Enzymatic corn wet milling: engineering process and cost model
© Ramírez et al; licensee BioMed Central Ltd. 2009
Received: 08 September 2008
Accepted: 21 January 2009
Published: 21 January 2009
Enzymatic corn wet milling (E-milling) is a process derived from conventional wet milling for the recovery and purification of starch and co-products using proteases to eliminate the need for sulfites and decrease the steeping time. In 2006, the total starch production in USA by conventional wet milling equaled 23 billion kilograms, including modified starches and starches used for sweeteners and ethanol production . Process engineering and cost models for an E-milling process have been developed for a processing plant with a capacity of 2.54 million kg of corn per day (100,000 bu/day). These models are based on the previously published models for a traditional wet milling plant with the same capacity. The E-milling process includes grain cleaning, pretreatment, enzymatic treatment, germ separation and recovery, fiber separation and recovery, gluten separation and recovery and starch separation. Information for the development of the conventional models was obtained from a variety of technical sources including commercial wet milling companies, industry experts and equipment suppliers. Additional information for the present models was obtained from our own experience with the development of the E-milling process and trials in the laboratory and at the pilot plant scale. The models were developed using process and cost simulation software (SuperPro Designer®) and include processing information such as composition and flow rates of the various process streams, descriptions of the various unit operations and detailed breakdowns of the operating and capital cost of the facility.
Based on the information from the model, we can estimate the cost of production per kilogram of starch using the input prices for corn, enzyme and other wet milling co-products. The work presented here describes the E-milling process and compares the process, the operation and costs with the conventional process.
The E-milling process was found to be cost competitive with the conventional process during periods of high corn feedstock costs since the enzymatic process enhances the yields of the products in a corn wet milling process. This model is available upon request from the authors for educational, research and non-commercial uses.
The conventional process for wet milling of corn involves chemically pretreating the corn in a solution of sulfurous acid (SO2 in water) followed by physical separation of the co-products and starch. This process is very energy and time consuming. Furthermore, it negatively affects the environment due to the high sulfur dioxide requirements during steeping. According to the Environmental Protection Agency, sulfur dioxide is one of the six most common air pollutants in the United States of America . Sulfur dioxide released to the atmosphere is associated with serious respiratory illnesses. At high levels, it particularly affects people with asthma . Also, oxidation of SO2 in the presence of other polluting gases in the atmosphere, such as nitrogen dioxide (NO2), forms sulfuric acid and causes the formation of acid rain.
Enzymatic wet milling (E-milling) was developed and proposed as an environmentally friendly alternative for conventional corn wet milling . We reported the optimization of conditions for E-milling [5, 6] and showed two important advantages to the use of enzymes in a modified two-stage procedure for wet milling; SO2 is reduced to levels sufficient to inhibit microbial activity and the time for soaking the corn kernel (steeping) is reduced six fold, from 36 to 6 hours.
The process was developed and tested in the laboratory using a batch process; however, a number of important questions were generated that could not be answered without being tested in a continuous system that included recycling streams. A continuous system with recycle streams cannot be tested on laboratory scale and commercial plant testing is required. Commercial plants are reluctant to evaluate the technology without knowing overall cost benefits of the process. There was a need to investigate the amount and cost of energy savings as a result of reducing steeping time to the much shorter pretreatment time. Furthermore, we needed to know if the cost of the enzyme for the pretreatment was going to make the process uneconomical or if perhaps the savings in energy could balance the cost of the enzyme. Finally, there was the proposed prediction that the recycle of the streams in the continuous process would lower the overall enzyme requirement. All of these questions were answered positively with the help of the process engineering and cost models.
Process model description
The process model was developed using process simulator software (SuperPro Designer®) in order to evaluate the continuous production of starch and co-products using E-milling. The model includes processing information such as composition and flow rates of the various process streams, descriptions of the various unit operations, mass and energy balances of each unit operation as well as detailed breakdowns of the operating and capital cost of the facility.
