Production of Starch with Very Low Protein Content from Soft and Hard Wheat Flours by Jet Milling and Air Classification

Jet milling is a fluid energy impact‐milling technique generally used for the ultrafine reduction of higher value materials. The efficiency of jet milling combined with air classification appears very efficient to separate starch from other wheat flour aggregate components and to produce wheat starch with very low residual protein content. Indeed, residual protein content of the starch‐rich fraction can be reduced to <2% db with a series of successive grinding and air classification operations. Lipid and pentosan contents were also reduced in the starch‐rich fraction. Nevertheless, jet milling cannot eliminate grinding differences observed between different types of wheat. Wheat hardness continues to have an effect on milling and classification yields and on the composition of air classification fractions. To obtain starch‐rich fraction with only 2% protein content, hard wheat flour required a series of at least five grinding steps, whereas only three steps are necessary for soft wheat flour. Under these conditions, hard wheat flours give 24% mass yield with 12% starch damage compared with 39% yield and a low starch damage content (6.4%) for soft wheat flour. These results highlight new prospects for the development of cereal flours, especially soft wheat flours.     

Ethanol Fermentation of Starch from Field Peas

Field peas (Pisum sativum) were evaluated as a potential feedstock for ethanol production. Ground peas were dry‐milled and separated into starch, protein, and fibrous fractions by air classification. Starch‐enriched fractions prepared from whole peas and dehulled peas contained 73.7% wt and 77.8% wt starch, respectively, a nearly two‐fold enrichment compared with whole peas. The fractions were liquefied and saccharified using industrial α‐amylase and glucoamylase at recommended enzyme loadings. A final ethanol concentration of 11.0% (w/v) was obtained in 48–52 hr, with yields of 0.43–0.48 g of ethanol/g of glucose. Starch present in whole ground peas was also saccharified and fermented, with 97% of the starch fermented when an autoclaving step was included in the liquefaction stage.     

Air Classification of Pin‐Milled Soybean Hulls

Commercial soybean hulls (14.6% crude protein) were pin milled and then air classified into five fractions to determine whether the hull structure can be disrupted and the protein constituents concentrated. The number of pin millings had only a small effect on the weight distribution of the five fractions. After one grinding, the sum of fractions 1 and 2 (<15–18 μm) represented only 3% of total hulls, and on three grindings, they amounted to only 6% of the hulls. Fraction 3 (19–24 μm) shifted from 16 to 20% with three grindings, while fraction 4 (25–30 μm) remained unchanged at 5%. Fraction 5 (>30 μm) shifted from 75 to 69% on three pin millings to compensate for the shifts noted in fractions 1–3. Scanning electron microscopy revealed that fractions 1 and 2 consisted of the parenchymal cell layers (innermost portion) of the hulls; these fractions had three times the protein content and more than twice the amount of lipids found in the ground starting material. Fraction 3 contained many of the hourglass cells typically found in the middle layer of the hulls plus parenchyma cell material and exhibited about twice the amount of protein and lipid found in the starting hulls. Fraction 4 included large hourglass cells plus globular material and contained about twice the amount of protein and about one and one‐half times the lipid of the starting material. Fraction 5 consisted primarily of clumps of palisade cells (outer cellular layer) adhering to each other and had a lower protein and lipid content than the starting hulls. Thus, pin milling causes some selective disruption of the hulls where parenchyma cell layers and hourglass cells are partially released. The palisade cells, however, are the most difficult to disrupt and constitute the fraction most resistant to pin milling. Our findings suggest that pin milling soybean hulls in combination with air classification can be used to concentrate the proteins and lipids in the fines fractions.     

