Warsaw University of Life Sciences

Annals Warsaw University of Life Sciences

Forestry and Wood Technology No 108 Warsaw 2019 Contents:

JOANNA WACHOWICZ, MARCIN ROSIŃSKI, RADOSŁAW ZIELIŃSKI, TOMASZ TRUSZKOWSKI “Influence of compaction and degassing on the properties of submicron WCCo produced by the PPS method.” 5

PIOTR BEER, PAWEŁ PACEK, IZABELA BURAWSKA-KUPNIEWSKA, SYLWIA OLEŃSKA, ANNA RÓŻAŃSKA “Influence of alder (Alnus glutinosa Gaerthn.) veneers on selected mechanical properties of layered pine (Pinus sylvestris L.) composites.” 13

IZABELA BETLEJ “Studies on the diversity of substrate composition in the culture medium of Kombucha microorganisms and its influence on the quality of synthesized cellulose.” 21

ŁUKASZ MATWIEJ, ROBERT KŁOS, MIROSŁAW BONOWSKI “Design of a snap connector to connect panel elements.” 26

KRZYSZTOF WIADEREK, IWONA WIADEREK, JULIA LANGE “Impact of upholstered furniture seat structures on the long-term use comfort.” 39

MAREK BARLAK, JACEK WILKOWSKI, KAROL SZYMANOWSKI, PAWEŁ CZARNIAK, PIOTR PODZIEWSKI, ZBIGNIEW WERNER, JERZY ZAGÓRSKI, BOGDAN STASZKIEWICZ “Influence of the ion implantation of nitrogen and selected metals on the lifetime of WC-Co indexable knives during MDF machining.” 45

IZABELA BETLEJ, KRZYSZTOF J. KRAJEWSKI “Bacterial cellulose synthesis by Kombucha microorganisms on a medium with a variable composition of nutrients.” 53

WOJCIECH KORYCIŃSKI, KRZYSZTOF J. KRAJEWSKI ,PAWEŁ KOZAKIEWICZ “Resistograph investigation of Scots pine wood utility poles in the State Museum at Majdanek.” 58

MAREK BARLAK, JACEK WILKOWSKI, ZBIGNIEW WERNER “Modelling of nitrogen implantation processes into WC-Co indexable knives for wood-based material machining using ion implanters with or without direct ion beam.” 68

JACEK WILKOWSKI, MAREK BARLAK, ROMAN BÖTTGER, ZBIGNIEW WERNER “The effect of nitrogen ion implantation on nano-scale hardness and elastic modulus of WC- Co indexable knives for wood materials machining.” 79

ADAM MAJEWSKI “Furniture use safety at early design stage.” 84

KATARZYNA ŚMIETAŃSKA, PIOTR PODZIEWSKI “A human quality control system in furniture manufacturing – a pilot study.” 93

KATARZYNA ŚMIETAŃSKA, PIOTR PODZIEWSKI “The use of the Krippendorff’s coefficient in determining intra-rater reliability in human visual quality control of furniture manufacturing processes.” 97 ALEKSANDRA NISKA, EDYTA MAŁACHOWSKA “The effect of the addition of primary fibers on the papermaking ability of wastepaper.” 104

EMILIA GRZEGORZEWSKA, JUSTYNA BIERNACKA, IZABELA PODOBAS “Trends in employment and labour productivity in the woodworking industry in selected EU countries.” 111

JUSTYNA BIERNACKA, EMILIA GRZEGORZEWSKA, IZABELA PODOBAS “The trends in employment and labour productivity in the pulp and paper industry in the selected EU countries.” 119

PIOTR BORYSIUK, ANNA TETELEWSKA, RADOSŁAW AURIGA, IZABELLA JENCZYK-TOŁŁOCZKO “The influence of temperature on selected strength properties of furniture particleboard.” 128

ADAM KRAJEWSKI, PIOTR WITOMSKI, ANNA OLEKSIEWICZ “Sinoxylon anale as wood borer insect and its parasites.” 135

MARTA GNACIŃSKA, ANDRZEJ RADOMSKI “The study of the impact of in situ polymerisation with styrene or acrylates on water absorbability and swelling of thermomechanically densified poplar wood.” 140

DARIA KAŹMIERCZAK, ANDRZEJ RADOMSKI “Studies on the suitability of oxidizing agents for discolouring lime and poplar wood in the first stage of transparent wood forming process.” 148 -

Scientific council:

Miroslav Rousek (Czech Republic) Ján Sedliačik (Slovakia) Nencho Deliiski (Bulgaria) Ladislav Dzurenda (Slovakia) Olena Pinchewska (Ukraine) Paik San H`ng (Malaysia) Loredana Badescu (Romania) Włodzimierz Prądzyński (Poland) Iskandar Alimov (Uzbekistan) Kazimierz Orłowski (Poland)

Board of reviewers:

Bogusław Andres Teresa Kłosińska Andrzej Antczak Grzegorz Kowaluk Bogusław Andres Paweł Kozakiewicz Piotr Beer Adam Krajewski Izabela Betlej Krzysztof Krajewski Justyna Biernacka Sławomir Krzosek Piotr Boruszewski Agnieszka Laskowska Piotr Borysiuk Mariusz Mamiński Izabela Burawska-Kupniewska Mateusz Niedbała Ewa Dobrowolska Piotr Przybysz Michał Drożdżek Anna Różańska Jarosław Górski Jacek Wilkowski Emila Grzegorzewska Piotr Witomski Agnieszka Jankowska Marcin Zbieć

Warsaw University of Life Sciences Press

e-mail: [email protected] SERIES EDITOR Ewa Dobrowolska – Editor in Chief ISSN 1898-5912 Anna Sekrecka-Belniak Mateusz Niedbała

PRINT: Drukarnia POZKAL Spółka z o.o. Spółka komandytowa 88-100 Inowrocław, ul. Cegielna 10 – 12

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 5-12 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Influence of compaction and degassing on the properties of submicron WCCo produced by the PPS method

JOANNA WACHOWICZ1, MARCIN ROSIŃSKI2, RADOSŁAW ZIELIŃSKI3, TOMASZ TRUSZKOWSKI4 1Department of Mechanical Processing of Wood, Warsaw University of Life Science – SGGW 2GeniCore, Poland 3The Institute of Advanced Manufacturing Technology, Poland 4Faculty of Materials Science and Engineering, Warsaw University of Technology

Abstract: Influence of compaction and degassing on the properties of submicron WCCo produced by the PPS method. The present study is concerned with the effect of the parameters of the degassing operation (temperature, load and heating rate) conducted at the initial stage of the Pulse Plasma Sintering (PPS) process and the sintering temperature at the final stage of the process, on the properties and microstructure of WCCo with a 6wt% cobalt content sintered by this method. The results of the study have shown that when the heating rate is too high, the material obtained is porous. In most experiments, the sintering temperature of 1050°C appeared to be too low to obtain WCCo composites with density close to the theoretical value (GT). Sintering at the temperature increased to 1070°C yielded sinters with density above 99%GT, with hardness of about 1900 -3/2 HV30 and fracture toughness KIC=9.3 MNm .

Keywords: Sintering; PPS (Pulse Plasma Sintering); Tungsten Carbide (WCCo); Composite

INTRODUCTION Although the composites of tungsten carbide with cobalt have long been known, they are still the most commonly used tool material, and more than a half of cutting tools at present are made of it [1]. Sintered carbides achieved this great success in the tool industry thanks to their properties, such as high hardness, very good frictional wear resistance, good thermal and electric conductivities, and high stability at elevated temperatures. Traditionally, tungsten carbide has been sintered freely at a temperature between 1400 and 1500ºC, depending on the cobalt content. The total duration of the sintering process, apart from the time taken by the necessary disintegration and mixing operations, amounts to a dozen or so hours [2]. The last two decades have seen a significant development of the sintering methods that use an electric field to activate the process. This permits an important reduction of the sintering time to less than 20 min (even several minutes). In the literature, these methods are known as EDC (Electric Discharge Compaction). Just as in conventional hot isostatic pressing (HIP), the sintering process is conducted under a uniaxial load, but obviates the drawbacks of HIP which requires a high temperature, a long process time, and in which the efficiency of heating the powder being consolidated is low. The electric-field activated processes differ from HIP in the way in which the heat energy is transmitted to the powder to be consolidated. PPS (Pulse Plasma Sintering) is a modern method belonging to the group of the EDC methods, that permits the sintering process to be conducted at a lower temperature and in a shorter time (about 10 min) [3,4]. The idea underlying this method consists in using electric pulses to heat the powder that is being pressed. The heating is realized through the Joul’s heat being dissipated at the contacts between the individual powder particles. Pulses of electric current are generated by discharging a 300 µF battery of capacitors charged to a voltage of several kV.

MATERIALS The WC6Co (wt%) composites were produced from mixtures of tungsten carbide (94wt%) powder with an average grain size of 0.4µm and ultra-fine-grained cobalt powder (6wt%). SEM images of these powders in the initial state are shown in Fig. 1. The powders were dry mixed in a Turbula-type mixer using carbide balls with a 1:1 ball-to-powder mass ratio. The mixing time was 5h.

a) b) Figure 1. SEM images of the powders (a) WC, (b) Co

The mixture of powders was subjected to initial compaction in a graphite die under a pressure from 5 to 100MPa, using a hand-operated press. The samples prepared thusly were sintered in a PPS apparatus [5]. All the sintering processes were conducted under a reduced pressure of 5x10-5 mbar in two stages. The first stage included degassing (aimed at removing the absorbed gasses), and the next stage consists of sintering. Degassing was conducted at a temperature (T1) between 600 and 1000oC under a load between 10 and 100MPa with heating at a rate from 700 to 1100°C/min. In the next stage of the process, the sintering temperature (T2) ranged from 1050 to 1100°C. Then, the samples were cooled to room temperature in vacuum of 5x10-5 mbar at a rate of 160°C/min under a load of 100MPa. The two stages of the process are illustrated schematically in Fig.2.

Figure 2. Schematic representation of the course of the sintering process

The microstructure of the sintered samples was observed in a HITACHI S-3500N scanning electron microscope. The images of the microstructures were used for determining the

6 average WC grain size in the sintered samples [6,7]. Their density was measured by the Archimedes method using a Radwag balance, and their hardness - by the Vickers method in a Matsuzawa VMT-7S hardness meter under a load of 294N (HV30).

RESULTS Effect of the degassing temperature and the heating rate on the properties of WCCo composites Figure 3 shows the relative density of the samples sintered at a temperature of 1050°C in dependence on the degassing temperature (600, 900, 1000°C) and the heating rate (700, 1100°C/min). At all the degassing temperatures examined in the present experiments, the samples heated at a rate of 1100°C/min had the relative density below the theoretical value. The highest density was achieved in the samples that were degassed at a temperature of 900°C. The reduction of the heating rate to 700°C/min resulted in a density that slightly exceeded the density of the samples heated with a rate of 1100°C/min, but still much below the theoretical value.

Figure 3. (a) Effect of the degassing temperature and heating rate on the relative density in samples sintered at 1050°C

Figure 4. Effect of the degassing temperature and heating rate on hardness in samples sintered at 1050°C

The measured values of hardness show a similar tendency (Fig.4), except in the samples degassed at 900°C with the heating rate of 700°C/min, which had the highest density and the best hardness (above 1900HV30) with the narrowest spread of the measured hardness

7 values (expressed as the standard deviation), compared to those of the other samples sintered at 1050°C. Figure 5a shows examples of the microstructure of a fracture of the WCCo composites heated (during the degassing stage) at a rate of 1100°C/min and sintered at a temperature of 1050oC. We can see that the crystalline faces of the WC grains are not fully shaped, with the shapes being similar to those in the starting WC powder (see Fig.1a), and that numerous pores are present in between the grains. Fig.5b shows the microstructure of a fracture of the WCCo composite degassed at a temperature of 900oC with a heating rate of 700°C/min, and sintered at a temperature of 1050°C. The WC grains are well shaped with sharp edges and cobalt distributed uniformly along them forming the so-called paths. No porosity can be observed. The structure is compact and homogeneous on the entire cross-section of the sample, which is confirmed by the hardness distribution.

a) b) Figure 5. SEM images of a fracture of the WCCo composites: (a) WCCo composite sintered at 1050oC, degassed at 900oC at a heating rate of 1100oC/min, (b) WCCo composite sintered at 1050oC, degassed at 900oC at a heating rate of 700oC/min

As our experiments have shown, heating at a rate of 1100°C/min seems to be too rapid for the material obtained to be dense, compact, and homogeneous. The processes of oxides reduction, evaporation, and condensation taking place during the first stage of the process and resulting in an increase of the inter-particle contact surface, appear to be hampered too much. The material obtained under these conditions is porous and its hardness is low. In the composite degassed at 600 and 1000°C, slightly better density and hardness were obtained with the heating rate decreased to 700°C/min, but it was only the degassing temperature of 900oC and the heating rate of 700°C/min that ensured a dense and homogeneous material.

Effect of the load (at the first stage) and sintering temperature (the second stage) on the properties of WCCo composites

Figure 6 shows how the load applied at the degassing stage affects the density of composites sintered at 1050 and 1070°C. The degassing process was conducted at a temperature of 600°C at the heating rate of 700°C/min. As the load increased, the relative density also slightly increased (from 97.1% to 98.7% of the theoretical value) in the samples sintered at 1050°C, whereas it remained unaffected (above 99% of the theoretical value) in the samples sintered at 1070°C.

Figure 6. Effect of the sintering parameters (load at the first stage and temperature at the second stage) on the relative density (percent of the theoretical value) of WCCo composites

Figure 7 shows the effect of the load applied during the degassing stage and the temperature at the sintering stage on the hardness of the composites. The lowest hardness (1830HV30) was obtained in the composites sintered at a temperature of 1050°C with the degassing load of 10MPa. Under a greater load (at the same temperature), the hardness slightly increases. All the composites sintered at 1050°C show a wide spread of the measured hardness values, which can be attributed to their poor porosity. The samples sintered at a temperature of 1070°C have a relatively high hardness of 1880HV30 (similar tendency as in the case of density) and the spread of the measured hardness values is narrow. The properties of these samples do not depend on the degassing load.

Figure 7. Effect of the sintering parameters (load at the first stage and temperature at the second stage) on the hardness of WCCo composites

Effect of the sintering temperature (II stage of the process) on the properties of WCCo composites Figure 8 shows how the sintering temperature affects the hardness and density of the composites. The degassing process was conducted at a temperature of 600°C with a heating rate of 700°C/min under a load of 50MPa. The samples sintered at 1050oC had a low density of 97.1% GT. The increase of the sintering temperature to 1070°C results in the density increasing above 99%GT. As to the hardness, it is relatively high (1905HV30) in all the samples sintered at 1050°C, but the spread of its measured values is wide. Only the samples

9 sintered at 1070°C and 1100°C are characterized by high density and high hardness of about 1800 HV30.

Figure 8. Effect of the sintering parameters (temperature at the 2nd stage of the process) on the hardness and relative density

The slightly lower hardness of the samples sintered at a temperature of 1100°C may be attributed to the presence of small regions (visible on the fracture of Fig.9) in which the grains are larger. The average grain size in these samples is 0.54 µm, whereas that in the samples sintered at 1070°C is 0.44 µm (close to the average grain size in the starting powder which is 0.4 µm).

Figure 9. SEM image of a fracture of the WCCo composite sintered at 1100oC, degassed at 600oC under a load of 50MPa

Figure 10 summarizes our results since it shows the effect of the sintering temperature on density (expressed as a percentage of the theoretical value) of the material at various parameters of the two stages of the process. It can be seen that below the sintering temperature of 1070°C, the density decreases. The results obtained for the sample degassed at 900°C at a heating rate of 700°C/min are shown in black in the figure.

Figure 10. Effect of the sintering temperature on the density of the WCCo composites

CONCLUSIONS WC6Co composites were produced by the PPS method. The process was conducted in two stages: the material was first degassed under load and then sintered. The effect of the process parameters on the density and hardness of the composites was examined. It appears that, irrespective of the degassing parameters such as load, temperature, and heating rate, the sintering temperature of 1050°C is too low to obtain a compact dense composite except for the samples degassed at 900°C with a heating rate of 700°C/min. Sintering at a temperature of 1070°C permits to produce composites with a density close to the theoretical value. It also appears that the higher the heating rate, the higher is the porosity of the material obtained and the worse its homogeneity. In the WCCo composites sintered at a temperature of 1070°C, no effect of the load applied during degassing on the properties of the composites was observed.

REFERENCES

1. TILLMANN W., 2000: Trends and market perspectives for diamond tools in the construction industry, International Journal of Refractory metals & Hard Materials 18 2. SOARES E. et al., 2011: Microstructures and properties of submicron carbides sintered with conventional technologies. Journal of the American Ceramic Society 3. ROSIŃSKI M. et al., 2012: Synthesis and characterization of the diamond/copper composites obtained by the pulse plasma sintering (PPS) method, Diamond & Related Materials 27–28 4. ROSIŃSKI M. et al., 2011: WC/Ti composite material enriched with cBN particles produced by pulse plasma sintering (PPS), Key Engineering Materials, 484 5. MICHALSKI A. et al., 2007: "Nanocrystalline cemented carbides sintered by the pulse plasma method." International Journal of Refractory Metals and Hard Materials 153- 158 6. WEJRZANOWSKI T., 2000: Computer Assisted Quantitative Description of the Functionally Graded Materials”. M.Sc. thesis, Warsaw University of Technology, Warsaw

7. CHA SI et al.: 2001 Microstructure and mechanical properties of nanocrystalline WC- 10Co cemented carbides. Scripta Materialia 44(8-9): 1535-1539.

Streszczenie: Wpływ ciśnienia prasowania oraz parametrów etapu odgazowania na właściwości submikronowego WCCo, wytwarzanego metodą PPS. Węgliki spiekane są cenionym materiałem narzędziowym stosowanym między innymi do obróbki materiałów drewnopochodnych. Tradycyjnie otrzymuje się je z użyciem wysokich temperatur oraz długich czasów. W ostatnich latach nastąpił rozwój nowych metod konsolidacji materiałów proszkowych (określanych, jako EDC - Electric Discharge Compaction), które umożliwiają otrzymanie spieków o dobrych właściwościach mechanicznych w krótkim czasie i niskiej temperaturze. W niniejszej pracy przedstawiono wpływ parametrów etapu odgazowania (temperatury, nacisku oraz szybkości nagrzewania) na właściwości i mikrostrukturę WCCo, o zawartości 6% (wag.) kobaltu, spiekanych metodą PPS. Badania wykazały, iż zbyt duża szybkość nagrzewania prowadzi do otrzymania porowatego materiału. Zastosowanie temperatury spiekania 1070°C pozwoliło uzyskać spieki o gęstości powyżej 99% gęstości teoretycznej oraz o twardości na poziomie 1900 HV30. Średni współczynnik koncentracji -3/2 naprężeń KIC dla otrzymanych materiałów wynosił 9,3 MNm .

Corresponding author:

Joanna Wachowicz, Warsaw University of Life Science – SGGW Faculty of Wood Technology 159 Nowoursynowska St. 02-787 Warsaw, Poland email: [email protected]

ORCID ID: Wachowicz Joanna: 0000-0002-7942-3959

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 13-20 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Influence of alder (Alnus glutinosa Gaerthn.) veneers on selected mechanical properties of layered pine (Pinus sylvestris L.) composites

PIOTR BEER1, PAWEŁ PACEK2, IZABELA BURAWSKA-KUPNIEWSKA1, SYLWIA OLEŃSKA1, ANNA RÓŻAŃSKA1 1 Department of Technology and Entrepreneurship in Wood Industry, Institute of Wood Sciences and Furniture, Warsaw University of Life Sciences WULS-SGGW, Poland 2 Stolarstwo Pacek, Poland

Abstract: Influence of alder (Alnus glutinosa Gaerthn.) veneers on selected mechanical properties of layered pine (Pinus sylvestris L.) composites. The aim of the study was to analyse the influence of using hardwood veneers in the base layer on selected mechanical properties of composites made of coniferous veneers dedicated for flooring applications. The modulus of elasticity and stiffness at three-point bending were determined in static, dynamic and fatigue tests. All tested mechanical properties of pine-alder composites showed, to a different extent, higher values than composites with a base layer made only of pine veneers.

Keywords: floor, composite, mechanical properties, veneer, hardwood, softwood

INTRODUCTION Wood is a verified and popular material used in floor production. It is used in public spaces and homes. Due to the different purposes of the spaces, floors have different requirements:  durability of use,  easy maintenance,  aesthetic finish. Nowadays, floors can be constructed using wood composites. Wood composites can have a layered cross-shaped construction, which results in greater dimensional stability of the structure. In addition, the layered construction makes it possible to use various materials in the top and base layers in the form of thin wooden boards or veneers. There are two-layer structures made of top and base layers (www.jaf-polska.pl/) and three-layer structures in which, apart from the top layer, the base layer is composed of an inner layer and a bottom layer (https://www.barlinek.com.pl). In all these structures, the base layer provides the composites with dimensional stability as well as mechanical properties. The presented work deals with two-layer structures. The main research trend concerning floors relates to the surface properties of the top (face) layer. In layered composites, it is built of solid or glued wood, as two-strip or three- strip wood flooring. Hence, this kind of research is focused on testing the durability of the top layer and conducted only on separated face layer material (without other layers), for example: hardness (Heräjärvi 2004; Holmberg 2000), strength (Song-Young and Hon-Lin 1999) or top layer examination after material modifications (Grześkiewicz and Krawiecki 2008). Research on the physical and mechanical properties of floor composites for floating floors and sports floors is mainly conducted in view of their elastic properties (Makowski and Noskowiak 2016). The aim of our study was to analyse the influence of hardwood veneers used in the base layer on selected mechanical properties of composites made of coniferous veneers dedicated for flooring applications.

MATERIALS Layered composites were produced for the study. Thin oak (Quercus L.) boards were used as the top layer – A; while pine (Pinus sylvestris L.) and alder (Alnus glutinosa Gaerthn.) veneers were used for the base layer – B (Fig. 1). The base layers were made of wood without defects. The oak boards had a thickness of 3 mm ± 0.2. Pine veneers with a thickness of 2.5 mm and 1.5 mm ± 0.1 mm and alder with a thickness of 1.5 mm ± 0.1 mm were arranged in a cross-shape construction. The mean value of modulus of elasticity of pine wood is 12000 MPa and the static bending strength, 100 MPa. The mean value of the modulus of elasticity of alder wood is 9700 MPa, and the static bending strength, 97 MPa [https://www.itd.poznan.pl].

Figure 1. Composite with a two-layer structure: A − top layer, B − cross-shaped base layer

The veneer sheets were cut to 350 mm x 180 mm in such a way that the even layers had fibres arranged perpendicularly to the length of the composite (┴), hence the odd layers had fibres arranged in parallel to the length of the composite (=). The composites prepared in this way were arranged according to the scheme presented in Table 1. 11 samples from each group were prepared for static and dynamic bending. One composite from each group was used to conduct the fatigue test.

Table 1. Layout scheme of individual veneers distinguished by thickness Oak thin Pine Pine Pine Pine Pine Pine Sample’s board veneer veneer veneer veneer veneer veneer mark [mm] [mm] [mm] [mm] [mm] [mm] [mm] = ┴ = ┴ = ┴ = 1.5 2.5 1.5 2.5 1.5 2.5 Alder Alder Alder Pine 3 veneer veneer veneer [mm] [mm] [mm] Pine/Alder 3 1.5 2.5 1.5 2.5 1.5 2.5

Urea-formaldehyde glue was used to bond the composite elements. The adhesive application was 180 g/m2. Pressing parameters were set as follows:  time 12 min,  temperature 120ºC,  pressure 1.2 MPa. After gluing, the composites were seasoned. The material was stored at a temperature of 22ºC, and 50% relative humidity, for 28 days. After this time, the geometry of composites was verified. The geometry verification of the composites was carried out using a mechanical plotter (Fig. 2). The device verifies the geometry in a vertical position. The composites were

14 supported at the base layer and the measurement was carried out on the top layer. The virtual measuring plane was fixed at three base points. The fourth point was adjustable to stabilize the test material during measurement. Measurement accuracy was 0.1 mm/m. Deviations from the theoretical plane of the composites’ surface were +/- 1.5 mm. a/ b/

Figure 2. Geometry verification of the composites: a - measuring probe path diagram, b – during measurement

METHODS The modulus of elasticity for static, dynamic and fatigue tests was specified based on the PN-EN 310:1994 standard, adapted to the samples’ dimensions (Boruszewski et al. 2013). The experiment was carried out on a TiraTest 2300 testing machine in the mode of a three- point test. The spacing of supports for the composites was 310 mm (Fig. 3).

a/ b/

Figure 3. Modulus of elasticity testing: a - testing machine during the test, b – sample of load diagram for desplacement

The measurements were made in the field of elastic deformations of composites, that is within Hooke's law. The formula (1) used to calculate the modulus of elasticity in bending tests was:

(1) where: Em [MPa] – modulus of elasticity, L1 [mm] – distance between support centres, F2-F1 [N] – load increase along the straight section of the curve,

F1 [N] – 10% of load, F2 [N] – 40% of load, b [mm] – sample width, a2 – a1 [mm] – increase of the desplacement arrow in the middle of the composite.

