Published online 8 January 2009
Published in Agron J 101:201-206 (2009)
DOI: 10.2134/agronj2007.0398
© 2009 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
Improving the Fermentation Characteristics of Corn through Agronomic and Processing Practices
Graig Reicks*,
Howard J. Woodard and
Anthony Bly
Plant Sci. Dep., South Dakota State Univ., Brookings, SD 57007-1096
* Corresponding author (Graig.Reicks{at}SDstate.edu).
 |
ABSTRACT
|
|---|
This study determined the influence of corn (Zea mays L.) hybrids, N availability, grain harvest moisture, and drying temperatures on dry-mill ethanol production. Six hybrids, ranging from 92 to 108 d in relative maturity (RM), were planted at two locations over 2 yr. One of four N fertilizer treatments were applied. Grain was hand-harvested at grain moistures of 20 and 25%. Grain was dried to about 15% moisture at either 25, 38, 52, or 60°C in 2003, and 38, 66, 75, or 93°C in 2004. Ethanol was measured after grain was subjected to a small-scale bench fermentation process. Grain yield increased at all four site-years as available N increased to the recommended N application rate. Relative ethanol concentration was generally not affected by normal N fertilizer rates. Significant reductions in relative ethanol concentration occurred at the both the highest and lowest N rates in one-of-four site years. Hybrids designated as high fermentable starch (HS) by the company did not necessarily yield more ethanol than other hybrids. Ethanol concentration was reduced by 0.3% at Brookings for grain that was subjected to a killing frost. Ethanol concentration generally did not differ between grain dried at 38 and 52°C in 2003. Ethanol from grain harvested at 25% moisture and dried at 25°C was 0.1 to 0.3% lower than when grain was dried at 38 or 52°C. Drying temperatures of 25 to 52°C had no influence on relative ethanol concentration when the grain was harvested at 20% moisture. However, ethanol concentration was lowered 0.1 to 0.4% when drying temperature increased to 93°C in 2004. These results suggest that producers should apply the recommended N rates for maximum economic yield, plant adapted hybrids, and dry corn grain between 38 and 52°C to maximize relative ethanol concentration.
Abbreviations: HS, high fermentable grain starch • R4, growth stage of corn when kernels are considered a dough consistency • RM, relative maturity
Received for publication December 13, 2007.
|
INTRODUCTION
|
|---|
THE INFLUENCE OF AGRONOMIC FACTORS on grain components important to ethanol production and fermentation efficiency has not been investigated completely. Previous research has focused on agronomic factors that influence grain composition. Uribelarrea et al. (2004) reported that starch concentration was negatively affected by protein concentration. Corn kernels have several types of protein, of which zein is the easiest to manipulate by soil N management. Tsai et al. (1978) observed that endosperm zein protein content increased from 1.86 to 6.21 mg for Pioneer 3369A hybrid in response to three different N application rates (66, 132, and 198 kg ha–1). Rendig and Broadbent (1979) found that N composition of the zein protein increased from 0.15 to 0.55% when the applied N rate increased from 0 to 352 kg ha–1.
Studies in vitro have shown that low N rates can also lower endosperm starch and that N is needed, up to a point, to produce endosperm starch (Singletary and Below, 1989 and (Singletary and Below,1990). Furthermore, inbreds (Wyss et al., 1991) and hybrids (Uribelarrea et al., 2004) can require different levels of N in growth medium for maximum kernel weight and maximum N concentration in the seed.
The RM is another factor that could affect the ratio of grain protein to starch. However, Jones et al. (1996) reported that RM affected the time of kernel development, with 10 and 16 d after pollination needed for maximum endosperm cell number in early- and late-maturing hybrids, respectively.
The relationship between grain starch and ethanol yield is not completely understood. The highest starch-producing hybrid was not the highest ethanol producer in one study (Dein et al., 2002). Predicting factors for a hybrid's ethanol production potential are not clear. Hybrids may have higher ethanol production potential under different climatic conditions, just as their grain yields vary by location and year. Therefore, ethanol production may not necessarily be strongly related to grain starch content.
