CASE STUDY: The Evaluation of a Vacuum Storage Method for High Moisture Distillers Grain and Its Effect as a Protein and Energy Supplement for Beef Cows

INTRODUCTION

High moisture wet distillers grain (DG) is an excellent feedstuff for inclusion into cattle diets. Because this feedstuff typically ranges in DM content between 30 and 60%, the length of storage time until feeding without spoilage is limited to 3 to 4 d in summer and to 1 week in winter (Garcia and Kalscheur, 2004). Conse- quently, cattle producers with smaller herd sizes and limited daily consump- tion of DG have had few choices for utilizing this feedstuff. Their options have included hauling smaller loads of DG from the plant to the farm to minimize spoilage, which ultimately results in higher transportation costs offsetting the lower purchase cost of DG, or ensiling the DG with another feedstuff such as dry corn silage or small grain straws. Blending DG with other feeds such as corn silage has resulted in a shift in fermentation patterns - from lactic acid fermenta- tion toward more production of acetic acid. Kalscheur et al. (2003) ensiled on an as-fed basis 50:50 and 75:25 corn silage to DG rations, respective- ly, and showed increased aerobic sta- bility upon exposure to air. They also reported increased feed stability at feed-out as one advantage of ensiling corn silage with DG. A practical ad- vantage of blending DG with fibrous residues such as corn silage is that the blend is easier to break and feed when frozen during the winter (Garcia and Kalschur, 2007) because of the larger particle size of the blend compared with the particle size of DG.

Kalscheur et al. (2002) evaluated the anaerobic fermentation characteristics of ensiling wet corn distillers grains alone or mixed with soy hulls (SH). Combining DG with SH resulted in immediate acidic conditions in the silage mass as a result of the initial low pH of the DG rather than from fermentation. Ensiling DG by itself has proven problematic in both upright and horizontal silos. With a modified bagger, DG can be stored in a silage bag, but the consistency of DG does not lend itself to all bagging equipment.

Among various low-cost approaches for ensiling DG is the vacuum system. Promma et al. (2000) used vacuum ensiling in polyethylene bags for small research samples and found that the system effectively prevented secondary fermentation. Achacoso et al. (1979) reported an advantage for ensiling corn silage via evacuation of air compared with conventional packing and sealing in steel drums. Vacuum ensiling systems of varying sizes have been used under farm conditions in Australia, Thailand and the United States (Moss et al, 2002). Moss et al. (2002) determined that a vacuum silo system may be a satisfactory method of ensiling direct chop alfalfa, especially during emergency situations, for short-term ensiling and for small herds. Vacuum preservation has not been evaluated for DG alone.

Distillers grain can serve as an excellent source of protein and fermentable fiber for ruminants (Davis, 2002). In general, there are 2 nutrition philosophies regarding the use of DG (Erickson and Klopfenstein, 2002). When fed at 15% or less of the diet DM, DG serves primarily as a source of supplemental protein. When fed at higher levels (greater than 15% of the diet DM), the primary role of DG is a source of energy. Numerous trials have been conducted feeding DG to growing and finishing feedlot cattle (Firkins et al., 1985; Ham et al., 1994; Perrin and Klopfenstein, 2001). Fewer studies have evaluated DG as a feedstuff for beef cows. Corners et al. (2002) reported DG supplementation increased ADG of beef heifers grazing fescue pasture. Kalscheur et al. (2004a) observed a decrease in milk production in dairy cows when DG was added to the diet above 10% of the DM. In contrast, Anderson et al. (2006) found that for dairy cows fed 20% DG, yields of milk, milk protein, and milk fat increased whereas DMI tended to decrease.

Several studies have ensiled DG with other feedstuffs such as corn silage, SH, wet beet pulp, and crop residues (Kalscheur et al, 2002, 2003, 2004b) for eventual use as a feedstuff. Ensiling DG by itself for use as a feedstuff for beef cows is a relatively unstudied concept. With smallerscale producers in mind, a study was designed to evaluate a vacuum ensiling method as a means to preserve wet distillers grains and to determine the subsequent feeding value of the preserved DG in corn silage diets for beef cows.

The purpose of this study was to evaluate one method of vacuum ensiling (preserving) DG for on-farm storage and utilization, and to determine the efficacy of vacuum-stored DG as a protein and energy supplement for late gestation and early lactation beef cows. The criteria evaluated included BCS changes of cows, BW changes of cows, and production performance, including lactation and reproductive performance.

