- Email: [email protected]
Industrial valorization of Quercus cerris bark: Pilot scale fractionation
Industrial Crops and Products 92 (2016) 42–49
Contents lists available at ScienceDirect
Industrial Crops and Products journal homepage: www.elsevier.com/locate/indcrop
Industrial valorization of Quercus cerris bark: Pilot scale fractionation Ali S¸en a,∗ , Carla Leite a , Leandro Lima a , Paulo Lopes b , Helena Pereira a a b
Centro de Estudos Florestais, Instituto Superior de Agronomia, Universidade de Lisboa, Tapada da Ajuda, 1349-017, Lisboa, Portugal Amorim & Irmãos, R&D Department, Rua de Meladas 380, P.O. Box 20, Mozelos, 4536-902, Portugal
a r t i c l e
i n f o
Article history: Received 19 January 2016 Received in revised form 7 July 2016 Accepted 27 July 2016 Keywords: Cork Bark Quercus cerris Fractionation Granulometric and densimetric separation
a b s t r a c t Cork-rich Quercus cerris bark collected from Turkey was fractionated in a laboratory and in a pilot-scale equipment to obtain cork and phloem fractions. After a primary trituration as a field post-harvest operation, the cork-rich bark granules were mainly concentrated in the big granules (>12 mm) indicating different mechanical properties of cork and phloem. The smaller granules (<4 mm) could be fractionated using water flotation by separating a floating cork fraction. The pilot-scale fractionation consisted of a mechanical grinding of the whole bark fractions followed by a granulometric and a densimetric separation. The operation was quite efficient and as a result, pure cork (8.4% wt) and cork-rich (18.5% wt) fractions were obtained. The colour analysis and FTIR spectroscopy showed the separation efficiency of the cork fractions. The results show the potential of Q. cerris bark for production of valuable cork fractions and the efficiency of combined granulation and fractionation. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Quercus cerris L., the Turkish or oriental oak, is a native tree species in the south and central Europe, Balkan Peninsula, Asia Minor, western Syria and Lebanon. This species has its major distribution in Turkey where it grows naturally in almost all parts of the country except in the easternmost regions (Kasapligil, 1981). The Q. cerris trees in south-eastern Turkey have a thick bark with a large rhytidome of several superposed periderms that contain substantial amounts of cork (S¸en et al., 2011a). The cork layers are conspicuous to the naked eye, making up patches of variable radial width from about 1 to 10 mm. Cork is a cellular material with structural and chemical features that give it an interesting combination of properties (Pereira, 2015). Cork is the raw material for an economic important industrial chain producing several cork products of which the wine cork stoppers are known worldwide and are the economic pillar of the industry (Pereira, 2007). The commercial cork that is used for wine stoppers is obtained from the cork oak (Quercus suber L.), a species with a geographical distribution restricted to the western Mediterranean basin and a total production limited to about 350 thousand tons (Pereira and Tomé, 2004). The idea of using cork-rich barks as an additional source of cork is tempting to enlarge the raw-material supply to the industry. Q.
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (A. S¸en). http://dx.doi.org/10.1016/j.indcrop.2016.07.044 0926-6690/© 2016 Elsevier B.V. All rights reserved.
cerris bark is an interesting candidate and as such has been the subject of recent studies. The cellular structure and chemical composition of Q. cerris cork proved to be similar to those of Q. suber cork (S¸en et al., 2010, 2011b), as well as its thermal behaviour (S¸en et al., 2012a, 2014). However Q. cerris bark has structural features i.e. anatomical composition, very different from those of Q. suber due to their different bark types (Sen et al., 2010). While Q. suber bark has no rhytidome and the cork forms a continuous and thick layer around the tree stem that may be removed and processed into cork products e.g. natural cork stoppers, in the case of Q. cerris the cork layers are included in a thick rhytidome with several periderms and are therefore separated by phloemic layers. This structure requires that bark has to be triturated and then undergo a fractionation process to efficiently separate the cork from the other lignocellulosic phloemic tissues. For the high value applications of cork e.g. for wine stoppers, it is necessary to have a cork-only fraction with granules of homogeneous granulometry which conditions the fractionation procedure. For instance, a laboratorial grinding and sieving followed by flotation was not sufficient to separate a pure cork fraction (S¸en et al., 2012a). It is therefore important to develop a fractionation methodology that could be adapted to industrial operational conditions and that yields cork granules with dimensions and purity adequate for the cork industry. The demonstration of the technological feasibility for separation of a pure cork fraction that may be used for high value agglomerated cork products is a fundamental step before further efforts towards Q. cerris bark valorisation. This was done here, where approximately one ton of Q. cerris bark was collected in the
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49
2.4. Laboratory separation by water flotation
Fraction
Triturated bark
> 12 mm Fraction 6-12 mm Fraction 4-6 mm Fraction 2-4 mm Fraction < 2 mm
43
1-2 mm 0.85-1.0 mm 0.425-0.850 mm 0.250-0.425 mm
A water flotation experiment was used to test its efficacy to obtain cork fractions with higher purity after laboratory scale fractioning. The fractions with 2–4 mm, 1–2 mm and 0.85–1.0 mm granules obtained from the initial post-harvest trituration were used. Approximately 30 g of each fraction were separated by density difference in distilled water with 15 min settling time after an initial quick mixing. The separation of the samples resulted in a floating fraction of cork-enriched granules (subsequently named supernatant) and a submerging fraction of phloem-enriched granules (subsequently named sediment). Both fractions were separated, dried and weighed. Two replicate experiments were performed for each fraction. 2.5. Pilot-scale fractioning
0.180-0.250 mm <0.180 mm Fig. 1. Schematic granulometric fractionation of the Quercus cerris bark after the post-harvest trituration obtained with laboratorial sieving equipment.
forest, triturated and fractionated in industrial pilot scale equipment, allowing to determine fraction yields and characteristics.
2. Material and methods 2.1. Samples The bark of Q. cerris was collected manually from the stumps of recently harvested trees at the Amanos Mountains in the Dörtyol province near the city of Hatay (Antioch), in the south of Turkey. The trees were cut to be used as firewood and for small-scale construction applications. The sampling site was a natural Q. cerris coppice stand with approximately 60-year-old trees.
2.2. Post-harvest trituration In order to facilitate transport, the bark samples were coarsely triturated locally in a hammer-type mill designed to granulate biomass for essential oil extraction. The trituration yielded bark pieces with varying dimensions, amounting to approximately 980 kg that were collected in four industrial big bags and shipped to our laboratory in Lisbon, Portugal.
2.3. Laboratory fractioning The result of the post-harvest trituration was evaluated upon arrival by fractionation in the laboratory. The triturated bark was manually fractionated using a vibratory sieving device with sieves with the following sizes: 12 mm, 6 mm, 4 mm and 2 mm. The fraction that passed the 2-mm sieve was further screened using a vibratory sieving apparatus (Retsch AS 200basic) with the following sieves: 1 mm, 0.850 mm, 0.425 mm, 0.250 mm and 0.180 mm. After sieving, the mass retained on each sieve was weighed and the corresponding mass fraction yields were determined (Fig. 1). Three independent samples with approximately 15 kg were taken and the result is shown as the mean. The moisture content of the bark particles was determined and averaged 12%.
The triturated bark samples were further fractionated in the pilot plant for cork processing of Cincork, Portugal. The pilot plant is designed to process the present commercial cork raw-materials and includes two cutting mills, sieve separation and densimetric tables; the equipment requires a minimum raw-material feedstock of approximately 600 kg. The Q. cerris bark that underwent the pilot scale fractionation amounted to 680 kg. A stepwise fractionation process was followed: the bark samples were passed through two mills, screened by a vibratory screens into five dimensional fractions (5–7 mm, 3–5 mm, 2–3 mm, 1–2 mm, 0.5–1 mm) followed by density separation on gravimetric separators into two density fractions (low density and high density) as schematically shown in Fig. 2. Fines were separated after each mill, and the oversized particles were recirculated in the second mill. After density separation, and following the practice used for the commercial cork, the granules were labelled as good (low-density and that can be directly used by the cork industry) and weak (high-density and rejected). 2.6. Bulk density The bulk densities of the fractionated bark samples were determined for each sieve fraction using a cylindrical glass container (25 mm diameter × 29 mm height) as the ratio of mass sample in the container to its volume. 2.7. UV/VIS spectroscopy and colour measurement The reflectance spectra of the laboratory-fractionated and pilot scale-fractionated bark fractions were measured with a Minolta CM-3630 (d/0◦ ) spectrophotometer at wavelengths between 360 and 720 nm with a 10 nm spectral resolution (S¸en et al., 2012b). The CIELAB colour parameters L*, a* and b* were determined for each granulometric fraction obtained by the two fractionation methods. Total colour differences (E*) between cork-rich and phloem-rich bark granules after pilot scale fractionation were calculated using the following formula (Bekhta and Niemz, 2003; Mononen et al., 2005; Fan et al., 2010):
2
2
E = L + a + b
2 1/2
The conversion of a reflectance spectrum into an absorbance spectrum was made by applying the Kubelka-Munk equation where R is the measured reflectance value, k is the light absorption coefficient and s is the light scattering coefficient. k = (1 − R)2 /2R s
44
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49
Fig. 2. Flow sheet for the pilot scale fractionation used in Cincork for cork samples.