Determination of the modulus of elasticity for dynamic bending was the key to determine whether the composite can be used in floating floor conditions and in the field of high dynamic loads such as parquets in sports halls. The tests consisted in dynamically applying a load to the composite. The speed of lowering and raising the traverse was set to 9 mm/s. In the fatigue tests, the traverse speed was set to 9 mm/s. The tested material was subjected to 50 dynamic load cycles. The study was analysed on the first and then on every tenth cycle.

Stiffness is the ability of a material, connection or structure to counteract deformations caused by external loads. It depends on the shape of the element, its elastic properties, type of load and boundary conditions. The stiffness of composites under static and dynamic loading conditions was determined from the formula (2) (PN-EN 1995-1-1):

(2) where: k [MNmm2] – stiffness, Em [MPa] – modulus of elasticity, b [mm] – sample width, t [mm] – sample thickness.

RESULTS Figure 4 presents the mean results of modulus of elasticity and stiffness in static tests. According to the analysis of modulus of elasticity results, it was slightly higher for the base layer built on a combination of pine and alder woods than for the base layer built only on pine wood veneers. As far as the stiffness is concerned, in case of the base layer built on a combination of pine and alder it was over 20% higher than for the base layer built only on pine wood veneers. a/ b/

Figure 4. Static bending test: a- modulus of elasticity, b- stiffness

Table 2 presents the mean values of standard deviation and coefficient of variation for the static bending tests. Results for the base layer built on a combination of pine and alder woods are slightly lower in all cases.

Table 2. Statistical data for static bending tests Pine Pine / Alder Em [MPa] k [MNmm2] Em [MPa] k [MNmm2] Standard deviation 940 33 710 28 Coefficient of variation 9% 9% 7% 6%

Figure 5 presents the mean results of modulus of elasticity and stiffness in dynamic tests. According to the analysis of themodulus of elasticity, it was slighlty higher for the base layer built on a combination of pine and alder than for the base layer built only on pine wood veneers. As far as the stiffness is concerned, in case of the base layer built on a combination of pine and alder it was over 20% higher than for the base layer built only on pine wood veneers. a/ b/

Figure 5. Dynamic bending test: a- modulus of elasticity, b- stiffness

Table 3 presents the mean values of standard deviation and coefficient of variation for the dynamic bending tests. The results for the base layer built on a combination of pine and alder woods are lower. Especially, the standard deviation of modulus of elasticity in dynamic tests is noticeably lower, as well as the coefficient of variation. It means that the application of alder veneers provides stability to the composite structure.

Table 3. Statistical data for dynamic bending tests Pine Pine / Alder Em [MPa] k [MNmm2] Em [MPa] k [MNmm2] Standard deviation 1027 31 644 30 Coefficient of variation 10% 8% 6% 6%

The results of verification of the modulus of elasticity and stiffness in fatigue tests are presented in Figure 6 and Figure 7. Despite the fact that the research was of exploratory nature, they provide important information for future studies. An analysis of the modulus of elasticity (Fig. 6) shows that it was ca. 17% higher for the base layer built on a combination of pine and alder woods than for the base layer built only on pine wood veneers. It is remarkable that during all the tests, up to 50 cycles, the achieved results show a very stable state for both composites.

Figure 6. The results of modulus of elasticity in the fatigue test with a breakdown into every 10 cycles including the first cycle

An analysis of the modulus of elasticity (Fig. 7) shows that it was higher for the base layer built on a combination of pine and alder woods than for the base layer built only on pine wood veneers. As far as the stiffness is concerned, it was over 20% higher for the base layer built on a combination of pine and alder woods than for the base layer built only on pine wood veneers.

Figure 7. Comparison of stiffness values achieved in the fatigue test

Table 4 presents the mean values of standard deviation and coefficient of variation for the fatigue bending tests. The standard deviation results for the base layer built on a combination of pine and alder woods are higher than for pure pine base layer. This results from much higher values of modulus of elasticity. However, the results of coefficient of variation do not confirm this.

Table 4. Statistical data for fatigue bending tests Pine Pine / Alder Em [MPa] k [MNmm2] Em [MPa] k [MNmm2] Standard deviation 26 0.9 96 4 Coefficient of variation 0.3% 0.3% 1% 1%

CONCLUSIONS The application of additional alder veneers in the base structure made of pine veneers has a positive effect on the tested mechanical properties of the composites. All the tested values of mechanical properties showed, to a different degree, higher values than in case of composites with a base layer made only of pine veneers, even despite the fact that the modulus of elasticity and bending strength of alder are lower. On this basis we can draw the conclusion that a more homogeneous structure of the diffuse-porous alder wood has a decisive influence on the properties of the composite.

Acknowledgements: The authors are grateful for the support of the National Centre for Research and Development, Poland, under "Environment, agriculture and forestry" – BIOSTRATEG strategic R&D program, agreement No BIOSTRATEG2/298950/1/NCBR/2016.

REFERENCES

1. BORUSZEWSKI P., BORYSIUK P., MAMIŃSKI M., NICEWICZ D., 2013: Przewodnik do ćwiczeń z podstaw technologii tworzyw drzewnych. Warszawa: Wydawnictwo SGGW (in Polish). 2. GRZEŚKIEWICZ M., KRAWIECKI J., 2008: Thermally modified ash and oak wood as materials for parquets – mechanical properties of the wood and its UV resistance for different kinds of wood finishing. Ann. WULS - SGGW, For. and Wood Technol. No 65, 2008: 93-97. 3. HERÄJÄRVI, H., 2004: Variation of basic density and Brinell hardness within mature Finnish Betula pendula and B. pubescens stems. Wood and Fiber Science 36 (2): 216- 227. 4. HOLMBERG H., 2000: Influence of grain angle on Brinell hardness of Scots pine (Pinus sylvestris L.). Holz als Roh- und Werkstoff. June 2000, Volume 58, Issue 1–2: 91–95. 5. MAKOWSKI A., NOSKOWIAK A., 2016: Empirical verification of a digital model of a basketball to assess elastic properties of sports floors. Ann. WULS - SGGW, For. and Wood Technol. No 95 2016: 227-230. 6. SONG-YUNG WANG, HON-LIN WANG., 1999: Effects of moisture content and specific gravity on static bending properties and hardness of six wood species. Journal of Wood Science, April 1999, Volume 45, Issue 2: 127–133. 7. BARLINEK, https://www.barlinek.com.pl/deska-barlinecka-warstwowa/ (October 2019). 8. JAF, https://www.jaf-polska.pl/firma/aktualnosci/Twoj-kawalek-podlogi-n4850319 (October 2019). 9. ITD, https://www.itd.poznan.pl/pl/instytut/infoteka/bazy-danych/vademecum (October 2019). 10. PN-EN 1995-1-1 Eurokod 5: Projektowanie konstrukcji drewnianych.

Streszczenie: Wpływ zastosowania obłogu olchowego (Alnus glutinosa Gaerthn.) na wybrane właściwości mechaniczne kompozytów warstwowych z drewna sosnowego (Pinus sylvestris L.). Celem badań była analiza wpływu zastosowania fornirów z drewna liściastego w warstwie podbudowy na wybrane właściwości mechaniczne kompozytów z fornirów iglastych przeznaczonych do aplikacji na podłogach. Określono moduł sprężystości oraz sztywność przy zginaniu trzypunktowym w testach: statycznym, dynamicznym oraz zmęczeniowym. Wszystkie badane właściwości mechaniczne kompozytów sosnowo- olchowych wykazały, w różnym stopniu, wyższe wartości od kompozytów o podstawie wykonanej jedynie z obłogów sosnowych.

Corresponding author:

Piotr Beer Institute of Wood Sciences and Furniture WULS-SGGW Nowoursynowska Str. 159 02-787 Warszawa, Poland w-mail: [email protected] phone: +48 22 59 38 526

ORCID ID: Beer Piotr 0000-0002-2906-1208 Burawska-Kupniewska Izabela 0000-0001-8636-5622 Oleńska Sylwia 0000-0002-8463-4814 Różańska Anna 0000-0003-1865-3571

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 21-25 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Studies on the diversity of substrate composition in the culture medium of Kombucha microorganisms and its influence on the quality of synthesized cellulose

IZABELA BETLEJ Department of Wood Science and Wood Protection, Faculty of Wood Technology, University of Life Science - SGGW

Abstract: Studies on the diversity of substrate composition in the culture medium of Kombucha microorganisms and its influence on the quality of synthesized cellulose. The paper presents the results of the assessment of the effect of nutrients, specifically different nitrogen concentrations in the growth medium of Kombucha microorganisms, on the morphology of cellulose produced and its sorption capacity. Analyzing the obtained research results, we found that polymers formed in different growth environments differ in morphological structure and swelling index. The polymers synthesized on a nitrogen-rich substrate were characterized by a multilayer structure and a lower swelling index than the polymers obtained on a nutrient-poor substrate.

Keywords: bacterial cellulose, Kombucha, carbon and nitrogen source

INTRODUCTION Bacterial cellulose is a polymer synthesized mainly by acetic fermentation microorganisms belonging to Gluconacetobacter, although other groups of bacteria such as Escherichia or Pseudomonas, and even fungi, also have the ability to synthesize this polymer [1, 2, 3]. Despite the fact that bacterial cellulose is chemically identical to plant cellulose, it has better physico-mechanical properties than plant cellulose. Better properties result from higher purity, uniform and continuous fiber network, higher degree of polymerization and crystallization, and a different ability to absorb and retain water [4, 5]. These features mean that bacterial cellulose has potentially wide application significance. Currently, outside the medical field, interest in using cellulose is centered around the paper industry [6, 7]. Recycled materials have significantly worse properties than those made of primary wood fiber. Thus, the use of bacterial cellulose, whose production method is relatively simple and short in time, may prove to be a highly important solution for the paper industry. A very important quality feature of bacterial cellulose is its high tensile strength. As Stanisławska [8] reports, films made on the basis of bacterial cellulose are more than three times more tear resistant than cellophane films and more than seven times larger than polypropylene films. The quality of bacterial cellulose depends on the type of culture of microorganisms. Numerous studies have found that the type of substrate ingredients, culture temperature, substrate pH, and drying method affect the efficiency of synthesized cellulose and its quality [9, 10]. Zhao et al. [11] compared the efficiency of cellulose synthesis on synthetic and undefined media derived from wastewater from fermentation of polysaccharides. The authors of the study determined the differences and similarities between samples of cellulose synthesized on different types of medium substrates. The influence of carbon and nitrogen sources on the thickness, shape, smoothness and tensile strength of polymers synthesized by Kombucha microorganisms was the subject of research by Yim et al. [12]. The authors of the study have also proved that cellulose tensile strength is twice as high as grain leather of the same thickness. This paper presents the results of research determining how the composition of Kombucha microbial growth medium affects the basic morphological characteristics and the swelling index of cellulose synthesized by these microorganisms.

MATERIALS AND METHODS The object of the research was bacterial cellulose obtained in the process of cultivating Kombucha microorganisms on the basis of various nutrient content. Each type of medium contained sucrose (2.5%) as a carbon source and various concentrations of vegetable peptone (0%, 0.1%, 0.25% and 0.5%), a nitrogen-rich component. Cellulose synthesis by Kombucha microorganisms was carried out in stationary culture conditions, in a heat incubator, at temperature-humidity conditions of 24˚C and 68 ± 2%, respectively. After 14 days of culturing, the obtained cellulose was weighed, purified, and then dried at 60˚C for 24h. The cellulose purification process consisted of washing the polymer with distilled water, followed by soaking for 30 min. in 0.1M NaOH, and another three washes in distilled water to stabilize the pH. For the synthesized polymers, the swelling index was determined based on the methodology presented in patent 227860 B1 [13]. The morphological properties of bacterial cellulose were investigated by means of emission scanning electron microscopy (SEM), FEI QUANTA 200 model, which is available at the SGGW analytical center. Observations were carried out in a low vacuum, with a magnification of 500-4000x. Dried cellulose samples were placed on a carbon belt without dusting.

RESULTS The composition of the culture medium had a clear effect on the appearance and morphology of cellulose synthesized by Kombucha microorganisms [Figure 1]. Cellulose obtained from cultivation on a nitrogen-rich medium was clearly whiter and coarser, despite having a significantly lower swelling rate. The microscopic image shows numerous layers of polymer superimposed. A 1000x magnification depicts cellulose layers in the form of webs that overlap [Figure 2]. There are also differences in the appearance of cellulose surfaces formed on different nutrient media. The structure of bacterial cellulose, formed on a nutrient- rich medium, is more folded [Figure, 2, 3]. This is probably the result of the cellulose fiber structure and their packing density.

Figure 1. Surface morphology of bacterial cellulose synthesized by Kombucha microorganisms on a medium containing A - 2.5% sucrose B - 2.5% sucrose and 0.5% peptone

Figure 2. Cellulose morphology synthesized by Kombucha microorganisms on a medium containing 2.5% sucrose and 0.5% peptone

Figure 3. Cellulose morphology synthesized by Kombucha microorganisms on a medium containing 2.5% sucrose and 0.1% peptone

Based on the results obtained, it can be concluded that the composition of the culture medium also clearly affects changes in the bacterial cellulose's swelling ability. Based on the obtained swelling index, it was found that the presence of nitrogen in the growth medium reduces the sorption potential of the biopolymer. In the nitrogen-rich medium (0.5%), the examined indicator was slightly over 850%. In the case of cellulose obtained from a culture

23 on a substrate with only sucrose as a nutrient, the swelling ratio was 2814.67% [Figure 4]. The swelling index on sucrose medium was over three times higher than the swelling value obtained for cellulose from the culture with the highest peptone content in the culture medium composition. The differences in the ability to absorb water may closely depend on the structure and packing of individual cellulose fibers in a three-dimensional network. Probably, the fibers in the multilayer structure obtained on a substrate rich in nitrogen create a more packed network than it is in a polymer synthesized on a substrate poor in nutrients.

Figure 4. Swelling index of bacterial cellulose obtained from various culture media A - 2.5% sucrose, A1- 2.5% sucrose + 0.1% peptone, A2 - 2.5% sucrose + 0.25% peptone, A3 - 2.5% sucrose + 0.5% peptone

CONCLUSION The following conclusions can be made based on the obtained results:  the composition of the growth medium of cellulose synthesizing microorganisms has an impact on the morphological characteristics of cellulose,  cellulose synthesized on nitrogen-rich medium has a lower swelling rate.

REFERENCES

1. LAZARINI S.C., YAMADA C., BARUD H.S., TROVATTI E., CORBI P. P. LUSTRI W.R., 2018: Influence of chemical and physical conditions in selection of Gluconacetobacter hansenii ATCC 23769 strains with high capacity to produce bacterial cellulose for application as sustained antimicrobial drug-release supports,” Applied Microbiology nr 125; 777-791. 2. MENENDEZ E., GARCIA-FRAILE P., RIVAS R., 2015: Biotechnological applications of bacterial cellulases, Bioengineering nr 2(3); 163-182. 3. SHARMA CH., BHARDWAJ N.K., 2019: Biotransformation of fermented black tea into bacterial nanocellulose via symbiotic interplay of microorganisms, International Journal of Biological Macromolecules nr 132; 166-177. 4. KHAN S., UL-ISLAM M., ULLAH M. W., ISRAR M., JANG J. H., PARK J.K., 2018: Nano-gold assisted highly conducting and biocompatible bacterial cellulose-PEDOT:PSS films for biology-device interface applications, International Journal of Biological Macromolecules nr 107; 865–873.

5. TAHARA N., TABUCHI M., WATANABE K., YANO H., MORINAGA Y.,TT AND F. YOSHINAGA 1997: Degree of Polymerization of Cellulose from Acetobacter xylinum BPR2001 Decreased by Cellulase Produced by the Strain Bioscience, Biotechnology and Biochemistry., 61 (II), 1862186S, 1997 6. EL-HOSENY S. M., BASMAJI P., DE OLYVEIRA G. M., MANZINE COSTA L. M., ALWAHEDI A. M., DA COSTA OLIVEIRA J. D., FRANCOZO G. B., 2015: Natural ECM-Bacterial Cellulose Wound Healing—Dubai Study Journal of Biomaterials and Nanobiotechnology, nr 6; 237-246. 7. SKOČAJ M., 2019: Bacterial nanocellulose in papermaking, Cellulose nr 26(11); 6477–6488. 8. STANISŁAWSKA A., 2016: Bacterial nanocellulose as a microbiological derived nanomaterial, Advances in Materials Science nr 16(4); 45-57. 9. CHEN G., WU G., CHEN L., WANG W., HONG F. F., JONSSON L. J., 2019: Comparison of productivity and quality of bacterial nanocellulose synthesized using culture media based on seven sugars from biomass, Microbial Biotechnology nr 12(4); 677–687. 10. ILLA M. P., SHARMA C. S., KHANDELWAL M., 2019: Tuning the physiochemical properties of bacterial cellulose: effect of drying conditions, Journal of Materials Science nr 54(18); 12024-12035. 11. ZHAO H., XIAC J., WANG J., YAN X., WANG C., LEI T., XIAN M., ZHANG H. 2018: Production of bacterial cellulose using polysaccharide fermentation wastewater as inexpensive nutrient sources, Biotechnology & Biotechnological Equipment nr 32(2); 50-356. 12. YIM S. M., SONG J. E. AND KIM H. R. 2017: Production and characterization of bacterial cellulose fabrics by nitrogen sources of tea and carbon sources of sugar, Process Biochemistry nr 59; 26-36. 13. FIJAŁKOWSKI K., RAKOCZY R., ŻYWICKA A., PEITLER D., DROZD R., KORDAS M., KONOPACKI M., JUNKA A., Patent: Sposób wytwarzania celulozy bakteryjnej, Polska, nr patent: 227860 B1, rok zgłoszenia 29.04.2015.

Streszczenie: Ocena zróżnicowania składu podłoża w hodowli mikroorganizmów Kombucha i jego wpływ na jakość syntetyzowanej celulozy. W pracy przedstawiono wyniki oceny wpływu składników pokarmowych, a dokładnie różnych stężeń azotu w podłożu wzrostu mikroorganizmów Kombucha, na morfologię oraz zdolność do pochłaniania wody celulozy syntetyzowanej przez te mikroorganizmy. Analizując uzyskane wyniki badań, stwierdzono, że polimery powstałe w różnych środowiskach wzrostu różnią się pod względem struktury morfologicznej i zdolności do pochłaniania wody, wyrażonej wskaźnikiem spęcznienia. Polimer syntetyzowany na podłożu zasobnym w azot odznaczał się wyraźną wielowarstwową strukturą i mniejszym wskaźnikiem spęcznienia niż polimer otrzymany na podłożu ubogim w składniki odżywcze.

Corresponding author:

Izabela Betlej ul. Nowoursynowska 159, 02-787 Warszawa email: [email protected]

ORCID ID: Betlej Izabela 0000-0001-6867-0383

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 26-38 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Design of a snap connector to connect panel elements

ŁUKASZ MATWIEJ1, ROBERT KŁOS1, MIROSŁAW BONOWSKI2 1Department of Furniture Design, Faculty of Wood Technology, Poznan University of Life Sciences 2LINDNER sp. z o.o., ul. Gnieźnieńska 67, 62-100 Wągrowiec, POLAND

Abstract:Design of a snap connector to connect panel elements. The aim of this study was to design, manufacture and verify the tensile strength of a prototype snap connector to be used to connect panel elements. Firstly, analyses were conducted on solutions of commercially available designs for connectors invisible from the cabinet’s outside and those with minimized visibility. While searching for the best concept of connector design, three proposals were prepared, of which – after thorough analysis of design – one concept was selected. In the next step, the adopted solution was improved so that the connector met the previously formulated design requirements. In the course of further analyses, the causes and effects of failure were verified in order to limit or eliminate potential defects. In the next stage of the study, numerical calculations were conducted for the nut and the connector, concerning tensile strength, using the Autodesk Simulation Multiphysics program. After a prototype connector was manufactured, tensile strength tests were conducted on the connector using a strength testing machine. Experiments verified the correctness of the developed design in terms of geometry and the physico-mechanical properties of materials of individual elements, and resulted in possible changes proposed in the design of the final connector product.

Keywords: connecting elements, snap connector, numerical calculations

INTRODUCTION At present, the furniture manufacturing industry accounts for approx. 2% GDP of the Polish market and comprises approx. 24 thousand enterprises. Thanks to such a considerable potential, in 2012 Poland ranked 4th in the European Union in terms of the value of furniture production at 6.9 billion Euro (Adamowicz and Wiktorski2013). Due to the high volume of furniture production, there is a great demand for all types of furniture hardware, including also connectors. Currently, many types of connectors used in furniture manufacturing are commercially available, while in terms of connections we may distinguish three main types: with mechanical connectors, fittings and adhesive inserts. The connections with mechanical connectors may be further subdivided into removable and permanent. The requirements imposed on connectors by the present day market are very high. They are expected to be reliable, very accurately and precisely manufactured, to have high strength parameters and a simple design. Moreover, final users of furniture consider the esthetic value of products to be of great importance. Thus we observe a growing trend to minimize the conspicuity of connectors. Many producers of furniture hardware worldwide strive to design connectors that would attract as little consumer attention as possible. Such R&D work resulted in the design of connectors for elements made from wood and wood-based materials such as Rafix, Tab 18 or Rasant Tab by Häfele, very popular Rastex connectors by Hettich or a very discrete Rostro connector by EffegiBrevetti. All these hardware elements combine functional properties (removable connectors) and esthetic value (minimized visibility from the cabinet’s outside). However, complete invisibility of the connectors – with some exceptions – may be provided only on condition that the connectors are permanent, i.e. devoid of the functionality of multiple use. At present, an exception in this respect is provided only by solutions based on the use of magnetic field, which considerably increases production costs of connector elements and is the primary cause for the limited interest in this design among furniture manufacturers.

Permanent connectors constitute a relatively small group of designs. They include, first of all, snap connectors. Relatively popular products available on the market are e.g. theTenso P-14 snap connector (Fig. 1) by Lamello, the MOD-EEZ self-locking fastener (Fig. 2) and the Titusonic special rivets (Fig. 3) by Titus.

Figure 1. Tenso P-14 connector (Source: Catalogue Lamello 2019)

Figure 2. MOD-EEZ connector: a – diagram of connection, b – drawing of a connector (Source: Catalogue MOD-EEZ2019)

Figure 3. A diagram of connection according to the WoodWelding method (Catalogue Titus 2019)

Although completely invisible after the elements are joined, these connectors are burdened with certain drawbacks. The primary disadvantages of these solutions include relatively large dimensions of connector elements, a complex design, and the need to use specialist tools when machining joined elements. Another drawback is also connected with the use of plastics, which due to their contents of released chemical compounds prevent disposal by incineration of furniture containing these connectors.

Thus, the primary aim of this study was to design and produce a prototype snap connector for joining panel elements that would have the following characteristics: ‒ simple design, ‒ small dimensions, ‒ use of materials allowing disposal by incineration, ‒ no need to use specialist tools during machining and assembly. The result of this study important from the point of view of pure science was to determine the tensile strength of the prototype connector. The objective important for practice was connected with the verification of the theoretical solution for the new connector design by FEM numerical analysis, and preparation of technical documentation.

