Fig. 1 to 3 summarizes and compares average ethanol yields from sorghum grain, sweet juice and biomass. Ethanol yields vary according to variety, geography, soil fertility and temperature.

Sweet sorghums usually yield from 50 up to 120 tons of stalks after the first cut. This feedstock contains 73% moisture, 13% soluble sugars, 5.3% cellulose, 3.7% hemicelluloses and 2.7% lignin. The stalks yield up to 70% sweet juice and 15.33 ton/ha of spent bagasse (Almodares & Hadi, 2009; Prasad et al., 2007).

Water added during extraction is considered part of he sweet juice yield (Fig. 1) and the sweet juice commonly contains around 14% soluble sugars. This substrate allows the production of 3,450 L ethanol/ha with a fermentation efficiency of 95%, similar to the result reported Kim & Day (2011) (3,296 L/ha). These last researchers did not consider losses that negatively affect fermentation efficiencies. Almodares & Hadi (2009), on the other hand, reported a yield of 3,000 L ethanol/ha directly when processing juice extracted from varieties that yielded from 39 to 128 ton stalks/ha. Although Wu et al. (2010b) did not report ethanol yields per hectare, the calculated ethanol production from the amount of total fermentable sugars extracted from a high yielding M81E cultivar planted at two different locations and bioconverted with a 95% of fermentation efficiency was in the range of 4,750 to 5,220 L/ha. These potential ethanol yields are equivalent to the bioconversion of 12 to 13 tons of maize kernels.

Experimental data obtained from sweet sorghums cultivated in Central Mexico indicated that these materials are capable of yielding 6.38 tons of sugar/ha/ cut. Consequently, when are adequately bioconverted have the potential of producing 4,132 L ethanol (unpublished data). Regarding to the lignocellulosic fraction, if 15.33 ton of bagasse/ha is obtained containing 29% cellulose and hemicellulose and 5.4% of remaining unextracted soluble sugars, up to 2,400 L of ethanol can be obtained (Fig. 3). This yield represents almost half of the 4,058 L/ha described by Kim & Day (2011) as theoretical ethanol.

In central Mexico, 42.5 ton of bagasse/ha with 50% fermentable sugars are commonly obtained. This biomass is capable of yielding 6,375 L ethanol with perfect conversion efficiency. However, experimental data where the acid-treated biomass was fermented with Issatchenkia orientalis indicated only 60% fermentation efficiency (3,865 L/ha) (unpublished
data). These results indicate that there are still many areas for potential improvements especially when processing spent biomass.

Almodares & Hadi (2009) reported that a yield up to 2 ton of grain/ha can be expected from sweet sorghum. If this material is milled, hydrolyzed and fermented, a final ethanol yield of 780 L can be expected. Nevertheless, the sweet sorghum grain during optimum harvesting is not fully matured and generally collected along the vegetative parts of the plant. Thus, the immature sweet sorghum kernels are usually processed with the bagasse and not fermented using grain technologies.

The biomass production per cultivated surface (Fig. 3) is the key and most important factor that affects ethanol yields indicating the importance of both plant breeding for the generation of new improved cultivars and the agronomic conditions mainly affected by soil fertility and water availability. The new biomass cultivars should adapt to marginal lands in order to minimize competition with basic grain production. The potential to obtain ethanol yields of 6630, 7000 and 10000 L/ha (with 95% of extraction and fermentation efficiency) can be achieved because yields of 50 to 120 tons of biomass/ha are reported. Comparatively Kim & Day (2011) indicated that the theoretical yield of maize kernels can be as high as 5,100 L/ha and up to 8,625 L/ha when the whole plant is bioconverted into ethanol (grain + corn stover). One of the most important factors to be addressed during yield calculation is indeed the energy required for ethanol production. Biomass and starch require more energy for hydrolysis compared to sweet sorghum juice. The technologies for starchy kernels and sweet juice are matured but the conversion and estimation of energy balances when processing lignocellulosic material will be critically important for the evaluation of economic advisability.