Difference spectra were calculated by subtracting the reflectance spectrum of the good-labelled sample from that of the corresponding weak-labelled sample (Chen et al., 2012). k k k = (lowgradecork) − (highgradecork) s s s 2.8. FTIR spectroscopy Fourier transform infrared spectroscopy (FTIR) analysis was performed using KBr pellets. Each ultra-milled sample was oven-dried at 60 ◦ C overnight. A 1.5 mg aliquot of each sample was mixed with 200 mg KBr and then milled in a mortar. The KBr pellets were prepared with a compressor Carver C40000-544. The KBr pellets were analyzed on a Bruker FT-IR spectrometer (Alpha 100382) equipped with a KBr pellet detector accessory (Alpha-T), collecting 24 scans per sample at a resolution of 4 cm−1 and wavenumber range of 4000–375 cm−1 . Data treatment was performed using OPUS 6.5 software from Bruker. In order to increase the sensitivity of the FTIR analysis, FTIR-ATR (attenuated total reflectance) method was also tested. The bark fractions obtained in the flotation experiments (supernatant and sediment) were evaluated using as a reference a pure cork fraction that was separated manually from the 2–4 mm fraction by visually sorting out cork-only granules. The samples were dried over-night at 60◦ C and grinded with a ball mill Retsch MM200 during 20 min. FTIR-ATR spectra were acquired for each sample that was dried at 60◦ C during 2 h before analyses. Spectra were recorded on a Bruker FT-IR spectrometer (Alpha 100382) with the following settings: 24 scans per sample, spectral resolution of 4 cm−1 and wavenumber range of 4000–375 cm−1 , using a diamond single reflection (Platinum-ATR) accessory and data was processed using the OPUS 6.5 software from Bruker. 3. Results and discussion 3.1. Post-harvest fractioning The preliminary coarse trituration of the Q. cerris bark yielded very different sized particles from large cm-sized pieces to fine material. The results of the granulometric separation (Table 1) showed that the major fraction corresponded to large pieces (44% of the total were retained in a 12 mm sieve) with the fine material under 2 mm representing 17%. A further separation of the <2 mm fraction showed that it was composed mainly of the 0.425–0.850 mm fraction followed by 0.250–0.425 mm and 1–2 mm fractions (Table 2).
Table 1 Mass yields of Quercus cerris bark fractions obtained with a post-harvest trituration. Fractions
Yield (%)
>12 mm 6–12 mm 4–6 mm 2–4 mm <2 mm Total
44 18 7 14 17 100
Table 2 Mass yields of the <2 mm fraction of Quercus cerris bark. Fractions
Yield (%)
1–2 mm 0.85–1 mm 0.425-0.850 mm 0.250-0.425 mm 0.180-0.250 mm <0.180 mm Total
16 9 45 18 5 6 99
The macroscopic appearance of the fractions was very different and a visual observation based on their colour differences could already show their differing contents in cork and phloem (Fig. 3). The large pieces were constituted mainly by portions of rhytidome with a high amount of cork (with its interspersed phloem) but also including strips of phloem with a long fibrous appearance (Fig. 3a). The 6–12 mm fraction contained cork granules still showing parts of the phloem, and the fibrous phloem now in thinner strips (Fig. 3b). The 4–6 mm fraction contained cork particles, although in a lesser proportion, but most granules were of phloem, including both elongated and more spherical particles (Fig. 3c). The 2–4 mm fraction had a similar aspect but with a smaller proportion of cork granules (Fig. 3d) and even less in the <2 mm fraction (Fig. 3e). The results obtained with the post-harvest trituration regarding the granulometric fractioning (Tables 1 and 2) and the visual aspect of the fractions (Fig. 3) clearly show that there is a selective fractioning and a diverse cork content in each fraction. The bulk density of the fractions may be used as an indicator of cork content since cork has a low density in the range of 0.13–0.38 g cm−3 (Oliveira et al., 2014) although it should be taken it account that bulk density is much lower than the material’s density given the intergranular empty spaces. The fraction (Table 3) with the lowest bulk density of 0.26 g cm−3 was the 4–6 mm fraction due to the higher content of cork as compared for instance with the 1–2 mm fraction with a bulk density of 0.60 g cm−3 .
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49
45
Fig. 3. Visual aspect of the Quercus cerris bark fractions with different particle dimensions: (a) >12 mm (b) 6–12 mm; (c) 4–6 mm; (d) 2–4 mm; (e) <2 mm.
Table 3 Bulk density of different Quercus cerris bark granulometric fractions after laboratory scale fractioning. Fractions
Bulk density (g cm−3 )
4–6 mm 2–4 mm 1–2 mm 0.85–1 mm 0.425–0.85 mm 0.250–0.425 mm 0.180–0.250 mm <0.180 mm
0.26 0.38 0.60 0.59 0.60 0.60 0.46 0.46
These results are explained by the structural heterogeneity of the bark (S¸en et al., 2011a) and the different mechanical properties of the tissues i.e. cork and phloem. The milling process used in this post-harvest trituration had a strong component of impact and compression stresses for the fracturing. The lignocellulosic phloemic tissues are brittle and fracture more easily, leading to smaller sized particles derived mostly from the heavily sclerified non-functional phloem and rhytidome. Also the strips of fibrous appearance correspond to phloem particles and result from the cluster arrangement of fibres (S¸en et al., 2011a). On the other side, cork does not fracture under compression and allows large deformations under stress (Gibson et al., 1981; Pereira, 2007, 2015). The cork tissue is therefore present mostly in the larger sized fractions where it acted like an impact absorber and is therefore still interspersed with the phloemic layers. The medium sized particles have a combined content of cork and phloem and the fines include mostly phloem. It is known that the bark structure influences the trituration behaviour and the particle size distribution and may be used to selectively enrich fractions in certain components (Baptista et al., 2013; Miranda et al., 2012, 2013, 2014). This selectivity is clearly favoured in the case of Q. cerris bark, because of its large proportion of cork in the thick rhytidome. Fines usually contain mechanically brittle lignified materials that are rich in extractives, while cork-enriched fractions have larger dimensions. This was also found recently in the fractionation of Pseodotsuga menziesii bark which has a structural arrangement similar to that of Q. cerris (Ferreira et al., 2015).
Fig. 4. Separation of Quercus cerris bark granules of different dimensions by water flotation into a cork floating layer and a phloem sedimenting layer.
The coarse trituration that was made here showed that it is possible to make a first step grinding as a post-harvest operation that allows separating cork-poor fractions of smaller dimensions i.e. 2–4 mm and <2 mm. The test to check if these fractions contained an amount of cork that would recommend their consideration was made by water floatation. The water flotation results for the granules below 4 mm are shown in Fig. 4. The yield obtained for the cork-enriched fraction (floating layer) is high for the 2–4 mm fraction (Fig. 3d) corresponding to 70% of the total, but it decreased to about half this value for all granulates smaller than 2 mm. The water flotation is a useful step in bark fractioning since it allows a preliminary separation of high-density phloem and lowdensity cork fractions (Pinto et al., 2009; S¸en et al., 2010; Ferreira et al., 2013). The water flotation results (Fig. 4) show that the yield obtained for the cork-enriched fraction is high for the 2–4 mm fraction (Fig. 3d) corresponding to 70% of the total, while significantly decreasing in all granulates smaller than 2 mm. Therefore, the material obtained from the post-harvest trituration may be sieved to screen out granulates below 2 mm when targeting for cork-rich fractions. This removal would reduce by 17% the transport cost of the raw-material (Table 1).
46
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49
60
Table 4 Mass yields of cork and cork-enriched fractions obtained after pilot scale trituration and fractioning of a total of 654 kg of Quercus cerris bark. Yield (%)
5–7 mm cork (Good) 3–5 mm cork (Good) Other combined cork-enriched fractions Reject (Fines) Process loss Total
6.5 1.9 18.5 60.4 12.7 100
5–7 mm 3–5 mm 2–3 mm 1–2 mm 0.5–1 mm
1-2 mm
High density
0.19 0.10 0.25 0.14 0.08
– 0.37 0.51 0.50 0.36
3-5 mm weak 2-3 mm weak
30
1-2 mm weak 0.5-1 mm weak
20
0
Wavelength (nm) Fig. 5. Variation of the spectral reflectance values between the fractionated Quercus cerris bark granules after pilot scale fractioning.