MATERIALS The authors assumed that the designed connector should be suited for various applications - not only in furniture, but also in other products made from wood and wood- based materials. For this reason, the following limitations were assumed for the designed connector: ‒ minimum thickness of joined elements: 16 mm, ‒ maximum recess depth for the connectorelement: 12 mm, ‒ assembly of the connector with no specialist tools required, ‒ no plastics used in theconnector, ‒ connector invisible from the outside. The design problem was solved based on a review of commercially available connectors and an analysis of various concepts for design solutions of the prototype connector. The optimal solution to the design problem was selected from among several analyzed design concepts. The selection criteria included simplicity of design, metal as the structural material and small dimensions. It was assumed that modification will be possible thanks to the improvement of the concept so that it would meet the previously stated requirements. The specially designed snap connector, meeting the specified assumptions, is composed of four elements: a double- threaded joint, a single-threaded joint with a stop, a nut and a screw (Fig. 4).

a b c d Figure 4. Elements of the snap connector. a – the M3 screw, b – the nut, c –the single-threaded joint with a stop, d – the double-threaded joint

A model of the selected design for the snap connector is presented in Fig. 5.In the double- threaded joint and in the single-threaded joint with a stop,on the outer cylinder surface, a metric thread M10 with apitchof 1.5 mm was used. The single-threaded joint with a stop has special ridges with a pitch of 0.3 mm to block thenut.The double-threaded joint has a seat for the metric M3 screw with a pitch of 0.5 mm. The M3 screw is standardized according to the PN-EN ISO 4762 (2006) specifications, which makes it readily accessible.

Figure 5. A model of the selected connector design

The screw has a capstan head with a hexagonal seat of the following dimensions: height k = 3 mm and diameter dk = 5.5 mm. The depth of thread in screw b (Fig. 6) corresponds to the length of the pin l = 12 mm. A specially bent square nut, 0.4 mm thick, is screwed on the M3 pin having a thread with a 0.5 mm pitch. Dimensions of the M3 screw selected for the connector are given in Fig. 6.

Figure 6. Dimensions of screws with flat, oval or round capstan heads used in the connector: dk = 3.5 – 5.6 mm, l = 12 mm, b = 12 mm, k = 3 mm

The difference in the height of the double-threaded joint and the single-threaded joint with a stop amounts to 0.5 mm. The difference in height was intentional, it was introduced to provide clearance to ensure proper matching of joints. When starting the assembly in the joinedelements, seats need to be made of 10 mm in depth and 8.5 mm in diameter. The diagram of assembly for the snap connector is presented in Fig. 7.

Figure 7. A diagram of assembly of the snap connector: a – element 1, b – single-threaded joint with a stop, c – screw, d – nut, e – double-threaded joint, f – element 2

The assembly operation needs to be performed following the successive steps: – step 1 – nut (d) is screwed on the M3 screw (c), – step 2 – the single-threaded joint with a stop (b) is screwed into the seat of the first element (a), – step 3 – the double-threaded joint (e) is screwed into the seat of the second element (f), – step 4 – the M3 screw (c) with the nut (d) is screwed into the double-threaded joint (e). At the moment of applyingadequate force, the connector snaps shut, as a result of the nut being blocked by the respective ridge in the single-threaded joint with a stop. According to the method proposed by Hamrol and Mantura(2012), in order to obtain information on the strengths and weaknesses of a product,the Failure Mode and Effect Analysis (FMEA)may be conducted already at the stage of preliminary design work on a new connector. This analysis makes it possible to verify the potential of occurrence of a given failure in the new product, and for this reason it is frequently performed at the first stage of designing novel products, before they reach the production stage. Thanks to this analysis, weaknesses of the product may be prevented, limited or eliminated. When designing the connector, FMEA was conducted as follows: – potential for failure in the final product was analyzed, – it was assessed according to the scale of importance, detectability and incidence rate of a given failure type, – the priority score was calculated for individual failures. In FMEA, a given failure of the product is assessed (Hamrol and Mantura2012) by assigning to it the so-called priority score. The assessment for the calculation of the priority score is conducted using three criteria: – importance of the failure – Z, where 1 means that the failure is of no importance, while 10 denotes great importance, – detectability of the failure – W, where 1 means that the detection of the failure is very easy and 10 means that the failure may not be detected, – possibility of the failure – R, where 1 means that the occurrence of the failure is improbable and 10 denotes inevitability of the failure.

The letter P denotes the priority score informing on the importance of a given failure type, calculated according to the following formula (1): (1)

At the design phase of a new product, an important stage consists in performing numerical calculations using the Finite Elements Method (FEM). It makes it possible to exclude or improve defective elements already at the design stage. The Autodesk Simulation Multiphysics environment was used for FEM calculations. During numerical calculations, the recorded values were compared to the respective strength grades of screws used in dynamometry (in accordance with the standardPN-EN ISO 898-1: 2013).In this manner, the connector was evaluated, so that it met requirements greater than or equivalent to grade 6.8. In the first stage of the study,a decision was made to conduct numerical calculations of the nut element. In order to obtain information whether the material may withstand the process of nut forming,the bending of a steel sheet element was simulated in the test. A model of 0.5 mm in thickness had a M3 tapped hole with a pitch of 0.5 mm. The examined element was placed on a support so that the action of forces resembled the real conditionsas closely as possible. The following settings were adopted in the program: – type ofelement: brick, – mesh type: bricks and tetrahedra, – mesh density of the nut: 25% , – mesh density of the support: 190%, – mesh size for the nut: 0.3 mm, – mesh size for the support: 1 mm, – assumed contact: bonded (between the nut and support), – nut material: AISI 1005 Steel, – support material: AISI Type 304 Stainlees Steel. In order to provide complete restraint of the model, the hole was assigned nodes preventing translation in the X, Y and Z axes. At each shoulder end of the nut element, a constant load of 10 N was exerted vertically downwards, following the Y axis (Fig. 8).

nut

10 N 10 N

support

Figure 8. A mesh model of the nut

In the next stage, stretching was simulated for the whole connector. When preparing a simplified model for numerical calculations,no bevelling of edges was applied in the double- threaded joint, the single-threaded joint with a stop or the screw. The bolt head was also simplified thanks to changes in its shape and elimination of rounded edges. Due to the axes of symmetry found in the connector, in order to shorten the numerical calculations, the analyses were conducted only on the model of a half of the connector. The following settings were adopted when preparing the model for calculations: – type ofelement: brick, – mesh type: bricks and tetrahedra, – mesh density of thedouble-threaded joint: 60 %, – mesh density of thesingle-threaded joint with a stop: 45 %, – mesh density of the nut: 45 %, – mesh density of the screw: 60 %, – mesh size for the double-threaded joint: 3 mm, – mesh size for the single-threaded joint with a stop: 0.3 mm, – mesh size for the nut: 0.3 mm, – mesh size for the screw: 3 mm, – contact at the interface of the elements: surface contact (coefficient of friction for steel on steel: 0.15), – material for all elements: AISI 1005 Steel. Elements were assigned the following movement constraints: – a constraint was put on translation towards the X, Y and Z axes for the screw and the double-threaded joint, – a constraint was put on translation towards the X axis for the nut, – a constraint was put on translation in one plane: X, Z for the single-threaded joint with a stop. Stretching of the connector was simulated in the course of the analyses. For this purpose,a surface load of 15 N directed vertically, opposite to the direction of the Y axis, was placed on the bottom of the seat in the single-threaded joint with a stop (Fig. 9).

double-threaded joint screw

nut

single-threaded joint with a stop

15 N

Figure 9. A mesh model of the connector

Prior to tensile strength testing of the connector, it was necessary to manufacture a prototype connector. A small batch of prototype elements was commissioned. A standardized screw was purchased in a single-line store. Table 1 presents material specifications. The double-threaded joint, the single-threaded joint with a stop and the screw were made to meet the requirements of the previously assumed quality grade (Table 1).

Table 1. Material specifications for connector elements Tensile strength Name of element Material Yield point Re [MPa] Rm[MPa] double-threaded joint Structural single-threaded joint with carbon steel 600-800 355-430 a stop C45 Galvanized nut steel 270-500 140-300 DX51D+Z275 Galvanized screw 800 640 steel grade 8.8

Due to the limited time of testing, the nut was made ofa lower grade material (grade 5.6) than it had been initially assumed (at least grade 6.8). The elements of the prototype connector are presented in Fig. 10.

a b c d

Figure 10. A prototype snap connector: a – double-threaded joint, b – M3 screw, c – nut, d – single-threaded joint with a stop

The tensile strength testing of the connector was conducted on a batch of 10 connectors using a Zwick 1445 strength testing machine. Prior to the tests, the connectors were assembled (snapped shut). During the tests, the connectors were mounted in the special clamps of the testing machine, parallel to the cross-bar motion axis. The connector was subjected to the tensile strength test at 10 mm/min. After the test was completed, the values of forces causing failure (uncoupling) of the connector were analyzed and the inflicted damage was assessed visually.

RESULTS Table 2 presents FMEA results, which gives a list of potential defects of the product together with the proposals for their elimination or limitation.Obtained priority scores constitute the basis for the introduction of changes, potentially leading to the improvement of the final product. According to FMEA, the nut is the element most susceptible to failure. The highest priority score was assigned to nut bending, for which it was 144. We proposed to eliminate the problem by using a different material or changingthe nut shape. Another defect

33 in this element was connected with the potential thread damage. This failure receiveda priority score of 96.

Table 2. FMEA for the designed snap connector Item Defect Proposal for elimination Z W R P Use of a different material /change of 1 Nut bending 6 6 4 144 nut shape 2 Connector off-size Increasing the scale of connector 7 4 4 112 Thread damage in the Use of a different material / 3 8 3 4 96 nut increasing nut thickness 4 Nut cracking Use of a different material 9 5 2 90 Insufficient pressure of double-threaded 5 joint to single- Reduction of pitch in stop ridges 8 2 5 80 threaded joint with a stop Chipping of stop 6 Change in stop ridge shape 8 4 2 64 ridges 7 Screw thread damage Use of a different material 8 3 2 48 Thread damage in 8 Use of a different material 8 3 2 48 double-threaded joint Axis displacement of double-threaded joint Reduction of clearance between 9 4 3 3 36 and the joint with a elements stop

Numerical calculations facilitated an analysis of distribution of reduced stresses appearing in the elements of the tested connector under the working load.When analyzing the distribution of stresses in the element, it needs to be stated that stresses accumulate mainly in the corners and around the nut (Fig. 11).

Figure 11. Distribution and values of reduced stresses in the nut

However, the Huber-Mises maximum reduced stresses (approx. 340 MPa) do not exceed admissible values for the used material. This result is satisfactory and suggests no damage during machining of the nut element in the process of steel sheet bending.

When analyzing the distribution of stresses, it needs to be stated that stresses accumulate mainly in the contact zones of mated elements: the nut and the stop. Nut corners are sites at the greatest risk of damage. Analysis of results showed that the greatest stresses were formed at the contact zone of the nut with the ridges of the single-threaded joint stop (Fig. 12), i.e. at its corners, and they amounted to approx. 580 MPa. On the surface of stop ridges, these stresses were much smaller, i.e. maximum approx. 230 MPa.

nut

single-threaded joint with a stop

Figure 12. Distribution and values of reduced stresses on the surface of the nut and the single-threaded joint with a stop

The analysis of the specifications of the PN-EN ISO 898-1 (2013) standard and the recorded testing results indicates that the nut would not meet the requirements of the minimal assumed quality grade 6.8. For this reason, the recommendation should be to have it made in strength grade 8.8. The other elements of the connector (the double-threaded joint, the single- threaded joint with a stop, the screw) should easily meet the assumed requirements. The strength criterion for the snap connector subjected to tensile testing was the value of force that would cause decoupling of the joined (snapped shut) connector.The results of tensile strength testing for the connector are presented in Fig. 13.

Figure 13. Tensile strength testing for the connector

Immediately after the onset of the test and exertion of load, a slight decrease in force was recorded, probably due to the so-called meshing of all nut shoulders. In the case of sample 5, a relatively low value of force was obtained due to the delicate bending of the nut element at the moment the connector elements snapped shut. Figure 14 presents the mean value of the recorded results with their scatter. Since some of the connectors failed even before the displacement of 0.76 mm, the presented results fall within the range of 0-0.58 mm. Maximum averaged force causing no displacement of connector elements was approx. 15 N. The greatest discrepancies in relation to the mean value were observed at displacements ranging from 0.09 to 0.18 mm, with a mean of 6.96 N. The standard deviation for all tests was 6.67 N. It was calculated based on the mean for all the maximum values that caused a complete failure of the tested connector, which amounted to 40.3 N.

Figure 14. Mean value and standard deviation of tested specimens

Upon thorough examination of uncoupled connectors, a truncation of nut shoulders was observed in each test (Fig. 15).

Figure 15. Damage on nut edges

This damage was caused by the nut rubbing againstthe stop ridges. No damage was found in single-threaded joints with stops, which may indicate better properties of the material of which these elements were made.

CONCLUSIONS Based on the collected information, the conducted analyses and tests, the following conclusions were formulated: 1. A review of the existing design solutions of removable and permanent connectors together withthe adopted assumptions made it possible to design and produce a prototype of a new snap connector. 2. The best possible design solution for the connector was selected thanks to ananalysis of possible causes and effects of defects and their verification in numerical calculations. 3. The numerical calculations indicated that the weakest structural element of the connector that was most likely to fail during operation was the nut. Therefore, a higher grade material needs to be selected when developing the final design of the new connector. 4. The tests of tensile strength of the prototype connector confirmed the accuracy of results provided by numerical calculations. Similarly as in the case of FEM, the nut proved to be the weakest element and its failure determined the tensile strength of the connector. Prior to launching the connector, it is recommended to manufacture elements with an adequate strength grade or to change the pitch of the stop ridge to a larger one, e.g. 0.5 mm. 5. The tests together with a detailed analysis of results made it possible to verify the accuracy of the theoretical design of the new connector andto develop of technical documentation in the form of working drawings and assembly drawings together with assembly instructions. 6. A small sized connector with a simple design may find extensive applications in connecting panel elements and may be successfully introduced in mass production as a competitive product, comparing to the ones commercially available at present.

REFERENCES

1. ADAMOWICZ M., WIKTORSKI T., 2013: Meble – polska inteligentna specjalizacja. http://www.meble.org.pl 2. Catalogue Lamello, 2019: Tenso P-14. Glue faster. http://www.lamello.com 3. Catalogue Mod-eez., 2019: Flexible joint structural fastening system. http://www.mod-eez.com

4. Catalogue Titus, 2019: Titusonic. Ultrasound Fastening http://www.titusplus.com 5. HAMROL A., MANTURA W., 2012: Zarządzanie jakością. Teoria i praktyka. Wydawnictwo naukowe PWN: Warszawa 6. PN-EN ISO 4762, 2006: Śruby z łbem walcowym z gniazdem sześciokątnym 7. PN-EN ISO 898-1, 2013: Własności mechaniczne części złącznych wykonanych ze stali węglowej oraz stopowej

Streszczenie: Projekt złącza zatrzaskowego do łączenia elementów płytowych. Celem niniejszej pracy było zaprojektowanie, wykonanie i sprawdzenie wytrzymałości na rozciąganie prototypu złącza zatrzaskowego przeznaczonego do łączenia elementów płytowych.Na wstępie dokonano wnikliwej analizy rozwiązań konstrukcji podobnych złączy znajdujących się na rynku. Dokonano także przeglądu wymagań dotyczących nowoprojektowanej konstrukcji złącza. W ramach poszukiwań najlepszej koncepcji rozwiązania konstrukcji złącza przygotowano trzy propozycje, z których – po przeprowadzeniu wnikliwej analizy konstrukcji – wybrano jedną koncepcję. W kolejnym etapie dokonano udoskonalenia przyjętego rozwiązania tak, aby złącze to spełniało postawione wcześniej wymagania projektowe. Podczas dalszych badańprzeprowadzono analizę rodzajów i skutków możliwych błędów (analiza FMEA), co miało posłużyć ograniczeniu lub wyeliminowaniu możliwych wad. W dalszej części pracy przeprowadzono obliczenia numeryczne przy wykorzystaniu programu AutodeskSimulationMultiphysics. Po wykonaniu prototypu złącza przeprowadzono badania wytrzymałości złącza na rozciąganie przy wykorzystaniu maszyny wytrzymałościowej. Badania eksperymentalne pozwoliły na zweryfikowanie poprawności opracowanej konstrukcji pod kątem geometrii oraz właściwości fizyko-mechanicznych materiału poszczególnych elementów i zaproponowanie ewentualnych zmian w konstrukcji finalnego złącza.

Corresponding author:

Łukasz Matwiej, Poznan University of Life Sciences Faculty of Wood Technology Department of Furniture Design Wojska Polskiego 38/42 60-627 Poznan, Poland email: [email protected] phone: +48 61 848 74 75

ORCID ID: Kłos Robert: 0000-0001-8583-6373

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 39-44 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Impact of upholstered furniture seat structures on the long-term use comfort

KRZYSZTOF WIADEREK1, IWONA WIADEREK2, JULIA LANGE1 1Department of Furniture Design, Faculty of Wood Technology, Poznań University of Life Sciences, 2Faculty of Finance and Banking, WSB University in Poznań

Abstract: Impact of upholstered furniture structures on the comfort of long-term use. The main objective of the study was to analyze the impact of changes to the structures of upholstered seating furniture to measure the comfort of new furniture and after long use. Tests were conducted with the use of a Force Sensitive Applications sensing mat to record contact pressure, and a profiled cavity pressed into the seat with a force of 760 N. The period of 5 years of long-term use was simulated by the cyclic load of 1000 N x 25,000 repetitions. Based on the analysis of the test results obtained, a decrease in the discomfort factor D by 12.7% for seat I and 11.5% for seat II was observed. This means an increase in the feeling of comfort in using these seats during the period of use. This is associated with a decrease in seat stiffness. Seats with less rigidity ("softer") cause less pressure on the human body due to the larger usable area.

Keywords: seat, upholstered furniture, comfort, contact pressure, foam.

INTRODUCTION The furniture industry is among the largest industries using foamed materials. In lounge furniture, polyurethane foams are still the main structural material. This material affects the comfort of use. Foam structures on lounge furniture are widely used mostly due to the technological and economic factors. However, a cheap and simple production does not always go hand in hand with quality. In the literature, a lot of attention is paid to issues related to the analysis of the selection and modeling of rigidity of seats made of traditional foams (Linder-Ganz et al. 2005; Schrodt et al. 2005; Vlaovic et al. 2008; Grujicic et al. 2009; Lusiak and Smardzewski 2010; Silber et al. 2010; Smardzewski et al. 2010a, 2010b; Wiaderek and Smardzewski 2010a, 2010b; Smardzewski and Matwiej 2013; Wiaderek et al. 2016). The construction material used is an important seat factor. From a user’s point of view, the most important is the impact of seat filling on comfort during use. The purpouse of the research is to determine the impact of long-term use of selected constructions of upholstered furniture seats for rest on the quality of use.

MATERIALS AND METHODS Polyurethane foams belong to a wide group of polyurethanes (PURs), which are characterized by wide durability, chemical and physical resistance as well as abrasion resistance. The properties of polyurethane foam depend primarily on their: - structure: cell sizes, their shape and structure, - density, - material constant values (Saha et al. 2005). The most commonly used types of elastic combinations of upholstered furniture were analyzed. The selected three most representative constructions are presented on Figure 1. The seats have fixed dimensions of 600 x 600 x 150 [mm]. All of them are based on a supporting structure in the form of wooden frames made of pine. Three seats were prepared for each variant. The load was applied to the seat using a numerically controlled Zwick testing machine. A force of 760 N was applied vertically downwards using a profiled indenter in accordance with the PN-EN 1728: 2012 standard. During the study, the displacement of the seat was recorded. Each seat was subjected to an

39 analysis of contact stress distribution (fig. 2). For this purpose, a 630 x 630 mm mFlex mat was used, equipped with 1024 sensors. The contact stress distribution on the seat surface was recorded with an accuracy of 0.01 kPa. The stress value was registered after 60 seconds from obtaining the assumed load force. This allowed to stabilize the foam deflection caused by the indenter.

Figure 1. Seat construction: a) 1-frame, 2-fibreboard, 3-wadding, 4- Bonell form, 5- T25180 polyurethane foam, 6-felt, 7-fabric, b) 1-frame, 2-spring, 3-felt, 4- T25180 polyurethane foam, 5-wadding, 6-fabric, c) 1-frame, 2-spring, 3-felt, 4- HDS55 higly elastic polyurethane foam, 5-wadding, 6-fabric

Figure 2. mFlex matting station

Seat comfort tests (Milivojevich et al. 2000) have shown a relationship between an even distribution of stress on the contact surface and a sense of comfort. One way to quantify the even distribution of pressure on the seat surface is to determine the percentage of SPD (Seat Pressure Distribution) contact stress distribution coefficient (Ahmadian et al. 2002) defined as:

where: n - number of sensors in which contact pressure has non-zero values, pi - contact pressure in any mat sensor [kPa], pm - average contact pressure for n sensors [kPa].

This method is used in conjunction with a system that illustrates the distribution of stress on the seat surface, such as mFlex system. The lower seat pressure distribution coefficient (SPD) value indicates a more favorable even distribution of stress on the seat surface. Thus, the seat pressure distribution coefficient (SPD), contact stress  , and contact surface have a direct impact on seating comfort. Unfortunately, this method does not exclude cases where the resulting stresses are too high for sitting comfort. In case of the discomfort factor D (Smardzewski et al. 2014), a high index will be obtained at high average contact stress on the pm sensors and low values of the contact surface A, and low values of the SPD factor. Low values of the D factor will objectively determine a high comfort of seat use. The discomfort factor D was determined according to the formula:

The next stage was subjecting the seats to long term tests of cyclic load modeled on the PN-EN12520 and PN-EN1728 standards. The tests consisted in a series of 25,000 repetitions of applying a force of 1000 N to the seat (which corresponds to about 5 years of home use). For this purpose, a profile indenter attached to a computer-controlled pneumatic cylinder was used (fig. 2).

RESULTS Laboratory tests have allowed to observe the load distribution and stress values for the analyzed structural variants of upholstered furniture seats. The results of the research were presented in the form of maps of stress distribution on the usable surface of the seats. A graphic comparison of test results before and after the cyclic loading process is presented in Table 1. To compare the changes in the tested values, Table 2 and a percentage graph were prepared (fig. 3).

Figure 3. Percentage comparison of the end results of the study

Table 1 Selected seat surface stress distribution maps:

Seat I Seat II Seat III

New

termuse

- Long

Table 2 Comparison of the end results of the study

Value before cyclic load Value after cyclic load application application Indication Symbol Unit

Seat I Seat II Seat III Seat I Seat II Seat III

Seat pressure distribution SPD % 17.02 14.38 22.24 19.54 16.97 20.94 coefficient

Discomfort factor D N/m4 14.13 13.89 9.28 12.33 12.29 9.36

Strain mm/mm cm 7.01 6.09 9.74 8.56 6.96 9.77

The analysis of results of the seat deformation test after the cyclic application of 760N, revealed that seat II is characterized with the highest rigidity at the values of deformation (from 6.09 to 6.96 mm/mm) in relation to the seat I (from 7.01 to 8.56 mm/mm) and seat III (from 9.74 to 9.77 mm/mm). However, the smallest decrease in stiffness was recorded for seat III, amounting to only 0.34%. The values of the contact stress distribution coefficient SPD indicate that seat III has the highest degree of comfort of use, followed by seat and then seat II. Values of the discomfort factor D also show that the most favorable in terms of comfort will be seat III. Although the difference between seat I and II is insignificant, seat I exhibits greater comfort than seat II. A greater discomfort factor indicates that the seat is not very comfortable. The smaller the discomfort factor, the more comfortable is the seat.

