KRAFT LIGNIN AS AN ADDITIVE IN PINE AND EUCALYPTUS PARTICLE COMPOSITION FOR BRIQUETTE PRODUCTION

ABSTRACT The objective of this study was to evaluate the properties of briquettes produced with different proportions of pine and eucalyptus wood, as well as to characterize the potential of kraft lignin as an additive in the composition of briquettes for energy generation. The treatments differed from one another in the pine and eucalyptus wood particle ratios (0, 25, 50, 75 and 100%), as well as for the kraft lignin content (0, 2, 4, 6, 8, 10, and 12%). The biomasses were characterized by their physical and chemical properties, and briquette properties were evaluated according to apparent density, lower calorific value (LCV), energy density, and maximum burst load. The results showed an increase in the apparent density as the proportion of lignin and eucalyptus in the briquettes increased. The particle composition of the briquettes had a higher influence on the energy density increase compared to the addition of kraft lignin, being more significant in briquettes produced with higher proportions of eucalyptus. It was also observed that the addition of lignin increased the resistance to the rupture load, and that there was a specific value at which this resistance was higher (at 7% or 11% of lignin, depending on the proportion of particles). Additionally, the briquettes made with 100% pine achieved greater mechanical resistance. In general, kraft lignin presented good potential for use as a briquette additive, contributing to improved energy and mechanical properties.

An attractive alternative for the pulp industry is to extract part of the lignin from the black liquor for commercialization as an additive, which can be used in the production of briquettes, phenolic resins, dispersants and surfactants (Lora et al., 2002;Gosselink et al., 2004).
Lignin is a phenolic macromolecule that forms part of the chemical composition of the wood, acting on the adhesion of fi bers. Alternative forms of lignin isolation to cooking the wood, entail signifi cant diff erences in its structures and physical and chemical properties, including its original shape in the wood, which changes its potential for application (Glaser, 1981). Among its advantages, kraft lignin has adhesive characteristics and can be used as a particulate binder for the production of briquettes. However, there are still few studies related to the topic (Gouvêa et al., 2010;Valadares et al., 2011).
Briquettes are products derived from plant biomass and have high potential to supply part of the energy consumed around the world through fossil fuels, whether in the form of thermal and/or electrical energy. Still, it is a renewable, easily transportable fuel, which is easy to handle, considering that it is a compacted product, with high energy density in comparison to other biomasses (Stolarski et al., 2013).
Compaction and particle adhesion are important factors that aff ect the mechanical strength and durability of briquettes. These factors can be infl uenced by the density of the constituent particles, as already observed in briquettes made of pine and eucalyptus, in which pine particles, being less dense, compact and agglutinate better. However, they are more expensive than eucalyptus particles, which are denser and more diffi cult to aggregate, which can result in briquettes with low mechanical resistance and a consequent reduction of their market value (Kaliyan and Morey, 2009).
Thus, mixing diff erent biomasses with diff erent properties may be an alternative to reduce costs and problems related to the low mechanical resistance caused by briquettes made from a single biomass, as in the case of eucalyptus wood. Additionally, the use of additives such as lignin may contribute to increase the mechanical resistance, as well as the calorifi c power and energy density of the briquette.
In this context, the objective of this work was to evaluate the properties of the briquettes produced with diff erent proportions of pine and eucalyptus wood, as well as to characterize the potential use of kraft lignin as an additive in the composition of briquettes for energy generation.
The lignin was isolated in the laboratory from the black kraft liquor, through the Lignoboost process (Berghel et al., 2013). Black liquor supplied from a Brazilian pulp plant was concentrated to 37% solids and acidifi ed with carbon dioxide to pH 8.5. Thereafter, sulfuric acid (20% v/v) was added to the acidifi ed liquor in a reactor at pH 2, and centrifuged for 15 minutes at a rotation of 5,000 rpm. The precipitated lignin was separated by a fi ltration system, followed by acid washing with its respective acids and then washed with hot distilled water. The retained material was then placed in the oven with forced ventilation at a temperature of 105 °C until completely dried. The dry lignin was subsequently milled and sieved to 100 mesh granulometry to ensure better homogeneity.

Characterization of wood and kraft lignin
Characterization of wood and kraft lignin followed the procedures described in Table 1. The moisture content, extractives, soluble and insoluble lignin, immediate chemical analysis and higher calorifi c value (HCV) were analyzed for the wood. In relation to lignin, the ash content, elemental chemical analysis and HCV were analyzed.

