ACTA FACULTATIS XYLOLOGIAE ZVOLEN

TECHNICKÁ UNIVERZITA VO ZVOLENE DREVÁRSKA FAKULTA

ACTA FACULTATIS XYLOLOGIAE ZVOLEN

VEDECKÝ ČASOPIS SCIENTIFIC JOURNAL

61 1/2019

Scientific journal Acta Facultatis Xylologiae Zvolen publishes peer-reviewed scientific papers covering the fields of wood: structure and properties, wood processing, machining and drying, wood modification and preservation, thermal stability, burning and fire protection of lignocelluloses materials, furniture design and construction, wooden constructions, economics and management in wood processing industry.

The journal is a platform for presenting reports and reviews of books of domestic and foreign authors.

Vedecký časopis Acta Facultatis Xylologiae Zvolen uverejňuje pôvodné recenzované vedecké práce z oblastí: štruktúra a vlastnosti dreva, procesy spracovania, obrábania, sušenia, modifikácie a ochrany dreva, termickej stability, horenia a protipožiarnej ochrany lignocelulózových materiálov, konštrukcie a dizajnu nábytku, drevených stavebných konštrukcií, ekonomiky a manažmentu drevospracujúceho priemyslu. Poskytuje priestor aj na prezentáciu názorov formou správ a recenzií kníh domácich a zahraničných autorov.

VEDECKÝ ČASOPIS DREVÁRSKEJ FAKULTY, TECHNICKEJ UNIVERZITY VO ZVOLENE 61 1/2019 SCIENTIFIC JOURNAL OF THE FACULTY OF WOOD SCIENCES AND TECHNOLOGY, TECHNICAL UNIVERSITY IN ZVOLEN 61 1/2019 Redakcia (Publisher and Editor’s Office): Drevárska fakulta (Faculty of Wood Sciences and Technology) T. G. Masaryka 24, SK-960 53 Zvolen, Slovakia Redakčná rada (Editorial Board): Predseda (Chairman): Vedecký redaktor (Editor-in-Chief): Členovia (Members):

Jazykový editor (Proofreader): Technický redaktor (Copy Editor):

prof. Ing. Ján Sedliačik, PhD prof. Ing. Ladislav Dzurenda, PhD. prof. RNDr. František Kačík, PhD. prof. RNDr. Danica Kačíková, PhD. prof. Ing. Jozef Kúdela, CSc. prof. Ing. Ladislav Reinprecht, CSc. prof. Ing. Jozef Štefko. CSc. doc. Ing. Pavol Joščák, CSc. doc. Ing. Hubert Paluš, PhD. Mgr. Žaneta Balážová, PhD. Antónia Malenká

Medzinárodný poradný zbor (International Advisory Editorial Board): Pavlo Bekhta (UA), Nencho Deliiski (BG), Vlado Goglia (HR), Denis Jelačić (HR), Bohumil Kasal (USA), Wojciech Lis (PL), Remy Marchal (FR), Miloslav Milichovský (CZ), Róbert Németh (HU), Peter Niemz (CH), Kazimierz A. Orlowski (PL), Franc Pohleven (SI), František Potůček (CZ), Włodzimierz Prądzyński (PL), Alfréd Teischinger (AT), Jerzy Smardzewski (PL), Mikuláš Šupín (SK), Richard P. Vlosky (USA), Rupert Wimmer (AT) Vydala (Published by): Technická univerzita vo Zvolene, T. G. Masaryka 2117/24, 960 53 Zvolen, IČO 00397440, 2019 Náklad (Edition) 150 výtlačkov, Rozsah (Pages) 165 strán, 13,42 AH, 13,52 VH Tlač (Printed by): Vydavateľstvo Technickej univerzity vo Zvolene Vydanie I. – apríl 2019 Periodikum s periodicitou dvakrát ročne Evidenčné číslo: 3860/09 Časopis Acta Facultatis Xylologiae Zvolen je registrovaný v databáze (Indexed in): Web of Science, SCOPUS, ProQuest, AGRICOLA, Russian Science Citation Index Všetky práva vyhradené. Nijaká časť textu ani ilustrácie nemôžu byť použité na ďalšie šírenie akoukoľvek formou bez predchádzajúceho súhlasu autorov alebo vydavateľa.

© Copyright by Technical University in Zvolen, Slovak Republic.

ISSN 1336–3824

CONTENTS

01. JOZEF KÚDELA – LADISLAV REINPRECHT – ZUZANA VIDHOLDOVÁ – MICHAL ANDREJKO: SURFACE PROPERTIES OF BEECH WOOD MODIFIED BY CO2 LASER ..................................................... 5 02. ANTON GEFFERT  EVA VÝBOHOVÁ  JARMILA GEFFERTOVÁ: CHANGES IN THE CHEMICAL COMPOSITION OF OAK WOOD DUE TO STEAMING .................................................................................................. 19 03 ZUZANA VIDHOLDOVÁ – ANNA SANDAK – JAKUB SANDAK: ASSESSMENT OF THE CHEMICAL CHANGE IN HEAT TREATED PINE WOOD BY NEAR INFRARED SPECTROSCOPY .......................................... 31 04. SERGIY KULMAN – LIUDMYLA BOIKO – DIANA HAMÁRY GUROVÁ – JÁN SEDLIAČIK: THE EFFECT OF TEMPERATURE AND MOISTURE CHANGES ON MODULUS OF ELASTICITY AND MODULUS OF RUPTURE OF PARTICLEBOARD ........................................ 43 05. IVAN KLEMENT  MIROSLAV UHRÍN  TATIANA VILKOVSKÁ: DRYING THE SPRUCE (PICEA ABIES L. KARST.) COMPRESSION WOOD .................................................................................................................................. 53 06. NENCHO DELIISKI – LADISLAV DZURENDA – DIMITAR ANGELSKI – NATALIA TUMBARKOVA: COMPUTING THE ENERGY FOR WARMING UP THE PRISMS FOR VENEER PRODUCTION DURING AUTOCLAVE STEAMING WITH A LIMITED POWER OF THE HEAT GENERATOR ........................................................... 63 07. RICHARD KMINIAK – ADRIAN BANSKI: GRANULOMETRIC ANALYSIS OF CHIPS FROM BEECH, OAK AND SPRUCE WOODTURNING BLANKS PRODUCED IN THE MILLING PROCESS USING 5-AXIAL CNC MACHINING CENTER ........................................... 75 08. KAZIMIERZ A. ORŁOWSKI – DANIEL CHUCHAŁA – TOMASZ MUZIŃSKI – JACEK BARAŃSKI – ADRIÁN BANSKI – TOMASZ ROGOZIŃSKI: THE EFFECT OF WOOD DRYING METHOD ON THE GRANULARITY OF SAWDUST OBTAINED DURING THE SAWING PROCESS USING THE FRAME SAWING MACHINE .................................. 83 09. VALENTIN ATANASOV – GEORGI KOVATCHEV: DETERMINATION OF THE CUTTING POWER DURING MILLING OF WOOD-BASED MATERIALS .......................................................................... 93 10. PAVLO LYUTYY – PAVLO BEKHTA – GALYNA ORTYNSKA – JÁN SEDLIAČIK: THE PROPERTIES OF REINFORCED LIGHTWEIGHT FLAT PRESSED WOOD PLASTIC COMPOSITES ........................................ 103

11. BARBARA FALATOVÁ  MARTA FERREIRO-GONZÁLEZ  DANICA KAČÍKOVÁ  ŠTEFAN GALLA  MIGUEL PALMA  CARMELO G. BARROSO: CHEMOMETRIC TOOLS USED IN THE PROCESS OF FIRE INVESTIGATION ............................................................ 111 12. ŠTEFAN BARCÍK  SERGEY UGRYUMOV  EVGENY RAZUMOV  RUSLAN SAFIN: STUDIES OF COMPONENT INTERCONNECTION IN A PLYWOOD STRUCTURE WITH INTERNAL LAYERS OF VENEER CHIPS .................................................................................................................. 121 13. DENISA LIZOŇOVÁ  ZUZANA TONČÍKOVÁ: EXPLORING THE APPLICATION OF NATURE-INSPIRED GEOMETRIC PRINCIPLES WHEN DESIGNING FURNITURE AND INTERIOR EQUIPMENT ............. 131 14. ROMAN NÔTA: THE IMPACT OF THE SPACER ON THE INTERIOR SURFACE TEMPERATURE IN THE DETAIL OF WOOD WINDOW GLAZING ............................................................................................................ 147 15. HUBERT PALUŠ – JÁN PAROBEK – MICHAL DZIAN – SAMUEL ŠIMO-SVRČEK – MARTINA KRAHULCOVÁ: HOW COMPANIES IN THE WOOD SUPPLY CHAIN PERCEIVE THE FOREST CERTIFICATION ............................................................................................... 155

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 5−18, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.01

SURFACE PROPERTIES OF BEECH WOOD MODIFIED

BY CO2 LASER Jozef Kúdela – Ladislav Reinprecht – Zuzana Vidholdová – Michal Andrejko ABSTRACT The paper is focused on the analysis of surface properties of beech wood modified by the CO2 laser. Colour and morphological changes in the beech wood surface subjected to laser treatment under different irradiation doses were monitored. Experiments were carried out to investigate the wetting surfaces treated with standard liquids. Free surface energy values were also determined for these surfaces. The results showed that increasing the irradiation dose over a range of 075 J∙cm² induced a significant increase in the total colour difference (E* = 45). This discolouration was caused by a significant decrease in brightness L* from 73 to 30 and also by the decreased colour coordinates a* a b*. Significant morphological changes were not detected except at the highest irradiation dose with the entire surface layer being carbonised and the surface roughness increased. An increase in irradiation doses resulted in a decrease in the surface free energy values of beech wood.

However, the wood surface modified this way still complied with the requirements for surface treatment with coating materials and for gluing.

Key words: beech wood, CO2 laser, surface properties, colour, roughness, waviness, wetting, surface free energy

INTRODUCTION In recent decades, there has been considerable interest in exploring the surface properties of wood in the context of its purpose-oriented modification. Various types of wood surface pre-treatments (thermal, thermo-hydro-mechanic, CO2 laser, plasma, nanoparticles and others) modify, in different ways, the wood structure and properties (VARGA and VAN DER ZEE 2008, ČERMÁK and DEJMAL 2013, KAČÍK and KUBOVSKÝ 2011, DZURENDA 2014, KÚDELA and ANDOR 2018, REINPRECHT et al. 2018), This may have considerable impacts on the quality of the surface treatment and also on the quality of the glued joints.

Electromagnetic radiation of various wavelengths induces progressive degradation of wood surfaces, which in the initial phases is manifested through discolouration. In this process, important factors are the form of the radiation (CO2 laser beam, UV, IR) and the wood species (PANDEY and VUORINEN 2008, KUBOVSKÝ and KAČÍK 2013, ZBOROWSKA et al. 2016, REINPRECHT 2016, KUBOVSKÝ et al. 2018, KÚDELA et al. 2018a). The purposeoriented modification of wood surface properties induced with CO2 laser beam, infrared (IR) and ultraviolet (UV) radiation is often performed based exclusively on empirical experience.

5

The technology of wood surface treatment with a laser beam, however, allows us to control the amount of energy supplied to the wood surface according to the required purposeoriented impact on the wood surface structure. The amount of energy delivered onto the wood surface with a laser beam and the conversion of this energy into heat depend on the laser's power, the speed of the laser head's movement, the focal distance and the raster density (GURAU et al. 2018, LI et al. 2018, KÚDELA et al. 2018a). The energy absorbed into the main wood components is reflected in changes to these components in colour, morphology and water absorbance/repellence of the wood surface (WUST et al. 2005, ALIGIZAKI et al.

2008, KAČÍK and KUBOVSKÝ 2011, HALLER et al. 2014, VIDHOLDOVÁ et al. 2017, GURAU et al. 2018, LI et al. 2018, SIKORA et al. 2018, KÚDELA et al. 2018a).

KAČÍK and KUBOVSKÝ (2011) demonstrated a drop in polysaccharide amounts in wood irradiated with a CO2 laser, depending on the amount of energy supplied. The degradation was especially evident in hemicelluloses and in part of the amorphous cellulose fraction. There were also changes to the lignin structure. WUST et al. (2005) also observed changes in the cellulose, hemicellulose and lignin distribution patterns across cell walls.

These authors suggested intervals for irradiation parameters that will guarantee melting without pyrolysis. The results of the X-ray photo-electron spectroscopy (XPS) indicated an increased number of non-polar bonds C–C and C–H, with C–O bonds maintained without change (DOLAN 2014). Besides irradiation parameters, the degradation of the main wood components also significantly depends on the wood species (WUST et al. 2005, KAČÍK and KUBOVSKÝ 2011, HALLER et al. 2014, DOLAN 2014).

These changes in the chemical structure of wood irradiated with different electromagnetic radiation types cause the wood to darken. This is especially typical for the light-coloured wood species. The changes in the wood surface chemistry are also reflected in the changes to the wood surface morphology. The results of microscopic observations point out (HALLER et al. 2014, DOLAN 2014) that treating the wood surface with a laser beam may make the surface smoother because the cells melt down to a depth of several micrometres without carbonisation. Carving a wood surface with a laser may lead to an opposite effect (GURAU et al. 2018). The form of electromagnetic radiation delivered onto the wood surface has considerable influence.

Energy supplied in various forms also affects the wood surface wettability. A liquid's capacity to wet a solid surface is assessed based on the contact angle value. The contact angle is an important indicator in predicting adhesion strength of glues and coating materials and in predicting the effectiveness of wood thermal and chemical modification. Contact angle values measured at the interphase with liquid standards are also a point for determining the thermodynamical characteristics of wood surface properties  surface free energy and its components (BLANCHARD et al. 2009, WANG and PIAO 2011, PETRIČ and OVEN 2015, HUBBE et al. 2015, LASKOWSKA and SOBCZAK 2018, JANKOWSKA et al. 2018) Contact angle values measured after different wood surface modifications are also indicative for prediction of whether the modified wood surface will be more or less waterabsorbent than the original.

HALLER et al. (2014) give an example of pine wood surface melt under effects of CO2 laser irradiation without carbonisation (treatment temperature below 200°C) and with worse water absorbance compared to the original, untreated surface (HALLER et al. 2014). The melt layer, which was several micrometres thick, substantially enhanced the surface water repellence, which was evident on the contact angle values exceeding 90° and on the lower rate of the drop soaking into the wood. DOLAN (2014) did not observe lower wetting performance for irradiated surfaces and even obtained some opposite results. Moreover, the laser-treated wood did not show significant changes in its surface energy. The total surface energy was low, with dominant disperse component. The polar (acid/base) ratio was 6

significantly reduced compared to the referential specimens. The Lewis base parameter of the acid/base component of surface free energy remained relatively high, while the Lewis acid parameter was lowered significantly, close to the zero value. The laser radiation parameters used in this case were not the same as those used by HALLER et al. (2014). The results of these two authors, however, point out that the purpose-oriented surface treatment can affect the surface water absorbance or repellence and the bond creation between the wood and film-forming material.

The aim of this study has been to investigate the selected surface properties of CO2 laser modified beech wood – discolouration, roughness variation, wettability and surface free energy.

MATERIALS AND METHODS Wood surface modification with CO2 laser The experimental measurements were carried out on beech wood test specimens with dimensions of 50  20  5 mm (Fig. 1a). The specimens were irradiated with a power laser LCS 400 (maximum power 400 W), following the methods designed by VIDHOLDOVÁ et al.

(2017). The specimen surface was situated 127 mm under the focal point of focusing lens (the same distance for all specimens). Under this setting, the laser beam diameter on the specimen surface was 10 mm, with an effective power of 45 W. The power was measured with an appliance Coherent Radiation Model 201. The laser beam was perpendicular-oriented to the specimen surface, and the laser head was moving along the surface at a determined speed.

To obtain uniform irradiation across the whole surface, the head was driven three times, parallel, with irradiated bands overlapping appropriately (Fig. 1b).

Fig. 1 Test specimen (a) and irradiation with a CO2 laser beam (b).

There were eight testing groups, each consisting of five specimens. These groups were irradiated with different values of irradiation dose H. In addition, there was a group of referential specimens for comparison (Table 1). The irradiation dose H was controlled by adjusting the moving speed of the laser head from 6 to 58 mm·s−1. The value was calculated according to the equation

7

đ??ť=

đ?‘ƒe ∙đ?œ? đ??´

=

đ?‘ƒe ∙đ?‘Ľ

(1)

đ??´Âˇđ?‘Ł

where Pe is the laser beam power on the specimen surface, ď ´ is irradiation time during one head passage (the ratio between the specimen dimension x and rate v), and A is the area irradiated during one passage. Table 1 shows irradiation dose values for different rates of laser beam passage.

Tab. 1 Irradiation dose for specific head movement rate.

Specimen Scanning rate v [mm¡s1] Irradiation dose H [J¡cm2]

Ref. 0 0

A 58 7.8

B 42 10.7

C 36 12.5

D 30 15.0

E 24 18.8

F 18 25.5

G 12 37.5

H 6 75.0

Wood surface morphology The changes to wood surface morphology induced with laser treatment were assessed based on roughness parameters Ra, Rz, and RSm and on the wood surface inspected with light microscopy. Roughness was measured on the radial surfaces parallel with and perpendicular to the grain with a profile-meter Surfcom 130A. The surface roughness was evaluated in this way: the waviness was filtered from the profile, and the resulting curve was transferred onto the baseline. The overall measured length consisted of the initial segment, five sampling length segments lr (cutoff Îťc), and the final segment lp. The first and last segment served to eliminate the possible vibrations associated with starting and stopping the measuring. The sampling length values ranged within 0.0258 mm in accordance with the preliminary measured values of the roughness parameters Ra and Rz. Wood surface colour In the test specimens, both the irradiated and referential ones, their colour space CIE L*a*b* was measured with a spectro-photometer Spectro-guide 45/0 gloss by BYK–GARDNER GmbH. The colour was measured at six spots on each specimen (uniformly distributed on the measured surface). The discolouration extent was expressed through the total colour difference ď „E*, calculated according to the equation: ∆đ??¸ ∗ = √∆đ??żâˆ—2 + ∆đ?‘Žâˆ—2 + ∆đ?‘? ∗2 ,

(2)

were ∆đ??żâˆ— = đ??żâˆ—2 − đ??żâˆ—1 ∆đ?‘Žâˆ— = đ?‘Ž2∗ − đ?‘Ž1∗ ∆đ?‘? ∗ = đ?‘?2∗ − đ?‘?1∗

(3) (4) (5)

where index 1 means the so-called referential value, measured on the non-irradiated wood surface, and index 2 indicates the coordinate value after the irradiation.

Wood surface wetting with liquids and wood surface free energy assessment The wood wetting process associated with the contact angle measurements up to the complete drop soaking into the substrate was performed with a goniometer Krßss DSA30 Standard. Two testing liquids differing in polarity were used – redistilled water and diiodomethane. Using these two liquids follows KÚDELA (2014). Diiodomethane is a nonpolar liquid with a non-polar component of surface free energy higher than the disperse component of wood. Redistilled water is a polar-apolar liquid with a polar component of surface free energy that is higher than the polar component of wood. Table 2 shows the testing liquids' parameters.

8

Tab. 2 Free surface energy of testing liquids L and its components.

Testing liquid

Liquid character

water diiodomethane

polar non-polar

L

LD

LP

+

--

25.5 0.0

25.5 0.0

2

[mJ·mm ] 72.8 50.8

21.8 50.8

51.0 0.0

From the moment of contact of the testing drop (volume 0.0018 ml) with the wood surface, the wood wetting and drop spreading along the fibre direction were inspected. The history of the drop shape, from the first contact up to the complete soaking, was recorded with a camera. The scanning frequency was set according to the wetting interval.

The drop shape was analysed, and the contact angle was determined based on two methods: drop perimeter (circle method) and drop height and diameter (height-width method, Fig. 2). a)

b)

Fig. 2 Determining the contact angle value by the circle method (a) and by the height-width method (b).

The contact angle values θ0 were measured at the beginning of the wetting process, meaning at the moment of the first contact between the drop and substrate. The drop width values d (drop diameter) were used for identifying the moment of reversion of the acceding contact angle into the receding one. The drop's contact angle at this moment was considered the "equilibrium" contact angle – e. The contact angle values 0 and e obtained in this way were used for calculating the contact angle value w corresponding to an ideal smooth surface, according to the methods described in LIPTÁKOVÁ and KÚDELA (1994).

Subsequently, this angle was used to calculate surface free energy and its components. The contact angle was measured at six different measuring spots on each specimen.

As the wood was wetted with two different liquids, the wood surface free energy was determined separately for wetting with water and with diiodomethane, according to the adjusted equation originally proposed by NEUMANN et al. (1974): cos  

(0,0137   S  2,00 )   S   L   L

(6)

 L  (0,0137   S   L  1)

with the disperse and polar components Sd and Sp calculated in the following way by KLOUBEK (1974).

1  cos   1  cos  2 p )  L  S ( ) 2 L 2

(7)

1  cos   S 1  cos  2 d ) L  ( ) .

2 L 2

(8)

 S d   Ld  ( S p  Lp (

The resulting surface energy of beech specimens irradiated with different radiation doses was determined as the sum of the energy polar component obtained with water and the disperse component obtained with diiodomethane.

9

RESULTS AND DISCUSSION Irradiation doses ranging from 7.8 to 75.0 Jcm2 caused major changes to the studied wood surface properties. The first one, detected visually, was surface discolouration. The results of the one-way variation analysis have unambiguously confirmed a significant influence of radiation intensity on the coordinate values L*, a*, and b*. Table 3 lists the basic statistical characteristics for these coordinates. The original beech colour coordinates values were from the interval reported by BABIAK et al. (2004) and VIDHOLDOVÁ et al. (2017). The average lightness in the referential specimens L* was 73.4, and this value decreased significantly with increasing radiation dose. The lowest lightness value (30) was measured in specimens irradiated with 37.5 Jcm2. The dose of 70 Jcm2 did not further decrease lightness, and even an opposite trend was observable. The surface of the specimens in this group was entirely carbonised, and the carbonised layer displayed a tendency for light reflexion.

With irradiation dose increase as great as 12.5 Jcm2, the values of coordinates a* and b* increased moderately, shifting towards red and yellow, which caused gradual colouring of the wood to dark brown. Further irradiation dose increases initiated a reverse trend. The values of both coordinates decreased noticeably, close to zero at the highest irradiation dose, with the coordinates shifting towards green and blue. The wood surface colour was modified from dark brown to black. Our results agree with those obtained by VIDHOLDOVÁ et al. (2017). This discolouration is documented in Table 3. The differences in the coordinates L*, a*, and b* and the total colour difference E* are illustrated in Fig. 3. The total colour difference ranged from the second to sixth degree of the six-degree colour evaluation scale designed by ALLEGRETTI et al. (2009). This means the discolouration ranges from just visually observable up to a completely new colour.

Qualitatively similar results for beech wood were also obtained by KAČÍK and KUBOVSKÝ (2011) and KÚDELA et al. (2018). In our case, there were more conspicuous changes in the colour coordinates a* and b* corresponding to the identical irradiation dose.

These deviation, however, may be due to the different directions of the laser head displacement parallel to the grain in our case, perpendicular to the grain in the literature referred to above and also due to the shorter distance between the specimens and the focusing lens in our case.

The figure in Table 3 illustrates that the wood surface discolouration was not homogeneous. We may suppose two reasons: first, uniform irradiation intensity was not possible across the whole specimen width, and second, the beech wood surface structure is heterogeneous. The irradiated surfaces were radial, with alternating early and late wood.

Despise being a diffuse-porous wood species, beech wood displays certain differences in structure and properties between early and late wood that play a major role in creating inhomogeneous colour patterns.

The microscopic observation revealed that the colour inhomogeneity was, apart from alternating early and late wood, also caused by pith rays consisting of thin parenchyma cells and representing in beech wood a 2027% share (KÚDELA and ČUNDERLÍK 2012). The pith rays were intensively darkening with increasing irradiation doses. In the case of group E, the pith rays exhibited marked dark spots and even carbonisation (Fig. 4).

The beech wood surface modification induced with CO2 laser treatment was manifested not only through discolouration but also through morphological changes evaluated based on the specified roughness parameters. Table 4 lists the basic statistical characteristics of the roughness parameters Ra, Rz, and RSm, measured parallel with and perpendicular to the grain, for all the irradiation modes.

10

Tab. 3 Beech specimen surface colour and basic statistical characteristics for the coordinates L*, a* and b*, after CO2 laser treatment with different irradiation doses. (number of the measurements n= 30).

Colour coordinates

Irradiation dose H (Jcm2) 7.8

10.7

12.5

15.0

18.8

25.5

37.5

75.0

Ref.

A

B

C

D

E

F

G

H

73.39

71.88

66.28

65.36

61.28

53.32

41.48

29.71

31.98

1.53

2.26

1.66

1.95

4.50

2.63

2.55

0.77

0.77

7.44

7.45

8.59

8.43

8.11

7.78

6.60

2.48

0.45

0.62

0.85

0.45

0.50

0.45

0.26

0.46

0.42

0.14

18.42

17.67

19.04

18.26

17.29

15.74

12.70

4.20

1.84

1.23 1.28 s  average, s  standard deviation

0.79

0.84

1.15

0.77

1.09

0.75

0.54

Basic statistical characteristics

0

L*

s

a*

s

b*

ΔL** ΔL

Δa** Δa

Δb* Δb*

50

0

40

-10

ΔE*

ΔL*, Δa*, Δb*

10

-20 -30

30 20 10

-40

0

-50 0

10

20

30

H

40

50

60

70

0

80

[J·cm-2]

10

20

30

H

40

50

60

70

80

[J·cm-2]

Fig. 3 Beech wood discolouration induced with different irradiation doses.

All the roughness parameters displayed high values and high variability. The roughness of our referential specimens was significantly higher than the roughness of milled or sanded wood as reported in KÚDELA et al. (2018b). The specimens discussed in this paper were cut with a fine circular saw, without any subsequent surface machining, which resulted in this difference.

The impacts of irradiation mode and anatomical direction were evaluated with the aid of a two-way variance analysis. The results have confirmed that both factors significantly affected all the roughness parameters. The roughness perpendicular to the grain was significantly higher than that parallel to the grain, but these differences were smaller than the analogical differences obtained for sanded or planed surfaces. The mechanical machining-sawing partially eliminated these differences.

11

Fig. 4 Beech wood surface discolouration and altered morphology resulting from CO2 laser irradiation with different amounts of supplied energy.

Table 4 indicates that the values of roughness parameters did not vary with varying irradiation doses in groups AG. The laser beam seemed to have smoothing effects because the fibres released during sawing were removed and the surface cells were melted. The results demonstrate, however, that this phenomenon was in many cases masked by the mechanical pre-treatment of the specimen surface. This is true for all roughness parameters in both anatomical directions. The maximum irradiation dose, however, enhanced roughness in both directions because the specimen surface was carbonised and cracked (Fig. 4).

In terms of roughness, in case of a natural surface, it is also necessary to include the technical parameters of the cutting tool used for machining the surface before CO2 laser treatment and the technological parameters of this laser treatment (CZANADY and MAGOSS 2011, GURAU 2013, FOTIN et al. 2013, KÚDELA et al. 2018a, b and others). The results of these authors as well as our results show that this process is intricate because it involves interactions between the wood surface and the mechanical cutting tool and the electromagnetic radiation within IR range. The wetting process in CO2 laser-modified wood was evaluated through contact angles 0, e, and w. The results revealed that the changes to the beech wood surface morphology and chemistry induced by CO2 laser treatment were also present where the wood surface was wetted with standard liquids. The water drop applied onto the wood surface was spreading continually over the surface, and, at the same time, it was soaking into the substrate.

12

Roughness parameters

Basic statistical character.

Tab. 4 Basic statistical characteristics of beech wood surface roughness parameters after treatment with a CO2 laser with different irradiation doses. (number of the measured segments n = 10).

Ra

[m]

Irradiation dose H (Jcm2)

[m] [m]

10.7

12.5

15

18.8

25.5

37.5

75

Ref.

A

B

C

D

E

F

G

H

8.231

7.578

7.222

8.261

7.397

7.276

8.168

7.093

11.867

1.297

1.750

2.199

1.821

1.192

2.054

1.388

0.787

3.132

52.729

51.711

48.188

56.680

48.279

44.260

59.263

45.318

75.525

8.905

14.130

17.030

13.970

11.343

12.378

13.315

8.387

26.443

800.279

804.770

717.811

723.610

734.087

37.897

739.712

661.479

627.802

94.588

168.755

139.810

144.987

117.519

104.693

243.112

157.336

158.642

s [m] RSm

7.8

Parallel to grain

s [m] Rz

0

s [m]

Perpendicular to grain 10.834

9.592

10.257

10.372

9.134

9.756

10.260

10.378

20.385

1.641

1.891

1.319

1.738

1.761

1.758

1.387

1.659

4.395

78.951

66.190

70.333

73.450

66.531

66.893

74.251

68.599

126.420

8.989

7.883

10.504

8.904

11.111

14.248

11.245

9.627

22.570

414.860

426.991

421.662

385.777

438.277

457.596

435.423

421.507

530.772

56.139

73.941

61.517

37.822

11.243

142.841

105.182

85.636

131.627

[m]

Ra

s [m] [m]

Rz

s [m] [m]

RSm

s [m]

 average, s  standard deviation

The average time necessary to reach the equilibrium te during wetting of beech wood radial surface with water ranged from 12 to 43 seconds for all variants investigated. The only exception was group H, with an average time te of 1.5 sec. The beech wood surface treatment with CO2 laser also significantly affected the contacts angle values θ0, θe, and θw. Fig. 5 illustrates their dependence on the irradiation intensity.

In all test groups, with the exception of H, the average values of the contact angle θ0 ranged from 60° to 65°. Despite significant differences detected between certain diameters, no dependence of the contact angle 0 values has been confirmed. Such dependence was found for the contact angles e and w. Fig. 5 shows that these two contact angles were considerably lower than 0 and that they increased proportionally with increasing irradiation intensity. In the case of H, all the contact angles were evidently lower. The specimens in this group had a continual carbonised layer on their surfaces, and this layer exhibited its own specific qualitative properties. water θθe e

θ0 θ0

θw θw

80

θ0, θe, θw [⁰]

θ0, θe, θw [⁰]

θ0 θ0

diiodomethane

60 40 20

θe θe

θw θw

80 60 40 20 0

0 0

0

10 20 30 40 50 60 70 80 H

10 20 30 40 50 60 70 80 H [J·cm-2]

[J·cm-2]

Fig. 5 Beech wood wetting dependence on irradiation dose.

13

Beech wood wetting with diiodomethane differed from water in quantity and quality.

The wetting rate in diiodomethane was much faster, and the average times te for the tested groups were about one second. The average contact angle 0 values in groups Ref to D did not show significant changes with increasing irradiation dose. Further radiation dose increases induced contact angle decrease, in the case of group H down to zero. Consequently, the wetting was perfect. The course of contact angle e values was qualitatively similar. The angle w calculated for an ideal smooth surface was evidently lower, with the average values ranging from 5.5° to 7°. In group H, this angle was also zero.

The values of contact angles 0 and e have been confirmed depending on the physicalchemical properties of the liquid and on the chemical properties of the wood surface. Supposing that the contact angle values w are exclusively the result of the chemical composition of the two neighbour phases (LIPTÁKOVÁ and KÚDELA 1994, KÚDELA and LIPTÁKOVÁ 2006), the different values corresponding to different irradiation doses confirm major chemical changes to the wood surface due to varied irradiation doses.

The contact angle values also express the extent of water absorbance/repellence.

Consequently, based on the contact angle 0 values, we may state that the water absorbance in the irradiated surface has not been reduced substantially. Similar results were obtained by DOLAN (2014), who investigated wetting of CO2 laser-modified poplar and pine wood surfaces with several liquids (water, form-amide, diiodomethane). This author observed that the best liquid for wood wetting was water, having the highest polar component, and the poorest one was diiodpomethane, being an non-polar substance. The same fact has been confirmed in our case.

However, the assessment of water absorbance/repellence based on the contact angle w values and on the time necessary for the drop to soak into the wood resulted in finding that the wood surface water absorbance decreased with increasing irradiation dose, which is in accord with HALLER et al. (2014). Nevertheless, in no case were our contact angles more than 90°, unlike the angles measured by the last cited authors.

The contact angle values θw were used for calculating the wood surface free energy S with its disperse and polar components γSd and γSp. The surface free energy calculated based on wetting with water was the highest in the case of the referential specimens, displaying an average value of 67.5 mJm2 – the same as the value obtained for milled radial beech wood surface by KÚDELA et al. (2016c). The polar component was somewhat higher (36.0 mJm2) than the disperse one (31.5 mJm2) (Fig. 6). With irradiation dose increasing, the surface free energy decreased proportionally, primarily due to the decrease in the polar component. The disperse component showed a moderate increase, just at the lowest irradiation dose. Later, however, no changes were detected. The surface free energy determined based on diiodomethane was evidently higher, consisting almost exclusively of the disperse component (Fig. 6).

Our results have confirmed that different liquid standards used for assessment of wood surface properties exhibited different performance at the wood-liquid interface. This was due to different surface free energy values and different disperse and polar components of these liquids. Following KÚDELA (2014), the final surface free energy was determined as the sum of the polar component derived from the wetting with water and disperse component obtained with diiodomethane. The surface free energy determined in this way was higher, with dominant disperse component, than the surface free energy calculated separately for the separate liquids (Table 5).

14

water

γγdd

γp γp

γγdd

γγ

γ, γd, γp [mJ·m-2]

γ

γ, γd, γp [mJ·m-2]

diiodomethane

80 60 40 20 0

γp γp

80 60 40 20 0

0

10

20

30

H

40

50

60

70

80

0

[J·cm-2]

10

20

30

H

40

50

60

70

80

[J·cm-2]

Fig. 6 Beech wood surface free energy and its disperse and polar component corresponding to different irradiation doses.

Tab. 5 The final values of surface free energy with the disperse and polar components for beech wood irradiated with varied energy amounts.

Free surface energy and its components

 d p

0 Ref. 86.39 50.43 35.96

7.8 A 78.18 50.41 27.77

Irradiation dose H (Jcm2) 10.7 12.5 15.0 18.8 25.5 B C D E F 76.61 75.20 75.62 74.15 73.60 50.46 50.46 50.45 50.50 50.54 26.15 24.73 25.17 23.65 23.06

37.5 G 69.25 50.48 18.77

75.5 H 78.17 50.80 27.37

The changes to beech wood surface properties discussed in this paper were caused by the structural changes in the main wood components. KAČÍK and KUBOVSKÝ (2011) found no significant degradation of saccharides in wood irradiated with a CO2 laser with a power less than 20 Jcm2 (corresponding to groups AE in our case). In this case, the wood surface modification does not show carbonisation symptoms. This means that the temperature in the surface layers did not exceed ca 200°C (HALLER et al. 2014). The irradiation dose of more than 25 Jcm2 caused a dramatic loss of polysaccharides as the result of degradation of hemicelluloses and a part of the amorphous cellulose fraction. The ratio between the cellulose and hemicelluloses increased significantly, too, and the wood surface was gradually noticeably carbonised. In our study, this was the case of specimens belonging to groups FH, with the surface layer of H specimens carbonised continually. The XPS results also indicate changes to the lignin structure (DOLAN 2014). Chemical reactions in lignin cause browning, especially in light-coloured wood species, including beech. Important agents are also extractive substances (CHANG et al. 2010, TOLVAJ et al. 2011, SIKORA et al.

2018), which are sensitive not only to UV radiation but potentially also to IR radiation (KÚDELA et al. 2018b). Identifying the nature of changes induced with specific irradiation types is not a simple problem. To address it, thorough chemical analyses on wood surfaces treated in this way are necessary.

CONCLUSION In beech wood surface modification with different irradiation doses, the energy from a CO2 laser was applied onto the wood surface, transformed into heat, and induced changes to the wood surface chemical structure, and, consequently, its properties.

15

The beech wood surface discolouration depended on the amount of supplied energy.

Increasing the irradiation dose within 075 J∙cm2 resulted in significant changes in the colour coordinates L*, a*, and b*. With increasing irradiation dose, lightness decreased (from 73 to 30), with the colour coordinates a* and b* clearly shifting towards green and blue, which also produced a dramatic increase in the total colour difference E*.

Important morphological changes, manifested in increased roughness, occurred only under the highest irradiation dose, mainly due to the surface layer carbonisation.

The beech wood surface free energy was calculated as the sum of polar component determined based on wood wetting with water and the disperse component determined based on wetting with diiodomethane. This energy decreased with increasing irradiation dose, especially due to the reduced polar component. Despite this fact, the laser-modified beech wood surfaces (except the carbonised one with a dose of 75 J∙cm2) seem to comply with the requirements for the surface treatment with coating materials and for gluing.

REFERENCES ALIGIZAKI, E. M., MELESSANAKI, K., 2, POURNOU, A. 2008: The use of lasers for the removal of shellac from wood. In e-PRESERVATIONScience 5: 3640.

ALLEGRETTI, O., TRAVAN, L., CIVIDINI, R., 2009: Drying techniques to obtain white Beech. Wood EDG Conference, 23rd April 2009, Bled, Slovenia. http://timberdry.net /downloads/ EDGSeminarBled/Presentation/EDG BABIAK, M., KUBOVSKÝ, I., MAMOŇOVÁ, M. 2004: Farebný priestor vybraných domácich drevín.

(Colour space of the selected domestic species.). In Interaction of wood with various form of energy. (Eds.: Kurjatko, S. and Kúdela, J.): Zvolen : Technical University in Zvolen, p.113–117.

BLANCHARD, V., BLANCHET, P., RIEDL, B. 2009: Surface energy modification by radiofrequency inductive and capacitive plasmas at low pressures on sugar maple: an exploratory study. In Wood Fiber Sci. 41(3): 245–254. ČERMÁK, P., DEJMAL, A. 2013: The effect of heat and ammonia treatment on colour response of oak wood (Quercus robur) and comparison of some physical and mechanical properties. Maderas: Ciencia y tecnologia 15(3): 375389 CSANADY, E., MAGOSS, E. 2011. Mechanics of wood machining. Sopron: University of West Hungary, 2011, 243 p.

DOLAN, J. A. 2014: Characterization of laser modified surfaces for wood adhesion. (Thesis for the degree of Master of Science In: Macromolecular Science and Engineering). The Faculty of Virginia Polytechnic Institute, Blacksburg, VA, 100 p.

DZURENDA, L. 2014: Colouring of beech wood during thermal treatment using saturated water steam.

In Acta Facultatis Xylologiae Zvolen 56(1):1322.

GURAU, L. 2013: Analyses of roughness of sanded oak and beech surface. In ProLigno, 9(4): 741–75.

GURAU, L., PETRU, A., VARODI, A., TIMAR, M.C., 2018: The influence of CO2 laser beam power output and scanning speed on surface quality of Norway maple (Acer platanoides). In BioResources 13(4): 81688183.

HALLER, P., BEYER, E., WIEDEMANN, G., PANZNER, M., WUST, H. 2014: Experimental study of the effect of a laser beam on the morphology of wood surfaces. https://www.researchgate.net/ publication/237543545 HUBBE, M. A., GARDNER, D. J., SHEN, W. 2015: Contact angles and wettability of cellulosic surfaces: a review of proposed mechanisms and test strategies. In BioResouces 10(4): 1–93.

CHANG, T. C., CHANG, H. T., WU, C. L., and CHANG, S. T. 2010: Influences of extractives on the photodegradation of wood. In Polym. Degrad. Stab. 95: 516−521.

JANKOWSKA, A., BORUSZEWSKI, P., DROŻDŻEK, M., RĘBKOWSKI, B., KACZMARCZYK, A., & SKOWROŃSKA, A. 2018: The role of extractives and wood anatomy in the wettability and free surface energy of hardwoods. In BioResources 13(2): 3082–3097.

16

KAČÍK, F., KUBOVSKÝ, I. 2011: Chemical changes of beech wood due to CO2 laser irradiation. In J. Photochem. Photobiol. A 222: 105–110.

KLOUBEK, J. 1974. Calculation of Surface Free Energy Components of ice according to its wettability by water, chlorbenzene and carbon disulfide. In J. Colloid Interface Sci. 46: 185–190.

KUBOVSKÝ, I., KAČÍK, F. 2013: Changes of the wood surface colour induced by CO2 laser and its durability after the xenon lamp exposure. In Wood Res.-Slovakia 58(4): 581590.

KUBOVSKÝ, I., OBERHOFNEROVÁ, E., KAČÍK, F., PÁNEK, M. 2018: Surface changes of selected hardwoods due to weather conditions. In Forests 2018 9(9): 557 KÚDELA, J. 2014: Wetting of wood surface by liquids of a different polarity. In Wood Res.-Slovakia 59(1): 1124.

KÚDELA, J:, ANDOR, T. 2018: Beech wood discoloration induced with specific modes of thermal treatment. In Ann. WULS-SGGW, For and Wood Technol. No 103: 64–69.

KÚDELA, J., ČUNDERLÍK, I. 2012: Bukové drevo – štruktúra, vlastnosti, použitie. (Beech wood – structure, properties and utilisation). Zvolen: Technická univerzita vo Zvolene, 152 p.

KÚDELA, J, JAVOREK, Ľ., MRENICA, L. 2016. Influence of milling and sanding on beech wood surface properties. Part II. Wetting and thermo-dynamical characteristics of wood surface. In Ann. WULS-SGGW, For and Wood Technol. No. 95: 154158.

KÚDELA, J., KUBOVSKÝ, I., ANDREJKO, M. 2018a. Impact of different radiation forms on beech wood discolouration. In Wood Research 63(6): 923934.

KÚDELA, J., LIPTÁKOVÁ, E. 2006: Adhesion of coating materials to wood. In J. Adhes. Sci. Technol.

20(8): 875895.

KÚDELA, J., MRENICA, L., JAVOREK, Ľ. 2018b. Influence of milling and sanding on wood surface morphology. In Acta Facultatis Xylologiae Zvolen 60(1): 71−83.

LASKOWSKA, A., & SOBCZAK, J. W. (2018). Surface chemical composition and roughness as factors affecting the wettability of thermo-mechanically modified oak (Quercus robur L.). In Holzforschung 72(11): 9931000.

Li, R., Xu, W., Wang, X.A., Wang, C. 2018. Modeling and predicting of the color changes of wood surface during CO2 laser modification, In J. Clean. Prod. 183:818823.

LIPTÁKOVÁ, E., KÚDELA, J. 1994. Analysis of the wood – wetting process 1994. In Holzforschung 48(2): 139–144.

NEUMANN, A. W., GOOD, R. J., HOPPE, C. J., SEJPAL, M. 1974. An equation of state approach to determine surface tensions of low-energy solids from contact angles. In J. Colloid Interface Sci. 49(2): 291–303.

PANDEY, K. K., VUORINEN, T. 2008. Comparative study of photo degradation of wood by a UV laser and a xenon light source. In Polym. Degrad. Stab. 93: 2138–2146.

PETRIČ, M., OVEN, P. 2015: Determination of wettability of wood and its significance in wood science and technology: a critical review. In Reviews of Adhesion and Adhesives 3(2): 121−187.

REINPRECHT, L., 2016: Wood deterioration, protection and maintenance. John Wiley & Sons, Chichester, 357 pp.

REINPRECHT, L., MAMOŇOVÁ, M., PÁNEK, M., KAČÍK, F. 2018. The impact of natural and artificial weathering on the visual, colour and structural changes of seven tropical woods. In Eur. J. Wood Prod. 76(1): 175–190.

SIKORA, A., KAČÍK, F., GAFF, M., VONDROVÁ, V., BUBENÍKOVÁ, T., KUBOVSKÝ, I. 2018. Impact of thermal modification on color and chemical changes of spruce and oak wood. In J. Wood Sci., 64: 406416.

TOLVAJ, L., PERSZE, L., ALBERT, L. 2011. Thermal degradation of wood during photodegradation.

In J. Photoch. Photobio. B 105(1): 90–93.

VARGA, D., van der ZEE, M. E. 2008. Influence of steaming on selected wood properties of four hardwood species. In Holz Roh- u. Werkstoff 66: 1118.

VIDHOLDOVÁ, Z., REINPRECHT, L., IGAZ. R. 2017. Mold on laser-treated beech. In BioResources 12(2): 41774186.

WANG, C., PIAO, C. 2011. From hydrophilicity to hydrophobicity: a critical review – part II. hydrophobic conversion. In Wood Fiber Sci. 43: 116.

17

WUST, H., HALLER, P., WIEDEMANN, G. 2005. Experimental study of the effect of a laser beam on the morphology of wood surfaces. The Second European Conference on Wood Modification, ECWM 2005, Göttingen: Univ. Göttinge, p. 367–370.

ZBOROWSKA, M., STACHOWIAK-WENCEK, A., WALISZEWSKA, B., PRADZYNSKI, W., 2016.

Colorimetric and FTIR ATR spectroscopy studies of degradative effects of ultraviolet light on the surface of exotic ipe (Tabebuia sp.) wood. In Cell. Chem. Technol. 50: 7176.

ACKNOWLEDGMENTS This work was supported by the Slovak Research and Development Agency under the contracts No. APVV-16-0177 and APVV-17-0583, and by the Scientific Grant Agency of the Ministry of Education SR and the Slovak Academy of Sciences Grant No. 1/0822/17.

The authors would like to acknowledge the support from Veronika Hyšková concerning the laboratory work.

ADDRESSES OF AUTHORS Prof. Ing. Jozef Kúdela, CSc. Ing. Michal Andrejko Technical University in Zvolen Faculty of Wood Sciences and Technology Department of Wood Science T. G. Masaryka 24 960 53 Zvolen Slovak Republic kudela@tuzvo.sk Prof. Ing. Ladislav Reinprecht, PhD. Ing. Zuzana Vidholdová, PhD. Technical University in Zvolen Faculty of Wood Sciences and Technology Department of Wood Technologies T. G. Masaryka 24 960 53 Zvolen Slovak Republic

18

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 19−29, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.02

CHANGES IN THE CHEMICAL COMPOSITION OF OAK WOOD DUE TO STEAMING Anton Geffert - Eva Výbohová - Jarmila Geffertová ABSTRACT Monitoring the chemical changes resulting from hydrothermal treatment of English oak (Quercus robur L.) through various steaming modes is presented in the paper. The greatest changes were observed in the polysaccharide content (holocellulose). In the most extreme steaming treatment in mode III (tmax 140 °C, total duration for 7.5 h), the decrease in their content was the most dramatic reaching approximately 36%. This decrease is the result of degradation of the most labile hemicelluloses. Only minimal changes in cellulose and lignin content were observed, and several concurrent effects result in the slight increase and decrease in individual modes. Firstly, the content of extractives decreased slightly, but with increasing temperature and extended steaming period a considerable increase in their content occurred. In the initial stages of steaming, new carbonyl and carboxyl groups in carbohydrates are formed by oxidation. Consequently, the deacetylation and degradation of hemicelluloses occurred. In addition, the decrease in the crystallinity of cellulose due to steaming was observed.

Key words: oak wood, steaming, extractives, holocellulose, cellulose, lignin, FTIR spectroscopy.

INTRODUCTION Hydrothermal treatment of wood by steaming or water vapor is a common industrial processing of wood, which is used to improving its properties. The wood after hydrothermal treatment is less sticky, less cracks, drying faster, having a more pleasant and uniform colour, increased durability and strength, and better stability (MELCER et al. 1983, DZURENDA and DELIISKI 2000, DZURENDA 2013).

Steaming causes changes in structural, physical, chemical and mechanical properties of wood. The extent of these changes depends on the hydrothermal treatment conditions (temperature, pressure, duration of action and other). In general, hydrothermal action on wood under mild conditions (shorter time, temperature below 80 °C) causes only minor changes in its main components. Deeper chemical changes occur with longer treatment times and temperatures above 80 °C, while mechanical strength of wood decreases (SOLÁR and MELCER 1990, 1992, MELCER et al. 1983, 1989, KAČÍK et al.1989, KAČÍK 1997, DZURENDA 2018a, 2018b).

The increase in the acidity of the condensate under hydrothermal treatment on wood is caused by the cleavage of acetyl and formyl groups of hemicelluloses and the formation of organic acids (particularly acetic acid and formic acid) which catalyze different 19

hydrolysis, dehydration, degradation, as well as condensation reactions of carbohydrates and their products. 5-hydroxymethyl-2-furaldehyde is formed by dehydration of hexoses, and 2furaldehyde is formed by degradation of the pentoses. Further decomposition of the furan derivatives produces levulinic acid and formic acid (JÖNSSON et al. 2013).

Hemicelluloses are heteropolysaccharides with branched and shorter chains of saccharide units. Because of its amorphous structure and the presence of acetyl groups, hemicelluloses are the most thermally labile of the wood polymeric components (HILL 2006).