Much of the research regarding the influence of harvest moisture and grain drying temperature on total starch available has focused on ethanol production in the wet-mill process. Few studies address the same issues with regard to the dry-mill ethanol production process. Studies with the wet-mill process indicated that decreasing grain harvest moisture will increase extractable starch (Weller et al., 1988) and high drying temperatures (above 90°C will decrease extractable starch (Haros et al., 2003; Paulsen et al., 2003; Weller et al., 1988).
Other research has shown that increased artificial drying temperatures caused the Maillard reaction to occur (Pizzoferrato et al., 1998; Yang et al., 1998). Starch begins to conjugate with proteins in the grain, possibly making the starch unavailable to the enzymes used for the hydrolysis reaction into glucose. Grain harvested at higher moisture levels will need more time and possibly higher temperatures to dry. This could increase the Maillard product level compared to grain harvested at lower moisture concentrations, potentially reducing efficiency of ethanol production. Since numerous production and processing factors have been shown to influence grain starch content, it would be important to study the individual contributions of N availability, hybrid, harvest moisture, and drying temperature on the dry mill process for ethanol production. A field study was conducted to investigate how corn hybrids, available soil N, and drying temperature interact to influence ethanol production.
|
MATERIALS AND METHODS
|
|---|
Two field sites were chosen for a 2-yr study: Brookings (44°18'37'' N, 96°47'50'' W) and Beresford (43°4'51'' N, 96°46'26'' W), in eastern South Dakota. Beresford is approximately 150 km south of Brookings and receives 75 to 125 mm more annual precipitation and from 160 to 280 more annual growing degree days (°C) per year than Brookings.
The field study at both locations arranged in a split-plot design compared six hybrids as main plots and four N rates at subplots. All treatments were replicated four times. In the Brookings and Beresford areas, producers plant hybrids with maturity averages of 97 and 102 d, respectively. Six corn hybrids ranging from 92 to 99 d in maturity length were planted at Brookings (Brandt, a fine-silty, mixed, superactive, frigid Calcic Hapludolls, 0–2% slopes). Six hybrids ranging from 94 to 108 d in maturity were planted at Beresford (Chancellor, a fine, smectitic, mesic Vertic Argiaquolls, 0–2% slope). Three of the six hybrids were grown at both locations. Three hybrids grown at each location were HS hybrids designated by the seed company. The HS hybrids selected are supposed to have grain characteristics that were preferred by dry-mill ethanol plants.
Twelve random soil cores were removed from uniform field sites from 0 to 15 cm and 15 to 60 cm depth increments at each site-year before planting. Samples were dried, crushed, and analyzed for extractable NO3–N and other nutrients. Fertilizer N needs were determined for typical yield goals of about 10,000 kg ha–1 for both locations. The recommended N rate was based on the soil test N recommendations for corn and considered legume or manure credits from the prior crop year (Gerwing and Gelderman, 2002). Fertilizer N treatments were applied at four rates: 0% or no fertilizer applied (control); 50% rate (half of the recommended N fertilizer rate); at 100% rate (recommended N fertilizer rate); or 200% rate (twice the recommended N fertilizer rate). All N fertilizer was broadcast as urea (46–0–0) after the planting operation was completed (Table 1
). Air and soil temperatures were cool during planting (10–12°C), and it was assumed that N volatilization was negligeable
View this table:
[in this window]
[in a new window]
|
Table 1. Composite soil test NO3–N level, estimated legume credits, applied fertilizer N, and estimated total available N for recommended N rate for site and year.
|
|
Corn was planted with a tractor-mounted, two-row no-till planter, in 76-cm rows with either six or eight rows per plot. All hybrids contained glyphosate resistance and Bacillus thuringiensis, spp. kurstaki to control European corn borer [Ostrinia nubilalis (Hübner)] transgenic traits. Insecticide (3.0% tefluthrin) was applied at a 2 kg ha–1 rate in a T band at planting in 2003. Insecticide (2.1% cyfluthrin) was applied at a 3.2 kg ha–1 rate in a T band at planting in 2004. Two 1168 mL ha–1 herbicide (48.8% glyphosate) applications were used for weed control during the growing season.
