Extraction Methods of Spanish Broom ( Spartium Junceum L . )

Effects of different extraction methods of the Spanish Broom shoots were measured and compared with the purpose of obtaining composite material. The content of cellulose, lignin, pentosan and ash in the Spanish Broom fi bers was determined. SEM analyses were performed.


INTRODUCTION 1. UVOD
The use of Spanish Broom and its diverse application has been known in the entire Mediterranean since the ancient times. Already in the ancient Rome, where Spanish Broom fi elds (genestium -or ager in later Latin) were cultivated as any other fi eld, and the plant or sprouts were sown in ploughed up furrows, as the Roman agronomist Columella describes. According to Vergil, hedges of Spanish Broom were planted in addition to willow, hazel, elder and other plants. Pliny writes that the sowing and planting of Spanish Broom is crucial for peasants. Amongst other things, its branches provide excellent material for tying vine and young trees. The Greeks, Romans and Carthagians used Spanish Broom as raw material for manufacturing ropes, nets, bags, sails, clothing and for covering roofs, whilst the fl ower of Spanish Broom produces exquisite honey.
During the '30s of the past century, foresters widely recommended the cultivation of Spanish Broom (Spartium junceum) in the coastal karst areas, more precisely M. Anić Ph. D in The Forest Journal in 1937 and engineer Stane Benko in his booklet on Spanish Broom, published by the Ministry of Industry and Mining, in Zagreb, 1946. The member of the Institute for Forestry Research Ante Premužić explains the possibility of using Spanish Broom for economic purposes in his report Systematic cultivation of Spanish broom in our karst areas, published in the Forest Journal. He wrote that diverse biological properties could be put to good use for forestation of poor soil of the coastal limestone area since Spanish Broom grows on clay and limestone soil of neutral, acid and alkali reaction. Positive sides of its cultivation were highly praised, giving these studies more of a propaganda character. The use of Spanish Broom along east Adriatic was described in detail in the publications of the Faculty of Philosophy, Zagreb University (Stojanović, 1962). Notwithstanding all that, the application of Spanish Broom, as raw material, is almost forgotten today. The entire activity has been reduced to a few associations more of a tourist signifi cance (Katović at al, 2011).
During the past several years the number of scientifi c work related to the production of Spanish Broom as raw material in the manufacture of composite materials considerably increased (Avella et al, 1998 The advantage of Spanish Broom over fl ax and hemp is that it can grow in the most unfavorable limestone soil; it is resilient to draught; and once planted it can be used during a period of up to twenty years, whilst fl ax and hemp demand high quality soil each year. In the past, natural fi bers were not taken into account as reinforcements for polymeric materials because of certain problems associated with their use. The lack of good interfacial adhesion, low degradation temperature, properties variability depending on harvest quality, age and body of the plant, and poor resistance towards moisture make the use of natural fi ber reinforced composites less attractive than the synthetic ones. However, the production of composites reinforced with synthetic fi bers and matrices (glass, carbon, aramid) requires a large amount of energy, and they are impossible or hard to recycle. Moreover, in case of fi re the presence of natural fi bers in composites decreases the content of toxic gases resulting from combustion.

Spanish Broom -biological properties 2.1. Brnistra -biološka svojstva
Spanish Broom is a shrub like plant from the family of legumes. It is the only species in the genus Spartium. Spanish Broom is a shrub that grows 1-1.5 meters tall and only old examples grow into smaller trees between 4 and 5 meters tall and 15 to 20 centimeters thick. Spanish Broom produces intensively yellow fl owers between May and July, and its legumes mature between August and October. Flowers have male and female organs pollinated by insects. It has many fl owers, one to two centimeters in diameter that cover hills, turning into a beautiful yellow color, whilst its pleasant fragrance is especially loved by women. Spanish Broom most often grows in fresh and suffi ciently moist soil, but it also grows quite well in very dry and rocky soil, stone falls and fl ood gravel pits. Shade does not suit it. It also has a special feature of xerophile plants. It grows small leaves on its shoots, which it discards during summer and drought and assimilates with long thin shoots, which have nuclei with chlorophyll under the cuticulum. Its roots are deep, and it binds the soil quite well. As a legume plant, it uses symbiosis to bind atmospheric nitrogen in the roots' lumps, thickening and enriching the soil (Bezić et al, 2003). After fi ber extraction, it can be used in the production of textile fi bers, just as hemp and fl ax. Its habitat is the Mediterranean area of the South Europe, southwest Asia and North West Africa. In Italy, in the Mediterranean area of olive groves, it climbs the altitude of 975 meters. In Turkey, Syria and Palestine, it reaches altitudes of 1,700 meters. It is regarded weed in the USA and New Zealand with a tendency to eradicate it in order to save indigenous plants.
Numerous anatomical adaptations to arid habitat can be noted in the microscopic structure of vegetative organs of Spanish Broom. The xerophytic (xerophytes -plants adapted to dry habitats) adaptation of the leaf is seen in its short life span as well as the transformation of its inner structure, with the dominating parenchyma chain. The top of the stem has taken over the leaf function, whilst the sclerenchyma fi ber and conductive elements occupy most of its secondary units (Cerchiara et al, 2010 A).
After the maturing of seeds, between mid July and autumn, shoots of the plant are cut with scissors directly near the main tree (Kovačević et al, 2010).