Degradation of especially non-cellulosic polysaccharides leads to the loss of holocellulose in hydrothermally treated wood (KAČÍK et al. 1990, 2001).

The resistance of different wood species to hydrothermal treatment is not the same (MELCER et al. 1983, 1989, KAČÍK et al. 1996). Hardwoods have a higher proportion of hemicelluloses, and the hemicelluloses of hardwoods have a higher content of acetyl groups compared to softwoods. Additionally, hardwood hemicelluloses are richer in pentosans, which are more susceptible to degradation than hexosans. Therefore, hardwoods are less thermally stable than softwoods (HILL 2006).

In lignocellulosic materials, the main components form the so-called “ligninsaccharide complex”. Cellulose microfibrils are covered with a heterogeneous hemicellulose polymer which is wrapped by amorphous lignin polymer (VOLYNETS et al. 2017). According to CHEN et al. (2010) during the initial phase of hydrothermal treatment lignin-free xylan is released, while lignin-bound xylan is dissolved in the later phase.

In contrast to hemicelluloses, the monomers of glucose in native cellulose form microfibrils stabilized by hydrogen bonds, thus making the macromolecule highly crystalline and more difficult to hydrolyze (TRAJANO and WYMAN 2013). The thermal stability of cellulose mainly influenced on its degree of crystallinity, crystallite size, and degree of polymerization (POLETTO et al. 2012, KIM et al. 2010). The rate of cellulose degradation is reduced if water is present, which is assumed to be due to the enhanced ability of the amorphous regions to change structure to produce more thermally stable crystalline regions (FENGEL and WEGENER 1989). With extended heating, chain scission of the cellulose occurs, producing oligosaccharides, with a concomitant decrease in the degree of polymerization as well as degree of crystallinity of cellulose (HILL 2006). According to other studies (BHUIYAN et al. 2000, KONG et al. 2017), the crystallinity increases at the initial stage and decreased at the later stages of heat treatment under moist conditions.

Several authors report that not only the carbohydrate but also the aromatic part of the wood (lignin) undergoes changes during the hydrothermal treatment. The depth of these changes depends primarily on the temperature and the time of action, as well as on the species of treated wood (SOLÁR and MELCER 1992, KAČÍK et al. 1989, KAČÍK et al. 1990).

The hydrothermal treatment causes also the formation of new chromophore structures in the lignin, which causes a change in the colour of the treated material (SOLÁR 1997).

The aim of this work was to determine and evaluate chemical changes occurring in the oak (Quercus robur L.) wood as a result of its modification in the different modes of steam, using conventional analytical methods as well as ATR-FTIR spectroscopy.

MATERIAL AND METHODS KLEMENT et al. (2010) characterized oak wood as hard, tough, solid, with poor impregnation and staining. In terms of physical, mechanical and technological properties it is an important raw material for industrial processing.

The samples of oak wood supplied from industrial plant Sundermann Ltd Banská Štiavnica were used to investigate chemical changes that occurred in different steaming 20

treatments. The samples with the dimensions 30 × 75 × 510 mm were thermally treated with saturated steam in the pressure autoclave APDZ 240 (DZURENDA 2018a). The modification mode of oak wood is given in Figure 1 and temperature of saturated water steam and duration of the technological process are shown in Table 1.

Fig. 1 Modification mode of oak wood with saturated water steam (DZURENDA 2018a).

Tab. 1 Thermal treatment of the oak wood.

Mode steaming I II III

Temperature of saturated water steam (°C) tmin tmax t4 110 115 100 125 130 100 135 140 100

Duration (h) 1 4.5 5.0 5.5

2 1.0 1.5 2.0

Disintegrated samples of the original oak wood and wood after steaming (Figure 2) were used to monitor the chemical changes.

Fig. 2 Samples of the original (0) and steamed oak wood samples (mode I, mode II, mode III).

The selected chemical characteristics in the samples of the original oak wood and the wood treated through various steaming techniques were determined in the fraction of sawdust from 0.5 mm to 1.0 mm prepared from completely disintegrated boards (including surface and center part): Tab. 2 Select chemical characteristics.

Ethanol-toluene solubility Polysaccharide fraction Cellulose Lignin Seifert´s cellulose

ASTM D 1107 – 96 Chlorite isolation method of Wise et al. (KAČÍK and SOLÁR 2000) Kürschner-Hoffer method (KAČÍK and SOLÁR 2000) ASTM D 1106 – 96 Acetylacetone method (Seifert 1956)

Note: The content of hemicelluloses was determined as the difference between the holocellulose and cellulose content.

21

The samples of wood, as well as isolated holocellulose and Seifert´s cellulose were analyzed using ATR-FTIR spectroscopy. The measurements were performed using a Nicolet iS10 FTIR spectrometer equipped with a Smart iTR attenuated total reflectance (ATR) sampling accessory with a diamond crystal (Thermo Fisher Scientific, Madison, WI). The resolution was set at 4 cm− 1 for 32 scans. The wavenumber range varied from 4000 to 650 cm− 1. Six analyses were performed per sample. OMNIC 8.0 software (Thermo Fisher Scientific, Madison, WI) was used to evaluate the spectra.

RESULTS AND DISCUSSION The results of the chemical analysis of the samples of the original oak wood and the wood after the steaming in each modes are depicted in Figure 3.

Fig. 3 Chemical characteristics of oak wood before and after steaming.

The observed values of the monitored characteristics for the original oak wood are within the ranges of values cited by various authors (KOLLMAN and FENGEL 1965, GEFFERTOVÁ et al. 2006, GEFFERTOVÁ and HANZEL 2007, GEFFERTOVÁ and GEFFERT 2007, ČABALOVÁ et al. 2018, LAUROVÁ et al. 2007) for various species of oak. The differences can be related not only to the type of wood, but also to the locations and places of sampling.

The content of extractives ranges from 2.3 to 5.6%, holocellulose from 73.2 to 83.3%, cellulose from 36.7 to 46.6 %, and lignin from 17.6 to 25.3%. According to the results of ČABALOVÁ et al. (2018) in English oak wood polysaccharides primarily glucose (48.30%) and xylose (21.44%), less mannose (4.08%), arabinose (1.97%) and galactose (1.17%) are present.

On the basis of the observed chemical characteristics of oak wood after various steaming modes it can be stated that the greatest changes were occurred in the polysaccharide content (holocellulose). At the lowest temperatures and the shortest period of steam (mode I) only a slight decrease in the polysaccharide fraction was recorded. By the following increasing of the intensity of the steam treatment, the polysaccharide content decreased by 10% in mode II and by 36% in mode III compared to the average value in the original oak wood.

22

According to FENGEL and WEGENER (1989), temperature influences first depolymerization of long hemicellulose chains into oligosaccharides and monosaccharides, which are dehydrate to form volatile compounds. At the same time, ongoing deacetylation affects the thermal stability of hemicelluloses (FENGEL 1966).

The content of cellulose in oak wood after steaming by mode I grew by 2.3%. This is a relative increase because of the reduced hemicellulose content in the sample. In other steaming modes, the cellulose content slightly decreased, whereby the decrease represents about 6% relative to the original sample. The slight increase and decrease in individual steaming modes is the result of several concurrent effects - degradation of hemicelluloses or amorphous cellulose and condensation reactions. Some authors report a relative increase in cellulose with a prolonged hydrothermal treatment time at a temperature range from 80 to 140 °C because of the loss of hemicelluloses and lignin (SOLÁR 1997).

According to KAČÍK (1997), the content of cellulose in hydrothermally treated wood does not usually change at temperatures up to 100 °C, but at temperatures above 100 °C, it increases because of the degradation of hemicelluloses. The results of some authors (SOLÁR 1997, MELCER et al. 1983, MELCER et al. 1989) confirm that during the hydrothermal treatment of the wood, there is first a relatively rapid release of hemicelluloses, then the slower release of water-soluble part of the lignin, and later also part of the amorphous cellulose.

While after steaming mode I the hemicellulose content decreased only by 1.1% compared to the original oak wood, after mode II it was already by 4.7%, and after the most intense steaming treatment in mode III it decreased by 22.9%, which means 75% reduction in hemicellulose content.

Considering the determined content of lignin in oak wood it can be stated that in mode I the water-soluble lignin content is likely to decrease. More intense steaming conditions caused degradation and condensation reactions of lignin and their synergistic effect was reflected by an increase in the lignin content by 3.5% compared to the original oak wood sample. SOLÁR (1997) states that depolymerization, reduction in the degree of lignin crosslinking and the disappearance of bonds in the lignin-polysaccharide system predominate in the early stages of hydrothermal wood treatment.

The content of extractives after steaming in mode I decreased from 9.4% to 8.4%, which is related to their release into the condensate from steaming or their decomposition.

In other steaming modes a considerable increase in the extractives was observed. After the most intense steaming (mode III), the content of extractives in the oak wood sample grew 2.3 times (from 9.4% to 21.8%) compared to their original oak wood content. This increase is already related to the release of decomposition products of other wood components into the extraction mixture.

Based on the determined chemical characteristics it can be concluded that the increased severity of the steaming conditions was reflected primarily in the change of content of holocellulose and extractives, less the content of cellulose and lignin.

In the FTIR spectra of wood absorption bands appertaining to all wood components are observable. Assignment of them is in the Table 3. Due to steaming several changes in their intensities are occurred (Table 4). In spectra of oak wood steamed by mode I intensities of characteristic absorption bands of lignin (at 1593 and 1504 cm-1) slightly decreased. After steaming at higher temperatures and with the increase of steaming time their intensities increased. These changes suggest that the content of lignin firstly decreased and with rising of treatment severity increased. Similar trend was found also by determination of lignin content using conventional method by ASTM D 1106 – 96. Similarly, the increase in intensity of characteristic absorption bands of lignin by thermal treatment of pedunculate oak (ČABALOVÁ et al. 2018) and eucalyptus wood (ESTEVES et al. 2013) was reported.

23

Tab. 3 Assignment of infrared absorption bands in wood spectrum (according to HON et al. 2001, PANDEY and PITMAN 2003).

Wavenumber (cm1) 3345 1735 1593 1504 1458 1422 1364 1232 1032 898

Peak assignment O–H stretching C=O stretching of acetyl, carboxylic groups and aldehydes Aromatic skeletal vibrations in lignin Aromatic skeletal vibrations in lignin C–H deformations in lignin and carbohydrates C–H deformations in lignin and carbohydrates C–H in-plane bending in carbohydrates syringyl ring and C–O stretch in lignin and xylan C-O stretching in polysaccharides stretching at the -(1,4)-glycosidic linkage, and C-H deformation

Tab. 4 Relative intensities of absorption bands Ai/A1032 of FTIR spectra of wood.

Wavenumber 3345 cm1 1735 cm1 1593 cm1 1504 cm1 1458 cm1 1422 cm1 1364 cm1 1232 cm1 898 cm1

wood - original 1.0157 0.3904 0.3099 0.1756 0.1740 0.0997 0.1185 0.2309 0.0697

wood – mode I 1.0256 0.3944 0.2838 0.1226 0.1589 0.0791 0.1160 0.2266 0.0701

wood – mode II 1.1890 0.5327 0.2998 0.1663 0.1637 0.0801 0.1399 0.2162 0.0696

wood – mode III 1.1392 0.3978 0.3192 0.1799 0.1612 0.0860 0.1289 0.2037 0.0806

During hydrothermal treatment on wood, more processes with different influence on the intensity of the absorption band around 1733 cm-1 run. The increasing its intensity may be due to opening of the glucopyranose ring, formation of new carbonyl and carboxyl groups, or cleavage of β-alkyl-aryl ether linkages in lignin. On the other hand, a decrease in its intensity may be caused by lignin condensation reactions, deacetylation of hemicelluloses and decomposition of aldehydes, carboxylic acids and their esters (ÖZGENC et al. 2017, ESTEVES et al. 2013, WINDEISSEN et al. 2009). In the FTIR spectrum of wood, the absorption bands of the different wood components overlap in this wavelength region. In order to better elucidate ongoing processes, we analyzed not only samples of wood, but also the samples of isolated holocellulose. Assignment of absorption bands in spectra of holocellulose is in the Table 5.

Tab. 5 Assignment of infrared absorption bands in holocellulose spectrum (according to HON et al.

2001, PANDEY and PITMAN 2003).

Wavenumber (cm-1) 3339 1732 1427 1371 1333 1317 1244 1161 1032 898

Peak assignment O-H stretching C=O stretching of acetyl or carboxylic groups C–H bending C–H in-plane bending OH in-plane bending CH2 wagging C–O stretching in xylan C-O-C asymmetric bridge stretching C-O stretching stretching at the -(1,4)-glycosidic linkage, and C-H deformation

24

In Table 6 relative intensities of absorption bands of FTIR spectra of holocellulose are shown. The intensity of absorption band at 1732 cm1 firstly increases due to increased temperature and extended period of steaming (mode I and mode II). In the hardest degree of steaming (mode III), the height of this peak decreased. We saw the same changes in the spectrum of wood. It can be concluded that changes in FTIR spectrum of wood in range 1770 to 1550 cm1are a sign of changes in polysaccharides.

Tab. 6 Relative intensities of absorption bands Ai/A1032 of FTIR spectra of holocellulose.

Wavenumber 3339 cm1 1732 cm1 1427 cm1 1371 cm1 1333 cm1 1317 cm1 1244 cm1 1161 cm1 898 cm1

HC - original 1.0241 0.2770 0.1039 0.0858 0.0555 0.0882 0.1959 0.1851 0.0664

HC – mode I 1.0396 0.3175 0.0896 0.0910 0.0568 0.0925 0.2207 0.1860 0.0811

HC – mode II 1.0630 0.3756 0.0934 0.1071 0.0570 0.0933 0.2621 0.1921 0.0890

HC – mode III 1.2282 0.1988 0.1155 0.1110 0.0629 0.1251 0.1334 0.2407 0.0917

In our experiment, the initial increase in intensity of absorption band at 1732 cm-1 may be due to formation of new carbonyl and carboxyl groups in carbohydrates by oxidation.

The followed decrease may be due to deacetylation and degradation of hemicelluloses. ÖZGENC et al. (2017) reported opposite changes due to heat-treatment in the intensity of peak at 17301732 cm-1 for deciduous and coniferous woods. While the bands at 17301732 cm1 increased for heat-treated beech wood, they decreased for heat-treated spruce and pine wood. SIKORA et al. (2018) reported the increase and shifting absorbance at around 1730 cm1 to smaller wavenumber with increasing treatment severity in the case of heat treatment of spruce and oak wood. ČABALOVÁ et al. (2018) found that the peak intensity at 1732 cm1 in spectrum of heat-treated oak wood initially increased, and then decreased as the treatment time increased. Based on the above it can be concluded that changes in the intensity of this peak depend on the hardness of the treatment as well as on the kind of wood.

Also it should be emphasized that samples of wood and not of isolated components were analyzed in the cited works. Therefore, the findings in these cases are the result of running the different processes in all wood components.

The changes in crystallinity were determined from FTIR spectra of Seifert´s cellulose as two parameters – the Total Crystallinity Index (TCI) according to NELSON and O´CONNOR (1964) and the ratio of intensities A1334/A1315 according to COLOM et al. (2003). Values of these parameters are shown in Table 7. The decrease in the crystallinity of cellulose due to steaming is obvious from the decrease in the total crystallinity index (TCI) and also from the increase in the ratio A1334/A1316.

Tab. 7 Parameters characterized cellulose crystallinity.

Steaming mode 0 I II III

TCI 0.4805 0.4676 0.4499 0.4488

A1334/A1316 0.7528 0.7572 0.7671 0.7702

KONG et al. (2017) monitored the effect of steaming time on the cellulose crystallinity in eucalyptus wood. The authors found that crystallinity increased, reaching a maximum 25

after 2 h, and then decreased. They examined the decrease in cellulose crystallinity due to increased numbers of chain scission reactions that increased the amorphous character of cellulose, and subsequently reduced the total amount of cellulose crystalline regions.

Furthermore, acetic acid which is formed by hydrolysis of hemicelluloses can cause the degradation of microfibrillar. Our findings are in agreement with cited work because in our experiment steaming times for all modes are longer, namely 5.5, 6.5 and 7.5 h.

CONCLUSION In this paper the chemical changes that occur from the hydrothermal treatment of English oak (Quercus robur L.) wood through various steaming modes were examined.

Increase in temperature and extension of the steaming period primarily affected the holocellulose and extractives contents, and less so the contents of cellulose and lignin.

The holocellulose content due to steaming decreased, the decrease reaching approximately 36% in the case of most extreme steaming treatment. Cellulose content in oak wood under the influence of steaming not greatly changed and its slight increase and decrease in individual modes is the result of several concurrent effects. Consequently, the decrease in holocellulose content is the result of degradation of most labile hemicelluloses.

The content of hemicelluloses reduces by about 75%.

The content of extractives first slightly decreased, but with increasing temperature and extended steaming period a considerable increase in their content was observed. This increase is already related to the release of decomposition products of other wood components into the extraction mixture.

The lignin content in steamed oak wood shows only minimal changes. In the early stages of hydrothermal treatment its content slightly decreased, but at more intense conditions the 3.5% increase in its content was observed.

Based on the results of FTIR analyses it can be concluded that in the initial stages of steaming new carbonyl and carboxyl groups are formed in carbohydrates by oxidation.

Consequently, the deacetylation and degradation of hemicelluloses are occured. In addition, the decrease in the crystallinity of cellulose due to steaming was observed.

As can be seen from Figure 2 the hydrothermal treatment of wood also resulted in darkening of wood samples. The intensity of change is dependent on the severity of conditions. The mechanism of colour change is complex and a number of overlapping reactions of the basic components of wood and their decomposing products are involved.

REFERENCES ASTM Standard D 1106 – 96: 1998. Standard Test Method for Acid Insoluble Lignin in Wood.

(TAPPI T T-13m-54).

ASTM Standard D 1107 – 96, Re-approved: 2001. Standard Test Method for Ethanol-Toluene Solubility of Wood. (TAPPI T 204 os-76).

BHUIYAN, T. R., HIRAI, N., SOBUE, N. 2000. Changes of crystallinity in wood cellulose by heat treatment under dried and moist conditions. In Journal of Wood Science, 2000, 46(6): 431436.

DOI: 10.1007/BF00765800.

CHEN, X., LAWOKO, M., HEININGEN, A. 2010. Kinetics and mechanism of autohydrolysis of hardwoods.

In Bioresource Technology, 2010, 101(20): 78127819.

DOI: 10.1016/j.biortech.2010.05.006.

26

COLOM, X., CARRILO, F., NOGUÉS, F., CARRIGA, P. 2003. Structural analysis of photodegraded wood by means of FTIR spectroscopy. In Polymer Degradation and Stability, 2003, 80(3): 543549. DOI: 10.1016/S0141-3910(03)00051-X. ČABALOVÁ, I., KAČÍK, F., LAGAŇA, R., VÝBOHOVÁ, E., BUBENÍKOVÁ, T., ČAŇOVÁ, I., ĎURKOVIČ, J. 2018. Effect of thermal treatment on the chemical, physical, and mechanical properties of pedunculate oak (Quercus robur L.) wood. In BioResources, 2018, (13)1: 157170.

DZURENDA, L., DELIISKI, N., 2000. Analysis of moisture content changes in beech wood in the steaming process with saturated water steam. In: Wood Research 45(4):18.

DZURENDA, L. 2013, Modification of wood colour of fagus sylvatica l. to a brown-pink shade caused by thermal treatment. In Wood Research 58(3):475482.

DZURENDA, L. 2018a. The Shades of Color of Quercus robur L. Wood Obtained through the Processes of Thermal Treatment with saturated Water Vapor. In BioResources, 2018, 13(1): 15251533.

DZURENDA, L. 2018b. Colour modification of Robinia pseudoacacia L. during the processes of heat treatment with saturated water steam. In Acta Facultatis Xylologiae Zvolen, 2018, 60(1): 6170.

ESTEVES, B., VELEZ MARQUES, A., DOMINGOS, I. PEREIRA, H. 2013. Chemical changes of heat treated pine and eucalypt wood monitored by FTIR. In Maderas, Cienc. Tecnol., 2013, 15(2): 245258.

FENGEL, D. 1966. Über die Veränderungen des Holzes und seiner Komponenten im Temperaturbereich bis 200 °C - Zweite Mitteilung: Die Hemicellulosen in unbehandeltem and in thermisch behandeltem Fichtenholz [About changes in wood and its components in the temperature range up to 200 °C - Second communication: Hemicelluloses in untreated and thermally treated spruce wood]. In Holz als Roh- und Werkstoff, 1966, 24(3): 98109. DOI: 10.1007/BF02608355.

FENGEL, D., WEGENER, G. 1989. Wood: Chemistry, Ultrastructure, Reactions. Berlin : Walter de Gruyter, 1989. 613 pp. ISBN 978-0-899-25593-4.

GEFFERTOVÁ, J., GEFFERT, A., FURTÁK, V. 2006. Vplyv rozdielnych charakteristík jadrového a beľového dubového dreva na vybrané charakteristiky sulfátových buničín. In Acta Facultatis Xylologiae Zvolen, 2006. 48(2): 2332.

GEFFERTOVÁ, J., GEFFERT, A. 2007. Porovnanie vybraných charakteristík sulfátových buničín pripravených z buka, duba a agáta. In VII. International symposium „Selected processes at the wood processing“. Zvolen : Technical University in Zvolen, 2007, 7 p.

GEFFERTOVÁ, J., HANZEL, P. 2007. Dub zimný a dub cérový v procese sulfátovej várky [Sessile oak and Turkey oak in the process of kraft cooking]. In VII. International symposium „Selected processes at the wood processing“. Zvolen : Technical University in Zvolen, 2007, 20 p.

HILL, C. A. S. 2006. Wood Modification: Chemical, Thermal, and Other Processes. Chichester : John Wiley & Sons, 2006. 260 pp. ISBN 978-0-470-02172-9.

HON, D. N., SHIRAISHI, N. 2001. Wood and Cellulosic Chemistry, Marcel Dekker, New York and Basel, 2001. 914 pp. ISBN 0-8247-0024-4.

JÖNSSON, L. J., ALRIKSSON, B., NILVEBRANT, N. O. 2013. Bioconversion of lignocellulose: Inhibitors and detoxification. In Biotechnology for Biofuels, 2013, 6(1): 16. DOI: 10.1186/17546834-6-16.

KAČÍK, F. 1997. Vplyv teploty a vlhkosti na zmeny sacharidov. Zvolen : Technical University in Zvolen, 1997. 68 pp. ISBN 80-228-0608-0.

KAČÍK, F. 2001. Tvorba a chemické zloženie hydrolyzátov v systéme drevo – voda – teplo [Creation and chemical composition of hydrolysates in the wood - water - heat system]. Zvolen : Technical University in Zvolen, 2001. 75 pp. ISBN 80-228-1098-3.

KAČÍK, F., SOLÁR, R. 2000. Analytická chémia dreva. Zvolen : Technical University in Zvolen, 2000.

369 pp. ISBN 80-228-0882-0.

KAČÍK, F., SOLÁR, R., MELCER, I. 1989. Štúdium hydrolyzátu po hydrotermickej úprave bukového dreva (Fagus silvatica L.), I. časť. In Zborník vedeckých prác Drevárskej fakulty, Zvolen : Technical University in Zvolen, 1989, pp. 3545.

KAČÍK, F., SOLÁR, R., MELCER, I. 1990. Štúdium hydrolyzátu po hydrotermickej úprave bukového dreva (Fagus silvatica L.), II. In Zborník vedeckých prác Drevárskej fakulty, Zvolen : Technical University in Zvolen, 1990, pp. 4556.

27

KAČÍK, F., SOLÁR, R., BALOGHOVÁ, D. 1996. Zmeny lignínu javorového dreva (Acer pseudoplatnus L.) vplyvom predhydrolýzy [Changes in lignin of maple wood (Acer pseudoplatnus L.) due to prehydrolysis]. In Selected Processes in the Chemical Processing of Wood, Zvolen : Technical University in Zvolen, pp. 173178.

KIM, U. J., EOM, S. H., WADA, M. 2010. Thermal decomposition of native cellulose: influence on crystallite size. In Polym Degrad Stab, 2010, 95(5): 778–781.

KLEMENT, I., RÉH, R., DETVAJ, J. 2010. Základné charakteristiky lesných drevín. Zvolen : NLC, 2010, 82 pp. ISBN 978-80-8093-112-4.

KOLLMAN, F., FENGEL, D. 1965. Änderungen der chemischen zusammensetzung von holz durch thermische behandlung. In Holz als Roh- und Werkstoff, 1965, 23(12): 461–468.

KONG, L., ZHAO, Z., HE, Z., YI, S. 2017. Effects of steaming treatment on crystallinity and glass transition temperature of Eucalyptuses grandis ᵡ E. urophylla. In Results in Physics, 2017, 7: 914– 919. DOI:10.1016/j.rinp.2017.02.017.

LAUROVÁ, M., VÝBOHOVÁ, E., MAMOŇOVÁ, M. 2007. Heartwood and sapwood lipophilic extractives of oak (Quercus petraea (Mattusch.) Liebl.). In Acta Facultatis Xylologiae Zvolen, 2007.

49(2): 17–23.

MELCER, I., MELCEROVÁ, A., SOLÁR, R., GAJDOŠ, E. 1983. Porovnávacia charakteristika zmien jaseňového dreva (Fraxinus excelsior L.) po jeho hydrotermickej úprave varením a parením. In Wood Research, 1983, 28(1): 3756.

MELCER, I., MELCEROVÁ, A., SOLÁR, R., KAČÍK, F. 1989. Chemizmus hydrotermickej úpravy listnatých drevín. Zvolen : University of Forestry and Wood Technology in Zvolen, 2/1989, 76 pp.

NELSON, M., L., O´CONNOR, R. T. 1964. Relation of certain infrared bands to cellulose crystallinity and crystal latticed typ. Part II. A New Infrared Ratio for Estimation of Crystallinity in Celluloses I and 11. In Journal of Applied Polymer Science, 1964, 8(3): 13251341. ÖZGENC, Ö., DURMAZ, S., BOYACI, I. H., EKSI-KOCAK, H. 2017. Determination of chemical changes in heat-treated wood using ATR-FTIR and FT Raman spectroscopy. In Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2017, 171: 395400.

PANDEY, K. K., PITMAN, A. J. 2003. FTIR studies of the changes in wood chemistry following decay by brown-rot and white-rot fungi. In International Biodeterioration & Biodegradation, 2003, 52(3): 151160. DOI: 10.1016/S0964-8305(03)00052-0 POLETTO, M., ZATTERA, A. J., FORTE, M. M., SANTANA, R. M. 2012. Thermal decomposition of wood: influence of wood components and cellulose crystallite size. In Bioresour Technol, 2012, 109(1): 148–53.

SEIFERT, K. 1956. Zur Frage der Cellulose – Schnellbestimmung nach der Acetylaceton – Methode [To the question of cellulose - rapid determination according to the acetylacetone method]. In Das Papier, 1956, 14(3): 104106.

SIKORA, A., KAČÍK, F., GAFF, M., VONDROVÁ, V., BUBENÍKOVÁ, T., KUBOVSKÝ I. 2018. Impact of thermal modification on color and chemical changes of spruce and oak wood. In Journal of Wood Science, 2018, 1-11. DOI: 10.1007/s10086-018-1721-0.

SOLÁR, R. 1997. Zmeny lignínu v procesoch hydrotermickej úpravy dreva. Zvolen : Technical University in Zvolen, 1997, 57 pp. ISBN 80-228-0599-8.

SOLÁR, R., MELCER, I. 1990. Chemical changes of the polysaccharidic part of hydrothermally treated oak wood their reflection in its mechanical properties. In Zborník vedeckých prác Drevárskej fakulty, Zvolen : Technical University in Zvolen, 1990, pp. 1526.

SOLÁR, R., MELCER, I. 1992. Structural land chemical alterations of lignin in the proces of oak wood hydrothermal treatment. In Wood Research, 1992, 12(3): 1123.

TRAJANO, H. L., WYMAN, C. E. 2013. Fundamentals of biomass pretreatment at low pH. In Aqueous Pretreatment of Plant Biomass for Biological and Chemical Conversion to Fuels and Chemicals.

C. E. Wyman (ed.), Medford : John Wiley & Sons, 2013, pp. 103128. DOI: 10.1002/9780470975831.ch6.

VOLYNETS, B., EIN-MOZAFFARI, F., DAHMAN, Y. 2017. Biomass processing into ethanol: Pretreatment, enzymatic hydrolysis, fermentation, rheology, and mixing. In Green Processing and Synthesis, 2017, 6(1): 122. DOI: 10.1515/gps-2016-0017.

28

WINDEISEN, E., BÄCHLE, H., ZIMMER, B., WEGENER, G. 2009. Relations between chemical changes and mechanical properties of thermally treated wood. In Holzforschung, 2009, 63: 773778.

ACKNOWLEDGMENTS This work was supported by the Slovak Research and Development Agency under the contract No. APVV-17-0456.

ADDRESSES OF THE AUTHORS prof. Ing. Anton Geffert, CSc. Ing. Eva Výbohová, PhD. doc. Ing. Jarmila Geffertová, PhD. Technical University in Zvolen Faculty of Wood Sciences and Technology Department of Chemistry and Chemical Technologies T.G. Masaryka 24 Slovakia geffert@tuzvo.sk vybohova@tuzvo.sk geffertova@tuzvo.sk

29

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 31−42, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.03

ASSESSMENT OF THE CHEMICAL CHANGE IN HEAT TREATED PINE WOOD BY NEAR INFRARED SPECTROSCOPY Zuzana Vidholdová – Anna Sandak – Jakub Sandak ABSTRACT Fourier-transform near-infrared spectroscopy (FT-NIR) was used as none-destructive method to determinate changes in the chemical structure of heat-treated wood. For this purpose, pine sapwood (Pinus sylvestris L.) was treated at different temperatures (from 100 °C to 240 °C) and for three durations (1, 3 or 5 hours). The effects of chemical changes on the FT-NIR spectra are linked to absorbance changes of functional groups (–OH, –CH, –CO and –CH2) of lignin, hemicelluloses and cellulose. Gradual degradation of amorphous portion of cellulose was caused by high temperature, while crystalline and semi-crystalline portions of cellulose seem to be less affected by the thermal treatment. The effect of various intensities of heat treatment on chemical changes of wood polymers varied depending on temperature and duration. Presentation of spectra in the form of the xylograms shows clear tendency of degradation kinetic. Evaluation of thermal stability of selected wood component and/or comparison of the influence of modification process parameters can be carried out.

Key words: pine, heat treatment, xylograms, FT-NIR.

INTRODUCTION Heat-treated wood has become an established commercial product possessing a number of advantages over the natural wood. Heat-treated wood is considered an ecofriendly alternative to chemically impregnated wood materials. This treatment reduces the hydrophilic behaviour of the wood by modifying the chemical structure of its components (hemicelluloses, cellulose and lignin) which results in changes of their properties. Some previous studies (GÉRARDIN 2016, KUČEROVÁ et al. 2016, SANDBERG & KUTNAR 2016, ČABALOVÁ et al. 2014, REINPRECHT & VIDHOLDOVÁ 2011, H ILL 2007, WELZBACHER et al.

2007) have reported that treatment temperature and its duration affect the chemical decomposition of wood. The most important positive effects of heat treatment of wood are: enhancement of resistance to biodegradation (ŠUŠTERŠIC et al. 2010, WELZBACHER & RAPP 2007, HAKKOU et al. 2006) improvement of the overall dimensional stability (VIITANIEMI et al. 1997, HILLIS 1984) and reduction of the heat transfer coefficient (MILITZ 2002). Heat treatment is lowering wood equilibrium moisture content (ALTGEN et al. 2016) and enhance the surface quality (PRIADI & HIZIROGLU 2013) in addition to bulk discoloration having attractive dark colour (TODOROVIC et al. 2012). High temperatures and long time of heat treatment decrease most of the mechanical properties of wood (YILDIZ et al. 2011). Thermal modification decreases the heat release rate and propensity for fire propagation in the flashover phase of some species (MARTINKA et al. 2016).

31

Fourier transform near-infrared spectroscopy (FT-NIR) is an efficient method for high-throughput non-destructive screening of chemical characteristics of different materials including the wood and wood based products. Energy of infrared light stimulates vibrations of -CO, -OH, -CH and -NH functional groups giving overtones and combination bands depending on the molecular structure, chemical composition or physical properties of the measured sample. A state of the art of the FT-NIR applications in wood and paper research has been published by TSUCHIKAWA & KOBORI (2015). Quality assessment of thermally treated wood by means of NIR was previously investigated by several researches (POPESCU et al. 2018, SANDAK et al. 2015, 2016, BÄCHLE et al. 2010, MEHROTRA et al. 2010, ESTEVES & PEREIRA 2008). NIR spectra can be pre-processes mathematically and evaluated by means of multivariate data analysis to obtain precise quantitative and qualitative information of physical-chemical nature of material.

In this study, the chemical changes due to heat treatment intensity were evaluated in pinewood by non-destructive FT-NIR spectroscopy. The chemical fingerprint of thermally modified wood is visualized by means of xylograms.

MATERIALS AND METHODS Material and wood treatment The defect-free pine sapwood (Pinus sylvestris L.) without cracks, knots or other growth inhomogeneity were used as experimental samples. The density in oven dry state ranged from 431 to 639 kg·m-3 with an average value of 506 kg·m-3. Specimens were heat treated under atmospheric pressure in the laboratory heating oven (Memmert UFB 500, Germany) at Department of Mechanical Wood Technology, FWST at Technical University in Zvolen, as shown in Table 1.

The heat treatment started by putting the samples at ambient temperature in oven with subsequent increasing of the temperature and without forced air circulation. The period to reach expected temperature varied from 15 minutes (for 100 °C) up to 60 minutes (for 240 °C). Duration of the heat treatment at fixed temperature was 1, 3 or 5 hours. Extensively treated wood was prepared at the temperature 240 °C during 8 hours. At the end of each treatment, samples were cooled down in desiccators in dry environment.

Tab. 1 Thermal modification set-up, treatment parameters, set size.

Species Pine – sapwood

Dimensions (R × L × T) (mm) 25 × 25 × 3

Treatment temperature (°C) 100 150 160 200 220 240

Treatment duration (h) 1 3 5 8*

Number of replica 4

Note: * The treatment duration of 8 hours was only used for preparation of extensively treated wood.

Wood degradation was monitored by measuring the mass loss (ML) and the CIE L a b colour coordinates. Mass loss percentage was determined on the representative samples set by means of their dry-weight change before and after heat treatment, determined after oven drying at 103 ± 2 °C to constant weight.

Colour of heat treated samples expressed in CIE L*a*b* system was measured on samples conditioned at room temperature of 20 ± 2 °C and relative humidity of 60 ± 5 %. * * *

32

Colour was measured using the Colour Reader CR-10 (Konica Minolta, Japan), with the illuminate type D65 light source and observer angle of 8 ° and approx. ϕ 8 mm measuring area.

FT-NIR measurements The FT-NIR spectrometer (VECTOR 22-N) produced by Bruker Optics GmbH (Germany) equipped with a fibre-optic probe was used for spectra collection. FT-NIR measurements were performed in a climatic chamber (20 °C, 60 % relative humidity), on the radial face of samples. The spectral range was between 4000 cm−1 and 12 000 cm−1 (2500 nm - 833 nm) and the resolution was set to 8 cm−1. Each spectrum was collected from 32 internal scans in the absorbance mode. Four measurements were performed on each sample and resulting spectra were averaged. All measurements and subsequent data evaluation were performed at Trees and Timber Institute CNR-IVALSA in San Michele all Adige (Italy).

Data evaluation Opus QUANT 6.5 (Bruker), PLS toolbox (Eigenvector) and LabVIEW 17 (National Instruments) software packages were used for spectral pre-processing and data mining. For the needs of this research, different evaluation methods were applied on pre-processed data (Table 2). Spectral bands (Table 3) were assigned according to SCHWANNINGER et al. (2011).

Tab. 2 Applied methods during FT – NIR measurements and evaluation.

Sample presentation Intact, manual around the surface

Acquisition mode absorbance

Regression method PLS

Spectral pre-treatment Attributes EMSC selected DT2nd components SNV Abbreviations: PLS = Partial Last Squares, EMSC = Extended Multiplicative Scatter Correction, DT2nd = second derivatives, SNV = Standard Normal Variate

Generation of xylogram Dedicated software for creation of xylograms was developed in LabVIEW. The spectral preprocessing included computation of second derivatives for all treated samples. The degradation coefficient of thermal treatment cdeg. TT was calculated according to modified formula which has been published in SANDAK et al. (2016) as equation (1).

cdeg. TT   

sMIN    sTT  

(1)

sMIN    sMAX  

where: sMIN – the value of DT2nd absorbance spectra of reference (untreated) wood at selected wavelength, sTT – the value of DT2nd absorbance spectra of treated wood at selected wavelength, sMAX– the value of DT2nd absorbance spectra of extensively treated wood (240 °C, 8 hours) at selected wavelength, λ – the wavelength of the infrared light corresponding to the particular functional group.

The outer perimeter corresponds to cdeg. TT = 0 and indicates negligible changes to the NIR spectra. All the results plotted within the central part of the xylogram indicate significant changes to the NIR spectra and extensive degradation of the corresponding component/functional group. A value of cdeg. TT = 1 indicates a fully degraded chemical component. The expected degradation pattern is that the cdeg. TT values gradually change from the outer to the inner part of the xylogram, following the acquired thermal treatment dose. It 33

was intended that the reference points were sorted according to wavenumber value but were not grouped according to chemical component and functional group.

RESULTS AND CONCLUSION Various intensity of heat treatment leads to mass loss and colour change due to modification process. The ML of wood during treatment is a key characteristic and it is often used for expressing the changes in the treated wood properties. The ML reflects the heat treatment process intensity as are shown in Figure 1. The ML varied from 0.0 % (100 °C, 1 h) to 35.8 % (240 °C, 5 h). ML depended on the temperature and duration what is in accordance with the state-of-the-art knowledge. Similar results were reported by ESTEVES et al. (2008a), who determined that mass loss varies between 0.2 % (170 °C, 1 h) up to 12.0 % (200 °C, 12 h). It was found that different coupled process parameters had comparable ML, for example: ML of ~ 6 % can be achieved when treat wood in 200 °C for 5 h or in 220 °C for 1 h. It has to be mentioned that wood degradation is more intense in the presence of atmospheric air due to extensive oxidation reactions. Moreover, acetic acid produced in such process acts as an additional depolymerisation catalyst. It was previously reported that there is a higher content of acetic acid released during wood thermal treatment in the oxidizing environment (ESTEVES et al. 2008a, 2007, STAMM 1956).

Fig. 1 Mass loss (%) of pine wood during its thermal treatment from 100 °C till 240 °C for 1, 3 and 5 hours.

The appearance and the average value of colour parameters (CIE L*, CIE a* and CIE b*) of heat treated wood are shown in Figures 3, 4, 5 - parts II and III. The CIE L* was the most sensitive parameter clearly related to the treatment intensity. There was a clear tendency of darkening with increasing of heat treatment temperature and time. In contrary, CIE a* and CIE b* parameters changed relatively slightly, when compare to CIE L*. The same tendency of CIE L*, CIE a* and CIE b* variations due to thermal treatment was reported

by TOKER et al. (2016), KAMPERIDOU et al. (2013), AKSOU et al. (2011) and BEKHTA & NIEMZ (2003). Colour is an essential wood property for the final consumer. Particularly, it is the determining factor for the selection of a specific wood product for the decorative/visual function ESTEVES et al. (2008b).

Figure 2 presents the second derivative of an averaged near infrared absorbance spectra for thermally treated and reference (untreated) wood. The range of variations is limited only to spectral bands that can be interpreted and associated with well-defined functional groups, which are listed in table 3. Independently to applied thermal treatment (from 100 °C for 1 hour to 240 °C for 5 hours), the spectra present similar trends with typical broad vibration 34

bands associated to the chemical components of wood. Consequently this trend was figured upon their range, which was determined as the minimum (Figure 2 - green line) and maximum (Figure 2 - red line) value. There was clean tendency of decreasing or increasing of absorbance at various bands. These spectral changes caused by heat treatment were well corresponding to previous reports for heated pine (RIDLEY-ELLIS et al. 2014), larch (YANG et al. 2018) and spruce (POPESCU et al. 2018, BÄCHLE et al. 2010).

Fig. 2 The range of 2ND derivative of absorbance of FT-NIR spectra on thermally treated and reference samples (Note: used bands 1-20 correspond to components listed in Table 3).

The thermal modification of pine sapwood results in changes of the chemical composition of wood. Closer examination of the xylograms together with changes of absorbance peaks provides additional information regarding kinetics of chemical changes due to treatment temperature and duration (Figures 3, 4, 5 - parts IV. and V.). The most evident variations in NIR spectra, being consequence of changes in chemical composition of functional groups, were observed for the treatments with temperature over 200 °C. On the other hand, cdeg. TT had small values when heating wood at 100 °C or other mild treatment temperatures (150 °C and 160 °C).

The absorption band at 4202 cm1 (band 1 in Table 3 and Figures 3, 4, 5 - part V.) is assigned to the second overtone of –OH deformation of holocellulose. A gradual decrease of the absorbance occurred for all treatment durations; however, the changes were more intense for 3 and 5 hours heat treatment.

Hemicelluloses are polysaccharide with lower degree of polymerization than cellulose.

The absorption bands present at the wavenumber 4403 cm-1 (3), 5882 cm-1 (12) and 5802 cm-1 (11) assigned to furanose/pyranose are due to –CH2 stretching and deformation and – CH stretching. A shift in the peak position towards the higher wavelength region occurred with increase of the temperature. This confirms that the physical-chemical structure of the hemicelluloses changes rapidly and its content decreases with the temperature increase (YILDIZ & GÜMÜŞKAYA 2007, SANDAK et al. 2016).

In wood, cellulose has a strong interaction with water due to three hydroxyl groups attached to the glucopyranose ring. The absorption bands assigned to the first overtone of the fundamental –OH stretching mode were identified at wavenumber 4403 cm-1 (3), 4748 cm-1 (6), 6140 cm1 (14), 6490 cm1 (16), 6622 cm1 (17), 6789 cm1 (18). The absorption band at wavenumber 7005 cm1 (19), assigned to -OH groups of amorphous regions of cellulose and water shows clear tendencies of its decrease with augmented temperature. The lower degradation intensity was observed for semi-crystalline (4806 cm1 35

(7), 5463 cm1 (9), 5590 cm1 (10)) and crystalline (6290 cm1 (15)), regions of cellulose.

Similar tendency of degradation kinetic was recorded previously in KAČÍK et al. (2015) and SIVONEN et al. (2002).

Lignin in wood is chemically and physically bonded to cellulose and hemicelluloses forming a three-dimensional polymer complex that contains acetal, α-phenyl-β-ether, phenyl-β-glucosidic and hydrogen bonds. The absorbance of the functional groups associated to lignin (4561 cm-1 (4) and 5982 cm-1 (13)) as well as assigned to lignin and extractives at 4679 cm-1 (5) was reduced in all investigated treatment configurations indicating continue lignin degradation and/or condensation.

The sorption and desorption of water is an important phenomenon which highly affects mechanical and physical properties of wood (e.g. dimensional stability, shrinkage and swelling). The NIR absorption bands at 5220 cm1 (8) and at 7005 cm1 (19) are assigned to combination of –OH stretching and –OH bending vibration modes in water. As expected, clear changes in both absorbance bands (8 and 19) were observed when the treatment temperature increased from 100 °C to 240 °C. Tab. 3 Band assignments of selected wood components after thermal modification (according to SCHWANNINGER et al. 2011).

Band 1 2 3 4 5 6 7

Wavenumber (cm1) 4202 4282 4403 4561 4679 4748 4806

8 9

5220 5463

10 11 12 13 14 15 16 17 18 19 20

5590 5802 5882 5982 6140 6290 6490 6622 6789 7005 7315

Chemical component; Bond vibration Holocellulose; O–H deformations (second overtone) Cellulose; C–H stretching, C–H2 deformation Cellulose, hemicellulose; C–H2 stretching, C–H2 deformation, Lignin C–H stretching, C=O stretching Lignin/extractives C–H stretching, C=C stretching Cellulose; O–H deformation, O–H stretching Cellulose semicrystalline and crystalline regions, O–H stretching, C–H deformations Water; O–H stretching, O–H deformations Cellulose semicrystalline and crystalline regions; O–H stretching, C–O stretching (second overtone) Cellulose semicrystalline and crystalline regions; C–H stretching (first overtone) Hemicellulose (furanose/pyranose); C–H stretching (first overtone) Hemicellulose; C–H stretching (first overtone) Lignin; C–H stretching (first overtone) Cellulose; O–H, stretching (first overtone) cellulose crystalline regions; O–H stretching (first overtone) Cellulose; O–H stretching (first overtone) Cellulose; O–H stretching (first overtone) Cellulose; O–H stretching (first overtone) Amorphous cellulose/water; O–H stretching (first overtone) Cellulose; C–H stretching (first overtone), C–H deformations

36

1 hour treatment I.

100 °C

150 °C

160 °C

200 °C

220 °C

240 °C

82.13 4.30 20.86

79.08 5.51 23.16

73.84 5.79 23.67

57.91 10.68 25.77

43.82 12.31 20.00

32.75 9.07 12.18

II.

III.

L* a* b*

IV.

V.

Note: I. Temperature of heat treatment, II. Appearance of heat treated wood, III. Colour of heat treated wood (colour coordinates after heat treatment – CIE L*, CIE a* and CIE b*), IV. Xylograms, V. Peak shift Fig. 3 Summary results for 1 hour heat treatment of pine sapwood.

37

3 hours treatment I.

100 °C

150 °C

160 °C

200 °C

220 °C

240 °C

81.82 4.45 22.11

75.09 6.05 24.73

73.81 6.00 23.97

47.45 10.16 22.13

34.61 9.05 13.27

30.26 5.00 5.36

II.

III.

L* a* b*

IV.

V.

Note: I. Temperature of heat treatment, II. Appearance of heat treated wood, III. Colour of heat treated wood (colour coordinates after heat treatment – CIE L*, CIE a* and CIE b*), IV. Xylograms, V. Peak shift Fig. 4 Summary results for 3 hours heat treatment of pine sapwood.

38

5 hours treatment I.

100 °C

150 °C

160 °C

200 °C

220 °C

240 °C

80.66 4.83 21.72

75.92 6.09 25.46

73.65 7.15 25.52

44.71 10.08 20.68

31.59 8.38 10.97

28.31 3.83 1.21

II.

III.

L* a* b*

IV.

V.

Note: I. Temperature of heat treatment, II. Appearance of heat treated wood, III. Colour of heat treated wood (colour coordinates after heat treatment – CIE L*, CIE a* and CIE b*), IV. Xylograms, V. Peak shift Fig. 5 Summary results for 5 hours heat treatment of pine sapwood.

39

CONCLUSSIONS Treatment duration and temperature, beside of the oxygen concentration, are principal factors that affect wood chemical change and its final appearance due to thermal modification. The analyses of FT-NIR spectra provided essential information about chemical changes of wood components after that process. It was confirmed that heattreatment of wood at elevated temperatures (over 200 °C) caused extensive destruction of hemicelluloses. Some extent of semi-crystalline cellulose and lignin degradation was also noticed but in a smaller degree. Therefore, the number of sorption sites available to link water with wood was dramatically reduced due to thermal treatment.

Profound understanding of chemical changes might be helpful for further optimization of the thermal treatment procedures at industrial scale. For that reason, xylograms are identified as a simple and illustrative method that might be highly suitable for visualization how thermal treatment effects on the chemical composition of wood. The same method may be implemented for studies on alternative modification/degradation processes of wood and other lignocellulosic materials.

REFERENCES ALTGEN, M., HOFMANN, T., MILITZ, H. 2016. Wood moisture content during the thermal modification process affects the improvement in hygroscopicity of Scots pine sapwood. In Wood science and technology, 50(6): 1181–1195.

AKSOU, A., DEVECI, M., BAYSAL, E., TOKER, H. 2011. Colour and gloss changes of Scots pine after heat modification. In Wood Research, 56(3): 329–336.

BÄCHLE, H., ZIMMER, B., WINDEISEN, E., WEGENER, G. 2010. Evaluation of thermally modified beech and spruce wood and their properties by FT-NIR spectroscopy. In Wood science and technology, 44(3): 421–433.

BEKHTA, P., NIEMZ, P. 2003. Effect of high temperature on the change in color, dimensional stability and mechanical properties of spruce wood. In Holzforschung 57: 539–546 ČABALOVÁ, I., KAČÍK, F., KAČÍKOVÁ, D., ORAVEC, M. 2014. The influence of cross-section of spruce wood specimens on saccharides changes at thermal loading. In Acta Facultatis Xylologiae Zvolen, 56(2): 81–86.

