Kevin J. Shinners

HARVEST AND STORAGE OF TWO PERENNIAL GRASSES AS BIOMASS FEEDSTOCKS

Some perennial grasses, such as reed canarygrass (RCG) and switchgrass (SWG), have prolific yield and low inputs, making them attractive as biomass feedstocks. When harvested as biomass, these grasses are more mature and have much greater yield than when harvested as animal forage. Much is unknown about how harvest equipment performance and storage characteristics are affected by these crop conditions. The objective of this research was to determine the crop yield and drying rate, baling rate, bale density, and bale storage characteristics of these grasses harvested as biomass feedstocks. After the establishment year, the three-year average yield of RCG was 21% less than SWG (7.70 vs. 9.69 Mg DM ha -1 ) using a single-cutting system that occurred in August. When the crops were left standing over winter and harvested in the spring, DM yields were reduced by 17% and 26% for SWG and RCG, respectively. When crop yield was similar, switchgrass tended to dry faster than reed canarygrass. Drying rates of these grasses were faster than typically experienced with forage crops like alfalfa. Bale density averaged 163 kg DM m -3 with no significant differences between crops or type of wrap (twine or net). Dry bales stored outdoors for 9 to 11 months averaged 3.8%, 4.8%, 7.5%, 8.7%, and 14.9% DM loss for bales wrapped with plastic film, breathable film, net wrap, plastic twine, and sisal twine, respectively. Bales stored under cover averaged 3.0% DM loss. The chemical and physical properties of bales stored outdoors were spatially variable. Preservation by ensiling in a tube produced average DM losses of 1.1% at average moisture of 39.9% (w.b.).

Optimizing on-farm pretreatment of perennial grasses for fuel ethanol production.

Switchgrass (Panicum virgatum L.) and reed canarygrass (Phalaris arundinacea L.) were pretreated under ambient temperature and pressure with sulfuric acid and calcium hydroxide in separate experiments. Chemical loadings from 0 to 100g (kg DM)(-1) and durations of anaerobic storage from 0 to 180days were investigated by way of a central composite design at two moisture contents (40% or 60% w.b.). Pretreated and untreated samples were fermented to ethanol by Saccharomyces cerevisiae D5A in the presence of a commercially available cellulase (Celluclast 1.5L) and beta-glucosidase (Novozyme 188). Xylose levels were also measured following fermentation because xylose is not metabolized by S. cerevisiae. After sulfuric acid pretreatment and anaerobic storage, conversion of cell wall glucose to ethanol for reed canarygrass ranged from 22% to 83% whereas switchgrass conversions ranged from 16% to 46%. Pretreatment duration had a positive effect on conversion but was mitigated with increased chemical loadings. Conversions after calcium hydroxide pretreatment and anaerobic storage ranged from 21% to 55% and 18% to 54% for reed canarygrass and switchgrass, respectively. The efficacy of lime pretreatment was found to be highly dependent on moisture content. Moreover, pretreatment duration was only found to be significant for reed canarygrass. Although significant levels of acetate and lactate were observed in the biomass after storage, S. cerevisiae was not found to be inhibited at a 10% solids loading.

Stage of maturity, processing, and hybrid effects on ruminal in situ disappearance of whole-plant corn silage