3.2. Pilot-scale fractioning
6 5
3-5 mm 2-3 mm
4
1-2 mm
∆ K/S
The pilot scale fractioning was made using the complete bark raw-material obtained after the post-harvest trituration (Fig. 2). The fines were screened out after the first and the second mills, and the separation was made by dimensions into the fractions (between 0.5 to 7 mm) used presently in the cork granulates industry, followed by a densimetric sorting into a lower density fraction (considered as high quality cork) and a higher density fraction (considered as weak quality cork). Pure cork particles with 3–5 mm and 5–7 mm dimension, as required for use in the production of agglomerated cork stoppers, could be obtained with a yield of 8.4% from the initial raw bark (Table 4). It is also noteworthy that a cork-rich fraction with 18.5% yield was also obtained that would be suitable for other types of cork agglomerates with less demanding purity requirements e.g. for surfacing or insulation agglomerates. The fine particles smaller than 0.5 mm were removed during the process. The yield of fines and rejects was 60.4%. The pilot scale fractioning using trituration and granulometric and densimetric separations could be applied to the bark of Q. cerris and allowed obtaining pure cork particles with 3–5 mm and 5–7 mm dimension (Table 4). The bulk density of these low density fractions, corresponding to values of 0.1–0.25 g cm−3 (Table 5), confirmed that they contain pure cork granules suitable for applications by the cork stoppers’ industry, as also checked by expert visual observation. Further a cork-rich fraction to be used in the production of cork agglomerates for other applications was obtained (Table 4). The bulk density of these high density fractions was in the range of 0.4–0.5 g cm−3 but the visual observation showed that very little contamination of phloemic material was present in these fractions, which therefore may be used for surfacing and insulation cork applications. The results show that the raw bark of Q. cerris yielded 26.9% of granulated fractions that may be used by the cork industry. The fines and rejects corresponded to 60.4% of the bark. This value was almost double of that reported for Q. suber cork granulation (35%) (Rives et al., 2012), indicating the original lower cork content in the Q. cerris bark. The equipment used was a pilot designed and adapted to the operational conditions of the present cork industry and to the characteristics of the Q. suber cork boards. It is therefore foreseeable
0.5-1 mm
10
Bulk density (g cm−3 ) Low density
2-3 mm
40
Table 5 Bulk density of Quercus cerris bark fractions after pilot scale fractioning classified by particle size and density class. Dimension fractions
3-5 mm
50
Reflectances (%)
Fractions
5-7 mm
0.5-1 mm
3 2 1 0 360
400
440
480
520
560
600
640
680
720
Wavelength (nm) Fig. 6. Absorption spectra k/s of cork granules of Quercus cerris of different particle size.
that optimization to other bark raw materials such as the present Q. cerris bark would improve the results. Nevertheless the present study proves that it is possible to obtain cork fractions with high purity and adequate particle size for use in high value cork products from Q. cerris bark using the current procedures and equipment of the cork industry. 3.3. UV/VIS spectroscopy and colour analysis The spectral reflectance values allowed to separate cork granules (Fig. 5). In the pilot scale fractionated samples it was possible to distinguish the cork-rich 5–7 mm, 3–5 mm and 0.5–1.0 mm fractions from the other fractions. There were differences in the absorption spectra of each granule class with the maximum absorption observed at 390, 400, 470 and 460 nm in the 0.5–1 mm, 2–3 mm, 1–2 mm and 3–5 mm granules, respectively (Fig. 6) The colour analysis showed three different groups of the granulometric classes: 3–5 mm and 5–7 mm fractions in the first group, 2–3 mm and 1–2 mm fractions in the second group and the 0.5–1.0 mm fraction in the third group that differed mainly by their lightness (L*) while yellowness (b*) and redness (a*) values were similar(Table 6). The total colour differences (E) between
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49 Table 6 CIELAB colour parameters of pilot scale fractionated Quercus cerris bark granules. Granulometric fractions
5–7 mm 3–5 mm 2–3 mm 1–2 mm 0.5–1.0 mm
L*
a*
b*
47
of Q. cerris cork are similar to those of corks from Q. suber and Pseudotsuga meziesii (Marques et al., 1996, 2006).
E*
Good
Weak
Good
Weak
Good
Weak
49.39 50.00 44.71 44.07 53.24
– 42.32 36.55 36.60 41.14
12.18 11.63 12.13 10.66 12.38
– 12.49 11.50 11.24 11.32
21.28 22.45 20.63 20.90 24.31
– 19.70 17.51 17.80 19.65
3.5. Economic considerations – 8.21 8.76 8.11 13.01
*E is the total colour difference between good and weak granules.
Table 7 CIELAB colour parameters of laboratory scale fractionated Quercus cerris bark granules. Granulometric fractions
L*
a*
b*
4–6 mm 2–4 mm 1–2 mm 0.85–1 mm 0.425–0.85 mm 0.250.0.425 mm 0.180–0.250 mm <0.180 mm
35.54 35.55 35.32 35.42 35.63 36.11 35.95 33.69
9.51 10.06 9.83 9.62 9.96 9.92 9.67 9.61
18.04 17.89 17.90 17.85 18.34 18.45 18.04 17.16
the good and weak quality fractions were similar although a higher colour difference was found for the 0.5–1.0 mm class. Colour was an indicator of differences between fractions related to the differing cork contents. Pure cork fractions have higher L* and lower b* in comparison with less pure cork fractions (Table 6) while fractions with substantial amounts of phloem have lower L*, a* and b* values (Table 7). The differences in the absorption spectra of each granule class e.g. in the maximum absorption (Fig. 6) imply different chemical compositions. The observed absorptions should be related to the suberin and lignin contents of the samples since lignin absorbs the uv/visible radiation up to 400 nm, with a peak at 280 nm (Pandey, 2005) while pure cellulose does not absorb radiation in the uv/vis region (Wondraczek et al., 2011).
3.4. FTIR observations The FTIR spectroscopy is a useful tool to monitor the chemical composition of bark (S¸en et al., 2012a,b). Fig. 7 shows the FTIR spectra of the cork-enriched granules separated by water flotation and includes also a manually separated cork sample. The bands of cork and phloem fractions were the same, only their intensity were different indicating a non-homogenous material. Nine intense bands were detected in both materials: the band at 3315 cm−1 refers to OH stretching, the bands at 2914 cm−1 and 2849 cm−1 are characteristic of symmetric and assymetric C H stretching bands of suberin, the band at 1735–1737 cm−1 is the characteristic C O band of suberin (Lopes et al., 2001). The intense bands at 1615 cm−1 , 1319 cm−1 , 777 cm−1 , and 513 cm−1 are assigned to C O stretching bands in calcium oxalate monohydrate (Conti et al., 2010; S¸en et al., 2012a,b). These bands were not observed in Q. suber cork spectrum. The intense band at 1033 cm−1 is characteristic of the C O valence vibration of polysaccharides (Colom et al., 2003; S¸en et al., 2012b) FTIR analysis may also be used as an indicator of cork content due to the differences of spectral features of cork and phloem, as given by the suberin specific bands at 2914, 2849 cm−1 and 1737 cm−1 . Low intensities of these bands are indicative of a fraction poor in suberin i.e. in cork, as it was the case in the phloem samples (sedimenting layer) after water flotation (Fig. 7). The spectral features
The usage of non-commercial corks such as Q. cerris cork by the cork industry is the first step in the valorisation of cork-rich barks. Since the cork industry is already established, with defined flow processes and operations, as well as with requirements for the specific cork products that are produced, it is necessary to separate adequate cork granulated fractions from the raw barks. These cork granulates can then be integrated to the cork granulate stream in the industry and follow the production process without any pretreatment. After evaluating the technical feasibility of obtaining cork fractions from the raw Q. cerris bark, using standard equipment from the cork industry, as it was made here, it is also important to look at the economic aspect, namely at the comparative prices of the raw-materials, as well as to its potential availability. Using the available statistical data of the Forestry Services for the region of sampling, an area of the Q. cerris dominated oak forests of over 100,000 ha, an annual wood production of 30000–50000 m3 may be estimated, allowing an estimation of the potential availability of bark of approximately 5000 m3 corresponding to over 2500 tons. The cost of Q. cerris bark is 15 D ton−1 in Turkey (GDF, 2015), a value that is lower by a 100 fac100factor to the price of raw cork in Portugal or Spain (1700 D ton−1 and 1100 D ton−1 respectively) (Sierra-Pérez et al., 2015). Considering that a 25% yield of cork fractions could be obtained from the bark, their raw price (excluding trituration and separation costs) would be 60 D ton−1 of cork fractions. Their final cost as an industrial raw-material will depend essentially on the transportation costs to the mill site. A rough estimate of locally performed trituration and separation costs, and of transport to the Portuguese cork industry was calculated as 300–400 D ton−1 . The very large price difference in relation to the Q. suber cork indicates that this new cork raw-material from Q. cerris bark should be economically interesting for the industry. It will also give a possibility of raw-material supply enlargement and diversification, which is a strategic advantage given the limited supply of Q. suber cork. It should also be noted that the use of Q. cerris bark is environmentally sustainable since it is associated to the overall forest management and tree exploitation. In fact the bark is obtained as a residue from the exploitation of Q. cerris trees that are periodically harvested from coppice and high forests directed mostly to local and regional use for energy and small scale construction wood. The bark is removed either at the forest without damaging the live trees or at processing sites. In any case the valorisation of Q. cerris bark will contribute to an overall sustainable resource management.