ESTEVES, B., DOMINGOS, I., PEREIRA, H. 2008a. Pine wood modification by heat treatment in air. In BioResources, 3(1): 142–154.

ESTEVES, B., MARQUES, A.V., DOMINGOS, I., PEREIRA H. 2007. Influence of steam heating on the properties of pine (Pinus pinaster) and eucalypt (Eucalyptus globulus) wood. In Wood Science and Technology, 41(3): 193–207.

ESTEVES, B., MARQUES, A. V., DOMINGOS, I., PEREIRA, H. 2008b. Heat-induced colour changes of pine (Pinus pinaster) and eucalypt (Eucalyptus globulus) wood. In Wood Science and Technology, 42(5): 369–384.

ESTEVES, B., PEREIRA, H. 2008. Quality assessment of heat-treated wood by NIR spectroscopy. In European Journal of Wood and Wood Products, 66(5): 323–332.

HILL, C. A. 2007. Wood modification: chemical, thermal and other processes (Vol. 5). John Wiley & Sons.

GÉRARDIN, P. 2016. New alternatives for wood preservation based on thermal and chemical modification of wood—a review. In Annals of forest science, 73(3): 559–570.

HAKKOU, M., PÉTRISSANS, M., GÉRARDIN, P., ZOULALIAN, A. 2006. Investigations of the reasons for fungal durability of heat-treated beech wood. Polymer degradation and stability, 91(2), 393-397.

HILLIS, W. E. 1984. High temperature and chemical effects on wood stability. In Wood Science and Technology, 18(4): 281–293.

40

KAČÍK, F., ĽUPTÁKOVÁ, J., ŠMÍRA, P., NASSWETTROVÁ, A., KAČÍKOVÁ, D., VACEK, V. 2016.

Chemical alterations of pine wood lignin during heat sterilization. In BioResources, 11(2): 3442– 3452.

KAMPERIDOU, V., BARBOUTIS, I., VASILEIOU, V. 2013. Response of colour and hygroscopic properties of Scots pine wood to thermal treatment. In Journal of Forestry Research, 24(3): 571–575.

KUČEROVÁ, V., LAGAŇA, R., VÝBOHOVÁ, E., HÝROŠOVÁ, T. 2016. The effect of chemical changes during heat treatment on the color and mechanical properties of fir wood. In BioResources, 11(4): 9079–9094.

MARTINKA, J., KAČÍKOVÁ, D., RANTUCH, P., BALOG, K. 2016. Investigation of the influence of spruce and oak wood heat treatment upon heat release rate and propensity for fire propagation in the flashover phase. In Acta Facultatis Xylologiae Zvolen. 58(1): 5–14.

MEHROTRA, R., SINGH, P., KANDPAL, H. 2010. Near infrared spectroscopic investigation of the thermal degradation of wood. In Thermochimica Acta, 507: 60–65.

MILITZ, H. 2002. Thermal treatment of wood: European processes and their background. In: International Research Group Wood Preservation, IRG/WP 02- 40241, Cardiff, Wales POPESCU, C. M., NAVI, P., PEÑA, M. I. P., POPESCU, M. C. 2018. Structural changes of wood during hydro-thermal and thermal treatments evaluated through NIR spectroscopy and principal component analysis. In Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 191: 405–412.

PRIADI, T., HIZIROGLU, S. 2013. Characterization of heat treated wood species. In Materials & Design, 49: 575–582.

REINPRECHT, L., VIDHOLDOVÁ, Z. 2011. Termodrevo. Ostrava : Šmíra-print. 89 p.

RIDLEY-ELLIS, D., POPESCU, C. M., KEATING, B., POPESCU, M. C., HILL, C. A. 2014. Stiffness changes during low temperature thermal treatment of Scots pine, assessed by acoustic NDT. In 7th European Conference on Wood Modification, Lisbon, Portugal, (Session 2B, Paper 1). 7 p.

SANDAK, A., SANDAK, J., ALLEGRETTI, O. 2015. Quality control of vacuum thermally modified wood with near infrared spectroscopy. In Vacuum, 114: 44–48.

SANDAK J., SANDAK A., ALLEGRETTI O. 2016. Chemical changes to woody polymers due to high temperature thermal treatment assessed with near infrared spectroscopy. In Journal of Near Infrared Spectroscopy, Wood NIR Special Issue 24, 6: 555–562.

SANDBERG, D., KUTNAR, A. 2016. Thermal modified timber (TMT): recent development in Europe and North America. In Wood and Fiber Science, 48: 28–39.

SCHWANNINGER, M., RODRIGUES, J.C., FACKLER, K. 2011. A review of band assignments in near infrared spectra of wood and wood components. In Journal of Near Infrared Spectroscopy 19: 287– 308. ŠUŠTERŠIC, Ž., MOHAREB, A., CHAOUCH, M., PÉTRISSANS, M., PETRIČ, M., GÉRARDIN, P. 2010.

Prediction of the decay resistance of heat treated wood on the basis of its elemental composition. In Polymer Degradation and Stability, 95(1): 94–97.

SIVONEN, H., MAUNU, S. L., SUNDHOLM, F., JÄMSÄ, S., VIITANIEMI, P. 2002. Magnetic resonance studies of thermally modified wood. In Holzforschung, 56(6): 648–654.

STAMM, A. J. 1956. Thermal degradation of wood and cellulose. In Industrial & Engineering Chemistry, 48(3): 413–417.

TODOROVIC, N. V., POPOVIĆ, Z., MILIĆ, G., POPADIĆ, R. 2012. Estimation of heat-treated beechwood properties by color change. In BioResources, 7(1): 0799–0815.

TOKER, H., BAYSAL, E., KOTEKLI, M., TURKOGLU, T. T., KART, S., SEN, T. F.,PEKER, T. H. 2016.

Surface characteristics of oriental beech and scots pine woods heat-treated above 200 C. In Wood Research, 61: 4354.

TSUCHIKAWA, S., KOBORI, H. 2015. A review of recent application of near infrared spectroscopy to wood science and technology. In Journal of Wood Science, 61(3): 213220.

YANG, S. Y., HAN, Y., CHANG, Y. S., PARK, J. H., PARK, Y., CHUNG, H., YEO, H. 2018. Classification of the hot air heat treatment degree of larch wood using a multivariate analysis of near-infrared spectroscopy. In Journal of Wood Science, 64(3): 220225.

YILDIZ, S., GÜMÜŞKAYA, E. 2007. The effects of thermal modification on crystalline structure of cellulose in soft and hardwood. In Building and Environment, 42(1): 62–67.

YILDIZ, S., YILDIZ, U. C., TOMAK, E. D. 2011. The effects of natural weathering on the properties of heat-treated alder wood. In BioResources, 6(3): 2504–2521.

41

VIITANIEMI, P., JAMSA, S., EK, P., VIITANEN, H. 1997. Method for improving biodegradation resistance and dimensional stability of cellulosic products. U.S. Patent No. 5,678,324. Washington, DC: U.S. Patent and Trademark Office.

WELZBACHER, R.C., RAPP, O.A. 2007. Durability of thermally modified timber from industrial-scale processes in different use classes: Results from laboratory and field tests. In Wood Material Science and Engineering, 2(1): 4–14.

WELZBACHER, R.C., BRISCHKE, C., RAPP, O.A. 2007. Influence of treatment temperature and duration on selected biological, mechanical, physical and optical properties of thermally modified timber. In Wood Material Science and Engineering, 2(2): 66–76.

ACKNOWLEDGEMENT This work was supported by the Slovak Research and Development Agency under the contract No. APVV-17-0583. Part of this research was conducted within BIO4ever (RBSI14Y7Y4) project funded within a call SIR by MIUR. The authors gratefully acknowledge the European Commission for funding the InnoRenew CoE project (Grant Agreement #739574) under the Horizon 2020 Widespread-Teaming program. We thank European mobility programmes Erasmus+ for financial support the mobility to CNR-IVALSA in San Michele all Adige (Italy) and COST action FP1407 for networking opportunity.

AUTHORS ADDRESS Zuzana Vidholdová Technical University in Zvolen Faculty of Wood Sciences and Technology Department of Mechanical Wood Technology T. G. Masaryka 24 960 53 Zvolen Slovak Republic Corresponding author: zuzana.vidholdova@tuzvo.sk Anna / Jakub Sandak Trees and Timber Institute CNR-IVALSA via Biasi 75 38010 San Michele all Adige Italy Corresponding author: anna.sandak@ivalsa.cnr.it Jakub Sandak InnoRenew CoE Livade 6, 6310 Izola Slovenia e-mail: jakub.sandak@innorenew.eu Jakub Sandak University of Primorska Andrej Marušič Institute Muzejski trg 2 SI-6000 Koper Slovenia

42

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 43−52, 2019 Zvolen, TechnickĂĄ univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.04

THE EFFECT OF TEMPERATURE AND MOISTURE CHANGES ON MODULUS OF ELASTICITY AND MODULUS OF RUPTURE OF PARTICLEBOARD Sergiy Kulman – Liudmyla Boiko – Diana HamĂĄry GurovĂĄ – JĂĄn SedliaÄ?ik ABSTRACT Taking into account nonlinear effects in the reaction of wood composite materials during the thermal, moisture and power loads, the strength phenomenological model can be created. Bending strength and modulus of elasticity of different types of particleboards were studied at the temperatures of 20, 40, 60, 80 and 100 °C and moisture content of 6, 8, 9, 11 and 15 % using the standard tensile testing machine. Based on the results of the tests, permanent members of the phenomenological model were determined describing adequately the strength and rigidity of particleboard.

Key words: non-linear effects, bending strength, modulus of elasticity in bending, strength phenomenological model

INTRODUCTION In order to utilize the full potential of wood composite materials (WCM), such as particleboard (PB) as building material, the changes of mechanical properties during exposure to different temperature and relative humidity must be known (AYRILMIS et al.

2010, BEKHTA and MARUTZKY 2007, BEKHTA and NIEMZ 2009). For reasons of safety, it is especially important to know the strengths of load bearing WCM structures under thermal, humidity and power conditions (DEXIN and Ă–STMAN 1983, SUZUKI and SAITO 1987, KULMAN et al. 2015). However, analytical methods are unavailable which would predict the performance of load structural WCM members during loads, yet. Therefore, the objective of this investigation is to develop a model which can be used to predict changes in the tensile and compressive properties of WCM during changes temperature and moisture content (MC) exposure.

Usually the strength calculations are carried out according to Hooke's law, which assumes a linear dependence of stresses and deformations. In the classical approach to the problem of strength accepted that failure occurs when a certain combination, which includes stress, strain, temperature, and some other parameters (describing the state of the material and its specific properties) reaches a critical value (PANASIUK 1988). It is considered that in the space of all possible values of these parameters there is a closed surface is described by the relation: đ?œ‘ = đ??š(đ?œŽ, đ?œ€, đ?‘‡, đ?‘Š, đ??śđ?‘› )

43

(1)

where Cn – parameters characterizing the properties of the macro real body volume determined experimentally. This limits the scope of permissible surface, in terms of strength, the material states.

The specific form of the phenomenological relation (1) for each material is established based on accepted postulates (hypotheses) about the destruction and the necessary experimental research on setting environments Cn. However, the generally accepted equations of limit states, which would explicitly take into account the effect of time, temperature, humidity, and the load is not created.

Since WCM for different thermal, moisture, power conditions will be in different states, tensile strength and modulus of elasticity (MOE) will differ its deviation from the ideal state in conditions in which the material has a maximum strength and stiffness. The phenomenological equation of state of the material can be expressed in the general form:

 T ,W ,t  f ( 0 , T ,W , t )

(2)

ET ,W ,t  f ( E0 , T ,W , t )

(3)

where σT,W,t – limit strength in its various states; σ0, E0 – the maximum possible value of the tensile strength and elastic modulus of the material under ideal conditions, i.e. in the absence of external influences, MPa; Т – thermostat temperature (ambient) K; W – moisture content, %; t – time, s.

In this paper, the equation of state without taking into account the time factor and scale factors is proposed in the form of an autonomous system. Then, at t = 0 the equation of state describes the short-term strength for the short quasi-static loading at a given temperature humidity conditions. When t  0 equation (2), (3) will be described by equations of state under long-term loading, i.e. the long-term strength.

Phenomenological model builds on the behaviour of the object as the result of a process, the essence of which is generally understood approximately, but details are not yet clear. In the model introduced some "constants", describing the specific behaviour of the object, with specification of the object, but without specifying the exact meaning of these “permanents” (PRIGOGINE and KONDEPUDI 1999).

The aim of this study is to develop a phenomenological model of strength and stiffness for composite materials based on wood under the influence of thermo – moisture – power loads, which takes into account the non–linear nature of the reaction this material to the external influences.

MATERIAL AND METHODS Three types of particleboard bonded with urea-formaldehyde resin commercially produced by Kronospan UA Ltd. were used in this study: melamine faced particleboard (MF PB), oak veneered particle board (VF PB) and particleboard P2 (P2 PB) according to EN 312 type P2. Test pieces with the thickness of 18 × 450 × 50 mm were cut from each type of board. Before testing, pieces were conditioned at 20 °C and 65% RH. The average densities of specimens were 757 kg/m3, 792 kg/m3, 733 kg/m3 and moisture content 5%.

Static 3-point bending tests were carried out in the universal test machine with temperature-controlled chamber at the speed deformation 2 mm/min. Investigated temperatures were 20 °C, 40 °C, 60 °C, 80 °C and 100 °C. Investigated moisture contents were 6 %, 8 %, 9%, 11% and 15 %.

Methods for determining the ultimate strength and modulus of elasticity in bending are described in detail in a previous paper (KULMAN et al. 2017). However, the methodology 44

for constructing phenomenological models of strength and rigidity was fundamentally different (KULMAN 2017).

As during model building considering the strength, only elastic deformation, the tensile strength and modulus of elasticity in the elastic deformation depends linearly on the external thermo – moisture – power of influence. Therefore, the rate of destruction in this area can be assumed as constant, and the process of maintaining strength (loss of strength), is considered analogous reaction of zero order. The elastic deformation order of reaction does not change. The statement that the order "of the reaction strength loss" does not change based on the curve also the "force – displacement" analysis when tests using a standard tensile strength machine. Character of the curve changes slightly for different temperatures (KULMAN et al. 2015). In addition, a kinetic measurement is to determine the long-term strength of particleboard (BOIKO et al. 2013) talking about the fact that the order of reaction process of losing long–term strength also does not change. And while there is the temperature dependence of the speed of this process following the simplest form, the Arrhenius equation (STILLER 1989):

k (T )  Ae

 EA RT

 Ae

TA T

 AeTe

(4)

where: А, EA – independent constants (or nearly independent) in the temperature range studied; EA – observed (imaginary) activation energy, kJ/mol; R = 0.0083 kJ/mol*K – universal gas constant; TA = EA / R – activation temperature, K. In this case, the activation temperature is equal to the upper temperature limit at which the body has the properties to withstand an external load, that is, the sublimation temperature TA  Tm ; Te  T / Tm – efficient temperature, which characterizes the deviation of the current temperature of the test by limiting the activation temperature in the range of operating temperatures T  Tmin ...Tmax . А – pre-exponential factor, which is postulated on the basis of the purpose for which is compiled by the Arrhenius equation. Its value can be determined based on conducted tests. In our case, this is the maximum possible strength of the material in the case of the minimum temperature, that is, when Tmin = 0 °K. Arrhenius equation shows that at constant temperature, constant speed k(T) of a process is determined by the activation energy. The higher the numerical value of the activation energy EA, the less active molecules (intermolecular bonds), the smaller the number of effective unbroken bonds and the thus lower the rate constant and the strength loss rate itself. The higher activation energy, the more difficult to break the bonds between the molecules and the higher strength. In modern interpretation, Arrhenius equation determines not only the temperature dependence of the process rate k, for example, the chemical reaction rate, but the rate of diffusion, longevity, relaxation period, the option of destruction. Thus in each case the value included in this equation have a different interpretation (STILLER 1989). Most important constant of integration А, (pre-exponential factor) is interpreted as a constant at the threshold included in the equation of variables that determine the nature of external influence (temperature, load, humidity, etc.). Moreover, A and EA – constants, independent (or nearly independent) in the investigated temperature range.

Taking all the above mentioned assumptions can be argued that the experimentally observed dependence should tensile strength and modulus of temperature in the form of the Arrhenius equation:

W ,T   0eT ,

(5)

e

45

where σW,T – actual, current strength, the strength at the current moisture content W (%) and temperature Т (°K); σ0 – constant equal to the maximum for the material tensile strength at W = 0,% and Т = 0,°K;  – constant coefficient, which characterizes the degree of influence of temperature taking into account other factors, and their interaction. In the case of recording only one factor influence, temperature:  W 0%  1 .

By analogy with the effect of temperature on the strength postulate influence of humidity on the strength of typing in the equation (2), (3) efficient moisture We. In addition, the construction of the model will take into account its non-linearity, entering into the equation the factor of interaction between effective temperature and moisture, in the form of their multiplication TeWe (KULMAN 2011, KULMAN and BOIKO 2016). The equation describing the phenomenological nonlinear model of strength and stiffness in the form of the Arrhenius equation takes the form:

W ,T   0eT eW eW / T

e

(6)

EW ,T  E0eTe eWe eWe / Te

(7)

e

e

e

where: σW,T – actual, current strength, that strength at the current moisture content W (%) and temperature Т (°K); We = (Wm – W)/Wm – effective moisture; Wm – maximum permissible moisture content of the material in which it has sufficient elastic properties for use, %; W - current moisture during operation, %; Te = (Tm – T)/Tm – effective temperature; Tm – temperature limit being material to take external loads sufficient for its operation, °K; T – current temperature material during its operation, °K; α, β, γ, δ, ε, θ – constant coefficients; α, δ – take into account the effect of temperature on the material tensile strength and modulus of elasticity; β, ε – take into account the effect of moisture content on the tensile strength and modulus of elasticity; γ, θ – the change in strength properties of the material in the joint action of humidity and temperature is not linear process of changing strength; EW,T – current modulus of elasticity, MPa; E0 – constant factor equal to the theoretically maximum possible for the material modulus of elasticity at W = 0,% ; Т = 0, °K. Moreover, the values of σ0, E0, α, β, γ, δ, ε, θ are determined by solving the system of equations: We = (Wm – W)/Wm

We1   ln  W1T1  ln  0  We1  Te1   T  e1   We 2   ln  W2T2  ln  0  We 2  Te 2   T   e2    We3  ln   ln    W   T   e3 e3 0  W3T3 Te 3    We 4  ln   ln  0  We 4  Te 4    W4T4 Te 4 

46

(8)

We1   ln EW1T1  ln E0  We1  Te1   T  e1   We 2   ln EW2T2  ln E0  We 2  Te 2   T   e2    ln E  ln E  W  T   We3  W3T3 e3 e3 0  Te3    ln E  ln E    T   We 4  W4T4 e4 0  We 4 Te 4 

(9)

where Te1, Te2, Te3, Te4 – effective temperature series of four tests, °K; We1, We2, We3, We4, – effective material moisture during the four-test series, %; σW1T1, σW2T2, σW3T3, σW4T4 – modulus of rupture at the appropriate temperature and MC, MPa; EW1T1, EW2T2, EW3T3, EW4T4 – modulus of elasticity at the appropriate temperature and MC, MPa.

RESULTS AND DISCUSSION The experimental factor levels and test results are shown in Table 1. For example summarized ruptures envelope for load-displacement curves in coordinates stress-rupture time for MF PB under different conditions shown in Fig.1. In all cases of tests carried out with increasing temperature and humidity, the strength and modulus of elasticity of the material decreased while increasing the time to failure.

Tab. 1 Experimental factors levels and test results for particleboards.

Test conditions Board type

MF PB

VF PB

P2 PB

a

Test group number 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

Temperature (°C)

Moisture content (%)

20 40 60 80 100 20 40 60 80 100 20 40 60 80 100

6 8 9 11 15 6 8 9 11 15 6 8 9 11 15

Test results modulus of bending elasticity in strength bending (MPa) (MPa) 16.80 ± 1.17a 1866.5 ± 95a 15.55 ± 0.93 1727.6 ± 88 13.80 ± 0.62 1533.2 ± 74 11.30 ± 0.15 1255.4 ± 36 8.00 ± 0.12 922.1 ± 22 19.68 ± 0.55 2582.0 ± 112 19.00 ± 0.49 2492.8 ± 95 17.50 ± 0.38 2296.0 ± 86 15.30 ± 0.33 2007.4 ± 78 12.00 ± 0.35 1674.4 ± 60 14.80 ± 0.44 1571.9 ± 65 13.40 ± 0.33 1423.2 ± 56 11.70 ± 0.21 1242.7 ± 48 9.50 ± 0.15 1009.0 ± 33 7.50 ± 0.11 755.3 ± 23

The confidence interval is indicated at p = 0.05 level.

For each species, analysis of variance (ANOVA) was conducted to study the effect of temperature and MC on the MOR and MOE at a 0.05 significance level. Results of ANOVA and multiple comparison statistical analysis for temperature and moisture content are shown in Table 2.

47

The significance value for MOR and MOE between 20 °C and 100 °C, and for moisture content between 6% and 15%, were less than 0.05, indicating that the effect of different temperatures on MOR and MOE for this pairs are statistically significant. The ANOVA results showed that the temperature had a more significant effect than MC on the MOR of MF PB, but MC had a more significant effect than temperature on the MOE. Tab. 2 ANOVA for bending strength and modulus of elasticity in bending for particleboards.

Dependent variable for board type Bending strength, MF PB Modulus of elasticity in bending, MF PB Bending strength, VF PB Modulus of elasticity in bending, VF PB Bending strength, P2 PB Modulus of elasticity in bending, P2 PB

Source

SS a

df b

MS c

F ratio

p value

Moisture content Temperature Error Total Moisture content Temperature Error Total Moisture content Temperature Error Total Moisture content Temperature Error Total Moisture content Temperature Error Total Moisture content Temperature Error Total

9.135 153.686 0.257 163.078 49612.5 2864606 10450 2924668.5 2.205 194.3 0.1 196.6 92820 2768343 3742 2864966 0.77 1723 0.184 1733 23312 2120947 2895 2149154

4 4 16 24 4 4 16 24 4 4 16 24 4 4 16 24 4 4 16 24 4 4 16 24

2.2839 38.4216 0.016

142.41 2395.73

0.000 0.000

12403.1 716151.5 653.125

19.99 1096.49

0.000 0.000

0.55 48.58 0.006

96.92 8540.7

0.000 0.000

23220 692086 2339

99.28 2959

0.000 0.000

0.19 431 0.0115

16.7 3742.4

0.000 0.000

6328 530236 180.9

34.9 2930

0.000 0.000

– sum of squares – degree of freedom c MS – mean square a SS b df

Based on the data obtained in the experiments, the constants in phenomenological models of MOR (6) and MOE (7) for each particleboard type were calculated using the system of equations (8) and (9). The results of calculations are presented in Tab. 3.

Tab. 3 Results of calculating the constants in phenomenological models of strength (MOR) and stiffness (MOE) for each particleboard types.

Constants in model of MOR

Board type

0

MF PB VF PB P2 PB

(MPa) 90.92 126.27 171.15

Constants in model of MOE

α

β

γ

E0

14.13 10.30 15.55

9.23 6.58 10.11

1.56 1.43 2.03

(MPa) 15675 21605 9238

ε

δ

θ

15.45 13.10 13.45

9.91 8.02 9.03

1.81 1.55 1.68

Using parameter is found strength using formulas (6), (7) calculate that a bootie design tensile strength and elastic modulus parts running on pure driving together when the temperature operating range from 20 to 100 °C and moisture content from 6% to 15%.

48

Fig. 1, 2 and 3 shows the results of calculations of strength and modulus of rupture depending on the temperature – moisture conditions for three types of particleboards.

Fig. 1 Changes mean bending strength (MPa) melamine faced particleboard depends from temperature (°C) and MC (%)

Fig. 2 Changes mean bending strength (MPa) veneered faced particleboard depends from temperature (°C) and MC (%)

Fig. 3 Changes mean bending strength (MPa) particleboard (P2 PB) depends from temperature (°C) and MC (%)

49

Fig. 4, 5 and 6 show the results of calculations of stiffness like as modulus of elasticity in bending depending on the temperature – moisture conditions for three types of particleboards.

Fig. 4. Changes mean modulus of elasticity in bending (MPa) melamine faced particleboards depends from temperature (°C) and MC (%)

Fig. 5. Changes mean modulus of elasticity in bending (MPa) veneered faced particleboard depends from temperature (°C) and MC (%)

Fig. 6. Changes mean modulus of elasticity in bending (MPa) of particleboard type P2 depends from temperature (°C) and MC (%)

50

Phenomenological model of strength (MOR) and stiffness (MOE) presented in Fig. 1 to 6 graphically limited in the variables coordinates a surface of the second order, the ultimate surface of strength and stiffness. Geometrically, it is the hyperbolic paraboloid. It is formed by the non-linear terms in the equations (6), (7). It characterizes the interaction of temperature and moisture on the strength and stiffness properties of the material.

Taking into account nonlinear effects in the reaction of wood composite materials during the thermal, humidity and power loads allows offering strength phenomenological model. Using the model to predict the tensile strength and modulus of elasticity wood composite materials for the specific thermo – moisture – power conditions of its use.

Based on the results of the tests, permanent members of the phenomenological model were determined that adequately describes the strength and stiffness of particleboards.

CONCLUSION The first created phenomenological models of strength (MOR) and stiffness (MOE) of particleboard graphically coordinate the surface of the boundary strength and stiffness in the form of a hyperbolic paraboloid that is formed by nonlinear terms in the model equations and characterizes the interaction of temperature and humidity with the strength properties of the material. Taking into account non-linear effects in the reaction of wood composite materials under thermal, humidity and power loads, it is possible to propose phenomenological strength models and apply them to predict the ultimate strength and modulus of elasticity of particleboard for specific operating conditions. Based on the test results, the permanent members of the phenomenological model were determined, which adequately describe the strength and rigidity of the particleboard when changing external influences in a wide range. The created method allows to significantly reduce the time required for testing and determine the strength parameters of the material when changing the thermo–moisture–strength loads and makes it possible to estimate the preservation of the strength properties of new materials and to predict the strength in the work of the structures already created.

REFERENCES AYRILMIS, N., UMIT, B., NUSRET, A.S. 2010. Bending strength and modulus of elasticity of woodbased panels at cold and moderate temperatures. In Cold. Reg. Sci. Technol., 63: 4043.

BEKHTA, P., MARUTZKY, R. 2007. Bending strength and modulus of elasticity of particleboards at various temperatures. In Holz Roh-Werkst., 65: 163165.

BEKHTA, P., NIEMZ, P. 2009. Effect of relative humidity on some physical and mechanical properties of different types of fiberboard. In Eur. J. Wood Prod., 67: 339349.

BOIKO, L.M., GRABAR, I.G., KULMAN, S.M. 2013. Durability particleboards in furniture. Osvita Ukrainy, Ukraine, 2013, 210 p.

DEXIN, Y., ÖSTMAN, B.A.L. 1983. Tensile strength properties of particle boards at different temperatures and moisture contents. In Holz Roh-Werkst., 41(7): 281286.

KULMAN, S. 2011. Nonlinear effects caused deformation and fracture composite materials based on wood. http://elibrary.nubip.edu.ua/13490/1/11ksm.pdf.

KULMAN, S., BOIKO L., ANTSYFEROVA A. 2015. Bending strength (modulus of rupture) and modulus of elasticity of MDF different density at various temperature. Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology 91: 101106.

51

KULMAN, S., BOIKO, L. 2016. Non-linear effects in the reaction of wood composite materials during the thermal, humidity and power loads. Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology 95: 159165.

KULMAN, S. 2017. Method for predicting the bending ultimate strength (MOR) and modulus of elasticity in bending (MOE) composite materials based on wood. Patent 114560 – UA, 2017.

KULMAN, S., BOIKO, L., PINCHEVSKA O., SEDLIAČIK J. 2017. Durability of wood-based panels predicted using bending strength results from accelerated treatments. In Acta Facultatis Xylologiae Zvolen, 59(2): 4152.

PANASIUK, B. 1988. Fracture mechanics and strength of materials: Handbook: Volume 1. Kiev Sciences Dumka. Kiev, Ukraine, 1988, 451 p.

PRIGOGINE, I., KONDEPUDI, D. 1999 Modern Thermodynamics. From Heat Engines to Dissipative structures. Edition Odil Jacob, NY, 1999, 652 p.

STILLER, W. 1989. Arrhenius Equation and Non-Equlibrium Kinetics. 100 Years Arrenius Equation.

BSB B. G. Teubener Verlagsgesellschaft, Leipzig, 1989, 354 p.

SUZUKI, S., SAITO, F.1987. Effects of environmental factors on the properties of particleboard. I. Effect of temperature on bending properties. In Mokuzai Gakkaishi, 33(4): 298303.

ACKNOWLEDGEMENTS This work was supported by the Ukrainian Ministry of Education and Science under Program No. 2201040: “The research, scientific and technological development, works for the state target programs for public order, training of scientific personnel, financial support scientific infrastructure, scientific press, scientific objects, which are national treasures, support of the State Fund for Fundamental Research”. The authors are grateful to Ministry of Education and Science of Ukrainian for financial support of this study.

This work was supported by the Slovak Research and Development Agency under the contracts No. APVV-14-0506 and APVV-17-0583.

This work was supported by the grant agency VEGA under the project No. 1/0626/16.

AUTHOR’S ADDRESS

Assoc. Prof. Sergiy Kulman, PhD. Assoc. Prof. Liudmyla Boiko, PhD. Zhytomyr National Agroecological University, ZNAEU Department of Wood Processing Blvd Stary 7 10008 Zhytomyr Ukraine Sergiy.Kulman@gmail.com lsdesign@ukr.net Prof. Ing. Ján Sedliačik, PhD. Ing. Diana Hamáry Gurová, PhD. Technical University in Zvolen Department of Furniture and Wood Products Masaryka 24 960 53 Zvolen Slovakia sedliacik@tuzvo.sk

52

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 53−61, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.05

DRYING THE SPRUCE (PICEA ABIES L. KARST.) COMPRESSION WOOD Ivan Klement  Miroslav Uhrín  Tatiana Vilkovská ABSTRACT The aim of the paper is to analyse differences in drying rate of compression and opposite spruce wood. Compression wood samples and their corresponding normal wood samples from the opposite part of a sample log were compared under conditions of two different drying modes (t = 60 °C and t = 90/120 °C). Also differences in density, initial and final moisture content, moisture gradient, intensity of water evaporation and case-hardening, both before and after drying process, were analysed. The differences in drying rate were noticeable mostly at the beginning of drying process. Slower drying rates, both above and under the fibre saturation point, were represented by compression wood. As expected, the differences in density of compression and opposite wood in oven dry state can be reported.

Differences in initial moisture content were not found and the differences in moisture gradient were not considered significant. The negative effect of compression wood on casehardening was not confirmed. Since the compression wood is the source of difficulties in wood processing, better knowledge of these different characteristics may have positive impact on spruce wood processing.

Key words: compression wood, opposite wood, drying, spruce, moisture gradient.

INTRODUCTION Reaction wood is created in coniferous and deciduous trees as a respond to the deviation of trunk from the vertical position. Reaction wood of coniferous trees is also called compression wood, because it is created on the compression side of the trunk. At a macroscopic level, compression wood is often recognised by its characteristic colour (TIMELL 1986). Compression wood appears dark because it absorbs more light, due to the high lignin content, and scatters less light, due to thick tracheid walls. As a general rule, the intensity of the colour of compression wood increases with increasing severity of compression wood (WARENSJÖ 2003). In cross sections, compression wood is characterised by thick cell walls and rounded outline of its tracheids, increased occurrence of spaces and the absence of the inner part of the secondary wall, the S3 layer (OLLINMAA 1959). The angle of microfibrils in S2 layer is 30°– 45° compared to 15°– 30° in normal wood (NEČESANÝ 1956, GORIŠEK and TORELLI 1999). OLLINMAA (1959) also found that in compression wood, tracheids are shorter compared to normal wood. There are also less bordered pits in compression wood (TARMIAN et al. 2011). From a chemical point of view, the compression wood has a higher content of lignin than normal wood (NEČESANÝ 1956, SAITO and FUKUSHIMA 2005).

53

Compression wood has properties which adversely affect its usefulness in wood products. Thicker cell walls coupled with the greater proportion of lignin in the cell wall makes the wood denser, more impermeable and stronger in compression. The larger microfibril angle in the S2 layer reduces the tensile strength and modulus of elasticity and increases the brittleness of the wood, making it unsuitable for uses in which it is likely to experience high stresses. The larger microfibril angle also means that the wood has a higher longitudinal shrinkage on drying, but a lower transverse shrinkage. This explains the distortion on drying of pieces of wood containing both normal and compression wood (BARNETT et al.). High longitudinal shrinkage of compression wood is source of problems during drying. The problem occurs when the dried material (board) contains both compression and normal wood. This may lead to non-uniform shrinkage causing distortion in the form of spring, bow or cup.

The drying characteristics of compression wood are still not well explained. Among rare studies on compression wood drying TARMIAN et al. (2009) investigated wood of Picea abies L. Karst containing a high proportion of well-developed compression wood and put forward that compression wood had a much lower drying rate than normal wood. They found out that noticeable difference in drying of compression and normal wood is in the domain above fibre saturation point (FSP), when the free water is removed. As the moisture content decreased to the bound water domain, the compression wood was dried comparatively fast as normal wood, and time to reach the same moisture content of 12% was similar because of the higher initial moisture content of the normal wood. STRAŽE and GORIŠEK (2006) also reported, that compression wood (radially oriented spruce samples) dried slower and also to higher equilibrium moisture content compared to normal wood samples.

The aim of presented article was to analyse differences in drying of compression and opposite spruce wood and to determine differences in density, moisture, moisture gradient, intensity of water evaporation and case-hardening. Better understanding of compression wood behaviour in drying process might have a positive effect on quality of final products.

EXPERIMENTAL Materials Spruce log (Picea abies L. Karst) with content of compression wood was selected from the forests in Sielnica-BrestovĂĄ (Slovakia). The diameter of the log was 38 cm on the narrow end and length was 4 m. Based on the distribution of compression wood along the whole log, one shorter log (1m long) was cut out. Thickness of compression wood zone on both ends of the log was not smaller than 8 cm.

Fig. 1 Spruce log with marked areas of compression wood-CW and opposite wood-OW (left) and sawing pattern (right).

54

Four boards were cut out from the sapwood zone of log, according to sawing pattern (Fig. 1). Two compression wood boards and two normal wood boards from opposite part of log (opposite wood). Subsequently, 4 drying samples were cut out from each board as shown in Figure 2 (4 compression wood samples and 4 opposite wood samples), with dimensions of 30 Ă— 100 Ă— 300 mm (thickness Ă— width Ă—length).

Fig. 2 Method of samples cut out. 1-drying sample, 2-part of board from which initial moisture content sample, moisture gradient sample, case-hardening sample and density sample were cut out, 3-waste.

Methods The process of drying was conducted using a laboratory kiln Memmert HCP 108 (Memmert GmbH + Co. KG, Schwabach, Germany) at the Department of Wood Technology, Technical University in Zvolen, Slovakia. Two drying modes were used.

Drying mode No.1 was conducted with maximum drying temperature of 60 °C and No.2 with maximum drying temperature of 120 °C (Table 1). The same psychrometric difference (Δt) of 2 °C was maintained in the stage above the FSP for both drying modes.

The psychrometric difference was increased to 12 °C when all samples reached fibre saturation point (approximately 30% of moisture content). During the drying mode No. 2, when the moisture content of all samples decreased below FSP, the drying temperature increased to 120 °C without regulation of relative humidity of the ambient air. Eight samples for both drying modes were placed in the drying chamber (four compression wood and four opposite wood samples) with their end grains sealed with silicone. Weight of each sample was continuously measured every 24 hours during drying. The gravimetric method was used to determine actual moisture content.

Tab. 1 Drying modes.

Drying mode 1 2

t (°C) 60 90

above FSP ď Ş (%) 91 94

Δt (°C) 2 2

t (C°) 60 120

below FSP ď Ş (%) 52 -

Δt (°C) 12 -

The density of wood at initial and final moisture content was measured according to STN EN 49 0108. The initial and final moisture content of wood was measured using the gravimetric method according to STN EN 49 0103. The intensity of water evaporation from the surface of samples was calculated using Eq. 1 ∆đ?‘š

đ?‘” = 2đ?‘™(đ?‘?+â„Ž)¡3600¡đ?œ? (kg ¡ m−2 ¡ s−1 )

(1)

where Δm is weight loss of wood during drying, l (length), b (width), h (thickness) of dried samples and τ is drying time interval.

Moisture gradient was measured before and after the drying process. The samples were cut into three layers according to Fig. 3. The moisture content of layers was measured 55

using gravimetric method according to STN EN 49 0103. Subsequently, the moisture gradient was calculated using Eq. 2 ∆đ?‘€đ??ś = đ?‘¤đ?‘š −

∑ đ?‘¤đ?‘ 2

(%)

(2)

where wm is moisture content of the middle layer and ws is moisture content of surface layers (%).

Fig. 3 Method of sample preparation for moisture gradient determining.

Case-hardening of wood before and after drying process was determined according to STN EN 49 0645. The method of sample preparation and case-hardening measurement is illustrated in Fig. 4. The level of case-hardening was determined according to the maximum space between the layers.

Mentioned properties were measured on the all samples.

Fig. 4 Method of case-hardening sample preparation (left) and realization of case-hardening method (right).

RESULTS AND DISCUSSION Table 2 shows density, initial and final moisture content and drying time of compared samples. The average density of compression wood measured at 0% MC was 562.2 kg.m-3 whilst the density of opposite wood was 479.2 kg.m-3. It is 14.7% higher value. KLEMENT and HURĂ KOVĂ (2015) also reported a higher density of spruce compression wood (527 kg.m-3 vs.

441 kg.m-3). The reason for higher value of compression wood density are thicker cell walls containing more lignin. Another reason is up to four times higher content of latewood (WARENSJĂ– 2003, GRYC 2005, DIAZ-VAZ et al. 2009). The density of compression wood is also affected by ratio of compression wood and position in the trunk.

Drying mode No. 1.

Initial moisture content of compression wood and opposite wood samples was approximately at the same level. The drying rate of compression wood was lower compared to opposite wood, both over and under FSP. The difference in drying rate was remarkable during first 48 hours of drying process. This is also associated with a slightly higher intensity of water evaporation during this phase (Table 3). Compression wood reached the required final moisture content of 10% in 241 hours whilst the opposite wood reached this moisture content in 189 hours (Fig. 5).

56

Tab. 2 Density, moisture content and drying time of samples Drying mode

Sample CW1 CW2 OW1 OW2 CW1 CW2 OW1 OW2

1

Moisture content [%]

2

Average Density at 0% MC ρ0 (kg·m3) 570 479 554 479

Initial Moisture Content wi (%) 83.20 84.06 83.21 82.48 89.78 88.89 87.59 86.76

Final Moisture Content wf (%) 10.50 10.63 10.21 10.13 9.60 9.71 9.59 9.38

Drying Time t(h) 241 189 150.5 149

100 90 80 70 60 50 40 30 20 10 0 0

50

100

t [h]

compression wood

150

200

250

opposite wood

Fig. 5 Drying rate curves for drying mode No. 1.

Tab. 3 Weight loss of samples and intensity of water evaporation for drying mode No. 1.

Drying Time t (h) 0 24 48 72 96 121 145 169 189 193 217 241

Drying Mode No. 1 Weight Loss of Samples m (kg) CW OW 0.046 0.034 0.030 0.027 0.023 0.020 0.019 0.043 0.005 0.021 0.012

0.048 0.036 0.031 0.026 0.022 0.017 0.016 0.040

Intensity of Water Evaporation g (kg· m-2·s-1) CW OW 6.84503E-06 5.11485E-06 4.39666E-06 3.96189E-06 3.26211E-06 2.98255E-06 2.81784E-06 7.58902E-06 4.42493E-06 3.06126E-06 1.71283E-06

7.1596E-06 5.26769E-06 4.56137E-06 3.91144E-06 3.07692E-06 2.58191E-06 2.37417E-06 7.1045E-06

Drying mode No. 2.

The differences in drying rate curves of compression and opposite wood were remarkable during first 24 hours of drying. The opposite wood reached higher intensity of water evaporation than the compression wood during this phase. Subsequently, the drying rate curves of both wood were similar. The overall time required to reach the desired final

57

moisture content was very similar. Compression wood reached the required final moisture of 10% in 150.5 hours and the opposite wood in 149 hours (Fig. 6).

moisture content [%]

100,00 90,00 80,00 70,00 60,00 50,00 40,00 30,00 20,00 10,00 0,00 0

50

100

t [h]

compression wood

150

200

250

opposite wood

Fig. 6 Drying rate curves for drying mode No. 2.

Tab. 4 Weight loss of samples and intensity of water evaporation for drying mode No. 2.

Drying Time t(h) 0 24 48 74 98 122 144 149 150.5

Weight Loss of Samples m (kg) CW OW 0.059 0.049 0.054 0.048 0.048 0.041 0.102 0.034

0.076 0.040 0.038 0.030 0.026 0.021 0.050

Intensity of Water Evaporation g (kg·m2·s1) CW OW 8.7755E-06 7.2264E-06 7.3759E-06 7.0854E-06 7.0557E-06 6.6531E-06 7.2365E-05 8.0926E-05

1.1277E-05 5.9963E-06 5.157E-06 4.4649E-06 3.8654E-06 3.4318E-06 3.583E-05

The difference in drying intensity of compression and opposite wood is due to differences in the anatomical structure (TARMIAN and PERRÉ 2009). Longitudinal and radial permeability of compression wood is lower compared to opposite wood (TARMIAN et al.

2009). The opposite wood dries remarkably faster compared to compression wood during the phase of free water removal due to larger and more numerous bordered pits of axial tracheids. Similar, the extended process of diffusion in compression wood is associated with a higher density of wood caused by less permeable, thick cell walls. Our results are consistent with TARMIAN et al. 2009. The moisture loss in compression wood over FSP was lower. WILLIAMS (1971) and DAVIS et al. (2002) mentioned that the differences in drying rate curves of compression and opposite wood gradually decrease as the moisture content approaches the FSP. This was not confirmed by our measurements. Compression wood dried slower even during the phase of bound water removal, this can be caused by a different structure of compression wood. Compression wood reached the required final moisture content substantially later than opposite wood under the conditions of drying mode No.1.

When the drying mode No. 2 was used, the compression wood dried slower than opposite wood in the area below the FSP. The time necessary to reach the required final moisture content was 1.5 hours longer.

58

The moisture gradient values are shown in Table 5. Values of moisture gradient were similar before the drying process. The negative moisture gradient of some samples was caused due to freezing of samples and subsequent condensation of water on sample surfaces after defrosting. Compression wood samples reached slightly higher values of moisture gradient after drying, under the conditions of both drying modes. Though, differences were small and can be considered insignificant.

The level of case-hardening of compression wood samples was lower compared to opposite wood samples before the drying process. This indicates that the samples were wellselected and consisted only pure compression wood (Table 6). Higher level of casehardening after drying process was showed in opposite samples. The samples reached higher level of case-hardening under the conditions of drying mode No. 2 (high-temperature) compared to drying mode No.1.

From the resulting data, the adverse effect of compression wood on the level of casehardening cannot be ascertained. Case-hardening level could be affected by combination of high temperature and high relative humidity.

Tab. 5 Moisture gradient.

Sample CW1 CW2 OW1 OW2 CW1 CW2 OW1 OW2

Drying Mode No. 1 Drying Mode No. 2 Moisture Content (%) Moisture Content (%) Ø ws Δw Ø ws Ws Wm Ws Ws Wm Ws Before Drying Process 73.23 81.74 82.53 77.88 3.86 79.96 80.97 83.44 81.70 87.01 87.12 81.80 84.40 2.72 91.61 86.16 84.55 88.08 71.85 80.70 80.35 76.10 4.60 88.25 84.51 77.92 83.09 72.06 76.24 79.14 75.60 0.64 93.80 95.66 97.77 95.78 After Drying Process 9.94 11.60 10.74 10.34 1.26 8.66 10.09 9.60 9.13 9.89 10.75 9.51 9.70 1.05 8.48 10.15 9.02 8.75 9.24 9.98 9.49 9.37 0.61 9.49 9.87 9.50 9.49 10.59 9.77 8.80 9.69 0.08 9.81 9.67 8.95 9.38

Δw 0.73 1.92 1.42 0.13 0.96 1.40 0.38 0.29

Tab. 6 The values of case-hardening.

Case-hardening Size of Gap (mm)

Before Drying Process After Drying Process

Drying Mode No. 1 CW1 CW2 OW1 OW2 0 0 1.3 1.46 1.04 1.44 1.27 2.29

CW1 0.45 6.54

Drying Mode No. 2 CW2 OW1 OW2 0 1.25 1.43 1.62 2.12 10.32

CONCLUSION Our measurements have shown the following findings:  Compression wood dried remarkably slower than opposite wood during both phases (over and under FSP). The time necessary to reach the required final moisture content of 10% was 52 hours longer when the drying mode No. 1 was used (t = 60 °C) and 1.5 hours longer when the drying mode No. 2 was used (t = 120 °C). Differences in drying rate and intensity of water evaporation were noticeable especially at the start of the drying process (first 24 to 48 hours).  The compression wood density in oven dry state was 14.7% higher than opposite wood.

59

  

Compression wood samples reached slightly higher values of moisture gradient after drying process under the conditions of both drying modes. But the differences were small and can be considered unnoticeably.

The adverse effect of the compression wood on the level of case-hardening after the drying process cannot be ascertained.

Better knowledge of these properties can help with the processing of coniferous wood under production conditions.

REFERENCES BARNETT, J., SARANPÄÄ, P., GRIL, J. 2014. The Biology of Reaction Wood. Berlin Heidelberg : Springer-Verlag, p. 7. ISBN 978-3-642-10813-6.

DAVIS, CP., CARRINGTON, CG., SUN, ZF. 2002. The influence of compression wood on the drying curves of Pinus radiata in dehumifier conditions. In Dry Technology, 2002, 20(4): 2005–2026.

DIAZ.VAZ, J. E. et al. 2009. Redalyc density and chemical composition madera de compresión en Pinus radiata II: Densidad y compuestos químicos, In Maderas. Ciencia y technologia, 2009, 11(2): 139151.

GORIŠEK, Ž., TORELLI, N. 1999. Microfibril Angle in Juvenile, Adult and Compression Wood of Spruce and Silver Fir. 1999, In Phyton (Buenos Aires), 39: 2730.

GRYC, V. 2005. Stavba a vybrané vlastnosti dřeva smrku (Picea abies L. Karst.) s výskytem reakčního dřeva. 2005, Doctoral Thesis. Brno : MZLU, p.192.

KLEMENT, I., HURÁKOVÁ, T. 2015. Vplyv sušenia na vlastnosti a kvalitu smrekového reziva s obsahom reakčného dreva, In Acta Facultatis Xylologiae Zvolen, 2015. 57(1): 7582. ISSN 13363824.

NEČESANÝ, V. 1956. Struktura reakčního dřeva. Praha : Preslia, 28: 6165. ISSN 0032-7786.

OLLINMAA, P. J. 1959. On certain physical properties of wood growing on drained swamps. In Acta Forestalia Fennica 72. 24 pp. (In Finnish with English summary).

SAITO, K., FUKUSHIMA, K. 2005. Distribution of lignin interunit bonds in the differentiating xylem of compression and normal woods of Pinus thunbergii. In Journal of Wood Science, 2005, 51(10): 246251.

STN 490 103: 1993. Wood. Determination of the moisture content of the physical and mechanical testing. Slovak Standards Institute, Bratislava, Slovakia.

STN 490 108: 1993. Wood. Determination of density. Slovak Standards Institute, Bratislava, Slovakia.

STN 490 645: 1993. Testing of quality of dried timber (test of case-hardening of timber) Slovak Standards Institute, Bratislava, Slovakia.

STRAŽE, A., GORIŠEK, Z. 2006. Drying characteristics of compression wood in Norway spruce (Picea abies Karst.). In Wood structure and properties. Zvolen : Arbora Publishers, pp. 399–403 TARMIAN, A., AZADFALLAH, M., GHOLAMIYAN, H., SHAHVERDI, M. 2011. Inter-Tracheid and Cross-Field Pitting in Compression Wood and Opposite Wood of Norway Spruce (Picea abies L.).

In Notulae Scientia Biologicae, 2011. 3(2): 145151.

TARMIAN, A., PERRÉ, P. 2009: Air permeability in longitudinal and radial directions of compression wood of Picea abies L. and tension wood of Fagus sylvatica L. In Holzforschung, 63(3): 352356.

