1 Introduction

An inherent characteristic of effluent treatment stations (ETSs), especially biological stations, is the production of sludge, the final disposal of which poses a major challenge, especially since the creation of Law 12305/10 (Brazil 2010), which banned its disposal in landfills. Due to significant contents of nitrogen and phosphorus, this residue can be used in agricultural areas as a soil conditioner or fertilizer, after receiving appropriate processing (Melo et al. 2001). The use of biosolids in agriculture can be an economic alternative to farmers, since it can totally or partially replace the use of inorganic fertilizers, besides being a more sustainable alternative for final disposal (Behling 2009).

Owing to its high levels of nitrogen, biosolid is a good alternative fertilizer for agricultural crops, including forage crops. The use of nitrogen fertilizers has been one of the limiting factors in producing and improving the economic performance of forage crops (Vitor 2006). The use of nitrogen fertilizers has positive effects in terms of both forage production and nutritional value (Andrade et al. 2003). Nabinger and Medeiros (1995) report that the nitrogen available determines the growth and development of a plant, with higher nitrogen availability corresponding to faster formation of axillary buds and, consequently, tillering of the respective buds, until the leaf area reaches a certain critical value, thereby changing the amount of light that reaches the late buds.

Among the several forages planted in Brazil, elephant grass—Cenchrus purpureus (Schumach.) Morrone—stands out because it is a tropical species that has been widely used in animal feed for meat and milk production, is present in all regions of Brazil, and exhibits easy adaptation, good production levels, and easy establishment (Vitor 2006; Magalhães et al. 2009). This forage originated from Subtropical Africa, and arrived in Brazil in the mid-1920s, being cultivated in all regions of the country (Oliveira et al. 2017; Quéno et al. 2011).

In addition to its use in animal feeding, elephant grass has been widely used as biomass for energy purposes. The biomass of elephant grass has potential for the production of biofuel, alcohol, and charcoal and can be burnt directly for methane production (Anderson et al. 2008; Jakob et al. 2009; Lee et al. 2010; Lima et al. 2011; Morais et al. 2012; Samson et al. 2005; Smeets et al. 2009; Strezov et al. 2008). This forage is a promising alternative for energy production because of its great potential for biomass production (Derezs et al. 2006). According to Samson et al. (2005) and Morais et al. (2011), elephant grass is one of the main forage species used for this purpose, as a result of its high production capacity, low input requirements in its production process, and good biomass quality.

Thus, the objective of this work was to analyze and compare the production parameters and bromatological parameters of elephant grass under increasing doses of biosolid and chemical fertilizers, thereby evaluating the possibility of using elephant grass as biomass, both as an energy source and as an animal feed.

2 Materials and Methods

The experiment was performed in an experimental farm in the city of Pedras de Fogo (Paraíba state, Northeast Brazil) in an area where biosolid was never used and that had not been cultivated for more than 6 years. The municipality is located in the microregion of the Southern Coast of Paraíba, which is classified as As—tropical climate (winter rains)—according to the Köppen-Geiger climate classification.

Five fertilization treatments (four with biosolid and one with chemical fertilization) were used in addition to the control plot, with two replicates for each treatment, totaling 18 plots of 2 m × 2 m each, chosen at random, as shown in Fig. 1.

Fig. 1
figure 1

Schematic representation of the experimental design applied. Source: Neves (2017)

Full size image

The soil was prepared before the experiment started. The area for forage planting and development was harrowed and plowed to facilitate the growth and development of the plant and its management (Italiano 2004). Manual cleaning was performed, and the soil was plowed using a tractor (Lopes 2004).

The biosolid used in the experiment was produced by a textile industry ETS of the industrial district of João Pessoa/PB, whose treatment is based on an extended-aeration, activated-sludge biological system. Its application was determined according to the Brazilian National Environment Council (better known by its acronym in Portuguese, Conama) (Brazil 2006), which defines criteria and procedures for the agricultural use of sewage sludge generated in sewage treatment plants and its byproducts and contains other provisions. For the calculation of the application rate, a physicochemical analysis of the biosolid was performed according to the same resolution.

The application rate according to Conama Resolution 375/06 (Brazil 2006) was calculated based on the available amount of nitrogen in the biosolid and the requirements of this element for the full development of the crop to be established. The application rate was calculated using Eq. 1.

$$ \mathrm{Application}\ \mathrm{rate}\ \left(\mathrm{t}/\mathrm{ha}\right)=\mathrm{N}\ \mathrm{recommended}\ \left(\mathrm{kg}/\mathrm{ha}\right)/\mathrm{N}\ \mathrm{available}\ \left(\mathrm{kg}/\mathrm{t}\right) $$
(1)

where

N recommended:

Quantity of nitrogen recommended for the crop, according to the State’s official recommendation;

N available:

Quantity of nitrogen calculated according to annex 3 of Conama Resolution 375/06 (Brazil 2006)

The dose of nitrogen fertilizer recommended for the elephant grass crop was determined using information collected from the literature, as fertilization data from the state of Paraíba were scarce. Saraiva and Carvalho (1991) did not observe significant effects on the production of elephant grass (dry matter) when a dose of 120 kg of N/ha combined with phosphate fertilization was applied, whereas Monteiro (1994) recommended doses ranging from 30 to 300 kg of N/ha. In the present experiment, a dose of 200 kg of N/ha was used.