Studies on the nutritional composition and antinutritional factors of three different germplasm seed materials of an under-utilized tropical legume, Mucuna pruriens var. utilis

Two different germplasms of a white variety and one germplasm of a black variety of Mucuna pruriens var. utilis were evaluated for their physicochemical properties as well as their nutritional and antinutritional characteristics. All germplasms had higher grain weight, density, hydration, and swelling capacity than other common legumes. The dehulled samples contained 303.2-335.5 g(-1) protein and 46.1-53.5 g x kg(-1) lipid, and these values were higher than the respective whole seeds. The levels of macro- and microelements in both whole and dehulled seeds were comparable to those in common pulses. All germplasms had a high dietary fiber content (18-19.5%), made up of mainly insoluble dietary fiber (DF). Seed lipids were high in unsaturated fatty acids (64.7-66.9%), specifically linoleic acid (48-49%). Whole and dehulled seeds of the white variety from Salem were particularly rich in sulfur-containing amino acids with significantly higher levels of in vitro protein digestibility than the other two germplasms. All germplasms had high levels of total phenols and phytate, trypsin, and chymotrypsin inhibitor activities, but were low in tannins, saponins, and alpha-amylase inhibitor activity. Only weak hemagglutinating activity against cow erythrocytes and no hemagglutinating activity against human erythrocytes (O) was observed in all the samples. Dehulled seeds were higher in total starch, including resistant starch and oligosaccharides (with verbascose as the major fraction) than the respective whole seeds. Both whole and dehulled samples of the white variety of Salem germplasm showed significantly lower concentrations of L-dopa, nonmethylated, and methylated tetrahydroisoquinolines than the respective whole and dehulled samples of other germplasms. In general, dehulling didn't affect the overall nutritional status in any of the presently investigated samples.     

Determination of free flavan-3-ol content in barley (Hordeum vulgare L.) air-classified flours: comparative study of HPLC-DAD/MS and spectrophotometric determinations

The determination of free flavan-3-ol compounds in barley flours (cv. Gotic) and two resulting milling fractions (fine fraction 57% and coarse fraction 43%, w/w) obtained by air classification of dehulled grain is herein described. The determinations were carried out using reversed phase high-performance liquid chromatography coupled with both diode array detection and ESI-MS compared with conventional spectrophotometric determinations (total phenolic compounds by the Folin-Ciocalteu method and free radical scavenging activity with the DPPH assay). Significant correlations among the HPLC quantification, the spectrophotometric data, and the antioxidant capacity of extracts were revealed by Pearson's analysis. Nine flavan-3-ols were identified by HPLC-MS. Catechins and their derivatives were found to make a substantial contribution to the antioxidant power of extracts. The coarse fraction showed larger concentrations of flavan-3-ols (221%) with respect to the fine fraction. This was confirmed by the antioxidant activity of the analyzed flours. The coarse fraction showed the greatest antioxidant activity (1200.1 +/- 66.2 micromol of Trolox equiv/100 g of flour) with respect to whole meal and fine fraction (1025.9 +/- 18.3 and 761.7 +/- 55.3 micromol of Trolox equiv/100 g of flour, respectively).     

Structure and Physicochemical Properties of Starches from Sieve Fractions of Oat Flour Compared with Whole and Pin-Milled Flour

One oat cultivar grown in Idaho (three field sites) was pin-milled and separated by sieving to investigate whether starch from oat bran differs from the remainder of kernel. Ground oat particles were classified into three sieve fractions: 300–850 μm, 150–300 μm and <150 μm). β-Glucan content in sieve fractions was analyzed and starch was extracted from kernels without milling and from kernels of each sieve fraction. β-Glucan contents of 300–850, 150–300, and <150 μm sieve fractions were 4.2, 2.3, and 0.8%, respectively. Therefore, starch in bran (300–850 μm sieve fraction) and endosperm (<150 μm sieve fraction) were separated. Starch isolated from entire kernels had significantly higher apparent and absolute amylose content than starch from the 300–850 μm sieve fraction. Starch from different sieve fractions was not significantly different in the apparent amylose, absolute amylose, amylopectin molecular weight, gyration radii, starch gelatinization, and amylose-lipid complex thermal transition temperatures. Starch from the 150–300 μm sieve fraction had significantly lower peak, final, and setback viscosity compared with the starch isolated from the 300–850 μm and <150 μm sieve fractions. Starch removed from the oat bran fraction during β-glucan enrichment may have different applications compared with starch obtained from other kernel compartments. Because pin-milling decreased apparent amylose content and shortened amylopectin branch chains, its potential to alter starch structure should be considered.     