4. Conclusions A stepwise fractionation scheme is proposed for the raw Q. cerris bark, as shown by Fig. 3. A first step of coarse trituration near the bark collection is advised, for better subsequent handling and transport with the sieving out of particles below 2 mm. The further processing at mill site should use two cutting mills separated by the sieving out of particles below 0.5 mm, with the granulated material going to sieving and densimetric tables for fraction separation. Particles above 7 mm should undergo a second cutting pass. With such trituration and fractionation design, it was possible to obtain pure cork fractions with a yield of 8.4% that can be used for
48
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49
Cork 2-4 mm Phloem 2-4 mm Cork manual separaon 2-4 mm Cork 0.85-1.0 mm Phloem 0.85-1.0 mm Cork 1-2 mm Phloem 1-2 mm
3875
3375
2875
2375
1875
1375
875
375
Wavenumber (cm-1) Fig. 7. FTIR spectra of cork-enriched and phloem-enriched granules separated by water separation.
production of cork agglomerated stoppers and fractions with a yield of 18.5% for production of surfacing and other cork agglomerates. This is an important valorisation route for the cork-rich bark of Q. cerris. Considering the comparative cost advantage of the Q. cerris cork, its integration into the present cork industrial production lines seems economically interesting, and will also allow to enlarge the raw-material supply base. Acknowledgements This work was supported by FEDER funds through the Operational Programme for Competitiveness Factors—COMPETE under the project NewCork 38363, and is part of the activities at the Strategic Project (UID/AGR/00239/2013) of Centro de Estudos Florestais, a research unit supported by the national funding of FCT—Fundac¸ão para a Ciência e a Tecnologia. We thank Cincork—Centro de Formac¸ão Profissional da Indústria de Cortic¸a (Portugal) for the pilot-scale granulation. The first author thanks FCT for a postdoctoral scholarship and also the Turkish General Directorate of Forestry, Kahramanmaras Regional Directorate of Forestry and Mr. Sebahattin Ozdemir and Ilker Ozdemir from Ozdrog Company for their kind help in sampling and treatment of Q. cerris barks. References Baptista, I., Miranda, I., Quilhó, T., Gominho, J., Pereira, H., 2013. Characterisation and fractioning of Tectona grandis bark in view of its valorisation as a biorefinery raw-material. Ind. Crop Prod. 50, 166–175. Bekhta, P., Niemz, P., 2003. Effect of high temperature on the change in color, dimensional stability and mechanical properties of spruce wood. Holzforschung 57 (5), 539–546. Chen, Y., Fan, Y., Gao, J., Stark, N.M., 2012. The effect of heat treatment on the chemical and color change of black locust (Robinia pseudoacacia) wood flour. Bioresources 7 (1), 1157–1170. Colom, X., Carrillo, F., Nogués, F., Garriga, P., 2003. Structural analysis of photodegraded wood by means of FTIR spectroscopy. Polym. Degrad. Stabil. 80 (3), 543–549. Conti, C., Brambilla, L., Colombo, C., Dellasega, D., Gatta, G.D., Realini, M., Zerbi, G., 2010. Stability and transformation mechanism of weddellite nanocrystals studied by X-ray diffraction and infrared spectroscopy. Phys. Chem. Chem. Phys. 12 (43), 14560–14566. Fan, Y., Gao, J., Chen, Y., 2010. Colour responses of black locust (Robinia pseudoacacia L.) to solvent extraction and heat treatment. Wood Sci. Technol. 44 (4), 667–678. Ferreira, R., Garcia, H., Sousa, A.F., Freire, C.S., Silvestre, A.J., Rebelo, L.P.N., Pereira, C.S., 2013. Isolation of suberin from birch outer bark and cork using ionic liquids: a new source of macromonomers. Ind. Crop Prod. 44, 520–527.
Ferreira, J.P., Miranda, I., Gominho, J., Pereira, H., 2015. Selective fractioning of Pseudotsuga menziesii bark and chemical characterization in view of an integrated valorization. Ind Crop Prod. 74, 998–1007. General Directory of Forestry (GDF) of Turkey, 2015. Non-wood forest products price plan 2014 (in Turkish). Retrieved from http://www.ogm.gov.tr/ ekutuphane/Dokumanlar/2014%20yili%20Tarife%20Bedeli%20Cetveli.pdf. Gibson, L. J., Easterling, K. E., Ashby, M. F. 1981. The structure and mechanics of cork. In Proceedings of the Royal Society of London A: Mathematical, Physical and Engineering Sciences (Vol. 377, No. 1769, pp. 99-117). The Royal Society. Kasapligil, B., 1981. Past and present oaks of Turkey. I. Phytologia 49, 95–146. Lopes, M.H., Barros, A.S., Pascoal Neto, C., Rutledge, D., Delgadillo, I., Gil, A.M., 2001. Variability of cork from Portuguese Quercus suber studied by solid-state 13C-NMR and FTIR spectroscopies. Biopolymers 62 (5), 268–277. Marques, A.V., Pereira, H., Meier, D., Faix, O., 1996. Isolation and characterization of a guaiacyl lignin from saponified cork of Quercus suber L. Holzforschung 50 (5), 393–400. Marques, A.V., Pereira, H., Rodrigues, J., Meier, D., Faix, O., 2006. Isolation and comparative characterization of a Björkman lignin from the saponified cork of Douglas-fir bark. J. Anal. Appl. Pyrol. 77 (2), 169–176. Miranda, I., Gominho, J., Mirra, I., Pereira, H., 2012. Chemical characterization of barks from Picea abies and Pinus sylvestris after fractioning into different particle sizes. Ind. Crop Prod. 36 (1), 395–400. Miranda, I., Gominho, J., Mirra, I., Pereira, H., 2013. Fractioning and chemical characterization of barks of Betula pendula and Eucalyptus globulus. Ind. Crop Prod. 41, 299–305. Miranda, I., Gominho, J., Ferreira-Dias, S., Pereira, H., 2014. Pattern recognition as a tool to discriminate softwood and hardwood bark fractions with different particle size. Wood Sci. Technol. 48 (6), 1197–1211. Mononen, K., Jääskeläinen, A.S., Alvila, L., Pakkanen, T.T., Vuorinen, T., 2005. Chemical changes in silver birch (Betula pendula Roth) wood caused by hydrogen peroxide bleaching and monitored by color measurement (CIELab) and UV-Vis, FTIR and UVRR spectroscopy. Holzforschung 59 (4), 381–388. Oliveira, V., Rosa, M.E., Pereira, H., 2014. Variability of the compression properties of cork. Wood Sci. Technol. 48 (5), 937–948. Pandey, K.K., 2005. Study of the effect of photo-irradiation on the surface chemistry of wood. Polym. Degrad. Stabil. 90 (1), 9–20. Pereira, H., Tomé, M., 2004. Cork oak. In: Encyclopedia of Forest Sciences. Elsevier Oxford, pp. 613–620. Pereira, H., 2007. Cork: Biology, Production and Uses: Biology, Production and Uses. Elsevier. Pereira, H., 2015. The rationale behind cork properties: a review of structure and chemistry. Bioresources 10 (3), 6207–6229. Pinto, P.C., Sousa, A.F., Silvestre, A.J., Neto, C.P., Gandini, A., Eckerman, C., Holmbom, B., 2009. Quercus suber and Betula pendula outer barks as renewable sources of oleochemicals: a comparative study. Ind. Crop Prod. 29 (1), 126–132. Rives, J., Fernandez-Rodriguez, I., Gabarrell, X., Rieradevall, J., 2012. Environmental analysis of cork granulate production in Catalonia–Northern Spain. Resour. Conserv. Recycl. 58, 132–142. S¸en, A., Miranda, I., Santos, S., Grac¸a, J., Pereira, H., 2010. The chemical composition of cork and phloem in the rhytidome of Quercus cerris bark. Ind. Crop Prod. 31 (2), 417–422. S¸en, A., Quilho, T., Pereira, H., 2011a. Bark anatomy of Quercus cerris L. var. cerris from Turkey. Turk. J. Bot. 35, 45–55.
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49 S¸en, A., Quilhó, T., Pereira, H., 2011b. The cellular structure of cork from Quercus cerris var. cerris bark in a materials’ perspective. Ind. Crop Prod. 34 (1), 929–936. S¸en, A., Miranda, I., Pereira, H., 2012a. Temperature-induced structural and chemical changes in cork from Quercus cerris. Ind. Crop Prod. 37 (1), 508–513. S¸en, A., Marques, A.V., Gominho, J., Pereira, H., 2012b. Study of thermochemical treatments of cork in the 150–400 ◦ C range using colour analysis and FTIR spectroscopy. Ind. Crop Prod. 38, 132–138.
S¸en, A., Van den Bulcke, J., Defoirdt, N., Van Acker, J., Pereira, H., 2014. Thermal behaviour of cork and cork components. Thermochim. Acta 582, 94–100. Sierra-Pérez, J., Boschmonart-Rives, J., Gabarrell, X., 2015. Production and trade analysis in the Iberian cork sector: economic characterization of a forest industry. Resour. Conserv. Recycl. 98, 55–66. Wondraczek, H., Kotiaho, A., Fardim, P., Heinze, T., 2011. Photoactive polysaccharides. Carbohydr. Polym. 83 (3), 1048–1061.