The calculation of the nitrogen available in the biosolid was performed according to annex 3 of Conama Resolution 375/06 (Brazil 2006) using Eq. 2.

$$ {\mathrm{N}}_{\mathrm{avail}}\ \left(\mathrm{mg}/\mathrm{kg}\right)=\left(\mathrm{MF}/100\right)\times \left({\mathrm{N}}_{\mathrm{Kj}}-{\mathrm{N}}_{\mathrm{N}\mathrm{H}4}\right)+0.5\ \mathrm{x}\ \left({\mathrm{N}}_{\mathrm{N}\mathrm{H}4}\right)+\left({\mathrm{N}}_{\mathrm{N}\mathrm{O}3}+{\mathrm{N}}_{\mathrm{N}\mathrm{O}2}\right) $$
(2)

where

MF:

Nitrogen mineralization fraction (%);

NKj:

Kjeldahl nitrogen (Kjeldahl nitrogen = total organic nitrogen + ammonia nitrogen (mg/kg);

NNH4:

Ammonia nitrogen (mg/kg);

NNO3 + NNO2:

Nitrate and nitrite nitrogen (mg/kg).

With the data collected in the physicochemical analysis of the biosolid, the following application rates were determined (Table 1) using the above equations, on a dry basis, for the four treatments that used the biosolid.

Table 1 Amount of biosolid applied
Full size table

The 1× biosolid dose was calculated according to the guidelines of Conama Resolution 375/06 (Brazil 2006). In the other treatments, namely, 2×, 4×, and 8× biosolid, the doses applied were 2×, 4×, and 8× the dose calculated according to the said Resolution, respectively.

The doses of biosolid were fractionated as follows: 15% of the total dose 1 day before elephant grass planting, 30% after 11 days, 30% after 18 days, and 25% after 34 days of planting. Before being applied, the biosolid underwent a stabilization process using hydrated lime. According to Conama Resolution 375/06 (Brazil 2006), it is mandatory to use a stabilization method to inactivate pathogens and reduce vector attraction. Laboratory tests were performed with crude biosolid mixed with lime in the proportions of 10, 20, 30, 40, and 50% of the dry weight of the biosolid, reaching a ratio of 30% lime to dry weight of the biosolid to reach pH 12 after 48 h and pH above 11.5 after 24 h to achieve alkaline stabilization of the biosolid. The lime was mixed with the biosolid on the day of application to the soil.

Chemical fertilization at elephant grass establishment, in addition to supplying nitrogen deficiencies, should provide sufficient phosphorus and potassium to meet the crop’s annual requirements; for this crop, 50 to 100 kg of P2O5/ha/year and 80 to 100 kg of KCl/ha/year are suggested, as many soils are potassium deficient (Werner 1986; Evangelista and Lima 2002). In the experiment, 100 kg/ha/year of both single superphosphate and potassium chloride was used. Thus, the fertilizer mix used had the following ratio: 200:100:100 (N:P:K). The fertilizer mix was applied only once, 1 day before planting the seedlings. The nitrogen source used was urea, which was also applied in a single application, 1 day before planting.

Planting was performed 1 day after application of chemical fertilizers and biosolid. In the soil preparation, approximately 15-cm-deep furrows were opened, which were spaced 1 m apart. The elephant grass seedlings were placed in the furrows, in pieces with three buds each, and spaced 1 m apart, according to the experience of the local farmers and based on the information in the scientific literature regarding elephant grass planting (Alcântara and Bufarah 1986; Gomide 1997; Martins and Fonseca 1998; Evangelista and Lima 2002).

The irrigation was performed using drippers, with a flow rate of 1.5 L/h, a spacing of 0.50 m between drippers, and one line of drippers per planting row.

The daily irrigation water depth was calculated using the methodology of Hargreaves and Samani (1982). For this purpose, a digital thermometer was installed in the experimental area to calculate the daily reference evapotranspiration from the maximum, minimum, and mean temperatures and the radiation using Eq. 3.

$$ {\mathrm{ET}}_0=0.0023\times \left(\mathrm{Tmean}+17.78\right)\times \left(\mathrm{Tmax}\hbox{--} \mathrm{Tmin}\right)0.5\times \left(\mathrm{RA}\times 0.408\right) $$
(3)

where

ET0:

Reference evapotranspiration (mm/h);

Tmean:

Mean temperature;

Tmax:

Maximum temperature;

Tmin:

Minimum temperature;

RA:

Extraterrestrial radiation (MJ/m2).