TARMIAN, A., REMOND, R., FAEZIPOUR, M., KARIMI, M., PERRÉ, P. 2009. Reaction wood drying kinetics: tension wood in Fagus sylvatica and compression wood in Picea abies. In Wood Science and Technology, 2009, 43(11): 113–130.

TIMELL, T. E. 1986. Compression wood in Gymnosperms. Vol. 1: Properties of compression wood.

Berlin : Springer Verlag, 1986. p. 2150. ISBN 3-540-15715-8.

WARENSJÖ, M. 2003. Compression wood in Scots pine and Norway spruce. Diss. (sammanfattning/summary) Umeå: Sveriges lantbruksuniv. In Acta Universitatis Agriculturae Sueciae. Silvestria, 2003, Doctoral Thesis.1401-6230;298 ISBN 91-576-6532-X.

60

WILLIAMS, D.H. 1971. A comparison of rates of drying of “compression” and “Normal” 4 in by 1 in Pinus radiate in N Z For. Res. Inst. Forest Product Report. 12: 303.

ACKNOWLEDGMENTS This work was supported by the Slovak Research and Development Agency under the contract no. APVV-17-0583.

ADDRESS OF AUTHORS Ivan Klement Miroslav Uhrín Tatiana Vilkovská Technical University in Zvolen Faculty of Wood Sciences and Technology Department of Wood Technology T. G. Masaryka 24 960 53 Zvolen klement@tuzvo.sk xuhrin@tuzvo.sk tatiana.vilkovska@tuzvo.sk

61

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 63−74, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.06

COMPUTING THE ENERGY FOR WARMING UP THE PRISMS FOR VENEER PRODUCTION DURING AUTOCLAVE STEAMING WITH A LIMITED POWER OF THE HEAT GENERATOR Nencho Deliiski – Ladislav Dzurenda – Dimitar Angelski – Natalia Tumbarkova ABSTRACT An approach for computing the specific energy needed for warming up the wooden prisms for veneer production during their autoclave steaming with a limited power of the heat generator is suggested. The approach is based on the integration of the numerical solutions of the created and verified 2-dimensional mathematical model for the transient non-linear heat conduction and energy consumption in frozen and non-frozen prismatic wood materials. An application of the suggested approach is shown in the paper for the case of computing the specific energy for warming up non-frozen and frozen beech prisms with cross-sections of 0.3 × 0.3 m, 0.4 × 0.4 m, 0.5 × 0.5 m and moisture content of 0.6 kg·kg-1 during their autoclave steaming aimed at their plasticizing in the veneer production. The power of the steam generator is limited and equal to 500 kW. The obtained results can be used to create the systems for optimized energy saving model based automatic control of the steaming process of wood materials.

Key words: autoclave steaming, energy consumption, wood materials, heat generator, limited power, model based control.

INTRODUCTION The steaming of wood materials with prismatic shape aimed at their plasticizing is an important part of the technological processes in the production of veneer and plywood (SHUBIN 1990, STEINHAGEN 1991, BURTIN et al. 2000, TREBULA – KLEMENT 2002, VIDELOV 2003, BEKHTA – NIEMZ 2003, PERVAN 2009, DELIISKI – DZURENDA 2010, DAGBRO et al.

2010, DELIISKI 2013b), etc.

The traditionally equipment and technologies used for steaming wood materials at atmospheric pressure of the processing medium are characterized by long durability (till some days) and low energy efficiency. During the last decades the utilization of intensive steaming of wood materials under increased pressure of the steam in autoclaves started (RIEHL et al. 2002, DELIISKI 2003, 2004, 2011a, VIDELOV 2003, DELIISKI – SOKOLOVSKI 2007, SOKOLOVSKI et al. 2007, DELIISKI et al. 2013a, 2013b).

In the specialized literature information about the energy consumption needed for warming up of wood materials during their autoclave steaming was given by DELIISKI (2003, 2009, 2013b), DELIISKI et al. (2010), DZURENDA – DELIISKI (2008, 2012), and DELIISKI –

63

DZURENDA (2010). The energy calculation in these publications were carried out for cases of unlimited generator power only.

The aim of the present work is to suggest an approach for the computation of the specific energy needed for warming up of wooden prisms for veneer production during their autoclave steaming with a limited power of the heat generator.

MATERIAL AND METHODS Modelling of the 2D heat distribution in the wood prisms subjected to steaming Mathematical models of 1D, 2D, and 3D heating processes of prismatic wood materials during their steaming at atmospheric and increased pressure of the processing medium have been created, solved and verified earlier by DELIISKI (2003, 2011b, 2013b, 2013c).

When the width of the wooden prisms does not exceed their thickness more than 3 times and simultaneously with this the length exceeds the thickness at least 5 times, then the heat transfer through the frontal sides of the prisms can be neglected, because it does not influence the change in temperature in the cross-section, which is equally distant from the frontal sides. In these cases for the calculation of the change in T in this section (i.e. only along the coordinates x and y) the following 2D model can be used: T x, y,      T x, y,    T x, y,    w -cr T , u, b , ufsp   w -cr T , u, b , ufsp   x  x y  y  





cw -е T , u, ufsp   w (b , u )





with an initial condition:





(1)

Tw x, y,0  Tw0

(2)

Т w 0, y,   Т w x,0,   Т m 

(3)

and the following boundary conditions:

According to eq. (3), the temperature at the prisms’ surfaces being in contact with the processing medium is equal to its temperature Tm due to the extremely high coefficient of heat transfer between the condensed saturated water steam on the wood materials during the whole regime of their thermal treatment processing (TTP).

Equations (1) to (3) represent a common form of a mathematical model of 2D heat distribution in prismatic wood materials subjected to steaming.

Mathematical description of the thermo-physical characteristics of the wood materials For solving and practical usage of eq. (1) it is needed to have mathematical descriptions of the thermal conductivity cross sectional to the fibers of the non-frozen and frozen wood, λwcr, of the effective specific heat capacity of the frozen and non-frozen wood, cw-e, and of the wood density above the hygroscopic range (i.e. when u > ufsp), ρw. For this purpose the description of λw-cr , cw-e, and ρw given in (DELIISKI 2011b, 2013b) and in (DELIISKI – DZURENDA 2010) can be used.

Mathematical descriptions of the thermal conductivity of non-frozen and frozen wood, w, and also of the specific heat capacity of the wood, cw, have been suggested by DELIISKI (1990, 1994, 2003, 2013a) using the experimentally determined in the dissertations by KANTER (1955) and CHUDINOV (1966) data for their change as a function of t and u. This

64

data for  w (t , u ) and сw (t , u) find a wide use in both the European (SHUBIN 1990, POŽGAJ et al. 1997, TREBULA – KLEMENT 2002, VIDELOV 2003, PERVAN 2009) and the American specialized literature (STEINHAGEN 1986, 1991, STEINHAGEN – LEE – LOEHNERTZ 1987, STEINHAGEN – LEE 1988, KHATTABI – STEINHAGEN 1992, 1993, 1995) when calculating various processes of the wood thermal treatment.

According to the mathematical description suggested in DELIISKI (1994, 2003, 2013a), the wood thermal conductivity during freezing of the wooden prisms can be calculated with the help of the following equations for  w (T , u, b , ufsp ) above the hygroscopic range: 272.15 λ w  λ w0  γ1  β(T  273 .15) @ u  ufsp & 213.15 K  T  423.15 K





λ w0  K ad  [0.165  1.39  3.8u   3.3.10  7 ρ 2b  1.015 .10 3 ρ b ]

v  0.1284  0.013u

(4) (5) (6)

In DELIISKI (2003, 2013b) the precise values of the coefficient Kad in eq. (5) for different wood species have been determined. For the discussed in this paper beech wood the following value of Kad-cr = Kad = 1.28 has been obtained.

The coefficients γ and  in equation (1) are calculated using the following equations: 272.15 and at the same time  For non-frozen wood when u  ufsp 272.15 K  T  423.15 K :

γ  1 .0

(7)

 579  β  3.65  0.124  10 3  ρb 

(8)

272.15 and at the same time  For frozen wood when u  ufsp

213.15 K  T  272.15 K :





γ  1  0.34[1.15 u  ufsp ]

(9)

 579  (10) β  0.002 (u  ufsp )  0.0038   0.124   ρb  where the fiber saturation point of the wood specie ufsp is calculated according to the equation 293.15 ufsp  ufsp  0.001(T  293 .15) (11) 272.15 is the fiber saturation point at T = 272.15 K (i.e. at t = –1 oC), kg·kg-1. At this and ufsp

temperature the melting of the frozen bound water in the wood is fully completed and the melting of the free water in the wood starts, (DELIISKI – TUMBARKOVA 2016); According to the suggested in DELIISKI (1990, 2011b, 2013b) mathematical description, the effective specific heat capacities the wood during TTP in an autoclave can be calculated with the help of the following equations for cw -e (T , u, ufsp ) above the hygroscopic range: 272.15 and at the same time the condition  For non-frozen wood when u  ufsp

273.15 K  T  423.15 K is fullfilld: cw -e  cw  nfr where 65

(12)

2862 u  555 5.49u  2.95 0.0036 2  T T (13) 1 u 1 u 1 u 272.15  For wood with frozen only free water in it when u  ufsp and at the same time cw nfr 

272.15 K  T  273.15 K : cw -e  cw nfr  cfw

(14)

where cfw  3.34  10 5

272.15 u  ufsp

1 u

(15)

272.15  For wood with frozen bound and free water in it when u  ufsp and at the same

time the condition 213.15 K  T  272.15 K is fullfilld:

cw -e  cw fr  cbwm

(16)

where 2 272 .15 0.00075 (T  272 .15) 526  2.95  0.0022 T  2261 u  1976 ufsp cw  fr  1.06  0.04u   1 u u 272 .15 fsp

(17) where



 exp0.05671Tu 272 .15

272 .15 cbwm  1.8938 10 4 ufsp  0.12 

272.15 293 .15 ufsp  ufsp  0.021

(18)

(19)

Eq. (19) is obtained from eq. (11) after substitution of T in it by T = 272.15 K. The wood density ρw, which participate in eq. (1), is determined above the hygroscopic range according to the below equation (CHUDINOV 1968, PERVAN 2009, DELIISKI 2011, DELIISKI et al. 2015b, HRČKA 2017)  w  b  (1  u) (20) Modelling of the heat energy needed for warming up of prisms during their steaming in an autoclave It is known that the specific heat energy consumption, which is needed for the warming up of 1 m3 of solid material, Qh, with an initial mass temperature T0 to a given average mass temperature Tavg is determined using the following equation (DELIISKI 2003, 2013b, DELIISKI – DZURENDA 2010): c  (21) Qh   Tavg  T0 3.6  10 6 Based on eq. (21), the specific energy needed for warming up of prismatic wood materials during their steaming can be calculated according to the following equation



n  Qhw



 c  n   w -e @ Ti, k  cw -e @ Tw 0 .   (Tin, j - Tw0 )dS w  6 2 3.6 10 S w S   w w

@ Tw0  Tin, k  Tw avg-end

66

(22)

where Twn  avg 

1 Sw

Sw 

 Ti,k dS w n

(23)

Sw

d b 4

(24)

The multiplier 3.6·106 in the denominator of eq. (22) ensures that the values of Qhw are obtained in kWh·m-3, instead of in J·m-3.

RESULTS AND DISCUSSION For numerical solution of the above presented mathematical model aimed at computation of the energy needed for warming up of prisms for veneer production during their autoclave steaming with a limited power of the heat generator a software package was prepared, which was input in the calculation environment of Visual Fortran Professional developed by Microsoft. For transformation of the model in a form suitable for programming an explicit form of the finite-difference method has been used (DELIISKI 2011b, 2013b).

With the help of the software package and of the approach suggested by the authors in DELIISKI et al. 2018, computations were made for the determination of Tm and also of the 2D non-stationary change of the temperature in 4 characteristic points of ¼ of the square cross section of beech prisms with thickness d and width b respectively, during their steaming in an autoclave with a diameter D = 2.4 m and length of its cylindrical part L = 9.0 m (DELIISKI – SOKOLOVSKI 2007, DELIISKI – DZURENDA 2010). The dimensions of the prisms’ cross sections were equal to 0.3 × 0.3 m, 0.4 × 0.4 m, 0.5 × 0.5 m, and the coordinates of their characteristic points were, as follow: Point 1: d/8, b/8; Point 2: d/4, b/4; Point 3: d/2, b/4; and Point 4: d/2, b/2.

During the solving of the models, the above presented mathematical descriptions of the thermophysical characteristics of beech wood (Fagus Sylvatica L.) with basic density ρb = 560 kg·m3 293.15 and fiber saturation point ufsp  0.31 kg  kg1 (NIKOLOV – VIDELOV 1987, DELIISKI –

DZURENDA 2010) were used. The initial temperature of the prisms was equal to 0 °C and –20 oC and their moisture content was 0.6 kg·kg1.

The increase of tm at the beginning of the 3-stage steaming regimes is calculated according to the approach given by the authors in DELIISKI et al. 2018 by taking in mind the available heat power of the generator that produces steam. During simulations the limited power of the generator qsource = 500 kW and loading level of the autoclave with filled in beech prisms for steaming ψ = 0.4 m3·m3 (i.e. ψ = 40%) were set.

Simultaneously with the solution of the model, computations of Tavg and Qhw have been carried out, using the value of the wood density ρw = 896 kg.m-3. This value of ρw is calculated according to eq. (20) for beech wood with u = 0.6 kg·kg-1 and ρb = 560 kg·m-3.

During the numerical simulations 3-stage TTP regimes for autoclave steaming of the prisms (see Fig. 1 to 4 below) were used, which form was presented in (DELIISKI et al. 2018).

As it was described in this source, during the first stage of the TTP regime input of water steam is accomplished in the autoclave, with situated inside wooden materials, until the temperature of the processing medium tm = 130 °C is reached. After reaching tm = 130 °C, this temperature is maintained unchanged by reducing the input of steam flux inside the autoclave until the calculated by the model average mass temperature of the wood, tavg, reaches a value of 90 °C. After reaching tavg = 90 °C the input of steam in the autoclave is 67

terminated and the second stage of the steaming regime begins. During this stage, by using the accumulated heat in the autoclave, the further heating and plasticizing of the prisms is accomplished, thus resulting in gradual reduction of the temperature tm for about 2 hours down to around 115 °C. Afterwards, the cranes directing the steam and condensed water out of the autoclave are opened, which initiates the third stage of the steaming regime. This stage ends after about one and half hour, when tm reaches approximate value of around 80 °С.

The calculations of Tavg have been carried out during the whole TTP regimes but the computations of Qhw have been conducted during only the first stages of these regimes, i.e. until reaching the set value of the average mass temperature of all studied prisms tavg = 90 °C. On Fig. 1 and Fig. 2 the calculated change in the temperature of the processing medium, tm, and also in the temperature in 4 characteristic points of 2 beech prisms with cross-section dimensions d × b = 0.3 × 0.3 m, initial temperature t0 = 0 °C and t0 = –20 °C during their TTP in an autoclave with loading level γ = 40 % is presented. The coordinates of the separate characteristic points are given in the legend of the figures.

140

o

Temperature t, C

120

tm

100

d/8, b/8

80 d/4, b/4

60 Beech: 0.3 x 0.3 m, t 0= 0 oC

40

d/4, b/2

u = 0.6 kg.kg-1 , g = 40 %

20

d/2. b/2

0 0

2

4

6

8

10

Time  , h

Fig. 1 Change in tm and t in 4 characteristic points of beech prisms with cross-section dimensions d × b = 0.3 × 0.3 m and t0 = 0 °C during their TTP in an autoclave at a loading of 40%.

140 tm

100

d/8, b/8

o

Temperature t, C

120

80 60

d/4, b/4 Beech: 0.3x0.3 m, t 0 = -20 oC

40

u = 0.6 kg.kg-1, g = 40 %

d/4, b/2

20 d/2. b/2

0 -20 0

2

4

6

8

10

12

Time  , h

Fig. 2 Change in tm and t in 4 characteristic points of beech prisms with cross-section dimensions d × b = 0.3 × 0.3 m and t0 = –20 °C during their TTP in an autoclave at a loading of 40%.

68

Figures 3 and 4 present the calculated change of tm in an autoclave and of tavg during TTP of the studied beech prisms with t0 = 0 oC and –20 oC respectively.

Figures 5 and 6 present the calculated change in tm and Qhw during the first stages of the TTP regimes for autoclave steaming with a limited power of the heat generator.

140

Temperature t , oC

120 100 80 tm: d x b = 0.3 x 0.3 m

60

tm: d x b = 0.4 x 0.4 m tm: d x b = 0.5 x 0.5 m

40

tavg: d x b = 0.3 x 0.3 m

Beech prisms: t 0 = 0 oC,

20

tavg: d x b = 0.4 x 0.4 m

u = 0.6 kg.kg-1, ψ = 40 %

tavg: d x b = 0.5 x 0.5 m

0 0

2

4

6

8

10

12

Time  , h

14

16

18

Fig. 3 Change in tm and tavg during the steaming of beech prisms with t0 = 0 °C in an autoclave at γ = 40%, depending on their cross-section dimensions.

140

Temperature t , oC

120 100 80 60 tm: d x b = 0.3 x 0.3 m tm: d x b = 0.4 x 0.4 m tm: d x b = 0.5 x 0.5 m tavg: d x b = 0.3 x 0.3 m

40 20 Beech prisms: t 0 = -20 oC,

0

tavg: d x b = 0.4 x 0.4 m tavg: d x b = 0.5 x 0.5 m

u = 0.6 kg.kg-1, ψ = 40 %

-20 0

2

4

6

8

10

12

14

16

18

20

22

24

Time  , h

Fig. 4 Change in tm and tavg during the steaming of beech prisms with t0 = –20 °C in an autoclave at γ = 40%, depending on their cross-section dimensions.

140

-3

Energy Q hw , kWh.m

Temperature t m , oC

120 100 80 60

tm: d x b = 0.3 x 0.3 m tm: d x b = 0.4 x 0.4 m tm: d x b = 0.5 x 0.5 m Qhw: d x b = 0.3 x 0.3 m Qhw: d x b = 0.4 x 0.4 m Qhw: d x b = 0.5 x 0.5 m

40 Beech prisms: t 0 = 0 o C,

20

u = 0.6 kg.kg-1, ψ = 40 %

0 0

2

4

6

8

10

12

14

Time  , h

Fig. 5 Change in tm and Qhw during the steaming of beech prisms with t0 = 0 °C in an autoclave at γ = 40%, depending on their cross-section dimensions.

69

140

Energy Q hw , kWh.m -3

Temperature t m , oC

120 100 80 60 tm: d x b = 0.3 x 0.3 m tm: d x b = 0.4 x 0.4 m tm: d x b = 0.5 x 0.5 m Qhw: d x b = 0.3 x 0.3 m Qhw: d x b = 0.4 x 0.4 m Qhw: d x b = 0.5 x 0.5 m

40 Beech prisms: t 0 = –20 oC,

20

u = 0.6 kg.kg-1, ψ = 40 %

0 0

2

4

6

8

10

12

14

16

18

20

Time  , h

Fig. 6 Change in tm and Qhw during the steaming of beech prisms with t0 = –20 °C in an autoclave at γ = 40%, depending on their cross-section dimensions.

The analysis of the obtained simulation results, part of which are presented on Fig. 1 to Fig. 6 lead to the following statements: 1. The non-stationary increasing of the temperature in the prisms’ characteristic points goes on according to very complex curves during the steaming process (Fig. 1 and Fig. 2).

2. When the water in the prisms is fully in a liquid state (i.e. at tw0 ≥ 0 oC), the increasing of tm causes a smoothly increasing of t in the characteristic points. The smoothness of the increasing of t depends proportionally on the distance of the points from the both prisms’ surfaces (Fig. 1).

3. When the subjected to steaming prisms are in frozen state, specific almost horizontal sections of retention of the temperature for a long period of time in the range from –1 oC to 0 oC can be seen, while in the points a complete melting of the frozen free water in the wood occurs (Fig. 2). As far the point is distanced from the prisms’ surfaces that much these sections with temperature retention are more extended. The reason of such a long retention of the wood temperature is the very low temperature conductivity of the wood during melting of the frozen free water in it (DELIISKI et al. 2015).

4. The smoothness of the increasing of tavg from tavg = tw0 to tavg = 90 oC during the 1st stage of TTP regimes depends proportionally on the dimensions of the prisms’ cross section.

During the 2nd stage of the regimes tavg remains practically unchanged and during the 3rd stage it decreases until reaching approximate value of around 82–83 oC (Fig. 3 and Fig. 4).

5. The increasing of the steaming time causes a smoothly increasing in the specific heat energy Qhw, which character is analogous to that of tavg. At the end of the 1st stage of TTP regimes, when tavg = 90 °C is reached, this energy reaches values of 65.4 kWh.m-3 and 96.6 kWh.m-3 for prisms with tw0 = 0 oC and tw0 = –20 oC respectively (Fig. 5 and Fig. 6). The larger value of Qhw at tw0 = –20 oC in comparison with that at tw0 = 0 oC is caused mainly by the participation in Qhw of the energy, which is needed for the melting of the frozen free water in the wood in the range from –1 oC to 0 oC (DELIISKI – DZURENDA 2010, DELIISKI et al. 2013a).

CONCLUSIONS The present paper describes the suggested approach for the computation of the specific heat energy Qhw, which is needed for warming up of wooden prisms for veneer production during their steaming in an autoclave at limited heat power of the steam generator. The 70

approach is based on the integration of the numerical solutions of personal 2D non-linear mathematical model of the steaming process of frozen and non-frozen prismatic wood materials, which are obtained using suggested by the authors in (DELIISKI et al. 2018) regimes for TTP with limited power of the heat generator.

For the solution of the model and practical application of the suggested approach for computation of Qhw, a software program was prepared in the calculation environment of Visual Fortran Professional developed by Microsoft.

The paper shows and analyses, as an example, diagrams of the non-stationary change in 2D temperature distribution and in dependant on it change in tavg and Qhw of beech prisms with cross-section dimensions 0.3 × 0.3 m, 0.4 × 0.4 m, and 0.5 × 0.5 m, initial temperatures of 0 °C and –20 °C, basic density of 560 kg·m-3, and moisture content of 0.6 kg·kg-1 during their steaming in an autoclave with a diameter of 2. 4 m, length of 9.0 m and loading with wood materials 40%, until reaching the average wood mass temperature of 90 °C at a limited heat power of the steam generator, equal to 500 kW. All diagrams are drawn using the results calculated by the model.

It has been determined, that the decrease of the cross-section dimensions of the prisms causes faster increase in tavg and Qhw during the 1st stage of the studied TTP regimes. At the end of this stage, i.e. at tavg = 90 °C, the energy Qhw reaches values of 65.4 kWh.m-3 and 96.6 kWh.m-3 for beech prisms with tw0 = 0 oC and tw0 = –20 oC respectively.

The obtained results can be used for science based computation of the energy saving optimized regimes for autoclave steaming of different wood materials at limited power of the steam generator. They will support the creation and improvement of systems for model based automatic realization of such regimes (DELIISKI 2003, 2011a, HADJISKI – DELIISKI 2016, HADJISKI at al. 2018).

Symbols b = width (m) c = specific heat capacity (J·kg-1·K-1) d = thickness (m) Q = specific heat energy (kWh·m-3) S = aria (m2) T = temperature (K) t = temperature (oC): t = T – 273.15 = moisture content (kg·kg-1 = %/100) u x = coordinate on the thickness: 0  x  d/2 (m) y = coordinate on the width: 0  y  b/2 (m) β = coefficients in the equations for determining of  γ = coefficients in the equations for determining of   = thermal conductivity (W·m-1·K-1)  = density (kg·m-3)  = time (s) ψ = loading level of the autoclave with filled in beech prisms for steaming (%) Δ = interval between time levels during the solving of the mathematical model (s) Subscripts ad = anatomical direction avg = average (for mass temperature of the prisms) b = basic (for density, based on dry mass divided to green volume) bw = bound water bwm = maximal possible amount of the bound water in the wood species cr = cross sectional to the fibers e = effective (for specific heat capacity)

71

end = end fsp = fiber saturation point fw = free water = heat h i = mesh point in the direction along the thickness for the prisms: 1, 2, 3, …, (d/Δx)+1 j = mesh point in the direction along the prisms’ width: 1, 2, 3, …, (b/Δx)+1 m = medium nfw = non-frozen water 0 = initial (for temperature) or at 0 oC (for ) p = process w = wood & = and simultaneously with this @ = at Superscripts n = time level during the solving of the mathematical model: n = 0, 1, 2, 3, …,р /Δ 272.15 = at 272.15 K, i.e. at –1 oC 293.15 = at 293.15 K, i.e. at 20 oC (for the standardized values of the wood fiber saturation point) REFERENCES

CHUDINOV, B. S. 1966. Teoreticheskiye issledovaniya teplofizicheskikh svoystv i termicheskoy obrabotki drevesiny, Disertatsiya na DSc, Krasnoyarsk, SSSR: SibLTI. BEKHTA, P., NIEMZ, P. 2003. Effect of High Temperature on the Change in Color, Dimensional Stability and Mechanical Properties of Spruce Wood. In Holzforschung, 57: 539546.

BURTIN P. et al. 2000. Wood Colour and Phenolic Composition under Various Steaming Conditions.

In Holzforschung, 54: 3338.

DAGBRO, O. et al. 2010. Colour Responses from Wood, Thermally Modified in Superheated Steam and Pressurized Steam Atmospheres. In Wood Material Science & Engineering, 5: 211219.

DELIISKI, N. 1990. Mathematische Beschreibung der spezifischen Wärmekapazität des aufgetauten und gefrorenen Holzes. In Proceedings of the VIIIth International Symposium on Fundamental Research of Wood. Warsaw, Poland: 229233.

DELIISKI, N. 1994. Mathematical Description of the Thermal Conductivity Coefficient of Non-frozen and Frozen Wood. In Wood Structure and Properties ’94, Zvolen : Technická univerzita vo Zvolene, p. 127134.

DELIISKI, N. 2003. Modelirane i tekhnologii za izparyavane na dŭrvesni materiali v avtoklavi.

Disertatsiya na DSc, Sofia: Lesotekhnicheski universitet, 358 s.

DELIISKI, N. 2004. Modelling and Automatic Control of Heat Energy Consumption Required for Thermal Treatment of Logs. In Drvna Industrija, 55(4): 181199.

DELIISKI, N. 2009. Computation of the 2-dimensional Transient Temperature Distribution and Heat Energy Consumption of Frozen and Non-frozen Logs. In Wood Research, 54(3): 67−78.

DELIISKI, N. 2011a. Model Based Automatic Control of the Wood Steaming Process in Autoclaves.

In Woodworking techniques, Prague, 710 September: 6772.

DELIISKI, N. 2011b. Transient Heat Conduction in Capillary Porous Bodies. In Ahsan A. (ed) Convection and Conduction Heat Transfer. Rieka : InTech Publishing House, 149176, http:// dx.doi.org/ 10.5772/21424.

DELIISKI, N. 2013a. Computation of the Wood Thermal Conductivity during Defrosting of the Wood.

In Wood research, 58(4): 637650.

DELIISKI, N. 2013b. Modelling of the Energy Needed for Heating of Capillary Porous Bodies in Frozen and Non-frozen States. Saarbrücken : Lambert Academic Publishing, Scholars’ Press, Germany, 106 pp.

DELIISKI, N. 2013c. 3D Modelling and Visualization of Non-Stationary Temperature Distribution

72

during Heating of Frozen Wood. In Drvna Industrija, 64(4): 293303.

DELIISKI, N., DZURENDA, L. 2010. Modelling of the Thermal Processes in the Technologies for Wood Thermal Treatment. Zvolen : TU vo Zvolene, Slovakia, 224 pp.

DELIISKI, N., DZURENDA, L., MILTCHEV, R. 2010. Computation and 3D Visualization of the Transient Temperature Distribution in Logs during Steaming. In Acta Facultatis Xylologiae Zvolen, 52(2): 2331, ISSN 1336-3824.

DELIISKI, N., DZURENDA, L., BREZIN, V. 2013a. Calculation of the Heat Energy Needed for Melting of the Ice in Wood Materials for Veneer Production. In Acta Facultatis Xylologiae Zvolen, 55(2): 2132, ISSN 1336-3824.

DELIISKI, N., DZURENDA, L., BREZIN, V., RAZUMOV, E. 2013b. Calculation of the Energy Needed for Heating of Frozen Wood until Melting of the Ice in it in the Veneer Production. In 10th International symposium “Selected processes at the wood processing”. September 1113, Ruzomberok, Slovakia, Volume of articles, 9 с., ISBN 978-80-228-2534-4.

DELIISKI, N., DZURENDA, L., TUMBARKOVA, N., ANGELSKI, D. 2015. Computation of the Temperature Conductivity of Frozen Wood during its Defrosting. In Drvna Industrija, 66(2): 8796.

DELIISKI, N., DZURENDA, L., ANGELSKI, D., TUMBARKOVA, N. 2018. An Approach to Computing Regimes for Autoclave Steaming of Prisms for Veneer Production with a Limited Power of the Heat Generator. In Acta Facultatis Xylologiae Zvolen, 60(1): 101112, DOI: 10.17423/ afx.2017.59.2.09.

DELIISKI, N., SOKOLOVSKI, S. 2007. Autoclaves for Intensive Resource Saving Steaming of Wood Materials. In 2nd International Scientific Conference “Woodworking techniques”, Zalesina, Croatia, 2007, p. 1926.

DELIISKI, N., TUMBARKOVA, N. 2016. A Methodology for Experimental Research of the Freezing Process of Logs. In Acta Silvatica et Lignaria Hungarica, 12(2): 145156, http://dx.doi.org /10.1515/aslh-2016-0013.

DZURENDA, L., DELIISKI, N. 2008. Mathematical Model for Evaluation of Heat Energy Consumption Norm at Color Homogenization of Beech Timber in Pressure Autoclave. In 6th International Science Conference “Chip- and Chipless Woodworking Processes”. Sturovo, Slovakia: 307314.

DZURENDA, L., N. DELIISKI, N. 2012. Drying of Beech Timber in Chamber Drying Kilns by Regimes Preserving the Original Colour of Wood. In In Acta Facultatis Xylologiae Zvolen, 54(1): 3342, ISSN 1336-3824.

HADJISKI, M., DELIISKI, N. 2016. Advanced Control of the Wood Thermal Treatment Processing. In Cybernetics and Information Technologies, Bulgarian Academy of Sciences, 16(2): 179197.

HADJISKI, M., DELIISKI, N., GRANCHAROVA, A. 2018. Spatiotemporal Parameter Estimation of Thermal Treatment Process via Initial Condition Reconstruction using Neural Networks: 5180. In Intuitionistic Fuzziness and Other Intelligent Theories and Their Applications, 193 pp., ISBN 9783-319-78930-9, Springer International Publishing, DOI: 10.1007/978-3-319-78931-6.

KHATTABI, A., STEINHAGEN, H. P. 1992. Numerical Solution to Two-dimensional Heating of Logs.

In Holz als Roh- und Werkstoff, 50(78): 308312, http://dx.doi.org/10.1007/ BF02615359.

KANTER, K. R. 1955. Issledovaniye teplovykh svoystv drevesiny. Dissertatsiya, Moskva, SSSR: MLTI. KHATTABI, A., STEINHAGEN, H. P. 1993. Analysis of Transient Non-linear Heat Conduction in Wood Using Finite-difference Solutions. In Holz als Roh- und Werkstoff, 51(4): 272278, http://dx.doi.org/ 10.1007/ BF02629373.

KHATTABI A., STEINHAGEN H. P. 1995. Update of “Numerical Solution to Two-dimensional Heating of Logs”. In Holz als Roh- und Werkstoff, 53(1): 9394, http://dx.doi.org/10.1007/ BF02716399.

PERVAN, S. 2009. Technology for Treatment of Wood with Water Steam. Zagreb : University in Zagreb.

POŽGAJ, A., CHOVANEC, D., KURJATKO, S., BABIAK, M. 1997. Štruktúra a vlastnosti dreva. 2. vydanie, Bratislava: Priroda a.s., 486 s.

RIEHL, T., WELLING, J., FRÜHWALD, A. 2002. Druckdämpfen von Schnittholz, Arbeitsbericht 2002/01: Institut für Holzphysik, Hamburg: Bundesforschungsanstalt für Forst- und Holzwirtschaft.

SHUBIN, G. S. 1990. Sushka i termicheskaya obrabotka drevesiny. Moskva: Lesnaya Promyshlennost'.

73

SOKOLOVSKI, S., DELIISKI, N., DZURENDA, L. 2007. Constructive Dimensioning of Autoclaves for Treatment of Wood Materials under Pressure. In 2nd International Scientific Conference “Woodworking techniques”, Zalesina, Croatia, 2007, p. 117126.

STEINHAGEN, H. P. 1986. Computerized Finite-difference Method to Calculate Transient Heat Conduction with Thawing. In Wood Fiber Science, 18(3): 460467.

STEINHAGEN H. P. 1991. Heat Transfer Computation for a Long, Frozen Log Heated in Agitated Water or Steam – A Practical Recipe. In Holz als Roh- und Werkstoff, 49(78): 287290, http:// dx.doi.org/10.1007/ BF02663790 STEINHAGEN H. P., LEE H. W. 1988. Enthalpy Method to Compute Radial Heating and Thawing of Logs. In Wood Fiber Science, 20(4): 415421.

STEINHAGEN, H. P., LEE, H. P., LOEHNERTZ, S. P. 1987. LOGHEAT: A Computer Program of Determining Log Heating Times for Frozen and Non-Frozen Logs. In Forest Products Journal, 37(1112): 6064.

TREBULA, P., KLEMENT, I. 2002. Sušenie a hydrotermická úprava dreva. Zvolen: TU v Zvolene, 449 s.

VIDELOV, H. 2003. Sushene i termichna obrabotka na dŭrvesina. Sofia: Lesotekhnicheski universitet, 335 s.

ACKNOWLEDGEMENTS This document was supported by the grant No BG05M2OP001-2.009-0034-C01 "Support for the Development of Scientific Capacity in the University of Forestry", financed by the Science and Education for Smart Growth Operational Program (20142020) and co-financed by the European Union through the European structural and investment funds.

This document was also supported by the APVV Grant Agency as part of the project: APVV-170456 as the result of work of authors and the considerable assistance of the APVV agency.

AUTHORS’ ADDRESSES Prof. Nencho Deliiski, DSc., PhD, University of Forestry Faculty of Forest Industry Kliment Ohridski Blvd. 10 1797 Sofia Bulgaria deliiski@netbg.com Prof. Ladislav Dzurenda, PhD, Technical University in Zvolen Faculty of Wood Science and Technology T. G. Masaryka 24 960 53 Zvolen Slovakia dzurenda@tuzvo.sk Assoc. Prof. Dimitar Angelski, PhD, Eng. Mag. Natalia Tumbarkova University of Forestry Faculty of Forest Industry Kliment Ohridski Blvd. 10 1797 Sofia, Bulgaria d.angelski@gmail.com ntumbarkova@abv.bg

74

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 75−82, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.07

GRANULOMETRIC ANALYSIS OF CHIPS FROM BEECH, OAK AND SPRUCE WOODTURNING BLANKS PRODUCED IN THE MILLING PROCESS USING 5-AXIAL CNC MACHINING CENTER Richard Kminiak – Adrian Banski ABSTRACT Granulometric composition of chips resulting from the milling process of beech, oak and spruce woodturning blanks with the thickness of 25 mm using a single milling cutter and a CNC machining center SCM TECH Z5 is presented in the paper. Granulometric composition of chips is observed in the range of commonly used technological conditions for given type of the milling cutter such as feed rate vf = 1 to 5 m·min1 and the thickness of removed layer e = 1, 3 and 5 mm. The fact that more than a half of the produced chips is a coarse fraction of fibrous chips with dimension more than 1 mm can be stated following the granulometric analysis. Isometric grains, i.e. chips with approximately the same size in all three dimensions are formed from dust fractions smaller than 500 μm. There is, on average, 2.38 % of inhalable dust with particles smaller than 125 μm. The finding that no particles in respirable dust with the size smaller than < 10 μm are formed can be stated.

Key words: CNC machining center, granulometric composition of chips, dust particles, respirable dust.

INTRODUCTION CNC technology has become an integral part of the woodprocessing industry so far. The CNC machining centers are among the most widely used ones. The 5 axial versions of these centers are trendy. In addition to the basic movements in the X, Y and Z axes, they can also rotate in the B axis (rotational movement around the Y axis) and C (rotational movement about the Z axis). 5-axial CNC machining centers are universal machines that make possible machining of the workpiece in its five basic surfaces and their combinations (the 6th surface is determined for the attachment of a workpiece during the machining process). Because of the versatility of the 5-axial CNC machining centers, the issue of suction of formed chips from the working environment started to be study. Collection bins are used to suck the chips; the bins are pressed against the workpiece surface and suck the sawdust into the ventilation system.

Collection bin in the 5-axial version of the machines can be of considerable dimensions with the volume in the range from 0.125 ÷ 0.350 m3 (significant pressure loss is created). At the same time, collection bin outlet is axial in terms of technical feasibility (while the movement of the produced chips is mostly radial). The problem arises when the workpiece side edges are machined with the tools producing bigger chips. During the milling, a rotating air stream is formed from separated chips. The energy of a mixture of rotating air and chips is so great that its significant part is not carried by the sucked air; it hits the walls of the collection bins and 75

then falls into the workspace of the CNC machine. The chips cannot be removed by a ventilation system. Therefore the worker “blows” out the workpiece after the final machining by the compressed air which causes secondary pollution of the working environment around the CNC machine (BANSKI and KMINIAK 2018, KMINIAK and BANSKI 2018).

The formed chip is a polydisperse bulk mass consisting of coarse, medium coarse and dust fractions. Wood dust with a grain size ranging from 1 ÷ 500 μm (HEJMA et al., 1981, DZURENDA et al. 2010, DZURENDA and ORLOVSKI 2011, OČKAJOVÁ and BANSKI 2013, HLÁSKOVA et al. 2016, PAŁUBICKI and ROGOZIŃSKI 2016, MARKOVÁ et al. 2018,) is a hygroscopic, low abrasive, explosive bulk mass. The ratio of dust particles depends on the characteristics of the processed material, the parameters of the tool as well as the technical and technological parameters of the machining process, (PALMQVIST and GUSTAFSSON 1999, DZURENDA 2002, DZURENDA et al. 2006, RATNASINGAM et al. 2010, FUJIMOTO et al. 2011, OČKAJOVÁ et al. 2016, HLÁSKOVA et al. 2015).

In terms of physiology and in accordance with international standards (USA - ACGIH, EPA and Europe - ISO, CEN, BMRC) the dust fractions smaller than 100 µm are divided as follows: breathable (inhalable) mass fraction < 100 µm, thoracal 5 ÷ 10 µm, tracheobronchial (respirable mass fraction) 2.5 ÷ 5 µm, high respirable mass fraction < 2.5 µm.

Wood dust from beech and oak, as stated by OČKAJOVÁ et al. (2006) is considered to be toxic and is classified as group 1 carcinogen. Pursuant to the Act of the Government Regulation of the Slovak Republic No. 83/2015 Coll. amending the Act of the Government Regulation of the Slovak Republic No. 356/2006 Coll. on minimum health and safety requirements for the protection of workers against the risks relating to the exposure to carcinogenic and mutagenic factors in the workplaces, as amended by SR government in Act No. 301/2007 Coll., the dust with a carcinogenic and mutagenic effect and the concentration of the toxic component of the aerosol cannot exceed the technical values for an existing factor (5 mg.m3) (the Act of the Government Regulation of the Slovak Republic No. 301/2007 Coll.; No. 471//2011 Coll.).

The aim of the paper is to determine a granulometric composition of the chips (produced chip – a real chip resulting from the machining process, sucked chip - a chip that can be removed from the workspace by a ventilation system) produced using the CNC machining center in the conditions associated with forming the chips with maximum size from beech, oak and spruce woodturning blanks.

METHODOLOGY Characteristics of the used material – natural furniture woodturning blanks with following parameters: wood species – European beech (Fagus sylvatica), English oak (Quercus robur), Norway spruce (Picea abies), texture – tangential sawn timber, parameters – thickness of 25 mm (± 0.5 mm), width of 80 mm (± 0.5 mm), length of 500 mm (± 1 mm), moisture content of 10 % (± 2 %) were used in the experiment.

Characteristics of the used machine – the experiment is carried out using the 5-axial CNC machining centre SCM Tech Z5 (Figure 1) supplied by SCM-group, Rimini, Italy.

Characteristics of the used tool – in the experiment, milling cutter – single-bladed designation KARNED 4451 by the manufacturer Karned Tools Ltd., Prague, Czech Republic was used (mentioned tool was selected due to the assumption that chips of maximum sizes copying the shape of produced chips are formed). Milling cutter was equipped with the reversible blade HW 49.5 / 9 / 1.5 from sintered carbide T10MG. Characteristics of the milling process – the workpiece was milled under following conditions: material removal rate – e = 1, 3 and 5 mm (the thickness of the removal layer is 76

based on the standard inputs of the processed materials), rotation speed – n = 20,000 min1 (the value recommended by the tool manufacturer), feed rate – vf = 1, 2, 3, 4 and 5 m·min1 (maximum feed rate recommended by a manufacturer is 5 m.min-1, but the operator adjusts / reduces the value to 1 m.min1 according to the local conditions of the milling process and the required quality of the formed surface). At least six samples were used in the milling process of each combination of parameters, until the chips with the dimensions of 3 × 50 g were formed.

Fig. 1 CNC machining center SCM Tech Z5.

Characteristics of the chip removal process – selected methodology of the chip removal process reflects the requirements to remove actually produced chips not only chips that can be removed by the suction unit as well as the expectation of forming the chips with maximum sizes causing the difficulties during isokinetic removal. During the experiment, the ventilation system of the machine was off and the collection bin was lifted over the working space. The ventilation system of the machine was replaced by the product see Figure 2.

Intake vacuum of the chips from the workspace was created by a mobile suction unit OP 1500 supplied by Proma SK, Zvolen, Slovakia (with suction capacity of 1,020 m3.hr-1, maximum intake vacuum of 1,400 Pa and suction pipe diameter of 110 mm). Sucked chips were collected on the cloth filter Hyundai VCP 200 by Hyundai Mobis Co. Ltd. South Korea (filter class G3 - STN EN 779/1822). The chips collected on the cloth filter were dropped into the collection bin and the weight of m = 50 g was reached; further analysis were conducted. The intake port for sucking the sawdust corresponded with the resulting direction of the sawdust stream. 3 samples with the weight of 50 g were analyzed for each combination of parameters.

Fig. 2 CNC part for removing the formed chips.

77

Characteristics of the granulometric analysis – the process of sifting was used to detect the granulometric composition of chips. For this purpose, special set of sieves arranged one above the other (2 mm, 1 mm, 0.5 mm, 0.25 mm, 0.125 mm, 0.063 mm, 0.032 mm and the bottom) placed on a vibration stand of the sifting machine Retsch AS 200c by the company Retsh GmbH, Haan, Germany was used. The parameters of sifting were as follows: frequency of sifting interruption – 20 seconds, amplitude of sieves deflection – 2 mm.g-1, sifting time - τ = 15 minutes, weighed sample – 50 g. The granulometric composition was obtained by weighing the portions remaining on the sieves after sifting using the electric laboratory scales Radwag 510/C/2 by the company Radwag Balances and Scales, Radom, Poland, with the weighing accuracy of 0.001 g. Three samples for each combination of parameters were analyzed in the process of sifting.

In order to specify the information about size of the smallest particles of fine fraction of dry sawdust, a microscopic analysis of granules of fraction of dry sawdust with the size smaller than 125 μm was carried out. Optical method – analysis of the picture obtained from the microscope Nikon Optiphot-2 with the lens Nikon 4×, was selected to conduct the proposed analysis of sawdust. Granules of sawdust were scanned by three low-cost television CCD cameras HITACHI HV-C20 (RGB 752 × 582 pixel), with horizontal resolution 700 TV lines and evaluated by the software LUCIA-G 4.0 (Laboratory Universal Computer Image Analysis), installed on a PC with the processor Pentium 90 (RAM 32 MB) with the graphic card VGA Matrox Magic under the operation system Windows NT 4.0 Workstation.

The analysis of picture LUCIA-G enables us to identify the individual particles of disintegrated wood material, quantitative determination of individual particles situated in the analyzed picture and basic information such as: width and length of particles, circularity expressing the measure of deviation of projection of a given chip shape from the projection of the shape of a circle according to the formula: ψ = (4*π*S)/P2,

(1)

where: S – surface of particle [m2], P – perimeter of particle [m]

RESULTS AND DISCUSSION Average values resulting from three sifting for each combination of examined parameters (wood species, feed rate and the thickness of removed layer) are summarized in Table 1.

Characteristics of produced mixture of chips Fraction chips with dimensions over 2 mm belong to a category of flat chips, i.e. the length and width of the chip significantly exceed its thickness. The chip has the shape of a removed layer of milled wood. When the feed rate was lower, the chips were without breaks. When the feed rate increases, the breaks occur more often. Fraction chips with the grain ranging from 2 mm to 500 μm belong to a category of fibrous chips, i.e. chips with a significant extension in one direction. Fractions smaller than 500 μm, can be characterized as isometric chips, i.e. chips with approximately the same size in all three dimensions. The 1.4 % of inhalable dust with fraction of particles smaller than 125 μm was formed in case of beech wood, 1.95 % in case of oak wood and 3.8 % in case of spruce wood analyzed. The CNC machining centre does not produce respirable fractions smaller than 10 μm.

The gathered data were statistically analyzed and the facts that the factor associated with wood species and the factor associated with the thickness of removed layer are considered statistically significant can be stated. The statistical effect was not confirmed in case of the factor – feed rate.

78

SPRUCE

OAK

BEECH

Wood species

Tab. 1 Average granulometric composition of chips for considered combinations of material removal rate and feed rate. sieves 2mm 1mm 500µm 250µm 125µm 63µm 32µm bottom 2mm 1mm 500µm 250µm 125µm 63µm 32µm bottom 2mm 1mm 500µm 250µm 125µm 63µm 32µm bottom

1 87.60 6.63 1.75 1.34 1.41 1.01 0.26 0.00 84.01 6.55 1.72 2.14 3.20 1.96 0.41 0.00 0.26 63.86 14.74 9.74 7.27 3.37 0.75 0.00

1 mm feed rate vf [m.min-1] 2 3 4 83.27 93.77 95.43 11.97 3.09 2.05 2.11 1.05 0.92 1.16 0.98 0.81 0.92 0.66 0.47 0.42 0.34 0.28 0.15 0.12 0.04 0.00 0.00 0.00 77.90 67.88 70.84 8.78 14.83 14.49 5.32 7.36 6.53 3.48 4.55 4.62 3.36 4.08 2.83 1.03 1.12 0.62 0.13 0.18 0.06 0.00 0.00 0.00 3.19 16.55 1.94 67.60 56.85 75.26 8.66 9.05 11.53 10.59 9.34 6.29 6.68 5.41 3.10 2.88 2.47 1.64 0.40 0.34 0.24 0.00 0.00 0.00

5 92.23 3.13 2.22 1.46 0.50 0.36 0.09 0.00 64.72 17.17 7.44 6.12 3.24 1.09 0.23 0.00 47.82 17.28 13.57 11.77 5.72 3.05 0.79 0.00

Thickness of removed layer e [mm] 3 mm feed rate vf [m.min-1] 1 2 3 4 5 84.93 72.39 66.94 63.46 55.87 5.65 6.65 8.56 9.65 10.98 3.43 9.89 10.87 11.58 15.04 2.52 6.42 8.22 10.73 14.45 2.04 3.62 4.66 3.79 2.43 1.10 0.84 0.58 0.62 0.97 0.32 0.20 0.17 0.16 0.27 0.00 0.00 0.00 0.00 0.00 64.95 74.10 64.77 51.72 44.83 9.90 9.77 15.95 16.84 19.36 9.87 7.83 9.12 12.82 15.96 7.98 4.68 4.41 12.71 14.51 5.10 2.75 4.36 4.75 3.84 1.78 0.71 1.13 0.95 1.18 0.42 0.15 0.26 0.21 0.32 0.00 0.00 0.00 0.00 0.00 41.74 0.29 27.93 21.19 31.06 25.56 58.92 29.17 38.27 28.63 10.29 16.85 19.24 19.87 22.54 10.87 9.62 12.99 11.39 10.69 6.29 7.51 6.21 5.47 4.49 4.05 5.75 3.33 2.74 2.15 1.20 1.06 1.14 1.07 0.45 0.00 0.00 0.00 0.00 0.00

1 52.97 14.25 11.42 9.85 7.46 3.40 0.65 0.00 44.01 14.90 14.49 11.57 9.17 4.78 1.04 0.04 46.85 18.31 9.12 13.05 7.17 4.16 1.34 0.00

5 mm feed rate vf [m.min-1] 2 3 4 46.06 37.70 40.48 11.92 9.54 10.60 13.68 16.50 18.11 13.18 20.19 21.34 12.08 13.91 7.60 2.61 1.76 1.48 0.47 0.40 0.38 0.00 0.00 0.00 30.85 33.03 36.10 16.36 17.93 15.74 17.16 16.78 17.32 16.50 18.59 21.51 14.42 11.34 7.81 4.11 1.91 1.20 0.56 0.40 0.32 0.03 0.01 0.00 45.05 29.06 42.66 14.26 22.98 18.25 16.44 23.45 21.16 12.98 14.09 10.60 6.76 6.66 4.68 3.57 2.79 2.06 0.95 0.97 0.59 0.00 0.00 0.00

5 37.50 9.44 22.26 25.12 4.04 1.26 0.37 0.00 41.76 13.75 16.57 22.33 4.67 0.70 0.22 0.00 34.12 29.21 20.64 10.56 3.75 1.37 0.35 0.00

The effect of wood species Three economically most important wood species in the Slovak Republic – beech, oak and spruce were used in the experiment. Following the statistical analysis of the gathered data, the fact that statistically significant difference among the wood species is in fractions with the dimension over 500 μm (Figure 3) can be stated. The fractions with the dimension over 2 mm can be observed especially in chips from beech wood. It is followed by oak and the smallest portion can be seen in the chips of spruce. A lower portion of the fraction over 2 mm is compensated for a higher proportion, in particular, fractions over 1 mm and fractions over 500 μm.