49
Contents lists available at ScienceDirect
Industrial Crops and Products journal homepage: www.elsevier.com/locate/indcrop
Industrial valorization of Quercus cerris bark: Pilot scale fractionation Ali S¸en a,∗ , Carla Leite a , Leandro Lima a , Paulo Lopes b , Helena Pereira a a b
Centro de Estudos Florestais, Instituto Superior de Agronomia, Universidade de Lisboa, Tapada da Ajuda, 1349-017, Lisboa, Portugal Amorim & Irmãos, R&D Department, Rua de Meladas 380, P.O. Box 20, Mozelos, 4536-902, Portugal
a r t i c l e
i n f o
Article history: Received 19 January 2016 Received in revised form 7 July 2016 Accepted 27 July 2016 Keywords: Cork Bark Quercus cerris Fractionation Granulometric and densimetric separation
a b s t r a c t Cork-rich Quercus cerris bark collected from Turkey was fractionated in a laboratory and in a pilot-scale equipment to obtain cork and phloem fractions. After a primary trituration as a field post-harvest operation, the cork-rich bark granules were mainly concentrated in the big granules (>12 mm) indicating different mechanical properties of cork and phloem. The smaller granules (<4 mm) could be fractionated using water flotation by separating a floating cork fraction. The pilot-scale fractionation consisted of a mechanical grinding of the whole bark fractions followed by a granulometric and a densimetric separation. The operation was quite efficient and as a result, pure cork (8.4% wt) and cork-rich (18.5% wt) fractions were obtained. The colour analysis and FTIR spectroscopy showed the separation efficiency of the cork fractions. The results show the potential of Q. cerris bark for production of valuable cork fractions and the efficiency of combined granulation and fractionation. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Quercus cerris L., the Turkish or oriental oak, is a native tree species in the south and central Europe, Balkan Peninsula, Asia Minor, western Syria and Lebanon. This species has its major distribution in Turkey where it grows naturally in almost all parts of the country except in the easternmost regions (Kasapligil, 1981). The Q. cerris trees in south-eastern Turkey have a thick bark with a large rhytidome of several superposed periderms that contain substantial amounts of cork (S¸en et al., 2011a). The cork layers are conspicuous to the naked eye, making up patches of variable radial width from about 1 to 10 mm. Cork is a cellular material with structural and chemical features that give it an interesting combination of properties (Pereira, 2015). Cork is the raw material for an economic important industrial chain producing several cork products of which the wine cork stoppers are known worldwide and are the economic pillar of the industry (Pereira, 2007). The commercial cork that is used for wine stoppers is obtained from the cork oak (Quercus suber L.), a species with a geographical distribution restricted to the western Mediterranean basin and a total production limited to about 350 thousand tons (Pereira and Tomé, 2004). The idea of using cork-rich barks as an additional source of cork is tempting to enlarge the raw-material supply to the industry. Q.
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (A. S¸en). http://dx.doi.org/10.1016/j.indcrop.2016.07.044 0926-6690/© 2016 Elsevier B.V. All rights reserved.
cerris bark is an interesting candidate and as such has been the subject of recent studies. The cellular structure and chemical composition of Q. cerris cork proved to be similar to those of Q. suber cork (S¸en et al., 2010, 2011b), as well as its thermal behaviour (S¸en et al., 2012a, 2014). However Q. cerris bark has structural features i.e. anatomical composition, very different from those of Q. suber due to their different bark types (Sen et al., 2010). While Q. suber bark has no rhytidome and the cork forms a continuous and thick layer around the tree stem that may be removed and processed into cork products e.g. natural cork stoppers, in the case of Q. cerris the cork layers are included in a thick rhytidome with several periderms and are therefore separated by phloemic layers. This structure requires that bark has to be triturated and then undergo a fractionation process to efficiently separate the cork from the other lignocellulosic phloemic tissues. For the high value applications of cork e.g. for wine stoppers, it is necessary to have a cork-only fraction with granules of homogeneous granulometry which conditions the fractionation procedure. For instance, a laboratorial grinding and sieving followed by flotation was not sufficient to separate a pure cork fraction (S¸en et al., 2012a). It is therefore important to develop a fractionation methodology that could be adapted to industrial operational conditions and that yields cork granules with dimensions and purity adequate for the cork industry. The demonstration of the technological feasibility for separation of a pure cork fraction that may be used for high value agglomerated cork products is a fundamental step before further efforts towards Q. cerris bark valorisation. This was done here, where approximately one ton of Q. cerris bark was collected in the
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49
2.4. Laboratory separation by water flotation
Fraction
Triturated bark
> 12 mm Fraction 6-12 mm Fraction 4-6 mm Fraction 2-4 mm Fraction < 2 mm
43
1-2 mm 0.85-1.0 mm 0.425-0.850 mm 0.250-0.425 mm
A water flotation experiment was used to test its efficacy to obtain cork fractions with higher purity after laboratory scale fractioning. The fractions with 2–4 mm, 1–2 mm and 0.85–1.0 mm granules obtained from the initial post-harvest trituration were used. Approximately 30 g of each fraction were separated by density difference in distilled water with 15 min settling time after an initial quick mixing. The separation of the samples resulted in a floating fraction of cork-enriched granules (subsequently named supernatant) and a submerging fraction of phloem-enriched granules (subsequently named sediment). Both fractions were separated, dried and weighed. Two replicate experiments were performed for each fraction. 2.5. Pilot-scale fractioning
0.180-0.250 mm <0.180 mm Fig. 1. Schematic granulometric fractionation of the Quercus cerris bark after the post-harvest trituration obtained with laboratorial sieving equipment.
forest, triturated and fractionated in industrial pilot scale equipment, allowing to determine fraction yields and characteristics.
2. Material and methods 2.1. Samples The bark of Q. cerris was collected manually from the stumps of recently harvested trees at the Amanos Mountains in the Dörtyol province near the city of Hatay (Antioch), in the south of Turkey. The trees were cut to be used as firewood and for small-scale construction applications. The sampling site was a natural Q. cerris coppice stand with approximately 60-year-old trees.
2.2. Post-harvest trituration In order to facilitate transport, the bark samples were coarsely triturated locally in a hammer-type mill designed to granulate biomass for essential oil extraction. The trituration yielded bark pieces with varying dimensions, amounting to approximately 980 kg that were collected in four industrial big bags and shipped to our laboratory in Lisbon, Portugal.
2.3. Laboratory fractioning The result of the post-harvest trituration was evaluated upon arrival by fractionation in the laboratory. The triturated bark was manually fractionated using a vibratory sieving device with sieves with the following sizes: 12 mm, 6 mm, 4 mm and 2 mm. The fraction that passed the 2-mm sieve was further screened using a vibratory sieving apparatus (Retsch AS 200basic) with the following sieves: 1 mm, 0.850 mm, 0.425 mm, 0.250 mm and 0.180 mm. After sieving, the mass retained on each sieve was weighed and the corresponding mass fraction yields were determined (Fig. 1). Three independent samples with approximately 15 kg were taken and the result is shown as the mean. The moisture content of the bark particles was determined and averaged 12%.
The triturated bark samples were further fractionated in the pilot plant for cork processing of Cincork, Portugal. The pilot plant is designed to process the present commercial cork raw-materials and includes two cutting mills, sieve separation and densimetric tables; the equipment requires a minimum raw-material feedstock of approximately 600 kg. The Q. cerris bark that underwent the pilot scale fractionation amounted to 680 kg. A stepwise fractionation process was followed: the bark samples were passed through two mills, screened by a vibratory screens into five dimensional fractions (5–7 mm, 3–5 mm, 2–3 mm, 1–2 mm, 0.5–1 mm) followed by density separation on gravimetric separators into two density fractions (low density and high density) as schematically shown in Fig. 2. Fines were separated after each mill, and the oversized particles were recirculated in the second mill. After density separation, and following the practice used for the commercial cork, the granules were labelled as good (low-density and that can be directly used by the cork industry) and weak (high-density and rejected). 2.6. Bulk density The bulk densities of the fractionated bark samples were determined for each sieve fraction using a cylindrical glass container (25 mm diameter × 29 mm height) as the ratio of mass sample in the container to its volume. 2.7. UV/VIS spectroscopy and colour measurement The reflectance spectra of the laboratory-fractionated and pilot scale-fractionated bark fractions were measured with a Minolta CM-3630 (d/0◦ ) spectrophotometer at wavelengths between 360 and 720 nm with a 10 nm spectral resolution (S¸en et al., 2012b). The CIELAB colour parameters L*, a* and b* were determined for each granulometric fraction obtained by the two fractionation methods. Total colour differences (E*) between cork-rich and phloem-rich bark granules after pilot scale fractionation were calculated using the following formula (Bekhta and Niemz, 2003; Mononen et al., 2005; Fan et al., 2010):
2
2
E = L + a + b
2 1/2
The conversion of a reflectance spectrum into an absorbance spectrum was made by applying the Kubelka-Munk equation where R is the measured reflectance value, k is the light absorption coefficient and s is the light scattering coefficient. k = (1 − R)2 /2R s
44
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49
Fig. 2. Flow sheet for the pilot scale fractionation used in Cincork for cork samples.