MATERIALS AND METHODS

Animal Care and Use

This study was approved by the Illinois State University Animal Care and Use Committee, protocol no. 262003.

Preservation Method

On November 5 and 6 freshly produced DG (36.53% DM), still warm from the processing plant, was placed (using 2 skid-steer loaders) on a 0.24-mm thick white plastic silo cover in 2 pyramid-shaped rows containing either 49 or 63.6 metric tons (t). Each row was either 3.0 or 4.6 m wide at the base, 1.5 m at the peak, and 22.9 m long. Before covering each row with plastic and sealing the plastic at the base with ground limestone, a 5.08-cm diameter × 15.2-m-long perforated, flexible, plastic tube was laid on each side of each DG row and was connected to a 5 HP, 45.4 L, 120-V shop vacuum (Shop Vac, Williamsport, PA). Subsequent to sealing, each vacuum (2 vacuums per covered row) was turned on 3 times/d for a 2-to 5-min duration at 0800, 1200, and 1600 h each day until the covered row was opened. Each vacuum was rated to have a positive displacement of 5.4 m^sup 3^/min of air and would pull the plastic cover tight to the DG within each 2 to 5 min of operation, displacing any air trapped between the cover and the pile of DG.

Samples of DG were collected at 0, 41, and 78 d post-sealing and analyzed for DM, pH, and the following VFA: acetic acid, propionic acid, butyric acid, and lactic acid. Samples were obtained by cutting a small hole in each bag at 6 locations, inserting a 1.52-m grain probe and resealing the hole with duct tape following removal of the grain probe and sample.

Animals and Diets

Eighty Simmental × Angus multiparous crossbred beef cows in the third trimester were blocked into 3 BCS groups (mean BCS = 8.0 for 10 cows, 6.7 for 30 cows, and 5.9 for 40 cows) and randomly allotted within a block to 10 treatment pens. Each pen contained 8 cows. Individual cows having similar BCS were shifted among treatment pens within a block to equalize mean pen starting weights. Within a block, cows within pens were randomly assigned to receive 1 of 2 diets (5 pens/diet). Control cows (CNTL) were fed corn silage-shelled corn (SC)-soybean meal (SBM)based TMR that were balanced to meet NRC estimated daily nutrient requirements (NRC, 1996). Preserved DG were initially included in treatment cow (TRT) diets to balance for CP replacing SBM, but were increased as needed over a 179-d feeding period to meet energy and protein requirements, substituting for SBM and SC.

Initially, CNTL were fed a diet containing 78.9% corn silage, 8.8% SC, and 9.0% SBM, providing 45.6% DM, 11.0% CP, 2.9% ether extract (EE) and 17.1% ADF (DM basis); TRT were fed a diet containing 81.5% corn silage and 14.8% DG, providing 41.5% DM, 11.3% CP, 4.1% EE, and 18.1% ADF (DM basis). When 67% of the cows in a pen had calved, dietary adjustments were made in SBM and DG to supply estimated CP requirements NRC (1996) for the cows in that individual pen. This dietary adjustment resulted in CNTL receiving a diet containing 65.4% corn silage, 20.4% SC, and 12.9% SBM, providing 48.7% DM, 12.8% CP, 3.0% EE, and 24.2% ADF (DM basis); TRT received a diet containing 72.7% corn silage and 23.3% DG, providing 43.9% DM 14.0% CP, 5.1% EE, and 29.2% ADF (DM basis). In addition, all cows were fed to maintain similar BCS and BW changes. To maintain similar BW and BCS changes of the cows across cattle pens, periodic changes in daily allotments of corn silage and either SC or DG were made. Therefore, the data regarding the mean nutrient and ingredient compositions represents that of the diets fed during the trial. Once every 21 d, subsequent to calving, the feeding technician visually evaluated the BCS of each cow in each pen and averaged the scores to obtain a pen-average BCS. The objective was to maintain the average BCS for each pen of cows in a BCS range of 5 to 6. Accordingly, separate adjustments were made in the amounts of each dietary ingredient included in each pen's TMR. For the purpose of dietary adjustments, energy of DG was assumed to have a ME value similar to SC. Dietary adjustments in TRT and CNTL diets were not formulated to be isonitrogenous. Shelled corn and DG were increased based on their estimated ME values to maintain similar cow BCS. A Rumensin (Elanco, Greenfield, IN) premix (fed to supply 200 mg/d per cow of monensin) was added to the TRT and CNTL TMR in the same percentage. Either NaH^sub 2^PO^sub 4^*H2O or CaHPO^sub 4^ and CaCO^sub 3^ was added to each TMR to provide similar amounts of dietary Ca and P. All cows were fed once daily. All feed ingredients were mixed into a TMR in a horizontal batch mixer before feeding. In addition to the TMR, all cows were allowed free choice access to a commercial mineral mixture. Calves were offered creep feed beginning 60 d after the birth of the first calf. In addition, calves were allowed access to CNTL and TRT diets at the bunk with their dams.