The extraterrestrial radiation was tabulated according to the latitude of the site and the month of the year. After the calculation of the reference evapotranspiration, the evapotranspiration of the crop was calculated using Eq. 4.

$$ {\mathrm{ET}}_{\mathrm{C}}={\mathrm{ET}}_0\times \mathrm{Kc} $$
(4)

where

ETC:

Crop evapotranspiration (mm/h);

Kc:

Crop coefficient.

The crop coefficient was tabulated. The values recommended by Alencar et al. (2009) for elephant grass were used.

Once the ETC was calculated, the daily irrigation depth was determined using Eq. 5.

$$ \mathrm{IW}=\left({\mathrm{ET}}_{\mathrm{C}}/\mathrm{Ei}\right)\times 100 $$
(5)

where.

IW:

Irrigation depth (mm/day);

Ei:

Irrigation water application efficiency (%).

The elephant grass was cut manually and the green matter was weighted per plot after 4 months of planting in the first cut and after 8 months of planting in the second cut. To estimate the plant height, three plants of the middle row of each plot were randomly selected, and the simple average of the three heights was calculated. The stem/leaf ratio was calculated using Eq. 6.

$$ \mathrm{Stem}/\mathrm{leaf}\ \mathrm{ratio}=\mathrm{Ws}/\left(\mathrm{Wgm}-\mathrm{Ws}\right) $$
(6)

where

Wgm:

Green matter weight of the middle row of each plot;

Ws:

Weight of stems without leaves;

Wgm − Ws:

Leaf weight.

Twenty random plants were collected from the middle row of each plot for the purpose of obtaining a composite sample to use in the bromatological analyses. Sixteen elephant grass plants were sufficient to perform the micronutrient analysis with a 10% error in the experiment of Primo et al. (2013) in Sobral (Ceará - Brazil). Once collected, the samples were sent directly to the laboratory responsible for the analyses, which used the methodology described by Bezerra Neto and Bareto (2004).

Using a randomized block design with six treatments (control plot, chemical fertilization, and 1, 2, 4, and 8× biosolid), with two replicates each, and totaling 18 experimental plots, statistical analysis of the data was performed using analysis of variance (ANOVA) through the F test and using the Tukey test to compare the means at the 1 and 5% levels of significance.

3 Results and Discussion

There was a significant effect (p < 0.01) for the treatments only for the total production variable in the first cut (TP1) (Table 2). The control and chemical fertilization treatments were statistically similar but different (p < 0.01) from the treatments 1× and 2× biosolid, which were also statistically similar. The treatments 4× and 8× biosolid differed from each other (p < 0.01) and from all other treatments in terms of the TP1 parameter. Evaluating the means using the Tukey test, there was a trend of increasing mean production in the first cut with the addition of biosolid in the soil for the control and chemical fertilization treatments. In the 1× biosolid treatment, as determined according to Conama Resolution 375 (Brazil 2006), the mean productions were greater than those obtained in the chemical fertilization and control treatments. Some authors, such as Pereira Jr et al. (1997), have found similar results for other crops; the aforementioned studies found that in a 3-year experiment, maize production was higher in treatments in which biosolids were used compared with chemical fertilization and the control lot.

Table 2 Summary of the ANOVA of the following variables: plant height in the first (PH1) and second (PH2) cut, stem/leaf ratio in the first (S/L_1) and second (S/L_2) cut, and total production in the first (TP1) and second (TP2) cut of the elephant grass
Full size table

Although it was not significant, a difference was observed in the mean production between the two cuts of elephant grass. A similar trend was expected with the first cut, which did not occur. It was observed that during the post-first cut period, in mid-April 2016, there was high rainfall in the experimental area, reaching cumulative volumes of 280.7 mm for that month alone (Aesa Executive Agency for Water Management of the State of Paraíba 2017). Thus, preferred paths of rainwater runoff among the experimental plots were observed, which may have caused leaching of materials from one plot to another. Thus, nutrients carried by this runoff may have caused the aforementioned difference in mean production between treatments in the two cuts. Na et al. (2016) found growth differences for three different grass types in different growth seasons.

Regarding the stem/leaf ratio, a parameter that is often evaluated for assessing the use of biomass as an energy source, there was a tendency of growth in the biosolid treatments compared with the others. All the mean values found in all treatments were greater than those found by Morais (2008) in an experiment in which they evaluated five different genotypes of elephant grass. Flores et al. (2012) did not find significant differences for the stem/leaf relationship between the treatments of their experiment for two species of elephant grass (fertilized and not fertilized).