The effect of the thickness of removed layer An increase in the thickness of removed layer means the length extension of the removed layer resulting in greater friability due to longer contact with the face of tool and thinner chips. As a result of chip friability, a decrease in the portion of fraction over 2 mm and subsequent increase in remaining fractions can be observed (Figure 4).

The effect of feed rate The effect of the feed rate was not proven despite the fact that an increase in feed rate means a decrease in the maximum thickness of the removed layer and an increase in the break occurrence.

79

90 80

Percentage representation

70 60 50 40 30 20 10 0 -10

2mm

1mm

500µm

250µm

125µm

63µm

32µm

bottom

Dimension of mesh sieve wood:

BEECH,

OAK,

SPRUCE.

Fig. 3 The effect of wood species on the average granulometric composition of chips.

100 90

Percentage representation

80 70 60 50 40 30 20 10

removal 3 mm

wood:

BEECH,

OAK,

32µm

bottom

63µm

125µm

250µm

1mm

500µm

2mm

Dimension of mesh sieve:

32µm

bottom

63µm

125µm

250µm

1mm

removal: 1 mm

500µm

2mm

Dimension of mesh sieve:

32µm

bottom

63µm

125µm

250µm

1mm

500µm

Dimension of mesh sieve:

-10

2mm

0

removal: 5 mm

SPRUCE.

Fig. 4 The effect of wood species and the thickness of removed layer on the average granulometric composition of chips.

80

90 80

Percentage representation

70 60 50 40 30 20 10

feed speed 1 m.min -1

feed speed 2 m.min -1

wood:

feed speed 3 m.min -1

BEECH,

OAK,

feed speed 4 m.min -1

Dimension of mesh sieve: 2mm 1mm 500µm 250µm 125µm 63µm 32µm bottom

Dimension of mesh sieve: 2mm 1mm 500µm 250µm 125µm 63µm 32µm bottom

Dimension of mesh sieve: 2mm 1mm 500µm 250µm 125µm 63µm 32µm bottom

Dimension of mesh sieve: 2mm 1mm 500µm 250µm 125µm 63µm 32µm bottom

-10

Dimension of mesh sieve: 2mm 1mm 500µm 250µm 125µm 63µm 32µm bottom

0

feed speed 5 m.min -1

SPRUCE.

Fig. 5 The effect of wood species and feed rate on the average granulometric composition of chips

CONCLUSION Following the experiments, we can draw the following conclusions:

• • • •

Fractions with dimension over 1 mm are considerable portion of sucked chips.

On average, 17.85 % of the sucked chips formed during the milling process using the CNC machining centre are dust particles smaller than 500 μm.

On average, 2.38 % of the sucked chips are inhalable dust particles smaller than 125 μm.

The occurrence of particles smaller than 32 μm and thus, of respirable particles was not proven.

References BANSKI, A., KMINIAK, R. 2018. Influence of the thickness of removed layer on granulometric composition of chips when milling oak blanks on the CNC machining center. In Trieskové a beztrieskové obrábanie dreva, 11(1): 2330. ISSN 2453-904X. DZURENDA, L. 2002. Vzduchotechnická doprava a separácia dezintegrovanej drevnej hmoty. Zvolen : TU vo Zvolene; p. 143. ISBN 80-228-1212-9.

DZURENDA, L., ORLOWSKI, K., GRZESKIEWICZ, M. 2010. Effect of thermel modification of oak wood on sawdust granularity. In Drvna industrija, 61(2): 89−94.

DZURENDA, L., WASIELEWSKI, R., ORLOWSKI, K. 2006. Granulometric analysis of dry sawdust from the sawing process on the frame sawing machine PRW15M. In Acta Facultatis Xylologiae Zvolen.

48(2): 5157.

DZURENDA, L., ORLOWSKI, K. 2011. The effect of thermal modification of ash wood on granularity

81

and homogeneity of sawdust in the sawing process on a sash gang saw PRW 15-M in view of its technological usefulness. In Drewno, 54(186): 2737.

FUJIMOTO, K., TAKANO, T., OKUMURA, S. 2011: Difference in mass concentration of airborne dust during circular sawing of five wood-based materials. In J Wood Sci 57: 149154. (https://doi.org/ 10.1007/s10086-010-1145-y) HEJMA, J. et al. 1981. Vzduchotechnika v dřevozpracovávajícím průmyslu. Praha : SNTL; p. 398.

HLÁSKOVÁ, L’., ROGOZINSKI, T., DOLNY, S., KOPECKỲ, Z., JEDINÁK, M. 2015. Content of respirable and inhalable fractions in dust created while sawing beech wood and its modifications. In Drewno 58(194): 135146.

HLÁSKOVÁ, L., ROGOZINSKI, T., KOPECKÝ, Z. 2016. Influence of feed speed on the content of fine dust during cutting of two-side-laminated particleboards. In Drvna Ind 67(1): 9-15 (DOI: 10.5552/ drind.2016.1417) KMINIAK, R., BANSKI, A. 2018. Separation of Exhausted Chips from a CNC Machining center in Filter FR - SP 50/4 with Finet PES 4 Fabric. AIP Conf. Proc. 2018; 020011-1 – 02011-4, (doi.org/10.1063/1.5049913) MARKOVÁ, I., LADOMERSKÝ, J., HRONCOVÁ, E., MRAČKOVÁ, E. 2018. Thermal parameters of beech wood dust. In BioRes. 13(2): 30983109. (DOI: 10.15376/biores.13.2.3098-3109).

OČKAJOVÁ, A., BANSKI, A. 2013. Granulometria drevného brúsneho prachu z úzko-pásovej brúsky.

In Acta Facultatis Xylologiae Zvolen, 55(1): 85−90. ISSN 1336-3824.

OČKAJOVÁ, A., BELJO LUČIĆ, R., ČAVLOVIĆ, A. TERENÒVÁ, J. 2006. Reduction of dustiness in sawing wood by universal circular saw. In Drvna industrija, 57(3): 119126.

OČKAJOVÁ, A., KUČERKA, M., BANSKI, A., ROGOZIŃSKI, T. 2016. Factors affecting the granularity of wood dust particles. In Chip and Chipless Woodworking Processes, 10(1): 137144.

PALMQVIST, J., GUSTAFSSON, S. I. 1999. Emission of dust in planning and milling of wood. In Holz als Roh- und Werkstoff, 57: 164−170.

PAŁUBICKI, B., ROGOZIŃSKI, T., 2016. Efficiency of chips removal during CNC machining of particleboard. In Wood Research, 61(5): 811818.

RATNASINGAM, J., SCHOLZ, F., NATTHONDAN, V. 2010. Particle size distribution of wood dust in rubberwood (Hevea Brasiliensis) furniture manufacturing. In European Journal of Wood and Wood Products, 68: 241242.

ACKNOWLEDGEMENT This article was created with the support of VEGA 1/0725/16 “Prediction of the quality of the generated surface during milling solid wood by razor endmills using CNC milling machines.” and support of VEGA 1/0485/18 “Machining strategy for special cutting models of agglomerated materials in nesting machining on a CNC machining center”.

AUTHORS ADDRESS Ing. Richard Kminiak, PhD. Ing. Adrián Banski. PhD. Technical University in Zvolen Faculty of Wood Sciences and Technology Department of Woodworking T. G. Masaryka 24 960 53 Zvolen Slovakia richard.kminiak@tuzvo.sk banski@tuzvo.sk 82

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 83−92, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.08

THE EFFECT OF WOOD DRYING METHOD ON THE GRANULARITY OF SAWDUST OBTAINED DURING THE SAWING PROCESS USING THE FRAME SAWING MACHINE Kazimierz A. Orłowski – Daniel Chuchała – Tomasz Muziński – Jacek Barański – Adrián Banski – Tomasz Rogoziński ABSTRACT The experimental results of the study focused on the effect of drying processes of warm air drying at the temperature of 6580°C and warm air-steam mixture drying at the temperature of 105°C of pine and beech wood to the size of sawdust grains created by cutting using RPW 15M frame saw is presented in the paper. Particle size analysis of dry sawdust was performed using two methods - screening method and optical method based on image analysis obtained from a microscope. The results showed that the drying mode did not affect the particle size distribution of the pine sawdust, but sawdust from beech wood dried with steam mixture at 105°C was characterized by finer particles.

Key words: pine wood, beech wood, warm air drying, warm air-steam mixture drying, frame sawing machine, granulometric sawdust analyses.

INTRODUCTION The increasing interest in sawdust, as a secondary raw material, in the last years, requires a proper specification of its physical properties as follows: granularity, geometric shapes and size of sawdust chips. The shape, dimensions and amount of chips depend on the form, physical and mechanical properties of the sawed wood (BELJO LUČIČ et al. 2009, DZURENDA et al. 2009, DZURENDA et al. 2010, DZURENDA and ORLOWSKI 2010, 2011, 2011a, FUJIMOTO et al. 2011), as well as on the shape, dimensions, and sharpness of the cutting blade, technical and technological conditions of the sawing process (DZURENDA et al. 2006, HLÁSKOVÁ et al. 2016, PAŁUBICKI and ROGOZIŃSKI 2016).

Sawdust is characterized as poly-dispersion bulk material consisting of coarse and medium- coarse fractions (HEJMA et al. 1981, DZURENDA 2009), i.e. bulk material with dimensions of grain over 0.3 mm, while the share of fine fractions with smaller dimensions of chips is not excluded. This fraction is fully covered by dust properties, including the specifics: dust explosion, low drop speed, slow sedimentation, long residence time in the air and filtration conditions (HLÁSKOVA et al. 2015, OČKAJOVA et al. 2016, RATNASINGAM et al. 2010, MARKOVA et al. 2018, KMINIAK 2018, KUČERKA and OČKAJOVÁ 2018).

Detailed results of particle-size analysis of sawdust described by OCKAJOVA et al.

(2006) have shown the influence of different parameters of a saw blade used for cutting on dimensions of particles created during sawing various types of wood. Comparison of the results of particle-size analysis of sawdust created in the sawing process using a universal 83

circular saw with triangular asymmetric spring set of teeth with the results of analysis of sawdust from sawing using a saw blade with tipped swaged anti-kickback teeth, with chip breaker and optimal chip clearance shown that smaller chips were formed during the use of the universal saw blade. This applies especially to the finest particles potentially hazardous to health.

There are many parameters affecting the characteristics of wood sawdust already analyzed. However, the influence of drying method of wood on the size of dust particles generated during processing is not yet analyzed. Thus, the objective of this work was to examine the effect of the drying method on the sawdust cerated during sawing on the frame sawing machine PRW15-15.

MATERIALS AND METHODS Materials The material used in the experiments was pine wood (Pinus sylvestris L.) and beech wood (Fagus sylvativca L.) which were dried with different technics in industial and labolatory conditions. The wood to be used in sawing experiments were blocks in dimensions b = 50 mm × Hp = 50 mm × Lp = 500 mm.

Drying process The drying process was conducted in the experimental facility of semi-industrial kiln, which is located at the Gdansk University of Technology (GUT). The experiments of drying process were described in details by Baranski (BARANSKI 2018). Drying modes for warm air drying as well as high temperature air-steam mixture drying take into account the ecommendations for drying the beech and pine timber listed in works (BASILICO at al. 1990; SEHLSTED-PERSSON 1995; PERVAN 2000; KLEMENT and SMILEK 2010; DZURENDA and DELIISKI 2012, 2012a; KLEMENT and DETVAJ 2013; BARAŃSKI and WIERZBOWSKI 2013; PINCHEVSKA et al. 2016).They were organized by the control system in two-stage drying, with modification in the first phase by increasing the drying medium temperature in the drying kiln to 65oC and in the second step to 80oC, while the high-temperature drying process was organized by the warm air-steam mixture was carried out at 105oC (Fig. 1).

Fig. 1 Variation of temperature and moisture content for different phases of drying processes (BARANSKI 2018).

84

The wood moisture content during the drying processes was measured inside material in eight positions of drying kiln. The sensors (eight pieces) were placed in the material so that it was possible to measure it in several characteristic points of the kiln’s volume, e.g., at the middle and at the outer layers of the wood stack. The moisture content of samples was measured using sensors operating on resistance of drying material, while temperature inside samples was measured using T type thermocouples (copper-constantan). They were put appropriately at the center of the sample, in which measured was the water content, while thermocouples were put into drilled holes from the top side of the sample to the depth about 40-50 mm according to the recommendations by (KLEMENT and DETVAJ 2013).

The moisture content of each sample was measured, before and after sawing, using the TANEL Co. (TANEL) moisture content meter type WRD 100 that has a measurement accuracy of ±2%.

Machine tool and tools Cutting tests were performed on the frame sawing machine PRW15M (Fig. 2) with a hybrid dynamically balanced driving system and elliptical teeth trajectory movement (WASIELEWSKI and ORLOWSKI 2002) at the Department of Manufacturing Engineering and Automation (GUT, PL).

Fig. 2 Narrow-kerf frame sawing machine (sash gang saw) PRW15–M.

The machine settings were as follows: number of strokes of saw frame per min (nF), 685 spm; saw frame stroke (HF), 162 mm; number of saws in the gang (n), 5; and average cutting speed (vc), 3.69 m·s1. The saw blades were sharp, with stellite tipped teeth: overall set (kerf width) (St), 2 mm; saw blade thickness (s), 0.9 mm; free length of the saw blade (L0), 318 mm; tension stresses of saws in the gang (σN), 300 MPa; blade width (b), 30 mm; tooth pitch (P), 13 mm; tool side rake (γf), 9°; and tool side clearance 85

(αf), 14°. The only varying cutting parameter was feed speed, which was applied at two levels: vf1 ≈ 0.9 m·min-1 and vf2 ≈ 1.72 m·min-1. This corresponds to a feed per tooth (fz) of ~0.105 mm and ~0.2 mm, respectively. Lamellae with thicknesses of 5 ± 0.2 mm were obtained as a result of the re-sawing process. The actual value of the feed per tooth were computed on the basis of the sawing time taken from the plots of time changes of electrical power consumption (Fig. 2). The mean value of feed per tooth for a sash gang saw is calculated as: 1000  v f  P (1) fz  n RP  H RP where: strokes number of the frame nRP = 685 1/min, stroke of the frame HRP = 162 mm P = 13 mm is a tooth pitch, and vf in m·min-1 is calculated as:

vf 

Lp

(2)

60  t c

where: Lp is length of the sample in m, and tc is the real cutting time taken from the plot, e.g.

Fig. 3.

Fig. 3 Time changes of electrical power consumption while sawing at two levels of feed speed vf1 and vf2 of beech samples BKS43 and BKS50, which were dried in the experimental kiln at the GUT.

Sawdust collection and sieve analyses For granulometric analyses, samples of pine sawdust and beech sawdust from samples dried with different technics were taken isokinetically from the exhaust pipe of a frame sawing machine PRW-15 in accordance with a standard ISO 9096.

The moisture content of both types of dust samples was MC ≈ 10%, and was determined by the weight method.

Sieve analysis was carried out at the Department of Woodworking of the Technical University in Zvolen on an automated vibratory screening machine Retsch AS 200 control; a set of control stainless steel sieves, diameter of sieve 200 mm, height 50 mm, diameter of sieve mesh 2 mm, 1 mm, 0.50 mm, 0.25 mm, 0.125 mm, 0.080 mm, 0.063 mm, and 0.032 mm. The residues on each sieves and bottom were weighed on a digital laboratory balance EP 200 (f. BOSCH) to an accuracy of 0.001 g. The sieving parameters were an amplitude 2 mm/(g), with an interval of 10 s, and a time of 20 min.

86

With the purpose of specifying of information about size of the smallest particles of fine fraction of dry sawdust a microscopic analysis of granules of fraction of dry sawdust with the size lower than 125 μm was realised. The proposed analysis of dry sawdust was carried out by an optical method – analysis of the picture obtained from the microscope Nikon Optiphot–2 with the objective Nikon 4×. Granules of sawdust were scanned by three low-cost television CCD cameras HITACHI HV-C20 (RGB 752 × 582 pixel), with horizontal resolution 700 TV lines and evaluated by a software LUCIA-G 4.0 (Laboratory Universal Computer Image Analysis), installed on a PC with the processor Pentium 90 (RAM 32 MB) with the graphic card VGA Matrox Magic under the operation system Windows NT 4.0 Workstation. The program of analysis of picture LUCIA-G enables to identify the individual particles of disintegrated wood material, quantitative determination of individual particles situated in the analysed picture and basic information such as: width and length of particles, circularity expressing the measure of deviation of projection of a given chip shape from the projection of the shape of a circle according to the scheme: ψ

4 π S O2

(3)

where: S – surface of particle in m2, O – perimeter of particle in m.

RESULTS AND DISCUSSION In Table 1 the symbols of samples, methods of drying, values of feed speeds vf and feed per teeth fz, which were used in the tests, are presented.

Tab. 1 Symbols of samples, types of raw materials, feeding parameters and methods of drying.

Symbol

Wood species

feed speed vf

feed per tooth fz mm 0.1064 0.2099

BKP-32 BKP-33

Beech (Fagus sylvatica L.) Beech (Fagus sylvatica L.)

m·min-1 0.906 1.788

BKS-43

Beech (Fagus sylvatica L.)

1.023

0.1202

BKS-50

Beech (Fagus sylvatica L.)

1.580

0.1855

SOP-37 SOP-38

Pine (Pinus sylvestris L.) Pine (Pinus sylvestris L.)

1.756 0.860

0.2062 0.1009

SOS-50

Pine (Pinus sylvestris L.)

0.977

0.1147

SOS-52

Pine (Pinus sylvestris L.)

1.580

0.1856

operating methods of drying warm air drying warm air drying warm air-steam mixture kiln drying at 105°C warm air-steam mixture kiln drying at 105°C warm air drying warm air drying warm air-steam mixture kiln drying at 105°C warm air-steam mixture kiln drying at 105°C

The results of the sieve analysis - size distribution of the dry chips of pine sawdust dried in different operating methods are given in Table 2. On the other hand, Table 3 shows the results of the sieve analysis - size distribution of the dry chips of beech sawdust dried in different operating methods.

Figure 4 presents the cumulative histogram residue granularity plots of sawdust obtained during the sawing process of pine wood dried in two different operating methods.

Furthermore, Figure 5 presents the cumulative histogram residue granularity plots of sawdust obtained during the sawing process of beech wood, which was dried in two different operating methods.

87

Tab. 2 Granulometric analysis of pine sawdust dried in two operating methods.

Measure Mark of sieve mesh of fraction [mm] 2.000 1.000 0.500 0.250 0,125 0.063 0.032 < 0.032

coarse medium coarse fine

Representation of fractions of dry pine wood [%] Operating drying methods of pine wood warm air drying warm air-steam mixture kiln drying at 105°C SOP-37 SOP-38 SOS-50 SOS-52 1.92 2.58 1.67 2.53 18.06 4.53 18.09 21.59 53.51 43.58 54.88 48.48 19.80 35.18 18.76 20.45 5.33 11.54 4.74 5.26 1.38 2.11 1.49 1.48 0.01 0.48 0.38 0.21 0.00 0.00 0.00 0.00

Tab. 3 Granulometric analysis of beech sawdust dried in two operating methods.

Measure of sieve mesh [mm]

Cumulative remainder of fraction, %

2.000 1.000 0.500 0.250 0,125 0.063 0.032 < 0.032

Mark of fraction coarse medium coarse fine

Representation of fractions in dry beech wood [%] Operating drying methods of beech wood warm air drying warm air-steam mixture kiln drying at 105°C BKP-32 BKP-33 BKS-43 BKS-50 2.28 2.24 1.75 5.05 25.72 17.04 3.17 22.01 29.48 40.16 30.61 26.12 25.93 32.56 42.91 28.50 13.84 6.21 17.91 12.08 2.19 1.37 2.96 5.61 0.57 0.42 0.69 0.63 0.00 0.00 0.00 0.00

100

SOP-37 (fz = 0.2) SOP-38 (fz = 0.1)

80

SOS-50 (fz = 0.1) SOS-52 (fz = 0.2)

60 40 20

0 0.000

0.032

0.063

0.125 0.250 Mesh size, mm

0.500

1.000

2.000

Fig. 4 Histograms of residue of pine sawdust while sawing on the sash gang saw PRW15M. Legend: close to the samples symbols in parentheses values of feed per tooth are provided; SOP - pine wood samples dried in the warm air; SOS - pine wood samples dried in kiln with warm air-stream mixture at 105°C.

88

Cumulative remainder of fraction, %

100

BKP-32 (fz = 0.1) BKP-33 (fz = 0.2)

80

BKS-43 (fz = 0.1) BKS-50 (fz = 0.2)

60 40

20 0 0.000

0.032

0.063

0.125 0.250 Mesh size, mm

0.500

1.000

2.000

Fig. 5 Histograms of residue of beech sawdust while sawing on the sash gang saw PRW15M. Legend: close to the samples symbols in parentheses values of feed per tooth are provided; BKP - beech wood samples dried in the warm air; BKS - beech wood samples dried in kiln with warm air-stream mixture at 105°C.

The largest of about 80% of the production of chips from pinewood and beech sawdust from the PRW 15 M saw blade process, regardless of the way the drying methods are being analyzed, are chips with dimensions above 0.5 mm. Image analyses of shapes and dimensions of sawdust particles of coarse and medium coarse fractions have shown that the preponderance of sawdust granules of these fractions created in the process of longitudinal feeding while sawing of wood belong to the group of polydisperse fibrous materials, which have stick shapes with considerable elongation in one of their dimensions. Microscopic analyses of sizes and shapes of fine fraction particles of dry sawdust have revealed that particles of this fraction with their shape belong to the group of isometric particles, what means particles with the same values in all 3 dimensions. This statement is consistent with the analysis of the shape of sawdust presented in the paper by (DZURENDA et al. 2006).

No chips with dimensions less than 32 μm were measured, which results in a sawing process of pine and beech wood to PRW 15 m, and respirable fractions less than 10 μm are not detected. The finnest particles propably are created during sawing, but methods used in this work are not suitable to detect them. The results described by (HLÁSKOVA et al. 2015) shown the content of such particles in sawdust. It is therefore necessary to carry out further research using the laser diffraction method for determination of the finnest particles content in sawdust.

Drying with warm air-steam mixture at 105°C has the effect on increasing the production of dust fraction in beech sawdust, as can be explained by the higher brittleness of beechwood at the higher feed rate of workpiece on the frame sawing machine.

CONCLUSIONS Based on the carried out analyses, it can be concluded that: • In the sawing process of pine and beech wood on a PWR 15 M frame, regardless of the method of drying, a chip with a predominantly medium and coarse fraction over 0.5 mm is formed.

• The fraction of the dust fraction with a grain size below 125 μm does not exceed 6.5% for both analyzed types of wood.

89

• A dust fraction below the 32 μm grain size was not demonstrated for both pine and beech wood.

• Drying of beech wood with warm air-steam mixture at 105° C resulted in a higher production of a finer (dust fraction) compared with air dry drying. In the case of pine wood, this phenomenon has not been observed.

Based on granulometric analyzes of sawdust from pine and beech wood, it can be stated that the influence of air drying or with warm air-steam mixture drying at 80°C does not affect the change in grain size that would limit the existing processes of sawdust processing in the production of agglomerated materials or biofuel in the form of briquettes and pellets.

REFERENCES BARANSKI, J. 2018. Moisture content during and after high- and normal-temperature drying processes of wood. In Drying Technology 2018, 36(6): 751761, doi.org/10.1080/07373937.

2017.1355319.

BARANSKI, J., KLEMENT, I., VILKOVSKÁ, T., KONOPKA, A. High Temperature Drying Process of Beech Wood (Fagus sylvatica L.) with Different Zones of Sapwood and Red False Heartwood. In BioResources 12(1), 1861-1870. DOI. 10.15376/biores.12.1.1861-1870.

BARAŃSKI, J., WIERZBOWSKI, M. 2013. Influence of High Temperature Air-Steam Mixture Application on Time Wood Drying Process. In 21st International Wood Machining Seminar, Tsukuba, Japan, August 4–7, 2013, pp 475–482.

BASILICO, C., GENEVAUX, J. M., MARTIN, M. 1990. High Temperature Drying of Wood SemiIndustrial Kiln Experiments. In Drying Technol. 8(4): 751–765. doi: 10.1080/ 07373939008959913 BELJO LUČIČ, R., ČAVLOVIČ, A., DUKIČ, I., JUG, M., IŠTVANIČ, J., ŠKALJIČ, N. 2009. Machining propertirs of thermally modified beech-wood compared to steamed beech-wood. In Woodworking technique. Zagreb : DENONA, p. 315324.

DZURENDA, L. 2007. Sypká drevná hmota, vzduchotechnická doprava a odlučovanie. Zvolen : TU vo Zvolene.

DZURENDA L., ORLOWSKI K., GRZEŚKIEWICZ M. 2010. Effect of thermal modification of oak wood on sawdust granularity. In Drvna Industrija. 61(2): 8994.

DZURENDA, L., WASIELEWSKI ,R., ORLOWSKI, K. 2006. Granulometric analysis of dry sawdust from the sawing process on the frame sawing machine PRW15M = Granulometrická analýza suchej piliny z procesu pílenia borovicového dreva na rámovej píle PRW-15M. In Acta Facultatis Xylologiae Zvolen. 48(2): 5157.

DZURENDA, L., ORLOWSKI, K. 2011. The effect of thermal modification of ash wood on granularity and homogeneity of sawdust in the sawing process on a sash gang saw PRW 15-M in view of its technological usefulness. In Drewno, 54(186).

DZURENDA, L., DELIISKI, N. 2012. Convective drying of beech lumber without color changes of wood. In Drvna industrija. 63(2): 95103.

DZURENDA, L., DELIISKI, N. 2012a. Drying of Beech Timber in Chamber Drying Kilns by Regimes Preserving the Original Color of Wood. In Acta Facultatis Xylologiae Zvolen, 54(1): 3342.

DZURENDA L., ORLOWSKI K. 2011a. Influence of feed rate on the granularity and homogenity of oak sawdust obtained during the sawing process on the frame sawing machine PRW15M. In Proceedings of the 4th International Science Conference Woodworking Techniques /ed. Barcik S., Dvorak J./, Czech University of Life Sciences Prague, Univ Zagreb. Prague: Czech University of Life Sciences Prague, Czech Republic.

FUJIMOTO, K., TAKANO, T., OKUMURA, S. 2011. Difference in mass concentration of airborne dust during circular sawing of five wood-based materials. In J Wood Sci 57: 149154. https://doi.org/10.1007/s10086-010-1145-y HLÁSKOVÁ, L’., ROGOZINSKI, T., DOLNY, S., KOPECKỲ, Z., JEDINÁK, M. 2015. Content of respirable and inhalable fractions in dust created while sawing beech wood and its modifications. In Drewno 58(194): 135146

90

HLÁSKOVÁ, L., ROGOZINSKI, T., KOPECKÝ, Z. 2016. Influence of feed speed on the content of fine dust during cutting of two-side-laminated particleboards. In Drvna Ind. 67(1): 915. DOI 10.5552 /drind.2016.1417).

KLEMENT, I., SMILEK, P. 2010. Vplyv teploty na proces vysokoteplotného sušenia bukového reziva.

(Temperature in fluence on the process of high temperature drying of beech lumber). In Acta Facultatis Xylologiae Zvolen, 52(2): 3441.

KLEMENT, I., DETVAJ, J. 2013. Sawmilling and wood drying. Zvolen : Technical University in Zvolen, 2013, 156 p., ISBN 978-80-228-2502-3.

KLEMENT, I., HURAKOVA, T. 2015. The influence of drying characteristics and quality of spruce timber with content of reaction wood. In Acta Facultatis Xylologiae Zvolen, 57(1): 7582.

KLEMENT, I., VILKOVSKÁ, T., BARANSKI, J., KONOPKA, A. 2018. The impact of drying and steaming processes on surface color changes of tension and normal beech wood. In Drying Technology 2018, doi.org/ 10.1080/07373937.2018.1509219.

KMINIAK, R., BANSKI, A. 2000. Separation of Exhausted Chips from a CNC Machining center in Filter FR - SP 50/4 with Finet PES 4 Fabric. AIP Conf. Proc. 2000, 020011-1 – 02011-4, doi.org/10.1063/1.5049913.

KUČERKA, M., OČKAJOVÁ, A. 2018. Thermowood and granularity of abrasive wood dust. In Acta Facultatis Xylologiae Zvolen, 60(2): 4351. DOI: 10.17423/afx.2018.60.2.04 MARKOVÁ, I., LADOMERSKÝ, J., HRONCOVÁ, E., MRAČKOVÁ, E. 2018. Thermal parameters of beech wood dust. In BioRes. 13(2): 3098-3109. DOI: 10.15376/biores.13.2.3098-3109.

OČKAJOVÁ, A., BELJO LUČIĆ, R., ČAVLOVIĆ, A. TERENÒVÁ, J. 2006. Reduction of dustiness in sawing wood by universal circular saw. In Drvna industrija, 57(3): 119126.

OČKAJOVÁ, A., KUČERKA, M., BANSKI, A., ROGOZIŃSKI, T. 2016. Factors affecting the granularity of wood dust particles. In Chip and Chipless Woodworking Processes, 10(1): 137144.

PAŁUBICKI, B., ROGOZIŃSKI, T. 2016. Efficiency of chips removal during CNC machining of particleboard. In Wood Research, 61(5): 811818.

PERVAN, S. 2000. Priručnik za tehničko sušenie drva. Zagreb: Sand, 272 p.

PINCHEVSKA, O., SPIROCHKIN, A., SEDLIACIK, J., OLIYNYK, R. 2016. Quality Assessment of Lumber after Low Temperature Drying from the View of Stochastic Process Characteristics. In Wood Research, 61(6): 871884.

RATNASINGAM, J., SCHOLZ, F., NATTHONDAN, V. 2010. Particle size distribution of wood dust in rubberwood (Hevea Brasiliensis) furniture manufacturing. In European Journal of Wood and Wood Products, 68: 241242.

SEHLSTED-PERSSON, M. 1995. High-Temperature Drying of Scots Pine: A Comparison Between HT- and LT-Drying. In Holz als Roh- und Werkstoff, 53(2): 95–99. doi: 10.1007/bf02716400.

WALKER J.C.F., BUTTERFIELD B.G., HARRIS J.M., LANGRISH T.A.G., UPRICHARD J.M. 2006.

Primary Wood Processing. Principles and practice. 2nd edition. In Springer Netherlands, X, p. 596 DOI: 10.1007/1-4020-4393-7.

WASIELEWSKI R., ORLOWSKI K. 2002. Hybrid dynamically balanced saw frame drive. In Holz als Roh- und Werkstoff, 60(3): 202–206. DOI: 10.1007/s00107-002-0290-4.

WIERZBOWSKI, M., BARAŃSKI, J. 2010. Application of Steam Gas Mixture for Wood Drying Purposes. In Ann. Warsaw Univ. Life Sci. For. Wood Technol., 72: 458–462.

ISO 9096: 2017. Stationary source emissions – Manual determination of mass concentration of particulate matter.

AUTHORS ADDRESSES Kazimierz Orlowski (ORCID id: 0000-0003-1998-521X) Daniel Chuchala (ORCID id: 0000-0001-6368-6810) Tomasz Muzinski Jacek Baranski (ORCID id: 0000-0001-9040-9181) Gdansk University of Technology Faculty of Mechanical Engineering 91

Narutowicza 11/12 80-233 Gdansk Poland Adriรกn Banski Technical University in Zvolen Faculty of Wood Science and Technology T. G. Masaryka 24 960 53 Zvolen Slovak Republic Tomasz Rogozinski (ORCID id: 0000-0003-4957-1042) Poznaล University of Life Sciences Faculty of Wood Technology 28 Wojska Polskiego st.

60-637 Poznan Poland

92

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 93−101, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.09

DETERMINATION OF THE CUTTING POWER DURING MILLING OF WOOD-BASED MATERIALS Valentin Atanasov – Georgi Kovatchev ABSTRACT Results of milling of wood-based materials used in furniture production like plywood and medium-density fibreboard (MDF) are presented in the paper. The experiments were performed using the wood shaper with lower spindle position FD-3 located in a laboratory at the Department of Woodworking Machines, University of Forestry – Sofia., The input power to the cutting mechanism was reported by measuring device US301EM – Unisyst Engineering Ltd. and its software. Accordingly, a planned two-factor regression analysis was performed to determine the influence of feed speed and cutting area. Following the experiment, regression equations were developed. They can be used in the analytical determination of the influence of the factors considered on the target function – cutting power. The results show that cutting power of the plywood reaches significant values exceeding those of MDF and commonly used wood species studied in previous research carried out by the authors.

Key words: MDF, plywood, milling, cutting power, power-energetic indicators.

INTRODUCTION Composite wood-based materials such as plywood and MDF are widely used in modern furniture production, although they are relatively new materials. Based on their good physical-mechanical performance, they are used to make cabinet furniture, armchairs, beds, chairs, and others (SIMEONOVA 2015, JIVKOV et al. 2013). This wide application requires their participation in a variety of technological operations in the production of the listed furniture types. One of them is milling.

Milling machines with a lower spindle position have a significant application in furniture production, manufacture of doors, windows, etc. This is mainly due to their universality – i.e. they can be used for a variety of wood operations. It is necessary to determine the cutting power when designing them. On its base, the electric motor that is required to drive the cutting mechanism must be selected (FILIPOV 1979, VLASEV 2007).

In recent years, experimental studies that relate to the definition of power-energetic indicators in milling were conducted. They concern power, force, specific work of cutting and specific electricity consumption for widespread in furniture production wood species such as beech (Fagus sylvatica L.), white pine (Pinus sylvestris L.), meranti (Shorea leprosula) and koto (Pterygota macrocarpa) (GOCHEV et al. 2017, GOCHEV et al. 2018, KUBŠ et al. 2016, KRAUSS et al. 2016, ATANASOV, KOVATCHEV 2018). There are also studies for the milling of wood species, which are used for other purposes – poplar wood (Populus 93

tremula L.) (BARCĂ?K et al. 2008), moreover, for other machines such as circular saws (KOVĂ Ä&#x152;, MIKLEĹ 2010, KOPECKY et al. 2014, ORLOWSKI, OCHRYMIUK 2017) and band saws (ATANASOV 2014, CHUCHALA, ORLOWSKI 2018). However, no results related to powerenergetic indicators of woodworking machines in processing of wood-based composite materials have been found in the literature. This is what determines the aim of this study: to conduct experimental research on the influence of key factors on cutting power in milling of widespread in furniture manufacturing materials such as plywood and MDF.

THEORETICAL BACKGROUND Cutting forces occur in the interaction of wood with the cutting tool. These forces require a certain power to overcome them. It is called cutting power. In milling, these forces have a variable character because the thickness of the chip â&#x20AC;&#x201C; when the cutting edge enters it is zero, and when it comes out it is maximum (when the feed direction is opposite to the cutting). In the theory, mostly based on past experiments conducted in the territory of the former Soviet republics, for the calculation of cutting power, some empirical formulas are used. By accepting the average values of the parameters involved, they can be simplified with practical purposes (BERSHADSKIY, TSVETKOVA 1975, IVANOVSKIY et al. 1972). One of them is đ?&#x2018; đ?&#x2018;? = đ?&#x2018;&#x2DC;đ?&#x2018;? đ?&#x2018;&#x17D;đ?&#x2018;? đ?&#x2018;&#x17D;đ?&#x2018;&#x2019; đ?&#x2018;Łđ?&#x2018;&#x201C; ,

(1)

where kc is the specific cutting resistance, N¡m-2; ap â&#x20AC;&#x201C; axial depth of cut (cutting width), m; ae â&#x20AC;&#x201C; radial depth of cut, m; vf â&#x20AC;&#x201C; feed speed, m.s-1.

Figure 1 shows a simplified scheme of the milling process showing the average tangential cutting force P, feed per tooth fz, angle range Ď&#x2020;, average uncut chip thickness hm, uncut chip thickness h (working engagement), cutting diameter Dc, feed speed vf and cutting speed vc directions. The latter direction is overlapped with a direction of the cutting force.

Fig. 1 Scheme of the milling process with a groove cutter.

In the literature, the specific cutting energy ec is defined as an amount of work required to convert a cubic meter of wood into sawdust (GRIGOROV 1992), and its value is equal to 94

the specific cutting resistance kc which is commonly used in the industrial practice.

Accordingly, it is noted that a number of factors can be influenced by, which can be assumed to have a direct effect on the cutting power. They are related to the physical and mechanical characteristics of the type of wood, its density, moisture content, temperature, etc. It is also noted that the kinematics of the process, the condition of the cutting tool, its linear and angular parameters, the type of cutting, etc., have an impact as well.

MATERIAL AND METHODS The experimental studies were conducted by a wood shaper with a lower spindle position, model FD-3 (ZDM Plovdiv, Bulgaria). Some of its more important technical parameters are: power and resolutions of the electric motor (AC, asynchronous) – Nm = 3000 W and nm = 2880 min-1, power supply voltage Np.s.= 3 x 380 V/50 Hz and diameter of the spindle Dm = 30 mm. The cutting tool is a groove cutter with the following basic parameters – cutting diameter Dc = 140 mm, thickness of the cutting plates s = 12 mm, front angle of cutting γ = 20 ̊, angle of sharpening β = 58 ̊, number of cutting teeth z = 6 pcs, material for hard-alloy plates – HW, weight m = 0,910 kg. The cutting tool is brand new and used only to conduct experiments. This gives reason to assume that in this case the impact of cutting edge wear is minimal and does not affect the process. To drive the spindle a V-ribbed belt was used. With the respective gear ratio, at diameters of the pulleys D1 = 190 mm (drive pulley) and D2 = 90 mm (driven pulley), taking into account the sliding coefficient, ns ≈ 6045 min-1 spindle resolutions were obtained, hence, calculated cutting speed is vc = 44,3 m.s-1.

The experimental samples are MDF blanks with a length L = 1200 mm, width B = 60 mm, thickness δ ≈ 20 mm and plywood blanks (made of beech veneer and urea formaldehyde adhesive) with the same dimensions. The density of the blanks was calculated based on theirs weight and volume. The weight is measured by electronic scale RADWAG WLC 1/A2 (Poland). The volume was determined by measuring their dimensions with a caliper and measuring tape. The general view of some of them can be seen in Fig. 2.

Fig. 2 Experimental samples of MDF and plywood.

In this study, the cutting power was determined experimentally. For this purpose, empirical equations such as formula 1 are not used. The cutting power is calculated by the formula 2. Previously, the efficiency coefficient of the cutting mechanism was determined (formula 3) (GOCHEV et al. 2017).

95

đ?&#x2018; đ?&#x2018;&#x2122;đ?&#x2018;&#x153;đ?&#x2018;&#x17D;đ?&#x2018;&#x2018; â&#x2C6;&#x2019; đ?&#x2018; đ?&#x2018;&#x2013;đ?&#x2018;&#x2018;đ?&#x2018;&#x2122;đ?&#x2018;&#x2019;

đ?&#x2018; đ?&#x2018;? = (

100

) đ?&#x153;&#x201A;,

(2)

where Nidle is input power of the cutting mechanism in idle condition, W; Nload â&#x20AC;&#x201C; input power of the cutting mechanism in load condition, W. đ?&#x153;&#x201A; = (1 â&#x2C6;&#x2019;

đ?&#x2018; đ?&#x2018;&#x2013;đ?&#x2018;&#x2018;đ?&#x2018;&#x2122;đ?&#x2018;&#x2019; đ?&#x2018; đ?&#x2018;&#x2122;đ?&#x2018;&#x153;đ?&#x2018;&#x17D;đ?&#x2018;&#x2018;

) 100.

(3)

To measure the input power of the cutting mechanism in load and idle conditions, the device US301EM â&#x20AC;&#x201C; Unisyst Ltd. (Bulgaria) was used. It allows measurement of active, reactive, full power, current, voltage, etc. â&#x20AC;&#x201C; in phases and in general. Three current CNCÂŽ CURRENT TRANSFORMER and three voltage transformers UNITRAF AD Ltd. were used to connect it to the electrical network of the machine. The required configuration was made as well. Using specialized software from the manufacturer, the results are automatically imported into Microsoft Excel and their average value was found.

The cutting power Nc (Y) is determined by conducting a planned two-factor regression analysis. As factors (input parameters), the feed speed vf (X1) and the area of cutting were selected Đ? (X2â&#x20AC;&#x201C; it is obtained by multiplying the thickness of the cutting part of the tooth s â&#x20AC;&#x201C; which is equal to the axial depth of cut ap (the width of the cut) and radial depth of cut ae â&#x20AC;&#x201C; the depth of the groove). The levels of variation of the relevant factors are determined by conducting preliminary experimental experiments. Furthermore, they are the same as those used in previous experiments by the authors in longitudinal milling of solid wood. The reason for this is the ability to perform a comparative analysis between the values of milling of composite materials and solid wood â&#x20AC;&#x201C; something which, due to the wide variation in experimental conditions, is difficult to accomplish with the part of the studies mentioned in Introduction. The levels of variation are X1 = 2, 6 and 10 m.min-1, ĐĽ2 = 48, 96 and 144 mm2.

Figure 3 is a graphical representation of the cutting process by a part of the cutting mechanism and machined detail.

Fig. 3 Experimental scheme.

Table 1 shows the experimental matrix with the combination of factors in explicit and encoded form. In addition, some additional experiments have been performed, the levels of which correspond to the middle of the factor space ĐĽ1= 0 and ĐĽ2 = 0. Due to their large 96

volume, the steps for the computation of the regression analysis are not described in this study. They can be seen in the literature on mathematical modelling of technological objects VUCHKOV (1986). To obtain the regression equations that describe the relevant processes and their verification, QstatLab5 and Microsoft Excel software products were used.

Tab. 1 Experimental Matrix. № 1.

2.

3.

4.

5.

6.

7.

8.

9.

X1 (vf) +1 +1 –1 –1 0 0 +1 0 –1

vf, m.min-1 10 10 2 2 6 6 10 6 2

X2 (A) +1 –1 +1 –1 0 +1 0 –1 0

A, mm2 144 48 144 48 96 144 96 48 96

RESULTS AND DISCUSSION When calculating the density of the test samples were obtained values for MDFρmdf = 585 kg.m-3 and plywood ρpl = 735 kg.m-3. As can be seen from the values obtained, the density of plywood is 150 kg.m-3 higher than that of MDF. The following regression equations were inferred from the processing of the obtained experimental results. After further calculations for the Fisher criteria and comparing it to the table value, it was proven that they are adequate and may be used for analysis of the respective process (factor levels are encoded: -1, 0, +1): MDF (4) Ncmdf(Y) = 0.258+0.145vf+0.203A+0.048vf2+0.022A2+0.122vfA; Plywood Ncpl(Y) = 0.965+0.399vf +0.593A-0.062vf 2-0.074A2+0.258vfA. (5) As it can be seen from the equations, the regression coefficient for the two materials is higher in front of the factor A – the area of milling. This means that it has a greater impact on cutting power. It is also seen that this factor is more dominant in plywood A = 0.593. The trend of the influence of factors is similar in longitudinal milling of solid wood – white pine (Pinus sylvestris L.), meranti (Shorea leprosula) and koto (Pterygota macrocarpa).

Figure 4 graphically shows the results after solving the MDF equation – for the three considered areas of milling. It is noted that at the lowest feed speed level 2 m .min1, the power values for milling areas 44 and 96 mm2 are approximately the same. Even during the experimental tests themselves, it was clearly felt that the load on the electric motor was minimal. Subsequently, after reaching the feed rate of about 4 m.min-1, it is clearly evident that the curve corresponding to a 96 mm2 milling area begins to rise more intensively. The feed speed has the most insignificant impact at a level of milling area 48 mm2 – the difference between the values at vf = 2 m.min1 and vf = 10 m.min-1 is minimum – 0.18 kW. The highest value obtained at vf = 10 m.min1 and A = 144 mm2 is approximately 0.8 kW, which is higher than expected – bearing in mind the homogeneous structure of this material. It can be concluded that the cutting power at milling of MDF, at the maximum levels of the factors considered, is close to that obtained in the longitudinal milling of wood species such as meranti (Shorea leprosula ≈ 0.9 kW) and koto (Pterygota macrocarpa ≈ 0.8 kW) (ATANASOV, KOVATCHEV 2018). As a reason for this, the adhesive added to the preparation of medium density wood fiber boards can be mentioned. Its potential abrasive impact and 97

the likely increase in the wear of the teeth during the experiments themselves can be mentioned as well. However, it can be argued that this is unlikely. The reason is that in previous studies it was found that the cutting speed factor Vc had the lowest impact on target function and 6045 spindle revolutions were determined to be optimal (Gochev et al. 2017, Gochev et al. 2018). For this reason, in the present study it is not included. A two-factor experiment that does not require a large number of tests was carried out – i.e. the influence of wear on cutting edges may be ignored.

Fig. 4 Influence of feed speed on cutting power at various areas of MDF milling.

The influence of the feed speed in milling of plywood is represented graphically in Figure 5. It is also seen here that at the smallest milling area the difference between the first and the last value (2 and 10 m.min-1) is the lowest– i.e. for A = 48 mm2 the analysis is identical to that for MDF. It is also evident from the curves that only in the lowest levels of variation of the factors vf = 2 m.min-1 and A = 48 mm2 the results for the two materials differ minimally. Subsequently, they increased significantly, and in milling areas of 96 and 144 mm2, the calculated values and the resulting curves for plywood exceed those for MDF approximately three times over their entire length. In addition, when comparing the plywood with previous studies conducted for the same conditions, the significant dominance in cutting power was noted. This is clearly expressed at the highest levels of variation of factors where the power significantly exceeds that of longitudinal milling of solid wood like meranti (Shorea leprosula), koto (Pterygota macrocarpa), white pine (Pinus sylvestris L.) and even approximately 2 times greater than that of beech (Fagus sylvatica L.) (ATANASOV, KOVATCHEV 2018, GOCHEV et al. 2017, GOCHEV et al. 2018). The significant load was also felt during the tests – by changing the noise of the engine. The reason for this can be found in the greater amount of glue. Moreover, taking into account the technology for obtaining these materials, the cutting here can be regarded as a more complex and energy intensive, as opposed to clear longitudinal cutting of solid wood. It should also be noted that the material used is beech veneer, which has greater density and strength. This determines the plywood as a material whose cutting requires high power. For this reason, it is not advisable to process it under severe cutting modes when the electric motor of the cutting unit has a nominal power of 3 kW (overloading is only allowed for short periods of time). In this case, if it is impossible or impractical to use a more powerful machine, it is advisable to cut larger areas in several passes through the machine or at low feed rates.

98

Fig. 5 Influence of feed speed on cutting power at various areas of plywood milling.

Figure 6 shows the effect of the more significant factor (A) on the cutting power at the highest feed rate (10 m.min-1) for both materials. From this figure, the trend that for each milling area the cutting power of the plywood is about 3 times greater than that of MDF is clearly visible.

Fig. 6 Influence of milling area on cutting power in milling of MDF and plywood.

CONCLUSIONS On the basis of the conducted experimental studies, the following more important conclusions and recommendations can be made: 1. Adequate regression equations that can be used to analyze the influence of feed speed and milling area on the cutting power of machining MDF and plywood were obtained.

This power can be considered as average, since at this cutting speed each of the teeth of the tool have passed through the material processed about 100 times per second. For this reason, it was practically very difficult to determine the influence of the thickness of the chip â&#x20AC;&#x201C; i.e. it is necessary to adopt an idealization of the cutting process.