DISCUSSION There is a clear impact of the seat structure on the distribution of usable stress, and thus on the comfort and quality of use. The largest decrease in stiffness was observed in the case of seat I equal to 22 %. It follows that the combination of standard foam T25180 and bonnell springs worsens long-term use. Seat II ranked second, with a decrease in rigidity of 14.2%. In this case, the decrease in seat rigidity results from a decrease in the rigidity of the foam itself. For seats I and II, a clear increase in the SPD contact stress distribution coefficient was observed, respectively 14.8% for I and 18% for II. A decrease in the discomfort factor D by 12.7% for I and 11.5% for seat II was also observed. This indicates an

42 increase in the feeling of comfort in using these seats. This is associated with a decrease in seat stiffness. Seats with less stiffness, "softer", put less pressure on the human body. In the case of seat III, no decrease was observed, which is due to the use of high quality HDS55 highly elastic foam. This seat does not show significant changes in the SPD coefficient (5.9%) and D factor (0.9%). The foam manufacturer's data are also confirmed, which states that seats made of highly flexible foam retain their properties for up to 10 years of use.

REFERENCES

1. AHMADIAN M., SEIGLER T.M., CLAPPER D., SPROUSE A., 2002: Alternative test methods for long term dynamic effects of vehicle seats. SAE Transactions Vol. 111, Section 2: Journal Of Commercial Vehicles (2002), pp. 684-692 2. GRUJICIC M., PANDURANGAN B., ARAKERE G., BELL W.C., HE T., XIE X., 2009: Seat-cushion and soft-tissue material modeling and a finite element investigation of the seating comfort for passenger-vehicle occupants. Materials and Design 30: 4273–4285. 3. LINDER-GANZ E., YARNITZKY G., PORTNOY S., YIZHAR Z., GEFEN A., 2005: Real-time finite element monitoring of internal stresses in the buttock during wheelchair sitting to prevent sores: verification and phantom results. II International Conference on Computational Bioengineering, pp. 14–16. 4. LUSIAK A., SMARDZEWSKI J., 2010: Creative thinking in designing furniture for pre-school children. Annals of Warsaw University of Life Sciences SGGW Forestry and Wood Technology 70: 270–278. 5. SMARDZEWSKI J., BARAŃSKA-WOŹNY J., WIADEREK K., PREKRAT S., GRBAC I., 2010a: Mechanical and biomechanical criteria in furniture designing for 60+ users. In: Proceedings of the International Conference Ambienta, Zagreb, pp. 113–122. 6. MILIVOJEVICH A., STANCIU R., RUSS A., BLAIR G.R., HEUMEN J.D., 2000: Investigating Psychrometric and Body Pressure Distribution Responses to Automotive Seating Comfort, SAE Technical Paper 2000-01-0626 7. SCHRODT M., BENDEROTH G., KUHHORN A., SILBER G., 2005: Hyperelastic Description of Polymer Soft Foams at Finite Deformations. Technische Mechanik 35 (3–4): 162–173. 8. SILBER G., ALIZADEH M., SALIMI M., 2010: Large Deformation Analysis for Soft Foams Based on Hyperelasticity. Journal of Mechanics vol.26: 327–334. 9. SMARDZEWSKI J., MATWIEJ Ł., 2013: Effects of Aging of Polyurethane Foams in the Context of Furniture Design. Drvna Industrija 64 (3): 201-209 10. SMARDZEWSKI J., PREKRAT S., PERVAN S., 2010b: Research of Contact Stresses between Seat Cushion and Human Body. Drvna Industrija 2: 95–101. 11. SMARDZEWSKI J., JASIŃSKA D., JANUS-MICHALSKA M., 2014: Structure and properties of composite seat with auxetic springs. Composite Structures 113: 354-361 12. VLAOVIC Z, BOGNER A, GRBAC I (2008) Comfort Evaluation as the Example of Anthropotechnical Furniture Design. Collegium Antropologicum 32 (1): 277–283. 13. WIADEREK K., MATWIEJ Ł., DETTLAFF M., 2016: Impact of structures of selected lounge furniture seats on the comfort of use. Annals of Warsaw University of Life Sciences SGGW Forestry and Wood Technology 96: 241-248. 14. WIADEREK K., SMARDZEWSKI J., 2010a: Numerical evaluation of seat hardness. Annals of Warsaw University of Life Sciences SGGW Forestry and Wood Technology 70: 305–311.

15. WIADEREK K., SMARDZEWSKI J., 2010b: Modeling of foam seats in terms of comfortable relaxation furniture design. Proceedings of the International Conference Ambienta, Zagreb, pp. 139–146. 16. PN-EN 1728 Meble mieszkaniowe. Meble do siedzenia. Metody badań wytrzymałości i trwałości / Furniture - Seating - Test methods for the determination of strength and durability. 17. PN-EN 12520 Meble. Wytrzymałość, trwałość i bezpieczeństwo. Wymagania dla siedzisk mieszkaniowych / Furniture. Strength, durability and safety. Requirements for domestic seating.

Streszczenie: Wpływ konstrukcji mebli tapicerowanych na komfort w okresie długotrwałego użytkowania. Głównym celem pracy była analiza wpływu zmian konstrukcji siedzisk mebli tapicerowanych na pomiar komfortu mebli nowych oraz po długotrwałym ich użytkowaniu. Badania przeprowadzone były przy użyciu maty sensorowej Force Sensitive Applications z odczytem naprężeń kontaktowych oraz profilowanego wgłębia wciskanego w siedzisko siłą 760 N. Okres 5 letniego, długotrwałego użytkowania zasymulowano działaniem cyklicznego obciążenia 1000N x 25000 powtórzeń. Na podstawie przeprowadzonych analiz uzyskanych wyników badań zaobserwowano niekorzystny wzrost współczynnika dyskomfortu D o około 0,8% przy zaobserwowanym spadku o 12,7 % dla siedziska I i 11,5% dla siedziska II. Oznacza to wzrost odczucia komfortu użytkowania siedzisk I i II w okresie użytkowania. Wiąże się to ze spadkiem sztywności siedzisk. Siedziska o niższej sztywności („bardziej miękkie”) wywołują mniejszy nacisk na ludzkie ciało z uwagi na większą powierzchnię użytkową.

Corresponding author:

Krzysztof Wiaderek Uniwersytet Przyrodniczy w Poznaniu, Wydział Technologii Drewna Wojska Polskiego 38/42 60-627 Poznań, [email protected]

ORCID ID: Wiaderek Krzysztof 0000-0001-5432-4738 Wiaderek Iwona 0000-0002-3267-3918 Lange Julia 0000-0002-3019-1499

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 45-52 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Influence of the ion implantation of nitrogen and selected metals on the lifetime of WC-Co indexable knives during MDF machining

MAREK BARLAK1, JACEK WILKOWSKI2, KAROL SZYMANOWSKI2, PAWEŁ CZARNIAK2, PIOTR PODZIEWSKI2, ZBIGNIEW WERNER1, JERZY ZAGÓRSKI1, BOGDAN STASZKIEWICZ1 1 Plasma/Ion Beam Technology Division (FM2), National Centre for Nuclear Research Świerk - NCBJ 2 Department of Mechanical Processing of Wood, Warsaw University of Life Sciences - SGGW

Abstract: Influence of the ion implantation of nitrogen and selected metals on the lifetime of WC-Co indexable knives during MDF machining. The paper presents the results of durability tests for WC-Co indexable knives during the machining of MDF. The knives were implanted with nitrogen, zirconium, molybdenum and tin, using MEVVA type implanter with non-mass separated beam. Additionally, the Monte Carlo simulation results of the main parameters of the depth profiles of the implanted elements are presented in this paper. A higher correlation of tool life with the project range and range straggling than with the parameter of the peak volume dopant concentration was demonstrated.

Keywords: WC-Co composites, ion implantation, modelling, MDF, milling

INTRODUCTION Cemented tungsten carbides - a combination of hard and brittle carbides and a relatively soft and ductile metallic binder - provide an exceptional combination of favourable properties such as strength, hardness, fracture toughness, refractoriness, stiffness, resistance to compressive deformation and wear resistance at room temperature as well as at higher temperatures up to 400°C (Milman et al. 1997, Sheikh-Ahmad and Bailey 1999, Pirso et al. 2004, Bonny et al. 2004, Choi et al. 2010, Olovsjö et al. 2013). Unfortunately, the durability of tools made of this material is still insufficient. There are few methods to improve this property, ion implantation being one of them (Barlak et al. 2016, Barlak et al. 2017). The present paper presents the influence of ion implantation of nitrogen and three metals, i.e. zirconium, molybdenum and tin, on the durability of WC-Co indexable knives, used for wood-based materials (in this case, for Medium Density Fiberboard - MDF) machining. Nitrogen is a typical element used to improve the durability/tribological properties (Kamiński and Budzyński 2017). The ion implantation of this gas to WC-Co substrate leads to the formation of nitrides- and a carbon-rich layer. The presence of nitride precipitates and the “lubricating” effect of carbon improves tribological properties of the modified material (Sun et al. 1997). Zirconium-based materials are used e.g. for sports technology (Khun et al. 2016), molybdenum oxides - in the system with aluminium bronze (Takeichi et al. 2009), and tin- based materials - in automotive industry or ocean tribology (Feyzullahoğlu et al. 2008, Wu et al. 2016).

MATERIAL AND METHODS The ion implantation processes were preceded by Monte Carlo simulations of the main parameters of the depth profiles of the implanted elements (like peak volume dopant concentration Nmax, projected range Rp and range straggling ΔRp), using freeware type code SRIM-2013.00 (The Stopping and Range of Ions in Matter) (SRIM 2019). The simulation was performed for 100 000 implanted ions. The modelled substrate material W-C-Co

(modelling codes treat the sample as a set of atoms that do not form chemical compounds) of a composition: 90.86% of tungsten, 5.94% of carbon and 3.2% of cobalt in weight percentages i.e. 47.4% of tungsten, 47.4% of carbon and 5.2% of cobalt in atomic percentages. The substrate material density, adopted for the simulation was 15.2 g/cm3. This value was declared by the knives supplier. In all cases, the simulations were performed for room temperature and for the total implanted fluence of 2e17 cm-2 and the acceleration voltage of 60 kV, including percentage charge state distribution data from Table 1. During the ion implantation processes using implanters with non-mass separated beam + + (with direct beam), the nitrogen beam includes two kinds of ions, i.e. N +N2 , in the ratio ~1:1, so there are two elementary charges per three atoms. In the case of the molecule N2 implanted with the acceleration voltage of 60 kV, each atom carries the energy of 30 keV, according to the law of energy conservation. Likewise, the average charge state is at the level of 0.67. The beam of metallic ions usually includes ions with a different degree of the ionization. Additionally, the percentage shares of individual ion types are also different. In this case, the average charge state can be determined by the addition of the values of multiplication of the percentage shares and the ionization degree. For example, for molybdenum it will be:

0.07·1 + 0.3·2 + 0.4·3 + 0.2·4 + 0.03·5 = 2.82 ≈ 2.8 (1)

Two ion charge states were used for the modelling, and then for the ion implantation of N and Sn, four for Zr and five for the Mo beam. The atomic radii of the implanted elements, determined from minimal-basis-set SCF functions are, in angstroms: 0.56 Å for N, 2.06 Å for Zr, 1.9 Å for Mo and 1.45 Å for Sn (Clementi et al. 1967). The modelling did not account for the phenomenon of substrate sputtering by the implanted ions.

Table 1. The percentage charge state distribution and average charge state of the ions implanted to WC-Co knives (Krivonosienko et al. 2001) Ion energy for acceleration voltage of 60 kV (keV) Average 30 60 120 180 240 300 Implanted ions charge Percentage charge state distribution (%) state 1+ 2+ 3+ 4+ 5+ + + N2 + N 67 33 - - - - 0.67 Zr+ + Zr2+ + Zr3+ + Zr4+ - 1 47 45 7 - 2.6 Mo+ + Mo2+ + Mo3+ + Mo4+ + Mo5+ - 7 30 40 20 3 2.8 Sn+ + Sn2+ - 47 53 - - - 1.5

The commercially available, WC-Co indexable knives, produced by Ceratizit company (KCR08), with dimensions 29.5×12×1.5 mm3 and presented in Fig. 1, were used in the investigations.

Figure 1. WC-Co indexable knives

Before processing, the knives were washed in high purity acetone under ultrasonic agitation. Next, the flank surfaces of the knives were implanted with nitrogen, zirconium, molybdenum and tin, using MEVVA (Metal Vapor Vacuum Arc) type implanter with non- mass separated beam, described in detail elsewhere (Bugaev et al. 1994). Nitrogen of 99.999% purity was used as the source of the implanted gaseous ions and as a working gas in the case of the implantation of metallic ions. The implanted fluence was at a level of 2e17 cm-2 and the acceleration voltage was 60 kV for all cases. The ion current densities were about 50 µA/cm2 and therefore, the estimated value of temperature of the implanted samples was about 300°C. The base pressure in the vacuum chamber was at a level from 2 to 5e-4 Pa. Virgin (non-modified, non-implanted) and implanted knives were tested for the lifetime. Four blades were tested in each modification group. Medium Density Fiberboard (MDF), produced by Pfleiderer company, with the thickness of 16 mm and the density of 743 kg/m3 was used for the experiments. The above-mentioned panels are a standard construction material, commonly used in furniture industry. Chosen properties of the cutting material are presented in Table 2.

Table 2. Selected mechanical and physical properties of the tested MDF Swelling after Flexural strength Modulus of elasticity Density Tensile strength Material 24h MOR MOE (kg/m3) (MPa) (%) (MPa) (MPa) MDF 16 mm 743 0.76 13.5 34.4 3367 thickness

Workpieces with dimensions of 2800×400×16 mm3 were milled on a CNC machining center Busellato Jet 130, equipped with a Faba FTS.07L4043.01 single-cutting milling head with a diameter of 40 mm. Grooves were made in the particleboard (with a width equal to the tool diameter, i.e. 40 mm) at a depth of 6 mm. During machining, constant cutting parameters were maintained (feed speed of 2.7 m/s, spindle speed of 18000 rpm and feed per tooth of 0.15 mm). The tool wear measurement was carried out after each pass (feed distance of 2.8 m), using a workshop microscope. The maximum wear width on the flank face of the VBmax blade was estimated. Machining stopped when the VBmax was equal to or greater than 0.2 mm. The cutting length was an indicator of the tool’s durability to achieve the tool’s wear criterion.

RESULTS AND DISCUSSION Fig. 2 and Table 3 present the results of the computer simulations of the main parameters and the depth profiles of ions implanted without mass separation to W-C-Co material. Additionally, a graphical definition of the main parameters of the depth profiles of the implanted elements is presented in the upper right-hand corner of Fig. 2.

Figure 2. The depth profiles of N, Zr, Mo and Sn ions implanted to W-C-Co material

The maximum value of the peak volume dopant concentration is observed for tin ions, and the minimum - for nitrogen. The difference is about threefold. These values are similar for two others elements, i.e. for zirconium and molybdenum and they are about 5e22 cm-3. The maximum values of the projected range and the range straggling are obtained for nitrogen ions and the minimum ones- for tin. In this case, the difference is more than twofold. These values are also very similar to zirconium and molybdenum (the depth profiles for both elements are very similar).

Table 3. Detailed parameters of N, Zr, Mo and Sn ion implantation to W-C-Co material Peak volume dopant Projected range Range straggling Implanted ions concentration -3 Rp (nm) ΔRp (nm) Nmax (cm ) + + N2 + N 3.44e22 36.8 21.8 Zr+ + Zr2+ + Zr3+ + Zr4+ 5.02e22 28.3 16.2 Mo+ + Mo2+ + Mo3+ + Mo4+ + Mo5+ 4.81e22 29.3 17.6 Sn+ + Sn2+ 9.17e22 15.6 9.3

The results of durability tests for the virgin samples and for the samples implanted with nitrogen, zirconium, molybdenum and tin are presented in Figs 3 and 4. Fig. 3 presents the values of the cutting length for all types of modification. Additionally, the standard deviation values are marked by error bars. The average values of the cutting length were: 51, 99, 81, 81 and 86 km for virgin samples and samples implanted with N, Zr, Mo and Sn, respectively. As we can see, the ion implantation resulted in an increased durability of the modified knives in each case. This effect was greatest in the case of nitrogen implantation (nearly 100% increase). The values of the cutting length for the implantation of metal ions increased by over 50% and, interestingly, were very similar in all cases. Unfortunately, it was worse in the case of uniformity of results. The value of standard deviation amounted to only 1 km for virgin knives, while for knives implanted with N, Zr, Mo and Sn these values were: 40, 25, 14 and 31, respectively. The most homogeneous values were obtained for Mo implantation.

Figure 3. The cutting length with standard error bars for virgin indexable knives and knives implanted with nitrogen, zirconium, molybdenum and tin

The maximum, average and minimum values of the relative tool life index for the indexable knives implanted with nitrogen, zirconium, molybdenum and tin are shown on Fig. 4. The best result of the maximum, i.e. nearly threefold increase in durability, was obtained for nitrogen implantation. The ion implantation of metallic ions increased the tool durability at the level of 100%. The worst result was observed for Zr implantation, but the worst results for tin and nitrogen implantation were similar to it.

Figure 4. The relative tool life index for indexable knives implanted with nitrogen, zirconium, molybdenum and tin

The correlation between the values of the main parameters of the depth implanted profiles and the cutting length (tool life) was also analysed (Fig. 5). The analysis revealed a 2 2 greater correlation with the projected range Rp (R ≈ 0.75) and the range straggling ΔRp (R ≈ 2 0.76), than with the peak volume dopant concentration Nmax (R ≈ 0.37). So, from the point of view of quality of the ion implantation process affecting the tools’ life, it is better to perform modification taking into account the projected range and the range straggling, than to obtain a higher peak volume dopant concentration.

Figure 5. The relationship between the projected range Rp, the range straggling ΔRp and the peak volume dopant concentration Nmax and the cutting length

CONCLUSION The average durability values of WC-Co indexable knives after ion implantation increased in all cases. The largest, on average almost two-fold increase was observed for tools implanted with nitrogen, while for the other elements it was at the level of 50%. The best results among metallic ions were achieved for tin. Zirconium and molybdenum are very comparable, both in terms of the implanted depth profiles and their influence on the durability of the modified knives; only the molybdenum effect is more homogeneous, i.e. the value range, from minimum, through average to maximum was the smallest for knives modified with this element. As one can easily see, there is no direct impact of the atomic radius of the implanted element on the durability of the tools into which it is introduced. However, despite the fact that durability depends on more factors, a higher correlation was demonstrated between the tool life and the depth parameters of the projected range and the range straggling than the parameter of peak volume dopant concentration. The large inhomogeneity of obtained results is a problem at this stage of investigations. There is a chance to further increase the average durability of modified tools after overcoming this problem.

REFERENCES

6. CLEMENTI E., RAIMONDI D.L., REINHARDT W.P., 1967: Atomic screening constants from SCF functions. II. Atoms with 37 to 86 electrons, The Journal of Chemical Physics 47, 1300-1307. 7. FEYZULLAHOĞLU E., ZEREN A., ZEREN M., 2008: Tribological behaviour of tin- based materials and brass in oil lubricated conditions, Materials and Design 29, 714- 720. 8. KAMIŃSKI M., BUDZYŃSKI P., 2016: Tribological properties of Stellite 6 cobalt alloy implanted with nitrogen ions determined in the tests conducted in engine fuel atmosphere, Advances in Science and Technology. Research Journal 11, 215-219. 9. KHUN N.W., YU H., CHONG Z.Z., TIAN P., TIAN Y., TOR S.B., LIU E., 2016: Mechanical and tribological properties of Zr-based bulk metallic glass for sports applications, Materials and Design 92, 667-673. 10. KRIVONOSIENKO A.W., NIKOLAEV A.G., LI S., 2001: Технические описание и инструкция по эксплуатаци ионного источника “Титан-3” (Technical descriptions and operating instructions of the ion source “Titan-3”), Российская Академия Наук - Институт Сильноточной Электроники, Tomsk, in Russian. 11. MILMAN. YU.V., CHUGUNOVA S., GONCHARUCK V., LUYCKX S., NORTHROP I.T., 1997: Low and high temperature hardness of WC-6 wt% Co alloys. International Journal of Refractory Metals and Hard Materials 15, 97-101. 12. OLOVSJÖ S., JOHANSON R., FALSAFI F., BEXELL U., OLSSON M., 2013: Surface failure and wear of cemented carbide rock drill buttons - The importance of sample preparation and optimized microscopy settings, Wear 302, 1546-1554. 13. PIRSO J., LETUNOVITŠ S., VILJUS M., 2004: Friction and wear behaviour of cemented carbides, Wear 257, 257-265. 14. SHEIKH-AHMAD J.Y., BAILEY J.A., 1999: High-temperature wear of cemented tungsten carbide tools while machining particleboard and fiberboard. Journal of Wood Science 45, 445-455. 15. SRIM. Interactions of ions with matter, http://www.srim.org/ 16. SUN J.S., YAN P., SUN X.B., LU G., LIU F., YE W., YANG J.Q., 1997: Tribological properties of nitrogen ion implanted WC-Co, Wear 213, 131-134. 17. WU H., ZHU S., CHENG J., QIAO Z., YANG J., 2016: Tribological Behavior of Tin- Based Babbitt Alloy Lubricated by Seawater, Tribology Transactions 59, 838-844 18. TAKEICHI Y., CHUJO T., OKAMOTO N., UEMURA M., 2009: Effects of molybdenum trioxide on the tribological properties of aluminium bronze under high temperature conditions, Tribology online 4, 135-139

Streszczenie: Wpływ implantacji jonów azotu i jonów wybranych metali na trwałość wymiennych noży WC-Co podczas obróbki płyt MDF. W artykule przedstawiono wyniki testów trwałościowych wymiennych noży WC-Co podczas obróbki płyt MDF. Noże były implantowane jonami azotu, cyrkonu, molibdenu i cyny, przy użyciu implantatora typu MEVVA bez separacji masowej. Ponadto, w artykule zamieszczono wyniki modelowania profili głębokościowych implantowanych pierwiastków, uzyskanych przy pomocy programu opartego o metodę Monte Carlo. Wykazano wyższe zależności korelacyjne trwałości ostrza z wartościami zasięgu rzutowanego i rozrzutu zasięgu niż z wartościami maksymalnej koncentracji objętościowej domieszki.

Corresponding author:

Marek Barlak e-mail: [email protected] National Centre for Nuclear Research Świerk - NCBJ Plasma/Ion Beam Technology Division (FM2) 7 Andrzeja Sołtana St. 05-400 Otwock, Poland

ORCID ID: Barlak Marek: 0000-0003-1416-7461 Wilkowski Jacek 0000-0001-5798-6761 Szymanowski Karol 0000-0003-0383-0897 Czarniak Paweł 0000-0001-8759-7679 Podziewski Piotr 0000-0002-2628-5062 Werner Zbigniew 0000-0003-1172-0268 Zagórski Jerzy 0000-0001-6311-947X Staszkiewicz Bogdan 0000-0002-7741-4362

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 53-57 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Bacterial cellulose synthesis by Kombucha microorganisms on a medium with a variable composition of nutrients

IZABELA BETLEJ, KRZYSZTOF J. KRAJEWSKI Department of Wood Science and Wood Protection, Faculty of Wood Technology, University of Life Science

Abstract: Bacterial cellulose synthesis by Kombucha microorganisms on a medium with a variable composition of nutrients. The paper presents the results of the assessment of the impact of various sucrose contents and the presence of various sources of nitrogen compounds in the growth medium of Kombucha microorganisms on the synthesis efficiency and the obtained bacterial cellulose mass. The analysis of obtained research results revealed that the efficiency of cellulose synthesis by Kombucha microorganisms depends on the quantity and quality of nutrients available in the growth medium.

