This disclosure relates to corn protein concentrate and methods of manufacturing the same.
For over 100 years, corn wet milling has been used to separate corn kernels into products such as starch, protein, fiber and oil. Corn wet milling is a two stage process that includes a steeping process to soften the corn kernel to facilitate the next wet milling process step that results in purified starch and different co-products such as oil, fiber, and protein. Further corn processing methods are now being investigated to further purify the protein co-product for incorporation into food-grade products, specifically.
Described herein is a corn protein concentrate comprising 55%-80% corn protein on a dry basis, an a* color value between about 0 and 4, and a b* color value between about 15 and 3, and less than about 2% oil on a dry basis.
Further described herein is a method of producing a corn protein concentrate, comprising providing a corn gluten meal, washing the corn gluten meal with a solvent comprising water and a water-miscible organic solvent to obtain a corn protein concentrate comprising 55%-80% corn protein on a dry basis, an a* color value between about 0 and 4, and a b* color value between about 15 and 3, and less than about 2% oil on a dry basis.
The process of producing a corn protein concentrate starts with a corn gluten meal typically comprising at least about 55 wt % protein on a dry basis (note that all reference to percentages herein are weight percentages unless stated otherwise). In most aspects, the starch in the corn gluten meal remains intact and does not undergo a destarching enzymatic hydrolysis process. Similarly, protein structure in the corn gluten remains intact, in most aspects, and does not undergo a denaturation/coagulation process under heat conditions.
The corn gluten meal may then be washed with a water-miscible solvent. In aspects of the present invention, the water-miscible solvent may be an ethanol-containing or isopropanol-containing solvent in concentrations ranging from about 85 wt % to about 99.5 wt %, preferably 85 wt % to about 98 wt % (ethanol or isopropanol), and more preferably in concentrations ranging from 85 wt % to about 95 wt % (ethanol or isopropanol).
A series of solvent washing steps may be performed to remove non-protein, non-starch components. In preferred aspects, there are no more than six solvent washing steps.
Surprisingly, the solvent washes described herein were found to remove many non-protein components (pigments, organic acids, oils, sulfites, etc.) from the starting corn gluten meal, thus enhancing the recovery of the corn protein concentrate as described in more detail below.
In certain aspects, both corn gluten meal and the solvent are introduced in a mixing tank and vigorously mixed for about 15 minutes. To reduce the amount of non-protein, non-starch components contained in the mixture, the mixture goes through an extraction and filtration step. Such extraction may be carried out using a batch stir tank, continuous stir tank reactor or by percolation or immersion extraction. In certain aspects, filtration is carried out using a Buchner funnel to filter out the non-protein, non-starch component-containing solvent and maintain the protein stream. It shall also be understood, however, that while filtration is used in an aspect of this process, other separation techniques such as drainage, percolation, centrifugation or decanting may be utilized to achieve the separation of the non-protein, non-starch component-containing solvent from the protein-containing stream.
The protein-containing stream undergoes another solvent washing, extraction and filtration step and, in preferred aspects, yet another solvent washing, extraction and filtration step therefore achieving three solvent washing steps. This solvent washing step is repeated once more before the protein-containing stream is dried in a desolventizer before recovering the corn protein concentrate.
A goal of the solvent washing process described above is to concentrate the corn protein-starch composition by removal of other non-protein components. Notably, the process described herein produces a corn protein concentrate product comprising 55-80 wt % corn protein on a dry basis (db), and in preferred aspects a corn protein concentrate product comprising 55-75 wt % (db) corn protein.
Another goal of the presently described process is the removal of residual oils, carbohydrates, organic acids, and pigment. The process described herein decreases the oil content so that it makes up less than 2 wt % (db) of the corn protein concentrate, more preferably less than 1.5 wt % (db) and more preferably less than 1 wt % (db).