Production and qualifi cation of briquettes
The pine and eucalyptus woods were transformed into particles using a hammer mill for grinding. Subsequently, the particles were classifi ed in a sieve with mesh 5.67 and 2.77 mm 2 , collecting the fraction that was retained in the last sieve.
After classifi cation, the particles were placed in a closed circulation oven at 60 ± 2 ° C until reaching a humidity of approximately 12 ± 2%. The moisture (dry basis) of the particles was weighed on halogen light scales.
The briquettes were produced in a Lippel brand laboratory briquette, LB-32 model, at 120 oC, with a compaction time of 5 minutes, a cooling time of 5 minutes and compaction pressure of 10.3 x 10 6 N.m -2 (1500 PSI). The total biomass used to produce each briquette was 16 grams.
The apparent density of the briquettes was determined by weighing and subsequent immersion in mercury, obtaining the volume displaced according to the hydrostatic balance method, described by Vital (1984). The energy density was obtained by the product of the lower calorifi c value (LCV) by the apparent density of the briquettes. The lower calorifi c value (LCV) was obtained taking into account the higher calorifi c value (HCV) and the proportion of the hydrogen element (H), according to equation 1. Thus, the fi nal LCV was calculated from the percentage proportions of eucalyptus (E), pine (P) and lignin (L) present in briquettes, according to equation 2.

Experimental design
The results were submitted to the Lilliefors test for normality, and the Cochran test for homogeneity of variance. The data obtained from the characterization of the particles and the kraft lignin was submitted to analysis of variance to verify the diff erences between the treatments. When signifi cant diff erences were observed, the t-test was applied at 95% signifi cance.
The data obtained for each briquette qualifi cation parameter, composed of diff erent proportions of pine and eucalyptus particles, and treated with diff erent percentages of lignin as an additive (0, 2, 4, 6, 8, 10, 12%) was statistically analyzed by means of regression analysis. The adjusted equations were compared by the F test, using the model identity test adopting a signifi cance of up to 5% probability, according to the Regazzi (1993) methodology for linear models, and Regazzi and Silva (2004) for nonlinear models. The choice of model was based on the highest coeffi cient of determination (R²), signifi cant F test, standard error of the estimate and on the graphical distribution of the residues. Table 1 shows the average values of the physical and chemical properties of the raw materials used to produce the briquettes. Note that kraft lignin has energy potential to be used as an additive in briquettes.  Figure 1 shows the eff ect of the proportion of eucalyptus particles on the apparent density of briquettes when diff erent percentages of kraft lignin were added. The model identity test for the variables being studied was signifi cant, so two equations were estimated for the apparent density of the briquettes, one as a function of the proportion of particles (Figure 1a), and the other as a function of the proportion of kraft lignin added to the briquettes (Figure 1b). The quadratic regression model was the one that best explained the eff ect of the apparent density of the briquettes with the addition of kraft lignin and the proportion of particles in the composition of the briquettes.  Figure 3a shows the eff ect of the addition of lignin on the energy density of briquettes made using diff erent biomass proportions. According to the model identity F test, the equality of curves hypothesis was rejected. Thus, fi ve models were adjusted for energy density as a function of the proportion of added kraft lignin. The quadratic regression model was the one that best explained the eff ect of the addition of lignin on the energy density of the briquettes. Figure 3b shows the eff ect of lignin addition on the maximum burst load of briquettes made with diff erent biomass proportions. The equation of curves hypothesis was also rejected according to the F model identity test. In this way, fi ve models were adjusted for the maximum rupture load as a function of the proportion of added kraft lignin. The quadratic regression model was the one that best explained the eff ect of lignin addition on the maximum burst load of the briquettes.

Physical and chemical characteristics of wood and kraft lignin
Although kraft lignin presented better energetic characteristics, such as higher apparent density and higher LCV, a high ash content was observed, considering that values greater than 1% of ash in kraft lignin extracted by the Lignoboost process are considered high (Tomani, 2010 The contents of soluble extractives in alcohol / toluene, the soluble, insoluble and total lignin content are expressed in wood base. The means followed by the same letter in parentheses (within the same row) did not diff er from each other, by the Tukey test at 5% probability. In the bulk density of forest biomass, the eucalyptus particles were greater than that observed for the pine particles. These values explain the diff erent compaction rates found in the briquettes, since they were made with the same amount of compaction mass and pressure.