Keywords: bacterial cellulose, Kombucha, carbon and nitrogen sources

INTRODUCTION From the chemical point of view, bacterial cellulose is the same polymer as plant cellulose, but it has additional, definitely better features than cellulose derived from plant tissues. First of all, it is characterized by high purity, which is guaranteed by the lack of lignin and hemicelluloses, high crystallinity, susceptibility to forming to any shape, high hygroscopicity and very high mechanical strength, as well as high biological compatibility [5, 8, 10]. These features guarantee great opportunities for the use of bacterial cellulose in various industries. Bacterial cellulose is already successfully used in medicine, as a dressing material or surgical implants, in optics as biosensors, and also in the food, pharmacy and paper industries [7]. In the paper industry, bacterial cellulose is mainly used for bleaching waste paper and as a filler for defects in printing [6]. The use of cellulose in the woodworking and packaging industries also seems to be prospective. Bacterial cellulose is synthesized by a large group of microorganisms, both bacteria and yeast. Among cellulogenic microorganisms, those belonging to the genera: Acetobacter, Aerobacter, Achromobacter, Agrobacterium, Psedomonas, and Sarcina have been identified [1]. These microorganisms often appear as a conglomerate, biofilm, often described as a “scoby”. Despite many unique physicochemical features and a very promising application perspective, the use of bacterial celluloses on a large scale poses some difficulties. This is mainly due to still high production costs and low productivity. A high yield of synthesis depends not only on the culture method, associated with the availability of nutrients for microorganisms, but also on the dynamic interaction of microorganisms with each other. The nutritional requirements of individual strains vary widely. Ramana and Singh [9] found that the best carbon source for the development of Acetobacter xylinum, the NUST4.1 strain, is glucose, and the growth of microorganisms and cellulose synthesis further increased in the presence of sodium alginate, while the growth of Acetobacter xylinum, the BRCS strain, is more dynamic with the additional presence of ethanol. An assessment of cellulose synthesis efficiency was conducted by Coban and Biyik [2] using variable sources of carbon and nitrogen. Fan et al. [3] evaluated the rate of synthesis and quality of bacterial cellulose on a substrate with the addition of waste from the food industry. In this work, attempts were made to assess the efficiency of bacterial cellulose synthesis by strains of microorganisms included in the Kombucha biofilm, using three types of media, differing in the amount of carbon source and the availability of nitrogen source. The

53 consumption rate of sucrose used as a carbon source was assessed in relation to cellulose synthesis efficiency.

MATERIALS AND METHODS Studies on the effect of nutrients on the intensity of cellulose synthesis by Kombucha microorganisms were carried out on three different media, the basic component of which was sucrose as a carbon source. Three types of media were prepared based on the literature data provided by Goh et al. [4] and Sharma and Behardwaj [3]. The sucrose content in the media ranged from 1.5% to 10%. The first type of medium contained only different sucrose concentrations. The second medium variant, apart from sucrose, contained a constant addition of vegetable peptone (0.25%), while the last medium variant contained a constant addition of tea extract (0.1%). The microorganisms were cultured under the temperature and humidity conditions of 24˚C and 68 ± 2%, respectively. The cultivation time of the microorganisms was 14 days. After a specified cultivation time, the synthesized cellulose was purified, weighed, then dried at 60˚C for 24h and weighed again. The dry mass and the percentage yield of polymer synthesis were determined based on the assumptions presented by Sharma and Behardwaj [10]: dry mass of polymer formed = mass of dried polymer [g]/volume of medium [l] percentage of synthesis [%] = (dry mass of polymer formed [g]/sum of the initial concentration of sucrose and additional component [g]) x 100

At the time of liquidation of the microorganism culture, the sucrose content was measured. The sucrose concentration in the post-culture fluid was determined polarimetrically, using a polarimeter model P3000 Kruss. After each measurement, the post-culture fluid was cleaned by syringe filters. The specific rotation of the sugar solution was measured in a 1 dm long glass tube. Due to the fact that the result of determinations obtained by the polarimetric method is the value of specific curvature, the concentration of sugar contained in the medium was determined from the following relationship: test concentration [%] = concentration of sugar solution at the beginning of the culture [%] x (specific torsion capacity of the test solution/specific torsion capacity of the solution of known concentration)

In order to determine the rate of consumption of the primary carbon source during cultivation, the sucrose consumption rate (g L-1 d-1), was determined, according to the assumption presented by Sharma and Behardwaj [10]: sucrose consumption rate (g L-1 d-1) = (sugar content in time 0 [g/l] - sugar content in final time K [g/l]) / cultivation time [d]

RESULTS Based on the results obtained, it can be concluded that the composition of the culture medium clearly affects the increase in dry matter of the biopolymer. In a medium richer in carbon source, the increase in dry matter of synthesized cellulose was clearly greater. On a medium containing 1.5% sucrose, the dry weight of the polymer produced was 14 times lower than the dry weight obtained on a medium containing as much as 10% sucrose (Table 1). Sucrose is the basic carbon source for yeast forming the Kombucha biofilm, used for the production of ethanol, which in turn is the main nutrient for acetic acid bacteria that synthesize cellulose. At the same time, it should be noted that the increase in bacterial cellulose dry matter also depends on the presence of nitrogen-rich components. The efficiency of cellulose synthesis in a nitrogen-rich environment was clearly higher than in a nitrogen-free environment. In the presence of peptone, the yield of cellulose synthesis was from 116% to

218% greater than in a medium without a nitrogen source (Table 2). Tea extract is also a nitrogen-rich component found in compounds such as alkali, peptides, and essential oils, however, the yield of bacterial cellulose synthesis in the presence of tea extract was significantly lower than in the presence of peptone. The reason for this phenomenon may be poorer availability of components present in the extract, but also their concentration or even the formation of by-products of their metabolism, which impede the synthesis of cellulose.

Table 1. The average value of obtained dry matter of cellulose synthesized by Kombucha microorganisms on various types of nutrient media Sucrose content [%] Peptone content [%] Tea extract content Dry mass of polymer [%] formed [g/l] 1.0 0.25 - 2.03 2.5 3.30 5.0 7.22 7.5 12.89 10.0 22.34 1.0 - 0.10 0.81 2.5 2.77 5.0 2.65 7.5 8.89 10.0 15.74 1.0 - - 0.2 2.5 1.36 5.0 1.08 7.5 2.24 10.0 2.81

Table 2. The average value of the obtained cellulose synthesis efficiency by Kombucha microorganisms on various types of nutrient media Sucrose content [%] peptone content [%] Tea extract content Polymer synthesis yield [%] [%] 1.0 0.25 - 116.0 2.5 120.0 5.0 137.0 7.5 166.0 10.0 218.0 1.0 - 0.10 50.6 2.5 106.5 5.0 51.9 7.5 117.0 10.0 155.8 1.0 - - 13.0 2.5 54.0 5.0 21.6 7.5 29.0 10.0 28.1

The increased efficiency of cellulose synthesis on substrates rich in nitrogen correlates with increased sucrose. On the medium containing peptone or tea extract, the sucrose content in the medium, after 14 days of culturing Kombucha microorganisms was lower than in the medium containing only sugar (Figure 1). Faster sugar consumption resulted in a greater increase in dry matter of synthesized cellulose. When determining the rate of sucrose consumption, it was found that in the case of microorganisms cultured on peptone-containing medium, it was from 0.050 to 0.380 g / L per day. The yield of cellulose synthesis with such daily sugar metabolism ranged from 116% to

218%. In a medium containingsucrose and tea extract, the sucrose consumption rate per day was lower than in the sucrose and peptone medium. Thus, a lower yield of cellulose synthesis was observed in this medium. The lowest sucrose consumption rate was found in the medium containing only sucrose as a nutrient for Kombucha microorganisms (Figure 2). Based on the above data indicated in Figures 1 and 2, the conversion of sugar to cellulose with the participation of Kombucha microorganisms was presented. The data shows that both the nutrient content with the necessary amount of carbon and the source of nitrogen have an impact on the mass of the synthesized polymer and its synthesis efficiency.

Figure 1. Sucrose content in growth media for Kombucha microorganisms

Figure 2. The rate of sucrose consumption by Kombucha microorganisms on various types of media

CONCLUSION The analysis of the obtained research results indicates that the yield of bacterial cellulose synthesis by Kombucha microorganisms is dependent on the quantity and quality of nutrients available in the growth medium. Both the amount of sucrose as a carbon source and the addition of substances rich in nitrogen affects the mass of cellulose obtained and its synthesis efficiency. In the presented work, it was found that the addition of peptone as a nitrogen source and an increased amount of sucrose in the medium clearly stimulate microorganisms for cellulose synthesis.

REFERENCES:

1. CAZÓN P., VELÁZQUEZ G., VÁZQUEZ M., 2019: Characterization of bacterial cellulose films combined with chitosan and polyvinyl alcohol: Evaluation of mechanical and barrier properties, Carbohydrate polymers nr 216; 72-85 2. COBAN E.P., BIYIK H., 2011: Effect of various carbon and nitrogen sources on cellulose synthesis by Acetobacter lovaniensis HBB5, African Journal Biotechnology nr. 10; 5346–5354 3. FAN X., GAO Y., HE W., HU H., TIAN M., WANG K., PAN S., 2016: Production of nano bacterial cellulose from beverage industrial waste of citrus peel and pomace using Komagataeibacter xylinus, Carbohydrate Polymers nr. 151; 1068-1072 4. GOH W.N., ROSMA A., KAUR B., FAZILAH A., KARIM A.A., BHAT R., 2012: Microstructure and physical properties of microbial cellulose produced during fermentation of black tea broth (Kombucha). II. International Food Research Journal nr. 19; 153–158 5. JEONG S. I., LEE S. E., YANG H., JIN Y.-H., PARK C.-S., & PARK Y. S. 2010: Toxicologic evaluation of bacterial synthesized cellulose in endothelial cells and animals, Molecular & Cellular Toxicology nr. 6(4); 370–377 6. URBINA L., CORCUERA M. Á., ECEIZA A., RETEGI A., 2019: Stiff all-bacterial cellulose nanopaper with enhanced mechanical and barrier properties, Materials Letters nr. 249; 67-70 7. MENENDEZ E., GARCIA-FRAILE P., RIVAS R., 2015: Biotechnological applications of bacterial cellulases, Bioengineering nr. 2(3) 163-182 8. ROUABHIA M., ASSELIN J. R. M., TAZI N., MESSADDEQ Y. S., LEVINSON D., ZHANG Z. 2014: Production of biocompatible and antimicrobial bacterial cellulose polymers functionalized by RGDC grafting groups and gentamicin, ACS Applied Materials & Interfaces nr. 6(3); 1439–1446 9. RAMANA K. V., SINGH L., 2000: Effect of various carbon and nitrogen sources on cellulose synthesis by Acetobacter xylinum, World Journal Microbiol. Biotechnol nr. 16 (3); 245–248 10. SHARMA CH., BHARDWAJ N.K., 2019: Biotransformation of fermented black tea into bacterial nanocellulose via symbiotic interplay of microorganisms, International Journal of Biological Macromolecules nr. 132; 166-177

Streszczenie: Synteza celulozy bakteryjnej przez mikroorganizmy Kombucha na podłożu przy zmiennym udziale składników pokarmowych. W pracy przedstawiono wyniki oceny wpływu różnych zawartości sacharozy oraz obecności różnych źródeł związków azotu w podłożu wzrostu mikroorganizmów Kombucha na wydajność syntezy oraz uzyskaną masę celulozy bakteryjnej. Analizując uzyskane wyniki badań, stwierdzono, że wydajność syntezy celulozy przez mikroorganizmy Kombucha jest zależna od ilości i jakości składników pokarmowych dostępnych w podłożu wzrostu.

Corresponding author:

Izabela Betlej, ul. Nowoursynowska 159, 02-787 Warszawa email: [email protected]

ORCID ID: Betlej Izabela 0000-0001-6867-0383 Krajewski J. Krzysztof 0000-0001-8716-1488

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 58-67 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Resistograph investigation of Scots pine wood utility poles in the State Museum at Majdanek

WOJCIECH KORYCIŃSKI1, KRZYSZTOF J. KRAJEWSKI 2,PAWEŁ KOZAKIEWICZ2 1 University of Life Sciences in Lublin, Faculty of Agrobioengineering, Department of Industrial and Medicinal Plants 2 Warsaw University of Life Science – SGGW, Institute of Wood Sciences and Furniture, Department of Wood Science and Wood Protection

Abstract: Resistograph investigation of Scots pine wood utility poles in the State Museum at Majdanek. Any activity relative to the protection of monuments is determined by the requirements of fidelity and authenticity in the preservation of the place and landscape. On the site of the State Museum at Majdanek, the former infrastructure of the concentration camp has been reconstructed. An element there of are pine wood utility poles.The present research project involved an assessment of their state of preservation with the method of resistography. The poles were subjected to inspection and preliminary acoustic assessment by means of tapping. Resistograph drillings were made radially, perpendicularly to the side surface of the poles, at various heights. A number of the poles have been found to be highly degraded in their sapwood part, which threatens their stability – these poles require immediate replacement. The principal cause of the degradation areactive feeding grounds of European house borer.The results of the research confirmed the effectiveness of resistographyin on- site assessment of the state of preservation of wooden poles.

Keywords: Scots pine wood, natural durability, resistography, wooden pole, protection of monuments, Majdanek

INTRODUCTION Scots pine ranks first among forest stands in Poland. As much as 58% of the total forest area (ca. 6 m. ha) corresponds to this species, which gives the stock of large timber at 950 m. m3 [Lasy w Polsce ‘Forests in Poland’ 2018].Pinewood, as the species of wood dominant on the Polish market for decades, was and still is widely used [Kozakiewicz 2019]. It served as construction timber for the purpose, among others, of constructing utility poles [Galewski and Korzeniowski 1958]. Pinewood has good mechanical properties; it shows, however, high variability depending also on the burden of inherent defects which result from the wood’s anatomical structure [Warywoda 1957, Galewski and Korzeniowski 1958, Wagenführ 2007]. For this reason, construction timber has to be sorted and has to comply with relevant standards. The requirements for utility poles are at present comprised in ISO 15206:2010; formerly they were stated in the overall technical requirements for utility poles to be applied in measurement and classification of timber in PGLLP (http://drewno.zilp.lasy.gov.pl/drewno/Normy/). Pinewood has medium natural durability (resistance to biotic factors). The resistance to fungi, defined and published in standard EN 350:2016, of pine hardwood zone amounts to 3-4, and that of sapwood to 5 (the lowest durability).In accordance with data quoted in the above standard, pine sapwood is also susceptible to insect attack, including that of European house borer. For this reason, it is advised that wood for external use should be impregnated. The effectiveness of the impregnation is determined, among others, by the penetration depth of the impregnating substance; in previous centuries it was enhanced by perforating the wood surface. Utility poles were impregnated in the same way [Bub-Bodmar and Tilger 1922, Kozakiewicz and Matejak, 2011]. Despite impregnation treatment, unfortunately, wooden utility poles often undergo degradation as a result of biotic factors. One of the non-destructive methods used to assess the technical condition of wood is resistography. This method was developed in 1985, and then gradually perfected, together

58 with the drilling appliances dedicated to the purpose [Rinn et al. 1990], and finally patented [Rinn 1990]. The usefulness of resistography in analyzing annual growth rings and density profiles of wood was confirmed in many research projects [e.g. Rinn et al. 1996, Winnistorfer et al. 1995, Kubus 2009, Rinn 2012], and so was its effectiveness in evaluating the health of wooden elements [e.g. Bernatowicz and Krajewski1998, Krajewski and Andres 2003]. The purpose of this work is to assess the state of preservation of pinewood utility poles on the site of State Museum at Majdanek by using resistography. The requirements of a museum exhibition and modern technical standards do not always meet. Even if a certain solution is not the best one from the viewpoint of construction health it is often forced in an element of landscape which materially affects the perception of museum space. For these reasons, some of the wooden construction elements in historical monuments are at a higher risk of suffering from the effects of exposure to destructive factors.

MATERIAL AND METHODS In July 2018, research into the state of preservation of wooden utility poles located on the site of the State Museum at Majdanek was carried out. The work was preceded by an analysis of museum documentation (approximation of the poles’ age) and confirmatory identification of wood species. The examination included a survey of the poles with regard to their compliance with the overall technical requirements for utility poles to be applied in the measurement and classification of timber in PGLLP (http://drewno.zilp.lasy.gov.pl/drewno/Normy/). So-called non-destructive tests were performed with IML 1410 resistograph which had the following characteristics: resolution0.04 mm, drilling depth up to 410 mm, feed rate 60 to 1500 mm/min. The Resistograph 1410 system consists of a drill, a bag with control panel, batteries, printer, and built-in memory. Printouts at 1:1 scale are made simultaneously with the drilling for which special dedicated drills of unique geometry are used, as presented in figure 1.

Figure 1. Shape and dimensions (in mm) of the tip of the resistograph steel drill

In total, several dozen bores were drilled in 20 out of 56 wooden poles located on the Museum site. All the drillings performed during the research were made at a feed rate of 200 mm/min. The resultant records of power input (drilling resistance) correlated with wood density were analyzed by using the B-Tools Pro 1.6 specialist program dedicated for IML resistographs. Measurements of wood moisture content were carried out with a MastechMS6900 hygrometer using the resistance method in accordance with EN 13183- 2:2002.

RESULTS AND DISCUSSION The utility poles which are at present located on the site of the State Museum at Majdanek do not date back to the time of World War II; the former poles were dismantled 59 immediately there after and used in rebuilding other objects damaged after the war. The new poles were installed ca. 1962 during the restoration of camp infrastructure. Until then, the power line had been only suspended directly between barrack roofs. a) b)

Figure 2. Majdanek camp - view of the entrance to Field III. a) immediately after the liberation in 1944 (reproduction of the plaque from the exhibition at the State Museum at Majdanek), b) current state photo of the same place (2019)

The poles were not reconstructed in the original places, which can be seen when comparing archive photographs (figures 2 – 4).It is not only the location of the poles that differs from the original. The present-day route of the power line is also different from that visible in photographs and archival camp electrification plans.

Figure 3. The camp at Majdanek after liberation in autumn 1944. - view of the western fence of the prisoner section with the original overhead power line visible (reproduction of the plaque from the exhibition at the State Museum at Majdanek).

a) b)

Figure 4. The State Museum at Majdanek - view of the western fence of the prisoner section a) condition from 1957 (photo: Edward Hartwig reproduction from PMM collection reference number XVII.7.4.20), b) current condition photo of the same place with a visible overhead power line (2019)

The present position of utility poles on the site of State Museum at Majdanek is shown in Figure 5. The location of new poles has been adapted to current needs and technical requirements. The poles that are in use at present were installed at different times and replaced gradually as damage appeared. They were made in accordance with the regulations and standards which were binding at the time of their reconstruction. The latest replacements were carried out in 2016; the five poles installed at that time were already supplied with the required CE certificates which confirmed their compliance with European standards (ISO 15206:2010).The former, original poles were sunk into the ground; at present they are all raised above ground level and fastened on reinforced concrete crutches. This prevents biodegradation of the poles’ lower sections. Moreover, the lower sections have been impregnated with bituminous substances. An analysis of the wood’s structural characteristics confirmed that the poles were made of round timbers of Scots pine (Pinus sylvestris L.). A survey of the poles with reference to the overall technical requirements for utility poles to be applied in measurement and classification of timber in PGLL Prevealed quality discrepancies in three of the 56 poles examined. Two poles had irregular twisting of fibers where the deviation from straight line exceeded 90° measured in the pole’s top half, one had two-sided curvature and a rotten knot whose diameter exceeded 60 mm (Fig.6). Most of the poles had marks of resin blazes, in none of them, however, excessively deep streaks were found.

Figure 5. The current arrangement of wooden utility poles on the site of the State Museum at Majdanek - damaged poles are marked in black and the remaining poles (examined with a resistograph) are marked in gray a) b)

Figure 6. Examples of poles with unacceptable defects: a) utility pole No. 5 with visible excessive twisting of fibers, b) utility pole No. 13 with visible curvature.

The length of electrodes permitted for the moisture content measurements to be taken only in the circumferential sections of the poles. The measurements were carried out in summer, in warm, dry weather. No increased moisture content was noted in any of the

62 examined poles. All the measurements fell within the scope of hygroscopic equilibrium of wood in conformity with the environment conditions. Wood moisture content ranged from 10.1% to 11.2%. The moisture content value from all measurements averaged 10.7% with variability ratio at 2.7%. Following a preliminary acoustic assessment by means of tapping, 20 poles were singled out for resistograph testing, several of them with distinct beat effect, several others with correct sound, for reference (figure 5). Resistograph tests disclosed pole damage to occur usually at the height ca. 1 m above ground level. The damage was caused by feeding larvae of European house borer (Hylotrupes bajulus L.). Despite the lack of serious external damage visible to the naked eye, resistograph tests revealed reduced wood density in eleven poles; in nine of them the reduction was very considerable. In all these cases the damaged area was the sapwood. In three poles, the sapwood at the place of measurement was almost entirely destroyed (no drilling power registered). A typical result for such a case (pole No. 8) is presented in figures 7 and 8. The side surface of the pole showed marks of a resin blaze as well as distinct cracks and a few exit holes of European house borers’ galleries. The total width of the damaged area was approximately 11 cm on a diameter of 22 cm. This corresponded approximately to the share of sapwood zone in the trunk of a mature pine tree (Dzbeński et. al. 2000). The damage to the sapwood zone varied in intensity with the height above ground level. Measurements at different heights showed diversed is tribution of density, well visible on the printouts of pole No. 33 tests (figures 9 and 10).Wood examined nearer its bottom end did not show changes in density. It was only above the concrete crutch holding the pole that reduced density in the sapwood part was clearly observable.

Figure 7. Fragment of utility pole No. 8 with the printout of drilling power correlated with the density profile along its entire diameter at a height (photo: Rafał Krajewski, 2018).

Figure 8. Analysis of the results of resistographic tests of utility pole No. 8 using the B-Tools Pro 1.6 program - the gray shade indicates fragments with typical wood density and black with reduced density

A lesser extent of damage, comprising only a fragment of the sapwood ring, was observed in eight poles, as exemplified in figure 10 (pole No. 33). Bores made at the same height but from different directions did not reveal losses in wood density. No changes in wood density were noted in the bottom sections of the poles (figure 9), despite the damage observed in the upperparts. Resistograph tests are effective in evaluating the state of preservation of wooden elements [Bernatowicz and Krajewski 1998, Krajewski and Andres 2003] but they require a comprehensive approach. In the analyzed case of utility poles, the visual inspection for the occurrence of defects and the test drillings were equally important.

Figure 9. Changes in wood density in utility pole No. 33 - a bore at a height of 100 mm from the bottom end shows no change in density - fragments with a typical (average for pine wood) density level are marked in gray.

Figure 10.Changes in wood density in utility pole No. 33 - a hole at a height of 700 mm from the bottom end shows local density reduction - the gray shade indicates typical wood density and black shade reduced wood density

It is significant that the sapwood damage observed in the poles begins at the depth ca. 1 cm from the side surface. The thin surface layer of sapwood provides a kind of thermal insulation for the stenothermic (Unger et al. 2013) larvae of the insect, which feed a little deeper. Some of the poles without visibly or audibly perceptible changes were preventively tested with the resistograph and showed no reduction in wood density. It is necessary to monitor the preservation of pinewood elements continuously, because of a high risk of them being infested by xylophages insects which cause considerable loss of material. The resistograph method is highly useful in conducting such tests, as it allows for a quick examination of the state of wood constructions.

CONCLUSIONS On the basis of visual examination and resistograph tests of pinewood utility poles at the site of the State Museum at Majdanek, the following conclusions have been drawn: 1. With regard to the presence of anatomical defects permissible in accordance with overall technical requirements for utility poles, three out of 56 poles examined did not meet the qualitative conditions. 2. The moisture content of the poles’ near-surface wood amounted to ca.10.7%, which conformed with the weather conditions existing at the time. By using the acoustic method of tapping, a possibility of internal destruction in more than ten poles was discovered. 3. As a result of resistograph tests, the presence of serious damage of sapwood areas infested by European house borer (Hylotrupes bajulus L.) was confirmed, and the extent of such damage was determined in 8 utility poles. 4. The effectiveness of resistograph tests in wooden structures with insect-caused damage requires selecting appropriate places for bores as well as making several diametrical trial drillings at various heights and from various directions.