Furthermore, the process described herein produces a corn protein concentrate wherein the total insoluble carbohydrate concentration ranging from about 15-18 g/kg, with insoluble carbohydrates, having a series of glucose polymers comprising three glucose units linked with alpha 1,4-glycosidic linkages (maltotriose or DP3) and greater than three glucose units (DP4) , comprising at least about 75% of the total insoluble carbohydrate concentration. In the same way that polar solvents favor the extraction of carbohydrates, they also favor the extraction of organic acids. As described herein, organic acids include citric acid, succinic acid, lactate, glycerol, acetate, and propionic acid. Steeping of corn gives rise to a variety of organic acids and some remain in the starting corn protein material used as the raw material for this process. The residual total organic acid concentration in the corn protein concentrate after solvent extraction is about 3.0 g/kg or less.
The starting corn gluten meal may be yellowish-orange in color because most of the corn pigments (luteins, zeaxanthins, cryptoxanthins, and carotenes) concentrate into the protein stream. This color is undesirable for most food-grade applications. Accordingly, the solvent washing step described herein eliminates a substantial amount of the color and provides a corn protein concentrate product having an “a*” color value between about 0 and 4 (and more preferably between 0 and 2), a “b*” color value between about 15 and 35 (and more preferably between 15 and 30), and an “L*” color value ranging between about 70 and 90 (and more preferably between 80 and 90).
There are also circumstances where the primary benefits of the corn protein concentrate are functional—particularly with respect to interactions with water. For example, a benefit of adding protein ingredients to processed meats is enhanced water holding through the cooking process and in such applications, the starch provides a potential benefit. Accordingly, the resulting corn protein concentrate also comprises starch levels ranging from 13 wt % to 23 wt % (db), and more preferably from 13 wt % to 16 wt % (db). Further, there's a desire to reduce the amount of free sulfite for food labeling purposes. The corn protein concentrate described herein has a free sulfite concentration less than 100 ppm.
Even more specifically, the presence of starch in the corn protein concentrate of the present invention provides desirable gelation properties in certain food applications, for example processed meat applications. The corn protein concentrate described herein has a gel strength ranging from about 0.15 to 0.20 N, and more preferably a gel strength ranging from about 0.15 to 0.20 N.
Raw materials for these experiments were collected using a pilot-scale vacuum drum filter to collect and dewater corn gluten meal (CGM) slurry from standard corn wet mill processing. The slurry was collected on the drum filter, rinsed with 1% H2O2 at a wash ratio of approximately 8% and after further draining was collected in plastic bags and frozen. Material was held frozen until use. Frozen CGM was thawed at room temperature before use—generally in the day preceding the work in some combination of room temperature and refrigerator conditions. Corn starch (Argo brand) was acquired from a local grocery.
For starch analysis, the method consists of boiling the CPC sample with aqueous calcium chloride solution to solubilize the starch and then measuring the optical activity of the solution with a polarimeter (derived from Method G-28 of the Corn Refiners Association Standard Analytical Methods).
A series of experiments were conducted to explore the effect of different solvent regimes on extraction in three steps.
Sample CPC070815-1: 200 g of corn gluten meal cake (58.4% moisture, from Cargill Corn Milling, Wahpeton) is suspended in 1000 g of absolute EtOH. After intense mixing with an immersion blender the suspension is left stirring for 15 minutes. The suspension is poured into an 18.5 cm Buchner funnel lined with VWR417 filter paper and drained under vacuum. When the drip rate is about 1/sec, the cake is collected and re-suspended in 1000 g 90% w/w EtOH and left stirring for 15 minutes. The entire process is repeated one more time for a total of three washes. The final cake is drained for approx. 2 minutes beyond the point when the surface solvent disappeared. The cake is spread out in a pie plate and desolventized in a hood initially then in a vacuum oven at about 65° C. overnight.