Cows were housed in an open-front cattle shed and concrete confinement lots designed to house 8 cow-calf pairs per pen. Cows were fed TMR once daily at fence line bunks according to treatment.

Cows were individually weighed on 2 successive days and the BW were averaged to obtain beginning, AI, and ending BW (d 78, 176, and 257 of the study, respectively). Body condition scores of the cows were determined by visual appraisal at the beginning, at AI, and at the end of the trial and were obtained by averaging the condition scores of 2 technicians working independently. Scores within the BCS system ranged from 1 to 9, where 1 = emaciated and 9 = fat.

All cows were synchronized according to the 2-injection Lutalyse (Pfizer Animal Health, New York, NY) regimen with the first injection occurring 75 d after the birth of the first calf. Following AI at a predetermined time, the cows were commingled into 2 groups (CNTL and TRT), allowed access to 2 earth lots, and exposed to bulls for a 55-d natural breeding season. Following natural mating, the cow-calf pairs were re-allotted and confined to their respective treatment pens until the trial was terminated.

Calf performance during the trial was measured by recording birth weight, calf survival at birth and during the remaining portion of the feeding trial, and calf weight at the end of the trial. Milk production of the cows was estimated by the weighsuckle-weigh technique. Milk production was estimated for all cows at 2 separate times 20 d apart, beginning 50 d after the birth of the first calf. The procedure utilized for estimating milk production, which was modified from that of Neville (1962), was as follows: 1) separate calves from cows for 4 h; 2) allow calves to nurse cows for 30 min; 3) separate calves from cows for 12 h; 4) weigh calves; 5) allow calves to nurse cows for 30 min; 6) weigh calves and multiply change in calf weight by 2 to estimate 24-h milk production.

Experimental Procedures

Volatile fatty acid analyses were conducted according to procedures described by AOCS (2003b) method Cg 4-94. During the feeding trial, samples of DG were collected once weekly in duplicate. Samples of the TMR for each pen were collected once weekly. All DG and TMR samples were dried at 55°C, ground in a Thomas- Wiley mill (Thomas-Scientific, Philadelphia, PA) to pass through a 1-mm mesh screen, and were frozen at - 4°C for later analysis. Samples of DG and TMR were analyzed for DM, ADF, EE, and CP. Acid detergent fiber was determined according to procedures described by Robertson and Van Soest (1981). Crude protein was calculated from percentage of N as determined by dry combustion with a Leco FP-528 Nitrogen-Protein Determinator (Leco Corporation, St. Joseph, MI). Ethyl Ether extract was determined by method AM5-04 (AOCS, 2003a) in an Ankom XTlO Extraction System (Ankom Technology, Macedon, NY).

Statistical Analysis

Data including dietary components (DM, ADF, EE, and CP), were analyzed using Statistical Package for Social Sciences procedures (SPSS, version 13, 1998; SPSS Inc., Chicago, IL). Data were analyzed for a randomized complete block design with unequal replication using least squares ANOVA. Dietary treatment, pen, and their interactions were the independent variables, and cow consumption data and cow and cowcalf performance were the dependent variables. Pen was the experimental unit when analyzing feed consumption data; however, individual animal unit served as the experimental unit for cow and calf-performance data analysis. Statistical differences were identified using a protected F-test. The residual mean square was used as the error term. Production performance data of cows expressed as percentages were analyzed utilizing chi-square procedures (Steel and Torrie, 1960). Milk production and calf performance data were analyzed using calf age as a covariate. Mean and SD of VFA were calculated for d 0, 41 and 78 and subjected to paired f-test analyses to compare pre- and post-preservation concentrations using SPSS software. Significance was set at (P < 0.05).

