APPLICABILITY EVALUATION OF Cupressus lusitanica FOR PULP PRODUCTION

This paper aimed to study Cupressus lusitanica wood tracheids biometrics along radial direction and chemical analysis in order to encourage the use of new forest species for pulp production in southern Brazil. Axial tracheids anatomical analysis was held in 3 positions, in the pith-periphery direction. Total length, total diameter and lumen diameter were analyzed, and wall thickness, flexibility coefficient, wall fraction, felting index and Runkel’s ratio were calculated based on these variables. Chemical analysis determined total extractives, sodium hydroxide attack degree, ash, lignin and holocellulose content. For anatomical variables, a statistically significant increase in the pith-periphery direction was observed for total length (L = 5,85mm), for total diameter (D = 47,18 μm), lumen diameter (d = 32,92μm) and wall thickness (e = 14,26μm) was observed stabilization from intermediate section. The species was characterized as with high rigidity tracheids (WF = 61,73%), although being considered “good” for wood pulp production (RI = 0,93). The chemical composition showed high lignin content (L = 36,21%) and low holocellulose content (H = 59,19%). Cupressus lusitanica wood showed appropriate anatomical characteristics to be part of the pulp industry raw material supply. However, chemical properties may compromise the species use due to high lignin content.


INTRODUCTION
Brazil was the 5 th largest world pulp producer (softwood and hardwood) in 2000, surpassed by the United States, Canada, Japan and Finland (ABRAF 2013).However, in the last decade, industry investment and national economic development exerted great impact on pulp production (BRACELPA 2013).It provided an increase of 87,7% in pulp domestic production and 190,4% in exports.Brazil's market share in the international pulp market went from 9,6% to significant 21% (ABRAF 2013).
According to IBÁ (2015), in 2014, Brazilian pulp production increase 8,8% when compared to 2013.During that year 16,46 million pulp tons were produced, evidencing Brazil as the 4 th largest world Pulp producer, only suprassed by United States, China and Canada.
The expected pulp domestic industry growth will require large forest areas planting, in addition to cold-resistant species studies, as southern Brazil states planting will be of great importance for the Likewise, Punches (2004) described that there is a need to study wood characteristics, since they are directly related to specific application and, depending on the final use, these properties are critical factors and must be taken into account in selecting the most suitable materials for industrial use.Clarke and Wessels (1995) stated that for the pulp and paper industry, the quality assessment process is not limited to the product analysis.In order to increase the process efficiency, all aspects that may influence the final product quality are analyzed, and a significant part lies in raw material improvement.However, there are still little available information about C. lusitanica wood pulp and paper properties, and little information touch on this species in the question of its features.
In wood pulping process studies, Gomide and Colodette (2007) referred to wood quality as the main parameter, since it is directly related to the wood anatomical and chemical structures.In the pulp and paper industry, the most popular species used as raw material for pulp production, has some contributing factors which include, excellent silvicultural characteristics for commercial planting and high yield rates.Besides wood quality, which is related to fiber dimensions as length and lumen diameter, wood specific gravity between 400 -550 kg/m³ and chemical compounds determination, highlighting cellulose and lignin (Punches 2004, Sansígolo andRamos 2011).
Thus, this work had as objective the study of biometrics of tracheid along the radial direction and chemical analysis of the species of C. lusitanica in order to direct the use of the species for the production of cellulose and aiming at the determination of the ideal point for harvest.

MATERIAL AND METHODS
The wood used to carry out the study came from a 19 years old Cupressus lusitanica MILL., experimental stand belonging to Florestal Gateados company, located in the municipality of Campo Belo do Sul, Santa Catarina state.

Sampling
42 logs with an average diameter ranging from 20 to 40 cm were obtained, and 5 trees out of them were selected according to COPANT 458 (1972) methodology, where the raw material used to carry out the chemical analysis was obtained from the first log.For C. lusitanica wood tracheid analyzes, the used material was derived from an approximately 5 cm thick disc located at the 5 selected trees BHD (1,30m) height.
Anatomical analysis of C. lusitanica wood aimed to obtain axial tracheids in the wood radial direction with 3 positions along the discs pith-periphery gradient.Samples were obtained in the pith, intermediate and periphery positions, which correspond to pith-periphery distance intervals of 0% (A1), 50% (A2) and 100% (A3), respectively, as shown in Figure 1.