Regarding the plant height parameter, the values found in both cuts were similar to those found by Deschamps (1999) in his experiment, which employed the same cutting period. Santos et al. (2001) for a cutting period of 3 months, using standardization cuts, found a mean plant height value of 1.63 m for an elephant grass cultivar.

Tables 3 and 4 present the results obtained for the bromatological parameters of elephant grass in the first and second cuts. It was observed that there were significant differences between treatments in terms of the following variables: total nitrogen (TN1) and total protein (p1) in the first cut and dry matter (DM2) in the second cut, all at p < 0.05, and acid detergent fiber (ADF1) in the first cut (p < 0.01).

Table 3 Summary of the ANOVA of the following variables: dry matter (DM1), mineral matter (MM1), total nitrogen (TN1), total protein (p1), acid detergent fiber (ADF1), and neutral detergent fiber (NDF1) in the first cut of the elephant grass
Full size table
Table 4 Summary of the ANOVA of the following variables: dry matter (DM2), mineral matter (MM2), total nitrogen (TN2), total protein (p2), acid detergent fiber (ADF2), and neutral detergent fiber (NDF2) in the second cut of the elephant grass
Full size table

There was a trend of increased mean values for the parameters total nitrogen and total protein with increased biosolid dose in both cuts compared with the chemical fertilization and control treatments. Some authors report for experiments with elephant grass that the highest levels of proteins in the plant are directly related to the nitrogen fertilizer doses (Ribeiro 1995; Andrade et al. 2003; Costa et al. 2004; Mistura et al. 2004). The results are similar to those found by Magalhães et al. (2009) who found that for three different cultivars of elephant grass, higher levels of protein were obtained in the treatments in which greater doses of nitrogen fertilizer were used.

Although some ADF values were less in the biosolid treatments than those in the chemical fertilization treatment and even in the control plot, all the mean values were greater than those reported by Nussio et al. (1998) and Kauter et al. (2006) for use of elephant grass as an energy source, for which fiber values must be greater than 30%. For neutral detergent fiber (NDF), all mean values were greater than those found by Andrade et al. (2003), which were 58.9 and 65.3% in the dry and rainy seasons, respectively, in the first and second cut.

The means found for the parameter mineral matter (MM) were satisfactory, although Vale et al. (2011) state that values greater than 7%, which occurred in all treatments in the two cuts, can compromise combustion when elephant grass is used for energy purposes. High ash content can cause energy losses, affect heat transfer, and decrease the calorific value of the biomass (Klautau 2008). Nevertheless, the mean values found in the experiment were very close to those found by Demirbas (2004) for rice husk, 11.3%, and greater than those found by the same author for sugarcane bagasse 22.3%, two biomasses used for burning and energy production.

The mean values for dry matter (DM) were greater than 94% in all treatments in both the first and second cuts. Rueda et al. (2016) did not find difference in the first 185 days of establishment of several species of elephant grass in relation to dry matter, with increase in this parameter only in the second and third years of the experiment. Some authors mention that DM production can be increased by combining biosolid with chemical fertilizers, depending on the biosolid or even the soil deficiencies, since both can have different characteristics. Anjos and Matiazzo (2000) added nitrogen and phosphorus fertilizers to the biosolid in their experiment, and Rangel et al. (2002) added potassium chloride to the biosolid.

As can be observed from Tables 3 and 4, there was a tendency of decreasing values of nitrogen and total protein and an increase in both NDF and ADF from one cut to the other. There was a tendency for fiber content to increase as the plant aged, whereas digestible components, such as protein and nitrogen, tended to decrease. Cabral et al. (2006) evaluated the NDF and crude protein levels in different cuts for an elephant grass genotype in the dry season and in the rainy season, finding higher NDF values and lower crude protein levels as the cutting interval increased; Martins-Costa et al. (2008) obtained similar results.

In all treatments evaluated in the present study, elephant grass suitable for energy use and animal supplementation was produced. The mean production in the treatment with the biosolid dose determined according to the Conama Resolution 375/06 (Brazil 2006) exhibited a trend that was very similar to that found for chemical fertilization. Regarding the treatments in which doses greater than the recommended level were used, new studies should be performed to better evaluate possible soil and plant contamination.

4 Conclusion

The results obtained in this study suggest that biosolid is able to supplement the soil with the nutrients necessary to produce high-quality elephant grass for both animal feed and energy use. The dose determined according to Conama Resolution 375/06 (Brazil 2006) enabled similar production to chemical fertilization when adding the two cuts, in addition to satisfactory bromatological characteristics for energy use and animal supplementation.

For doses greater than the level recommended by Conama Resolution 375/06 (Brazil 2006), in spite of the excellent results achieved, new studies should focus on possible soil and ground water contamination, mainly by heavy metals, pathogenic microorganisms, and even excess nitrogen, regardless of the levels being within the required limits determined according to the physicochemical and microbiological characterization of the biosolid.