Difference spectra were calculated by subtracting the reflectance spectrum of the good-labelled sample from that of the corresponding weak-labelled sample (Chen et al., 2012). k k k = (lowgradecork) − (highgradecork) s s s 2.8. FTIR spectroscopy Fourier transform infrared spectroscopy (FTIR) analysis was performed using KBr pellets. Each ultra-milled sample was oven-dried at 60 ◦ C overnight. A 1.5 mg aliquot of each sample was mixed with 200 mg KBr and then milled in a mortar. The KBr pellets were prepared with a compressor Carver C40000-544. The KBr pellets were analyzed on a Bruker FT-IR spectrometer (Alpha 100382) equipped with a KBr pellet detector accessory (Alpha-T), collecting 24 scans per sample at a resolution of 4 cm−1 and wavenumber range of 4000–375 cm−1 . Data treatment was performed using OPUS 6.5 software from Bruker. In order to increase the sensitivity of the FTIR analysis, FTIR-ATR (attenuated total reflectance) method was also tested. The bark fractions obtained in the flotation experiments (supernatant and sediment) were evaluated using as a reference a pure cork fraction that was separated manually from the 2–4 mm fraction by visually sorting out cork-only granules. The samples were dried over-night at 60◦ C and grinded with a ball mill Retsch MM200 during 20 min. FTIR-ATR spectra were acquired for each sample that was dried at 60◦ C during 2 h before analyses. Spectra were recorded on a Bruker FT-IR spectrometer (Alpha 100382) with the following settings: 24 scans per sample, spectral resolution of 4 cm−1 and wavenumber range of 4000–375 cm−1 , using a diamond single reflection (Platinum-ATR) accessory and data was processed using the OPUS 6.5 software from Bruker. 3. Results and discussion 3.1. Post-harvest fractioning The preliminary coarse trituration of the Q. cerris bark yielded very different sized particles from large cm-sized pieces to fine material. The results of the granulometric separation (Table 1) showed that the major fraction corresponded to large pieces (44% of the total were retained in a 12 mm sieve) with the fine material under 2 mm representing 17%. A further separation of the <2 mm fraction showed that it was composed mainly of the 0.425–0.850 mm fraction followed by 0.250–0.425 mm and 1–2 mm fractions (Table 2).
Table 1 Mass yields of Quercus cerris bark fractions obtained with a post-harvest trituration. Fractions
Yield (%)
>12 mm 6–12 mm 4–6 mm 2–4 mm <2 mm Total
44 18 7 14 17 100
Table 2 Mass yields of the <2 mm fraction of Quercus cerris bark. Fractions
Yield (%)
1–2 mm 0.85–1 mm 0.425-0.850 mm 0.250-0.425 mm 0.180-0.250 mm <0.180 mm Total
16 9 45 18 5 6 99
The macroscopic appearance of the fractions was very different and a visual observation based on their colour differences could already show their differing contents in cork and phloem (Fig. 3). The large pieces were constituted mainly by portions of rhytidome with a high amount of cork (with its interspersed phloem) but also including strips of phloem with a long fibrous appearance (Fig. 3a). The 6–12 mm fraction contained cork granules still showing parts of the phloem, and the fibrous phloem now in thinner strips (Fig. 3b). The 4–6 mm fraction contained cork particles, although in a lesser proportion, but most granules were of phloem, including both elongated and more spherical particles (Fig. 3c). The 2–4 mm fraction had a similar aspect but with a smaller proportion of cork granules (Fig. 3d) and even less in the <2 mm fraction (Fig. 3e). The results obtained with the post-harvest trituration regarding the granulometric fractioning (Tables 1 and 2) and the visual aspect of the fractions (Fig. 3) clearly show that there is a selective fractioning and a diverse cork content in each fraction. The bulk density of the fractions may be used as an indicator of cork content since cork has a low density in the range of 0.13–0.38 g cm−3 (Oliveira et al., 2014) although it should be taken it account that bulk density is much lower than the material’s density given the intergranular empty spaces. The fraction (Table 3) with the lowest bulk density of 0.26 g cm−3 was the 4–6 mm fraction due to the higher content of cork as compared for instance with the 1–2 mm fraction with a bulk density of 0.60 g cm−3 .
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49
45
Fig. 3. Visual aspect of the Quercus cerris bark fractions with different particle dimensions: (a) >12 mm (b) 6–12 mm; (c) 4–6 mm; (d) 2–4 mm; (e) <2 mm.
Table 3 Bulk density of different Quercus cerris bark granulometric fractions after laboratory scale fractioning. Fractions
Bulk density (g cm−3 )
4–6 mm 2–4 mm 1–2 mm 0.85–1 mm 0.425–0.85 mm 0.250–0.425 mm 0.180–0.250 mm <0.180 mm
0.26 0.38 0.60 0.59 0.60 0.60 0.46 0.46
These results are explained by the structural heterogeneity of the bark (S¸en et al., 2011a) and the different mechanical properties of the tissues i.e. cork and phloem. The milling process used in this post-harvest trituration had a strong component of impact and compression stresses for the fracturing. The lignocellulosic phloemic tissues are brittle and fracture more easily, leading to smaller sized particles derived mostly from the heavily sclerified non-functional phloem and rhytidome. Also the strips of fibrous appearance correspond to phloem particles and result from the cluster arrangement of fibres (S¸en et al., 2011a). On the other side, cork does not fracture under compression and allows large deformations under stress (Gibson et al., 1981; Pereira, 2007, 2015). The cork tissue is therefore present mostly in the larger sized fractions where it acted like an impact absorber and is therefore still interspersed with the phloemic layers. The medium sized particles have a combined content of cork and phloem and the fines include mostly phloem. It is known that the bark structure influences the trituration behaviour and the particle size distribution and may be used to selectively enrich fractions in certain components (Baptista et al., 2013; Miranda et al., 2012, 2013, 2014). This selectivity is clearly favoured in the case of Q. cerris bark, because of its large proportion of cork in the thick rhytidome. Fines usually contain mechanically brittle lignified materials that are rich in extractives, while cork-enriched fractions have larger dimensions. This was also found recently in the fractionation of Pseodotsuga menziesii bark which has a structural arrangement similar to that of Q. cerris (Ferreira et al., 2015).
Fig. 4. Separation of Quercus cerris bark granules of different dimensions by water flotation into a cork floating layer and a phloem sedimenting layer.
The coarse trituration that was made here showed that it is possible to make a first step grinding as a post-harvest operation that allows separating cork-poor fractions of smaller dimensions i.e. 2–4 mm and <2 mm. The test to check if these fractions contained an amount of cork that would recommend their consideration was made by water floatation. The water flotation results for the granules below 4 mm are shown in Fig. 4. The yield obtained for the cork-enriched fraction (floating layer) is high for the 2–4 mm fraction (Fig. 3d) corresponding to 70% of the total, but it decreased to about half this value for all granulates smaller than 2 mm. The water flotation is a useful step in bark fractioning since it allows a preliminary separation of high-density phloem and lowdensity cork fractions (Pinto et al., 2009; S¸en et al., 2010; Ferreira et al., 2013). The water flotation results (Fig. 4) show that the yield obtained for the cork-enriched fraction is high for the 2–4 mm fraction (Fig. 3d) corresponding to 70% of the total, while significantly decreasing in all granulates smaller than 2 mm. Therefore, the material obtained from the post-harvest trituration may be sieved to screen out granulates below 2 mm when targeting for cork-rich fractions. This removal would reduce by 17% the transport cost of the raw-material (Table 1).
46
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49
60
Table 4 Mass yields of cork and cork-enriched fractions obtained after pilot scale trituration and fractioning of a total of 654 kg of Quercus cerris bark. Yield (%)
5–7 mm cork (Good) 3–5 mm cork (Good) Other combined cork-enriched fractions Reject (Fines) Process loss Total
6.5 1.9 18.5 60.4 12.7 100
5–7 mm 3–5 mm 2–3 mm 1–2 mm 0.5–1 mm
1-2 mm
High density
0.19 0.10 0.25 0.14 0.08
– 0.37 0.51 0.50 0.36
3-5 mm weak 2-3 mm weak
30
1-2 mm weak 0.5-1 mm weak
20
0
Wavelength (nm) Fig. 5. Variation of the spectral reflectance values between the fractionated Quercus cerris bark granules after pilot scale fractioning.