Grain moisture concentration was monitored on a weekly basis in the field with a hand-held moisture tester after the grain reached physiological maturity (black layer). To compare both harvest grain moisture and grain drying temperature effect on relative ethanol concentration, whole ears were hand-harvested from the 100% N rate plots when grain moisture concentrations reached 250 g kg–1 and again at 200 g kg–1. Sixty ears were randomly harvested from the rows adjacent to the middle two rows of each plot. The 60 ears were randomly split into four 15-ear lots, and each lot was placed into a two-gallon plastic bag. The 15-ear lots were stored at 4°C to be used in a drying experiment to be described later. Grain was harvested from the two center rows with a two-row plot combine at about 150 g kg–1 grain moisture to determine grain yield.
Grain was dried on the cobs in this study because it did not shell effectively when harvested at 250 g kg–1 moisture. Cobs harvested for relative ethanol concentration were dried in a large forced-air dryer in 2003. Ears were removed from their respective sealed plastic bags and placed into paper bags with open tops. One 15-ear lot from each plot was dried at either 25 (control), 38, 52, and 60°C in 2003. During the drying procedure, the grain moisture level was tested several times daily with the same hand-held moisture tester used in the field. Because the 2003 ethanol yields showed little response to drying temperature, the levels of drying temperature were increased to 38, 66, 80, and 93°C for grain harvested in 2004. A drying oven was used to dry the cobs because the force draft air dryer used in 2003 was not able to accommodate the higher drying temperature ranges. Cob moisture from one grain sample from each drying rack was checked several times daily until grain moisture reached about 150 g kg–1. Grain was then shelled from each cob, blown clean with a low pressure air source, and sieved to remove foreign matter before weighing and processing.
Grain was distilled by a 54-h fermentation bench test similar to the ethanol production process of Singh and Graeber (2005). The ethanol concentration of the resulting solution was measured using high-performance liquid chromatography and expressed as volume of ethanol per volume of liquid filtrate as a fraction ratio. The 54-h fermentation method was used by the seed company to rank their hybrids for relative ethanol concentration potential.
A PROC MIXED model was applied as the statistical analysis procedure on the data (SAS Institute, 2000) to determine the influence of the main effects (N treatment, hybrid, drying temperature) and interactions on grain yield and ethanol concentration. The drying temperature and harvest moisture portion of the study was considered a split-split plot design, as hybrid was the main plot and harvest moisture and drying temperature were the subplots.
|
RESULTS AND DISCUSSION
|
|---|
Grain Yield
Mean grain yield response to N treatment increased significantly for each increment of applied N from the control (0% N treatment) to the 100% N treatment level for all site-years (Fig. 1
). Three of four site-years for which the 100% N treatment was applied, grain yield exceeded the goal of 10,000 kg ha–1 indicating that soil test methodology for this N treatment was appropriate. Only at the Brookings site in 2004 was this grain yield of 10,000 kg ha–1 not achieved at this recommended N treatment rate. Brookings experienced an early frost that year and the yield fell short by about 6% less than the yield goal (data not shown). Additional grain yield increases above the 100% N treatment were measured as N treatment increased to 200% for both sites in 2004 only. During this year, mean grain yield increased at Brookings and Beresford, respectively, by 11.5% and 9.4% (data not shown), over the 100% N treatment.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 1. Mean grain yield response (pooling all hybrids) to N treatment (% of recommended fertilizer N) at Brookings and Beresford for 2003 and 2004. Grain yield mean for a particular N treatment with a different letter statistically significant from another N treatment within a particular site-year when analyzed by the Student's t test at the alpha = 0.05 probability level. Dotted line at 10,000 kg ha–1 yield represents the yield goal for which N was applied at the 100% N treatment for all site-years.
