Fig. 1. The Piedmont harvest system. A whole-stalk harvester cuts sorghum and deposits it in windrows, and a field loader dumps the stalks onto a trailer for transport to the field processing site.
Last update September 12, 1997 aw
Both herbaceous and woody crops represent possible sources of fiber for conversion to ethanol. Herbaceous crops harvested as hay are generally field-dried and stored outside in round bales. Woody crops have the advantage of a relatively long harvest season (40 weeks annually) which reduces storage requirements. They too are dried naturally and stored in open-air storage facilities.
Sweet sorghum can produce large quantities of both readily fermentable carbohydrate, and fiber for conversion via enzymatic hydrolysis, per unit land area. In fact, on the average, sweet sorghum produces more carbohydrate per unit land area than maize in the drought-prone southeastern Piedmont of the United States (Parrish et al. 1985). Unlike maize, sweet sorghum does not concentrate carbohydrates in grain, but stores them in the stalk. Many tons of high moisture content material must be handled to collect the fermentable constituent, and equipment and transportation costs are directly related to tonnage handled. Also, the harvest season is short, only 4 to 6 weeks. The challenge is to harvest the crop, separate it into juice and fiber, and utilize each constituent for year-round production of ethanol.
This study determines the cost, expressed in $/liter of ethanol, to deliver sweet sorghum to an ethanol production facility in the Piedmont. Options exist for each step of the process--harvesting and field processing, use of rind-leaf fraction, and final processing to ethanol--and costs vary according to option.
In developing a concept for a sweet sorghum-for-ethanol industry in the Piedmont, an attempt was made to encourage the involvement of a large number of growers with varying production units, perhaps as small as 10 hectares. It is hypothesized that a centralized ethanol production plant will buy whole-stalk sorghum standing in the field and will be responsible for harvesting, processing, and transporting the crop. The grower will provide bunk silo space to store the fiber constituent. The central plant could own the necessary harvesting equipment, or perhaps will contract with a harvesting company. In either case, farmers will not be required to own harvesting equipment, used only a small fraction of the year, to harvest their crop.
The following is envisioned for the forage chopper option: whole-stalk sorghum is chopped with a conventional forage chopper, blown into forage wagons or trucks, and transported to a truck-mounted screw press parked inside a bunk silo. Chopped sorghum is passed through the press to express the juice, and the residue is conveyed immediately into the silo. Juice is collected in a storage tank which is emptied periodically (twice daily, or more if indicated) by a tanker truck. At the ethanol production plant, juice is fermented directly. This system, hereafter referred to as the "forage chopper" system, could be implemented today with existing technology. Performance parameters for the various pieces of equipment in the forage chopper system are known, except for the screw press.
Research conducted by Cundiff and Worley (1991) and Crandell et al. (1989) indicates that a forage chopper could be modified to collect sweet sorghum pith and drop rind-leaf back on the field. This machine, referred to as a "pith combine", is envisioned as an assembly of the following subsystems: a forage chopper pickup mechanism, slightly modified forage chopper feed rolls and chopper assembly, a set of straw walkers mounted behind the chopper, and a conveyor to load pith into a forage wagon. For this study, a "pith combine" system is defined by replacing the forage chopper in the forage chopper system with a pith combine. All features of the forage chopper and pith combine systems are identical except the pith combine leaves a rind-leaf fraction equal to 10% of the whole-stalk mass on the field.
A whole-stalk harvester is being developed for the Piedmont (Rains et al. 1990). A system, hereafter referred to as the Piedmont system (Fig. 1), is expected to operate as follows: The whole-stalk harvester cuts stalks and deposits them in windrows. A field loader dumps stalks onto trailers for transporting and stockpiling at a processing site adjacent to a bunk silo. At some later time, perhaps after 30 to 60 days storage, stalks are loaded into a processor consisting of feeder, chopper, and pith separator. The processor is mounted on a flat-bed trailer for transport from farm to farm. Stalks are fed into the chopper/separator, which operates like the pith combine, except that it separates out a rind-leaf fraction equal to 30% of the whole-stalk mass. After passage through the screw press to capture juice, pith presscake is recombined with the rind-leaf fraction and conveyed into the bunk silo. Resulting silage is identified as "combination" silage to differentiate it from silage produced with the pith combine system, which does not include the rind-leaf fraction. Juice produced by the Piedmont system is handled in the same manner as the other two systems.