Composition,Thermal Behavior,and Gel Texture of Prime and Tailings Starches from Garbanzo Beans and Peas

Prime and tailings starches of garbanzo beans and peas were separated and the chemical composition, physical properties, thermal behavior, and gel properties were determined. Starch granules <35 μm were 85% in garbanzo beans, 66.8% in a smooth pea cv. Latah, and only 18.4% in a smooth pea cv. SS Alaska. Amylose content of prime starch was 35.9% in garbanzo beans, 44.5–48.8% in smooth peas, and 86.0% in wrinkled pea cv. Scout. Tailings starch amylose content was at least 8% higher than the corresponding prime starch. The endothermic enthalpy value of garbanzo bean and two smooth pea prime starches ranged from 12.1 to 14.2 J/g, while prime starch from wrinkled peas gave a distinctly lower enthalpy value of 1.1 J/g. Differential scanning calorimetry endothermic enthalpy and amylograph pasting properties of prime starch were significantly related to its amylose content (P < 0.05). Prime starches of garbanzo beans and smooth peas produced highly cohesive elastic gels. Wrinkled pea prime starch formed the strongest (though brittle) gel, as indicated by high hardness (21.8 N), low cohesiveness (0.29), and low springiness (0.82). Hardness of gel stored at 22°C and at 4°C was positively correlated with amylose content of starch.     

Macromolecular interactions and rheological properties of buckwheat-based dough obtained from differently processed grains

The physicochemical properties of the protein and starch fractions of flour obtained from buckwheat grains that were previously dehulled or puffed after dehulling were investigated. Dehulling removed most of the nonprotein, nonstarch components of the grain, without affecting the chemical and structural features of the protein and starch components, as made evident by microstructural and spectroscopic measurements. Puffing resulted in extensive modifications of the interprotein network as well as in most of the properties of the buckwheat starch. Flours obtained from dehulled or puffed after dehulling grains were blended with 60-80% wheat flour and tested for their dough-making ability. Blends containing dehulled and puffed buckwheat flours gave dough of much lower quality than dehulled, but had water-holding properties that may be of interest for the shelf life of baked products.     

Quality of Flours from Waxy and Nonwaxy Barley for Production of Baked Products

One nonwaxy (covered) and two waxy (hull-less) barleys, whole grain and commercially abraded, were milled to break flour, reduction flour, and the bran fraction with a roller mill under optimized conditions. The flour yield range was 55.3–61.8% in whole grain and increased by 9–11% by abrasion before milling. Break flours contained the highest starch content (≤85.8%) independent of type of barley and abrasion level. Reduction flours contained less starch, but more protein, ash, free lipids, and total β-glucans than break flours. The bran fraction contained the highest content of ash, free lipids, protein, and total β-glucans but the lowest content of starch. Break flours milled from whole grain contained 82–91% particles <106 μm, and reduction flours contained ≈80% particles <106 μm. Abrasion significantly increased the amount of particles <38 μm in break and reduction flours in both types of barley. Viscosity of hot paste prepared with barley flour or bran at 8% concentration was strongly affected by barley type and abrasion level. In cv. Waxbar, the viscosity in bran fractions increased from 428 to 1,770 BU, and in break flours viscosity increased from 408 to 725 BU due to abrasion. Sugar snap cookies made from nonwaxy barley had larger diameter than cookies prepared from waxy barley. Cookies made from break flours were larger than those made from reduction flours, independent of type of barley. Quick bread baked from nonwaxy barley had a loaf volume similar to that of wheat bread, whereas waxy barley bread had a smaller loaf volume. Replacement of 20% of wheat flour by both waxy and nonwaxy barley flour or bran did not significantly affect the loaf volume but did decrease the hardness of quick bread crumb.     

Nutrient Content in Buckwheat Milling Fractions

Buckwheat seeds (Fagopyrum esculentum Moench) were milled into 23 fractions: seven fine flours, three coarse flours, four small semolina, two big semolina, six bran, and one husk fraction. A considerable variation in gross chemical composition was found among the milling fractions. The protein content varied from 4.4 to 11.9% (db) in flours and from 19.2 to 31.3% in bran fractions; starch varied from 91.7 to 70.4% in flours and from 42.6 to 20.3 in bran. The percentage of soluble dietary fiber contained in total dietary fiber was higher in flours than in semolina and bran fractions. Ash, Fe, P, tannin, phytate content, and color were also investigated. A unique distribution of phytate was found in starch. Correlation is significantly positive in husk, bran, and semolina fractions, while correlation is significantly negative in flour fractions. Depending on technological or nutritional demands, appropriate fractions may be chosen to achieve the desired end‐use product.     

