Fermentation characteristics and bromatological composition of sweet sorghum bagasse silages

Universidade Federal dos Vales do Jequitinhonha e Mucuri, Programa de Pós-Graduação em Biocombustíveis, Diamantina, Minas Gerais, Brasil. Universidade Federal dos Vales do Jequitinhonha e Mucuri, Programa de Pós-Graduação em Zootecnia, Unaí, Minas Gerais, Brasil. Universidade Federal de Minas Gerais, Escola Veterinária, Programa de Pós-Graduação em Zootecnia, Belo Horizonte, Minas Gerais, Brasil. *Endereço para correspondência: francoise@zootecnista.com.br


INTRODUCTION
Sweet sorghum (Shorghum bicolor L. Moench) is an important crop used as bioethanol source (ROHOWSKY et al., 2012).Due to its C 4 photosynthetic system, it presents fast dry matter accumulation, resulting in high biomass production (TURHOLLOW et al., 2010).Different sweet sorghum cultivars are available to be used as bioenergy resource, resulting in technology and conversion (MARTINS et al., 2017).However, there is concern about the disposal of the remaining bagasse after juice extraction for biofuel production.The supply of this bagasse for ruminants allows the association of agroenergy and livestock chains, minimizing environmental damages and undue discard of these products in nature (ABDALLA et al., 2008).As ensiling is a suitable method for forage conservation (NAEINI et al., 2014), it was performed with the intention of preserving this by-product for a longer period, providing potentially edible feed for ruminants.Thus, each cultivar provides bagasse with differentiated characteristics and variations in yield and microorganisms at the ensiling process, which directly interferes with the fermentation pattern and results in different silages qualities (OLIVEIRA et al., 2009).Therefore, this study aimed at evaluating dry matter recovery, gas losses, bromatological composition and fermentative characteristics of bagasse silages from different sweet sorghum cultivars.

MATERIALS AND METHODS
The cultivation of sorghum cultivars was conducted at the 2011/12 summer crop at Fazenda Rio Manso, an experimental unit of the Federal University of Jequitinhonha Seeding was performed in January and in May, 102 days later, harvest was carried out.The average height of the cultivars were 2.7 meters.The whole plants went through the electric mill for juice extraction and bioethanol production, generating bagasse.Maintaining the experimental arrangement of the treatments in their respective experimental units, the bagasse was milled for ensiling, added of its panicles, with an average particle size of 1 to 2 centimeters, using a stationary silage machine.The bagasse of the cultivars was ensiled in experimental silos composed of PVC tubes, with capacity of 2.5 kg, provided with Bunsen valve adapted to their lids, to allow fermentation gases to escape.The forage compaction was performed manually with a stick, obtaining an average specific mass of 600 kg/m 3 of natural matter.The silos were opened in September 2013.They were weighed again and the dry matter recovery, gas losses and the average particle size of the silages were evaluated, according to the methodology described by Jobim et al. (2007).The samples used for analysis of the fermentation characteristics, bromatological composition and average particle size were collected by separating the contaminated parts with molds in the upper part and contaminated with effluent in the lower part, using only the silage from the middle part of the silo, which was homogenized and sampled.For the determination of the bromatological composition, samples of approximately 300 g of each silo were collected, which were pre-dried in a forced-air oven at 55º for 72 hour, and the samples milled in Willey mill with a 1 mm sieve.The dry matter (DM) contents of bagasse and silage, crude protein (CP), neutral detergent insoluble fiber (NDF) and acid detergent insoluble fiber (ADF) were analyzed in the milled samples (DETMANN et al., 2012).At the same time, samples of 25 g of fresh silage from each silo were collected and 250 mL of distilled water were added.They remained in resting for 1h to read pH (CHERNEY & CHERNEY, 2003) and to determine electrical conductivity (KRAUS et al,. 1997).A fast reading water activity meter was used to analyze water activity (Aw), according to Greenhill (1964).To analyze the ammoniacal nitrogen, part of the fresh material was pressed in a hydraulic press for extraction of the silage juice, which was preserved with 0.036 N (0.32%) sulfuric acid and frozen for further determination of the ammoniacal nitrogen content through distillation with magnesium oxide (AOAC, 1995).Data were submitted to analysis of variance, using the statistical program GENES (Cruz, 2006), and the means were grouped and compared by the Scott and Knott test at 5% significance.