2. When cutting plywood, the cutting power exceeds 2 kW at higher levels of variation of the factors considered. Such values exceed significantly those obtained under the same

99

conditions but in longitudinal milling of solid wood – white pine (Pinus sylvestris L.), beech (Fagus sylvatica L.), meranti (Shorea leprosula), koto (Pterygota macrocarpa) etc.

3. The cutting power during the processing of MDF, considering its homogeneous structure and a relatively low density, is approximately equal to that obtained with the aforementioned tropical wood species – meranti (Shorea leprosula), koto (Pterygota macrocarpa).

REFERENCES ATANASOV, V. 2014. Izsledvane na operativnite pokazateli za mobilni khorizontalni lentovi trioni Disertatsiya na Ph.D, Sofia : Lesotekhnicheski universitet, 196 s.

ATANASOV, V., KOVATCHEV, G. 2018. Determination of the cutting power in processing some deciduous wood species. In The 8th Conference on Hardwood Research and Utilisation in Europe – Sopron, 2018, 8: 53-55. ISBN 978-963-359-096-6. ISSN 2631-004X. BARCÍK, ST., PIVOLUSKOVÁ, E., KMINIAK, R. 2008. Effect of Technological Parameters and Wood Properties on Cutting Power in Plane Milling of Juvenile Poplar Wood. In Drvna industrija, 2008, 59(3): 107112. ISSN: 0012-6772. EISSN: 1847-1153.

BERSHADSKY, A., TSVETKOVA, N. 1975. Rezka dereva. Minsk, 1975, 304 s.

CHUCHALA, D., ORLOWSKI, K. 2018. Forecasting values of cutting power for the sawing process of impregnated pine wood on band sawing machine. In Mechanik. 2018, (89): 766768. DOI: 10.17814/mechanik.2018.8-9.128.

FILIPOV, G. 1979. Mashini za proizvodstvo na mebeli i obzavezhdane. Sofia, 1979, 462 s.

GOCHEV, ZH., VUKOV, G., VITCHEV, P., ATANASOV, V., KOVATCHEV, G. 2017. Modelirane i eksperimentalno izsledvane na protsesite pri nadlŭzhno frezovane na masivno dŭrvo. In Tema № 22, Sofia : NIS / LTU, 2017, 76 s.

GOCHEV, ZH., VUKOV, G., ATANASOV, V., VICHEV, P., KOVACHEV, G. 2018. Study on the Power – Energetic Indicators of a Universal Milling Machine. In The VIII-th International Scientific and Technical Conference Innovations in Forest Industry and Engineering Design. Sofia, 2018, 1: 1824.

ISSN 1314-6149. e-ISSN 2367-6663.

GRIGOROV, P. 1992. Ryazane na dŭrvo. Zemizdat. Sofia, 1992, 319 s.

IVANOVSKY, E., VASILEVSKAYA, P., LAUTIER, M. 1972. Novyye issledovaniya lesozagotovok. In Lesnaya promyshlennost', 1972, 128 s.

JIVKOV, V., SIMEONOVA, R., MARINOVA, A. 2013. Influence оf the Veneer Quality and Load Direction on the Strength Properties of Beech Plywood as Structural Material for Furniture. In Innovation in Woodworking Industry and Engineering Design, 2013, 1: 86-92. ISSN 1314-6149. eISSN 2367-6663.

KOPECKY, Z., HLASKOVA, L., ORŁOWSKI, K. 2014. An Innovative Approach to Prediction Energetic Effects of Wood Cutting Process with Circular-Saw Blades. In Wood research, 2014, 59(5): 827834. ISSN: 1336-4561.

KOVÁČ, J., MIKLEŠ, M. 2010. Research on Individual Parameters for Cutting Power of Woodcutting Process by Circular Saws. In Journal of forest science, 2010, 56 (6): 271–277. ISSN 1212-4834 (Print). ISSN 1805-935X (On-line).

KRAUSS, А., PIERNIK, M, PINKOWSKI, G. 2016. Cutting Power during Milling of Thermally Modified Pine Wood. In Drvna industrija, 2016, 67 (3): 215-222. ISSN: 0012-6772. EISSN: 1847-1153.

KUBŠ, J., GAFF, M., BARCÍK, ST. 2016. Factors Affecting the Consumption of Energy during the Milling of Thermally Modified and Unmodified Beech Wood. In Bio Resources, 2016, 11(1): 736747. ISSN: 1930-2126.

ORLOWSKI, K., OCHRYMIUK, T. 2017. A newly-developed model for predicting cutting power during wood sawing with circular saw blades. In Maderas-ciencia y tecnologia. 19(2): 149162. DOI: 10.4067/S0718-221X2017005000013 SIMEONOVA, R. 2015. Kharakteristiki na yakostta i deformatsiyata na ŭglovite sŭedineniya na strukturni elementi, izraboteni ot shperplat. Disertatsiya na Ph.D, Sofia : Lesotekhnicheski universitet, 166 s.

100

VLASEV, V. 2007. Uprazhnenie za uprazhnyavane na dŭrvoobrabotvashti mashini. Sofia : Lesotekhnicheski universitet,78 s.

VUCHKOV, I., STOYANOV, S. 1986. Matematichesko modelirane i optimizatsiya na tekhnologichni obekti. Sofia : Tekhnika na Dŭrzhavnoto izdatelstvo. 341 s.

ACKNOWLEDGEMENTS This document was supported by the National Program "Young Scientists and Postdoctoral Students", Institution - University of Forestry, Faculty of Forest Industry (FFI).

AUTHORS’ ADDRESSES Chief Assist. Prof. Valentin Atanasov, PhD, Assist. Prof. Georgi Kovatchev, PhD, University of Forestry Faculty of Forest Industry Kliment Ohridski Blvd. 10, 1797 Sofia Bulgaria vatanasov_2000@ltu.bg g.kovatchev@gmail.com

101

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 103−110, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.10

THE PROPERTIES OF REINFORCED LIGHTWEIGHT FLAT PRESSED WOOD PLASTIC COMPOSITES Pavlo Lyutyy – Pavlo Bekhta – Galyna Ortynska – Ján Sedliačik ABSTRACT The properties of reinforced lightweight flat pressed wood plastic boards are examined in this study. The laboratory shredded low density polyethylene (rLDPE) particles, wood particles (WP) and expanded polystyrene (EPS) were used for making WPC boards. The fiberglass mesh and polypropylene fibers were used for reinforcing the lightweight WPC. The bending strength (MOR), modulus of elasticity (MOE), tensile strength perpendicular to the plane of the board (or internal bond - IB) and thickness swelling of reinforced lightweight WPC after immersions in water for 24 hours (TS) were evaluated. The results showed that the strength properties of WPC boards were improved with the addition of polypropylene fibers and fiberglass mesh. The fact that the values of MOR/MOE increased with adding the polypropylene fiber and fiberglass mesh by 11.6%/7.7% and 77.5%/51.3%, was found. The IB of lightweigth WPC reinforced with polypropylene fiber and fiberglass mesh increased up to 7.4% and 11.1% compared to unreinforced lightweight WPC. The TS was not affected by the addition of reinforcing materials. However, the obtained values of MOR, MOE, IB of the reinforced lightweight WPC meet the requirements of the standards EN 16368 and EN 312.

Key words: lightweight wood plastic composites, expanded polystyrene, fiberglass mesh, polypropylene fibers, reinforced materials.

INTRODUCTION One of the problems of increasing the lightweight boards development is the shortage of raw material and the need to reduce costs in the wood-based composites industry. Also the fast-growing market of knockdown furniture, reduce material weight, customers’ packaging and transportation demands leads to production of new products and production concepts which increase the resource efficiency without compromising the mechanical properties of the composites (BARBU and VAN RIET 2008).

The scientists have been working to develop the various methods (tendency) for reducing the density of wood composite materials, particularly those intended for the manufacture of furniture, such as particleboards and fibreboards. So, ultra-low density fiberboards with the density 55 kg/m3 are produced without applying any pressing pressure (YONGQUN et al. 2011). But mechanical properties of ultra-low density fiberboards still remain low in comparison with medium density boards (MDF) due to their extremely low density.

103

Among the production techniques, extrusion is another way to manufacture of extruded low density particleboards (KOLLMANN et al. 1975). The density of such boards varies from 210 to 460 kg/m3 according to “Sauerland-spanplatte” technical characteristic.

But, the bending strength (MOR) of these boards is unsatisfactory. Also there were the researchers of using of different agricultural plants (hemp, sunflower, topinambur, maize and miscanthus) such as light particles for the production of lightweight particleboards.

However the lightweight boards bending strength does not meet the requirement of EN 312 type P2 (BALDUCCI et al. 2008).

Different foamable polystyrene and already foamed polystyrene particles could be used for the production of lightweight flat pressed wood plastic composites (LYUTYY et al.

2018). But all investigated boards produced with different compositions do not comply with the requirements of EN 312. That’s why the modification additives need to be added to the lightweight WPC for the manufacture of composites with more homogeneous properties.

Lightweight boards could be laminated by different materials such as high press laminate (HPL), MDF, high density fiberboard (HDF) and plywood for the improvement of their mechanical and physical properties (JIVKOV et al. 2012).

There have been different attempts to improve the physical and mechanical properties of flat pressed WPC boards namely using right size of raw material, optimum mixture and preparation of the elements in the WPC board or using reinforcement materials.

Reinforcements for the composites can be fibers, fabrics particles or whiskers. The points to be noted in selecting the reinforcements include compatibility with matrix material, thermal stability, density, melting temperature etc. The efficiency of discontinuously reinforced composites is dependent on tensile strength and density of reinforcing phases (BLEDZKI et al. 2002). Flat pressed WPC boards were surface-reinforced by two different types of thermoplastic face layers [a commingled fabric made of glass and polypropylene filaments (TWINTEX ) and a glass fabric reinforced polypropylene laminate (S-TEX)] to improve flexural properties (SCHMIDT et al. 2013). The reinforced WPC boards exhibited greatly improved flexural properties, with MOE/MOR values up to nearly 10000/90 N/mm2). RIZVI and SEMERALUL (2008) added 5% glass fibers to WPC specimen with varying amounts of wood fiber content and found significant improvement in bonding strength and modulus of elasticity. Improvement of MOR and MOE up to 60% was achieved by applying the reinforcement on both surfaces of the WPC deck boards.

Nowadays, there is very little research on the the reinforcing of WPCs and no study has been reported concerning the reinforcing of lightweight WPCs. But lightweight WPCs still suffer from lack of strength and toughness, which can be improved by adding of reinforcement materials. Therefore, the objective of this study was to investigate the possibility of improvement of mechanical properties of lightweight flat pressed wood plastic composites by reinforcement materials using.

MATERIALS AND METHODS The WPC boards were made using laboratory shredded recycled low density polyethylene (rLDPE) and wood particles (WP) with moisture content of 2-3 % and expanded polystyrene (EPS). The WP were commercially produced for particleboard mill.

The rLDPE particles were used as the polymer matrix. The fraction analysis of rLDPE and WP particles is presented in Table 1. The properties of EPS granules were as follows: diameter – 24 mm; the bulk density – in the range from 6 to 10 kg/m3. The fiberglass mesh A-125 (Fig. 1, a) and polypropylene fibers HLV-52 (Fig. 1,b) were used for reinforcing lightweight WPC. The density of the polypropylene fibers HLV-52 was 91 g/cm3; melting temperature 104

– 160 °C; fibers length - 12 mm and fibers diametr – 18 microns. The density of fiberglass mesh A-125 was 0.125 kg/m3; mesh thickness – 0.47 mm; cell size – 4×5 mm.

The ratio of WP to rLDPE was 60:40. The content of EPS was 2 % from the weight of the WP/rLDPE composition.

Tab. 1 Fraction analysis (by % weight).

Components WP rLDPE

(a)

/5 4.75 9.53

5/4 12.2 3.04

4/2 15.79 53.14

Screen hole size (mm) 2/1 1/0.63 0.63/0.315 40.28 15.67 9.13 32.45 1.83 

0.315/0 2.18 

(b) Fig. 1 Reinforcing materials: (a) – fiberglass mesh; (b) – polypropylene fibers.

The manufacture of lightweight WPC boards reinforced by polypropylene fibers. The content of polypropylene fibers was 7.5% from the weight of the WP/rLDPE composition.

WP, rLDPE, EPS and polypropylene fibers (in the natural dry state) were mixed by hand during 10 minutes. The WPC composition was formed into the packet at the open form and thereafter the packet was transferred into hot press (Figure 2).

Fig. 2 Manufacturing process of lightweight WPC boards reinforced by polypropylene fibers.

The manufacture of lightweight WPC boards reinforced by fiberglass mesh. After mixing of WPC composition (WP/rLDPE/EPS), it was divided into three equal parts. Then first part was formed in the open press form and fiberglass mesh was putted above.

105

Whereupon was formed middle layer and above was formed fiberglass mesh and the last layer of WPC composition (Figure 3).

Fig. 3 Manufacturing process of lightweight WPC boards reinforced by fiberglass mesh.

The WPC packets were hot pressed under pressure 3.5 MPa at temperature 180 °C and duration 1 min/mm in a one-step process. At the end of the hot-pressing cycle, the WPC board was immediately moved from the hot press into the cold press with the temperature 20 °C for the cooling to the temperature 3040 °C. WPC boards with 8 mm thickness were trimmed to a final size of 250 × 230 mm. The target densities of lightweight WPC boards were 600 kg/m3. WPC boards with the same target density but without reinforced materials were manufactured at the same pressing parameters.

Finally, the manufactured WPC boards were conditioned in a climate room with the relative humidity 65 ± 5% and temperature 20 ± 2 °C before being cut into test specimens.

Bending strength (MOR), modulus of elasticity (MOE), tensile strength perpendicular to the plane of the board or internal bond (IB) and thickness swelling after immersions in water 24 hours (TS) of the lightweight WPC boards were evaluated according to EN 310, EN 317 and EN 319 standards respectively.

RESULTS AND DISCUSSION Statistical analysis of variance (ANOVA) was conducted to determine whether there was a significant difference between mechanical and physical properties of reinforced lightweight WPC with type of reinforcing material (Table 2). It was found that the type of reinforcing material significantly influenced the mechanical properties but had no effect on the thickness swelling of WPC boards. The greatest effect on the MOR, MOE and IB values was observed for the fiberglass mesh.

106

Tab. 2 The test results of ANOVA for properties of reinforced lightweight WPC samples (Îą=0.05).

Source Type of additive

Dependent variable MOR MOE IB TS

Sum of Squares

df

206.647 702736.000 .004 19.043

Mean Square

2 2 2 2

103.323 351368.000 .002 9.521

F 208.649 150.738 36.580 1.352

Sig. .000 .000 .000 .283

The highest values of MOR and MOE were observed in lightweight WPC boards with using fiberglass mesh (Figure 4). The fiberglass mesh create more structural construction of the composition. The use of polypropylene fibers did not allow to create homogeneous and completely correct placement of fibers in the composite. As a confirmation of this the Eglass fiber for two different fiber lengths (5 and 10 mm) were used (ASHRAFI et al. 2011).

Fiber reinforcement did not have significant effect on the elastic modulus of WPC. However, the flexural strength of WPC reinforced with E-glass fibers is considerably lower than the unreinforced one. Again, the presence of fibers does not have significant effect on the elastic modulus, while even 5% addition of fibers results in significant reduction in the flexural strength of WPCs. Orientation of the fibers in the direction of the load is critical for enhancing the mechanical properties of composites (ASHRAFI et al. 2011).

However, it was difficult to form a packet where all the reinforced fibers will be directed perpendicular to the load action. That's why the values MOR and MOE were significantly higher in lightweigth composites with fiberglass mesh in composition. The MOR of WPC reinforced with fiberglass mesh is considerably higher than the unreinforced one. Any way, the using of fiberglass mesh and polypropylene fibers leads to increase of MOR and MOE. In particular, the values of MOR/MOE increased by 11.6%/7.7% and 77.5%/51.3%, recpectively with adding of polypropylene fiber and fiberglass mesh.

Figure 5(a) shows the effect of the type of reinforcing materials on the IB of lightweight WPC. It was found that the IB of lightweight WPC reinforced with fiberglass mesh significantly increases up to 11.1% and reinforced with polypropylene fiber increases up to 7.4 % in comparison with control.

Reinforcement with fiberglass mesh and polypropylene fibers did not have a significant effect on TS (Figure 5(b)). The values of TS for control and TS for reinforced composites are approximately on the same level.

1200

16

1016.7 c

13.51 c

1100

14

1000 8.49 b

10 8

MOE (MPa)

MOR (MPa)

12

7.61 a

800

724 b 672 a

700

6

600

4

500

400

2

(a)

900

control

polypropylene fiber

fiberglass mesh

(b)

control

polypropylene fiber fiberglass mesh

Fig. 4 The influence of the type of reinforcing material on the MOR (a) and MOE (b) of lightweight WPC boards. (Vertical bar: standard deviation; means followed by different letter denotes that they are statistically different according to the Duncan test at Îą=0.05).

107

0.35

26

0.30

22

0.27 a

19.7 b

17.5 ab

20

TS (%)

IB (MPa)

19.2 ab

24

0.30 c 0.29 b

0.25

0.20

18

16 14 12

0.15

(a)

control

polypropylene fiber fiberglass mesh

10 control

(b)

polypropylene fiber fiberglass mesh

Fig. 5 The influence of the type of reinforcing material on the IB (a) and TS (b) of lightweight WPC boards. (Vertical bar: standard deviation; means followed by different letter denotes that they are statistically different according to the Duncan test at Îą=0.05).

Lightweight WPC boards with polypropylene fibers could be classified as type LP1 (EN 16368), LD-1 and LD-2 (ANSI A208.1) (Table 3). However, boards with fiberglass mesh also meet the requirements LP2 (EN 16368) P1 (EN 312) and do not comply with the P2 (EN 312) norms regarding to the values of MOE. According to ANSI A208.1 the investigated WPC boards could be used as the door core.

Tab. 3 The requirements to the properties of lightweight and conventional particleboards according to EN standards and their comparison with the investigated reinforced lightweight WPC boards.

Board Type Reinforced lightweight WPC boards: Polypropylene fiber Fiberglass mesh Control lightweight WPC boards Lightweight particleboards (EN 16368): LP1 LP2 Particleboards (EN 312): P1 P2 P3 Low density particleboards (ANSI A208.1) LD-1 LD-2

MOR/MOE (MPa)

IB (MPa)

TS (%)

8.49 / 724 13.51 / 1016.7 7.61 / 672

0.29 0.30 0.27

17.5 19.2 19.7

4.0 / 550 8.0 / 1000

0.28 0.40

-

10.5 11.0 / 1800 15.0 / 2050

0.28 0.40 0.45

17.0

2.8 / 500 2.8 / 500

0.10 0.14

-

CONCLUSIONS The using of polypropylene fibers and fiberglass mesh creates the possibility of producing reinforced lightweight WPC boards in the laboratory scale. The influence of the type of reinforcing material on the properties of lightweight WPC boards was studied. The values of MOR/MOE increased by 11.6%/7.7% and 77.5%/51.3%, recpectively with adding of polypropylene fiber and fiberglass mesh. It was found that the highest values of MOR and MOE were observed in lightweight WPC boards with using fiberglass mesh. The obtained values of TS of the reinforced composites and control lightweight WPC are not significantly different. Based on the obtained results, lightweight WPC boards with polypropylene fibers could be classified as type LP1 (EN 16368), LD-1 and LD-2 (ANSI A208.1). However, boards with fiberglass mesh also meet the requirements LP2 (EN 16368) P1 (EN 312). These 108

positive results from the laboratory experiments give the base to plan experiments in industrial conditions.

REFERENCES ANSI A208.1. 2009: Particleboard. American Natioanal Standard. Composite Panel Association.

ASHRAFI, M., VAZIRI, A., NAYEB-HASHEMI, H. 2011. Effect of processing variables and fiber reinforcement on the mechanical properties of wood plastic composites. In Journal of Reinforced Plastics and Composites 30(23): 19391945.

BALDUCCI, F., HARPER, C., MEINLSCHMIDT, P., DIX, B., SANASI, A. 2008. Development of innovative particleboard panels. In Drvna Industrija, 59(3): 131136.

BARBU, MC, VAN RIET, C. 2008. European panels market developments - current situation and trends. In The Proceeding of the SWST Annual Convention, 1012 Nov. Concepción, Chile.

BLEDZKI A.K., SPERBER V.E., FARUK O. 2002. Natural and wood fibre reinforcement in polymers. In Rapra Review Reports, 13(8): 158 р.

EN 16368: 2014. Lightweight Particleboards – Specifications. European Committee for Standardization, Brussels.

EN 310: 1993. Wood-based panels – Determination of modulus of elasticity in bending and of bending strength. European Committee for Standardization, Brussels.

EN 312: 2010. Particleboards – Specifications. European Committee for Standardization, Brussels.

EN 317: 1993. Particleboards and fibreboards – Determination of swelling in thickness after immersion in water. European Committee for Standardization, Brussels.

EN 319: 1993. Particleboards and fibreboards – Determination of tensile strength perpendicular to the plane of the board. European Committee for Standardization, Brussels.

JIVKOV, V., SIMEONOVA, R., KAMENOV, P., MARINOVA, A. 2012. Strength properties of new lightweight panels for furniture and interiors. 23rd International scientific conference. In Wood is good – with knowledge and technology to a competitive forestry and wood technology sector.

Zagreb.

KOLLMANN, FP., KUENZI, EW., STAMM, AJ. 1975. Principles of wood science and technology: II Wood Based Materials. Berlin Heidelberg : Springer-Verlag, 1975. 703 p. IBSN 978-3-642-87931-9.

LYUTYY, P., BEKHTA, P., ORTYNSKA, G. 2018. Lightweight flat pressed wood plastic composites: possibility of manufacture and properties. In Drvna Industrija, 69(1): 5562.

RIZVI, G., SEMERALUL, H. 2008. Glass‐fiber‐reinforced wood/plastic composites. Journal of Vinyl and Additive Technology; 14(1): 3942.

SCHMIDT, H., BENTHIEN, J., THOEMEN, H. 2013. Processing and flexural properties of surface reinforced flat pressed WPC panels. In European Journal of Wood and Wood Products, 2013, 71: 591–597.

ACKNOWLEDGEMENTS This work was supported by the Slovak Research and Development Agency under the contracts No. APVV-14-0506 and APVV-17-0583.

Author’s address Assist. Prof. Pavlo Lyutyy, CSc. Prof. Ing. Pavlo Bekhta, DrSc. Assist. Prof. Galyna Ortynska, CSc. Ukrainian National Forestry University Department of Wood-Based Composites, Cellulose and Paper Zaliznyaka 11 79057 Lviv 109

Ukraine pawa_lyutyj@ukr.net bekhta@ukr.net ortynskag@gmail.com Prof. Ing. Jรกn Sedliaฤ ik, PhD. Technical University in Zvolen Department of Furniture and Wood Products T. G. Masaryka 24 960 53 Zvolen Slovakia sedliacik@tuzvo.sk

110

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 111−119, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.11

CHEMOMETRIC TOOLS USED IN THE PROCESS OF FIRE INVESTIGATION Barbara Falatová  Marta Ferreiro-González  Danica Kačíková  Štefan Galla  Miguel Palma  Carmelo G. Barroso ABSTRACT Fire debris analysis is one of the most challenging steps in fire investigation. Data analysis is subjected to human interpretation by comparing with a reference database. In recent years, chemometric tools have been successfully applied to data in order to avoid misunderstandings and make data interpretation less subjective. In the present study, two ignitable liquids (ILs) (diesel and ethanol) as well as two substrates (cork from Quercus suber bark and cotton sheet) were used. Fire debris samples were prepared following the modified procedure of Destructive Distillation Method for Burning. Sampling was performed in delayed times. In accordance with a progressive approach to data, multivariate statistical analyses, such as unsupervised hierarchical cluster analysis (HCA) and principal component analysis (PCA) as well as supervised linear discriminant analysis (LDA) were applied. The aim of the current research is to investigate whether HS-MS eNose is able to detect different ILs among fire debris samples containing different substrates after delayed times of sampling.

Key words: HS-MS eNose, Chemometrics, Fire Debris Analysis, Total Ion Spectrum.

INTRODUCTION In many criminal activities, such as arsons, ignitable liquids (ILs) are usually used as accelerants. Gasoline and diesel are the most commonly identified accelerants reported by American forensic laboratories since they are easy to obtain (Mach 1977). Samples collected from fire scenes are further analyzed in a laboratory to determine the presence of ignitable liquid residues (ILRs) (STAUFFER et al. 2008). Interpretation of laboratory analysis of fire debris is considered as one of the most complicated analysis among forensic sciences (STAUFFER et al.

2003). Forensic chemists are tasked with target compounds analysis. The presence of particular target compounds characterizes ignitable liquid. The targets that remain are detectable even when IL is evaporated, diluted or contaminated (KETO and WINEMAN 1991; ALMIRALL and FURTON 2004). The determination of ILRs in fire debris might be the difference between fires that were deliberately or unintentionally set (BAERNCOPF et al. 2011).

The American Society for Testing and Materials (ASTM) published standard methods for fire debris analysis and created a classification system for ignitable liquids consisting of eight classes of products. Each class has a precise description (STAUFFER and LENTINI 2003).

Currently, The ASTM E1618 (ASTM 2014) describes the standard method for identification of ILRs in extracts from fire debris by Gas Chromatography-Mass Spectrometry (GC-MS). This standard is the most widely used in the globe. According to BORUSIEVITZ (2004), this 111

methodology is subjected mainly to human interpretation as it is based on the evaluation of total ion chromatogram (TIC) or extracted ion chromatogram (EIC) of the major compounds by visual pattern comparison to a reference database (BORUSIEWICZ et al. 2004). Moreover, this methodology is time-consuming and the interpretation of results becomes more complicated as samples are not neat ILS but collected from fire debris.

Problems related to data interpretation are: weathering, microbiological activity and interfering compounds (BAERNCOPF et al. 2011). In recent years, several studies have alerted about acid alteration of ILs (MARTÍN-ALBERCA et al. 2016). MARTÍN-ALBERCA et al. (2016, 2015b, 2015a) verified acidification and acid modifications on ILs. The authors declared, that the interpretation of the compounds of interests are affected, when gasoline is mixed with concentrated sulphuric acid (STAUFFER et al. 2003). In addition, it is complicated to classify post-burn samples due to the presence of volatile compounds resulting from substrate backgrounds, combustion or pyrolysis products (ALMIRALL and FURTON 2004; FERNANDES et al. 2002).

However, during last years, some improvements on this methodology and new alternatives approaches have been reporting to overcome some of the drawbacks that this methodology presents and to face the new challenges that fire debris examiners have (MARTÍN-ALBERCA et al. 2016).

WADDELL et al. (2014) stated, that one of the strain of fire debris classification from different laboratories may be caused by comparison of chromatographic data, specifically TICs. Total ion spectrum (TIS) provides an alternative approach for data analysis (SIGMAN et al. 2008). The TIS is identical to an average mass spectrum (MS) covering the complete chromatographic range.

Nevertheless, the most of the forensic laboratories perform fire debris analysis by GCMS, several authors have been working on alternative methodologies that can complement or may become an alternative to traditional methods because of advantages they provide. In this sense, FERREIRO-GONZÁLEZ et al. (2014) successfully optimized and validated an electronic nose based on mass spectrometry (HS-MS eNose). The HS-MS eNose was successfully applied for thermal desorption of ILR from carbon strips as an alternative to the use of CS2 solvent (FERREIRO-GONZÁLEZ et al. 2015). The technique was applied to weathered samples (ALIAÑO-GONZÁLEZ et al. 2018) also optimized for analysis of fire debris samples without any pre-treatment (FERREIRO-GONZÁLEZ et al. 2016) and the results were validated by comparison to those obtained by the GC-MS reference method (FERREIROGONZÁLEZ et al. 2017).

The HS-MS eNose is an analytical technique that gives specific responses to an entire aroma in a way similar to humans without previous separation. For this reason, it performs an overall mass spectrum (MS) of an volatile profile characteristic of each sample (PÉREZ PAVÓN et al. 2006).

The aim of the present study is to investigate whether HS-MS eNose together with chemometric tools are able to clearly detect the presence /absence of different ILs from simulated fire debris samples. Two different materials were used as substrates and sampling was performed after delayed times. Total ion mass spectrum (TIS) from burned samples was analyzed by applying chemometric tools by using IBM SPSS Statistics 22 software.

MATERIAL AND METHODS Fire debris samples Two different substrates, cotton sheet and cork (a part of Quercus suber bark) and to different ignitable liquids (IL) were used for investigation in this study. Fire debris preparation 112

followed the modified procedure of Destructive Distillation Method for Burning (WILLIAMS et al. 2012). One piece of a substrate (5 × 5 cm) was replaced by six small pieces (1 × 4 cm) and placed on the bottom of a metal can (Fig. 1). 0. 5 mL of ignitable liquid were applied onto surfaces. Respectively, diesel and ethanol were used. Subsequently, the can with vented lit was placed above a propane torch. When a smoke appeared the samples were burned for approximately two additional minutes. The can was then removed from the flame and allowed to cool down. After a cooling time of approximately three minutes one of the two fire suppression agents was applied by spraying it onto the carpet surface and the sample was covered and re-lit. The sampling of prepared fire debris was performed in delayed times – 10 minutes, 1 and 6 hours. The samples were labelled by following codes: fire debris (FD), cotton sheet (CS), cork (CO) and delayed time (0h, 1h, 6h). All the combinations of variations were used for fire debris preparation. Burned samples were denoted by the substrate code followed by a liquid code: diesel (DIE) and ethanol (ETH). For instance: for burned cotton sheet substrate without IL when sampling was performed 1 hour after burning the code was FD – CS _1h, for burned cork with diesel when sampling was performed 6 hours after burning, the code was FD-CO+DIE_6h. The simulated fire debris were placed directly into vials and analyzed by HS – MS eNose. a) Side view

b) Above view

Vented lit Metal can

10

Bottom of metal can

40

Samples

Propane torch

Fig. 1 Schematics of side (a) and above (b) view of the experimental setup. All measurements are in mm.

HS-MS eNose spectra acquisition Analysis of the samples was performed on an Alpha Moss (Toulouse, France) electronic nose based on mass detector system composed of an HS 100 static headspace autosampler and a Kronos quadrupole mass spectrometer. Nitrogen was used as a carried gas. Samples in 10 mL sealed vials (Agilent Crosslab, Santa Clara, CA, USA) were placed in the autosampler oven and heated. Headspace was taken from the vial by a gas syringe. To avoid condensation, the syringe was heated above the sample temperature (+5 °C) and consequently injected into the mass spectrometer. Between each sample injection, the gas syringe was flushed with nitrogen to avoid cross-contamination. The experimental conditions used for the headspace sampler were optimized in another study (FERREIROGONZÁLEZ et al. 2016) and consisted in the following conditions: incubation temperature 115 °C, incubation time 10 min, agitation speed 500 rpm, syringe type 5 mL, syringe temperature 125 °C, flushing time 120 s, fill speed 100 μL/s, injection volume 4.5 mL and injection speed 75 μL/s. The total time per sample was 15 min. The components in the headspace of the vials were passed directly to the mass detector without any chromatographic separation or sample pre-treatment. In this way, the resulting total ion mass spectrum (TIS) gives a fingerprint of the sample. The electron ionization spectra were recorded in the range of 45–200 mass-to-charge ratios (m/z). The instrument control was

113

achieved using Residual Gas Analysis software and Alpha Soft 7.01 software package (Alpha Moss, Toulouse, France).

Data analysis Total ion mass spectrum (TIS) from burned samples was arrange in a data matrix named D m x n, where m is the number of m/z intensities in the range of 45200 and n is the number of fire debris samples. The intensities of each m/z were taken as independent variables, standardized by feature scaling.

Multivariate statistical analyses represented by HCA, PCA and LDA were performed by using IBM SPSS Statistics 22 software (Armonk, NY, USA).

RESULTS AND DISCUSSION First of all, the exploratory multivariate analysis technique HCA was applied with the aim to investigate natural clustering among fire debris samples. HCA was carried out by using all the m/z intensities as independent variables for forming clusters (D 36 x 156). Ward’s hierarchical agglomerative clustering method and squared Euclidean distances were employed for the HCA. The dendrogram resulting from the HCA is represented in Fig. 2.

As can be clearly seen form Fig. 2, fire debris samples are divided into two major clusters. The major cluster A contains 100 % of samples burned with ILs. This cluster is widely distributed into two subclusters A1 and A2 that fully discriminate according to the type of used IL. Based on the results, fire debris samples are clustered due to their chemical composition – alcohols and diesel. The alcohol subcluster (A1) consists of samples solely containing ethanol. The subcluster A2 is formed by fire debris samples with diesel. Based on the presence of any type of IL, ethanol and diesel are joint in the same cluster at shorter distance than the cluster B that includes burned samples without any IL. The second major cluster B consists of fire debris samples without ILs. The cluster further discriminates between substrates used in fire debris samples. While the subcluster B1 contains solely samples with cork, the subcluster B2 is formed by fire debris samples with cotton sheet. A tendency related to the sampling time was not observed.

Fig. 2 Dendrogram obtained from the HCA for all the fire debris samples using the signal from the HSMS eNose (45200 m/z).

114

In order to further investigate the tendencies proposed by HCA - particular ILs and distinguishing them, further analyses were required. The unsupervised pattern recognition PCA method was employed to the whole data set D36 x 156 of fire debris samples. Nine principal components (PCs) were extracted with the eigenvalues greater than 1.0. The most of the variability in the data is explained by first four PCs (92.22 % of the accumulative variance). Thus, the data matrix has effectively been reduced to 2 dimensions, while still remaining 86.35 % of the information. Fig. 3 displays score scatter plot of first two PCs. The following general observations can be made from the visual inspection:

Fig. 3 Scores for the fire debris samples (n=36) in the two dimensional PC1-PC2 space.

Plotting PC1 versus PC2 allows seeing a grouping of samples due to the presence of any of the ILs. Whereas PC1 refers to the overall results and separate samples in accordance with the presence/absence of ILs, PC2 is related to the type of IL Almost all the m/z show high influence on the PC1 with positive values, except m/z 55 and m/z 57, that show loadings with low negative values, respectively -.080 and -.0.95.

PC2 allows the full discrimination regarding the ILs (diesel or ethanol). M/z with the highest influence in PC2 with positive values above .08 are the following: m/z 55, m/z 56, m/z 57, m/z 69 and m/z 71. The high negative value is represented by m/z 45 (-.88). These results suggest that the data from the HS-MS eNose are mainly related to those compounds responsible first for the discrimination of the fire debris due to the presence/absence of ILs and second to the type of used IL. Based on the tendency showed in the unsupervised techniques, it was preceded with the application of a supervised technique named LDA to identify whether there are specific 115

m/z values in the TIS that are more significant than the others when discriminating samples according to the type of IL, a supervised chemometric technique LDA was applied to the data set D36 x 156. Three groups (samples without ILs, samples burned with ethanol and samples burned with diesel) were established a priori. Stepwise method was applied. Based on LDA results, two canonical functions were obtained that explain, sequentially, 83.90 % and 16.10 % of the variance. The full discrimination were obtained (100 % of cross-validated grouped cases were correctly classified). Groups of predicted variables will make predictions that are statistically significant (p= .00) in their accuracy. The m/z values selected for classification in the accordance of discriminant functions were: m/z 45, m/z 60, m/z 70, m/z 93, m/z 124, m/z 139, m/z 143, m/z 155, m/z 162, m/z 166, m/z 185 and m/z 189.

45 189

1

60

0,8 0,6

185

70

0,4

WITHOUT IL

0,2

166

93

0

DIESEL ETHANOL

162

124 155

139 143

Fig. 4 Fingerprinting of the burned samples with and without the presence of ILs.

When only intensity values of the signals (m/z) selected by the LDA for developing the Fisher's linear discriminant functions are displayed, a different fingerprint for burned samples with and without ILs is obtained (Fig. 4). All the m/z values were normalized to the base peak at 100%.

As the Fig. 4 displays, a characteristic fingerprint for each type of IL was also obtained.

Fire debris samples without any of IL do not represent any characteristic m/z. Samples burned with diesel present only one characteristic m/z of the highest intensity, which is found in samples with both substrates. M/z 70 that refers to alkanes that are abundantly presented in petroleum distillate products such as diesel. Samples burned with ethanol are related to only m/z 45 with the highest value that refers to alcohols.

CONCLUSION The presented study focuses on identification ILs from simulated fire debris samples.

A progressive approach to data analysis is represented by applying multivariate statistical analysis. Based on results, chemometric tools successfully discriminate samples regarding the presence of any ILs used. According to tendencies obtained from results from HCA, PCA as unsupervised pattern recognition was further applied. Based on the results, it is suggested that data obtained from HS-MS eNose are firstly related to compounds that discriminate fire debris samples due to the presence / absence of the ILs. In addition, results from supervised LDA clearly demonstrate different fingerprints among simulated fire debris samples that 116

might be used on purpose to preliminary determine the presence of the ILs. To sum up, the results obtained from this study support the idea of previous studies that HS-MS eNose is able to detect and identify particular ILs from fire debris samples and might be used as complementary analytical method to reference method broadly used in forensic laboratories.

The useful results could be also obtained in the case of other natural substrates on the base of wood, not only for cork. Further research should be performed to deal with other aspects of HS-MS eNose to be considered as an alternative method to fire debris analysis.

REFERENCES ALIAÑO-GONZÁLEZ, MARÍA, MARTA FERREIRO-GONZÁLEZ, GERARDO BARBERO, JESÚS AYUSO, MIGUEL PALMA, CARMELO BARROSO 2018. Study of the Weathering Process of Gasoline by eNose.

In Sensors 18(1):139. https://doi.org/10.3390/s18010139.

ALMIRALL, JOSÉ R., KENNETH G. FURTON 2004. Characterization of Background and Pyrolysis Products That May Interfere with the Forensic Analysis of Fire Debris. In Journal of Analytical and Applied Pyrolysis, 71(1): 51–67. https://doi.org/10.1016/S0165-2370(03)00098-6.

ASTM 2014. ASTM E1618 - 14 Standard Test Method for Ignitable Liquid Residues in Extracts from Fire Debris Samples by Gas Chromatography–Mass Spectrometry, issued 2014. www.astm.org.

BAERNCOPF, JAMIE M., VICTORIA L. MCGUFFIN, RUTH W. SMITH 2011. Association of Ignitable Liquid Residues to Neat Ignitable Liquids in the Presence of Matrix Interferences Using Chemometric Procedures. In Journal of Forensic Sciences, 56 (1): 70–81. https://doi.org/ 10.1111/j.1556-4029.2010.01563.x.

BoruSIEWICZ, RAFAŁ, ZADORA GRZEGORZ, JANINA ZIĘBA-PALUS 2004. Application of Head-Space Analysis with Passive Adsorption for Forensic Purposes in the Automated Thermal Desorption-Gas Chromatography-Mass Spectrometry System. In Chromatographia, 60(S1): 133–42. https://doi.org/ 10.1365/s10337-004-0299-4.

FERNANDES, M. S., C. M. LAU, W. C. WONG 2002. The Effect of Volatile Residues in Burnt Household Items on the Detection of Fire Accelerants. In Science and Justice - Journal of the Forensic Science Society, 42(1): 7–15. https://doi.org/10.1016/S1355-0306(02)71791-7.

FERREIRO-GONZÁLEZ, MARTA, JESÚS AYUSO, JOSÉ A. ÁLVAREZ, MIGUEL PALMA, CARMELO G. BARROSO 2015. Application of an HS-MS for the Detection of Ignitable Liquids from Fire Debris.

In Talanta 142. Elsevier:150–56. https://doi.org/10.1016/j.talanta.2015.04.030.

FERREIRO-GONZÁLEZ, MARTA, GERARDO F. BARBERO, JESÚS AYUSO, JOSÉ A. ÁLVAREZ, MIGUEL PALMA, CARMELO G. BARROSO 2017. Validation of an HS-MS Method for Direct Determination and Classification of Ignitable Liquids. In Microchemical Journal 132. Elsevier B.V.: 358–364. https://doi.org/10.1016/j.microc.2017.02.022.

FERREIRO-GONZÁLEZ, MARTA, GERARDO F. BARBERO, MIGUEL PALMA, JESÚS AYUSO, JOSÉ A. ÁLVAREZ, CARMELO G. BARROSO 2016. Determination of Ignitable Liquids in Fire Debris: Direct Analysis by Electronic Nose. In Sensors (Switzerland) 16 (5). https://doi.org/10.3390/s16050695.

FERREIRO-GONZÁLEZ M., AYUSO, J., ÁLVAREZ, J. A., PALMA, M., BARROSO, C. G. 2014. New Headspace-Mass Spectrometry Method for the Discrimination of Commercial Gasoline Samples with Different Research Octane Numbers. In Energy & Fuels, 28(10): 6249–6254. https://doi.org/10.1021/ef5013775.

KETO, RAYMOND O., PHILIP L. WINEMAN 1991. Detection of Petroleum-Based Accelerants in Fire Debris by Target Compound Gas Chromatography/Mass Spectrometry. In Analytical Chemistry, 63(18): 1964–1971. https://doi.org/10.1021/ac00018a013.

MACH, M. 1977. Gas Chromatography-Mass Spectrometry of Simulated Arson Residue Using Gasoline as an Accelerant. In Journal of Forensic Sciences, 22: 348–357. https://doi.org/ 10.1520/JFS10596J. MARTÍN-ALBERCA, CARLOS, CARMEN GARCÍA-RUIZ, OLIVIER DELÉMONT 2015a. Study of Acidified Ignitable Liquid Residues in Fire Debris by Solid-Phase Microextraction with Gas Chromatography and Mass Spectrometry. In Journal of Separation Science, 38(18): 3218–3227.

117

https://doi.org/10.1002/jssc.201500337.

MARTÍN-ALBERCA, CARLOS, CARMEN GARCÍA-RUIZ, OLIVIER DELÉMONT 2015b. Study of Chemical Modifications in Acidified Ignitable Liquids Analysed by GC-MS. In Science and Justice, 55(6): 446–55. https://doi.org/10.1016/j.scijus.2015.06.006.

MARTÍN-ALBERCA, CARLOS, FERNANDO ERNESTO ORTEGA OJEDA, CARMEN GARCÍA-RUIZ 2016.

Study of Spectral Modifications in Acidified Ignitable Liquids by Attenuated Total Reflection Fourier Transform Infrared Spectroscopy. In Applied Spectroscopy, 7(3): 520–530. https: //doi.org/10.1177/0003702815626681.

MARTÍN-ALBERCA, CARLOS, FERNANDO ERNESTO ORTEGA-OJEDA, CARMEN GARCÍA-RUIZ 2016.

Analytical Tools for the Analysis of Fire Debris. A Review: 2008–2015. In Analytica Chimica Acta, 928. Elsevier Ltd:1–19. https://doi.org/10.1016/j.aca.2016.04.056.

PÉREZ PAVÓN, JOSÉ LUIS, MIGUEL DEL NOGAL SÁNCHEZ, CARMELO GARCÍA PINTO, MA ESTHER FERNÁNDEZ LAESPADA, BERNARDO MORENO CORDERO, ARMANDO GUERRERO PEÑA. 2006.

Strategies for Qualitative and Quantitative Analyses with Mass Spectrometry-Based Electronic Noses. In TrAC - Trends in Analytical Chemistry, 25(3): 257–266. https://doi.org/ 10.1016/j.trac.2005.09.003.

SIGMAN, M.E., WILLIAMS, M.R., CASTELBUONO, J.A., COLCA, J.G., CLARK, C.D. 2008. Ignitable Liquid Classification and Identification Using the Summed-Ion Mass Spectrum. In Instrum. Sci. Technol. https://doi.org/36:375–393.

STAUFFER, E. et al. 2008. Fire Debris Analysis. Boston : Academic Press.

STAUFFER, E. 2003. Concept of Pyrolysis for Fire Debris Analysts. In Science and Justice - Journal of the Forensic Science Society, 43(1): 29–40. https://doi.org/10.1016/S1355-0306(03)71738-9.

STAUFFER, ÉRIC, JOHN J. LENTINI 2003. ASTM Standards for Fire Debris Analysis: A Review. In Forensic Science International, 132(1): 63–67. https://doi.org/10.1016/S0379-0738(02)00459-0.

WADDELL, ERIN E., MARY R. WILLIAMS, MICHAEL E. SIGMAN 2014. Progress toward the Determination of Correct Classification Rates in Fire Debris Analysis II: Utilizing Soft Independent Modeling of Class Analogy (SIMCA). In Journal of Forensic Sciences, 59(4): 927–35. https://doi.org/10.1111/1556-4029.12417.

WILLIAMS, MARY R., MICHAEL E. SIGMAN, JENNIFER LEWIS, KELLY MCHUGH PITAN 2012.

Combined Target Factor Analysis and Bayesian Soft-Classification of Interference-Contaminated Samples: Forensic Fire Debris Analysis. In Forensic Science International, 222 (1–3). Elsevier Ireland Ltd: 373–86. https://doi.org/10.1016/j.forsciint.2012.07.021.

ACKNOWLEDGEMENT This work was supported by the Slovak Research and Development Agency under the contract no.

APVV-17-0005 (50 %). This work was supported by the Slovak Research and Development Agency under the contract No. APVV-0057-12 (50 %).

AUTHORS ADDRESS Barbara Falatová Danica Kačíková Technical university in Zvolen Faculty of Wood Sciences and Technology Department of Fire Protection T. G. Masaryka 2117/24 960 53 Zvolen Slovak Republic barbara.falatova@gmail.com danica.kacikova@tuzvo.sk

118

Marta Ferreiro-González Miguel Palma Carmelo G. Barroso Department of Analytical Chemistry Faculty of Sciences University of Cadiz Agrifood Campus of International Excellence (ceiA3) IVAGRO, P.O. Box 40 11510 Puerto Real Cadiz Spain marta.ferreiro@uca.es miguel.palma@uca.es carmelo.garcia@uca.es Štefan Galla Fire Research Institute of the Ministry of Interior Rožňavská 11 831 04 Bratislava Slovak Republic stefan.galla@gmail.com

119

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 121−129, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.12

STUDIES OF COMPONENT INTERCONNECTION IN A PLYWOOD STRUCTURE WITH INTERNAL LAYERS OF VENEER CHIPS Štefan Barcík  Sergey Ugryumov  Evgeny Razumov  Ruslan Safin ABSTRACT The impact of the assembly diagrams of plywood package with internal layers of veneer chips on the performance of component interconnection, i.e. surface soundness and interfacial interaction in response to equal cleavage and chipping is presented in the paper.

Rational design of plywood package assembly providing the required physical and mechanical properties is described as well. The fact that cohesive destruction of wood particles occurs regardless of the assembly scheme when veneer layers are torn off and when plywood is cut along the boundary layers of plywood with internal layers on the basis of wood particles was found. Soundness of the joint at the interface between the layers of veneer and wood particles is due to the presence of an adhesive layer with an increased glue content.

The maximum pull-off strength is observed in plywood samples with a central veneer layer.

Therefore its use in critical structures is recommended. This type of plywood can be used effectively as a structural material in various fields.

Keywords: plywood, veneer sheets, veneer chips, bond quality, surface soundness, interfacial interaction

INTRODUCTION The refuse wood is an inevitable part of the plywood production. Inclusion of wood processing waste into the production of wood composite materials helps reduce the cost of the end products and utilize them more effectively (STRELKOVA & NOVIKOVA 1993; UGRUMOV & SMIRNOV 2006).

It is highly recommended to use glued wood particles of milled refuse, accompanying plywood production, for the manufacture of the inner layers of plywood. The basis of the strength of such a material is a peeled veneer, and a composition based on wood particles, mixed with a synthetic binder, serves as a filler (UGRUMOV et al. 2007; MALISHEVA 2013).

Plywood with inner filling of veneer chips can effectively dispose the generated refuse wood, reduce production costs, and expand the range of products, while maintaining their quality and competitiveness, which is relevant for the woodworking industry.

It is important to predict strength properties of the plywood, and the degree of interfacial interaction between the veneer sheets and an inner filling, because it allows to adjust the assembly diagram and the composition depending upon the desired properties and fields of application.

Any system of adhesive (liquid) - the substrate (solid) can be characterized by the adhesion and fracture mode (disbond mode among the components). Knowing the 121

weaknesses of the material, can make it easier to improve its efficiency and durability (BERLIN 1990; UGRUMOV & SVESHNIKOV 2010). The interfacial interaction in plywood with internal layers of veneer chips can be better observed at phase boundaries: liquid (binder) solid (wood particles); liquid (binder) - solid (peeled veneer).

We will consider as the structure of the plywood package with internal layers of veneer chips the number of layers of veneer of a certain thickness, alternating with the internal layers of the glued wood particles procured from milled refuse of plywood production. The paper discusses the efficient structures of plywood package filled with wood particles and their physical and mechanical properties.