|
|
Ethanol Production
The variables of site, year, and N treatment interacted to influence ethanol concentration (P = 0.049) (Table 2
). The significant reduction in ethanol concentration at the 0% N rate at Brookings in 2004 relative to Beresford (Table 3
and Fig. 2
) appears partially-responsible for this three-way interaction. This may have been caused by unusually frigid weather in Brookings on 19 Aug. 2004, during the R4 (dough) growth stage, as night-time temperatures dropped to 0°C. The cold temperature damaged, but did not kill the crop. Apparently, lack of available N combined with the frost damage altered grain composition to adversely affect ethanol concentration.
View this table:
[in this window]
[in a new window]
|
Table 2. PROC MIXED analysis of ethanol concentration and grain yield influenced by Site (S), Year (Y), Hybrid (H), and N treatment (N).
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Mean responses of grain yield, ethanol concentration, grain starch, and grain protein at Brookings and Beresford for 2003 and 2004. Means with a different letter were significantly different from another site year response with the same N treatment when analyzed by the Student's t test at the alpha = 0.05 probability level.
|
|
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 2. Mean grain ethanol concentration responses to N treatment (percent of recommended N application) at Brookings and Beresford for 2003 and 2004. Ethanol concentration means for a particular N treatment with a different letter was significantly different from another site-year response for the same N treatment when analyzed by the Student's t test at the alpha = 0.05 probability level.
|
|
Ethanol concentration decreased at the highest N rate at the Brookings site in 2003 (Fig. 2), which also explains the three-way interaction between site, year, and N treatment. Ethanol concentration was similar for the 50 and 100% N rates, but decreased by 0.19% as N rate increased to 200%. Increased N rates were expected to depress relative ethanol concentration by causing excessive grain protein accumulation (Singletary and Below, 1989; Uribelarrea et al., 2004). However, grain from the other three site-years also experienced similar protein increases at the 200% N rate, but without the depression in relative ethanol concentration (Table 3). Excessive grain protein accumulations alone cannot explain decreases in relative ethanol concentration.
The variables of site and hybrid also interacted to influence ethanol concentration (P = 0.019) (Table 2). Hybrid X4, a hybrid common to both sites, reached physiological maturity at Beresford during both years, but only accumulated one-half of its hard starch layer at Brookings in 2004 (Table 4
). However, mean ethanol concentration for this hybrid was similar for the Brookings and Beresford sites, respectively (Table 5
). Hybrid X6, another hybrid common to both sites, reached physiological maturity at Beresford during both years, but only accumulated one-third of its hard starch layer at Brookings in 2004 (Table 4). Mean relative ethanol concentration of this hybrid was significantly higher at Beresford than Brookings (Table 5). This suggests that relative ethanol concentration potential may be reduced when kernel development is arrested at or before one-half of its hard starch layer has accumulated. Afuakwa and Crookston (1984) reported dramatic increases in kernel weight, beginning at the onset of the R5 dent stage and continuing until the one-half milk-line stage (one-half of hard starch layer formed). Increases in kernel weight were less dramatic as grain matured from the one-half milk line stage until physiological maturity. Terminating kernel filling before the one-half milk line may reduce grain components necessary to maximize relative ethanol concentration.