Sample CPC070815-2: 200 g of corn gluten meal cake (58.4% moisture, from Cargill Corn Milling, Wahpeton, N. Dak.) is suspended in 1000 g of isopropanol. After intense mixing with an immersion blender the suspension is left stirring for 15 minutes. The suspension is poured into an 18.5 cm Buchner funnel lined with VWR417 filter paper and drained under vacuum. When the drip rate is about 1/sec, the cake is collected and re-suspended in 1000 g 90% w/w isopropanol and left stirring for 15 minutes. The entire process is repeated one more time for a total of three washes. The final cake is drained for approx. 2 minutes beyond the point when the surface solvent disappeared. The cake is spread out in a pie plate and desolventized in a hood initially then in a vacuum oven at about 65° C. overnight.
Sample CPC070815-3: 200 g of corn gluten meal cake (58.4% moisture, from Cargill Corn Milling, Wahpeton, N. Dak.) is suspended in 1000 g of solvent comprising 80% w/w ethyl acetate, 20% w/w EtOH. After intense mixing with an immersion blender the suspension is left stirring for 15 minutes. The suspension is poured into an 18.5 cm Buchner funnel lined with VWR417 filter paper and drained under vacuum. When the drip rate is about 1/sec, the cake is collected and re-suspended in 1000 g 72% w/w ethyl acetate, 18% w/w ethanol, 10% w/w water and left stirring for 15 minutes. The entire process is repeated one more time for a total of three washes. The final cake is drained for approx. 2 minutes beyond the point when the surface solvent disappeared. The cake is spread out in a pie plate and desolventized in a hood initially then in a vacuum oven at about 65° C. overnight.
Sample CPC070815-4: 90 g of freeze dried CGM (2.93% moisture by moisture balance, from Cargill Corn Milling, Wahpeton, N. Dak.) is extracted in 1000 g of hexane. After intense mixing with an immersion blender the suspension is left stirring for 15 minutes. The suspension is poured into an 18.5 cm Buchner funnel lined with VWR417 filter paper and drained under vacuum. When the drip rate is about 1/sec, the cake is collected and re-suspended in 1000 g hexane and left stirring for 15 minutes. The entire process is repeated one more time for a total of three washes. The final cake is drained for approx. 2 minutes beyond the point when the surface solvent disappeared. The cake is spread out in a pie plate and desolventized in a hood initially then in a vacuum oven at about 65° C. overnight. The cake is conspicuously yellow-orange and the extraction solution is relatively pale yellow.
Sample CPC070815-5: 90 g of freeze dried CGM (2.93% moisture by moisture balance, from Cargill Corn Milling, Wahpeton, N. Dak.) is extracted in 1000 g of absolute EtOH. After intense mixing with an immersion blender the suspension is left stirring for 15 minutes. The suspension is poured into an 18.5 cm Buchner funnel lined with VWR417 filter paper and drained under vacuum. When the drip rate is about 1/sec, the cake is collected and re-suspended in 1000 g absolute EtOH and left stirring for 15 minutes. The entire process is repeated one more time for a total of three washes. The final cake is drained for approx. 2 minutes beyond the point when the surface solvent disappeared. The cake is spread out in a pie plate and desolventized in a hood initially then in a vacuum oven at about 65° C. overnight. The cake is conspicuously yellow-orange and the extraction solution was relatively pale yellow. These samples filtered very fast compared to the usual extractions in 90% EtOH.
Sample CPC070815-6: 200 g of wet CGM cake (58.4% moisture, from Cargill Corn Milling, Wahpeton, N. Dak.) is extracted in 1000 g of absolute EtOH. After intense mixing with an immersion blender the suspension is left stirring for 15 minutes. The suspension is poured into an 18.5 cm Buchner funnel lined with VWR417 filter paper and drained under vacuum. When the drip rate is about 1/sec, the cake is collected and re-suspended in 1000 g 90% w/w EtOH in water and left stirring for 15 minutes. The entire process is repeated one more time for a total of three washes. The final cake is drained for approx. 2 minutes beyond the point when the surface solvent disappeared. The cake is spread out in a pie plate and desolventized in a hood initially then in a vacuum oven at about 65° C. overnight. The cake is conspicuously yellow-orange and the extraction solution is relatively pale yellow.