RESULTS AND DISCUSSION

The DG used in this study contained the solubles fraction and was produced using a dry-grind procedure. Fresh DG at the time of delivery to the farm for preservation was at pH 4.6 and was found to contain 36.18 ± 0.48% DM, 32.03 ± 0.54% CP, 7.08 ± 0.6% EE, and 20.6 ± 3.60% ADF (Table 1). The nutrient content of DG used in this trial is similar to the mean values previously reported for DG of 29.7% CP, 19.7% ADF, and 10.0% fat (NRC, 2001). At the time of preservation (d 0) DG contained 0.45 ± 0.07 mg/g acetic acid, 0.09 ± 0.04 mg/g propionic acid, 0.05 ± 0.03 mg/g butyric acid and 1.01 ± 0.34 mg/g lactic acid. Over the first 78 d of storage, VFA concentration decreased (P < 0.05) 86.7, 66.7, and 60.0% for acidic acid, propionic acid, and butyric acid, respectively, and 95% for lactic acid. Volatile fatty acid concentration of the vacuum exhaust was not measured. The volume of VFA that were exhausted because of volatilization under vacuum compared with what was metabolized by yeast or fungi was not determined. It is possible that vacuum may have increased the amount of VFA volatilized and lost in the exhaust.

The data of this trial suggests the vacuum storage procedure used did not ensile the DG. Daily air temperatures were obtained from the Illinois State University Facilities Management Department for the time period of this trial. The mean air temperature during the first 41 d was 3.9 ± 10.0°C. The mean 24-h daily high temperature during the 41 d was 4.4°C. These ambient temperatures (Figure 1) should have been high enough to allow ensiling to occur and should not have been a limiting factor (Ensminger and Olentine, 1978). The low pH and warm relative temperature of the DG at the time of initial storage may have provided an unsuitable environment for ensiling to occur.

The time required to unload, move, cover, and seal each pile was 2 h. The cost of the plastic for each sealed pile was $180. The cost of the 2 vacuums and accessories required for each sealed pile was $200. The plastic for each bag was not reusable; therefore, the bag was a one-use expense. Each vacuum could be used again and the expected life for each vacuum is 5 yr. The total prorated equipment cost per sealed pile was $1.95/t DG.

Bag 1 was opened on d 78 (January 21) and no spoilage of DG was observed. Though no color chart comparison was made, the appearance (golden yellow color) of the DG at d78 was similar to the DG appearance at d 0. Approximately 200 kg of DG (5 kg/d per cow, 40 cows/d) were fed daily. Limited amounts of the DG were frozen on the perimeter of the pile within bag 1 when it was opened. Figure 2 shows the mean air temperature during the time period when bag 1 was fed. Beginning on d 148 (70 d after opening bag 1 = April 20) limited amounts of spoilage were observed. Spoiled DG was separated daily from the face of the open pile, weighed, and discarded. After acquiring each day's feed, bag 1 was resealed by rolling up the plastic and applying vacuum to the bag. Spoilage progressively increased subsequent to d 148. Bag 1 was emptied on d 190 (May 14). Of the 63.6 t stored in bag 1, 12.6 t (19.8%) were considered spoiled by the end of the first 112 d of feeding.

Bag 2 was opened on d 190 postsealing (May 14) and was fed for 67 d following the same procedure used for bag 1. No spoilage was observed in the DG when bag 2 was opened. However, limited amounts of spoilage were observed beginning the second day after opening and progressively increased after the first 30 d after opening. Feeding was terminated on July 20 because of spoilage of the remaining DG. Of the 49.1 t DG stored in bag 2, 27.6% (13.6 t) was considered spoiled and unfit for feeding. Mean ± SD weekly air temperatures during the 67 d feeding period for bag 2 are shown in Figure 3. The mean temperature during this time period was 21.9 ± 12.7°C.