Overview of selected E-milling process equipment
55.556 kg/s m loading rate/belt width
6 h residence time
600 ppm of SO2 in pretreatment tanks
Enzymatic treatment tank
3 h residence time
Mill Starch (MS) thickener
7939 l/min throughput
25% (w/w) solids in underflow
3218 l/min throughput
33% (w/w) solids in underflow
3112 l/min throughput
17% (w/w) solids in underflow
3945 l/min throughput
28% (w/w) solids in underflow
Last stage of starch washing
4558 l/min throughput
1.3 kg fresh water/kg of dry corn
Overall material balance for E-milling model
Corn (15% water)
Dry germ (3% water)
Gluten feed (10% water)
Gluten meal (10% water)
Starch slurry (60% water)
Corn wet milling product yields (conventional and enzymatic) derived from the process models
Conventional yield (%)1
Enzymatic yield (%)1
Gluten feed (Soak water solids plus fiber)
Cost model description
The cost model for the E-milling process was developed using the cost analysis capabilities of the software (SuperPro Designer®). The model includes economic information such as user-supplied equipment purchase and operating costs of the various unit operations, fixed capital investments for the plant, raw materials and consumable costs as well as detailed breakdowns of the operating and capital cost of the facility. The data for our previous model  was obtained from operators of wet milling facilities, equipment suppliers, pricing and cost data reported by trade organizations and government agencies and relevant publications. Inputs from technology suppliers were incorporated into this E-milling study where required. Supplier inputs were obtained from all the major equipment items from suppliers once the process flow diagrams were developed and equipment sizing could be determined. The assembling and analysis of this data was done using the cost estimating program in Superpro Designer®, using generally accepted methods for conducting conceptual economic evaluations for industrial processes . Cost levels in both models were adjusted to reflect economic conditions in the first half of 2007.
Results and discussion
The corn is received, weighed, cleaned and stored in silos. The silo in our model is sized to hold enough corn for three days of operation.
The steeping step of the conventional process is substituted for a short soaking pretreatment during the E-milling process, long enough to increase the moisture content to 50% in the corn kernel prior to grinding. This pretreatment is very important to preserve the integrity of the germ during grinding. In our model, the corn is soaked in a group of three stainless-steel tanks and held in the soaking solution for a total of 6 h at 55°C. The SO2 concentration is 600 ppm for the soaking solution compared with 2000 ppm for the conventional steeping solution. The SO2 is used mainly for microbial control in the E-milling process, not as a chemical processing agent. The reduction on sulfur consumption annually is equal to 461,926 kg for a processing plant with a capacity of 2.54 million kg of corn per day. The soaking is done in a semi-continuous countercurrent system, in the same way the steeping is done in conventional wet milling. During the soaking process, about 46% of the soluble solids are removed and carried in the soak water. The soak water is concentrated, mixed with the corn fiber later in the process and dried to produce corn gluten feed. After soaking, the hydrated corn is submitted to a coarse grinding (first degermination) prior to the enzymatic treatment to allow better penetration of the enzyme.
The ground corn along with the overflows of hydrocyclones used for germ separation (except from the A cyclone of primary germ separation unit) is incubated in a reactor tank with a commercial protease (Prosteep™) for 3 h at a controlled temperature of 50°C and pH of 4.5. The amount of enzyme needed for the treatment is based on experimental data and was calculated as 1 mL/kg of corn in the tank (1117–1210 SAPU/kg of corn). The activity of the protease is expressed in Spectophotometric Acid Protease Units (SAPU). One SAPU is the amount of enzyme that liberates one micromole of tyrosine per minute from the casein substrate under the conditions of the assay. Considering the solid content of the 'fresh' corn and the amount of solids being treated (as part of the recycle), the concentration of enzyme in the treatment tank is set to 1.392 g of enzyme/kg of solid material. The process model shows a consumption of 12.12 kg of protease/h or 0.117 g of protease/kg of fresh corn.
During this step, the protease hydrolyzes the protein matrix (gluten) that surrounds the starch granules. The enzymatic treatment disrupts starch-gluten interactions so that the starch and gluten can be separated. The proteolysis must not be so extensive as to completely degrade the gluten matrix (preventing gluten recovery), which is why not all proteases found to allow starch recovery are acceptable for this process. The proteolysis treatment and the conventional process using sulfites both allow the starch to be isolated by disrupting the starch-gluten interactions; however, the specific chemical mechanism of the two pretreatments is not identical and there can be some additive benefits of using them together .