Wet Fractionation of Garbanzo Bean and Pea Flours

Garbanzo bean and pea flours originating from the central part of the cotyledons were higher in starch but lower in protein and fiber than those from the outer layer of the cotyledon. These flours were fractionated by the wet process into prime starch, tailing starch, and solubles. The patented wet-fractionation method was successfully modified to reduce the total amount of water during the process. The modified process produced comparable yield and purity of separated fractions. Under the recycling water method, the yields of prime starch were 46.7, 33.6, and 41.1%, respectively, in garbanzo bean and two smooth pea cultivars (Latah and SS Alaska). Isolated prime starches were <0.4% in protein and <0.19% in ash, indicating high purity.     

Optimization of a Laboratory Dehulling Process for Lentil (Lens culinaris)

Red lentil (Lens culinaris) is mainly processed into dehulled and split forms before human consumption and characteristics such as dehulling efficiency (DE), which is the sum of percent dehulled whole seed (PDW) and percent dehulled split seed (PDS), are important to lentil breeders, processors, and exporters. A laboratory Satake dehuller was used to evaluate the dehulling characteristics of red lentil. The effects of dehulling conditions (abrasive wheel speed, dehulling time, and seed moisture content) were investigated using response surface methodology. Increasing dehulling time and seed moisture content decreased DE. Increasing seed moisture content decreased powder and broken fractions but increased the undehulled whole seed fraction. PDW was decreased but PDS was increased as dehulling time was increased. Percent hull removed during dehulling process decreased as seed moisture content was increased but increased as abrasive wheel speed or dehulling time was increased. The optimum dehulling conditions for the laboratory dehuller, based on maximizing DE and percent hull removed while minimizing powder fraction (loss), were established. Good agreement was found between experimental values for dehulling characteristics (DE, PDW, PDS, hull removed, and powder produced) obtained at optimum dehulling conditions and predicted values for those characteristics obtained using the models developed.     

Wet Processing of Barley Grains into Concentrates of Proteins, β‐Glucan,and Starch

An improved wet method was developed to process barley into fractions concentrated in protein, (1‐3)(1‐4)‐β‐d ‐glucan (BG), starch, or other carbohydrates (CHO). Alkaline concentration, solvent to barley flour ratio (SFR), and extraction temperature were evaluated for their effects on concentration and recovery of protein, BG, starch, oil, ash, and other CHO in each fraction type. Results show that the three parameters and their interactions all had significant effects, resulting in varying nutrient concentrations and recovery rates in each type of fractions. For protein fractions, protein content varied from 37.7 to 75.2%, protein recovery from 8.5 to 75.7%, and increasing alkaline concentration and SFR improved nutrient recovery. For BG fractions, BG content ranged from 21.5 to 87.0%, BG recovery from 28.6 to 78.0%, and increasing alkaline concentration decreased BG content but increased its recovery significantly. For starch fractions, starch content varied from 76.9 to 93.9%, starch recovery from 33.6 to 63.9%, and all parameters had little effect on the nutrient concentrations, but alkaline concentration and SFR improved recovery of starch, other CHO, and mass. Overall, the improved wet method was effective in concentrating the major nutrients from barley into their respective fractions, but process optimization through manipulating the three parameters is necessary to achieve a maximum concentration or recovery rate of a nutrient of interest in a specific fraction.     