Sample CPC070815-7: 200 g of wet CGM cake (58.4% moisture, from Cargill Corn Milling, Wahpeton, N. Dak.) is extracted in 1000 g of absolute EtOH. After intense mixing with an immersion blender the suspension is left stirring for 15 minutes. The suspension is poured into an 18.5 cm Buchner funnel lined with VWR417 filter paper and drained under vacuum. When the drip rate is about 1/sec, the cake is collected and re-suspended in 1000 g 90% w/w EtOH in water and left stirring for 15 minutes. The entire process is repeated one more time in this manner The cake is then washed two times for ten minutes with 1000 g 90% w/w EtOH in water. The final cake is drained for approx. 2 minutes beyond the point when the surface solvent disappeared. The cake is spread out in a pie plate and desolventized in a hood briefly then in a vacuum oven at about 65° C. overnight. The cake is conspicuously yellow-orange and the extraction solution is relatively pale yellow.
Before extraction, CGM is about 67% protein (db) and 4-6% oil (db). All of the solvents tested decreased the oil content by 80% or more (Table 1). Protein concentrations are equal or somewhat higher after extraction. The color metric L* increases considerably when wet CGM cake is used as starting material, but less so when freeze-dried material is used. The color measures a* and b* are decreased in almost all cases, with the greatest effect apparent with 90% isopropanol. Some protein may have been lost during extraction as some corn proteins are solvent soluble, especially in the absence of prior heat treatment. In contrast, starch is insoluble in these solvents and is concentrated through the process.
Two pilot trials were conducted to test initial conditions for processing of CGM. The behavior of CGM in the extractor and desolventizer created some issues with stickiness and solvent removal, but are attributed to the presence of starch and partial protein solubilization. Subsequent trials were conducted to produce samples for functional testing. The various operating conditions are shown in Tables 2 and 3.
The first two attempts to produce a CPC at pilot scale are primarily concerned with understanding processing issues. The first attempt resulted in a higher protein, higher oil composition than the second attempt (Table 4), but demonstrates that the process could be used to significantly decrease the oil content and pigment of CGM (compare to data in Table 5).
In the subsequent pilot trials, the protein concentrations were lower and starch concentrations were higher, partly due to higher initial starch concentrations (Table 5). Without being bound by any theories, one explanation may be that extraction conditions dissolved sufficient protein (mainly alpha-zein) to reduce the final concentration because starch was not extracted in a proportional manner Oil concentration is decreased more than 95%. The input EtOH concentration used is higher in the production of CPC120 than CPC119, but there is about the same final protein concentration. In a complementary manner, the starch concentration is higher in the finished product than the starting material.
Table 6 shows that extraction decreased monosaccharides, had almost no impact on di- and trisaccharides, while “concentrating” higher oligomers. Extraction also removed 90% or more of the lactic acid (Table 7). It is unclear whether there was any change in the succinic acid or citric acid concentrations.
Removal of pigments during pilot extraction results in an overall lighter product (Table 8) with significant declines in the residual pigments contributing to yellow (a*) and red (b*). The higher EtOH concentration used with CPC120 seems to result in a less intensely colored product across all three measures.
One of the hypothesis behind development of a lower protein variant is that the starch present from the CGM will have a strong affinity for water and can be used to hold water through a cooking cycle. The samples from the earliest attempts were never tested for functionality, but samples CPC119 and CPC120 are suitable for functional testing. Two measures of functional behavior were particularly notable: viscosity and gelation.
The two pilot corn protein concentrate samples have very high viscosities compared to an equal concentration of corn protein isolate (CPI) (Table 9 and
Dried CGM (whether in the vacuum oven or production dryer) has a relatively low viscosity at each measuring point and shows relatively little structure in the response (
Two features of the viscosity profiles for CPC samples should be noted. First, they reach a peak viscosity later in the heating cycle than the more pure protein variant. Second, upon cooling they initially lose some viscosity, but then regain all or more of the lost viscosity to finish near the peak viscosity. Samples CPC119 and CPC120 differ in peak viscosity. This seems like the most dramatic consequence of the differing solvent compositions used in extraction.
