estimate (SEE), standard error of prediction (SEP), and the SD_x/SEP ratio showed that calibration of the instrument with whole flaxseed was equal in precision to that obtained with the ground flaxseed. Growth location or seed moisture content had no effect on oil content of whole flaxseed determined by the NIR. The whole seed calibration allowed rapid, nondestructive screening for oil in flaxseed at greatly reduced cost.     

Rapid and nondestructive determination of seed oil by pulsed nuclear magnetic resonance technique

The pulsed NMR technique for rapid and nondestructive determination of oil in oilseeds has been developed. The effects of spin-lattice relaxation time, spin-spin relaxation time, seed moisture, angular position of the seeds, sample tube thickness, and sample height upon the magnitude and reproducibility of the NMR signal were studied. Based upon these studies, various parameters for seed oil analysis have been fixed. The oil content of Brassica, peanut, and sunflower seeds was determined. The reproducibility of the measurement is ± 1 %. The technique was tested by measuring the oil content of the same seeds by the cold percolation method (CCl₄ extraction). It was further tested by determining the oil content of 60 Brassica seed samples independently at three laboratories. The results of these tests are given.     

Effect of sprouting on the quality and composition of canola seed and oil

Sprouting has been considered as a damage factor in grading canola. This project deals with the evaluation of the effect of sprouting on the quality and composition of canola seed and oil. Sprouted seeds had lower oil content than nonsprouted seeds as determined by exhaustive petroleum ether extraction. The difference, although statistically significant, was small, less than 0.1% oil at the maximum level of sprouting allowed in topgrade canola. There were no differences in chlorophyll contents or moisture contents between sound and sprouted seeds. Sprouted seeds had significantly higher levels of FFA and crude protein than sound seeds. Oxidation parameters (diene and aldehyde) were higher in oils from sound seeds than oils from sprouted seeds, but there was no statistically significant difference in PV. Sprouted seeds had higher levels of tocopherols and sucrose, but lower levels of raffinose, stachyose, and total sugars than sound seeds. There was no difference in overall FA composition of the oil between sound and sprouted seeds. The second extraction of the Federation of Oils Seeds and Fat Associations (FOSFA) extraction method, which allowed the extraction of more polar lipids, contained significantly more saturated FA. However, this was not significant in the overall FA composition of the oils because this fraction counted for about 2% of the total lipid content. The presence of sprouted seed had an effect on results for oil and crude protein determined by NIR as compared with results by FOSFA extraction, or pulsed NMR for oil and Dumas combustion for crude protein. Addition of sprouted seed samples to the NIR, calibration set overcame this problem. These results suggested that sprouting did not have a highly damaging effect on the quality and composition of canola seed and oil when less than 10% of the seeds in a sample were sprouting.     

Estimating the content of free fatty acids in high‐oleic sunflower seeds by near‐infrared spectroscopy

The content of free fatty acids (FFA) in vegetable oils represents an important quality factor in oil crops. The objective of the investigation was to develop a near‐infrared (NIR) calibration for estimating the FFA content in high‐oleic sunflower seeds. A sample set of different varieties from the harvest of 2004 as well as of 2005 from two locations in Germany was used; additionally seeds from 2003 were stored under unsuitable conditions to obtain samples utilised for calibration with an extended FFA range. A direct titration method for FFA determination was developed and adjusted to the official AOCS method. The modified method is sufficiently reliable, much faster than the AOCS method and therefore suitable for use in the calibration of NIR spectrometers. The developed NIR spectroscopy (NIRS) calibration was calculated with a modified partial least square algorithm, standard normal variate and detrend scatter correction and the 2^nd derivative of the spectra of ground sunflower seeds. The standard error of prediction of the validated calibration was 0.20, and the multiple coefficient of determination (RSQ_val) reached 0.94. The obtained results demonstrated clearly the efficiency and how cost effective the NIRS method is for the estimation of FFA content in sunflower seeds.     

An analysis scheme for estimation of crude oil quality

A seed analysis scheme was designed to rapidly estimate the quality of extracted oil. Factors of crude oil quality evaluated were: free fatty acids, oxidative status (Totox value), color, and phosphatides (soybean) or wax (sunflower). Soybean and sunflower seeds subjected to extended storage at varying moisture contents were sampled at incremental time periods to yield fifty storage-damaged samples of each oilseed. Oil was extracted from 50-g lots of each sample and analyzed for the crude oil quality factors according to standard methods. Alternative instrumental and chemical analyses of the quality factors were correlated with the standard methods. Hexanal content, measured by headspace-gas Chromatographie analysis of the ground full-fat meal, was correlated to the oxidative status. Crude oils recovered by rapid extraction, using sonication, and desolventation were monitored by spectrophotometry for color correlation. Free fatty acid content was determined by titration methods and monitored by spectrophotometry. Modified turbidimetric methods estimated the phosphatides (soybean) or wax (sunflower seed) contents. The analysis scheme provides for the rapid estimation of oil quality as impacted by various pre- and post-harvest events that cause deterioration of oilseeds. The mention of firm names or trade products does not imply that they arc endorsed or recommended over other firms or similar products not mentioned.     