Physicochemical Changes in Nontraditional Pasta During Cooking

Physicochemical changes in the components of nontraditional spaghetti during cooking were reflected in the quality of the cooked product. Spaghetti formulations used were semolina (100%), whole wheat flour (100%), semolina/whole wheat flour (49:51), semolina/flaxseed flour (90:10), whole wheat flour/flaxseed flour (90:10), and semolina/whole wheat flour/flaxseed flour (39:51:10). Spaghetti quality was determined as cooking loss, cooked weight, and cooked firmness. Physicochemical analyses included total starch, starch damage, pasting properties, and protein quality and quantity of the flour mixes and spaghetti cooked for 0, 2, 4, 10, and 18 min. As cooking time progressed, total starch content decreased up to 5.7% units, starch damage increased up to 11.7% units, and both pasting parameters and protein solubility decreased significantly in all six formulations. Changes in the starch damage level, total starch content, and pasting properties of spaghetti correlated significantly (P < 0.05) with the cooking loss, cooked weight, and cooked firmness values recorded for the spaghetti. High levels of glutenin polymers and low levels of the albumin and globulin fractions were associated with low cooking losses and cooked weight and with high cooked firmness, indicating the involvement of these proteins in the cooked quality of nontraditional spaghetti.     

Soft Wheat Starch Pasting Behavior in Relation to A‐ and B‐type Granule Content and Composition

Flours of two soft wheat cultivars were fractionated into native, prime, tailing, A‐, and B‐type starch fractions. Starch fractions of each cultivar were characterized with respect to A/B‐type granule ratio, amylose content, phosphorus level (lysophospholipid), and pasting properties to investigate factors related to wheat starch pasting behavior. While both cultivars exhibited similar starch characteristics, a range of A‐type (5.7– 97.9%, db) and B‐type granule (2.1–94.3%, db) contents were observed across the five starch fractions. Though starch fractions displayed only subtle mean differences (<1%) in total amylose, they exhibited a range of mean phosphorus (446–540 μg/g), apparent amylose (18.7–23%), and lipid‐complexed amylose (2.8–7.5%) values, which were significantly correlated with their respective A‐ and B‐type granule contents. A‐type (compared with B‐type) granules exhibited lower levels of phosphorus, lipid‐complexed amylose, and apparent amylose, though variability for the latter was primarily attributed to starch lipid content. While starch phosphorus and lipid‐complexed amylose contents exhibited negative correlation with fraction pasting attributes, they did not adequately account for starch fraction pasting behavior, which was best explained by the A/B‐type granule ratio. Fraction A‐type granule content was positively correlated with starch pasting attributes, which might suggest that granule size itself could contribute to wheat starch pasting behavior.     

Recovery of Starch and Protein from Wet-Milled Corn Fiber

Physical and chemical methods were used to recover starch and protein from wet-milled corn fiber. A single milling of the fiber produced an 18% yield of mill starch. By separating the mill starch with a starch table, 68% of this material was recovered as starch with a protein contamination of 0.66%. Milling increased fine fiber from 4.5% in the starting material to 11.5% after a single grind. Successive additional milling passes modestly increased the mill starch and fine fiber yields with a corresponding decrease in the coarse fiber yield. Pretreatment with combinations of lactic and sulfurous acids had only a small effect on the distribution and composition of the recovered fractions.     

Starch characteristics and their influences on in vitro and pig prececal starch digestion

The main objective of this research was to study the characteristics of starch granules and their influences on in vitro and pig prececal starch digestion of corn, dehulled barley, wheat, and potato. Scanning electron microscopy was used to study the starch endosperm structure in the parent material as well as in vitro starch digestion. The results showed that corn starch granules were polyhedral, with a diameter ranging from 2 to 10 μm, whereas those of dehulled barley and wheat were spherical, with a diameter ranging from 5 to 20 μm. Potato had the largest starch granules among starch sources reported herein, with oval spheres of 10-50 μm in diameter. In vitro starch hydrolysis showed that starch granules of corn degraded faster than the starch of dehulled barley and wheat, with the potato starch being degraded the slowest. The in vivo digestibility trial using ileal-cannulated pigs confirmed the starch degradation of grains. The in vitro (x, %) and in vivo (y, %) digestibility were highly correlated [y = 6.5304x - 538.48 (R(2) = 0.9924)]. On the basis of the results, in vitro starch hydrolysis might be useful in predicting in vivo prececal starch digestibility. The digestion kinetic characteristics of different starch sources might be employed to evaluate the starch digestive rate at the pig ileum.     