Axial tracheids Analysis
The C. lusitanica small wooden fillets were obtained 40 samples for maceration technique application, following the methodology described by Franklin (1945).
Macerate samples were measured through the following variables; total length (L), total diameter (D) and lumen diameter (d) and wall thickness (e), in EUROMEX microscope, measured with the aid of lens with microns (µm) graduated reticule.According to the variables, important pulp and paper production anatomical parameters were assessed, such as: flexibility coefficient (FC), wall fraction (WF), felting index (FI) and Runkel's index (RI).

Samples chemical analysis
Analyses were performed using C. lusitanica 40-60 mesh sawdust (dry basis) following TAPPI standards recommendations -Technical Association of Pulp and Paper Industry (TAPPI) and ABNT-NBR, with these being described in Table 1.

Statistical analysis
Data were submitted to the Kolmogorov-Smirnov's test at 5% probability.Studied material chemical variables means, standard deviation and coefficient of variation (CV%) analyses were carried out when data normality was shown.As for tracheid analysis assessment, Tukey's test (TSD) at 5% probability was applied.

Dimensional tracheids analysis
Total length (L), total diameter (D), lumen diameter (d) and wall thickness (e) of wood anatomical characteristics results along radial direction are shown in Table 2.For the analyzed variables, a statistically significant increase was observed in the pith-periphery direction for tracheid length only, wall thickness has only statistical difference between pith and periphery sections.Total diameter has pith section lower than intermediate and periphery sections.For lumen diameter, tracheids in wood intermediate portion differ from pith section and is statistically equal as periphery region.
Applicability evaluation of ..: de Almeida et al.
Species that were closer to C. lusitanica were P. tecunumanii, with 10,00 µm (Shimoyama and Wiecheteck 1993), P. taeda, with 9,33 µm (Castelo et al. 2008) Thus, due to fiber length and wall thickness high values and reduced total diameter and lumen diameter values, higher paper strength and higher pulp yield may be expected.However, very thick fibers tend to keep their original form, what weaken interfiber bonding, decreasing strength and burst resistance, although it means an increase in tear resistance (Shimoyama and Wiecheteck 1993).
In addition to wood characteristics, anatomical relation indexes are essential for the selection of most suitable species to certain products and uses.Pulp and paper production influential parameters of C. lusitanica wood is described in Table 3. C. lusitanica wood tracheids relation indexes variation in pith-periphery direction showed statistical significance for the felting index only.Among these indexes, the flexibility coefficient relates lumen diameter and fiber length, and the higher its value, the more flexible will be the fiber, with more connections being likely to occur between them, what would increase tension and burst resistance (Shimoyama and Wiecheteck 1993).Thus, C. lusitanica wood has a mean FC value (69,13%), being higher than values observed for P. taeda (66,71%) (Castelo et al. 2008) and P. tecunumanii (64,00%) (Shimoyama and Wiecheteck 1993).
Wall fraction is a relation that must be assessed in the reverse of flexibility coefficient.Values below 40% are expected, so that fibers are not too rigid (Nisgoski et al. 2011).However, the material under study had a mean value of 61,73%; which is much higher than P. tecunumanii, 36 % (Shimoyama and Wiecheteck 1993) and P. taeda 33,29% (Castelo et al. 2008), meaning that C. lusitanica fibers are extremely rigid, inflexible and with interconnection difficulties (Rocha et al. 2004).
Regarding, felting index tracheids showed higher quality along the pith-periphery gradient, so that the wood next to the periphery region has higher resistance to tearing and rapture than wood pith and intermediate sections.In order to produce good quality papers, higher than 50 felting index values are necessary (Nisgoski et al. 2011), what is reached on C lusitanica wood.
Runkel's index showed no variation along positions, with a mean value of 0,93.The Runkel ratio is the ratio of fiber cell wall thickness to its lumen that determines the suitability of a fibrous material for pulp and paper production.If a wood species has a high Runkel ratio, its fiber will be stiff and less flexible and poor bonding ability.High Runkel ratio fibers produce bulkier paper than fibers with low Runkel ratio (Kiaei et al. 2014).
Fiber quality can be considered as "good", Rocha et al. (2004) mentions that fiber quality with up to 0,25 index is considered excellent for paper; from 0,25 to 0,50 is very good; from 0,50 to 1 is good; from 1 to 2 is regular, and above 2 may not be used for paper.
Applicability evaluation of ..: de Almeida et al.
The value observed was higher than described by Anoop et al. (2014) wich maximum values were below 0,60 for P. caribaea, P. patula and P. oocarpa on the same sections evaluated.
Even with high wall thickness and low flexibility coefficient, C. lusitanica wood showed suitable characteristics to supply the pulp and paper industry, with better RI values than some commonly used species in industry, such as P. taeda, with 0,52 (Castelo et al. 2008), P. tecunumanii, with 0,57 (Shimoyama and Wiecheteck 1993), and Eucalyptus grandis, 0,60 (Rocha et al. 2004).
However, contrary to Foelkel and Zvinakevicius (1975) and Pereira and Higa (2003), C. lusitanica wood has suitable anatomical characteristics for kraft pulp production, especially higher tear resistance compared to Pinus sp.species, which are commonly grown for pulp and paper production in southern Brazil.
The results also demonstrate a stabilization in the morphological parameters from the intermediate section and suggest the anticipation in the harvest of the afforestation, reducing the rotation duration.Total extractives content showed values ranging from 3,56 to 4,73%; and the maximum range was of 1,17%.Klock et al. (2005) reported that this content usually ranges from 5 to 8% in coniferous woods.
Wood extractives influence is of fundamental importance for species establishment, especially for forest-based industry unusual species, with main emphasis in the pulp and paper industry production processes.
Thus, extractives presence negatively affects productive pulping aspects, besides also causing increased reagents consumption, delignification reaction inhibition, equipment corrosion, pulp quality reduction and cooking liquor recovery hindering (Fengel and Wegener 1989, MIMMS 1993, IPT 1988, Sarto 2010).Therefore, species used for pulp and paper production need to have low extractive levels in its constitution, and C. lusitanica has acceptable extractive levels compared to already used Pinus spp.species.
Holocellulose is the constituent with the highest wood proportion in C. lusitanica, totaling 59,19%.However, the holocellulose content obtained was lower than what is expected for the species, and also low when compared to other conifer species.Okino et al. (2010) reported that, for the same species and with similar age 71,8% holocellulose values were observed, corroborating with Foelkel and Zvinakevicius (1975), who obtained 71,6% holocellulose and Pereira and Higa (2003), using 56 years old C. lusitanica wood obtained 62,6% holocellulose.