FTIR (Furierov transformirajući infracrveni spektrofotometar)
The instrument is used in infrared spectroscopy for the identifi cation of unknown material, for determining the ratio of individual components in the sample, its quality and consistency. Infrared beams pass through the sample. Some of the beams are absorbed in the sample and others pass through it, creating a spectrum unique to each element. Perkin-Elmer FT-IR Spectrum 100 was used for the purpose of work. Before the measuring, Spanish Broom samples were turned into powder, using a grinder with an oscillating steel bowl, model MM 400, of the Retsch company. The measuring was performed by FTIR spectrometer using ATR technique.

TA (termogravimetar)
The pyrolysis characteristics of Spanish broom, hemp and fl ax were analyzed using a thermogravimetric analyzer PerkinElmer Pyris 1 TGA. The samples, weighing between 4 and 5 mg were stacked in an open platinum sample pan and the experiment was conducted in air atmosphere. All samples for TGA were measured from 50 °C to 800 °C at the heating rate of 10 °C /min.

SEM (skenirajući elektronski mikroskop)
Morphological structures of Spanish Broom shoots were recorded with the FE-SEM (Field Emission-Scanning Electron Microscope) of the company TESCAN, with a steamer and unit for EDX analysis (Energy Dispersive X-Ray Analysis). Before of the microscopic recording, samples were processed for 180 seconds in a steamer with gold and palladium, using operative voltage of 5-15 kV.

Metode istraživanja
The Spanish broom fi bers were extracted from the plant branches harvested in Dalmatia (Šibenik area) which contained 35% of humidity in relation to the branches dried under standard conditions. Fig. 1 clearly shows that Spanish Broom shoots have two basic layers: rigid and woody inner layer, porous in the middle, which makes the plant quite light; and an outer layer, or rather skin consisting of stringy fi bers. Here, the lignin is a typical chemical and morphological part of the fi ber, which provides rigidity and conducts fl uids.
Methods for determining the content of cellulose, lignin, woody polyoses (hemicellulose), extracting substances and ash content were conducted in compliance with the regulations previously described in this journal (Antonović et al, 2007).
Effects of different types of the Spanish Broom shoots extraction were examined in the work in terms of quality and quantity of the obtained fi bers: A. Fiber extraction in seawater: Freshly picked Spanish Broom shoots were tied in a bundle, weighing 100 g and then soaked in water for 21 days. After resting in seawater, at the average temperature of 22 °C, the fi bers were separated from the woody part by rubbing against stone plates. They were rinsed in seawater and then the fi bers started to separate from the woody part. Before the examination, the fi bers were dried in standard conditions. The content of the obtained fi bers stood at approximately 12% of the total weight of the dry shoots.
B. Samples extracted in a solution of 15 g/l sodium alkali: Freshly picked Spanish Broom samples were tied in bundles (100 g) and processed during three hours at 120 °C. After the processing, the fi bers easily separated from the woody part. Before the examination, the fi bers were dried in standard conditions. The content of the obtained fi bers stood at approximately 30% of the total weight of the dry shoots. (Cerchiara et  al, 2010 B) C. Samples extracted using the DiCoDe procedure: Freshly picked Spanish Broom samples were tied in bundles (100 g) and processed during 15 minutes at 100 °C. After being processed in the alkali, the fi bers were rinsed in distilled water to the level of neutral, and then put in moist conditions in an autoclave at 120 °C and pressure of 10 atmospheres during three hours. Quick decompression followed. Afterwards, the fi bers easily separated from the woody part. Before the examination, the fi bers were dried in standard conditions. The content of the obtained fi bers stood at approximately 30% of the total weight of the dry shoots.
D. Extraction by microwave application: This is a new patented extraction procedure, developed at the Institute for Textile Chemical Technology and Ecology of the Zagreb University -Faculty of Textile and Technology. For the purpose of this study, a Tefl on reactor, inert to microwave activity, of the total volume of 500 ml, was constructed. 50 grams of Spanish Broom shoots and 300 ml of 15% NaOH solution were put in the Tefl on reactor and then in a microwave oven set to 900 W, for a period of 10 minutes. After being processed under the infl uence of the microwaves, the samples were rinsed in hot and then cold distilled water. After the rinsing, the fi bers easily separated from the woody part. The content of the obtained fi bers stood at approximately 30% of the total weight of the dry shoots.