This might be the expected consequence of replacing some of the protein with starch, but
The viscosity results suggest that samples CPC119 and CPC120 should form good gels and this is what was observed.
CGM that was dried in a vacuum oven, ground and then tested in the gel method formed a solid at the tube's tip where the particulates had settled (
Corn gluten cake is collected on a rotary drum vacuum filter with rinsing. The destarched slurry is fed to the drum at 1.2 gal/min at a density of about 1.016 g/mL. The rinse water supplemented with hydrogen peroxide to a concentration of 0.31% w/w active hydrogen peroxide is applied at 0.12 gal/min. Upon completion of the vacuum dewatering, the treated cake is frozen until further use.
10 kg of peroxide-treated, corn gluten cake with 60-65% moisture is processed through a dual rotor crusher with a 0.125-inch screen to generate a uniformly sized particle for homogeneous extraction. The cake is fed to a Crown Iron Works Model IV immersion extractor using a drag conveyor dropping through a crossover screw and then a delumper (for a better understanding, an illustration of the Crown Iron Works Model IV immersion extractor may be found on the crowniron.com website) into the extractor. The extractor includes a series of inclined drag conveyors arranged so that the lower end of the conveyor is submerged in the extraction solvent and the upper end was above the solvent. The conveyor carried the solids forward such that the material is initially submerged in solvent and then the material emerged from the solvent and excess solvent drained back into the solvent stream. At the end of the conveyor, the solids dropped onto another conveyor with a similar arrangement. The model IV extractor has six extraction stages. Fresh solvent is introduced at the discharge end and flowed towards the inlet end and is ultimately discharged at a point preceding the solids introduction.
After the final solvent contact, the solids are conveyed up a long section to allow more extensive draining before falling into a crossover screw for transport to desolventizing. The solvent is fed into the system at 0.445 kg/min and the solids are introduced at 0.027 kg/min (based on a calibrated volumetric feeder) and the solvent is maintained at 56° C. by in situ heat exchangers. Total solvent to solids ratio is about 16 and total contact time is about 60 minutes. The water of the extraction system is introduced through a combination of carryover water from the input material and water in the fresh solvent. The composition of the feed solvent to contact the extracted destarched com gluten is approximately 92.2% ethanol and 7.8% water. Consequently, the composition of the solvent varied across the extractor, but the final solvent concentration is about 92% ethanol.
Desolventizing occurred in a Bepex Solidaire dryer operated with a surface temperature of about 155-160° C. and an absolute pressure from about 270-330 millibar (with a target of about 300 millibar).
The resulting corn protein concentrate product is about 54.9% protein (dry basis). Further, the oil is less than 0.5% on a dry basis, the product color, as measured using the Hunter colorimeter, has “L*” color equal to 85.5, “a*” color equal to 1.1 and “b*” color equal to 20.5. The free sulfite is 89 mg/kg (dry basis).
Corn gluten cake is collected on a rotary drum vacuum filter with rinsing. The destarched slurry is fed to the drum at 1.2 gal/min at a density of about 1.016 g/mL. The rinse water supplemented with hydrogen peroxide to a concentration of with 1% w/w is applied at 0.12 gal/min. Upon completion of the vacuum dewatering, the treated cake is frozen until further use.