View this table:
[in this window]
[in a new window]
|
Table 4. Maximum kernel maturity of corn hybrids grown at Brookings and Beresford, South Dakota, after a killing frost in 2004.
|
|
View this table:
[in this window]
[in a new window]
|
Table 5. Student's t test comparison of mean grain ethanol concentration response to Hybrid (H) grown at Brookings and Beresford, SD in 2003 and 2004.
|
|
Year was an important single main effect (P < 0.001) influencing ethanol concentration (Table 2) but site was not significant (P = 0.269). In 2003, all hybrids reached physiological maturity in Brookings and Beresford before grain was harvested. In 2004, none of the hybrids in Brookings reached physiological maturity before the first killing frost (Table 4). A few very cool nights in late August (3–5°C) may have slowed maturing. At Beresford, four of six hybrids reached physiological maturity in 2004. Ethanol concentration was reduced an average of 1.8% from 2003 to 2004 (Table 3). Hybrids X1 through X6, grown in Brookings, and hybrids X3, X4, X6, X7, X8, and X9, grown in Beresford represent rankings of increasing greater maturity length (Table 5). Kernel maturity was delayed commensurate with hybrid maturity at Brookings and Beresford after the recorded killing frost in 2004. The milk line advancement for the hybrids in Brookings progressed further for earlier hybrids than later maturing hybrids, which no doubt influenced grain components important to relative ethanol concentration compared to 2003. The killing frost in 2004 was probably also the main reason that the grain yield at Brookings fell about 6% short of the yield goal of 10000 kg ha–1 (Fig. 1).
The X7 hybrid was designated as a HS hybrid, and when compared to other hybrids grown at Beresford, X7 had a higher relative ethanol concentration than X3 (Table 5), which was also designated as a HS hybrid. The longest maturing hybrids, X8 and X9, were both non-HS hybrids and had significantly less relative ethanol concentration than the other hybrids. Their kernel development was arrested at the R5 (dent) stage with two-thirds to three-fourths of their visible hard starch layers formed when the first killing frost occurred at Beresford in 2004 (Table 4).
Influence of Grain Drying Temperature
The overall range for the grain drying temperature was increased from 2003 to 2004, so it was more appropriate to perform a statistical analysis on the data by separate years. The 25°C control was eliminated in 2004 to add a higher temperature treatment because grain is more-commonly dried with artificial heat at temperatures as high as 93°C than temperatures near 25°C. In addition, grain drying advanced very quickly after physiological maturity had been attained at Beresford in 2004, so only one hybrid moisture level (200 g kg–1) could be analyzed for all six hybrids. However, data was collected for three of the six hybrids for both hybrid moisture levels (200 and 250 g kg–1) for this site-year. So it was more appropriate to analyze the data by separate sites for the influence of harvest moisture, drying temperature, and hybrid on ethanol concentration.
Drying temperature significantly affected relative ethanol concentration for all four site-years (Table 6
). A significant interaction between hybrid and drying temperature (P = 0.034) impacted relative ethanol concentration at Brookings in 2003 (Table 7
). There was no drying temperature effect on the earliest maturing hybrids (X1 and X2). Relative ethanol concentration was reduced somewhat at the highest drying temperatures for hybrids X3 and X4. However, this reduction may not be relevant as ethanol concentration at these temperatures were equivalent statistically to the ethanol for the lowest drying temperature. Likewise, the ethanol concentration of the latest maturing hybrids (X5 and X6) were both unaffected by the highest drying temperatures during either growing season because mean ethanol concentration was statistically equivalent to the ethanol concentration produced at the lower drying temperatures.
View this table:
[in this window]
[in a new window]
|
Table 6. PROC MIXED analysis of ethanol concentration influenced by harvest moisture (M), grain drying temperature (T), and hybrids (H) by site and year.
|
|
View this table:
[in this window]
[in a new window]
|
Table 7. Student's t test comparison of mean of the hybrid x drying Temperature (T x H) interaction for mean ethanol concentration of corn grain dried to 155 g kg–1 moisture harvested from Brookings, South Dakota, in 2003.
|
|
Harvest moisture and drying temperature interacted in 2003 to significantly affect relative ethanol concentration at Brookings (P < 0.001) and Beresford (P = 0.006) (Table 6). This interaction can be explained by two events. In the first event, lower relative ethanol concentrations were measured from grain harvested at 250 g kg–1 moisture and then dried at the low temperature (25°C). When grain at 250 g kg–1 moisture was dried at 25°C, respiration from microbial activity reduced grain carbon levels lowering fermentable starch (Steele et al., 1969). No fungi or other microbial activity was evaluated/observed in this study.