The literature contains little data regarding storage losses of DG for which to compare the spoilage losses observed with vacuum preservation of DG. According to a summary of the literature (Jurgens and Bregendahl, 2007), average storage losses for corn silage under good management without additives (preservatives) for stacks is 10.8% with a range of 12.0 to 25.0% and for bunker silos is 15.0% with a range of 10 to 25%. Kalscheur et al. (2005) observed DM loss attributed to visual spoilage was greater (P < 0.01) in untreated (no preservative) DG (3.1%) than in DG treated with a preservative (1.3%) when stored outdoors in open boxes. Loy and Strohbehn (2007) reported total losses from transportation, storage, and feeding amounted to 16.7% for modified DG stored in a plastic silo bag, of which 5.0% were attributed to spoilage during storage. In the study reported here, no spoilage was observed for DG during storage until the vacuum bag was opened. Upon opening, larger losses (19.8 and 27.6%) were observed than reported for other methods of storing high moisture feedstuffs. Organic matter loss because of shrinkage during storage or feeding was not measured. Because no spoilage was visually observed in either bag when first opened, but considerable spoilage occurred subsequent to opening, the data suggests that spoilage was due to exposure to air and was not a function of length of storage time. Once the bag was opened, spoilage appeared to occur more rapidly as the daily high temperature increased above 15°C.

The feeding trial began on January 22 and continued for 179 d until July 20. Because of the removal or death of animals during the trial, data were analyzed by least squares procedures for unequal subclass numbers where appropriate. Two cows were removed from the trial and one cow died from causes not related to the diet.

Table 2 shows the analyzed nutrient composition of the CNTL and TRT diets. The TRT diet contained significantly greater CP and moisture contents than the CNTL diets because of the higher percentage of CP and moisture contents of the DG. The DG fed during the trial was analyzed (n = 26) to contain 37.1 ± 0.4% DM, 32.0 ± 0.5% CP, 7.1 ± 0.6% EE, and 20.6 ± 3.6% ADF. Distillers grains (DM basis) represented 21.9 ± 2.3% of the DMI for TRT cows and the combination of SC and SBM supple- mentation represented 25.3 ± 4.6% of CNTL cows DMI (Table 3). No differences (P > 0.05) were observed between CNTL and TRT regarding DMI or ADFI. No significant differ- ences in starting, AI, or ending cow BW or BCS were observed (Table 4). Also, mean production performance of the cows (Table 5) was similar between TRT and CNTL cows. No significant differences were observed in adjusted 205-d calf BW, percentage of calf crop saved at birth or weaned, cow conception rates, or estimated 24h milk production. The cows did lose BW and BCS from the beginning of the feeding trial until AI but the changes represented normal BW and BCS changes experienced by cows during this phase of the productive cycle (Walker et al. 1998, 2004).

IMPLICATIONS

This study found the vacuum storing procedure could preserve DG for prolonged periods, but the vacuum procedure used did not ensile DG. The vacuum storage method works well as long as the bag is sealed. Upon opening and exposure to air, limited visual spoilage can be expected as long as the air temperature is below 15°C. As the air temperature increases above 15°C, increased spoilage can be expected. Because spoilage can occur upon exposure to air, a recommendation is to limit storage amounts under one covered pile to an amount that can be fed within 30 d, especially during the warmer months. In this study, spoilage of the DG did not occur until the bag was opened for feeding. For a herd of 40 cows fed 2.7 to 4.6 kg (6 to 10 lb) of DG/d per cow, each covered pile should contain 4 to 6 t of DG. One shop vacuum will displace the air from several individual 4- to 6-t piles. This study suggests that the feeding value of vacuum stored DG as a protein and energy supplement is similar to the feeding value of SC and SBM supplementation for beef cows fed corn silage-based diets. Vacuum stored DG supplementation resulted in similar cow performance as did SC and SBM supplementation. Vacuum storage of DG can serve as a low-cost preservation method for beef cow-calf producers regardless of the size of the operation. This storage procedure allows smaller scale operations the opportunity to take advantage of pricing strategies for utilizing high moisture corn processing co-products generally limited to larger operations.

ACKNOWLEDGMENTS

This study was jointly funded by the Archer Daniels Midland Company, Decatur, IL; Illinois Council on Food and Agriculture Research, Urbana, IL; and Illinois State University, Normal, IL. The authors are grateful to the University of Missouri Columbia Experiment Station Chemical Laboratories and Thomas Mawhinney who conducted the VFA analyses.

© 2008 American Registry of Professional Animal Scientists Provided by ProQuest LLC. All Rights Reserved.