Five in situ trials with whole-plant corn silage (WPCS) were conducted in two ruminally-cannulated Holstein cows to determine 24-h ruminal disappearance of dry matter (DM), starch, and neutral detergent fiber (NDF). In Trial 1, the effect of maturity of WPCS on ruminal nutrient disappearance was evaluated. Treatments were early dent (ED), 1/4 milk-line (1/4 ML), 2/3 milk-line (2/3 ML), and black-layer (BL) stage of maturity. Ruminal disappearance of DM was lower (p<0.01) for BL (474 g/kg) than ED (547 g/kg), 1/4 ML (579 g/kg), or 2/3 ML (530 g/kg). Ruminal starch disappearance was lower (p<0.01) for BL (862 g/kg) than ED, 1/4 ML, or 2/3 ML which averaged 950 g/kg. In Trial 2, the effect of mechanical processing of mature and immature WPCS and stover silage at harvest was evaluated. Processing increased (p<0.01) ruminal starch disappearance for both immature (844 vs. 664 g/kg) and mature (790 vs. 525 g/kg) WPCS. In Trial 3, two WPCS hybrids (grain vs. leafy) at two plant populations (low=59,000 or high=79,000 plants/ha) were evaluated. Type of hybrid or plant population did not affect DM or NDF disappearance. However, starch disappearance was higher (p<0.01) for the leafy hybrid (872 vs. 731 g/kg). In Trial 4, brown-midrib (BMR) corn harvested as WPCS was evaluated for ruminal nutrient disappearance versus a conventional grain hybrid. Ruminal DM (602 vs. 561 g/kg) and NDF (326 vs. 220 g/kg) disappearance were higher (p<0.01) for BMR. In Trial 5, ruminal nutrient disappearance of high- and low-NDF corn silage hybrids were compared. Ruminal DM (662 vs. 620 g/kg) and starch (987 vs. 950 g/kg) disappearance were higher, but NDF disappearance was lower (176 vs. 225 g/kg) for the low-NDF hybrid (p<0.001). In summary, delaying silage harvest to BL reduced ruminal nutrient disappearance. Mechanical processing increased ruminal disappearance of WPCS, primarily through greater ruminal starch disappearance. Ruminal starch and NDF disappearance were higher for leafy and BMR hybrids, respectively, relative to a conventional grain hybrid. Ruminal DM disappearance was increased for a low-NDF hybrid, but effects were positive for starch and negative for NDF disappearance.

Single-pass, split-stream harvest of corn grain and stover.

A grain combine was equipped with a whole-plant corn head and modified to produce single-pass, whole-plant corn harvesting with two crop streams: grain and stover. Capture of potential stover DM varied from 48% to 89% for leaves, from 49% to 92% for stalks, and was greater than 90% for husks and cobs, depending on corn head height. Stover aggregate moisture varied between 36% and 50% (w.b.), and area capacity ranged between 1.6 and 2.6 ha h-1, depending on corn head height. Whole-plant harvesting reduced area capacity by nearly 50% compared to harvesting with a conventional ear-snapping head. Single-pass stover had an average particle size of 69 mm and bulk densities of 51 and 110 kg DM m-3 in the wagon and bag silo, respectively. Estimated ethanol yield ranged between 2600 and 3945 L ha-1, depending on corn head height. Fermentation of single-pass stover in a bag silo was adequate, with average losses of 6% of total DM.

Single-pass harvest of corn grain and stover: performance of three harvester configurations.

Corn stover was harvested with a modified combine that simultaneously harvested grain and stover in separate streams. The harvester was used to collect the following stover fractions using three different heads: cob and husk (ear-snapper head); stalk and leaves (stalk-gathering head); and stalks, leaves, husk, and cob (whole-plant head). Cob and husk were also collected when using the stalk-gathering head, but in a separate stream from the stalks and leaves. Material harvested with the ear-snapper, whole-plant, or stalk-gathering head had average moisture of 38.2%, 45.0%, and 46.7% (w.b.); particle size of 14, 22, and 90 mm; and density in the transport container of 98, 64, and 40 kg DM m-3, respectively. Area productivity was 3.4, 1.5, and 1.9 ha h-1; fraction of available stover DM actually harvested was 18%, 64%, and 49%; and total harvester specific fuel use was 1.46, 2.07, and 1.83 L Mg-1 DM or 17.0, 33.4, and 27.4 L ha-1 for the ear-snapper, whole-plant, and stalk-gathering head configurations, respectively. The untilled ground cover in the fall and spring was greater than the minimum requirement of 30% when using any of the three heads. Chisel plowing in the fall with twisted shovels buried too much residue no matter which harvester configuration was used. Chisel plowing in the spring with sweeps left sufficient residue when stover was harvested with either the ear-snapper or stalk-gathering heads. Material harvested with the ear-snapper, stalk-gathering, and whole-plant heads had an average density in a bag silo of 261, 111, and 160 kg DM m-3, respectively. Average loss in a bag silo was less than 4.3% of total stover DM after nine months of storage. Based on estimated cellulose and hemicellulose content, ethanol yield was 868, 1474, and 2804 L ha-1 from materials harvested with the ear-snapper, stalk-gathering, and whole-plant heads, respectively.