Germ separation and washing
Protein content, lipid content and unit price of the co-products derived from the wet-milling process models (conventional and enzymatic)
Fiber separation and recovery
This stage of the process remains as the conventional treatment where the degermed corn slurry is passed over the grit screen to separate water, loose starch and gluten (together known as mill starch) from the fiber and bound starch and gluten. The mill starch is sent further in the process, for the separation of gluten and starch. The remaining solids are finely ground to complete the dispersion of the starch and the ground slurry is washed and separated in countercurrent fashion over a set of screens. The clean fiber is dewatered by a screen and a screw press to a final moisture of 60%. This fiber is combined with the concentrated soak water, dried to 10% moisture and sold as corn gluten feed. The corn gluten feed flow is 18,074 kg/h and has approximately the same protein content compared with traditional wet milling (Table 4).
Gluten separation and recovery
As in the conventional process, the gluten is separated from the starch by density differences in a series of three centrifuges where the underflow of the middle one, known as the primary separator, is sent to the starch washing process. The last centrifuge (gluten thickener), along with a rotary vacuum belt filter and a ring dryer, concentrates the gluten to a final moisture of 10%. The gluten is sold as corn gluten meal. The final corn gluten meal (6,285 kg/h in our model) has approximately the same protein content on a dry weight basis as the conventional process (Table 4).
Starch washing and recovery
The washing and recovery of the starch is done in 12 stages in a countercurrent fashion, as it is done in the conventional wet-milling process. This is the only part of the process where fresh water is used to wash the product. The water usage is essentially the same for E-milling and conventional processes (2.3 kg water/kg starch produced). The final starch slurry (144,385 kg/h) contains 60% moisture content with less than 1% of impurities.
Equipment and capital costs
Capital costs by section
(US$ × 100)
(US$ × 100)
Steeping (or pretreatment)
Furthermore, the use of enzymes in the process has the effect of keeping more of the soluble solids with the gluten, fiber and germ and less with the corn soak liquor. The higher soluble loading translates into more material being processed in the gluten, fiber and germ areas and consequently larger equipment capacities and costs. Overall, the combination of the above factors results in a lower capital cost for the E-milling facility of about $4,400,000 or about a 5.5% reduction in costs over a conventional wet milling line.
The operating costs for an E-milling facility are similar to the operating costs for a conventional wet milling facility. We have estimated that the cost of producing a clean starch slurry for further processing by both the wet milling process and the E-milling process would be about $0.193 per kilogram.
The reductions in the capital cost of an E-milling facility are described above. When these savings are spread over a 10-year period the operating cost is reduced by approximately $440,000 per year. Reductions in insurances, taxes and maintenance fees, which are all related to the estimated capital cost savings, provide a reduction in costs of approximately $190,000 per year.
The throughput to the evaporator is limited to a concentration of 50% solids in the syrup leaving the evaporator. Since the concentration of solids in the steep water to the evaporator is lower in the E-milling case, a higher volume of water can be removed in the evaporator at a lower cost than would be achieved in a dryer which results in an additional cost saving of about $40,000 per year.
The lower concentration of sulfur needed results in cost savings of $15,000 per year while the inclusion of the enzymes required for the process adds $1,440,000 per year to the operating costs. The need for sulfuric acid in the enzymatic process for pH adjustment adds $49,000 per year to the operation costs.
The starch co-products produced in a wet milling facility include two protein-based animal feeds (corn gluten meal and corn gluten feed) that are valued for their protein content and a third co-product, corn germ, whose price is a function of its protein content and its lipid content . Table 4 shows the protein and lipid content on a dry weight basis in the co-products for both processes. In E-milling, the protein content slightly decreases for corn gluten meal and corn gluten feed and increases for corn germ. The lipid content in the germ decreases by 1%. The difference in composition of the co-products creates a difference in the unit price for co-products. The unit prices were calculated using the method describe by Johnston et al .