REFERENCES

1. BERNATOWICZ G., KRAJEWSKI K.J., 1998: Wykorzystanie rezystografu do oceny stanu technicznego drewnianych podwalin kościoła w Boguszycach. Ochrona Drewna – XIX Sympozjum Jadwisin 15-17 września 1998 r. Wydawnictwo SGGW. Warszawa, 163-172. 2. BUB-BODMAR F., TILGER B. 1922: Die Konservierung des Holzes in Theorie und Praxis. Berlin. Verlagsbuchhandlung Paul Parey. 3. DZBEŃSKI W., KOZAKIEWICZ P., KRUTUL D., HROL J., BELKOVA L., 2000: Some physical and mechanical properties of Scots pine (Pinus sylvestris L.) from Rogów as a comparative material for research on the Latvian provenance pine. Materials of the 14th Conference WTD SGGW „Wood material of all times”. Rogów 2000, 13-15 November 2000. pp.: 31-36. 4. EN 13556:2003 Round and sawn timber – Nomenclature of timbers used in Europe. 5. EN 350:2016 Durability of wood and wood-based products – Testing and classification of the durability to biological agents of wood and wood-based materials. 6. GALEWSKI W., KORZENIOWSKI A., 1958: Atlas najważniejszych gatunków drewna. Państwowe Wydawnictwo Rolnicze i Leśne. 7. http://drewno.zilp.lasy.gov.pl/drewno/Normy/ (ramowe warunki techniczne na słupy teletechniczne mające zastosowanie do pomiaru i klasyfikacji surowca drzewnego w PGLLP). 8. EN 13183-2:2002 Moisture content of a piece of sawn timber. Estimation by electrical resistance method. 9. ISO 15206:2010 Timber poles - Basic requirements and test methods. 10. KOZAKIEWICZ P., 2019: Sosna zwyczajna (Pinus sylvestris L.) - polskie drewno. Przemysł Drzewny Research& Development nr 2/2019 (25), str. 72-77. 11. KOZAKIEWICZ P., MATEJAK M., 2011: Nakłuwanie słupów i podkładów. Przemysł Drzewny nr 4 (Rok LXII). S.35-36. 12. KRAJEWSKI K.J., ANDRES B., 2003: Przydatność tomografii impulsowej i metody rezystograficznej do oceny stopnia uszkodzenia drewnianych elementów konstrukcyjnych w budownictwie. Ochrona Przed Korozją. Miesięcznik Stowarzyszenia Inżynierów i Techników Przemysłu Chemicznego Numer Specjalny 10s/A (Rok XLVI), 109 – 114. 13. KUBUS M., 2009: The evaluation of using resistograph when specifying the health condition of a monumental tree. Not. Bot. Hort. Agrobot. Cluj 37 (1), 175-164. 14. LASY W POLSCE 2018. Centrum Informacyjne Lasów Państwowych. ORWLP w Bedoniu. Warszawa. 15. RINN F., 1990: Device for material testing, especially wood, by drill resistance measurements. German Patent 4122494. 16. RINN F., 2012: Basics of typical resistance-drilling profiles. Western Arborist (Winter 2012): 30-36. 17. RINN F., BECKER B., KROMER B., 1990: Ein neues Verfahren zur direkten Messung der Holzdichte bei Laub- und Nadelhölzern. Dendrochronologia 7, 159-168. 18. RINN F., SCHWEINGRUBER F.H., SCHÄR E., 1996: Resistograph and X-ray density charts of wood comparative evaluation of drill resistance profiles and Z-ray density charts of different wood species. Holzforschung 50: 303-311. 19. UNGER A., SCHNIEWIND A.P., Unger W., 2013: Conservation of Wood Artifacts: A Handbook. Springer-Verlag Berlin Heidelberg New York.

20. WAGENFÜHR R., 2007: Holzatlas, neu bearbeitete und erweitere Auflage. Mit 890 zum Teil mehrfarbigen Bildern. Fachbuchverlag Leipzig im Carl Hanser Verlag., München, Germany. 21. WARYWODA A., 1957: Encyklopedia techniczna. Drzewa użytkowe w architekturze przestrzennej i przemyśle. Krakowski Zespół Pracowników Naukowych. Krakowskie Zakłady Graficzne. Kraków. 22. WINNISTORFER P.M., XU W., WIMMER R., 1995: Application of a drill resistance technique for density profile measurement in wood composite panels. Forest Products Journal 45(6), s. 90-93.

Streszczenie: Badania rezystograficzne sosnowych słupów teletechnicznych na terenie Państwowego Muzeum na Majdanku. Na terenie Państwowego Muzeum na Majdanku częściowo odtworzono dawną infrastrukturę obozu. Jednym z jej elementów są sosnowe słupy teletechniczne. W ramach niniejszej pracy oceniono stan ich zachowania z wykorzystaniem metody rezystograficznej. W pierwszej kolejności obiekty poddano oględzinom pod kątem spełnienia ramowych warunków technicznych na słupy teletechniczne oraz wstępnej ocenie metodą dźwiękową przez opukiwanie, a także pomiarom wilgotności drewna. W wytypowanych 20 słupach dokonano średnicowych wierceń rezystografem w ich przekrojach poprzecznych na różnych wysokościach. Na postawie analizy danych z rezystografu stwierdzono, że część słupów jest silnie zdegradowana w części bielastej, co grozi utratą ich stateczności – słupy te wymagają natychmiastowej wymiany. Głównym czynnikiem powodującym degradację są aktywne żerowiska larw spuszczela pospolitego, które znalazły tam dogodne warunki bytowania. W ramach podjętych badań potwierdzono przydatność i skuteczność metody resystograficznej do terenowej oceny stanu zachowania słupów drewnianych.

Corresponding author:

Paweł Kozakiewicz Department of Wood Sciences and Wood Preservation Institute of Wood Sciences and Furniture Warsaw University of Life Sciences – SGGW 166 Nowoursynowska St. 02-787 Warsaw, Poland email: [email protected] http://pawel_kozakiewicz.users.sggw.pl phone: +48 22 59 38 647 fax: +48 22 59 386 59

ORCID ID: Krajewski J., Krzysztof 0000-0001-8716-1488 Kozakiewicz Paweł 0000-0002-2285-2912

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 68-78 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Modelling of nitrogen implantation processes into WC-Co indexable knives for wood-based material machining using ion implanters with or without direct ion beam

MAREK BARLAK1, JACEK WILKOWSKI2, ZBIGNIEW WERNER1 1 Plasma/Ion Beam Technology Division (FM2), National Centre for Nuclear Research Świerk - NCBJ 2 Department of Mechanical Processing of Wood, Warsaw University of Life Sciences - SGGW

Abstract: Modelling of nitrogen implantation processes into WC-Co indexable knives for wood-based material machining using ion implanters with or without direct ion beam. The paper presents the results of modelling of the depth profiles of nitrogen implanted to W-C-Co material using ion implanters with or without direct ion beam. The modelling was performed using the Monte Carlo method. The results were obtained for one fluence of the implanted ions and three different values of the acceleration voltage.

Keywords: WC-Co composites, ion implantation, modelling, Monte Carlo method

INTRODUCTION Cemented tungsten carbides - a combination of hard and brittle carbides and a relatively soft and ductile metallic binder, gives rise to an exceptional combination of attractive properties such as strength, hardness, fracture toughness, refractoriness, stiffness, resistance to compressive deformation and wear resistance at room as well as at higher temperatures up to 400°C (Milman et al. 1997, Sheikh-Ahmad and Bailey 1999, Pirso et al. 2004, Bonny et al. 2010, Choi et al. 2010, Olovsjö et al. 2013). Unfortunately, the durability of tools made of this material is still insufficient. There are several methods to improve this property, ion implantation being one of them (Barlak et al. 2016, Barlak et al. 2017). This low temperature method is used in material engineering processes to modify/change the physical and/or chemical properties of the near-surface regions of solid materials (targets) through the introduction of foreign element(s) ions. The introduced ions are accelerated in the electric field to an energy from several dozen to several hundred kiloelectronvolts, which corresponds to their velocity from hundreds to thousands of kilometres per second. The ion beam interacts with the modified material, introduces new atoms, damages its crystal lattice, creates vacancies and other defects up to total amorphisation. The substrate atoms and early implanted atoms are ejected from the surface (Fig. 1). The modified region is not an additional layer, hence no adhesion problem occurs (no delamination), and the change of dimensions and of the surface finish of the implanted material is negligible.

Figure 1. Diagram of the ion implantation process

There are two main types of ion implanters, i.e. implanters without mass separation (with direct ion beam) and with mass separation (without direct ion beam) - Fig. 2. A separating magnet is used for mass separation of the ion beam in order to obtain a homogenous beam in terms of constant e/m ratio.

Figure 2. Diagram of an ion implanter without mass separation (left) and with mass separation (right)

The ion beam generated by an ion implanter without mass separation contains several kinds of ions with different degrees of ionization. Additionally, the percentage shares of individual ion types are also different (Table 1). The use of all data for the selected element is usually cumbersome, therefore the average charge state (ACS) is a more popular parameter for the calculation/modelling as an equivalent of the sum of profiles for the individual ion 69 kinds. The average charge state is determined by adding the values of multiplication of the percentage shares and the ionization’s degree. For example, for yttrium it will be:

0.07·1 + 0.63·2 + 0.29·3 + 0.01·4 = 2.24 ≈ 2.2 (1)

The situation is slightly different in the case of ion implantation of nitrogen. Nitrogen + + beam contains two kinds of ions, i.e. N +N2 , in the ratio ~1:1, so there are two elementary charges per three atoms. With the acceleration voltage of 50 kV, one per three nitrogen atoms (i.e. 0.33 of atoms) carries the energy of 50 keV and two atoms (nitrogen molecule, i.e. 0.67 of atoms) carry the energy of 25 keV each. Likewise, the average charge state is at the level of 0.67.

Table 1. The percentage charge state distribution and average charge state of several elements (Krivonosienko et al. 2001 and the author’s own result, unpublished) Ion energy for acceleration voltage of 50 kV (keV) Average Implanted ions 25 50 100 150 200 250 charge Percentage charge state distribution (%) state 1+ 2+ 3+ 4+ 5+ Li+ - 100 - - - - 1.0 + + N2 + N 67 33 - - - - 0.67 Y+ + Y2+ + Y3+ + Y4+ - 7 63 29 1 - 2,2 Re+ + Re2+ + Re3+ + Re4+ + Re5+ - 2 26 40 26 3 3.0 Pb+ + Pb2+ - 40 60 - - - 1.6

Fig. 3 shows an example modelling of ion implantation to W-C-Co substrate material (modelling codes treat the sample as a set of atoms that do not form chemical compounds), without taking into account the phenomenon of sputtering, for two selected elements, from Table 1, i.e. for lead with two charge state ions and rhenium with five ones, for the acceleration voltage of 50 kV. The profiles named Pb+ + Pb2+ and Re+ + Re2+ + Re3+ + Re4+ + Re5+ were modelled for an implantation without mass separation. The ACS profiles are the equivalent of the previous profiles modelled for the average charge state. The single profiles Pb+ and Pb2+ for lead and Re+, Re2+, Re3+, Re4+ and Re5+ for rhenium were modelled for the individual charge state and the percentage share in the ion beam. The sum of these individual profiles should give the suitable profile Pb+ + Pb2+ or Re+ + Re2+ + Re3+ + Re4+ + Re5+.

Figure 3. The modelled profiles for the ion implantation of lead and rhenium to W-C-Co substrate

All the profiles presented above have the shape close to a Gaussian curve (known also as normal distribution or bell curve) and can be defined with three main parameters, i.e. the peak -2 volume dopant concentrations (Nmax, cm ), the projected range (Rp, nm) and the range straggling (ΔRp, nm) - Fig. 4 (upper, left). Sometimes, two additional parameters, i.e. the kurtosis and the skewness, are used to describe the divergence in relation to normal distribution (measures of shape). The coefficient of kurtosis is a measure of the degree of peakedness/flatness in variable distribution - Fig. 4 (upper, right). A normal distribution is a mesokurtic distribution. A leptokurtic distribution (positive kurtosis) has a high degree of peakedness (higher peak) and heavier tails, and vice versa, a platykurtic distribution (negative kurtosis) has a low degree of peakedness (lower peak) and lighter tails. The coefficient of skewness is a measure of the degree of asymmetry in variable distribution - Fig. 4 (lower). A positively skewed distribution, named also skewed to the right, has a tail pulled in the positive direction, and vice versa, a negatively skewed distribution, named also skewed to the left, has a tail pulled in the negative direction (Surbhi 2017).

Figure 4. The graphical definitions of the main peak parameters, the kurtosis and the skewness

The present paper provides information about the modelling of nitrogen ion implantation (a typical element used to improve the durability/tribological properties) to W-C-Co material, using implanters without or with mass separation. The obtained results can be applied for example for the design of ion implantation processes of WC-Co indexable knives for wood- based material machining, as presented in Fig. 5. This paper is an extension of the information presented in Ref. (Barlak 2019).

Figure 5. WC-Co indexable knives

MATERIAL AND METHODS The Stopping and Range of Ions in Matter (SRIM-2013.00), freeware type code was used for the modelling of the depth profiles of nitrogen implanted to W-C-Co material (SRIM 2019). SRIM is a group of software packages which calculate many features of the transport of ions in matter and it includes quick calculations to generate e.g. tables of stopping powers, range and straggling distributions for any ion at any energy in any elemental target. The SRIM code is based on a Monte Carlo simulation method. It is a method used for mathematic modelling of processes that are too complex to predict their results using an analytical approach. An important role in this method is played by draw (random selection) of quantities characterizing the process, with the random selection being carried out in accordance with a distribution that must be known (MCM 2014). The modelling was performed for 100 000 implanted ions of nitrogen. The composition of the W-C-Co material was: 90.86% of tungsten, 5.94% of carbon and 3.2% of cobalt in weight percentages i.e. 47.4% of tungsten, 47.4% of carbon and 5.2% of cobalt in atomic percentages was defined as a target. Its density adopted for the simulation was 15.2 g/cm3. This value was declared by the knives supplier (KCR08 type knives, by Ceratizit, Austria). The simulations were performed for room temperature and for the total implanted fluence of 1e17 cm-2, including percentage charge state distribution data from Table 1. The acceleration voltage was 5, 50 and 500 kV. The modelling did not take into account the phenomenon of substrate sputtering by the implanted ions. The values of peak volume dopant concentrations Nmax, projected range Rp, range + + straggling ΔRp, kurtosis and skewness were determined for all nitrogen profiles, i.e. N + N2 , + + ACS, N and N2 , and in the next step - they were compared with the values obtained for the + + N + N2 profiles.

RESULTS AND DISCUSSION + + + Fig. 6 presents the depth profiles for all kinds of nitrogen ions i.e. N + N2 , ACS, N + -2 and N2 , for the acceleration voltage of 5, 50 and 500 kV and the fluence of 1e17 cm . It is + + evident that N + N2 and ACS profiles are similar for 5 and 50 kV. A similar situation is + + + + observed when comparing the N + N2 and N2 profiles, but in this case N2 peak is higher + + and narrower than N + N2 peak, also for 5 and 50 kV. The greatest difference is observed + + + + + + between N and N + N2 peaks. N peak is lower and wider than N + N2 peak for the first two cases. The situation is different for 500 kV. In this case, the differences are large due to + + the saddle character of the curve of N + N2 .

Figure 6. The depth profiles of nitrogen ions implanted to W-C-Co material

All detailed data, i.e. the values of peak volume dopant concentrations, projected range, range straggling, kurtosis and skewness, characterizing the modelled depth profiles of the implanted nitrogen, have been presented in Table 2.

Table 2. Data characterizing the modelled depth profiles of the implanted nitrogen Parameter Peak volume Projected Range Implanted ions dopant range straggling Kurtosis Skewness concentration -3 Rp (nm) ΔRp (nm) Nmax (cm ) 5 kV N+ + without mass 1.07e23 5.2 6.6 4.2934 1.0169 N + separated beam 2 ACS 1.05e23 5.3 6 2.8386 0.6298 with mass N+ 7.99e22 7 7.8 2.9655 0.557 + separated beam N2 1.28e23 4.3 4.8 1.1341 1.0947 50 kV N+ + without mass 2.07e22 31.6 37.8 3.6237 0.8576 N + 2 separated beam ACS 1.99e22 31.8 32.4 2.4251 0.3096 with mass N+ 1.44e22 44.4 43.8 2.4435 0.1964 + separated beam N2 2.43e22 25 26 2.5837 0.3286 500 kV N+ + without mass 4.22e21 230.1 210.8 1.5395 0.6113 N + 2 separated beam ACS 4.97e21 233 162.8 2.7197 -0.4509 with mass N+ 3.98e21 325.2 200.8 1.6609 -0.4737 + separated beam N2 5.56e21 181.4 136.4 2.2233 -0.2815

The percentage change of the determined values of all the above parameters in + + comparison to the values obtained for N + N2 profiles is presented in Table 3 and additionally, for better visualization, in Fig. 7. The character of changes of all the parameters is generally the same for each kind of ion implantation. The values of peak volume dopant concentrations and the values of kurtosis increase together with the energy of the implanted ions. On the contrary, the values of range straggling and skewness decrease together with ion energy, The values of the projected range are approximately constant for the ACS case, slightly increase for N+ ions and slightly + decrease for N2 ions. The change in the value of peak volume dopant concentrations is close to zero for + + ACS, negative for N ions and positive for N2 . The range of the change for all profiles spans from -25% to 32%. On the contrary, the change in the projected range is positive for ACS and N+ ions, + and negative for N2 . In this case, the changes for all data are in the range from about -21% to 41%. + + The range straggling is negative for ACS and N2 , and mixed for N , in the range from about -35% to 18%. The kurtosis and the skewness show the largest changes in value, especially for ACS + and N2 ions. The change in kurtosis is from about -74% to 77%, and the change in skewness is from about -178% to 8% (nearly 190% for all the data).

+ + Table 3. The percentage change of depth profiles data in comparison to the values obtained for N + N2 profiles Change of value (%) Peak volume Implanted ions Projected Range dopant Kurtosis Skewness range straggling concentration 5 kV + + without mass N + N2 - - - - - separated beam ACS -1.9 1.9 -9.1 -33.9 -38.1 with mass separated N+ -25.3 34.6 18.2 -30.9 -45.2 + beam N2 19.6 -17.3 -27.3 -73.6 7.7 50 kV + + without mass N + N2 - - - - - separated beam ACS -3.9 0.6 -14.3 -33.1 -63.9 with mass separated N+ -30.4 40.5 15.9 -32.6 -77.1 + beam N2 17.4 -20.9 -31.2 -28.7 -61.7 500 kV + + without mass N + N2 - - - - - separated beam ACS 17.8 1.3 -22.8 76.7 -173.8 with mass separated N+ -5.7 41.3 -4.7 7.9 -177.5 + beam N2 31.8 -21.2 -35.3 44.4 -146

Figure 7. The graphical presentation of changes in the value of peak volume dopant concentrations Nmax, projected range Rp, range straggling ΔRp, kurtosis and skewness

Globally, the value of the change in peak volume dopant concentrations, projected range and range straggling is relatively low for ACS for all ion energies (in comparison to the + + + ion implantation of N + N2 ). A larger change of these parameters appears for N2 and for N+. The peak shape changes in all cases. The largest value of the change is for an ion energy of 500 keV.

CONCLUSION Based on the comparison of the modelled depth profiles of nitrogen implanted to W-C-Co material, with the use of implanters with or without direct beam, the following conclusions can be formulated: + + + 1. There are no great differences between the ion implantation with N + N2 and N2 beam, especially for the acceleration voltage at the level of 5 kV. This situation is only slightly

76 worse for the acceleration voltage at the level of 50 kV, and completely different for 500 kV, where much greater distances and changes in the shape of the profile are observed. 2. The ion implantation with N+ beam can’t be an equivalent of the implantation with + + N + N2 beam in any case. + + 3. The ACS profile can be a good equivalent of the implantation with N + N2 beam for the ion energy range from several to several dozen kilovolts.

REFERENCES

1. BARLAK M., WILKOWSKI J., WERNER Z., 2016: Ion implantation changes of tribological and corrosion resistance properties of materials used in wood industry, Annals of Warsaw University of Life Science - SGGW, Forestry and Wood Technology 94, 19-27. 2. BARLAK M., WILKOWSKI J., BORUSZEWSKI P., WERNER Z., PAŁUBICKI B., 2017: Changes of functional properties of materials used in wood industry after ion implantation processes, Annals of Warsaw University of Life Science - SGGW, Forestry and Wood Technology 97, 133-139. 3. BARLAK M., WILKOWSKI J., WERNER Z., 2019: Modelling of the ion implantation modification of WC-Co indexable knives for wood machining, Annals of Warsaw University of Life Science - SGGW, Forestry and Wood Technology 106, 57- 61. 4. BONNY K., DE BAETS P., PEREZ Y., VLEUGELS J., LAUWERS B., 2010: Friction and wear characteristics of WC-Co cemented carbides in dry reciprocating sliding contact, Wear 268, 1504-1517. 5. CHOI S.-H., KANG S.-D., KWON Y.S., LIM S.G., CHO K.K., AHN I.-S., 2010: The effect of sintering conditions on the properties of WC-10wt%Co PIM compacts, Research on Chemical Intermediates 36, 743-748. 6. KRIVONOSIENKO A.W., NIKOLAEV A.G., LI S., 2001: Технические описание и инструкция по эксплуатаци ионного источника “Титан-3” (Technical descriptions and operating instructions of the ion source “Titan-3”), Российская Академия Наук - Институт Сильноточной Электроники, Tomsk, in Russian. 7. MCM, 2014: Monte Carlo method, https://www.sciencedirect.com/topics/engineering/monte-carlo-method 8. MILMAN. YU.V., CHUGUNOVA S., GONCHARUCK V., LUYCKX S., NORTHROP I.T., 1997: Low and high temperature hardness of WC-6 wt%Co alloys. International Journal of Refractory Metals and Hard Materials 15, 97-101. 9. OLOVSJÖ S., JOHANSON R., FALSAFI F., BEXELL U., OLSSON M., 2013: Surface failure and wear of cemented carbide rock drill buttons - The importance of sample preparation and optimized microscopy settings, Wear 302, 1546-1554. 10. PIRSO J., LETUNOVITŠ S., VILJUS M., 2004: Friction and wear behaviour of cemented carbides, Wear 257, 257-265. 11. SHEIKH-AHMAD J.Y., BAILEY J.A., 1999: High-temperature wear of cemented tungsten carbide tools while machining particleboard and fiberboard. Journal of Wood Science 45, 445-455. 12. SRIM, 2019: Interactions of ions with matter, http://www.srim.org/ 13. SURBHI S., 2017: Differences Between Skewness and Kurtosis, https://keydifferences.com/ differences-between-skewness-and-kurtosis.html

Streszczenie: Modelowanie procesów implantacji jonów azotu do stosowanych w obróbce materiałów drzewnych wymiennych noży WC-Co, przy użyciu implantatorów bez lub z wiązką bezpośrednią. W artykule przedstawiono wyniki modelowania głębokościowych profili azotu, implantowanego do materiału W-C-Co, przy użyciu implantatorów bez lub z wiązką bezpośrednią. Modelowanie przeprowadzono przy użyciu metody Monte Carlo, dla jednej wartości dawki implantowanych jonów oraz trzech różnych wartości napięcia przyspieszającego.

Corresponding author:

Marek Barlak e-mail: [email protected] National Centre for Nuclear Research Świerk - NCBJ Plasma/Ion Beam Technology Division (FM2) 7 Andrzeja Sołtana St. 05-400 Otwock, Poland

ORCID ID: Barlak Marek 0000-0003-1416-7461 Wilkowski Jacek 0000-0001-5798-6761 Werner Zbigniew 0000-0003-1172-0268

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 79-83 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

The effect of nitrogen ion implantation on nano-scale hardness and elastic modulus of WC-Co indexable knives for wood materials machining

JACEK WILKOWSKI1, MAREK BARLAK2, ROMAN BÖTTGER3, ZBIGNIEW WERNER2 1Department of Mechanical Processing of Wood, Warsaw University of Life Sciences - SGGW 2 Plasma and Ion Technology Division (FM2), National Centre for Nuclear Research Świerk – NCBJ 3Institute of Ion Beam Physics and Materials Research, Helmholtz-Zentrum Dresden-Rossendorf – HZDR

Abstract: The effect of nitrogen ion implantation on nano-scale hardness and elastic modulus of WC-Co indexable knives for wood materials machining. The paper presents the results of a study investigating the effects of nitrogen ion implantation into the surface layer of WC-Co blades, used for machining wood materials, on their hardness and modulus of elasticity at the nano-scale. The modified blades were analyzed in six variants of implantation process parameters, and compared with control blades (virgin, unmodified). Three energies of accelerating nitrogen ions were used in the implantation process (5, 50 and 500 keV) and two doses of implanted ions (1e17 and 5e17 cm-2). The nano-hardness and elastic modulus of all blades were measured using an Anton Paar TriTec (UNHT) hardness tester.