Automatic karl fischer titration of moisture in sunflower seed

An automatic Karl Fischer (KF) titrator of the motor-driven buret type was applied to the determination of moisture in sunflower seed. A study of the effect of sample size on KF moisture analysis showed a significant decrease in moisture content with increase in sample size from 1 to 5 g. In the moisture range of 5.5–10.5%, a sample size of 3–4 g gave moisture values closest to those obtained by the AOCS official oven method for sunflower seed. Comparison of KF moisture analysis with oven methods on 6 samples with moisture contents ranging from 5.4 to 12.7% showed that KF moisture values were not significantly different from air oven and vacuum oven methods. The mean standard deviation of KF determinations of whole sunflower seed was 0.11% moisture content, whereas for forced draft oven moistures, the mean standard deviation was 0.05%. KF moisture values generally were slightly lower than the air oven moisture values. Using KF and vacuum oven methods to measure moisture contents, sunflower seed were found to lose 0.6% moisture from a sample containing 8.9% moisture when grinding the sample with Hyflo Super Cel as in the AOCS official method.     

Determination of tocopherols and phytosterols in sunflower seeds by NIR spectrometry

The objective of this work was to develop a near‐infrared reflectance spectrometry (NIRS) calibration estimating the tocopherol and phytosterol contents in sunflower seeds. Approximately 1000 samples of grinded sunflower kernels were scanned by NIRS at 2‐nm intervals from 400 to 2500 nm. For each sample, standard measurements of tocopherol and phytosterol contents were performed. The total tocopherol content was obtained by high‐performance liquid chromatography coupled with a fluorescence detector, while the total phytosterol content was assessed by gas chromatography. For tocopherol, the calibration data set ranged from 175 to 1005 mg/kg oil (mean value around 510 ± 140 mg/kg oil), whereas for the phytosterol content, the calibration data set ranged from 180 to 470 mg/100 g oil (mean value of 320 ± 50 mg/100 g oil). The NIRS calibration showed a relatively good correlation (R² = 0.64) between predicted by NIRS and real values for the total tocopherol content but a poor correlation for the total phytosterol content (R² = 0.27). These results indicate that NIRS could be useful to classify samples with high and low tocopherol content. In contrast, the estimation of phytosterol contents by NIRS needs further investigation. Moreover, in this study, calibration was obtained by a modified partial least‐squares method; the use of other mathematical treatments can be suitable, particularly for total phytosterol content estimation.     

Nondestructive single-kernel oil determination of maize by near-infrared transmission spectroscopy

The utility of near-infrared transmission spectroscopy (NITS) for the nondestructive prediction of oil content in single maize kernels was explored. Calibration models were developed from spectral information gathered between 850 and 1050 nm. Nuclear magnetic resonance (NMR) spectroscopy was employed as a reference method to determine the actual oil content of samples used for calibration development and testing. Various positionings of the kernels in the light path and calibration math treatments were explored. The best NITS calibration yielded a 1.2% standard error of cross validation, which was over four times the standard error of NMR reproducibility. Although not as accurate as NMR, NITS does have utility in selecting kernels with the highest oil content from a segregating population.     

Jaocs news feature: Collaborative study of the determination of total oil in sunflower seed

A collaborative study was made of two methods for determining the total oil content of sunflower seed. The seed of three sunflower varieties of low, medium, and high oil content were analyzed on different days by 12 collaborators. In one method, the sunflower seeds were ground with an equivalent wt of diatomaceous earth. In the second method, the seeds were ground with a high-speed grinder without diatomaceous earth. The method using diatomaceous earth in the grinding of the seed gave a significantly higher oil content for all three varieties. It was recommended that total oil on a per cent dry wt basis be calculated from the moisture determination of the whole seed and not from the moisture of ground sample, which was too variable.     