LEACHATE CONDUCTIVITY AS AN INDEX FOR QUANTIFYING LEVEL OF FORAGE CONDITIONING

A method based upon measuring the conductivity of the leachate (LC) from mechanically conditioned forage was evaluated and used as an index for quantifying the extent of mechanical damage caused by various mechanical treatments. Relative to a Surface Area Index (SAI) method previously used, the LC method was simple, fast, and could be completed with readily available laboratory equipment (orbital shaker table, Waring® blender, and conductivity meter). The LC method differentiated various mechanical conditioning treatments more often and with greater sensitivity than the SAI method. The consistency of different Waring blender treatments was assessed to determine if it could be used as a standard treatment for normalizing leachate conductivity values.

An overview of physical treatments of lucerne performed at Madison, Wisconsin, for improving properties

Macerated (shredded) forage has been shown to have two important properties: drying rates several times those of conventional forage and dry matter digestibilities 10–15% higher than conventional forage. A field-drying process was developed which consists of mowing, macerating, pressing the forage into thin cohesive mats, and placing these on the stubble. A small prototype machine was constructed to carry out these processes concurrently. Equipment was also developed to harvest the forage mats either for dry hay or for silage. Ensiled macerated lucerne fermented twice as fast as conventional lucerne. Feeding trials with sheep, goats, and dairy cattle verified the higher digestibility of macerated as compared with unmacerated lucerne. Heat treatment of hay increased the “escape” or “bypass” protein of lucerne, which is expected to improve efficiency of protein utilization by ruminants. The increase in escape protein due to heat treatment was approximately 50% greater for macerated lucerne than for conventional lucerne.

Pilot-scale on-farm pretreatment of perennial grasses with dilute acid and alkali for fuel ethanol production.

Switchgrass (Panicum virgatum L.) and reed canarygrass (Phalaris arundinacea L.) were pretreated during anaerobic wet storage by adding sulfuric acid or calcium hydroxide (50 g kg-1 DM). Experiments were conducted at both laboratory scale (250 g DM) and pilot scale (250 kg DM) for either 60 or 180 days. Pretreated and untreated samples were fermented to ethanol by Saccharomyces cerevisiae in the presence of a commercially available cellulase (Celluclast 1.5L) and s-glucosidase (Novozyme 188). With acid pretreatment, conversion of cellulose to ethanol was 35 and 12 percentage units higher than for untreated controls for reed canarygrass and switchgrass, respectively. Similarly, lime-pretreated reed canarygrass and switchgrass out-yielded controls by 3 and 13 percentage units, respectively. Cellulose and lignin contents were largely unaffected by pretreatment and anaerobic storage, but hemicelluloses were lower for both pretreatments.

Drying Rates of Macerated Alfalfa Mats

ABSTRACT ALFALFA was macerated and then formed into thin mats which were deposited onto the forage stubble for field drying. Drying rates of mats and control swaths having equal dry matter per unit area were determined. The macerated alfalfa mats dried to 20% (w.b.) moisture content in 6 h or less, while the control swaths did not reach this moisture content during the first day. Mat drying rate is roughly proportional to mat dry matter per unit area. The evaporation rate of water from a standard evaporation pan poorly models the moisture evaporation rate from mats or swaths.

PROCESSING WHOLE-PLANT CORN SILAGE WITH CROP PROCESSING ROLLS ON A PULL-TYPE FORAGE HARVESTER

Whole-plant corn silage (WPCS) was harvested with a pull-type forage harvester equipped with crop processing rolls. Variables considered were crop maturity, theoretical length-of-cut (TLC), processor roll speed difference and roll clearance. Whole-plant particle size, whole cob and coarse fiber fractions, level of kernel breakage, and kernel fraction particle-size were used to quantify crop physical properties. Compared to unprocessed WPCS cut at 9.5 mm TLC, WPCS cut at 19 mm TLC and then processed had greater whole-plant particle size, less whole cob fraction, fewer unbroken kernels, smaller kernel fraction particle size, and greater estimated surface area of the kernel fraction. The processed treatments at 19 mm TLC and 3 or 5 mm roll clearance required similar specific energy and produced similar harvesting rates compared to the control unprocessed treatment at 9.5 mm TLC. Feeding dairy cattle WPCS (~33% of DMI) increased fat-corrected milk yield by about 4% and fat yield by about 2%. Compared to unprocessed WPCS cut at 9.5 mm TLC, WPCS cut at 19 mm TLC and processed with 1 mm roll clearance increased in situ DM disappearance by seven percentage units. Based on crop physical properties, harvester energy requirements and dairy cattle lactation performance, the recommended settings when harvesting WPCS with an on-board crop processor would be 19 mm TLC and 1 to 3 mm roll clearance.