Another issue is the estimation of the value of the surface soundness and interfacial interactions at phase boundaries (layers of peeled veneer sheets and veneer chips) in a plywood structure.

MATERIALS AND METHODS The objective of the paper is to estimate the value of the surface soundness and interfacial interactions at the boundaries of peeled veneer sheets and veneer chips composition depending on the assembly diagram of the plywood package and recommend effective schemes for assembling a plywood package.

Issues to be solved are as follows: to estimate the interfacial interaction and surface soundness of veneer chips depending on the package structure, and to recommend the efficient structure of plywood package.

Plywood samples of the format 400x400 mm were produced according to various schemes of assembly in a hydraulic laboratory press P100-400. During the experimental assembly the urea-formaldehyde resin adhesive was applied; birch peeled veneer with 1.5 mm thickness was used in the outer and central layers, and specially sorted birch chips were used in the internal filling (the chips passed through a 10 mm-hole diameter sieve and remained on a 5 mm-hole diameter sieve); the production process was carried out at following constant factors: - plywood board thickness, mm â&#x20AC;&#x201C; 16; - temperature of press - 130Ë&#x161;C; - specific pressing pressure - 2 MPa; - exposure time under a pressure - 16 min.

The mass of wood particles in the formation of internal layers was calculated in such a way that, after pressing, the density of plywood samples was 700 kg/m3.

The production of experimental samples was carried out according to a one-stage scheme. In accordance with the assembly scheme, samples consisting of the necessary number of layers of veneer and tarred wood particles were formed, further they were pressed in a cold press to increase transportability, and then pressed in a hot press with 16 mm thick stoppers.

Ready-assembled plywood samples were conditioned for 1 day, and then were cut to the appropriate samples for testing.

The method of estimating the value of the surface soundness in the cleavage of peeled veneer sheets according to GOST (State Standards of Russia) 27325 and the method of determining the strength of chipping in accordance with GOST 9624 were basic in the research.

The summary of the method of evaluating the surface soundness in the cleavage layer is a separation of an area of face layer (sheets of peeled veneer) from the base coat (the inner layers of the wood-adhesive composition) in a direction perpendicular to the latter, 122

depending on the assembly design of the plywood package retaining the fracture load.

50 Ă&#x2014; 50 mm plywood samples were used to determine surface soundness. The surface of the birch cylinder was smoothly covered with epoxy adhesive. Further, the cylinder was glued perpendicular to the horizontal plane on the center of the sample, and exposed it for 24 hours under normal conditions. Upon exposure, the surface of the coatingwas drilled to the base coat around the cylinder, allowing the emergence of the drilling footprint on the inner layer of wood-adhesive composition. The test piece of plywood with internal layers of wood-adhesive composition is shown in Fig. 1.

Fig. 1 The image of the sample to determine the surface soundness.

The tearing machine P-5 was used for the tests. The test samples were mounted in a test fixture. The test device was attached to the upper jaw of the machine through a cylinder according to the scheme shown in Fig. 2.

Fig. 2 Scheme of the sample mounting in the tearing machine: 1 - sample capture; 2 - sample; 3 - birch cylinder.

123

The tests were performed for the outer and central peeled veneer sheets to determine the effect of the structure of the assembly package on the value of the adhesive strength and interfacial interaction.

Figure 3 shows the cylinder installation diagrams when testing the surface soundness of outer and central peeled veneer sheets in the package, depending on the design of the assembly.

a)

b)

c)

d)

e)

f) Fig. 3 Cylinder installation diagrams during the test: a) assembly diagram No 1 (outer veneer sheets); b) assembly diagram No 2 (outer veneer sheets); c) assembly diagram No 2 (central veneer sheets); d) assembly diagram No 3 (outer veneer sheets); e) assembly diagram No 3 (central veneer sheets); f) assembly diagram No 4 (outer veneer sheets). 1 - a layer of epoxy glue; 2 - birch cylinder; 3 - outer veneer sheets; 4 â&#x20AC;&#x201C; the central veneer sheet.

124

The ultimate split strength and fracture pattern were measured by the results of the tests.

Samples shown in Figure 4 were manufactured to determine the strength of chipping along the phase boundaries.

a)

b)

b)

d)

e) Fig. 4 The form of the samples to determine the chipping strength: a) assembly diagram No1a (upper veneer sheets); b) assembly diagram No 1b (bottom veneer sheets); c) assembly diagram No 2; d) assembly diagram No 3; e) assembly diagram No 4.

The length of the sample was 85 mm, and the thickness - 16 mm. The cutting width was determined depending on the thickness of the saw cut and ranged from 8 to 15 mm to ensure whole capture by the test device. The cutting depth h was determined according to the package structure. The length of the chipping plane 1 was (12.5 Âą 0.5) mm.

The test machine P-5, equipped with wedge grips, was used to determine wood ultimate split strength at the phase boundary. The fracture load was recorded for each sample, further the ultimate split strength and the nature of the fractures were determined.

125

RESULTS In most cases, the fractures can be classified as follows: adhesive (adhesive is entirely separated from the substrate), cohesion (failure occurs on the adhesive layer or on the base coat), and mixed (adhesive-cohesive, in which there is a partial separation of the adhesive (adhesive layer) from the substrate (base coat) or partial fracture of the substrate and the partial fracture of the adhesive).

The adhesion strength for the main plywood assembly diagrams has been determined experimentally in response to equal cleavage of outer and central veneer sheets.

Figure 5 shows the dominant character of the samples fracture upon the completion of the cleavage test on veneer sheets.

Fig. 5 . The typical plywood sample fracture in response to cleavage.

Table 1 represents the results of experimental studies to determine the adhesive strength of plywood with internal layers of wood-adhesive composition in response to cleavage of veneer sheets.

Average value

Standard deviation

Coefficient of variation, %

1.59

0.032

1.87

-

-

-

2

1.63

0.051

3.35

1.25

0.033

2.40

3

1.67

0.050

3.23

1.23

0.015

0.81

4

1.65

0.112

6.82

-

-

-

126

cohesively on the inner layer of veneer chips mixed type (veneer / wood particles) cohesively on the inner layer of veneer chips mixed type (veneer / wood particles)

Central ply sheet

Coefficient of variation, %

1

Type of fracture of the sample during the test

outer ply

Standard deviation

Soundness of plywood when tearing off veneer sheets from internal layers, MPa outer veneer sheets central veneer sheets Average value

No of the assembly diagram

Tab. 1 Soundness of plywood in response to tearing off veneer sheets from internal layers.

mixed type (veneer / wood particles) mixed type (veneer / wood particles) -

The obtained data show that when veneer sheets are removed, a cohesive destruction along veneer chips is observed, or a mixed character of destruction, that is, simultaneous destruction of veneer and wood particles, which indicates a high interfacial bond strength.

The nature of the interfacial interaction of liquid (binder) with a solid body (wood particles and veneer) at the boundary of their interaction, depending on the plywood assembly diagram, was determined by cleaving along the boundary of the contact of the outer veneer sheet and the inner layer of veneer chips, as well as the inner layer of veneer chips and the central veneer sheet.

For example, Figure 6 shows some typical samples at chipping fractures depending on the package structure.

Table 2 represents the results of experimental studies to determine the strength of plywood with internal layers of veneer chips in response to chipping.

a)

b)

c) Fig. 6 The typical chipping fractures of the samples: a) assembly diagram No 1a; b) assembly diagram No 3; c) assembly digram No 4.

Tab. 2 The strength of plywood in response to chipping on the adhesive layer.

Average value

Standard deviation

Coefficient of variation

No of the assembly diagram

Chipping strength in layers, MPa

1Đ°

3.05

0.331

10.86

1b

3.1

0.192

6.20

2 3 4

4.21 4.12 3.15

0.162 0.172 0.158

3.86 4.18 5.02

The main type of the fracture of the samples during the test

Cohesively on the veneer chips Mixed on glue line between the layers of veneer with tear-outs of wood fibers from the surface of the veneer Cohesively the veneer chips Cohesively on the veneer chips Cohesively on the veneer chips

127

Thus, mainly cohesive on veneer chips or mixed nature of fracture was observed, which indicates a high interfacial strength of bonding.

DISCUSSION We can mainly observe cohesive fracture on the veneer chips, regardless of the assembly diagram, in the process of the adhesive cleaving of veneer layers and chipping at the boundary layers of plywood with internal layers of venner chips. High adhesion strength at the phase interface (veneer layers and veneer chips) can be explained by the presence of an adhesive layer with a high content of glue, which contacts with pitched veneer sheet and glued wood particles.

Slight differences in the values of surface soundness at the phase boundaries, depending on their location, can be explained by the difference in the degree of cure of the binder along the cut of plywood - during the pressing the temperature of the central layers is less than of the outer ones, therefore, the completeness of binder cure in the inner layers is smaller than in the outer layers.

The maximum split strength was observed in the samples from the plywood with assembly diagrams No 2, and No 3, with the inner veneer layer, so these were recommended for the use in critical structures. The assembly diagrams No 1, and No 4 are regarded as simple from the perspective of the process of package formation.

The obtained results of the estimation of the physical and mechanical properties of plywood with internal layers of venner chips exceed the properties of analog materials. Thus, the obtained strength in the separation of veneer sheets from internal layers of wood chips, ranging from 1.23 to 1.67 MPa, depending on the assembly scheme, significantly exceeds the standardized and statistical values for analogue materials – chipboards and oriented stand boards (OSB) [ 8], and the strength at shearing along the layers, ranging from 3.05 to 4.21 MPa, exceeds the normalized and statistical values for general-purpose plywood [9], which indicates a high bond strength and a high reliability prediction of the material.

CONCLUSION High level of surface soundness in response to cleavage of veneer sheets and deep interfacial interactions at the phase boundaries between layers of veneer and layers of woodadhesive composition let us conclude that plywood with internal layers of veneer chips is reliable and possesses good capacities, and that it can be effectively used as a structural material in various fields.

LITERATURE BERLIN, А. А. 1990. Printsipy sozdaniya kompozitsionnykh polimernykh materialov. Moskva : Khimiya Publ.

MALISHEVA, G.V. 2013. Prognozirovanie resursa kleevikh soedineniy. In Klei. Germetiki.

Tekhnologii. 8. pp. 3134.

SEDLIAČIK, J., BEKHTA, P., POTAPOVA, O. Technology of low-temperature production of plywood bonded with modified phenol-formaldehyde resin. In Wood Research. 2010. 55(4): 123130.

STRELKOVA, V.P., NOVIKOVA, О.М. 1993. Linii maloy moshchnosti dlya proizvodstva plit i drugikh pressovannykh izdeliy iz drevesnykh i selskokhozyaystvennykh. In Derevoobrabatyvayushchaya promyshlennost. 6: 21–22.

128

TREML, S., TRÖGER, F. 2011. On the manufacture of oriented strand board (OSB) from original durable wood species for use outdoors freely exposed to weathering. In Holz als Roh-und Werkstoff.

2011. 69(1): 163165.

UGRUMOV, S.А., BOROVKOV, Е. А., SHCHERBAKOV, А. Е. 2007. Razrabotka tekhnologicheskoy posledovatelnosti proizvodstva kompozitsionnoy fanery. In Lesnoy Vestnik: Vestnik MGUL. 6. pp.

120–123.

UGRUMOV, S.А., SMIRNOV, А.А. 2006. Organizatsiya tekhnologicheskogo protsessa proizvodstva kompozitsionnoy fanery. In Lesnoy Vestnik: Vestnik MGUL. 3: 123126.

UGRUMOV, S.А., SVESHNIKOV A.S. 2010. Kompleksnoe issledovanie svoystv kompozitsionnoy fanery. In Lesnoy Vestnik: Vestnik MGUL. 6. pp. 163165.

ACKNOWLEDGMENT The paper was written within the project: VEGA 1/0315/17 “Research of relevant properties of thermally modified wood at a contact effects in the machining process with the prediction of obtaining an optimal surface”.

AUTHORS’ ADDRESSES Štefan Barcík Technical University in Zvolen Faculty of Environmental and Manufacturing Technology Department of Environmental and Forestry Machinery Department of Machinery Control and Automation Slovak Republic Sergey Ugryumov St. Petersburg State Forest Engineering University Faculty of Technological Machines and Timber Transport Department of Technological Processes and Machines of the Forest complex St. Petersburg Russian Federation Evgeny Razumov Czech University of Life Sciences Faculty of Forestry and Wood Sciences, Czech Republic, Prague.

Ruslan Safin Kazan National Research Technological University Faculty of Power Engineering and Technological Equipment Department of Architecture and Design of Forest products Kazan Republic of Tatarstan Russian Federation

129

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 131−145, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.13

EXPLORING THE APPLICATION OF NATURE-INSPIRED GEOMETRIC PRINCIPLES WHEN DESIGNING FURNITURE AND INTERIOR EQUIPMENT

Denisa Lizoňová  Zuzana Tončíková ABSTRACT Since ancient time the nature has been the essence of human existence, and people were trying to look for inspiration and advice for solving various problems in the nature.

Most natural objects contain basic geometric principles creating a sort of systematic character and order in them. Natural structures and shapes represent a perfect design and architectural suggestions refined over the course of 3.8 billion of years of evolution. These unique and functional shape and space solutions create an ingenious system providing, through detailed study and transformation, a creative way of looking for inspiration for making innovative products with high added value in the form of nature being the mentor.

The aim of the paper is to show how the biomimicry principles can be used and applied for the creation of new objects. Theoretical basis, the methodology and outputs of the workshops “Discovering Geometric Principles in the Nature as the Source of Inspiration” organized under the guidance of the authors for the students of the study program Design at the Technical University in Zvolen within the selective course of Geometric Composition is presented in the paper. The aim of the workshop was to apply the geometric principles from the selected natural shapes into the design of a unique solitaire.

Key words: design, geometry, natural aesthetics, biomimicry.

INTRODUCTION Nature and the world surrounding us symbolize an unlimited source of ideas and inspiration for creative activities. Although the forms of natural elements are various at first sight, most of them have certain features and principles in common, and these can be observed and defined in more detail. The precise natural shapes have developed and changed throughout several millennia and include the essence that is natural for human and evoke pleasant feelings. Selected geometric procedures as well as mathematical models can represent a tool for more exact expression and description of such natural forms.

People have been observing and describing the surrounding world and searching for principles and rules since time immemorial. Already early Greek philosophers Plato and Pythagoras studied pattern attempting to explain order in nature. Plato (c.427-347 BC) in his book Timaeus describes the geometric creation of the world and he presents there his idea that the Creator created the visible world similar to a geometric progression. The

131

Platonic solids – Tetrahedron, Cube, Octahedron, Dodecahedron, Icosahedron – make up the four elements and heaven.

In the 19th century, Belgian physicist Joseph Plateau examined soap films, leading him to formulate the concept of a minimal surface. Plateau’s original problem dealt with surfaces of minimum area without pressure differences (so H = 0) and with boundaries that are simple closed curves.

German biologist Ernst Haeckel played an important role in connecting the sciences and art. He discovered, described and named thousands of new species, mapped a genealogical tree relating all life forms, and coined many terms in biology. As an artist he painted hundreds of organisms to emphasize their symmetry. Most notable is his book Kunstformen der Natur (Art Forms in Nature). Haeckel was outstanding as a scientific artist.

Instead of drawing just a front view, he also illustrated the other side if visible through gaps and holes in the skeletons. The result was a three-dimensional picture – rarely seen until then.

Haeckel’s Art Forms in Nature is not merely a set of examples, which with each detail reveals part of the whole. It demonstrates naturalness itself. By revealing the form of nature, knowledge of nature may be ascertained, which, according to Haeckel, should not be restricted to branches of natural science following experimental agendas. Knowledge of nature is “natural aesthetics”. Accordingly, aesthetics is nothing more than reflections of nature itself. Nature, which develops out of and into itself, is beautiful.

Haeckel’s illustrations inspired many artists of the time. One of them was the Paris architect René Binet. He used natural form of a radiolarian as the basis for his monumental entrance gate to the Paris World Exposition in1900.

1a

1b

Fig. 1a René Binet’s entrance gate to the Paris World Expositionin1900 (http://architectuul.com/ architecture/the-monumental-gate), 1b. Ernst Haeckel’s illustration of radiolarian (HAECKEL 1998).

Cooperation between the Hungarian biologist Aristid Lindenmayer and French American mathematician Benoît Mandelbrot was another significant contribution to the research into nature and its mathematical and geometrical interpretation. In cooperation they showed how the mathematics of fractals could create plant growth patterns. Mandelbrot believed that fractals, far from being unnatural, were in many ways more intuitive and natural than the artificially smooth objects of traditional Euclidean geometry. “Clouds are not spheres, mountains are not cones, coastlines are not circles, and bark is not smooth, nor does lightning travel in a straight line” (MANDELBROT 1982).

132

Natural forms, characterized by specific functional and shape solutions, offer the opportunity to take advantage of their uniqueness. Their exact exploration and subsequent transformation can be a unique inspiration for the design of new innovative products.

The aim of the paper is to show how to apply the principles of biomimicry in creating new objects. Research of the chosen problem related to the application of bio-inspired approach in design of furniture and interior equipment was based on the initial observation of selected natural shapes. Then the selected objects were abstracted using geometric knowledge and constructions. These constructions have been thoroughly analyzed and have become the basis for product development.

Within the education in the field of study 2.2.6 Design at the Technical University in Zvolen, the students have the possibility of enrolling into the course Geometric composition.

This paper presents the methods and outputs of a workshop, whose aim was to introduce the students to the issue of geometric principles in the nature, their rules, construction possibilities and use. Further aim was to apply the acquired knowledge and design an object – a unique solitaire inspired by the forms, patterns and geometrical relations derived from natural principles. The methodical process was based on Biomimicry procedures (BENYUS 1997) representing the most complex designer processes for creating bio-inspired innovations.

Beauty is a large part of why biomimicry resonates. Our search for mentors brings us back into contact with the living world, a place we were tuned to appreciate. Having spent 99.9% of our planetary tenure woven deep into the wild, we humans naturally admire the weaverbird’s nest, the conch’s shell, the scales of a shimmering trout. In fact, there are few things more beautiful to the human soul than good design (BAUMEISTER 2013).

The core idea is that nature has already solved many of the problems we are grappling with. Animals, plants, and microbes are the consummate engineers. After billions of years of research and development, failures are fossils, and what surrounds us is the secret to survival. (BENYUS 1997, REED 2004, BAR-COHEN 2006). The biomimicry practice follows a well-organized but flexible transdisciplinary team-based design process applicable to any kind of tangible or intangible design challenge (ROWLAND 2017). The first level of biomimicry is the mimicking of natural form. Deeper biomimicry adds a second level, which is the mimicking of natural process, or how a thing is made. At the third level is the mimicking of natural ecosystems (BAUMEISTER 2013).

To the expansion of biomimetics, education must play a significant role. It should be included in the education syllabus of architecture and design degrees to make them aware of the potential of the approach (MAHMOUD 2012). For the needs of our intention, the workshop methodology was based on the first level of Biomimicry, i.e. “mimicking of natural form”.

EXPERIMANTAL AND THEORETICAL PART Biomimicry is an approach to innovation that seeks sustainable solutions to human challenges by emulating nature’s time-tested patterns and strategies. The goal is to create products, processes, and policies—new ways of living—that are well-adapted to life on earth over the long haul.

When an interior designer says that a design is influenced by nature, he or she is most likely talking about its appearance: it has an organic shape. Nature is a good teacher in this regard, but imitating or being inspired by natural-looking forms, textures and colors alone is not biomimetics. To quote Dr. Julian Vincent ‘biomimetics has to have some biology in it.’

By which he means that a design should in some way be informed by nature’s science, not 133

just its look to be truly biomimetic. However, perhaps the key to understanding the role of biomimicry in interior architecture is the fact that the reason for the success of any design is not that it can trace its roots back to a natural principle but that it is an example of good design! Biomimicry is a philosophical approach that can lead to novel ideas and innovative solutions that have many potential advantages, for example, from functional or sustainability perspectives (MAHMOUD 2012).

Workshop Methodology The first workshop stage covered complex mapping of the state of the art of geometric principles found in the nature, as well as mapping the most significant examples of applying the nature inspired principles in art, architecture and design.

The second project stage comprised the individual scientific and artistic experimental research. The task of each participating student was to present three products of nature, which they brought to the studio. The presentation was grounded not only on observing the product of nature; moreover, it covered also geometrical and mathematical analysis of the element focused on identifying the aforementioned geometric principles and drawing the specific product of nature in identified geometric condensation. Design students were subsequently divided into groups, based on the same or similar elements they analyzed.

The creative experiment itself comprised the application of the knowledge into an object design in the form of conceptual solitaires inspired by the forms, patterns and geometrical relations derived from the natural principles. The project outputs were the scale models, sketches, geometrical analyzes and photos of scale prototypes.

As an example of good practice were students familiarized with examples of the current similar bio-inspired project such as biomimetic research project based on the biological principles of conifer cones. They presented some designs inspired by the method that cones protect the seeds inside; the spines close up to protect the seeds inside in the rainy weather and open up to improve the chances of the seeds escaping at the dry weather, see Fig. 2. One of these designs is the FAZ Pavilion which located in the city centre of Frankfurt; the summer pavilion provides an interior extension of this popular public space.

Fig. 2 FAZ Pavilion project ( MAHMOUD 2012).

Bio-inspired furniture as an example for design studies Mantis Table (Fig. 3) designed by Alvaro Uribe. Inspired by insect body parts and adding a light and elegant touch, the tableâ&#x20AC;&#x2122;s base is structured to mimic the small and delicate legs of a praying mantis which are uniquely angled to support the insectâ&#x20AC;&#x2122;s disproportionately long and heavy body. The design uses bio-inspired concepts adding to the structureâ&#x20AC;&#x2122;s lightness while still encompassing high-strength properties to support the heavy glass surface. The table uses the least amount of aluminium still being able to hold six times its own weight.

Its shape is dynamic and natural.

134

Fig. 3 Alvaro Uribe - Mantis table (http://arthitectural.com).

Biomimicry chair- 3D Printed Soft Seat (Fig.4) designed by Lilian Van Daal that uses plant cell structure. By manipulating the constructions, she has succeeded in realizing not only the base framework but also micro and macro support, ventilation, and the skin of the product using just one substance. It is designed as an alternative to conventional soft seating, which requires several different materials and processes to create the frame, padding, and covers.

Fig. 4 Lilian Van Daal – Biomimicry Chair- 3D Printed Soft Seat (https://www.lilianvandaal.com).

Radiolaria - small life-forms were for Lilian van Daal’s basic inspiration to create Biomimicry Chair - Radolaria #1(Fig.5). The structure of Radiolaria amplified by 3D printing affords various levels of flexibility and comfort without using different types of foam like in common soft seating. The lattice of connections within Bryozoa skeletons inspired Van Daal to create a system of connection points for assembly of the chair without extra materials such as glue.

Fig. 5 Lilian Van Daal – Biomimicry Chair - Radolaria #1 ((https://www.lilianvandaal.com).

135

Visible Patterns in Nature from the Geometrical Perspective As mentioned above, the natural elements contain certain geometric rules causing that people perceive such object as pretty. When talking about an object which is pretty, i.e. it has artistic qualities; it means that it drew our attention in a certain way, captivated us or pleased us.

Regarding the formal aspect, it contains some kind of order that can be expressed as pretty (CRHÁK 2012).

We seek order. The gestalt psychologist Wolfgang Metzger spoke in this connection of a love of order of the sense. Corresponding to love of order of the sense are figures that stand out of against a background and that are as forms clearly recognizable levels and axes of symmetry. This is one reason why we find crystals and certain types of organisms so beautiful, whether they are bilaterally or radially symmetrical.

Geometric rules found in many natural shapes can be characterized as “patterns in nature”. According to Stevens, 1976, patterns in nature are visible regularities of form found in the natural world. These patterns recur in different contexts and can sometimes be modelled mathematically. Natural patterns include symmetries, spirals, meanders, waves, trees, tessellations, foams, cracks and stripes. From the viewpoint of using the natural geometry in design, symmetry and curves play the most significant role.

Symmetry According to Martin, 1996, a geometric object has symmetry if there is an “operation” or “transformation” (such as an isometry or affine map) that maps the figure or object onto itself; i.e. the object has an invariance under the transform. The exact symmetry, however, is strictly limited to mathematics and geometry only. Real object show only approximate symmetry, which is processed by brain into an ideal symmetric shape. Symmetry creates a class of patterns in nature, where the near-repetition of the pattern element is by reflection or rotation.

The main types of symmetries are: point symmetry, linear symmetry and rotational symmetry.

A figure has point symmetry (point reflection) if it is built around a point, called the center, such that for every point on figure there is another point directly opposite and at the same distance from the center.

A set of points has linear symmetry (bilateral symmetry, mirror symmetry) if and only if there is a line “l”, such that reflection through “l” of each point in the set is also a point in the set. Certain figures can be mapped onto themselves by a reflection in their lines of symmetry. It is possible for a figure to have more than one line of symmetry.

Let’s show for example of real function to illustrate the symmetry of its graphs.

Let f(x) be a real-valued function of a real variable. Then f is even if the following equation holds for all x and –x in the domain of f. f(x) = f(-x) Geometrically speaking, the graph face of an even function is symmetric with respect to the y-axis, meaning that its graph remains unchanged after reflection about the y-axis.

A geometrical figure has rotational symmetry if the figure appears unchanged after a rotation around a point by an angle whose measure is strictly between 0º and 360º. The angles of 0º and 360º are excluded since they represent the original position (nothing new happens).

The angles of rotational symmetry will be factors of 360.

136

Fig. 5 Natural objectâ&#x20AC;&#x2122;s symmetry.

Curves and Spirals The straight line can be found rarely in the nature. The growth and changes of the environment modelled the living organisms into rounded shapes based on curves. Many natural curves are excellent real models of mathematical curves. Mathematical curves most often found in the nature are e.g. circle, spiral, helix or regular sinuous curve - meander.

Spiral is a plane curve described by the point P on a line p turning around a fixed point O, while the distance OP does not change. The most famous spirals are: Archimedean spiral r = a + bθ logarithmic spiral r = a. e bθ (approximations of this are found in nature as Fibonnaci and golden spiral) Fermatâ&#x20AC;&#x2122;s spiral r = θ1/2 hyperbolic spiral r = a/θ Spirals are common in plants and in some animals, notably mollusks. For example, in the nautilus, a cephalopod mollusk, each chamber of its shell is an approximate copy of the next one, scaled by a constant factor and arranged in a logarithmic spiral. Other plant spirals can be seen in phyllotaxis, the arrangement of leaves on a stem, and in the arrangement of other parts as in composite flower heads and seed heads like the sunflower or fruit structures like the pineapple, as well as in the pattern of scales in pine cones, where multiple spirals run both clockwise and anticlockwise.

A model for the pattern of florets in the head of a sunflower was proposed by H. Vogel. This has the form: θ = n . 137,5°, r = c â&#x2C6;&#x161;đ?&#x2018;&#x203A; where n is the index number of the floret and c is a constant scaling factor, and is a form of Fermat's spiral. The angle 137.5° is the golden angle which is related to the golden ratio and gives a close packing of florets.

These phyllotaxis spirals can be generated mathematically from Fibonacci sequence which approximates the golden section (golden ratio).

The â&#x20AC;&#x153;golden sectionâ&#x20AC;? (lat. sectio aurea) is a division of an abscissa into two parts so that the longer part divided by the smaller part is also equal to the whole length divided by the longer partâ&#x20AC;? (Ĺ&#x2DC;Ă?MAN1987). Expressed algebraically, for quantities a and b with a > b > 0, a / b = (a+b) / a The Golden Ratio describes the perfectly proportional relationship between two proportions. Mathematicians since Euclid have studied the properties of the golden ratio, including its appearance in the dimensions of a regular pentagon and in a golden rectangle,

137

which may be cut into a square and a smaller rectangle with the same aspect ratio. The golden ratio appears in some patterns in nature.

Helix is a solid curve, which can be defined as a line on a rotary plane created by a point that is moving while the ratio between the axial shift and corresponding angle rotation is constant.

Fig. 6 Natural curves and spirals.

Natural Forms and Principles in Architecture and Design of the 20th and 21st Century The most significant style of the 20th century playing and important role in creating our current perception of what we call inspiration by the natural motives, is the Art Nouveau.

The tendency of architecture with natural shapes with its smooth flowing volumes, consistency with the natural environment and use of natural materials was established by the architect of Modern style A. Gaudi. (Fig: a) Asymmetry and irregularity of architectural shapes have become a feature of the work of F. Hundertwasser. (Fig. b) In his Vienna house architect realizes his ideals of beauty: the absence of straight lines, rich polychromatic facades (colored majolica tiles and plaster), diversity of vegetation, different configuration of window and door openings (KAZANTSEVA, MYHAL 2014).

The style which applied geometrically stylized motives in a similar way is Art Deco.

This movement connected various styles of the early 20th century including Constructivism, Cubism, Modernism, Bauhaus, Art Nouveau and Futurism. Art Deco follows the early Neoclassicism with the elements of exotic motives, stylized animals, leaves and sunrays.

Organic design was initiated by Frank Lloyd Wright, who believed in creating harmony between people and nature and considered architecture a medium for creating the perfect equilibrium between the artificial and natural world. This belief was expressed by the use of natural materials and smooth curved forms. His philosophy is clearly mirrored in his most famous buildings like the house over the waterfall “Fallingwater” or Guggenheim museum in New York.

In Scandinavia this trend was overlapping with the “Democratic design” being a way of improving the everyday life of people.

In the 70’s design, the most prominent design personality inspired by the natural forms is the designer Luigi Colani (Fig. c). “Whenever we talk about biodesign we should simply bear in mind just how amazingly superior a spider’s web is to any load-bearing structure man has made – and then derive from this insight that we should look to the superiority of nature for the solutions. If we want to tackle a new task in the studio, then it’s best to go outside first and look at what millennia-old answers there may already be to the problem.”

(COLANI 2018) In the case of Colani’s design, the outcome is evidently very successful, but even more is possible, looking at nature as a role model (GRUBER 2008).

138

4a

4b

4c

Fig. 7 a. A. Gaudi - Cassa Batlo, b. Hundertwasser – Vienna House, c. Luigi Colani - Coal mine Achenbach (https:// sk.pinterest.com/pin/370069294373125298/?1p=true).

The works of his successors like Zaha Hadid, Ron Arad, Mark Newson are strongly marked by the connection to curves inspired by bionic forms found in the nature. Especially significant was the contribution of the designer Ross Lovegrove. Inspired by the logic and beauty of nature his design possess a trinity between technology, materials science and intelligent organic form, creating what many industrial leaders see as the new aesthetic expression for the 21st century. There is always embedded a deeply human and resourceful approach in his designs, which project an optimism, and innovative vitality in everything he touches from cameras to cars to trains, aviation and architecture (LOVEGROVE 2018).

RESULTS AND DISCUSSION Witin the workshop, the students worked under the expert guidance of the authors of this article with selected natural elements that formed the basis of their design. The first phase of the workshop was aimed at acquaintance the students with the theoretical basics of Biommimicry work and the basic geometric design principles that are contained in nature.

Than the selected natural elements were studied from a natural and geometric point of view.

The selected natural element was analyzed and its geometrical properties were graphically described There was also a geometric scheme (symmetry, golden rules, spirals,…) that express the specific relationships worked up. Geometric elements and basic natural shape have become the basis for designing of new products. The methodical process was based on Biomimicry procedures representing the most complex designer processes for creating bioinspired innovations. All the design phases were constantly under the guidance of the authors of the article, and the students were led to correctly apply the theoretical approaches.

The students within the workshop designed unique solitaires with a specific function.

The outputs were in the form of sketches, visualizations, photos and models.

False Shamrock – Oxalis triangularis (authors: Dobešová Daniela, Froncová Martina, consutants: Lizoňová Denisa, Tončíková Zuzana, Melicherčíková Iveta) The inspiration for this design was the plant Oxalis triangularis with a distinctive deepmaroon to purple leaves of a triangular shape. At night the leaves resembling butterfly wings close towards the stem. From the geometry aspect, the shape is based on isosceles triangle.

The whole object is created on the principle of rotational symmetry with the angle of 120° (Fig. 8).

139

Fig. 8 Oxalis triangularis.

This geometrically interesting plant inspired the creation of a sun umbrella with additional lighting. The shapes of the umbrella are derived from the plant’s organic structure and the folding of the umbrella is inspired by the way of closing the leaves. The surface is made of translucent, water resistant fabric. The sun umbrella is supplemented by LED lighting.

African Violet (authors: Borbélyová Jozefína, Hlavatá Miriam, consutants: Lizoňová Denisa, Tončíková Zuzana, Melicherčíková Iveta) When designing this project, the inspiration was the African violet, specifically the leaf structure and arrangement. From the geometry aspect, the violet leaves are characterized by the axial symmetry. The arrangement of leaf venation supports the symmetry even more.

The leaf shape can vary with various cultivars, however, the selected element has a curve similar to the shape of the golden spiral (Fig. 9).

The principle of leaf growth was the inspiration for creating an interior wall lamp. The typical leaf surface was pronounced by the use of soft textile material pleasant to touch. The textile has the function of a lamp shade. Optical fibers were used for the lamp production.

The light source is inserted into a paper cover resembling a rock. Two branches of leaves having the function of two lightings at various heights come out of the source.

Fig. 9 African Violet.

Grain ear (authors: Horváthová Alexandra, Jack Dominik, Verbovancová Natália, consutants: Lizoňová Denisa, Tončíková Zuzana, Melicherčíková Iveta) The shape of the following object was inspired by the grain ear, its interesting shape and the 140

grain arrangement. Considering geometry, the grain ear is typical by its oblique axial symmetry. The composition is dominated by the grain arrangement rhythm (Fig.10).

The shape and internal arrangement of individual grain ear parts were the basis for designing group seating. The supporting aluminum construction is supplemented by wooden elements creating the seat body. Upholstery is designed from natural fabric.

Fig. 10 Grain ear.

Scallop shell (authors: Javorková Veronika, Kováčiková Terézia, Kristof Albert,, consutants: Lizoňová Denisa, Tončíková Zuzana, Melicherčíková Iveta) Another inspiration from the nature selected by students was the scallop shell. The geometry of the shell profile is characterized by a soft curve resembling the golden spiral in the end part. In addition it features also axial symmetry supported by the regular ribbing pattern creating a king of rhythm (Fig.11).

The design object is a shelf with book stands. Its fan arrangement separates the books.

The shelf, to which the book stands are attached, is created by several gradually decreasing parts creating a 3D effect and representing the other half of the closed shell.

Fig. 11 Scallop shell.

Sea urchin (authors: Bačíková Dominika, Klinovská Kristína, consutants: Lizoňová Denisa, Tončíková Zuzana, Melicherčíková Iveta) The inspiration in this case was the test of a sea urchin Echinus living on the European seabed. During the life the body of the sea urchins is covered by moving spines. From the geometry aspect, adults are characterized by a distinctive five-fold symmetry. Individual segments with similar color are arranged at the angle of 72°. The cross-section of the shape has the proportions of the golden ratio with a soft shape resembling the golden spiral (Fig. 12).

The designed objects represent massage seating and backrests for a relax zone or a pool inspired by the shape and structure of the test. The designed objects can be used in

141

exterior, too, e.g. in a coral bay. The soft oval shape is supplemented by water massage nozzles.

Fig. 12 Sea urchin.

Pine cone (authors: Debnárová Mária, Magerová Júlia, consutants: Lizoňová Denisa, Tončíková Zuzana, Melicherčíková Iveta) The inspiration from nature was the pine cone and the spiral in its arrangement and the ornament created this way. The female cone has two types of scale: the bract scales, derived from a modified leaf, and the seed scales. The bract scales are spirally arranged around a central axis. The cone geometry is interesting due to several curves visible on its surface.

Firstly, it is the arrangement of the bract wood scales in the shape resembling the golden spiral, which can be observed when looking at the cone upside down. Secondly, the bract woody scales are arranged in the shape of a helix (Fig. 13).

The interesting shape created by the nature provided a form suitable for further design inspiration. The spiral curve was preserved and the spaces between individual bract scales were used as storage space for designing a cooking spoon holder.

Fig. 13 Pine cone.

Citrus fruit (authors: Gondová Ivana, Hanesová Simona, consutants: Lizoňová Denisa, Tončíková Zuzana, Melicherčíková Iveta)

142

The design inspiration was the citrus fruit and its cross-section and regular segment structure.

The segments are also called “liths”, and the space inside each lith is a locule filled with juice vesicles. From the aspect of geometry, the cross-section of the citrus fruit features strong radial symmetry. The individual segments create endless rhythm. Various types of citrus fruits are divided into different number of liths, a feature depending on the variety of the fruit.

Fig. 14 Citrus fruit.

The designed object is a fruit holder having a shape of citrus fruits. It is created by three main segments that can be separated if required. Inside parts of the segments are further divided by plastic elements in order to enable separate storage of individual fruit types without touching (Fig.14).

Humans are a part of the nature, and therefore, they perceive the arrangement and shapes as something natural and one’s own and such shapes and arrangements are often defined as beautiful. Within education and artistic aims of the project all partial aims were fulfilled: to focus on and map the theory of geometric principles found in the nature and to use the obtained theoretical analyses for creating new design concepts. Within acquiring the information we supplemented the theoretical input of geometric and mathematical analyses of living products of nature with specific examples showing how architects, designers and artists were and still are inspired by the nature and its forms. Most students considered the method “mimicking of natural forms” very inspiring and creative. When comparing this workshop with other workshops focused on artistic geometry and other topics, the students showed more creative potential and were able to create very remarkable concepts.

Within the project evaluation we have realized few deficiencies caused by the time limitation for the instructors’ theoretical inputs. Similarly, the absence of a biologist was considered a negative, since they would be able to immediately answer any questions on the shape and mainly the function of a certain geometric form from the organism evolution point of view. It was encouraging that several students were eager and became absorbed in the topic in order to understand the geometrical shape of the product of nature not only from the formal, but also from the functional aspect. This was partially fulfilled in the case of the false shamrock design, whose lamp shade opening mechanism was inspired by the way how the plant closes its leaves at night. Also the students inspired by the “citrus fruit” used not only the shape of the fruit segment, but also the internal structure reinforcing the bowl construction. The results can be formally compared with similar short-term workshops, when the designers applied the biomimicry procedures, e.g. “Furniture Design” course given to Karadeniz Technical University Interior Architecture 3rd grade students. (TAVSAN, SONMEZ 143

2015), or findings by (DE PAUW et al. 2012). The findings indicate that NIDS inspire the students, encourage out-of-the-box thinking, and provide absolute- instead of relativedesign principles to guide the concept development.

Our results indicate a strong potential of the biomimicry procedure for encouraging innovative and creative designer thinking focused on creating furniture and interiors. Since it is them who are the new generation influencing the creation of space that surrounds us in the most intimate sphere – in our homes. Our experience with applying the biomimicry procedures is similar to those of other universities. It shows that universities should make effort to create conditions for availability of similar courses and workshops dealing with the methods suitable for creating new nature inspired innovations.

CONCLUSION Nature has been a huge laboratory for research, development and design for 3.8 billion years and is still innovating. Since time immemorial people have turned to nature where they have been looking for inspiration, advice or ideas. Almost all aspects of human progress have been associated with observing and studying the natural structures and phenomena.

People, led and inspired by the natural laws, were able to create cathedrals and bridges, brilliant works of arts and inventions. Many of the world’s history thinkers like Leonardo da Vinci or Albert Einstein were inspired by natural systems, structures and creations. A framework for understanding one form of biomimicry has been experimentally verified on real student’s projects. The case studies emphasize that integrating biomimicry within interior environments requires introducing the approach at the primary stages of the design process, ideally before any preliminary ideas have even been formed. Nature opens the designers, architects, scientists and engineers a completely new world and enables them, through studying the capabilities, structures, shapes and processes developed in plants and animals, to design better, stronger, more ecological and sustainable products for our life and future. Within the short-time workshop with design students we managed to create a compact collection of solitaires inspired by the geometric forms found in living products of nature.

At first sight an elementary task showed to be very positive, enlightening and creative experience providing a proof of what a perfect and harmonic world has the evolution created.

We believe that such experience will stay in the design students’ minds and will help them to look for the answers in the nature also in the future when they face important design challenges.

REFERENCES ALI, MAHMOUD 2012. Biomimicry as a Problem Solving Methodology in Interior Architecture.

Procedia - Social and Behavioral Sciences. 50. 502–512. 10.1016/j.sbspro.2012.08.054.

BAR-COHEN, Y. 2006. Biomimetics—using nature to inspire human innovation. In Bioinspiration & Biomimetics, 2006, 112 BAUMEISTER, D, TOCKE, R, DWYER, J, RITTER, S, BENYUS, J. 2013. The Biomimicry Resource Handbook: A Seed Bank of Best Practices. In Biomimicry 3.8: Missoula. 280 p.

BENYUS, J. 1997. Biomimicry: Innovation inspired by nature. New York : William Morrow, 1997, 324p. ISBN 978-0060533229.

CRHÁK, F. 2012. Výtvarná geometrie plus. Brno : VUTIUM, 2012.186 p. ISBN 978-80-214-3767-8

144

DE PAUW, I. C., KARANA, E., KANDACHAR, P. V. 2012. Nature-inspired design strategies in sustainable product development: A case-study of student projects. In DS 70: Proceedings of DESIGN 2012, the 12th International Design Conference, Dubrovnik, Croatia, 787796.

ELAM, K. 2011. Geometry of design: Studies in proportion and composition. New York : Princeton Architectural Press, 2011. 143 p. ISBN 978-16-1689-036-0 GRUBER, P. 2008. The signs of life in architecture. In Bioinspiration & biomimetics, 2008, 3(2), p. 023001.

HAECKEL, E. 1998. Atr forms in nature. Munich : Prestel-Verlag, 1998.139 p. ISBN 3-7913-1990.

KAZANTSEVA, T., MYHAL, S. 2014. Aesthetic tendencies in the architectural and landscape design driven by natural shapes. In Przestrzeń i Forma, 22(1): 91104.

MANDELBROT, B. 1982. The fractal geometry of nature. New York : W. H. Freeman & Co, 1982.

480 p. ISBN 0-7167-1186-9.

METZGER, W. 2006. Laws os Seeing. Cambridge : MIT Press, 2006. 194 p. ISBN 0-262-13467-5.

PLATO. TIMAEUS. 1975. Minneapolis : Wizard’s Book-shelf, 1975.

REED, P. A. 2004. A Paradigm shift: Biomimicry. The technology teacher. December/January. 2004.

23–27. ŘÍMAN, J. 1987. Malá československá encyklopédia. Praha : Academia, 1987. 927 p.

ROWLAND, R. 2017. Biomimicry step-by-step. In Bioinspired, Biomimetic and Nanobiomaterials, 2017, 6(2): 102112.

STEVENS, P. S. 1976. Patterns in Nature. London : Little, Brown & Co. 1976. 256 p. ISBN 978-01405-5114-3.

TAVSAN, F., SONMEZ, E. 2015. Biomimicry in furniture design. In Procedia-social and behavioral sciences, 2015, 197: 22852292.

VOŘÁČKOVÁ, Š. 2012. Atlas geometrie- Geometrie krásna a užitečná. Praha : Academia, 2012.

252 p. ISBN 978-80-200-1575-4 Available online: www.colani.ch/ (09/2018).

Available online: www.rosslovegrove.com/index.php/about-us/ (09/2018).

AUTHORS’ ADDRESSES Ing. Denisa Lizoňová, ArtD. Technical University in Zvolen Department of Mathematics and Descriptive Geometry T. G. Masaryka 24 960 53 Zvolen Slovakia denisa.lizonova@tuzvo.sk Ing. Zuzana Tončíková, ArtD. Technical University in Zvolen Department of Furniture Design and Interior T. G. Masaryka 24 960 53 Zvolen Slovakia zuzana.toncikova@tuzvo.sk

145

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 147154, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.14

THE IMPACT OF THE SPACER ON THE INTERIOR SURFACE TEMPERATURE IN THE DETAIL OF WOOD WINDOW GLAZING Roman Nôta ABSTRACT The impact of the spacer of insulated glazing on the surface temperature in the glazing detail of a wooden window is discussed in the paper. Six warm edge profiles from three manufacturers were assessed in the study, the model glazing with parameters 4-12-4-12-4 with the U-factor of 0.7 W/m2K. The detailed model of the spacer was compared also to the Two Box model. Subsequently, the Psi values of glazing (Ψg) for these spacers were determined and compared with the values from the Data sheet Psi values for windows (BUNDESVERBAND FLACHGLAS E.V.) and calculated values of Uw according to the STN EN ISO 10077. Following the calculation, minimum surface temperatures ranging from Δθsi of approx. 0.10 °C for the detailed model to Δθsi of approx. 0.15 °C for the Two Box model was observed. When comparing the detailed models with the Two Box models, the minimum surface temperatures were in the range from approx. 0.5 °C representing 3.40 %. This difference was also observed in case of the exterior temperature resulting in condensation on the test window. In such case the difference between the detailed and Two Box model was at the value of 16.94%, while the Uw value for all studied variants was the same 1.7 W/m2K. Key words: wood windows, surface temperature, IGU spacer, Psi values.

INTRODUCTION Hygienic requirements affect the minimum required interior surface temperature of a window, since condensation can occur on the construction surface leading to mould growth on these surfaces. Condensed water on the material surfaces is more significant for the mould growth than the atmospheric humidity. Approximately 7080 % of moulds occurring in the environment can produce mycotoxins affecting negatively the human health (UVZ SR, 2014).

Condensation on the material surface occurs when the surface temperature decreases below the dew point temperature. Moulds grow and develop with higher relative atmospheric humidity. The standard STN 73 0540-3 defines the critical surface temperature for mould development for various combinations of temperature and atmospheric humidity.

The coldest place on the surface of a window construction is the detail of glass system fitting. It is caused by the major thermal bridge created by the shape of the construction as well as by the impact of the spacer in the insulation glazing unit (IGU). It is a place on the glass where the surface temperature changes from the lowest value to the value which equals to the surface temperature in the centre of IGU. This region, according to VAN DER BERGH 147

et al. (2013), can reach up to approx. 102 mm. It can be affected by a number of factors, from the thermal characteristics of the environment, glazing thickness, to the type and thickness of the window construction etc. However, the type of the used spacer affects the surface temperature in the glass fitting detail most. Its impact in calculating the window heat transfer coefficient Uw is expressed by the linear thermal transmittance of the glazing – Ψg (Psi value).

Nowadays, spacers with improved thermal performance, so called warm edge spacers, are used most often. These are according to the standard STN EN ISO 10077-1 defined by the equation: Σ (d x λ) ≤ 0,007 (01) Where d is the spacer wall thickness and λ is the thermal conductivity coefficient of the material.

For these spacers the Ψg value, for plastic window and insulation double glazing, is less than 0.051 W/mK (MEYER-QUEL 2017).

Due to the complex nature of calculating the thermal performance of the spacers and a number of materials a simplified calculation model Two Box model was created. The principle is in substituting the spacer by a pair of rectangles (box), where one represents the sealing substance (polysulfide) and the other one represents the spacer construction. The dimensions of the rectangles are the same as those of the spacer, and the thermal performance for the calculation is substituted by the equivalent value of the thermal conductivity coefficient (SVENSEN et al.

2005, IFT ROSENHAIM 2015).

THEORETICAL – EXPERIMENTAL PART Calculation of thermal transmittance and surface temperature through window frame was made and based on EN ISO 10077 Thermal performance of windows, doors and shutters — Calculation of thermal transmittance. Part 1: General and Part 2: Numerical method for frames. Glazing model is derived from the programme WINDOW 7.6 (HUIZENGA et al.

2017B). It has been done by modelling in computer programme THERM 7.6 (HUIZENGA et al. 2017A). Boundary conditions for the calculation by EN ISO 10077-2. Reason for this is using windows in less favourable conditions.

Reference temperature:

θi = 20 °C θe = 0 °C

internal external

Reference surface resistance: internal Rsi = 0.13 m2·K /W internal - reduced radiation Rsi = 0.20 m2·K /W external Rse = 0.04 m2·K/W To calculate the θsi of materials, the values of thermal conductivity (λ [W/m·K]) according to the Tab. 1 and Tab. 2 were used. The values are taken from STN EN ISO 100772:2018 which gives us the characteristics of the materials most commonly used for production of windows.

A model of wooden window construction with construction depth of 88 mm was used for the modelling (see Fig. 1).

148

Tab. 1 Coefficient of thermal conductivity of window frame materials.