The forage chopper system has one key disadvantage compared to the pith combine or Piedmont system. Passing chopped whole stalk through the press reduces press capacity and juice yield. Little sugar is contained in the fibrous leaf and rind, but it absorbs juice, thus reducing the total juice that can be expressed. Because juice expression is a relatively expensive processing step, it is important to investigate options which maximize juice yield per hour of press operating time.
Using the assumption that sugar content of juice remaining in the presscake (fiber that exits the screw press), is equal to sugar content of the expressed juice, Cundiff (1991) presents a procedure to calculate the sugar collected in the juice for several whole-stalk fractionation options. Using the expression ratio (defined as juice mass divided by input mass to press) and the juice Brix, quantity of sugar collected in the juice can be calculated for each system. Crandell et al. (1989) conducted a single experiment for a 65 to 75% pith fraction and found that the expression ratio was 0.625 and 0.60, respectively. Cundiff and Rains (1991) later completed a replicated experiment and found that the expression ratio for chopped whole stalks was 0.36, and for a 90% pith fraction it was 0.46. Based on these limited data, the following expression ratios were assumed for the material produced by the three systems: forage chopper (0.35), pith combine (0.45), and Piedmont (0.55).
Stripping away rind-leaf mass equal to 10% of the whole stalks mass (pith combine system) increases the sugar yield per Mg of input to the press by 29%. When a 30% rind-leaf fraction is eliminated (Piedmont system), sugar yield per Mg input is increased by 57%. The increase in whole-stalk sugar captured in the juice is 16% for the pith combine system and 10% for the Piedmont system. If the press has the same capacity (Mg/h) for the pith fractions as the chopped whole stalk, then press operating cost per unit of sugar captured in the juice is minimized for the 70% pith fraction. The 90% pith fraction represents a compromise choice which maximizes whole stalks sugar yield in the juice, and still achieves a 29% increase in screw press performance.
The forage chopper system requires the lowest equipment investment and the Piedmont system requires the greatest. A key question is, does increased yield of juice fermentables pay for the additional equipment investment?
The Piedmont system has one key advantage over the forage chopper and pith combine systems--it allows for whole-stalk storage. Without whole-stalk storage, sorghum must be processed (juice expressed and residue ensiled) as it is harvested. Therefore, harvest and juice expression operations are tied together--if one machine breaks down, the entire system is delayed. Whole-stalk storage allows the harvest season to be extended at least 30 days, and perhaps up to 60 days, without significant degradation of sugars. Extension of harvest season results in cost savings at an ethanol production plant by reducing peak capacity requirement. If the harvest season can be extended from 8 to 12 weeks, processing equipment capacity can be reduced by one-third.
Worley and Cundiff (1991) developed a systems model of sweet sorghum harvesting, and estimated costs for harvesting and juice expression via forage chopper, pith combine, and Piedmont systems. Costs were $13.25/Mg whole stalk (forage chopper), $9.05 (pith combine), and $16.20 (Piedmont). Assuming a 40 Mg whole stalk/ha yield, costs were $530/ha (forage chopper), $362 (pith combine), and $648 (Piedmont).
Potential ethanol yield from the fiber is more difficult to predict. Emerging enzymatic hydrolysis technology has not been proven on a commercial scale. Since sweet sorghum is not currently a commercial crop, and is not expected to compete with maize as an ethanol feedstock until fiber conversion is a commercial option, it is appropriate to use the projected conversion efficiency for the mid-1990s, i.e., 420 liter/dry Mg (Norman Hinman, Manger, Biofuels Program, Solar Energy Research Institute, 1617 Cole Boulevard, Golden, CO 80401-3393), which equates to 147 liter/Mg silage at 35% dry matter. Based on an average yield of 40 Mg whole stalks/ha, the total potential ethanol yield per hectare from both juice and fiber is 5,025 (forage chopper), 4385 (pith combine), and 4790 (Piedmont). These results were based on expected silage yields of 0.617 Mg silage/Mg whole stalks (forage chopper), 0.47 (pith combine), and 0.553 (Piedmont). Juice yields were 332 liter/Mg whole stalks (forage chopper), 385 (pith combine), and 365 (Piedmont).