The relative quantities of the co-products also vary from a conventional wet milling facility to an E-milling facility. The result is that the net value of the co-products is 0.3% greater for an E-milling facility than a conventional wet milling plant.
Annual and unit production costs
Corn starch, as a water slurry, is the principal product of the wet milling and the E-milling processes. A comparison of the economics of these processes is then achieved by comparing the unit cost of producing the starch in each case.
Annual operating and production costs
Corn – kg
Other raw materials
Facility related costs
Total operating costs
Corn gluten meal
Corn gluten feed
Total co-product credits
Annual starch production 1
Unit starch production cost ($/kg)
Cost comparison with conventional wet milling
The E-milling process can be economically competitive with the conventional wet milling process and our models have indicated a slight, but not significant, cost advantage to the E-milling process over the conventional wet milling process under the economic conditions that existed in the first half of 2007.
A Technical Cost Model was developed for an enzymatic corn wet milling processing plant with a capacity to process 2.54 million kg of corn per day. This model was used as a tool to understand the differences between the E-milling and conventional wet milling processes, and the cost issues associated with it. We used the model to conduct sensitivity studies using modifications in the price of corn and enzyme. The model allows the user to predict the impact of those modifications in the operating, annual and unit production costs. Our comparison shows that due to the significant recycle of enzyme within the process (quantified using the process model), a significant reduction in the quantity of enzyme necessary over a batch process is possible even if some unaccounted activity losses due to adsorption or inactivation were to occur. It was also found that under current corn and enzyme costs, the E-milling process is slightly more economic on a unit starch production cost; however, it was also shown that under high corn and/or reduced enzyme costs the process can be significantly more economical than the conventional process.
Additionally, the reduction in sulfur consumption was found to be 461,926 kg/year for the model size of 2.54 million kg/year (100,000 bu/day). If adapted industry wide within the United States of America, this would translate into a reduction of 12.6 million kg of sulfur per year.
This model is available upon request from the authors for educational uses and non-commercial research to study the enzymatic wet milling process and to show the impact of changes in the costs of starch, enzyme and co-products. It is not intended to replace a customized process design package. The model requires the use of SuperPro Designer®, Version 7.0, build 17 or later. A free copy of this program can be used to view the model and may be downloaded form the Intelligen website http://www.intelligen.com.
We thank Kevin Hicks for reviewing the process model and manuscript. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not constitute an endorsement by the U.S. Department of Agriculture.
- Corn Refiners Association Inc: Corn Annual. Washington DC. 2007.
- United States Environmental Protection Agency: SO2. How Sulfur Dioxide Affects the Way We Live and Breathe. Office of Air Quality Planning and Standards; 2000.
- Boushey H: Asthma, sulfur dioxide and the Clean Air Act. West J Med 1982, 136: 129-135.
- Johnston DB, Singh V: Use of proteases to reduce steep time and so 2 requirements in a corn wet-milling process. Cereal Chem 2001, 78: 405-411. 10.1094/CCHEM.2001.78.4.405View Article
- Johnston DB, Singh V: Enzymatic milling of corn: optimization of soaking, grinding, and enzyme incubation steps. Cereal Chem 2004,81(5):626-632. 10.1094/CCHEM.2004.81.5.626View Article
- Johnston DB, Singh V: Enzymatic milling product yield comparison with reduced levels of bromelain and varying levels of sulfur dioxide. Cereal Chem 2005, 82: 523-527. 10.1094/CC-82-0523View Article
- Ramírez EC, Johnston DB, McAloon AJ, Yee W, Singh V: Engineering process and cost model for a conventional wet milling facility. Ind Crops Prod 2008, 27: 91-97. 10.1016/j.indcrop.2007.08.002View Article
- Association for the Advancement of Cost Engineering International: Conducting Technical and Economic Evaluations in the Process and Utility Industries. In AACE Recommended Practices and Standards. Morgantown, WV: AACE Inc; 1990.
- Johnston DB, McAloon AJ, Moreau RA, Hicks KB, Singh V: Composition and economic comparison of germ fractions from modified corn processing technologies. JAOCS 2005, 82: 603-608. 10.1007/s11746-005-1116-z
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.