Keywords: ion implantation, WC-Co indexable knives, nitrogen ion implantation, nano-scale mechanical behavior, hardness, elastic modulus, wood materials machining

INTRODUCTION The basic wear mechanisms of cutting blades during the machining of wood materials are: abrasive wear, strength, erosion and oxidation. With few exceptions, the intensity of abrasive mechanisms such as micro-cutting, grooving (plastic deformation) and cracking is inversely proportional to the hardness of the blade material. A blade with high hardness is also characterized by increased fragility, and therefore low plasticity in the conditions of the microcutting process [Ndlovu 2009]. Strength mechanism-related chipping and cracking result from exceeding the constant static load strength or fatigue due to cyclic loads during friction, causing a change in compressive and tensile loads during cutting [Kupczyk 2009]. The erosive mechanism during chipboard machining results from the impact of mineral particles (sand), which remove fragments of the blade material under the influence of momentum. Oxidation, being one of the mechanisms of chemical wear, when machining with WC- Co cemented carbide blades, proceeds intensely at temperatures above 970 K (≈700°C). Chemical synthesis with oxygen mainly affects the matrix through cobalt (Co), thus increasing the porosity of the blade material and extending the effect of oxidation to a greater depth. As a result of this process, brittle phases are formed, which are easily removed by sliding friction [Kupczyk 2009]. Therefore, in order to limit the intensity of the above mechanisms, it is necessary to ensure high hardness of the blade material (also at elevated temperatures) while maintaining its low brittleness and high strength (toughness) for impact loads, as well as a low coefficient of friction of the blade against the workpiece [Kupczyk 2009]. The relations between hardness and modulus of elasticity, and between friction and wear can be investigated on the microstructural scale [Ndlovu 2009, Staedler and Schiffmann 2001, Wilkowski et al. 2019].

The aim of the work was to determine the impact of nitrogen ion implantation, with selected process parameters, on the nano-scale hardness and modulus of elasticity of the surface layer of WC-Co indexable knives dedicated for wood materials machining.

MATERIAL AND METHODS The WC-Co indexable knives with dimensions of 29.5×12.0×1.5 mm3, produced by Ceratizit Co (Austria) were used for the tests (Fig. 1). Selected properties of the tested WC-Co composites are shown in Table 1.

Table 1. Properties of the tested WC-Co indexable knives [www.ceratizit.com] Binder Bending WC grain Density Hardness Material symbol content strength size [µm] [g/cm3] HV30 Co [%] [MPa] KCR08 0,5-0,8 3,2 15,20 1790 2300

The modification of the flank surface of WC-Co blades in the process of nitrogen ion implantation was carried out at Helmholtz-Zentrum Dresden Rossendorf (HZDR) according to the plan shown in Table 2. In this way, six variants of modification of indexable knives were obtained. They were compared with unmodified (virgin, control) knives.

Table 2. Main parameters of the nitrogen ion implantation process Energy Dose Temperature (keV) (cm-2) (°C) 5 1e17 <20°C 5 5e17 <20°C 50 1e17 <50°C 50 5e17 <50°C 500 1e17 <200°C 500 5e17 <200°C

Hardness and modulus are calculated from the load-displacement data for each indentation by the Oliver and Pharr method [Oliver and Pharr 2003, Kempf 2002, Wilkowski et al. 2009]. The local properties were measured using an Anton Paar TriTec Ultra Nano Hardness Tester (UNHT) (Fig. 2).

Figure 1. WC-Co indexable knives for wood Figure 2. Anton Paar TriTec Ultra Nano Hardness materials milling Tester (UNHT)

The Berkovich indenter was used. The measurement parameters were as follows: maximum load: 1 mN; loading / unloading speed: 2 mN/min; pause: 5 s.The measurement was repeated five times on each sample.

RESULTS AND DISCUSSION Fig. 3 shows the results of the nano-scale hardness (top) and modulus of elasticity (bottom) of the examined indexable knives. Bar charts with standard deviation indicate changes in hardness and modulus of elasticity as a result of the nitrogen ion implantation process. The highest values of hardness and modulus of elasticity were obtained for modifications with the following parameters: ion energy: 5 keV and ion dose: 5e17 cm-2. The values obtained were higher than those for virgin knives. With an increase in ion energy (50 and 500 keV), the properties of the WC-Co surface layer measured at the nano-scale decreased, irrespectively of the ion dose. The lowest hardness and modulus of elasticity were obtained for the following modification parameters: ion energy of 500 keV and ion dose of 5e17 cm-2 for the lowest hardness, and dose 1e17 cm-2 for the lowest modulus of elasticity.

Figure 3. Nano-scale hardness and elastic modulus of tested WC-Co indexable knives The values of the tested properties are lower when the implantation energy is increased to 50 and 500 keV, which happens to a greater extent for hardness than for modulus of elasticity.viable interpretation is that this is due to the fact that nitrogen implantation produces nitrides (e.g. WN, W2N, W3N), which are not harder than WC tungsten carbide, but exhibit higher impact strength and fracture toughness. This is explained by the Samsonov theory based on the configurational solid state model.

Samsonov's research showed that the hardness of nitrides is less than that of carbides of the same metals (including tungsten) and, at the same time, their plasticity is greater. This is due to differences between nitrides and carbides based on a smaller statistical weight of atoms in a stable s2p6 configuration [Samsonov and Kosolapova 1975]. Abrasion resistance generally increases with hardness. According to Samsonov, this relation is not confirmed in the case of carbides. This is attributed largely to the fragility of these compounds, manifested by their minimal plasticity in the conditions of the micro-cutting process [Wilkowski et al. 2018].

CONCLUSION Based on the results obtained, the following conclusions can be drawn: 1. The highest values of hardness and modulus of elasticity were obtained for modifications with the following parameters: ion energy of 5 keV and ion dose of 5e17 cm-2. The values obtained were higher than those for virgin knives. 2. Together with an increase in ion energy (50 and 500 keV), the properties of the WC- Co surface layer measured at the nano-scale decreased, irrespectively of the ion dose.

Acknowledgments. This scientific work was financed from funds for science in 2017-2018, granted for the implementation of an internationally co-financed project “Influence of nitrogen ions implantation on WC-Co composites used in wood based materials machining” (grant of the Polish Ministry for Science and Higher Education No W83/HZDR/2017). Parts of this research were supported by the Ion Beam Center at the Helmholtz-Zentrum Dresden- Rossendorf e. V., a member of the Helmholtz Association (Proposal 17001078-ST).

REFERENCES

1. KEMPF M., 2002: Nanohärtemessungen mit dem Rasterkraftmikroskop. Der Andere Verlag, Osnabrück, 160 pp. 2. KUPCZYK M.J., 2009: Wytwarzanie i eksploatacja narzędzi skrawających z powłokami przeciwzużyciowymi. Wydawnictwo Politechniki Poznańskiej, 445 pp. 3. NDLOVU S., 2009: The wear properties of tungsten carbide-cobalt hardmetals from the nanoscale up to the macroscopic scale. Doctoral thesis, 134 pp. 4. OLIVER W.C., PHARR G.M., 2003: Measurement of hardness and elastic modulus by instrumented indentation: Advances in understanding and refinements to methodology. J. Mater. Res. 19/1: 3-20 5. SAMSONOV G.V., KOSOLAPOVA T.J., 1975: Vysokotemperaturnye karbidy. Naukova Dumka 6. STAEDLER T., SCHIFFMANN K., 2001: Correlation of nanomechanical and nanotribological behavior of thin DLC coatings on different substrates. Surface Science 482-485: 1125-1129. 7. WILKOWSKI J., BARLAK M., WERNER Z., BÖTTGER R., KONARSKI P., PISAREK M., 2018: The effect of nitrogen ion implantation on the properties of WC- Co composites used in wood-based materials machining. Poster presentation at Yucomat 2018 Conference, Herceg Novi, Montenegro, 3-7 September, 2018 8. WILKOWSKI J., BARLAK M., WACHOWICZ J., BÖTTGER R., WERNER Z., 2019: Nano-scale hardness and elastic modulus of WC-Co composites and their relationship to the tools life during particleboard milling. Ann. WULS - SGGW, For. and Wood Technol. 106: 62-66

Streszczenie: Wpływ implantacji jonów azotu na twardość i moduł sprężystości w nano-skali noży wymiennych WC-Co do obróbki materiałów drzewnych. W artykule przedstawiono wyniki wpływu procesu implantacji jonów azotu do warstwy wierzchniej ostrzy WC-Co do obróbki materiałów drzewnych na ich twardość i moduł sprężystości w nano-skali. Analizowano ostrza modyfikowane w sześciu wariantach parametrów procesu implantacji, oraz w celach porównawczych, ostrza kontrolne (niemodyfikowane). Zastosowano trzy energie przyspieszenia jonów azotu w procesie implantacji (5, 50 i 500 keV) oraz dwie dawki implantowanych jonów (1e17 oraz 5e17 cm-2). Pomiaru twardości i modułu sprężystości w nano-skali wszystkich ostrzy dokonano przy użyciu twardościomierza Anton Paar TriTec (UNHT).

Corresponding author:

Jacek Wilkowski e-mail: [email protected] Warsaw University of Life Sciences – SGGW Instytut Nauk Drzewnych i Meblarstwa Faculty of Wood Technology 159 Nowoursynowska St. 02-787 Warsaw, Poland

ORCID ID: Wilkowski Jacek 0000-0001-5798-6761 Barlak Marek 0000-0003-1416-7461 Werner Zbigniew 0000-0003-1172-0268

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 107, 2019: 84-92 (Ann. WULS - SGGW, For. and Wood Technol. 107, 2019)

Furniture use safety at early design stage

ADAM MAJEWSKI Department of Furniture Design, Faculty of Wood Technology, Poznan University of Life Sciences

Abstract: Furniture use safety at early design stage. This paper presents a virtual method of determining the usage safety of a bed frame. The aim of the study was to investigate the influence of different external loads dictated by standards on the newly designed bed structure. Furthermore, it was decided to take into consideration unusual ways of acting load. Finally, the test of dynamic load was carried out. The analysis was executed by means of Finite Element Method in Computer Aided Engineering software. Results of stress distribution and displacements were compared to previously investigated mechanical properties of the used materials, such as elastic plywood and pine wood rigidity. Based on compared data, the safe use of bed frame was determined. The numerical calculations performed proved that there are possibilities of design changes to improve user safety and/or reduce production costs while ensuring sufficient mechanical properties of the furniture. Conclusions of the analysis suggested a possibility to sub-optimize the tested solution.

Keywords: FEM, bed, stress, displacement, dynamic

INTRODUCTION

Each product on the market should guarantee safety to customers in accordance with the General Product Safety Directive (2001/95/WE). A safe piece of furniture is designed and produced in line with the requirements and conditions described by European standards. Obviously, the safety of furniture depends primarily on such factors as shape of the piece, the used materials and their mechanical characteristics or the kinds of joints. On the one hand, during the design process there is a number of ideas connected with technology and materials. Engineers are obliged to immediately find out the option that provides the demanded stiffness and durability. On the other hand, making physical models and conducting physical tests takes lots of time and has a significant influence on production costs increase. That is why there is a need for virtual tools allowing designers to verify their construction solutions in a quick and efficient way at early stages of the design process. The Finite Element Method (FEM), developed more than 50 years ago, allows engineers to predict the behaviour of the designed structures during usage, with the help of numerical calculations. The essence of FEM is to discretize the analyzed object. It is a process of transferring continuous models and equations into discrete counterparts - finite elements. The use of software with FEM ensures a fast and convenient way of verifying the mechanical properties of new construction designs, although specialist knowledge is required. Over a dozen years after the appearance of FEM, numerous computer analysis were undertaken in terms of furniture durability. Subsequent promising results of chair construction test executed by Smardzewski (1990) with a simple application PANDA-1 paved a new and constantly evolving direction in the development of furniture design. In the following studies, a numerical model designed for determining the durability of skeleton furniture constructions was also developed by Smardzewski. Smardzewski and Prekrat (2002) used FEM software to evaluate stress distribution in disconnected furniture joints. Smardzewski and Papuga (2004) assessed stress distribution in angle joints of skeleton furniture subsequently. Smardzewski and Ożarska (2005) tried to investigate the rigidity of cabinet furniture with semi-rigid joints of the confirmat type. Smardzewski and Kłos (2011) presented alternative methods of numerical modelling of dowel joint rigidity of board elements, using for this purpose nodes of substitute linear elasticity modulus. Strength analysis methods included in the design phase

84 contributed effectively to optimizing the construction. Smardzewski and Gawroński (2001) presented FEM algorithms for chair construction optimization. Subsequently, they developed a method of gradient optimization for skeleton furniture with different connections. Smardzewski and Prekrat (2009) successfully performed the optimization of a sofa frame in the integrated CAD-CAE environment. While designing, every aspect of furniture construction has to be taken into consideration at an early stage of prototyping. In relation to attempts of suboptimization, reference was made to the criteria described by Ostwald (1987) and Smardzewski (2008). The idea of this paper is to pay attention to a method that allows to determine the mechanical properties of a structure globally, at an early stage of the design process and, based on the results, propose design changes in order to suboptimize the construction.

MATERIALS AND METHODS

Material characteristics The bed frame shown in figure 1 was made mainly of pine wood elements. Only the head and foot of the bed were made of glued and bent elastic plywood veneers made of Fuma (exotic wood species) - the head was made of three veneers 8 mm thick and the foot was made of four veneers 5 mm thick. In numerical analysis wood was considered as an orthotropic material. Elastic plywood was analyzed in regard to two major axes of orthotropy. Thus, in a 3D model of the bed, there were spatially oriented elements, and the values of their mechanical properties depend on the anatomical direction of the wood and wood-based material.

Figure 1. Tested bed structure

Hence, it was necessary to investigate the mechanical properties of the used materials. Samples of pine wood were prepared accordingly, in order to determine their flexural rigidity (Rg) in three point bending and their Modulus of Elasticity (MOE) in the longitudinal anatomical direction (longitudinal-tangential - LT and longitudinal-radial - LR surfaces). There were ten samples for each surface series. Flexural rigidity of wood and MOE tests were carried out in accordance with PN-77/D – 04103 and PN-63/D-04117 standards, respectively. The tests of flexural rigidity and MOE of plywood were conducted along with the PN-EN 310 standard, although it was impossible to determine the flexural rigidity of the samples in which the grain was oriented in parallel to the acting force. Examined specimens did not shutter under the acting load even when a small bending radius value was obtained. Ten samples were tested for each material - elastic plywood 5 mm and 8 mm thick with grains running in

85 parallel and in perpendicular to the acting force. Figure 2 presents the stiffness characteristics of pine wood samples during flexural tests.

Figure 2. Stiffness of the pine wood used in the experiment

Clearly, pine wood in LT surface is characterized by maximum stiffness value about 37% higher than LR surface. Comparable data gained from flexural experiment conducted on elastic plywood are presented in figure 3. The elastic plywood with 8 mm thickness is stiffer by about 42% than the 5 mm plywood.

Figure 3. Stiffness of the elastic plywood used in the experiment

The characteristics of pine wood and plywood presented above are not sufficient for a full description of orthotropic properties. They refer to elastic properties in the longitudinal direction (wood) and two major directions of orthotropy (flexible plywood). The heterogeneity of the used materials is characterized by the three main directions of orthotropy. In addition, it was necessary to specify the value of Poisson's ratios in the individual

86 anatomical planes. Based on the data gathered by Hearmon (1948) and mentioned by Smardzewski (2008), the missing properties of wood were adopted. Figure 4 shows the main axes of orthotropy of flexible plywood conventionally referred to wood anatomical directions - longitudinal (L), tangential (T) and radial (R).

Figure 4. Adopted main axes of orthotropy for elastic plywood

Noted the fact that elastic properties of plywood take extreme values in the main two mutually perpendicular directions, it was assumed that the most important Poisson's ratios will be determined in plane LT. It was also assumed that the omission of Poisson's ratios for the planes containing the direction of the thickness (R) will not have a significant impact on the quality of the results. To fulfill those assumptions, it was relevant to use the equation mentioned by Smardzewski (2008) and Gerrard (1987). Adjusted to the requirements of this paper, it presents as follows:

(1) Where: - Poisson’s ratio for LT surface, MOELT - Modulus of Elasticity for longitudinal direction, - Poisson’s ratio for TL surface, MOETL - Modulus of Elasticity for tangential direction.

The shear modulus was calculated in accordance with the equation presented by Smardzewski:

Gyx = (MPa) (2) Where: Gyx – Shear modulus in chosen surface (MPa), MOEy – Modulus of Elasticity in chosen direction (MPa), νyx – Poisson’s ratio for chosen surface.

Table 1. Selected properties of pine wood used in the tests Modulus of Density Rigidity, Rg Modulus of Elasticity, MOE Shear modulus, G Poisson's ratio ν [-] Rupture, [kg/m3] [MPa] [MPa] [MPa] MOR [MPa]

dynamic LT LR LT LR R T LT LR RT LR LT RT TR RL TL test Pine 550 97 82 12809 12758 1118 584 693 1181 70 0.4 0.5 0.7 0.3 0.04 0.02 89 wood

Moreover, for the dynamic test, the Modulus of Rupture (MOR) was also determined. The equation by use of which the MOR is determined, is important for the elastic limit range.

Based on the relationship between the MOR ratio and literature (Bergman et. al. 2010), it was assumed that the ratio of MOR to MOE of pine is about 0.007. The full characteristics of materials gathered both from experiments and literature are shown in table 1 and table 2.

Table 2. Selected properties of elastic plywood used in the tests Density Modulus of Elasticity, MOE Shear modulus, G Rigidity, Rg [MPa] Poisson's ratio [-] [kg/m3] [MPa] [MPa] grains grains grains grains LT LR RT LR LT RT TR RL TL parallel perpendicular parallel perpendicular elastic plywood 5 mm 350 - 32 124 4237 1471 2119 62 - 0.4 - - - 0.01 elastic plywood 8 mm 350 - 31 44 4511 1566 2256 23 - 0.4 - - - 0.01

Load diagrams and support Based on the PN-EN 1725 standard and heuristic thoughts exhibiting customer's use, static diagrams were adopted as shown in figure 5.

Figure 5. Static load diagrams Loads dictated by standards were applied as force P1, P2, P3, and P4. Custom ways of usage were reflected by forces P5 (left upper corner of top is being pushed) and P6 (two persons are resting simultaneously). The dynamic experiment was simulated by dropping a steel element weighing 4.2 kg on the side of bed directly (Fig. 6).

Figure 6. Dynamic load diagram The bed frame was supported in a way to prevent its movement under the influence of the forces. In the dynamic test, the divided side of bed was supported so as to simulate the connection with the rest of construction.

Numerical models Numerical models were prepared directly from the imported 3D bed model created in CAD software (Inventor Professional). The discretization process proceeded automatically in the CAE application (Nastran In-CAD). More attention was paid to proper attribution of material characteristics to model elements in relation to orthotropic and orthotropic axes. To do it correctly, the groups of different spatially oriented parts were separated as presented in figure 7.

Figure 7. Major axes of orthotropy in relation to global axes of CAE software

Table 3. Orthotropy axes settings in CAE software CAE axes settings No. Global X-direction Global Y-direction Global Z-direction 1 X Y Z 2 Y Z X 3 Z X Y

The main axes of orthotropy were attributed to the global axes (1, 2, 3) set in the CAE program (table 3). Then, the material characteristics were implemented and the bed structure was loaded and supported. For the dynamic durability test, the contact between surfaces of the steel element and the top of bed side was defined. The gravity force was applied to the steel element. After adding support, the model was completely prepared for the dynamic test.

RESULTS AND DISCUSSION The stiffness and durability of the bed frame were assessed on the basis of values of displacement and stress distribution in the loaded elements. The results of computer calculations were analyzed precisely and the elements with maximum stress and displacement values were indicated. In the next stage, the data obtained by numerical computation were compared to the results from previously conducted physical experiments. It led to determine whether the stress occurring in the structure results in damage or destruction of intensive elements. Figure 8 shows an example of the displacements occurred under the influence of an external load.

Figure 8. Displacement of bed frame under applied force This variant presents the maximum deflection value which is remarkably higher than the effects of other forces acting. Apart from this case, based on the results (table 4) it was concluded that the bed frame remains stiff enough under the influence of other acting forces.

Table 4. Results of numerical analysis Maximum Maximum Max. stress in Forced element Test displacemen stress percentage of material material t [mm] [MPa] rigidity [%] Load diagram 1 1.03 12.29 Pine wood [LR] 14.99 Load diagram 2 0.77 7.16 Pine wood [LR] 8.73 Load diagram 3 0.66 11.32 Elastic plywood* 35.38 Load diagram 4 0.71 13.05 Elastic plywood* 40.78 Load diagram 5 24.33 15.41 Elastic plywood** 49.71 Load diagram 6 3.50 10.93 Pine wood [LT] 11.27 Dynamic test 16.16 43.70 Pine wood [LR[ 53.29 * - element made of glued 5 mm elastic plywood sheets ** - element made of glued 8 mm elastic plywood sheets

The presented displacement values ranged from nearly 0.66 mm to 3.50 mm. They prove the satisfactory mechanical properties of the tested bed frame and contribute to attempt its suboptimization. For the dynamic test, it was determined that the steel element caused a 16.16 mm displacement while hitting the top surface of the bed side element. The biggest value of displacement in the fifth load diagram indicated also the biggest stress result. The obtained value of 15.41 MPa corresponds to nearly 50% of the maximum rigidity of the tested 8 mm thick flexible plywood. Other stress effects in directly loaded elements did not exceed 41% of maximum rigidity of the materials that they were made of. The calculated percentage relationships indicate that structural elements are strong and successfully carry external loads. Despite the large deflections, the top front does not show a tendency for damage or destruction under the influence of the load. This is due to the good elastic characteristics of the material from which it was made. In dynamic tests, the calculated stress value (43.70 MPa) corresponds to 53% of the determined Modulus of Rupture. It appears that this element is strong enough. The results indicate that the impact does not cause serious deformities or damage. Moreover, the calculated stress reached values that were nearly three times higher than the results of the static tests.

CONCLUSION In general, relatively low values of displacement and stress distribution point out that the bed structure is sufficiently rigid and stiff. The analysis of the bed frame showed that the attempts of suboptimization can be freely taken. The performed numerical calculations proved that there are possibilities of design changes to improve user safety and/or reduce production costs while ensuring sufficient mechanical properties of the furniture. Suggested solutions include:  Reducing the number of elastic plywood sheets assembling the bed foot.  Changing the number of supporting ribs in order to ensure better support in 1/3 of the bed head length.  Modifying the technology of supporting ribs - these elements (made of wood) could be cut as a single MDF elements.  Changing the material and reducing the thickness of elements in the bed sides to 18 mm. By use of the finite element method, there is a possibility to perform multiple bed tests considering changes of the design. This can be done without any additional financial and material losses. The time devoted to this activity results in the optimal variant at the virtual modelling stage, reducing the number of prototypes produced.