The second event that contributed to the interaction between harvest moisture and drying temperature involved a significant reduction in ethanol concentration as the drying temperature of grain harvested at 250 g kg–1 moisture was increased from 38°to 60°C (Table 8
). When the initial grain harvest moisture level was 200 g kg–1, mean ethanol concentration was not reduced as drying temperature increased from 38°to 60°C. When the 38°C grain drying temperature was increased to 66°C for grain grown at Beresford in 2004, relative ethanol concentration decreased, regardless of grain moisture (Table 9
). This provides evidence that reductions in relative ethanol concentration begin at drying temperatures near 60°C, especially from grain of a higher moisture content that required longer heat exposure.
View this table:
[in this window]
[in a new window]
|
Table 8. Student's t test comparison of harvest moisture x drying temperature (M x T) interaction of grain ethanol concentration of corn dried to 155 g kg–1moisture from two sites in South Dakota in 2003.
|
|
View this table:
[in this window]
[in a new window]
|
Table 9. Student's t test comparison of mean drying temperature (T) influence of grain dried to 155 g kg–1 moisture on mean grain ethanol concentration from two sites in 2004.
|
|
The Maillard reaction likely caused the lowered ethanol concentrations for all four site-years (Tables 8 and 9) as drying temperature increased. This reaction occurs when heat causes the carbonyl group of a sugar (glucose) to conjugate with the amino group of an amino acid to produce N-substituted glycosylamine and water. Yang et al. (1998) found lower digestability of a modified corn starch by 
-amylase when the starches were conjugated with lysine or poly(lysine) to form Maillard reaction products. Pizzoferrato et al. (1998) also reported that 100°C heat reduced potato starch release of glucose compared to room temperature (25°C).
Other researchers have struggled to find effective predictors for ethanol production from corn grain. Dein et al. (2002) reported that ethanol concentration was not exclusively dependent on starch content after considering several grain components. They found the fermentation efficiencies of six different hybrids to vary between 87 and 96%, even though the starch concentrations of the hybrids did not differ significantly. Fermentation efficiency was defined as the theoretical ethanol concentration based on the starch content of each hybrid. Singh and Graeber (2005) also found almost no correlation between ethanol concentration and extractable starch as determined by a laboratory wet milling process. These individuals also found almost no correlation between ethanol concentration and nondestructive NIR analyses of starch, protein, or oil. Future research should attempt to find correlations between ethanol production and other grain characteristics.
|
CONCLUSION
|
|---|
Hybrids in which the grain filling period was shortened by a killing frost produced significantly less ethanol concentration than hybrids that reached physiological maturity. Producers should always plant hybrids which are adapted to the local growing conditions that will usually reach physiological maturity before the first killing frost.
Several HS hybrids were selected for this study and compared to other hybrids without that designation. Only one of four HS hybrids actually produced more ethanol, bringing the entire HS evaluation system into question.
Ethanol concentration was not influenced by higher N treatments even though the grain yield doubled with the recommended N treatment compared to the 0% N treatment. Higher N treatment probably increased grain protein content at the higher N treatments (although this was not measured), but did not lower dry-mill ethanol concentration. There is some evidence from other studies that higher N treatments could have produced an amino acid composition more favorable to yeast fermentation than lower low N rates.
In most comparisons, drying temperatures of 38 and 52°C produced high ethanol concentrations, regardless of grain moisture at harvest. Ethanol response to other drying temperatures was less consistent, but tended to be lower for temperatures below 38°C and lower for temperatures above 52°C.
|
NOTES
|
|---|
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.
|
REFERENCES
|
|---|
- Afuakwa, J.J., and R.K. Crookston. 1984. Using the kernel milk line to visually monitor grain maturity in maize. Crop Sci. 24:687–691.[Abstract/Free Full Text]
- Dein, B.S., R.J. Bothast, L.B. Iten, L. Barrios, and S.R. Eckhoff. 2002. Fate of Bt protein and influence of corn hybrid on ethanol production. Cereal Chem. 79:582–585.