RESULTS AND DISCUSSION
It was observed that the silages of the varieties BR505; CMSXS629; CMSXS635; CMSXS639; CMSXS643; CMSXS646; V82391 and V82392 presented lower rate of total DM recovery and consequently higher gas losses (Table 1).For Pahlow et al. ( 2003), the higher the recovery percentage of DM, the greater the efficiency of microorganisms that are desirable for the fermentation, reducing the total losses of DM.Gas losses are unavoidable and difficult to reduce because they occur during the fermentation process of forage ensiling, and can be caused by the manifestation of bacteria of the genus Clostridium, which act on lactate or sugars, producing butyric acid and CO 2 (JOBIM et al., 2007).The overall mean of APS was 12.52 mm (P> 0.05), which is an ideal size for good quality fermentation (Table 1).In addition, small particles in the manufacture of silages, facilitate the compaction in the silage process and decrease physical losses during the silos opening.
Tabela 1. Mean values of dry matter recovery; gas losses and average particle size (APS) of sweet sorghum bagasse silages From the results presented in Table 2, it can be observed that there was no difference between the dry matter content (% DM) of sorghum bagasse, with overall mean of 32.80% (P>0.05).It can be justified by the juice extraction from the plant to produce ethanol.The dry matter content of the forage is one of the main factors in the silage process, since it will determine the type of fermentation that will develop inside the silo (OLIVEIRA et al., 2009).The MS content is considered satisfactory for a well-conserved silage (Table 1), with mean of 28.55% (P>0.05),close to that found by Vidya et al. (2016) that obtained content of 34.83%, and lower than that recorded by Kumari et al. (2011), who observed a content of 42.10% for DM in exclusive bagasse silages.The cultivars BR501; XBSW80147; BR505; CMSXS629, CMSXS633, CMSXS634; CMSXS635, CMSXS639, CMSXS643, CMSXS644, CMSXS648, BRS601, Sugargraze, V82391, V82392 and V82393, presented higher crude protein content: 7.14; 7.15; 6.93; 6.50; 6.51; 6.67; 6.98; 6.79; 6.67; 6.75; 6.64; 6.91; 6.82; 6.54; 6.62 and 6.69 respectively (P<0.05).These values are close to the ones found by Kumari et al. (2011), with mean of 7.27.The cultivar CMSXS642 presented lower neutral and acid detergent fiber content: 39.23 and 19.23%, respectively.The cultivar XBSW80147 had lower acid detergent fiber content.Ávila et al. ( 2013) reported levels of 63% NDF and 41.4% ADF in the bagasse silages of sorghum cultivars.There was no significant differences among the silages for pH values, with mean of 3.49 (P>0.05)(Table 3), which leads to acid silages, which may have helped to maintain an adjusted internal situation which may have helped to preserve an internal situation adjusted for maintenance of the nutrients until the silos opening.Acidity is an important point in the process of assessing the quality of silages.According to Evangelista et al. (2005), the silages are classified as excellent when they present pH lower than 4.6 for the range of 26 to 35% DM content.
The silages of the cultivars BR501 and BRS601 presented higher values of electrical conductivity (EC), 0.93 and 0.95 S / cm (P<0.05)respectively (Table 3), demonstrating that there was a greater cell lysis, and consequently, higher ions loss in effluents.The measurement of EC is the ability of water to conduct electric current and evaluates the intensity of the cellular rupture in the forage subjected to the cut and the consequent extravasation of ions to the solution (KRAUS et al., 1997).There was no significant difference in water activity (Aw) in this study (P> 0.05), resulting in mean value of 0.96 (Table 3).The highest value of Aw indicates that the action of microorganisms can be intensified, considering that the microbial digestion of the nutrients occurs in the extracellular environment.Water is an important vehicle for this hydrolysis process.According to Ditchfield (2000), the term water activity was created to determine water available for microbial growth and reactions that could deteriorate food.The value of Aw indicates the water level in its free form in the materials and is expressed in a scale from 0 (zero) to 1.0.Zero is considered for water-free materials and 1.0 for water in its liquid form.Consequently, the mean of 0.96 favors the microbial activity in these silages.A significant difference (P<0.05) was observed for ammoniacal nitrogen (N-NH 3 ) contents only in cultivars BR501; CMSXS629; CMSXS636; CMSXS648 and V82392 higher levels were observed: 4.20, 3.83, 4.15 and 4.36%, respectively.The silages of this study can be classified as of very good quality, according to Ribeiro et al. (2007), since they presented a relation-NH 3 NT lower than 10%.Sweet sorghum bagasse silages presented an adequate conservation of the material inside the silos.Minimum losses, which helped to maintain a good bromatological composition and stable fermentative characteristics, were recorded until the silos opening.

To
Coordination for the Improvement of Higher Education Personnel (CAPES), to National Council for Scientific and Technological Development ( CNPq), to Embrapa Maize and Sorghum and University of the Jequitinhonha and Mucuri Valleys (UFVJM).
Means in the same column followed by different lowercase letters differ from each other within each factor by the Scott and Knott test, at 5% probability.

Table 2 .
Mean values of: Dry matter (DM) of the bagasse and silage; Means in the same column, followed by different lowercase letters, differ among each other within each factor by the Scott and Knott test at the 5% probability.

Table 3 .
Mean values of pH, eletrical conductivity (EC); wateractivity (Aw) and Ammoniacal nitrogen (%N-NH3/N-Total) of sweet sorghum bagasse silages Means in the same column, followed by different lowercase letters, differ from each other within each factor by the Scott and Knott test at 5% probability.