Figure 1 .
Figure 1.Specimen withdrawal position scheme for tracheids biometrics variation study.
Figure 2 are related to C. lusitanica wood chemical composition values through the following variables: total extractives, ash content, lignin content and holocellulose content.
found 7,3% total extractives in 56 years old C. lusitanica plantations, while Okino et al. (2010) found a mean value of 4,1% for total extractives, a very close value to the one obtained in this study.In a study with C. lusitanica, C. arizonica and C. sempervirens, Santos et al. (2014) observed total extractives values of 3,87; 2,46 and 3,34% respectively.

Table 1 .
Wood chemical analysis and standards list.
*Technical Association of Pulp and Paper Industry procedures **Brazilian Standards.

Table 3 .
Pulp and paper production dimensional relations.
(2012) al. (2011)4%) and E. dunnii (16,49%) biomass studies.However, results were lower than those obtained byTrugilho et al. (2005)using Eucalyptus clones, in which 21,69% values were shown for sodium hydroxide extractives, what states that C. lusitanica wood has lower low molecular weight carbohydrates removal susceptibility (hemicellulose and degraded cellulose) than some Eucalyptus clones used for pulp production.Inorganic material proportion in this study was shown through ash content, which had a mean of 0,51% what is close to commonly found values for conifers.According toAló et al. (2011), Pinus sp.wood ash content is of 0,62% while Souza et al.(2012)obtained 0,45% content for P. taeda.Lignin results of C. lusitanica had mean values of 36,21% which was higher when compared to lignin contents shown to the same specie and similar age by Okino et al. (2010) 33,5% and Santos et al. (2014) 34,69%.It also corroborate with