Near-infrared spectroscopic analysis of deterioration indices of soybeans for process control in oil milling plant

The feasibility of near infrared (NIR) spectroscopy was examined for analyzing quality criteria or deterioration induces of soybeans, such as the nitrogen solubility index (NSI), acid value, digestibility and percentage germination, as well as the contents of their major constituents (moisture, oil and total nitrogen). Because all the criteria are closely related to yield and quality of the final product, their rapid analysis is required for process control in the oil milling plant. NIR diffuse-reflectace spectra were measured on ground samples in an InfraAlyzer 500 (Bran+Luebbe Co., Norderstedt, Germany). As for the major constituents, the standard errors of prediction (SEP) were 0.34% (moisture), 0.24% (oil) and 0.067% (total nitrogen). There results show the ability of NIR for analyzing these parameters, as many researchers reported previously. As for the other quality criteria, the SEP values were 2.40 (NSI), 5.58 (digestibility), 0.29 (acid value) and 15.27 (germinatio percentage). Further, NIR analysis for some of these criteria of the extracted residue and defatted soybeans was also examined, and almost the same level of SEP values was obtained. Although it is still necessary to improve the accuracy, we concluded that the NIR method has the feasibility to measure quality criteria for the purpose of process control in the plant in place of the time-consuming chemical analysis that are conventionally used. Part of this study was presented at 84th AOCS Annual Meeting & Expo, April 27, 1993, Anaheim, California.     

Determination of In-Shell Peanut Oil and Fatty Acid Composition Using Near-Infrared Reflectance Spectroscopy

NIR reflectance spectroscopy was used to analyze the total oil and fatty acid concentration of Virginia and Valencia types of in-shell peanuts rapidly and nondestructively. NIR absorbance spectra were collected in the wavelength range from 400 to 2,500 nm using a NIR instrument. Average total oil concentrations of all samples were determined by a standard Soxtec extraction method. Fatty acids were converted to the corresponding methyl esters and measured using gas chromatography. Partial least squares analysis was performed on the calibration set, and models were developed for predicting total oil and fatty acids. The best model was selected based on coefficient of determination (R ²), standard error of prediction, and residual percent deviation (RPD) values. Virginia-type in-shell peanuts had RPD values of >5.0 for both absorbance and reflectance models, indicating that the method could be used for quality control and analysis. Valencia peanuts had an RPD value of 3.01, which indicates that the model is good for initial screening purposes. For both types of peanuts, fatty acid prediction gave RPD values of <5 for all the models, indicating they could be used for initial screening purposes.     

The effect of hydrogen content on estimation of seed oil by pulsed nuclear magnetic resonance

In the nondestructive estimation of seed oil by pulsed nuclear magnetic resonance (NMR), an assumption generally is made that the hydrogen content of the oil in the seed sample under investigation is the same as that of the oil standard or that of the oil in the standard seed samples. The hydrogen content is defined as the number of hydrogen atoms per unit mass of oil. The validity of this assumption has been investigated by; (i) calculating the Hydrogen contents of various oils on the basis, of their reported fatty acid composition, and (ii) experimentally determining the hydrogen contents of cotton and mustard oils obtained from different varieties of seeds. The FID method was used to monitor the NMR signal intensities. Both calculated and experimental values show that that hydrogen contents can be different not only for different oils but also for the same oil extracted from different varieties of the seeds. This variation in hydrogen contents is shown to introduce an error in the oil content values as estimated by NMR methods. The magnitude of this error increases with oil content, and also with the difference in the hydrogen contents.     

Controlling Argan Seed Quality by NIR

The suitability of using visible/near infrared spectroscopy (Vis/NIR), as a rapid and non‐destructive technique for monitoring the quality of argan seeds (Argania spinosa Skeels) was studied. The analyzed parameters were the fatty acid composition of argan seed oil, seed moisture content, seed oil content and oil stability index (OSI). The ratio between major unsaturated and saturated fatty acids (U/S) during the oxidation assay at constant temperature was studied. Values from infrared drying were used as a laboratory reference for the moisture. Argan seed oil content was determined by Soxhlet extraction. A fatty acid analysis was carried out by gas chromatography and the OSI was determined by the Rancimat test. Predictive models of argan seed moisture, ratio U/S and OSI showed good accuracy. Therefore, Vis/NIR measurements can be used for controlling several argan seed quality parameters. This procedure might be of interest to the argan oil industry, which is currently in the process of modernization and expansion.     

Identification of adulterants in olive oils

The application of discriminant analysis for identifying and quantifying adulterants in extra virgin olive oils is presented. Three adulterants were used (sunflower oil, rapessed oil, and soybean oil) and were present in the range 5–95%. Near-infrared spectroscopy and principal components analysis were used to develop a discriminant analysis equation that could identify correctly the type of seed oil present in extra virgin olive oil in 90% of cases. Partial least squares analysis was used to develop a calibration equation that could predict the level of adulteration. Cross validation suggested that it was possible to measure the level of adulteration to an accuracy of ±0.9%. External validation of the derived calibation equation gave a standard error of performance of ±2.77%.     