3.2. Pilot-scale fractioning
6 5
3-5 mm 2-3 mm
4
1-2 mm
∆ K/S
The pilot scale fractioning was made using the complete bark raw-material obtained after the post-harvest trituration (Fig. 2). The fines were screened out after the first and the second mills, and the separation was made by dimensions into the fractions (between 0.5 to 7 mm) used presently in the cork granulates industry, followed by a densimetric sorting into a lower density fraction (considered as high quality cork) and a higher density fraction (considered as weak quality cork). Pure cork particles with 3–5 mm and 5–7 mm dimension, as required for use in the production of agglomerated cork stoppers, could be obtained with a yield of 8.4% from the initial raw bark (Table 4). It is also noteworthy that a cork-rich fraction with 18.5% yield was also obtained that would be suitable for other types of cork agglomerates with less demanding purity requirements e.g. for surfacing or insulation agglomerates. The fine particles smaller than 0.5 mm were removed during the process. The yield of fines and rejects was 60.4%. The pilot scale fractioning using trituration and granulometric and densimetric separations could be applied to the bark of Q. cerris and allowed obtaining pure cork particles with 3–5 mm and 5–7 mm dimension (Table 4). The bulk density of these low density fractions, corresponding to values of 0.1–0.25 g cm−3 (Table 5), confirmed that they contain pure cork granules suitable for applications by the cork stoppers’ industry, as also checked by expert visual observation. Further a cork-rich fraction to be used in the production of cork agglomerates for other applications was obtained (Table 4). The bulk density of these high density fractions was in the range of 0.4–0.5 g cm−3 but the visual observation showed that very little contamination of phloemic material was present in these fractions, which therefore may be used for surfacing and insulation cork applications. The results show that the raw bark of Q. cerris yielded 26.9% of granulated fractions that may be used by the cork industry. The fines and rejects corresponded to 60.4% of the bark. This value was almost double of that reported for Q. suber cork granulation (35%) (Rives et al., 2012), indicating the original lower cork content in the Q. cerris bark. The equipment used was a pilot designed and adapted to the operational conditions of the present cork industry and to the characteristics of the Q. suber cork boards. It is therefore foreseeable
0.5-1 mm
10
Bulk density (g cm−3 ) Low density
2-3 mm
40
Table 5 Bulk density of Quercus cerris bark fractions after pilot scale fractioning classified by particle size and density class. Dimension fractions
3-5 mm
50
Reflectances (%)
Fractions
5-7 mm
0.5-1 mm
3 2 1 0 360
400
440
480
520
560
600
640
680
720
Wavelength (nm) Fig. 6. Absorption spectra k/s of cork granules of Quercus cerris of different particle size.
that optimization to other bark raw materials such as the present Q. cerris bark would improve the results. Nevertheless the present study proves that it is possible to obtain cork fractions with high purity and adequate particle size for use in high value cork products from Q. cerris bark using the current procedures and equipment of the cork industry. 3.3. UV/VIS spectroscopy and colour analysis The spectral reflectance values allowed to separate cork granules (Fig. 5). In the pilot scale fractionated samples it was possible to distinguish the cork-rich 5–7 mm, 3–5 mm and 0.5–1.0 mm fractions from the other fractions. There were differences in the absorption spectra of each granule class with the maximum absorption observed at 390, 400, 470 and 460 nm in the 0.5–1 mm, 2–3 mm, 1–2 mm and 3–5 mm granules, respectively (Fig. 6) The colour analysis showed three different groups of the granulometric classes: 3–5 mm and 5–7 mm fractions in the first group, 2–3 mm and 1–2 mm fractions in the second group and the 0.5–1.0 mm fraction in the third group that differed mainly by their lightness (L*) while yellowness (b*) and redness (a*) values were similar(Table 6). The total colour differences (E) between
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49 Table 6 CIELAB colour parameters of pilot scale fractionated Quercus cerris bark granules. Granulometric fractions
5–7 mm 3–5 mm 2–3 mm 1–2 mm 0.5–1.0 mm
L*
a*
b*
47
of Q. cerris cork are similar to those of corks from Q. suber and Pseudotsuga meziesii (Marques et al., 1996, 2006).
E*
Good
Weak
Good
Weak
Good
Weak
49.39 50.00 44.71 44.07 53.24
– 42.32 36.55 36.60 41.14
12.18 11.63 12.13 10.66 12.38
– 12.49 11.50 11.24 11.32
21.28 22.45 20.63 20.90 24.31
– 19.70 17.51 17.80 19.65
3.5. Economic considerations – 8.21 8.76 8.11 13.01
*E is the total colour difference between good and weak granules.
Table 7 CIELAB colour parameters of laboratory scale fractionated Quercus cerris bark granules. Granulometric fractions
L*
a*
b*
4–6 mm 2–4 mm 1–2 mm 0.85–1 mm 0.425–0.85 mm 0.250.0.425 mm 0.180–0.250 mm <0.180 mm
35.54 35.55 35.32 35.42 35.63 36.11 35.95 33.69
9.51 10.06 9.83 9.62 9.96 9.92 9.67 9.61
18.04 17.89 17.90 17.85 18.34 18.45 18.04 17.16
the good and weak quality fractions were similar although a higher colour difference was found for the 0.5–1.0 mm class. Colour was an indicator of differences between fractions related to the differing cork contents. Pure cork fractions have higher L* and lower b* in comparison with less pure cork fractions (Table 6) while fractions with substantial amounts of phloem have lower L*, a* and b* values (Table 7). The differences in the absorption spectra of each granule class e.g. in the maximum absorption (Fig. 6) imply different chemical compositions. The observed absorptions should be related to the suberin and lignin contents of the samples since lignin absorbs the uv/visible radiation up to 400 nm, with a peak at 280 nm (Pandey, 2005) while pure cellulose does not absorb radiation in the uv/vis region (Wondraczek et al., 2011).
3.4. FTIR observations The FTIR spectroscopy is a useful tool to monitor the chemical composition of bark (S¸en et al., 2012a,b). Fig. 7 shows the FTIR spectra of the cork-enriched granules separated by water flotation and includes also a manually separated cork sample. The bands of cork and phloem fractions were the same, only their intensity were different indicating a non-homogenous material. Nine intense bands were detected in both materials: the band at 3315 cm−1 refers to OH stretching, the bands at 2914 cm−1 and 2849 cm−1 are characteristic of symmetric and assymetric C H stretching bands of suberin, the band at 1735–1737 cm−1 is the characteristic C O band of suberin (Lopes et al., 2001). The intense bands at 1615 cm−1 , 1319 cm−1 , 777 cm−1 , and 513 cm−1 are assigned to C O stretching bands in calcium oxalate monohydrate (Conti et al., 2010; S¸en et al., 2012a,b). These bands were not observed in Q. suber cork spectrum. The intense band at 1033 cm−1 is characteristic of the C O valence vibration of polysaccharides (Colom et al., 2003; S¸en et al., 2012b) FTIR analysis may also be used as an indicator of cork content due to the differences of spectral features of cork and phloem, as given by the suberin specific bands at 2914, 2849 cm−1 and 1737 cm−1 . Low intensities of these bands are indicative of a fraction poor in suberin i.e. in cork, as it was the case in the phloem samples (sedimenting layer) after water flotation (Fig. 7). The spectral features
The usage of non-commercial corks such as Q. cerris cork by the cork industry is the first step in the valorisation of cork-rich barks. Since the cork industry is already established, with defined flow processes and operations, as well as with requirements for the specific cork products that are produced, it is necessary to separate adequate cork granulated fractions from the raw barks. These cork granulates can then be integrated to the cork granulate stream in the industry and follow the production process without any pretreatment. After evaluating the technical feasibility of obtaining cork fractions from the raw Q. cerris bark, using standard equipment from the cork industry, as it was made here, it is also important to look at the economic aspect, namely at the comparative prices of the raw-materials, as well as to its potential availability. Using the available statistical data of the Forestry Services for the region of sampling, an area of the Q. cerris dominated oak forests of over 100,000 ha, an annual wood production of 30000–50000 m3 may be estimated, allowing an estimation of the potential availability of bark of approximately 5000 m3 corresponding to over 2500 tons. The cost of Q. cerris bark is 15 D ton−1 in Turkey (GDF, 2015), a value that is lower by a 100 fac100factor to the price of raw cork in Portugal or Spain (1700 D ton−1 and 1100 D ton−1 respectively) (Sierra-Pérez et al., 2015). Considering that a 25% yield of cork fractions could be obtained from the bark, their raw price (excluding trituration and separation costs) would be 60 D ton−1 of cork fractions. Their final cost as an industrial raw-material will depend essentially on the transportation costs to the mill site. A rough estimate of locally performed trituration and separation costs, and of transport to the Portuguese cork industry was calculated as 300–400 D ton−1 . The very large price difference in relation to the Q. suber cork indicates that this new cork raw-material from Q. cerris bark should be economically interesting for the industry. It will also give a possibility of raw-material supply enlargement and diversification, which is a strategic advantage given the limited supply of Q. suber cork. It should also be noted that the use of Q. cerris bark is environmentally sustainable since it is associated to the overall forest management and tree exploitation. In fact the bark is obtained as a residue from the exploitation of Q. cerris trees that are periodically harvested from coppice and high forests directed mostly to local and regional use for energy and small scale construction wood. The bark is removed either at the forest without damaging the live trees or at processing sites. In any case the valorisation of Q. cerris bark will contribute to an overall sustainable resource management.