Thermal conductivity (λ [W/m·K]) 0.25 50.00 0.11 0.35 160.00 0.17

Material ethylene propylene diene monomer (EPDM) steel (oxidized) Picea Abies (L.) silicone alloy aluminium polyvinylchloride (PVC) Flexible, with % softener

Equivalent thermal conductivity (λeq) air cavities has been determined according to the algorithms in the software program THERM, modelled using the ISO 15099 (Thermal performance of windows, doors and shading devices – Detailed calculations) cavity Model

(a)

(c)

(e)

(b)

(d)

(f)

Fig. 1 Model of wooden window and detailed models of spacers ((a) Chromatech Ultra F, (b) Chromatech ultra S, (c) Thermix TX.N plus, (d) Thermix TX pro, (e) TGI Spacer, (f) TGI Spacer M).

Tab. 2 Coefficient of thermal conductivity of spacers.

Material Desiccant - Silicagel * Silicone sealant (DC 3362 HD)** - secondary seal Polyisobutylene (PIB) * – primary seal Glass Gas – gap 1 (10% air- 90% argon - EN 673) *** Gas – gap 2 (10% air- 90% argon - EN 673) *** Chromatech Ultra F 1 Stainless steel 1.4372 Polyvinylchloride (PVC) Chromatech Ultra S 2 Stainless steel 1.4372 Polypropylene (PP) Thermix TX.N plus 3 Thermix TX Pro 4 Polypropylene (PP) - no glass fibers Steel Stainless steel 1.4372 TGI Spacer 5 Polypropylene (PP) with talcum powder Stainless steel 1.4301 TGI Spacer M 6 Polypropylene (PP) with talcum powder Steel C72D2 Stainless steel 1.4372

Thermal conductivity (λ) [W/m·K] 0.13 0.26 0.20 1.00 λeq 0.020 λeq 0.021 15 0.17 15 0.25 0.22 50.00 15 0.193 15 0.193 47.30 15

Document Technique d´Application 6/15-2256, 2 Document Technique d´Application 6/17-2365_V1, 3,4 Document Technique d´Application 6/16-2348, 5,6 Document Technique d´Application 6/16-2305_V1, * STN EN ISO 10077-2, ** Product Information, Dow Corning® 3362 HD Insulating Glass Sealant, *** λeq by HUIZENGA et al. 2017B 1

149

After evaluating the minimum surface temperature, the temperature index calculated according to the equation (02) was established. “The temperature index is non-dimensional, and represents the interior surface temperature relative to the interior and exterior air temperatures.

The use of the temperature indexes offers the opportunity to compare the thermal performance of samples subjected to different boundary conditions.” (MAREF et al. 2011) θ -θ

fRsi(θ) = θsi-θ e i

(02)

e

Via its modification (Shin, 2017), the exterior temperature, which will cause condensation on the model window, can be determined (equation (03)). θe,dp−min =

θdp - fRsi(θ) θi

(03)

1 - fRsi(θ)

Where θdp represents the dew point temperature corresponding to the calculation of the interior air temperature at its relative humidity. Our conditions correspond to the conditions of STN 730540-2 (θi = 20°C a φi = 50%), while the dew point temperature in such conditions is θdp = 9.26°C. The Ψg value calculations were performed according to the procedure outlined in “Calculating Fenestration Product Performance in WINDOW 6 and THERM 6 According to EN 673 and EN 10077” (LBNL 2012) and the associated spreadsheet, with the following exceptions. The surface transfer coefficient at internal/external surface used in the models was 25 W/m2K for the external surface and 7.69 and 5 W/m2K for the internal and internal - reduced radiation surface according to the ISO 10077, instead of 23 and 3.6+(4.4*ε/0.9) W/m2K shown in the document.

RESULTS AND DISCUSSION

18 16 14

Surface temperature on the warm side [°C]

20

The interior surface temperature on the window structure obtained when modeling with the Chromatek Ultra F spacer is shown in the graphs in Fig. 2. For the other spacers the temperature course is the same. The lowest temperature is in the place of IGU contact with the window frame. This place is represented by 0 on the x axis (see fig. 3). In this place, condensation occurs most often causing a high risk of mould creation. The temperature in this place can be seen in Table 3. The table provides the temperatures for the detailed model of the spacer as well as for the Two Box model.

-220 -200 -180 -160 -140 -120 -100 -80 -60 -40 -20

0

20 40 60 80 100 120 140 160 180

Distance of the window [mm] Fig. 2 Interior surface temperature of the wooden window frame and IGU whit Chromatech ultra F spacers - detailed model. Θsi min = 14.65 °C.

150

190 value on the x-axis (fig. 2 to 7)

0 value on the x-axis (fig. 2 to 7), The lowest temperature

-220 value on the x-axis (fig. 2 to 7)

Fig. 3 Distance of the windows – value on the x-axis in Fig. 2.

The biggest difference between surface temperatures was recorded in the glazing detail reaching the value of 0.09 °C, representing 0.64%.

Tab. 3 Interior surface temperature of the edge of glass. spacer Chromatech ultra F Chromatech ultra S Thermix TX pro Thermix TX.N plus TGI Spacer TGI Spacer M

detailed model 2B model detailed model 2B model detailed model 2B model detailed model 2B model detailed model 2B model detailed model 2B model

Θsi min [°C] 14.65 14.17 14.61 14.03 14.60 14.11 14.56 14.11 14.61 14.13 14.60 14.11

ΔΘsi min [°C] DM vs. 2B

0.48 3.30% 0.58 3.97% 0.49 3.36% 0.45 3.11% 0.48 3.30% 0.49 3.37%

fRsiΘ [-] 0.733 0.709 0.730 0.701 0.730 0.706 0.728 0.705 0.731 0.707 0.730 0.706

Θe pd-min [°C] 20.17 16.84 19.84 15.97 19.80 16.48 19.48 16.45 19.88 16.61 19.80 16.47

More significant difference for the wooden window was recorded in comparing the detailed model of the spacer with the Two Box model. This represented approx. 0.5°C being 3.40%. From this aspect the difference is not significant. However, if the thermal coefficient is used to determine the minimum exterior temperature for condensation (θ e, pd-min), the temperature difference between the detailed and Two Box model is Δ θ e, pd-min 3.36°C, representing 16.94%.

151

-22,00

Detailed model

Exterior temperature [°C]

-20,17

-19,84

-19,80

-16,84

-19,48

-16,48

-15,97

-19,88

2B model

-19,80

-16,61

-16,45

-16,47

-16,00

-10,00 Chromatech ultra F

Chromatech ultra S

Thermix TX Thermix TX.N TGI Spacer pro plus

TGI Spacer M

Fig. 4 Calculated exterior air temperature at which the condensation start to occur.

Following these values, we determined values of linear thermal transmittance for glazing Ψg for the detailed model and Two Box model for individual model situations and compared them with data from the Data sheet Psi values for windows (Table 4).

The difference between the detailed and Two Box model was probably caused by the used values of coefficient of thermal conductivity of individual materials used for calculations. For instance, with secondary seal representing approx. 30.29% of the insulation glazing spacer area, the value used for the detailed model is 0.26 W/mK (Dow Corning Corporation 2013) and for Two Box model 0.40 W/mK (BUNDESVERBAND FLACHGLAS E.V.) representing performance better by 35%. The properties of the spacers were discussed in the studies of VAN DER BERGH et al. (2013), SVENSEN (2005) and ELMAHDY (2003). Their calculation and the transfer of the detailed models to the Two Box models are described in the Guideline ift Rosenheim (WA-08/3 and WA-17/1).

Tab. 4 Comparison of psi value, BUNDESVERBAND FLACHGLAS E.V. vs. calculated value. spacer Chromatech ultra F Chromatech ultra S Thermix TX pro Thermix TX.N plus TGI Spacer TGI Spacer M

detailed model 2B model detailed model 2B model detailed model 2B model detailed model 2B model detailed model 2B model detailed model 2B model

Ψg

Ψg

ΔΨg

ΔΨg

BF

cal.(LBNL)

BF vs. cal.

2B vs. detailed

 0.038  0.041  0.039  0.040  0.039  0.039

0.026 0.035 0.027 0.038 0.027 0.037 0.028 0.037 0.027 0.036 0.026 0.037

0.003 6.71 % 0.003 7.01 % 0.002 6.34 % 0.003 6.74 % 0.003 7.28 % 0.002 6.22 %

0.012

67.86%

0.014

64.95%

0.012

68.63%

0.012

68.81%

0.012

68.30%

0.013

67.16%

Table 5 illustrates the Uw values calculated according to the STN EN ISO 10077 for the variants of linear thermal transmittance coefficient according to the Bundesverband Flachglas (BF) (UwΨgBF), calculated with the Two Box model (UwΨg2B) and with the detailed model (UwΨgDM), as well as comparisons of the differences between the calculated values and values provided by the Data sheet Psi values for windows (BUNDESVERBAND FLACHGLAS E.V.).

152

Tab. 5 Comparison of U-value of wooden windows with different psi-value. spacer

Uw

Uw

Uw

ΨgBF

Ψg2B

ΨgDM

2

2

[W/m K] [W/m K] Chromatech ultra F Chromatech ultra S Thermix TX pro Thermix TX.N plus TGI Spacer TGI Spacer M

1.729 1.736 1.732 1.734 1.731 1.732

1.694 1.700 1.696 1.698 1.695 1.696

2

[W/m K] 1.671 1.673 1.673 1.675 1.673 1.672

ΔUw

ΔUw

Ψg2B vs. DM.

Ψg BF vs. DM.

[W/m2K]

[W/m2K]

0.023 0.027 0.023 0.023 0.023 0.025

1.36% 1.62% 1.38% 1.38% 1.34% 1.46%

0,059 0,064 0.059 0.060 0.058 0.061

3.41% 3.68% 3.42% 3.43% 3.38% 3.50%

Uw Ψg BF,2B,DM

by EN ISO

1,7 1,7 1,7 1,7 1,7 1,7

According to the calculations, the heat transfer coefficient for various spacers varied in thousandths of W/m2K and did not exceed 0.40%.When comparing individual models and thus also the Psi factor values, the difference was in the range of hundredths of W/m2K and did not exceed 3.68%. When rounding according to the paragraph 7.2.3 of STN EN ISO 10077-1, the Uw values do not vary.

CONCLUSIONS AND FUTURE WORK According to the calculations carried out using six detailed models of insulation glazing spacers and Two Box models of the corresponding spacers, the impact on the minimum surface temperature is insignificant for individual spacer types, on average 0.64%. Nevertheless, the most significant difference occurred when comparing the detailed models with Two Box models – 16.94%. Eventually, such differences do not affect the value of heat transfer coefficient of window.

It can be supposed that this difference occurred due to the impact of thermal performance of the materials used in the production of individual spacers mentioned in the CCFAT documents. Since the study dealt with the theoretical calculation of the surface temperatures, confirming such assumption will be discussed in subsequent studies. Laboratory measurements will be required in order to confirm the temperature causing condensation in various types of spacers.

REFERENCES BUNDESVERBAND FLACHGLAS E.V. 2017. Data sheet Psi values for windows, based on determination of the equivalent thermal conductivity of spacers by measurement, for the Chromatech ultra F (No. W 16 2-05/2016), Chromatech utra S (Nr. W35 12/2017), Thermix TX pro (No. W34 01/2017), Thermix TX.N plus (No. W10 4-05/2016), TGI Spacer (No. W9 4-05/2016), TGI Spacer M (No. W20 1-5/2016).

LBNL 2012. Calculating fenestration product performance in Window 6 and Therm 6 according to EN 673 and EN 10077, Online: https://windows.lbl.gov/tools/knowledge-base/articles/en-673-iso10077, [cit.: 15.7.2018], Lawrence Berkeley National Laboratory, Windows and Daylighting, Building Technology & Urban Systems Division, 2009.

CCFAT 2018A: Document Technique d´Applocation. Référence á l´Avis Technique 6/16-2302_V1 TGI®-Spacer, TGI®-Spacer M, CSTB, Champs-sur-Marne, France. 2018.

CCFAT 2017A: Document Technique d´Applocation. Référence á l´Avis Technique 6/16-2348 THERMIX TX.N plus, THERMIX TX Pro, CSTB, Champs-sur-Marne, France. 2017.

CCFAT 2018B: Document Technique d´Applocation. Référence á l´Avis Technique 6/17-2365_V1 CHROMATECH ULTRA S, CSTB, Champs-sur-Marne, France. 2018.

153

CCFAT 2017B: Document Technique d´Applocation. Référence á l´Avis Technique 6/16-2347 CHROMATECH ULTRA F2, CSTB, Champs-sur-Marne, France. 2017.

DOW CORNINING CORPORATION 2013. Product Information, High Performance Building, Dow Corning® 3362 HD Insulating Glass Sealant, Form No. 62-1635D01, 2.10.2016.

ELMAHDY, A.H. 2003. Effects of improved spacer bar design on window performance. In National research Council of Canada, September 2003, ISSN 1206-1220.

HUIZENGA, CH. et al. 2017A. THERM Fine Element Simulator v7.6.1.0: Program description. A PC program for analysing the two-dimensional heat transfer through building products. Berkeley, California : University of California 2017.

HUIZENGA, CH. et al. 2017B. BERKELY LAB WINDOW v7.6.4.0: Program description. A PC program for NFRC Certification and modelling Complex Glazing Systems. Berkeley California : University of California 2017.

IFT ROSENHEIM 2015. Thermally improved spacer, Part 1, Determination of representative Ψ-values for profile sections of windows, ift-Gudeline WA-08/3 February 20015, ift Rosenheim GmbH, Rosenheim, 2015.

IFT ROSENHEIM 2013. Thermally improved spacer, Part 2, Determination of equivalent thermal conductivity by means of measurement, ift- Gudeline WA-17/1 October 2013, ift Rosenheim GmbH, Rosenheim, 2013.

MAREF, W. et al. 2011. Condensation risk assessment on box windows: the effect of the windowwall interface. In Journal of Building Physics, 36(1): 3556, Sage Publications, London, 2017 ISSN 1744-2591, online ISSN 1744-2583, 2012, DOI: 10.1177/1744259111411653.

MEYER-QUEL, I. 2017. Update in Sachen Warme Kante. In Glaswelt: Fenster, Fassade, Glas 4/2017, 9698, online: https://www.warmekanteberater.de/wp-content/uploads/2017/04/Glaswelt_042017_Update-Warme-Kante_KORR.pdf, [cit.: 20.8.2018], Alfons W. Gentner Verlag Gmbh & Co. KG, Stuttgart, 2017, ISSN: 0017-1107.

SHIN, M.-S., RHEE, K.-N., YU, J.-Y., JUNG, G.-J. 2017. Determination of Equivalent Thermal Conductivity of Window Spacers in Consideration of Condensation Prevention and Energy Saving Performance. In Energies, 10(5): 717, MDPI AG, Basel, Switzerland, 2017, EISSN 1996-1073 DOI: 10.3390/en10050717.

SVENSEN, S., LAUSTSEN, J. B., KRAGH, J., 2005. Linear thermal transmittance of the assembly of the glazing and the frame in windows. In Proceedings of the 7th Symposium on Building Physics in the Nordic Countries 2, pp. 995-1002, IBRI, Keldnaholti, IS-112, Reykjavik Iceland, 2005.

STN 73 0540-2/2012, Thermal protection of buildings. Thermal performance of buildings and components. Part 2: Functional requirements, ÚNMS SR, 2012 STN EN ISO 10077-1/2018. Thermal performance of windows, doors and shutters – Calculation of thermal transmittance – Part 1: General, ÚNMS SR, 2018 STN EN ISO 10077-2/2018. Thermal performance of windows, doors and shutters – Calculation of thermal transmittance – Part 2: Numerical method for frames, ÚNMS SR, 2018 VAN DEN BERGH, S., HART, R., PETTER JELLE, B., GUSTAVSEN, A. 2013.Window Spacer and Edge Seals in Insulating Glass Units. A state-of-the-Art review and Future Perspective. In Energy and Buildings 58: 263280. Elsevier B.V. 2013 ISSN: 0378-7788. DOI: 10.1016/j-enbuild.2012.10.006.

UVZSR 2014. Užitočné informácie o plesniach (Useful information about funnels), online, (http://www.uvzsr.sk/index.php?option=com_content&view=article&id=2107:uitone-informacieonplesniach&catid=101:vnutorne-prostredie-a-zdravie), [cit. 24.5.2017] Úrad verejného zdravotníctva Slovenskej republiky (Public Health Authority of the Slovak Republic), 2014.

AUTHOR ADDRESS Ing. Roman Nôta, PhD. Technical University in Zvolen Department of Furniture and Interior Design T.G. Masaryka 24 960 53 Zvolen nota@tuzvo.sk 154

ACTA FACULTATIS XYLOLOGIAE ZVOLEN, 61(1): 155−165, 2019 Zvolen, Technická univerzita vo Zvolene DOI: 10.17423/afx.2019.61.1.15

HOW COMPANIES IN THE WOOD SUPPLY CHAIN PERCEIVE THE FOREST CERTIFICATION Hubert Paluš – Ján Parobek – Michal Dzian – Samuel Šimo-Svrček – Martina Krahulcová ABSTRACT The idea of forest certification, a means of tackling deforestation and forest degradation emerged in the 1990s. Forest certification is a voluntary process whereby the quality of forest management and production is assessed by an independent third party.

Besides the general information on the implementation of forest certification systems, the study is focused on the description of the main roles of forest certification. In particular, how roles are perceived by certified entities and the differences in perception of forest certification between the forest owners and managers and the chain of custody certified companies in the wood supply chain are analysed. Respondents are people with high level of understanding the certification concept and considering the certification primarily as a non-economic tool. Subsequently, they understand certification as an economic tool as well.

Forest owners and certified companies tend to have same incentives for entering the certification process, however, with different priorities among these incentives. Finally, the main problems related to certification are identified in the paper. In contrast to certified wood processing industries and traders, forest owners do not consider certification costs to be the main problem. On the other hand, main problems of the certified supply chain identified by chain of custody companies is those connected to the sufficient quantity of certified forest products and overpriced certified material inputs. The fact that positive aspects of certification are better valued than its shortcomings are confirmed by the main results.

Key words: wood supply chain, forest certification, chain of custody, certified companies.

INTRODUCTION Forest certification was initially introduced as a voluntary mechanism by environmental groups to ameliorate the consequences of tropical deforestation and forest degradation (RAMETSTEINER and SIMULA 2003). It represents a mechanism based on thirdparty auditing of compliance with established standards. This mechanism was quickly accepted as a means to promote sustainable forest management (DURST et al. 2005, PERERA et al. 2007, SIRY et al. 2005) and directly influenced forest management practices (AULD et al. 2008, LEWIS and DAVIS 2015, MACDICKEN et al. 2015, MOORE et al. 2012). Through certification as a soft policy instrument (NUSSBAUM et al. 2005, ŠÁLKA et al. 2017), it is possible to provide credible assurance to customers about the effective compliance of forest management with sound social, environmental, and economic principles (CABARLE et al.

1995, JOHNSON and WALCK 2004, RICKENBACH and OVERDEVEST 2006).

155

Since their launch, forest certification programs have increasingly become an instrument of governmental procurement policies, obligatory requirements for awarding ecolabels, corporate policies of private companies, requirements for green building initiatives, and acceptance as a tool for proving the legality of timber origin. In addition, perceived pressure from shareholders, firm size, financial health, past environmental performance, and regulatory threats have been linked to firms’ decisions to meet environmental standards voluntarily. For some certified companies the implementation of forest certification provides the satisfaction of supporting the sustainability of natural forest resources and society as a whole (WWF 2000). It may also serve to improve their corporate images and access to markets (HANSEN and PUNCHES 1999, HUBBARD and BOWE 2004) or may be part of business system innovations (GILANI et al. 2016).

NUSSBAUM and SIMULA (2005) identified several policy issues having relevance to certification such as responsible or sustainable forest management, the balance between economic, social and environmental concerns for forest management, illegal harvesting, conservation of biodiversity, timber markets, etc. At the forest management level, forest certification is a process by which forest owners voluntarily submit their forests for inspection by an independent certification body to determine whether their management practices meet clearly defined standards, particularly those regarding sustainability (PECK 2006). Certified forest consider certification to be a tool representing a commitment to environmental responsibility that improves external company image, promotes sustainable utilisation of forest resources, and improves forest management practices (PALUŠ et al.

2018a).

Chain of custody (CoC) is a mechanism that provides assurance that wood and woodbased products originate from sustainably managed forests. It becomes one of the factors in determining leadership position in the forest and wood-based sector, especially under economic crisis conditions (POTKAŃSKI et al. 2011). TUPPURA (2016) found out that incentives for adopting forest certification among the world’s leading forestry companies are more often external rather than internal, and more market driven than regulation driven.

Immature markets, the indirect nature of most benefits, and certification being an unfamiliar concept are commonly cited reasons for a lack of manufacturer support or involvement (JAYASINGHE et al. 2007). Forest certification is considered important for indicating a company’s sense of responsibility, for keeping market share and for selling products in an existing market (OWARI et al. 2006).

RICKENBACH and OVERDEVEST (2006) view forest certification as a market-based incentive for forestry enterprises as firms adopting certification practices expect direct market benefits.

Advantages of market access shall also offer sufficient incentives for suppliers to bear the costs of certification GALATI et al. (2017) argue that influence of internal drivers to adopt voluntary certification linked to a pro-environmental behaviour of owners and managers, such as a signalling mechanism and moral and ethical reasons, is more important than economic or market incentives. Empirical results by NAKAMURA et al. (2001) and TAKAHASHI (2001) revealed that the market economic and social models explained participation in forest certification.

The adoption of voluntary forest certification by forest owners may result in additional costs related to the standards implementation, initial and surveillance conformity assessment through internal and external audits, and cost resulting from the changes to the traditional forest management practices caused by the certification requirements (PALUŠ 2013).

Increased costs of certification are, thus, one of the main barriers for the adoption of forest certification for forest owners as well as for the chain of custody level (MOORE et al. 2012, PALUŠ and KAPUTA 2009, CARLSEN et al. 2012, CUBBAGE et al. 2010). The unit cost is relatively higher for small compared to large forest owners (CUBBAGE et al. 2009); however, 156

certified forest owners believe that certification benefits exceed costs (MIKULKOVÁ et al.

2015, CUBBAGE et al. 2009).

Many survey-based studies have evaluated the willingness to pay for certified wood products (AGUILAR and CAI 2010, AGUILAR and VLOSKY 2007, KOZAK et al. 2004, OZANNE and VLOSKY 2000, OZANNE and VLOSKY 1997, PALUŠ et al. 2017, PALUŠ et al. 2018b, VEISTEN 2007, VLOSKY et al. 2009, YAMAMOTO et al. 2014) and indicated a level of price premium that consumers were willing to pay for different wood products in specific market segments. Apart from the promotion of sustainable forest management practices and the satisfaction of supporting the sustainability of natural forest resources and society as a whole (WWF 2000), benefits following from the adoption of forest certification are related to the improvement of the external companies’ image associated to their environmental performance (MOORE et al. 2012, PALUŠ et al. 2017, PALUŠ et al. 2018b, VLOSKY et al. 2009, PERERA 2008, MIKULKOVÁ et al. 2015, PALUŠ and KAPUTA 2009) and environmental communication and consumer relations (OWARI and SAWANOBORI 2008).

Benefits of CoC certification include improved supply chain management performance, communications in supply chains, inventory controls, market knowledge, transparency, and profitability (VIDAL et al. 2005), lower overall costs (MILES and COVIN 2000, PEFC 2017), wood legality assurance, company’s image and competitiveness of wood products (TRISHKIN et al. 2014), good reputation and international recognition (Halalisan et al. 2013) and sales increase (TOLUNAY and TÜRKOĞLU 2014). Regarding timber legality issues, CoC certificates are an acceptable measure for the legality verification of timber products required by the European Timber Regulation (EUTR), in particular concerning risk assessment and risk mitigation procedures as a part of an operator’s due diligence system (EUROPEAN COMMISSION 2016). Implementing forest and CoC certification as an assurance for timber legality could contribute to reduced costs and administrative work for operators required to establish due diligence systems according to the EUTR. Only minor additional operator costs are required to make these systems fully compatible with EUTR requirements (EFI 2011). Role of forest certification as a tool to proof timber legality is one of the most important incentives that motives that CoC certified companies to enter certified forest products market (PALUŠ et al. 2017, 2018b).

Having introduced the meaning and main roles of forest certification the main objective of this paper is to analyse how these roles are perceived by certified entities and, in particular, what are the differences in perception of forest certification by forest owners and managers and CoC certified companies in the wood supply chain.

METODOLOGY This study is based on previous studies (PALUŠ et al. 2017, 2018a, 2018b) which was oriented to the state and perception of forest certification in Slovakia and selected countries of the Central and South Europe. One of the studies was focused on the issues of sustainable forest management certification and two studies on chain of custody certification in the wood supply chain. Both studies were carried out using a mail questionnaire survey. This study builds on the results of the three previous surveys. Forest owners and managers have been selected for the survey from the national register of forest owners and managers, national PEFC database and international FSC database. To determine the minimum sample size, a 5% margin error was assumed. Based on this assumption, the total number of forest owners was determined to 369. On the other hand, the companies in the wood supply chain were identified from the international registers of CoC holders of the PEFC and FSC certification schemes. Totally, 487 wood processing companies and traders were contacted in the survey.

157

Data of above mentioned surveys were analysed using the statistical analysis software SPSS and the non-parametric Mann-Whitney U test to determine the differences between groups of forest owners and CoC certified companies in terms certification scheme used and company size. Additionally, the Chi-square test was applied to identify all the group and between group distributions.

The analyses are oriented to the three main areas. The first section contained issues aimed at the examination of the level of understanding the concept of forest certification and CoC certification based on the objectives of PEFC and FSC certification. The research also determined the level of agreement with the basic certification statements. The main certification statements referred to the main objectives and purposes of certification. One part of questions dealt with the issues such as the use of sustainable resources, commitment to environmental responsibility, improvement of forest owners’ image and legality issues.

Second part of certification statement was focused on market access, profit margin, improved communication, as well as the improvement of the internal efficiency of management. The second area of the research surveyed internal information about involvement in the certification process, namely the expectations motivating forest owners and companies to implement certification. The last area was oriented towards difficulties regarding certification, including cost related to certification.

For the analyses a five–point Likert scale was used to measure the level of understanding of the certification concepts and level of agreement with principal certification statements, were 1 corresponded to strongly disagree or do not understand and 5 was strongly agree or completely understand. The level of agreement indicated by number 3 represented a neutral response level of all respondents. The data of current study was analysed using a comparison method. This method is developed as a grounded theory to understood the systematic comparison of a relatively small number of cases, focusing specifically on its relationship to experimental and statistical approaches.

RESULTS Based on the previous studies two groups of respondents were examined. The first group was represented by forest owners; the second group was represented by CoC certified wood processing companies and traders. The group of forest owners consisted by all types of ownership. The area of managed forest was used as an indicator of company size for forest owners. Small owners (up to 500 ha) represent 35% of the respondents, followed by medium-size (501–10,000 ha) owners (55%). Only 10% of respondents represented large owners managing more than 10,000 ha of forests. On the other hand, the size of wood processing companies was determined by the number of employees. Small companies (11– 50 employees) represent 41% of respondents, followed by equal representation (27% each) of micro (1–10 employees) and medium-size companies (51–250 employees). Only 5% of respondents represented large companies (over 251 employees).

First of all, the level of understanding of certification concepts and agreement with basic certification statements were examined. Table 1 present the mean value of the level of understanding of certification. Both group of respondents understand the certification concept at adequate level.

The main differences are observed in answers regarding the certification statements.

Wood processing companies consider the ensurance of legal origins of wood (4.13) as a most important role of certification. Both groups of respondents strongly believe that the certification represent promotion of sustainable utilization of forest resource. Forest owners further consider the environmental responsibility as a very important role of certification.

158

Both groups of respondents see the certification as a tool to improve market access.

Respondents expressed the lowest level of agreement with the statements regarding the economic issues and those related to the improvement of the internal efficiency that forest certification may bring to companies. On the other hand, wood processing companies consider the ensurance of legal origin of wood as a most important statement of certification.

Certification as a tool for the improvement of market access, communication, or the internal company efficiency was considered as the least important. Table 2 shows the result of all certification statements for the both forest owners and wood processing companies.

Tab. 1 Understanding of certification concept.

Forest owners

Understanding of certification Understanding of forest management (SFM) concept Understanding of PEFC objectives Understanding of FSC objectives

4.11 3.76 3.12

CoC certified companies Mean 3.56 3.94 3.92

Tab. 2 Level of agreement with certification statements.

CoC certified companies

Forest owners

Certification statements

Mean Promotion of sustainable utilisation of forest resource Prevention from illegal logging Improvement in efficiency of corporate management Improvement of communication with customers Improvement of market access Improves management efficiency Commitment to environmental responsibility Improves external company image Improves forest management practices Increases profit margins Ensurance of legal origin of wood Confidentiality in sourcing of timber Improvement in efficiency of internal material flow systems Improvement in efficiency of corporate management

3.90 3.05 3.01 3.16 3.27 3.01 4.03 4.02 3.61 2.87 n/a n/a n/a n/a

4.08 3.60 3.42 3.72 3.54 3.79 n/a n/a n/a n/a 4.13 3.56 2.94 2.89

The second section of the survey was focused on the expectations motivating forest owners and CoC certified companies in the wood supply chain to implement certification.

As the table 3 shows, the improvement of external company image was identified as the most agreed statement by all respondents. Other expectations motivating forest owners to be certified were linked to the demonstration of sustainable forest management practices, commitment to environmental issues, improvement of forest management, and the search to obtain new customers. Expectations such as increase profit margins and increase sales volume were marked as the least motivating factor. In contrast the very important reasons why wood processing companies enter the certified market were the expectations of increase of sales (3.63), penetrating new markets (3.53), commitment to environmental issue (3.52) and seeking to expand market share (3.44). The least significant reasons to enter the certified market were the seeking to expand market share and to diversify product line.

When differences in certification schemes were examined the Mann-Whitney U test proved the significant differences in the factor focused on the demonstration of sustainable forest management practices (U = 162.0, p = 0.001), where the PEFC certificate holders are 159

more convinced than FSC ones. Further results show that the PEFC certified companies are more identified with the aforementioned benefits of forest certification than the FSC certified forest owners. This significant difference were observed only in the case of forest owners.

On the other hand, in the case of CoC certified companies, there was a significant influence of certification scheme on the statement relating to penetration of new markets identified (FSC companies are more motivated to enter certified forest products market than PEFC).

Tab. 3 The level of agreement with main motivation factors of implementation of certification.

CoC certified companies

Forest owners

Motivation factors

Mean Improvement of external company image

4.30

4.11

Commitment to environmental issues

4.14

3.52

Demonstration of sustainable forest management practices

4.23

n/a

Seeking to increase sales volume

2.68

3.63

Seeking to increase profit margins

2.73

2.98

Seeking to obtain new customers

3.03

n/a

Improvement of forest management

3.57

n/a

Penetrate new markets

n/a

3.53

Seeking to expand market share

n/a

3.44

Seeking to diversify product line

n/a

2.58

The last area of the analysis was oriented towards difficulties regarding certification.

Forest owners consider the compliance with the certification criteria (2.60) as the most difficulty issue. Other important difficulties regarded the administrative burden associated with the performance of internal audit (2.51), and the implementation of documentation (2.50). In the same way, the higher cost related to certification (2.45) were identified as a difficulty. Other certification related issues, for example, external audit (2.27), personnel and training (2.22), or communication with the group entity (1.85) seems to be perceived as less problematic. In the case of CoC certified companies the respondentsâ&#x20AC;&#x2122; perception of the problems connected to the purchase of certified products were also examined. These are connected to the insufficient quantity (2.81) of certified products followed by small margins (2.81) and overpriced inputs (2.80).

DISSCUSION Based on the results, it can be stated that both groups (forest owners and CoC certified companies) of respondents well understand the meaning of certification and the role of forest certification. Both groups of respondents primarily consider the certification as a noneconomic tool and subsequently they understood it as an economic tool. Their opinions were also similar in many other cases. Both perceive certification as a tool for the improvement of forest management, ensurance of the sustainability of forest management and last, but not least, determination of the origin of the harvested wood. Moreover, the two monitored groups of respondents understand certification also as a tool for improving the image of the company and improvement of communication with customers and as a way to penetrate new markets. Our results correspond with the conclusions of another authors (OWARI et al. 2006; RICKENBACH and OVERDEVEST 2006). Their research shows that forest certification is 160

consider as tool of indicating a company sense of responsibility, for keeping market share and for selling products.

Both groups tend to have same incentives for entering the certification process. These motives are very closely related to the level of certification perception. However, the order of importance of these incentives is opposite in case of the certified forest owners and certified companies. Forest owners enter certification in order to improve forest management, ensure sustainability of forest management and to deal with questions regarding the current environmental issues. On the other hand, it is natural that their incentives are also oriented towards economic and marketing areas. Decision of forest owners to enter the certification process is also conditioned by expectations to obtain new customers, increase profit and sales volume. In contrast, wood processing companies enter to the certification system with clear motives, which are primarily oriented inward the company. Therefore, their basic incentives to enter the certification system are to increase sales, penetration to the new markets or to seek expansion of market share. In contrary, environmental motives are less important for these companies. Based on the study by TUPPURA (2016) many companies enter the certification system to penetrate new markets, gain price premium for certified goods and obtain new customers. Similarly, the results of China´s study suggested that the expected price premium from certified timber is the most important motive for forest owners. Other motives such as increased timber growth, public recognition and environmentally friendly harvest were insignificant (TIAN et al. 2018).

These results are very similar to our current study in Slovakia. Our results are further supported by study of GALATI (2017) who determined the main motivation factor to the enter to certification system in Italy. This research show that the main motivations are those to meet the demand expressed by consumers and the aim to increase the market competitiveness.

CONCLUSION The objective of the research was to compare the opinions of the forest owners and CoC certified companies, particularly aimed at the analysis of the understanding of the concept and the role of the certification, expectations of companies following from implementation and difficulties connected to certification. In general, positive aspects of certification are better valued by respondents than its shortcomings. The following conclusions can be drawn: • respondents (forest owners and CoC certified companies) have a high level of understanding of the certification concept and perceive certification as a tool for the improvement of forest management, ensurance of the sustainability of forest management and last, but not least, determination of the origin of the harvested wood; • forest owners and CoC certified companies tend to have same incentives for entering the certification process. However, the order of importance of these incentives is opposite. Forest owners enter certification in order to improve forest management, ensure sustainability of forest management and to deal with issues regarding the current environmental problems. In contrast, CoC certified companies enter to the certification system with clear motives, which are primarily oriented inward the company; • forest owners do not consider certification costs as the main problem in contrast to the system implementation and its maintenance as well as external audit performance. On the other hand, CoC certified companies consider the key problems

161

related to the certified supply chain connected to the sufficient quantity of certified forest products, low margins, and overpriced certified material inputs; REFERENCES AGUILAR, F. X., VLOSKY, R. P. 2007. Consumer willingness to pay price premiums for environmentally certified wood products in the U.S. In Forest Policy and Economics, 2007, Vol. 9, Issue 8, p. 1100-1112.

AGUILAR, F. X., CAI, Z. 2010. Conjoint effect of environmental labelling, disclosure of forest of origin and price on consumer preferences for wood products in the US and UK. In Ecological Economics, 2010, Vol. 70, Issue 2, p. 308–316.

AULD, G., GULBRANDSEN, L. H., MCDERMOTT, C. L. 2008. Certification Schemes and the Impacts on Forests and Forestry. In Annual Review of Environment Resources, 2008, Vol. 33, p. 187–211.

BUTTERFIELD, R. P., HANSEN, E., FLETCHER, R., NIKINMAA, H. 2005. Forest Certification and Small Forest Enterprises: Key trends and impacts - benefits and barriers. Forest Trends and the Rainforest Alliance, 2005, Washington, DC. CABARLE, B., HRUBES, R. J., ELLIOT, C., SYNNOTT, T. 1995. Certification Accreditation: The Need for Credible Claims. In Journal of Forestry, 1995, Vol. 93, Issue 4, p. 12–16.

CARLSEN, K., HANSEN, C. P., LUND, J. F. 2012. Factors affecting certification uptake - Perspectives from the timber industry in Ghana. In Forest Policy and Economics, 1995, Vol. 25, p. 83–92.

CUBBAGE, F. W., MORE, S.; HENDERSON, T., ARAUJO, M. M. F. C. 2009. Costs and Benefits of Forest Certification in the Americas. In Natural Resources: Management, Economic Development and Protection, 2009; p. 155–185.

CUBBAGE, F., DIAZ, D., YAPURA, P., DUBE, F. 2010. Impacts of forest management certification in Argentina and Chile. In Forest Policy and Economics, 2010, Vol. 12, Issue 7, p. 497–504.

DURST, P. B., MCKENZIE, P. J., BROWN, C. L., APPANAH, S. 2006. Challenges facing certification and eco–labelling of forest products in developing countries. In International Forestry Review, 2006, Vol. 8, Issue 2, p. 193–200.

EFI. 2011. Support study for development of the non-legislative acts provided for in the Regulation of the European Parliament and of the Council laying down the obligations of operators who place timber and timber products on the market. Final report. [online]. [cit. 2017-10-4]. Available online: <http://ec.europa.eu/environment/forests/pdf/EUTR-Final_Report.pdf>.

EUROPEAN COMMISSION. 2016. Regulation (EU) No 995/2010 of the European Parliament and of the Council of 20 October 2010 laying down the obligations of operators who place timber and timber products on the market. [online]. [cit. 2017-10-4]. Available online: <http://eur-lex.europa.eu/legalcontent/EN/TXT/ PDF/?uri=CELEX:32010R0995&from=EN>.

GALATI, A., GIANGUZZI, G., TINERVIA, S., CRESCIMANNO, M., LA MELA VECA, D. S. 2017.

Motivations, adoption and impact of voluntary environmental certification in the Italian Forest based industry: The case of the FSC standard. In Forest Policy and Economics, 2017, Vol. 83, p. 169–176.

GILANI, H. R., KOZAK, R. A., INNES, J. L. 2016. The State of Innovation in the British Columbia Value-Added Wood Products Sector: The Example of Chain of Custody Certification. In Canadian Journal of Forest Research, 2016, Vol. 46, p. 106775.

HALALISAN, A. F., MARINCHESCU, M., POPA, B., ABRUDAN, I. V. 2013. Chain of Custody certification in Romania: profile and perceptions of FSC certified companies. In International Forestry Review, 2013, Vol. 15, Issue 3, p. 305–314.

HANSEN, E.. PUNCHES, J. 1999. Developing Markets for Certified Forest Products: A Case Study of Collins Pine Company. In Forest products journal, 1999, Vol. 49, Issue 1, p. 30–35.

HUBBARD, S, BOWE, S. 2004. Putting Value on Certified Forest Products: Views from Wisconsin. In Wood Wood Prod, 2004, p. 57–62.

JAYASINGHE, P.S., ALLEN, D., BULL, G. Q., KOZAK, R. A. 2007. The status of forest certification in the Canadian value-added wood products manufacturing sector. In Forestry Chronicles, 2007, Vol. 83, Issue 1, p. 113–125.

162

JOHNSON, D., WALCK, C. 2004. Certified Success: Integrating Sustainability into Corporate Management Systems. In Journal of Forestry, 2004, Vol. 102, Issue 5, p. 32–39.

KOZAK, R. A., COHEN, D. H., LERNER, J., BULL, G. Q. 2004. Western Canadian consumer attitudes toward certified value-added wood products: an exploratory assessment. In Forest Products Journal, 2004, Vol. 54, p. 21–24.

LEWIS, R. A., DAVIS, S. R. 2015. Forest certification, institutional capacity, and learning: An analysis of the impacts of the Malaysian Timber Certification Scheme. In Forest Policy and Economics, 2015, Vol. 52, p. 18–26.

MACDICKEN, K. G., SOLA, P., HALL, J. E., SABOGAL, C., TADOUM, M., DE WASSEIGE, C. 2015.

Global progress toward sustainable forest management. In Forest Ecology and Management, 2015, Vol. 352, p. 47–56.

MIKULKOVÁ, A., HÁJEK, M., ŠTĚPÁNKOVÁ, M., ŠEVČÍK, M. 2015. Forest certification as a tool to support sustainable development in forest management. In Journal of Forest Science, 2015, Vol. 61, Issue 8, p. 359–368.

MILES, M. P., COVIN, J. G. 2000. Environmental marketing: a source of reputational, competitive, and financial advantage. In Journal of Business Ethics, 2000, Vol. 23, Issue 3, p. 299– 311.

MOORE, S. E., CUBBAGE, F., EICHELDINGER, C. 2012. Impacts of Forest Stewardship Council (FSC) and Sustainable Forestry Initiative (SFI) Forest Certification in North America. In Journal of Forestry, 2012, Vol. 110, Issue 2, p. 79–88.

NAKAMURA, M., TAKAHASHI, T., VERTINSKY, I. 2001. Why Japanese firms choose to certify: A study of managerial responses to environmental issues. In Journal of Environmental Economics and Management, 2001, Vol. 42, Issue 1, p. 23–52.

NUSSBAUM, R., SIMULA, M. 2005. The Forest Certification Handbook; Earthscan: London, 2005.

ISBN 1844071235.

OWARI, T., JUSLIN, H., RUMMUKAINEN, A., YOSHIMURA, T. 2006. Strategies, functions and benefits of forest certification in wood products marketing: Perspectives of Finnish suppliers. In Forest Policy Economics, 2006, Vol. 9, Issue 4, p. 380– 391.

OWARI, T., SAWANOBORI, Y. 2006. Market Benefits of Chain of Custody Certification Perspectives of Japanese Suppliers. In Forest Resource Management and Mathematical Modelling, 2006, Vol 7, p. 121–132.

OZANNE, L. K., VLOSKY, R. P. 2003. Certification from the U.S. consumer perspective: a comparison from 1995 and 2000. In Forest Products Journal, 2003, Vol. 53, p. 13–21.

OZANNE, L. K., VLOSKY, R. P. 1997. Willingness to pay for environmentally certified wood products: a consumer perspective. In Forest Products Journal, 1997, Vol. 47, p. 39–48.

PALUŠ, H. 2013. Trh a obchod s drevom a výrobkami z dreva (Wood and Wood Products Market and Trade), Zvolen: Technical University in Zvolen, 2013. ISBN 978-80-228-2587-0.

PALUŠ, H., KAPUTA, V. 2009. Survey of attitudes towards forest and chain of custody certification in the Slovak Republic. In Drewno, 2009, Vol. 52, Issue 182, p. 65–81.

PALUŠ, H., PAROBEK, J., DUDÍK, R., ŠUPÍN, M. 2017. Assessment of chain-of-custody certification in the Czech and Slovak Republic. In Sustainability, 2017, Vol. 9, p. 1–13.

PALUŠ, H., PAROBEK, J., ŠULEK, R., LICHÝ, J., ŠÁLKA, J. 2018a. Understanding Sustainable Forest Management Certification in Slovakia: Forest Owners’ Perception of Expectations, Benefits and Problems. In Sustainability, 2018, Vol. 10, Issue 7, 17 p.

PALUŠ, H., PAROBEK, J., VLOSKY, R. P., MOTIK, D., OBLAK, L., JOŠT, M., GLAVONJIĆ, B., DUDÍK, R., WANAT, L. 2018b. The status of chain-of-custody certification in the countries of Central and South Europe. In European Journal of Wood and Wood Products, 2018, Vol. 76, Issue 2, p. 699– 710.

PECK, T. 2001. The International Timber Trade. Cambridge: Woodhead Publishing Ltd, 2001. ISBN 9781855731905.

PEFC. 2017. PEFC Chain of Custody Certification. The Key to Sell Certified Products. [online]. [cit.

2017-10-4-]. Available online: <https://www.pefc.org/images/documents/brochures/PEFC_Chain_of_Custody_Certification.pdf>.

PERERA, P., VLOSKY, P. R., HUGHES, G., DUNN, M. A. 2007. What do Louisiana and Mississippi Nonindustrial Private Forest Landowners Thing about Forest Certification?. Southern Journal of Applied Forestry, 2007, Vol. 31, Issue 4, p. 170–175.

163

PERERA, P., VLOSKY, R. P., DUNN, M. A., HUGHES, G. 2008. U.S. home-center retailer attitudes, perceptions and behaviors regarding forest certification. In Forest Products Journal, 2008, Vol. 58, p. 21–25.

POTKAŃSKI, T., WANAT, L., CHUDOBIECKI, J. 2011. Leadership in time of crisis or crisis of leadership? Implications for regional development. In Intercathedra, 2011, Vol. 27, p. 45–52.

RAMETSTEINER, E., SIMULA, M. 2003. Forest certification - An instrument to promote sustainable forest management? In Journal of Environmental Management, 2003, Vol. 67, Issue 1, p. 87–98.

RICKENBACH, M.; OVERDEVEST, C. 2006. More than Markets: Assessing Forest Stewardship Council (FSC) Certification as a Policy Tool. In Journal of Forestry, 2006, Vol. 104, Issue 3, p. 143– 147.

SIRY, J. P.; CUBBAGE, F. W.; AHMED, M. R. 2005. Sustainable forest management: Global trends and opportunities. Forest Policy and Economics, 2005, Vol. 7, Issue 4, p. 551–561. ŠÁLKA, J., DOBŠINSKÁ, Z., SARVASOVA, Z., ŠTĚRBOVÁ, M., PALUŠ, H. 2017. Lesnícka politika (Forest Policy); Zvolen: Technical University in Zvolen, 2017. ISBN 978-80-228-3008-9.

TAKAHASHI, T. 2001. Why firms participate in environmental voluntary initiatives: Case studies in Japan and Canada, Vancouver: University of British Columbia, 2001.

TIAN, N., LU, F., JOSHI, O., POUDYAL, N. C. 2018. Segmenting Landowners of Shandong, China Based on Their Attitudes towards Forest Certification. In Forests, 2018, Vol. 9, p. 1–14.

TOLUNAY, A., TÜRKOĞLU, T. 2014. Perspectives and Attitudes of Forest Products Industry Companies on the Chain of Custody Certification: A Case Study from Turkey. In Sustainability, 2014, Vol. 6, p. 857–871.

TRISHKIN, M., LOPATIN, E., KARJALAINEN. T. 2014. Assessment of motivation and attitudes of forest industry companies toward forest certification in northwestern Russia. In Scandinavian Journal of Forest Research, 2014, Vol. 29, p. 283–293.

TUPPURA, A., TOPPINEN, A., PUUMALAINEN, K. 2016. Forest Certification and ISO 14001: Current State and Motivation in Forest Companies. In Business Strategy and the Environment, 2016, Vol. 25, p. 355368.

VEISTEN, K. 2007. Willingness to pay for eco-labelled wood furniture: choice-based conjoint analysis versus open-ended contingent valuation. In Journal of Forest Economics, 2007, Vol. 13, Issue 1, p.

29–48.

VIDAL, N., KOZAK, R., COHEN, D. 2005. Chain of custody certification: an assessment of the North American solid wood sector. In Forest Policy and Economics, 2005, Vol. 7, Issue 3, p. 345–355.

VLOSKY, R. P., OZANNE, L. K. 1998. Environmental certification of wood products: The U.S. manfacturers’ perspective. In Forest Products Journal, 1998, Vol. 48, p. 21–26.

VLOSKY, R., GAZO, R., CASSENS, D., PERERA, P. 2009. Changes in value-added wood product manufacturer perceptions about certification in the United States from 2002 to 2008. In Drvna Industrija, 2009, Vol. 60, p. 89–94.

WWF. 2000. The Forest Industry in the 21st Century. Report prepared by the WWF’s Forests for Life Campaign. Branksome House, Godalming, Surrey.

YAMAMOTO, Y., TAKEUCHI, K., SHINKUMA, T. 2014. Is there a price premium for certified wood?

Empirical evidence from log auction data in Japan. In Forest Policy and Economics, 2014, Vol. 38, p. 168–172.

ACKNOWLEDGEMENTS The authors are grateful for the support of the Scientific Grant Agency of the Ministry of Education, Science, Research and Sport of the Slovak Republic, and the Slovak Academy of Sciences, Grant No. 1/0666/19 “Determination of the development of a wood-based bioeconomy”, Grant No. 1/0674/19 “Proposal of a model for the eco-innovation integration into the innovation process of companies in Slovakia in order to increase their performance” and the Slovak Research and Development Agency, Grant APVV-14-0869 “Research of the utilization of wood as a renewable raw material in the context of a green economy”.

164

AUTHOR ̓ S ADDRESS Hubert Paluš Ján Parobek Michal Dzian Samuel Šimo-Svrček Technical University in Zvolen Faculty of Wood Sciences and Technology T. G. Masaryka 24 960 53 Zvolen Slovakia palus@tuzvo.sk parobek@tuzvo.sk michal.dzian@tuzvo.sk xsimosvrcek@is.tuzvo.sk Martina Krahulcová Faculty of Forestry Technical University in Zvolen Department of Marketing, Trade and World Forestry T. G. Masaryka 24 960 53 Zvolen Slovakia xkrahulcovam@is.tuzvo.sk

165