Extrusion of Cross‐Linked Hydroxypropylated Corn Starches II. Morphological and Molecular Characterization

A series of cross‐linked hydroxypropylated corn starches were extruded with a Leistritz micro‐18 co‐rotating extruder. Extrusion process variables including moisture (30, 35, and 40%), barrel temperature (60, 80, and 100°C), and screw design (low, medium, and high shear) were investigated. Scanning electron microscopy (SEM) of extruded starches showed a gel phase with distorted granules and granule fragments after extrusion at 60°C. After extrusion at 100°C only a gel phase was observed with no granular structures remaining. High performance size exclusion chromatography (HPSEC) equipped with multiangle laser light‐scattering (MALLS) and refractive index (RI) detectors showed extruded starches degraded to different extents, depending on extrusion conditions. The average molecular weight of the amylopectin of unextruded native corn starch was 7.7 × 10⁸. Extrusion at 30% moisture, 100°C, and high shear reduced the molecular weight of amylopectin to 1.0 × 10⁸. Hydroxypropylated normal corn starch extruded at identical conditions showed greater decreases in amylopectin molecular weight. With the addition of cross‐linking, the amylopectin fractions of the extruded starches were less degraded than those of their native and hydroxypropylated corn starch counterparts. Similarly, increasing moisture content during extrusion lowered amylopectin degradation in the extruded starches. Increasing temperature during extrusion of cross‐linked hydroxypropylated starches at high moisture content (e.g., 40%) lowered amylopectin molecular weights of the extruded starches, whereas increasing extrusion temperature at low moisture content (30%) resulted in less degraded molecules. This difference was attributed to the higher glass transition temperatures of the cross‐linked starches.     

Effects of Varying Weight Ratios of Large and Small Wheat Starch Granules on Experimental Straight‐Dough Bread

One commercial bread wheat flour with medium strength (11.3% protein content, 14% mb) was fractionated into starch, gluten, and water solubles by hand‐washing. The starch fraction was separated further into large and small granules by repeated sedimentation. Large (10–40 μm diameter) and small (1–15 μm diameter) starch fractions were examined. Flour fractions were reconstituted to original levels in the flour using composites of varying weight percentages of starch granules: 0% small granules (100% large granules), 30, 60, and 100% (0% large granules). A modified straight‐dough method was used in an experimental baking test. Crumb grain and texture were significantly affected. The bread made from the reconstituted flour with 30% small granules and 70% large granules starch had the highest crumb grain score (4.0, subjective method), the highest peak fineness value (1,029), and the second‐highest elongation ratio (1.55). Inferior crumb grain scores and low fineness and elongation ratios were observed in breads made from flours with starch fractions with 100% small granules or 100% large granules. As the proportion of small granules increased in the reconstituted flour, it yielded bread with softer texture that was better maintained than the bread made from the reconstituted reference flour during storage.     

Effect of Single‐Pass and Multipass Milling Systems on Whole Wheat Durum Flour and Whole Wheat Pasta Quality

Single‐pass and multipass milling systems were evaluated for the quality of whole wheat durum flour (WWF) and the subsequent whole wheat (WW) spaghetti they produced. The multipass system used a roller mill with two purifiers to produce semolina and bran/germ and shorts (bran fraction). The single‐pass system used an ultracentrifugal mill with two configurations (fine grind, 15,000 rpm with 250 μm mill screen aperture; and coarse grind, 12,000 rpm with 1,000 μm mill screen aperture) to direct grind durum wheat grain into WWF or to regrind the bran fraction, which was blended with semolina to produce a reconstituted WWF. Particle size, starch damage, and pasting properties were similar for direct finely ground WWF and multipass reconstituted durum flour/fine bran blend and for direct coarsely ground WWF and multipass reconstituted semolina/coarse bran blend. The semolina/fine bran blend had low starch damage and had desirable pasting properties for pasta cooking. WW spaghetti was better when made with WWF produced using the multipass than single‐pass milling system. Mechanical strength was greatest with spaghetti made from the semolina/fine bran or durum flour/fine bran blends. The semolina/fine bran and semolina/coarse bran blends made spaghetti with high cooked firmness and low cooking loss.