Road networks in the Piedmont are such that, within a 15-km radius, estimated average trucking distance to bring feedstock to a central facility is about 15 km. Energy in the diesel fuel to truck juice 15 km is equivalent to 1.5% of the energy in the ethanol produced. With silage, the transport energy is 1% of the energy in the ethanol. For comparison, the average energy to move petroleum from the wellhead to a retail outlet is 4% of the energy in the petroleum.
Trucking cost was determined by contacting companies engaged in trucking operations similar to the needed operations. A molasses hauler reported a cost of $0.93/km for short deliveries. (All trucking costs are presented as a per-km charge for travel out empty and return loaded). A logging contractor reported a trucking cost of $0.78/km, and a feed mill reported $0.75/km. Cost to haul silage was taken to be $0.78/km and cost to haul juice was taken to be $0.93/km.
Potential return at $44/dry Mg is $499/ha. A fiber conversion plant would have to pay $79/dry Mg for a grower to break even on sweet sorghum. These costs are shown in Table 1 on a per-liter-of-expected-ethanol-yield basis for comparison with the other systems.
The pith combine and Piedmont system options were developed in an attempt to reduce the cost of fermentables collected in the juice. Using the assumption of 420 liter/dry Mg for fiber conversion of the silage, the yield from the juice is a minor part of the total yield per ton of whole stalks, 28% (forage chopper), 37% (pith combine), and 32% (Piedmont). Differences between these systems and the forage chopper system are obscured by the silage yield. The analysis does show a slightly lower cost for the pith combine, suggesting that return of a 10% rind-leaf fraction to the field as a contribution to sustainable agriculture may be a viable option. The Piedmont system offers the advantage of whole-stalk storage, and subsequent extension of the harvest season from 8 to 12 weeks. Since this system only has a cost 14.4% higher than the forage chopper system, it also merits further study.
Net feedstock cost in a maize wet milling plant typically ranges 20 to 25% of the total ethanol production cost (Fuel Ethanol Cost-Effectiveness Study 1987). Remaining costs (all other than feedstock) range from $0.159 to $0.359/liter, and if costs at a fiber conversion plant are similar, total cost of producing fuel ethanol from sweet sorghum will range from $0.200 (feedstock) + $0.159 (conversion) = $0.359 (total) to $0.246 (feedstock) + $0.359 (conversion) = $0.605 (total). Current selling price of ethanol as a fuel additive is $0.30 to $0.35 per liter (Oxyfuel News 1992); consequently, the market must change before sweet sorghum can be competitive.
There is so little difference in the projected cost for the three harvest systems, none should be excluded from further study at this time. Should the market change, and a central plant be built to operate on sweet sorghum, it is probable that a mix of all three harvesting systems would be used.
Two forage chopper options were considered, harvesting the crop for silage only, and harvesting for juice expression and silage. Feedstock cost, computed up to the point conversion begins, ranged from $79 to $90/dry Mg. If the objective is simply the delivery of a Mg of fiber for a conversion process, it does not appear that a high moisture crop like sweet sorghum, which must be stored by ensiling, can be delivered at a cost competitive with high-yielding perennial grasses which are harvested and stored like hay. Ensiling does provide an opportunity for biochemical modification of the fiber during storage, and this advantage may increase the competitiveness of an ensiled crop.
| Cost ($/liter expected ethanol yield) | ||||
| Forage chopper | ||||
| Operation | Fiber only | Juice+fiber | Pith combine | Piedmont system |
| Field production | 0.099 | 0.073 | 0.083 | 0.076 |
| Harvest/field processing | 0.066 | 0.105 | 0.083 | 0.135 |
| Storage | 0.038 | 0.028 | 0.024 | 0.026 |
| Transportation | 0.008 | 0.009 | 0.010 | 0.009 |
| Total | 0.211 | 0.215 | 0.200 | 0.246 |