4. Conclusions A stepwise fractionation scheme is proposed for the raw Q. cerris bark, as shown by Fig. 3. A first step of coarse trituration near the bark collection is advised, for better subsequent handling and transport with the sieving out of particles below 2 mm. The further processing at mill site should use two cutting mills separated by the sieving out of particles below 0.5 mm, with the granulated material going to sieving and densimetric tables for fraction separation. Particles above 7 mm should undergo a second cutting pass. With such trituration and fractionation design, it was possible to obtain pure cork fractions with a yield of 8.4% that can be used for
48
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49
Cork 2-4 mm Phloem 2-4 mm Cork manual separaon 2-4 mm Cork 0.85-1.0 mm Phloem 0.85-1.0 mm Cork 1-2 mm Phloem 1-2 mm
3875
3375
2875
2375
1875
1375
875
375
Wavenumber (cm-1) Fig. 7. FTIR spectra of cork-enriched and phloem-enriched granules separated by water separation.
production of cork agglomerated stoppers and fractions with a yield of 18.5% for production of surfacing and other cork agglomerates. This is an important valorisation route for the cork-rich bark of Q. cerris. Considering the comparative cost advantage of the Q. cerris cork, its integration into the present cork industrial production lines seems economically interesting, and will also allow to enlarge the raw-material supply base. Acknowledgements This work was supported by FEDER funds through the Operational Programme for Competitiveness Factors—COMPETE under the project NewCork 38363, and is part of the activities at the Strategic Project (UID/AGR/00239/2013) of Centro de Estudos Florestais, a research unit supported by the national funding of FCT—Fundac¸ão para a Ciência e a Tecnologia. We thank Cincork—Centro de Formac¸ão Profissional da Indústria de Cortic¸a (Portugal) for the pilot-scale granulation. The first author thanks FCT for a postdoctoral scholarship and also the Turkish General Directorate of Forestry, Kahramanmaras Regional Directorate of Forestry and Mr. Sebahattin Ozdemir and Ilker Ozdemir from Ozdrog Company for their kind help in sampling and treatment of Q. cerris barks. References Baptista, I., Miranda, I., Quilhó, T., Gominho, J., Pereira, H., 2013. Characterisation and fractioning of Tectona grandis bark in view of its valorisation as a biorefinery raw-material. Ind. Crop Prod. 50, 166–175. Bekhta, P., Niemz, P., 2003. Effect of high temperature on the change in color, dimensional stability and mechanical properties of spruce wood. Holzforschung 57 (5), 539–546. Chen, Y., Fan, Y., Gao, J., Stark, N.M., 2012. The effect of heat treatment on the chemical and color change of black locust (Robinia pseudoacacia) wood flour. Bioresources 7 (1), 1157–1170. Colom, X., Carrillo, F., Nogués, F., Garriga, P., 2003. Structural analysis of photodegraded wood by means of FTIR spectroscopy. Polym. Degrad. Stabil. 80 (3), 543–549. Conti, C., Brambilla, L., Colombo, C., Dellasega, D., Gatta, G.D., Realini, M., Zerbi, G., 2010. Stability and transformation mechanism of weddellite nanocrystals studied by X-ray diffraction and infrared spectroscopy. Phys. Chem. Chem. Phys. 12 (43), 14560–14566. Fan, Y., Gao, J., Chen, Y., 2010. Colour responses of black locust (Robinia pseudoacacia L.) to solvent extraction and heat treatment. Wood Sci. Technol. 44 (4), 667–678. Ferreira, R., Garcia, H., Sousa, A.F., Freire, C.S., Silvestre, A.J., Rebelo, L.P.N., Pereira, C.S., 2013. Isolation of suberin from birch outer bark and cork using ionic liquids: a new source of macromonomers. Ind. Crop Prod. 44, 520–527.
Ferreira, J.P., Miranda, I., Gominho, J., Pereira, H., 2015. Selective fractioning of Pseudotsuga menziesii bark and chemical characterization in view of an integrated valorization. Ind Crop Prod. 74, 998–1007. General Directory of Forestry (GDF) of Turkey, 2015. Non-wood forest products price plan 2014 (in Turkish). Retrieved from http://www.ogm.gov.tr/ ekutuphane/Dokumanlar/2014%20yili%20Tarife%20Bedeli%20Cetveli.pdf. Gibson, L. J., Easterling, K. E., Ashby, M. F. 1981. The structure and mechanics of cork. In Proceedings of the Royal Society of London A: Mathematical, Physical and Engineering Sciences (Vol. 377, No. 1769, pp. 99-117). The Royal Society. Kasapligil, B., 1981. Past and present oaks of Turkey. I. Phytologia 49, 95–146. Lopes, M.H., Barros, A.S., Pascoal Neto, C., Rutledge, D., Delgadillo, I., Gil, A.M., 2001. Variability of cork from Portuguese Quercus suber studied by solid-state 13C-NMR and FTIR spectroscopies. Biopolymers 62 (5), 268–277. Marques, A.V., Pereira, H., Meier, D., Faix, O., 1996. Isolation and characterization of a guaiacyl lignin from saponified cork of Quercus suber L. Holzforschung 50 (5), 393–400. Marques, A.V., Pereira, H., Rodrigues, J., Meier, D., Faix, O., 2006. Isolation and comparative characterization of a Björkman lignin from the saponified cork of Douglas-fir bark. J. Anal. Appl. Pyrol. 77 (2), 169–176. Miranda, I., Gominho, J., Mirra, I., Pereira, H., 2012. Chemical characterization of barks from Picea abies and Pinus sylvestris after fractioning into different particle sizes. Ind. Crop Prod. 36 (1), 395–400. Miranda, I., Gominho, J., Mirra, I., Pereira, H., 2013. Fractioning and chemical characterization of barks of Betula pendula and Eucalyptus globulus. Ind. Crop Prod. 41, 299–305. Miranda, I., Gominho, J., Ferreira-Dias, S., Pereira, H., 2014. Pattern recognition as a tool to discriminate softwood and hardwood bark fractions with different particle size. Wood Sci. Technol. 48 (6), 1197–1211. Mononen, K., Jääskeläinen, A.S., Alvila, L., Pakkanen, T.T., Vuorinen, T., 2005. Chemical changes in silver birch (Betula pendula Roth) wood caused by hydrogen peroxide bleaching and monitored by color measurement (CIELab) and UV-Vis, FTIR and UVRR spectroscopy. Holzforschung 59 (4), 381–388. Oliveira, V., Rosa, M.E., Pereira, H., 2014. Variability of the compression properties of cork. Wood Sci. Technol. 48 (5), 937–948. Pandey, K.K., 2005. Study of the effect of photo-irradiation on the surface chemistry of wood. Polym. Degrad. Stabil. 90 (1), 9–20. Pereira, H., Tomé, M., 2004. Cork oak. In: Encyclopedia of Forest Sciences. Elsevier Oxford, pp. 613–620. Pereira, H., 2007. Cork: Biology, Production and Uses: Biology, Production and Uses. Elsevier. Pereira, H., 2015. The rationale behind cork properties: a review of structure and chemistry. Bioresources 10 (3), 6207–6229. Pinto, P.C., Sousa, A.F., Silvestre, A.J., Neto, C.P., Gandini, A., Eckerman, C., Holmbom, B., 2009. Quercus suber and Betula pendula outer barks as renewable sources of oleochemicals: a comparative study. Ind. Crop Prod. 29 (1), 126–132. Rives, J., Fernandez-Rodriguez, I., Gabarrell, X., Rieradevall, J., 2012. Environmental analysis of cork granulate production in Catalonia–Northern Spain. Resour. Conserv. Recycl. 58, 132–142. S¸en, A., Miranda, I., Santos, S., Grac¸a, J., Pereira, H., 2010. The chemical composition of cork and phloem in the rhytidome of Quercus cerris bark. Ind. Crop Prod. 31 (2), 417–422. S¸en, A., Quilho, T., Pereira, H., 2011a. Bark anatomy of Quercus cerris L. var. cerris from Turkey. Turk. J. Bot. 35, 45–55.
A. S¸en et al. / Industrial Crops and Products 92 (2016) 42–49 S¸en, A., Quilhó, T., Pereira, H., 2011b. The cellular structure of cork from Quercus cerris var. cerris bark in a materials’ perspective. Ind. Crop Prod. 34 (1), 929–936. S¸en, A., Miranda, I., Pereira, H., 2012a. Temperature-induced structural and chemical changes in cork from Quercus cerris. Ind. Crop Prod. 37 (1), 508–513. S¸en, A., Marques, A.V., Gominho, J., Pereira, H., 2012b. Study of thermochemical treatments of cork in the 150–400 ◦ C range using colour analysis and FTIR spectroscopy. Ind. Crop Prod. 38, 132–138.
S¸en, A., Van den Bulcke, J., Defoirdt, N., Van Acker, J., Pereira, H., 2014. Thermal behaviour of cork and cork components. Thermochim. Acta 582, 94–100. Sierra-Pérez, J., Boschmonart-Rives, J., Gabarrell, X., 2015. Production and trade analysis in the Iberian cork sector: economic characterization of a forest industry. Resour. Conserv. Recycl. 98, 55–66. Wondraczek, H., Kotiaho, A., Fardim, P., Heinze, T., 2011. Photoactive polysaccharides. Carbohydr. Polym. 83 (3), 1048–1061.
49