Briquettes properties
The maximum apparent density was reached at a proportion of 75% eucalyptus particles (1102 kg.m -3 ). In relation to the proportion of kraft lignin in the briquette composition, the best result was obtained at 12%. These increases were 3% and 4% in relation to the eucalyptus particles and to the addition of lignin, respectively, therefore being considered low. The explanation may be due to the diff erent compaction rates obtained as a function of the same pressure used for all briquettes.
According to Iwakiri et al. (2008), the apparent density of the compacted biomass is greater the lower the density of the source material. However, briquettes made with a higher proportion of eucalyptus particles, which have a higher bulk density in comparison to pine, did not present lower apparent density values. This contradiction can be explained by the diff erent compression ratio for the diff erent briquettes produced, because they are produced with the same amount of mass, with diff erent percentages of diff erent bulk density (pine, eucalyptus and lignin) and compacted under the same pressure.
In a study conducted by Gouvêa (2012) on the production of briquettes from a mixture of kraft lignin  and residues from the furniture industry pressed at diff erent temperatures (60 °C, 75° C and 90 °C), there was an increase in the apparent density of the briquettes with the increase of lignin content, reaching a higher value when 20% of lignin was added at all compaction temperatures.
In this study, a direct relationship between lignin increase and apparent density in the briquettes was also observed, given that the amount of lignin added to the particle composition caused small changes in the volume of the briquettes, since the mass was the same for all the treatments.
It is observed in Figure 3a that the addition of eucalyptus particles in the briquettes resulted in an inverse proportional eff ect for the lower calorifi c value, given that the LCV of eucalyptus (4418 kcal. kg -1 ) is the lowest amongst the components, with pine being (4480 kcal.kg -1 ) and kraft lignin (4896 kcal.kg -1 ). Thus, increasing the proportion of eucalyptus content reduces the proportion of the other components, and consequently reduces the LCV of briquettes. In Figure  3b, the addition of kraft lignin had a direct proportional eff ect on the LCV of the briquettes, which was to be expected, given that its contribution to the calorifi c value is greater.
As for energy density, it is observed in Figure 3a that the amount of energy stored per unit volume in the briquettes had a greater infl uence on the composition of the particles than the addition of kraft lignin. The values for energy density indicated that the best results were seen in briquettes made with 50% of each biomass (pine and eucalyptus), followed by briquettes composed of 75 and 100% eucalyptus, respectively. These values are explained by the variation of apparent density in the briquettes in relation to the proportion of eucalyptus particles, which had a signifi cantly positive relation, presenting a correlation coeffi cient of 0.49.
Another point to be observed in energy density is the positive eff ect of the addition of kraft lignin. Note that there was an increase for most treatments, except in briquettes made with 25% eucalyptus. The most signifi cant increases were 12, 11 and 9% in relation to the control treatment (0% kraft lignin), with the addition of 6, 8 and 12% kraft lignin in briquettes made with 50% and 75% eucalyptus, and 100% pine, respectively.
In addition to energy density, another property influenced by the composition of the particles and the addition of kraft lignin was the mechanical resistance of the briquettes. The results showed that with 25% eucalyptus particles there was a decrease in the mechanical resistance of briquettes. It is observed in Figure 3b that the highest maximum bursting load reached for briquettes with 100% pine, this is because, with its being a less dense wood than eucalyptus, there is a higher compaction rate due to the greater cohesion strength between the particles. Concerning the briquettes made with a mixture of pine and eucalyptus, the composition of 50% of eucalyptus was the one that presented better results.
Additionally, the addition of kraft lignin gave the briquettes greater strength, which was expected. Therefore, the compaction temperature (120 °C) was higher than the glass transition temperature of the lignin (Kaliyan and Morey, 2010;Stelte et al., 2011). This added to the moisture content of the biomass, with a softening of the kraft lignin, which spread throughout the particles during compaction. After cooling, the lignin became rigid again, acting as a bonding agent between the particles, which gave greater mechanical resistance to the briquettes.
For briquettes made only with eucalyptus and with 50% eucalyptus, the addition of 7% lignin was the point of greatest mechanical resistance. For the other treatments, with up to 12% lignin added, the mechanical resistance increased as the proportion of lignin in the briquettes increased. In general, it can be inferred that, depending on the proportion of biomass in the briquettes, there is an optimum lignin content, which gives the briquette maximum resistance, and that a high content of this component can reduce its mechanical resistance. The mechanical resistance of briquettes is important for usual movement, to avoid crushing or breaking during transport or fall.
It was found after the production of the briquettes, as well as by Gouvêa (2012), that they presented a hydrophobic film on its external surface, which may promote a reduction of its hygroscopic equilibrium moisture. This factor is important, since a lower moisture content in the briquette gives it a greater useful calorific value.

CONCLUSIONS
The apparent density in the briquettes increased with the lignin content and, with the increase of up to 75% of eucalyptus particles in its composition. The lower calorifi c value of the briquettes was proportional to the increase of the lignin content, and to the increase of the percentage of pine particles.
The composition of the briquette particles had a greater infl uence on the increase in the energetic density in comparison to the addition of kraft lignin, being more notable in briquettes made with a greater proportion of eucalyptus.
The increase of the lignin content gave the briquettes a higher bursting load. However, for briquettes made with 100% eucalyptus and 50% eucalyptus, the mechanical resistance presented a maximum value for the addition of 7% of lignin. For briquettes made with 100% pine, 25% eucalyptus and 75% eucalyptus, the maximum burst load was achieved by adding 11% lignin. In addition, briquettes made with only pine achieved a higher burst load.
In general, kraft lignin presented a good potential for use as a briquette additive leading to improved energy and mechanical properties. However, due to the variation of the values of its properties in relation to the proportion of lignin used, further research to test kraft lignin with a lower ash content is recommended.