REZULTATI I DISKUSIJA
The IR spectra of the fi ber sample obtained after alkaline extraction was recorded on the ATR-FTIR spectrometer (Spectrum 100, Perkin Elmer), while cotton, fl ax and hemp fi bers were taken as reference samples. From the recorded FT-IR spectra of the fi ber samples, it can be seen that all fi bers have peak characteristic for cellulose although they present some differences (Fig. 2.). The intensity of the signal at 1734 cm -1 corresponding to the C=C esther band related to pectin is higher in fl ax than in hemp fi bers, while this band is not found in Spanish broom fi ber sample. On the contrary, only a weak peak at ca. 1500 cm -1 corresponding to the C=C in-plane aromatic vibrations from lignin can be observed in the case of Spanish broom fi bers, providing an ulterior evidence that the applied chemical treatment was adequate for the almost complete removal of the non-cellulosic compounds from the broom fi bers. On the other hand, the two little sharp peaks observed over a broader peak in the area of 2850 to 2950 cm -1 , attributed to the CH 2 and CH groups of long alkyl chains of waxes, are present in the spectra of fl ax and hemp.  Fig. 3 shows TG spectra together with the corresponding DTG curves of some bast fi bers. The TG curve can be divided with respect to the three temperature ranges. A slight weight loss of the material is characteristic for temperatures under 200 °C, which is due to water loss in the form of absorbed moisture. In this area, weight loss of the examined material is under 10%. When the temperature is between 200 and 550 °C, signifi cant weight loss is evident, which results from thermal decomposition of hemicellulose, cellulose and lignins. In this area, the weight of the examined material decreased by 70 -80 %. When the temperature exceeds 550 °C, the weight loss is not as considerable as in the previous area. This weight loss is a consequence of thermal disintegration of other heavy components. Lignin-cellulose fi ber structure can be qualitatively identifi ed using DTG curves. In other words, hemicellulose, cellulose and lignin can be distinguished from the intensity distribution of the weight loss of the tested material. Thermal disintegration of the hemicellulose takes place at temperatures between 150 and 350 °C. The cellulose disintegrates at temperatures between 275 and 350 °C, whilst the lignin disintegrates at temperatures between 250 and 550 °C (Nekkaa et al, 2006;Chen et al, 2010).
The FTIR spectra of Spanish Broom fi bers obtained after four different extraction procedures are presented in Fig. 4.
According to the previously highlighted peaks that point to the presence of pectin, lignin and/or wax, it can be seen that the procedure B resulted in the best pectins and maybe not total elimination of lignin as seen from the broadening of the peak at 1640 cm -1 (see the shoulder towards the lower wave number values). However, the latter treatment seems very promising, especially from the economical point of view.
The chemical composition of fi bers extracted by different procedures is shown in Fig. 5. The only signi- fi cant difference is observed in the case of the procedure A for which the lowest amount of cellulose is found and the resulting high amounts of lignin, polyoses as well as extractives confi rm the observation made from the FTIR spectrum of the Spanish Broom fi bers obtained by sea water extraction procedure. The results of the remaining three procedures are equally good with respect to the cellulose content of the fi ber, approxima-tely 80%, whilst the content of the residuals differs from procedure to procedure. In the case of procedure B a kind of equality in the amounts of lignin, polyoses and extractives is found out, while the fi bers obtained by applying the procedures C and D show different distribution among lignin, polyose and extractive amounts. In the case of the procedure C, the amounts of lignin and extractives are almost equivalent; whilst the con- tent of polyoses is relatively high i.e. two times the former two components. For the microwave procedure (D), the lignin and polyoses amounts are comparable whereas the extractive content is very low. Samples of the remaining material after the extraction of fi bers from the Spanish Broom shoots were also examined. Fig. 6 clearly shows that wax is the major component of the cuticula (vermenes membrane), i.e. two sharp peaks at 2850 and 2920 cm -1 , respectively. Other samples present more or less pectin and/or lignin, while the predominant part are cellulosic components having a lower degree of crystallinity as it appears from the shape of the main peak at ca. 1030 cm -1 , one of the C-O stretching modes characteristic of cellulose, which is smoother presenting only shoulders at ca. 1155 cm -1 and 1105 cm -1 corresponding to the C-C ring breathing and the C-O-C glycosidic ether band.
The evaluation of the FTIR spectra is confi rmed by the chemical composition analyses presented in Fig.  7. The presented results are similar to those found in literature i.e. with respect to the fi bre extraction procedure, the cellulose amount changes from 66.9 to 91.7% in the case of mechanical and physicochemical procedure, respectively (Cerchiara et al, 2010 B) or 74.1% as reported by Angelini et al. (2000) for mechanically obtained fi bres. The same trend is valid for the lignin amount, which is always lower for the fi bers extracted by non-mechanical procedures.

ZAKLJUČCI
The comparison between water fi ber extraction and NaOH solution proved that the procedures in NaOH solution result with a signifi cantly higher quantity of fiber bundles in relation to water fi ber extraction. In all three cases of solution related processing, the quantity of fi bers and the remainder after the extraction is similar. Extraction by microwaves requires the smallest amount of energy and the shortest processing period.
The structure of the Spanish Broom fi bers can be qualitatively identifi ed using DTG curves. Hemicelluloses, celluloses and lignin can be distinguished from the intensity distribution of the weight loss of the tested material since thermal decomposition of cellulose occurs at different temperatures.
FT-IR specters of the fi bers cotton, hemp, fl ax and Spanish Broom have peak characteristic for cellulose although they present some differences.