Possibilities of Application of NIR Technique in the Analysis of oilseeds and their Derivatives - Examination of Used Frying Oils

During deep fat frying process, several compounds arise in the oil. The determination of their quantity by the classical methods is a very long and laborious work. For this purpose, quick instrumental examinations are needed. Therefore, the authors have investigated the possibilities of applicability of the NIR technique for determining the polar, polymer and linoleic acid contents of the sunflower oil heated at 170–180°C and 200–210°C, resp., for different times. They examined the equation forms describing the relationship between the parameters and values measured at given wavelengths of the NIR spectra, the characteristic wavelengths, the K values and the accuracy of the determination. They established that the NIR Technique can give successful results by the quick determination of the examined parameters. It was found that the standard error of calibration (SEC) of polar content at the thermal load of 170–180°C was 0.991 % and at that of 200–210°C 0.915%, that for polymer content 1.21% and 1.65%, resp., that for linoleic acid content 0.436% and 0.446%, resp., and the multiple correlation coefficient was ranging between 0.994 and 0.999.     

Solvent winterization of sunflower seed oil

Samples of oil from whole and dehulled sunflower seed were solvent winterized. The solvent mixture, 85% acetone, 15% hexane (^v/v), was used at solvent-in-oil concentrations of 20, 40, and 70% by wt and the samples winterized at 0, −5, −10, and −15 ± .01 C for 4 hr. Generally, sunflower oils from whole seed remained free from cloud formation longer on refrigeration when the oils were winterized at lower temperatures and at lower solvent-in-oil concentrations. With oil from the dehulled samples, no winterization condition produced an oil with a predictable clouding time. However, correlations were significant between residual wax content after winterization and clouding time of the oils from whole seed. Oils from dehulled seed were not as highly correlated with wax content as oils from whole seed. This study indicates that crude sunflower seed oil might be winterized with the aid of solvents and that decortication prior to extraction might not be necessary for effective winterization.     

Relation of days after flowering to chemical composition and physiological maturity of sunflower seed

Hybrid sunflower seed (achene) were collected from plants at 7-day intervals after the initiation of flowering which occurred 58 days after planting. The seed were analyzed for moisture, total oil, free fatty acids, lipid classes, and fatty acid composition. Seed dry weight, oil and triglyceride contents were maximum 35 days after the initiation of flowering (DAF) when the seed moisture content was about 36%. This point was defined as “physiological maturity” for sunflowers. The fatty acid composition of the oil extracted from the seed was determined at each stage of maturity. Total saturated fatty acids were 27% at 7 DAF and then decreased to a constant 9% by 35 DAF. At 7 DAF, linolenic acid content was 10.7% then decreased to less than 0.1% by 28 DAF. Oleic acid was about 12% at 7 DAF, increased to 59.6% at 14 DAF, and then gradually decreased to 31.4% by 56 DAF. On the other hand, linoleic acid was about 48% at 7 DAF, decreased to 23% by 14 DAF, but then gradually increased to 59.2% by 56 DAF. An analysis of variance of linoleic and oleic acid contents from 21 DAF to 70 DAF showed a highly significant change in composition with maturation time. The changes in the composition of these fatty acids from 21 DAF to 70 DAF appeared to be related to the environmental temperature which gradually decreased until 56 DAF. Increase in free fatty acids after physiological maturity indicated that deterioration of seed oil was beginning to occur.     

Comparison of three whole seed near-infrared analyzers for measuring quality components of canola seed

Whole seed near-infrared (NIR) analyzers are capable of high-speed compositional analysis of oilseed commodities. This study compared the PerCon Inframatic 8144 (Perten Instruments, North America Inc., Reno, NV), the Tecator Infratec 1225 (Tecator AB, Hoganas, Sweden) and the NIR-Systems 6500 (NIR Systems, Inc., Silver Spring, MD) analyzers for measurement of oil, protein, chlorophyll and glucosinolates in intact canola seed of composite samples from the Grain Research Laboratory's (Winnipeg, Manitoba, Canada) annual Western Canada Harvest Surveys (1985–1989) for assembly of calibration and prediction sets. No significant differences were found between the three instruments for oil [standard error of prediction (SEP 0.43–0.55%)], protein (SEP 0.35–0.42%) and glucosinolates (SEP 2.4–3.8 mM/g). Neither the Tecator nor the PerCon instruments were effective for determining chlorophyll. By combining oil content and fatty acid composition data to give an estimate of the total level of each fatty acid in the sample, high correlations were obtained for total saturates, linolenic acid, and linoleic acid although the RPD (ratio of the S.E. of prediction to the S.D. of the original data) values were not high enough to enable routine use of the method to predict results. Presented at the 84th AOCS Annual Meeting & Expo, April 27, 1993, Anaheim, California.