- Gerwing, J., and R. Gelderman. 2002. Fertilizer recommendation guide. EC750. South Dakota State Univ., Coop. Ext. Serv., Brookings, SD.
- Haros, M., M.P. Tolba, and C. Suàres. 2003. Influence of corn drying on its quality for the wet-milling process. J. Food Eng. 60:177–184.
- Jones, R.J., B.M.N. Schreiber, and J.A. Roessler. 1996. Kernel sink capacity in maize: Genotypic and maternal regulation. Crop Sci. 36:301–306.[Abstract/Free Full Text]
- Paulsen, M.R., S.W. Mbuvi, A.E. Haken, B. Ye, and R.K. Stewart. 2003. Extractable starch as a quality measurement of dried corn. Appl. Eng. Agric. 19:211–217.
- Pizzoferrato, L.M. Paci, and G. Rotilio. 1998. Structural modification and bioavailability of starch components as related to the extent of maillard reaction: An enzymatic degradation and solid-state 13C CPMAS NMR Study. J. Agric. Food Chem. 46:438–441.[Medline]
- Rendig, V.V., and F.E. Broadbent. 1979. Proteins and amino acids in grain of maize grown with various levels of applied N. Agron. J. 71:509–512.[Abstract/Free Full Text]
- Ritchie, S.W., and J.J. Hanway. 1982. How a corn plant develops. Spec. Rep. 48. Iowa State Univ. Coop. Ext. Serv., Ames.
- SAS Institute. 2000. SAS user's guide. SAS Inst., Cary, NC.
- Singh, V., and J.V. Graeber. 2005. Effect of corn hybrid variability and planting location on dry grind ethanol production. Trans. Am. Soc. Agric. Eng. 48:709–714.
- Singletary, G.W., and F.E. Below. 1989. Growth and composition of maize kernels cultured in vitro with varying supplies of carbon and nitrogen. Plant Physiol. 89:341–346.[Abstract/Free Full Text]
- Singletary, G.W., and F.E. Below. 1990. Nitrogen-induced changes in the growth and metabolism of developing maize kernels grown in vitro. Plant Physiol. 92:160–167.[Abstract/Free Full Text]
- Steele, J.L., R.A. Saul, and W.V. Hukill. 1969. Deterioration of shelled corn as measured by carbon dioxide production. Trans. Am. Soc. Agric. Eng. 33:685–689.
- Tsai, C.Y., D.M. Huber, and H.L. Warren. 1978. Relationship of the kernel sink for N to maize productivity. Crop Sci. 18:399–404.[Abstract/Free Full Text]
- Uribelarrea, M., F.E. Below, and S.P. Moose. 2004. Grain composition and productivity of maize hybrids derived from Illinois protein strains in response to variable nitrogen supply. Crop Sci. 44:1593–1600.[Abstract/Free Full Text]
- Weller, C.L., M.R. Paulsen, and M.P. Steinberg. 1988. Correlation of starch recovery with assorted quality factors of four corn hybrids. Cereal Chem. 65:392–397.
- Wyss, C.S., J.R. Czyzewicz, and F.E. Below. 1991. Source-sink control of grain composition in maize strains divergently selected for protein concentration. Crop Sci. 31:761–766.[Abstract/Free Full Text]
- Yang, W., M. Hattori, T. Kawaguchi, and K. Takahashi. 1998. Properties of starches conjugated with lysine and poly(lysine) by the Maillard reaction. J. Agric. Food Chem. 46:442–445.[Medline]
TOP
ABSTRACT
INTRODUCTION