10 kg of peroxide-treated, corn gluten cake with 60-65%% moisture is processed through a dual rotor crusher with a 0.125-inch screen to generate a uniformly sized particle for homogeneous extraction. The cake is fed to a Crown Iron Works Model IV immersion extractor using a drag conveyor dropping through a crossover screw and then a delumper (for a better understanding, an illustration of the Crown Iron Works Model IV immersion extractor may be found on the crowniron.com website) into the extractor. The extractor includes a series of inclined drag conveyors arranged so that the lower end of the conveyor was submerged in the extraction solvent and the upper end was above the solvent. The conveyor carries the solids forward such that the material was initially submerged in solvent and then the material emerged from the solvent and excess solvent drained back into the solvent stream. At the end of the conveyor, the solids dropped onto another conveyor with a similar arrangement. The model IV extractor had six extraction stages. Fresh solvent is introduced at the discharge end and flows towards the inlet end and is ultimately discharged at a point preceding the solids introduction. After the final solvent contact, the solids are conveyed up a long section to allow more extensive draining before falling into a crossover screw for transport to desolventizing. The solvent is fed into the system at 0.445 kg/min and the solids were introduced at 0.027 kg/min (based on a volumetric feeder) and the solvent is maintained at 59° C. by in situ heat exchangers. Total solvent to solids ratio is about 16 and total contact time is about 60 minutes. The water of the extraction system is introduced through a combination of carryover water from the input material and water in the fresh solvent. The composition of the feed solvent to contact the extracted destarched com gluten is approximately 97.3% ethanol and 2.7% water. Consequently, the composition of the solvent varied across the extractor, but the final solvent concentration was about 97% ethanol.
Desolventizing occurred in a Bepex Solidaire dryer operated with a surface temperature of about 155-160° C. and an absolute pressure from about 270-330 millibar (with a target of about 300 millibar).
The resulting corn protein concentrate product is about 57.5% protein (dry basis). Further, the oil is less than 0.5% on a dry basis, the product color, as measured using the Hunter colorimeter, has “L*” color equal to 89.9, “a*” color equal to 0.5 and “b*” color equal to 17.2. The free sulfite is 89 mg/kg (dry basis).
The formation of a viscous dispersion or gel during and after heating can be useful in many food systems, often improving the texture or yield of the food. One possible non-limiting example of this functionality can be seen in a model system based on a beef frank, which is a kind of emulsified meat product. The model system was adapted from Paulson et al. (1984) Can. Inst. Food Sci. Technol. J. 17:202-208.
A 36 g sample of 93% lean ground beef is weighed into a dish and stored at about 4° C. until use. A 45 g sample of lard (Armour) is weighed into a separate dish and stored at ambient temperature (20-25° C.) until use. 25 g of cold tap water is weighed into a centrifuge tube and stored at 4° C. until use. Another 33 g of tap water is weighed into a cup and stored at 4° C. until use. Salt (4.5 g) is weighed into a small dish and protein additive (4 g) is weighed into another small dish. Both of the latter are stored at ambient (20-25° C.) temperature until use.
A Cuisinart mixing bowl is mounted onto the base (Cuisinart Little Pro Plus). The protein for the batch is added to a tube containing 25 g of water, shaken and left to hydrate at room temperature for 2-4 minutes. The pre-weighed meat is added to the Cuisinart bowl and pulsed 2-3 times to break up the chunks. The salt is added and pulsed a few times. The hydrated protein and remaining water are added to the bowl and pulsed 2-3 times. Finally, the lard is added to the bowl and pulsed 2-3 times. The Cuisinart is run with constant mixing for 1 minute, the sides are scraped down, and the mixer run another one minute. Two 30 g samples are removed and placed into 50 mL plastic centrifuge tubes with screw top closures. After vigorous tapping to settle the material, the tubes are centrifuged for 1 minute at 3000 g to force out entrained air. The tubes are placed into a 75° C. water bath for 35 minutes to cook. At the end of the heating, the tubes are removed from the bath, allowed to partially cool, and the liquid is decanted into pre-weighed aluminum dishes and weighed. The liquid lost is subtracted from the initial weight and used to calculate the mean yield. A reference is prepared in the same way but without the protein added to the 25 g of water. The protein ingredient provides a substantial yield boost to the finished product. Results are highlighted in Table 11.
This application claims priority to U.S. Provisional Application No. 62/312,867, filed Mar. 24, 2016, which is hereby incorporated by reference in its entirety.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/US2017/023999 | 3/24/2017 | WO | 00 |
Number | Date | Country | |
---|---|---|---|
62312867 | Mar 2016 | US |