REFERENCES

1. BERGMAN R. ET AL., 2010: Wood Handbook, Wood as an Engineering Material. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory. 2. GERRARD CH., 1987: The equivalent orthotropic properties of plywood. Wood Sci Technol 21: 335 – 348. 3. HEARMON R.F.S, 1948: Elasticity of Wood and plywood. Forest Products Research, Department of Scientific and Industrial Research. Special raport no.7 London. 4. OSTWALD M., 1987: Optymalizacja konstrukcji. Wydawnictwo Politechniki Poznańskiej. (in Polish) 5. SMARDZEWSKI J., 1990: Numeryczna analiza konstrukcji mebli metodą elementów skończonych. Przemysł Drzewny 41 (7): 1-5. (in Polish)

6. SMARDZEWSKI J., GAWROŃSKI T., 2001: FEM algorithm for chair optimization. EJPAU. Vol. 4. Issue 2. 7. SMARDZEWSKI J., PREKRAD S., 2002: Stress distribution in disconnected furniture joints. EJPAU. Vol. 5. Issue 2. 8. SMARDZEWSKI J., GAWROŃSKI T., 2003: Gradient optimisation of skeleton furniture with different connections. EJPAU. Vol. 6. Issue 1. 9. SMARDZEWSKI J., PAPUGA T., 2004: Stress distribution in angle joints of skeleton furniture. EJPAU. Vol. 7. Issue 1. 10. SMARDZEWSKI J., OŻARSKA B., 2005: Rigidity of cabinet furniture with semi rigid joints of the confirmat type. EJPAU. Vol.8. Issue 2. 11. SMARDZEWSKI J., 2008: Projektowanie mebli. Wydawnictwo PWRiL Poznań. (in Polish) 12. SMARDZEWSKI J., PREKRAT S. 2009: Optimisation of a sofa frame in the integrated CAD-CAE environment. EJPAU. Vol. 12. Issue 4. 13. SMARDZEWSKI J., KŁOS R., 2011: Modeling of joint substitutive rigidity of board elements. Annals of Warsaw University of Life Sciences – SGGW. Forestry and Wood Technology No 73, 2011: 7-15.

Streszczenie: Bezpieczeństwo użytkowania mebla na wczesnym etapie projektowania. Zgodnie z Dyrektywą 2001/95/WE Parlamentu Europejskiego i Rady z dnia 3 grudnia 2001 r. w sprawie ogólnego bezpieczeństwa produktów wymaga się, aby sposób użycia wyrobu ani on sam nie stwarzał zagrożenia dla zdrowia i życia użytkowników. Dobór materiałów na elementy mebla i sposób ich połączenia stanowią, zatem kluczowe działania w kontekście inżynierskiego projektowania mebli. Jednym z zadań wieloetapowego procesu projektowania jest kosztowne wykonywanie prototypów z materiałów docelowych. Dzięki oprogramowaniu wspierającemu projektowanie inżynierskie - CAE, można dokonać wstępnej oceny wytrzymałości konstrukcji przez numeryczne obliczenia metodą elementów skończonych. W niniejszej pracy przedstawiono wirtualne symulacje sposobu obciążenia nowej konstrukcji łóżka, podjęte już na etapie przygotowania bryły 3D. Elementy modelowano, jako wykonane z drewna sosny zwyczajnej oraz sklejki elastycznej FUMA o grubościach 5 mm i 8 mm. Wartości sił oraz punkty ich przyłożenia dobrano na podstawie odpowiednich norm. Obciążenie przykładano także do potencjalnie newralgicznych węzłów konstrukcji. Przeprowadzono ponadto symulację dynamicznego obciążenia boku łóżka. Na podstawie uzyskanych wyników ustalono, że zaproponowana konstrukcja jest bezpieczna, bowiem naprężenia w obciążanych statycznie elementach nie przekraczają 50% wytrzymałości użytego materiału. Szczyt łóżka, pomimo relatywnie dużego ugięcia, nie wykazywał tendencji do poważnych deformacji i uszkodzenia. Ostatnią część pracy poświęcono suboptymalizacji analizowanej konstrukcji łóżka.

Corresponding author:

Adam Majewski Wojska Polskiego 38/42, 60-637, Poznan, Poland email: [email protected] phone: 61 848 74 75

ORCID ID: Majewski Adam 0000-0003-1645-9871

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 93-96 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

A human quality control system in furniture manufacturing – a pilot study

KATARZYNA ŚMIETAŃSKA, PIOTR PODZIEWSKI Department of Mechanical Processing of Wood, Faculty of Wood Technology, WarsawUniversity of Life Sciences– SGGW

Abstract: A human quality control system in furniture manufacturing – a pilot study. The article presents the results of pilot studies in which an attempt was made to check what the subjectivity of visual inspection looks like during the control of MDF delamination process in the milling process. 100% match of assessors’ scores was obtained only with low (VB = 0–0.08mm) and high (VB = 0.3–0.33mm) tool wear. In addition, the impact of gender on the results was significant. Women rated items less restrictively. Based on the results obtained, one can expect a problem with subjectivity when using a single-person human quality control system.

Keywords: human visual quality control, melamine-faced MDF, milling

INTRODUCTION Quality control of production plants is an extremely important issue at every stage of product manufacturing. As a basic element of the organization’s management system (next to planning, organizing and leading), it aims to check the compliance of the process or product with the customer’s requirements, by eliminating or minimizing defects and incompatibility (Hamrol et al. 2011, Kang et al. 2018) Despite the development of measurement methods and the widespread automation along with the technological progress, the dominant method of quality control in many companies is the use of human senses only (organoleptic control). A special example of such control is visual control (visual inspection), which includes: simplicity of carrying out in comparison with other types of control, low costs, speed, low number of samples. This method does not require specialized measuring equipment, the role of which is played by human vision, and is a non- destructive method (Kujawińska and Vogt 2015, Kang et al. 2018). However, visual control does not guarantee an unequivocal, fully correct assessment. The complexity of the problem of credibility of a human visual assessment results from the fact that its effectiveness is influenced by many factors (Kujawińska and Vogt 2015): - technical (type of defects, defect visibility, quality level, standards (tests), control automation), - psychophysical (age, sex, observation skills, experience, temperament, creativity), - organizational (training, scope of decision making, feedback, precise instructions), - workplace environment (light, noise, temperature, work time, workstation organization), - social (team communication, pressure, isolation). It is, therefore, necessary to regularly check the effectiveness of human visual quality control in manufacturing plants. Appearance of conformity assessment errors, in the form of incorrect product classification, may contribute to deterioration of process efficiency. Due to the high degree of subjectivity of visual inspection, human quality control system should set themselves the task of not only focusing on finding and eliminating the root causes of errors, but also monitor the effectiveness of the work of raters. This will allow creating a reliable source of feedback on the production process, which will improve production and reduce manufacturing costs. The article attempts to determine what the subjectivity of visual control looks like when controlling the MDF delamination process in the milling process.

MATERIALS AND METHODS The assessment procedure required special preparation of samples for testing and ensuring appropriate conditions. The experiment uses elements made of a MDF laminated board (one of the basic materials in the furniture industry). A step was milled in each of them, resulting in a k1 edge for evaluation of machining quality. The dimensions and the shape of the object are shown in Fig. 1. In the experiment, one brand new tool was used, which was gradually blunted to obtain the desired state of wear of the VB blades (0mm, 0.08mm, 0.09mm, 0.11mm, 0.20mm, 0.22mm, 0.30mm, 0.31mmand 0.33mm). Dulling of the tool took place in a way that imitated factory conditions, i.e. on various wood-based materials. Five items were made for each tool condition.

Fig.1. Dimensions and shape of samples (in mm) with a marked edge k1.

Six assessors(3 women – J1, J2, J3; 3 men – J4, J5, J6) took part in the study, furniture engineers, after short-term internships in furniture factories. The results obtained in the study are 45 assessments for each assessor. They were always held in the same room, during one uninterrupted session, with similar lighting and sample presentation (all elements were displayed simultaneously). During the tests, the assessor (under time pressure, which allowed to simulate working conditions in industrial conditions) made a single assessment of the k1 edge of the subject: 1 – high quality, 0.5 – conditionally acceptable quality, 0 – clearly bad quality. The assessor should not try to pretend to be a professional (e.g. factory) quality controller but be guided by his own “intuition” (as if he were an ordinary customer who watches the goods in the store). Possible damages on the first and last 5mmof the edges may be due to non-processing reasons and as such should not affect the quality assessment of the entire edge, so they should be ignored by the assessor.

RESULTS AND DISCUSSION The obtained results of assessments by individual assessors are presented in Table 1. On its basis, the agreement table whose rows are the subjects from the rating table where any row with fewer than two ratings are dropped was determined. Analyzing the received data (Tab.1, Tab.2.), ambiguity of assessors judgments was found in 31% of cases, withinwhich 11% differentiation occurred in a perfectly symmetrical way (three assessors gave the same higher grade, the other lower grade). 100% compliance was observed for VB in the ranges of 0–0.08mm and 0.30–0.33mm. In addition, the impact of gender turned out to be unambiguously significant – women rated subjects less restrictively (giving a higher grade). Also within the same sex, more or more severe assessors were noticed (J6 turned out to be the most radical). The above conclusions suggest that a human quality control system based on one person would not be an objective tool for assessing manufactured products. Despite the fact that there was no extreme case of divergence of grades (0 and 1 for the same subject),

94 a definite problem with the subjectivity of grades during the control of the MDF delamination process in the milling process should be stated.

Table 1. Rating table Tab. 2. Agreement table VB J1 J2 J3 J4 J5 J6 VB 0 0.5 1 1 1 1 1 1 1 0 0 6 1 1 1 1 1 1 0 0 6 0 mm 1 1 1 1 1 1 0 mm 0 0 6 1 1 1 1 1 1 0 0 6 1 1 1 1 1 1 0 0 6 1 1 1 1 1 1 0 0 6 1 1 1 1 1 1 0 0 6 0.08 mm 1 1 1 1 1 1 0.08 mm 0 0 6 1 1 1 1 1 1 0 0 6 1 1 1 1 1 1 0 0 6 1 1 1 1 1 1 0 0 6 1 1 1 1 1 1 0 0 6 0.09 mm 1 1 1 1 1 1 0.09 mm 0 0 6 1 1 1 1 1 1 0 0 6 0.5 0.5 0.5 0.5 1 0.5 0 4 2 1 0.5 1 0.5 1 0.5 0 2 4 1 0.5 1 0.5 1 0.5 0 2 4 0.11 mm 1 0.5 1 0.5 1 0.5 0.11 mm 0 3 3 0.5 0.5 0.5 0.5 0.5 0.5 0 6 0 0.5 0.5 0.5 0.5 0.5 0.5 0 6 0 0.5 0 0.5 0 0.5 0 3 3 0 0.5 0.5 0 0 0.5 0 4 2 0 0.20 mm 0.5 0.5 0.5 0.5 0.5 0 0.20 mm 2 4 0 0.5 0.5 0.5 0.5 0.5 0 2 4 0 0.5 0.5 0.5 0.5 0.5 0 2 4 0 0.5 0.5 0.5 0.5 0.5 0 2 4 0 0.5 0.5 0.5 0 0.5 0 3 3 0 0.22 mm 0.5 0 0 0 0.5 0 0.22 mm 4 2 0 0.5 0.5 0.5 0.5 0.5 0 2 4 0 0 0 0 0 0.5 0 5 1 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 0 6 0 0 0.30 mm 0 0 0 0 0 0 0.30 mm 6 0 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 0 6 0 0 0.31 mm 0 0 0 0 0 0 0.31 mm 6 0 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 0 6 0 0 0.33 mm 0 0 0 0 0 0 0.33 mm 6 0 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 0 6 0 0

There are many questions that still need to be asked regarding this topic, such as: how would a person evaluate the same samples on a different day or in different working conditions? In this case, one would also expect a lack of repeatability. This problem requires further investigation. The presented article presents only pilot studies.

CONCLUSIONS 1. 100% match of assessors scores was obtained only with low (VB = 0-0.08mm) and high (VB = 0.3-0.33mm) tool wear. 2. The influence of gender on the results turned out to be significant. Women rated items less restrictively. 3. Based on the results obtained, one can expect a problem with subjectivity when using a single-person human quality control system.

REFERENCES

1. HAMROL, A., KOWALIK, D., KUJAWIŃSKA, A. 2011: “Impact of selected work condition factors on quality of manual assembly”, Human Factors and Ergonomics in Manufacturing 21(2), pp. 156–162. 2. KANG C.W., RAMZAN B.M., SARKAR B., IMRAN M., 2018: “Effect of inspection performance in smart manufacturing system based on human quality control system”,The International Journal of Advanced Manufacturing Technology, February 2018, Volume 94, Issue 9–12, pp. 4351–4364, https://doi.org/10.1007/s00170-017- 1069-4 3. KUJAWIŃSKA, K., VOGT, K., 2015, “Human factors in visual quality control”, Management and Production Engineering Review6(2), pp. 25–31.

Streszczenie: Kontrola jakości przez człowieka w produkcji mebli - badanie pilotażowe. W artykule zaprezentowano wyniki badań pilotażowych, w których podjęto próbę sprawdzenia jak wygląda problem podmiotowość kontroli wzrokowej podczas kontroli procesu delaminacji MDF w procesie frezowania. Stuprocentową zgodność ocen sędziów otrzymano jedynie przy niskim (VB=0–0,08mm) oraz wysokim (VB=0,3–0,33 mm) zużyciu narzędzia. Ponadto wpływ płci na wyniki okazał się istotny. Kobiety oceniały przedmioty mniej restrykcyjnie. Na podstawie otrzymanych wyników można spodziewać się problemu z subiektywnością w przypadku zastosowania jednoosobowego systemu kontroli jakości przez człowieka.

Corresponding authors:

Katarzyna Śmietańska, Piotr Podziewski, Faculty of Wood Technology – SGGW, Department of Mechanical Processing of Wood, ul. Nowoursynowska 159, 02-776 Warsaw, Poland e-mail: [email protected] e-mail: [email protected]

ORCID ID Śmietańska Katarzyna 0000-0001-8705-3700 Podziewski Piotr 0000-0002-2628-5062

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 97-103 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

The use of the Krippendorff’s coefficient in determining intra-rater reliability in human visual quality control of furniture manufacturing processes KATARZYNA ŚMIETAŃSKA, PIOTR PODZIEWSKI Department of Mechanical Processing of Wood, Faculty of Wood Technology, Warsaw University of Life Sciences – SGGW

Abstract: The use of the Krippendorff’s Coefficient in determining intra-rater reliability in human visual quality control of furniture manufacturing processes. The article presents the results of research on the effectiveness of human visual quality control of furniture manufacturing processes. The aim of this experiment was to check the inter-rater reliability of three elements: the judge, the instruction of which the quality of individual objects made of laminated MDF was determined and the quality assessment by a human. The Krippendorff’s Coefficient was used as a measure of visual control effectiveness. The results of the experiment showed that the use of human visual quality control of furniture manufacturing processes can be a good solution. The values of all three coefficients K1, K2 and K3 gave alpha> 0.8, which is considered a guarantee of high compliance. Therefore, the tested judge turned out to be reliable and competent, the instructions were clear and sufficient, and the figures obtained were reliable. It turned out that the Krippendorff’s Coefficient values differ depending on the number of categories (for two categories K2 = 0.904, K3 = 0.902; for three categories K1 = 0.849). The way the categories were created did not affect the stability of the assessment. The alpha coefficient can undoubtedly be considered a good, convenient measure of monitoring the reliability of the measurement system, because by giving a specific numerical rating, it allows one to control the effectiveness of implemented improvement actions.

Keywords: Krippendorff’s Coefficient, intra-rater reliability, human visual quality control, melamine-faced MDF, milling

INTRODUCTION The aim of quality control of production plants is to check the compliance of the process or product with the requirements of the internal or external customer (Hamrol 2007) by eliminating or minimizing defects and incompatibility in the product. Most often quality control is performed by direct measurement or observation, and its results are data enabling interpretations of the assessed process or product state. In many industries (also in the furniture industry), there is a need for control that will accurately describe the quality of manufactured products, provide information about the process capability and indicate any areas for improvement. Despite the development of measuring methods based on increasingly objective measuring instruments and the widespread automation of technology (most operations in the manufacturing process are carried out by machines), man is still indispensable, and in many companies the dominant quality control method is organoleptic. Machines collect information in the form of measuring specific product features, but making decisions regarding control activities (dispositions) is a problematic stage. The main reason is the continuous increase in the number and diversity of customer needs, which often requires an unconventional approach to product quality control in the production process or acceptance control. In such cases, the automatic system is not able to ensure an adequate level of repeatability and reproducibility of the inspection and visual inspection carried out by man is the only and most appropriate solution (Hamrol 2007). A man, thanks to their knowledge and experience, is able to respond in a flexible manner, adequate to the situation. In contrast, a machine does not have the ability to learn from experience and improve its work, as well as to improvise or analyse alternative solutions.

Visual inspection is particularly important for processes whose repeatability and reproducibility is limited and the process results differ and require an individual approach when assessing the quality of their performance. The presence of a man in control operations of technological processes is indispensable, especially due to the increasing complexity of manufactured products. Human participation in visual control is valued for its ability to recognize new cases, flexibility in non-standard situations. Making decisions regarding the quality of controlled products requires not only a specific industry knowledge package but often an individual approach to each controlled piece and high sensitivity to incompatibility and limited confidence in the manufactured product. It should be emphasized that visual quality control carried out by man is characterized by a much higher level of subjectivity, which becomes extremely important in the case of objects and devices of everyday use, such as furniture, interior design elements, cars and household appliances. Only a human is able to subjectively decide which of the elements in the further process of use will satisfy him. The advantages of visual control (visual inspection) due to which it is often used in production include: simplicity of carrying out in comparison with other types of control, low costs, speed, and a low number of samples. This method does not require specialized measuring equipment, the role of which is played by human vision, and is a non-destructive method. Therefore, manufacturers of products from various industries (automotive (Vogt et al. 2015), electronic (Vogt et al. 2010)) commonly use traditional visual quality control methods, supported only by auxiliary measurements using measuring devices, despite the awareness of the imperfections of the organoleptic method and the significant risk of not detecting an incompatibility or its inappropriate assessment by an employee (controller) (Giesko et al. 2011). However, you should be aware of the fact that visual control does not guarantee a clear, correct assessment, the primary reason being the limited human reliability. The complexity of the problem of the credibility of a human visual assessment is due to the fact that its effectiveness is influenced by many factors, both organizational and directly related to the human being (Drury et al. 1986; Hamrol et al. 2011). Those factors can be divided into 5 categories (Vogt et al. 2015): - technical (type of defects, defect visibility, quality level, standards (tests), control automation, other), - psychophysical (age, sex, observation skills, experience, temperament, creativity, other), - organizational (training, scope of decision making, feedback, precise instructions, other), - workplace environment (light, noise, temperature, work time, workstation organization, other), and - social (team communication, pressure, isolation, other). There are three sub-types of reliability in literature (Weber 1990; Krippendorff 1980): 1. stability (intra-rater reliability) – consists in re-encoding the same data by the same people; largely refers to the coder’s skills, 2. accuracy – measures the compliance of the coding of the material with respect to the standard, established by a group of experts or based on previous research, 3. reproducibility (inter-reliability) – involves checking the degree of consistency in coding the same material by several people; the measurement is based on an estimate of the proportion of consistent categorizations between judges to all their decisions; is defined as the degree of agreement between judges or reliability between judges. During visual control in industrial practice, the intra-rater reliability becomes the biggest, basic limitation (problem), which is the result of the consistency of several assessments of a given controller. As one knows, products may not be the only the subject of non-compliance in production; it can also be the results of the work of the evaluating

98 controller. They may result from limited sensitivity to errors, limited perception as well as experience or skills. The causes can be divided into: direct (ignorance or limited opportunities to perform work properly) and indirect (non-ergonomic workstation or incorrect operation of machines). It is, therefore, necessary to regularly check the effectiveness of human visual quality control in manufacturing plants. The appearance of conformity assessment errors in the form of incorrect product classification may contribute to deterioration of process efficiency. The basic, most frequently used tools for measuring reliability are (Lombard et al. 2004, Krejtz et al. 2005, Krippendorf 2004, Neuendorf 2002, Scott 1955, Rosenfield 1986, Fleiss 1971, Cohen 1960, Hayes 2007, Holsti 1969): - joint-probability of agreement (percent agreement), - Holsti’s method (Holsti’s CR), -Scott’s (1955)  (pi), - Cohen’s  (Kappa), - Krippendorff’s , - Kendall’s W. Therefore, quality systems in industry should set themselves the task of not only focusing on finding and eliminating the root causes of errors but also monitoring the effectiveness of the work of raters. This will allow creating a reliable source of feedback on the production process, which will improve production and reduce manufacturing costs. These tools allow you to assess the capabilities of a measuring system consisting of quality controllers. They are a good solution for monitoring the reliability of the measurement system because by giving a specific numerical rating, it allows you to assess the effectiveness of implemented improvement actions.

MATERIALS AND METHODS The research used Krippendorff’s Coefficient – the coefficient of agreement between individual assessments as the most appropriate statistic for assessing intra-rater reliability in human visual quality control. It allows to determine the degree of credibility (compliance) of the results obtained between multiple repetitions of the same set of objects assessment by one observer, giving the repeatability and credibility of the visual assessment of one judge. Krippendorff’s alpha allows uniform reliability standards to be applied to a great diversity of data (Krippendorff 2004, Krippendorff 2011): - it can be used for any number of values per variable - can be used for any number of observers - can be used for small and large samples small and large sample sizes - to several metrics (scales of measurements) – nominal, ordinal, interval, ratio, and more - applies to data sets with missing values to data with missing values The coefficient is defined in the simplest form [1]:

α = 1-Do/De [1] Whrere: Do – a measure of the observed disagreement De – a measure of the disagreement that can be expected when chance prevails When agreement is observed to be perfect and disagreement is, therefore, absent, Do = 0 and α = 1, indicating perfect reliability. When agreement and disagreement are a matter of chance and observed and expected disagreements are equal, De = Do and α = 0, indicating the absence of reliability.

The assessment procedure required special preparation of test samples and appropriate conditions. Samples were elements made of a MDF laminated board (one of the basic materials in the furniture industry). In each element, a groove and a step were milled, forming three edges for the evaluation of machining quality (k1, k2, k3). The dimensions and shape of the object are shown in Fig. 1. All assessments always take place in the same room, during one uninterrupted session, with similar lighting and sample presentation (all elements were displayed simultaneously). During the tests, the judge (under time pressure, which allowed to simulate working conditions in industrial conditions), evaluated the subject edges four times (each separately) on a scale of 1–3. The following is the content of the instructions: 1. The judge assesses not the quality of the whole item but the quality of each of the 3 indicated edges separately (each separately). The edges are marked as k1, k2, k3 (as shown in the drawing below).

a b

Figure 1. The shape of the element (a) and the location of the individual edges to be assessed (b).

2. The judge should not pay attention to the condition of the corners (short zones at the beginning and end of each edge). Possible damages on the first and last 5 mm edges may have non-processing reasons and as such should not affect the quality assessment of the entire edge. The length of ignored zones (5 mm) is given with a significant margin and can be treated as an estimate (you do not need to use a ruler, so-called “eye- balling” is sufficient). 3. The judge should not attempt to pretend to be a professional (e.g. factory) quality controller but be guided by his own “intuition” (as if he were an ordinary customer who watches the goods in a store). 4. The judge has three ratings to choose from:  rating 1 = very good, good or at least satisfactory quality without any major “aesthetic discomfort”  rating 2 = quality creating some moderate “aesthetic discomfort” but conditionally acceptable, e.g. if it was less visible edges of the furniture or in the case of a very attractive product price  rating 3 = clearly unambiguous quality, definitely unacceptable from an “aesthetic point of view”.

100

RESULTS AND DISCUSSION The obtained numerical data (assessments of the “JD” judge) were recorded in a tabular form. Based on the results of the conducted tests, three values of the measurable Krippendorff’s numerical coefficient  were determined as follows (without distinguishing between individual edges k1, k2, k3):  K1 – the value of Krippendorff’s Coefficient was calculated for a three-grade evaluation – the quality of the elements was assessed on the basis of three categories (three grades 1, 2, 3, in accordance with the instructions)  K2 – Krippendorff’s Coefficient value was calculated for a two-stage evaluation – the results of two better categories (grades 1 and 2 in accordance with the instructions) were combined into one  K3 – Krippendorff’s Coefficient value was calculated for a two-stage evaluation – the results of two worse categories (grades 2 and 3 according to the instructions) were combined into one The obtained results are presented on the graph (Fig. 2).

Figure 2. Krippendorff’s Coefficient values for different assessment variants

CONCLUSIONS 1. The use of human visual quality control of furniture manufacturing processes proved to be a good solution. The values of all three alpha coefficients (for K1, K2 and K3) exceeded 0.8, which is considered a guarantee of high compliance. Therefore, the JD judge turned out to be reliable and competent, and the figures obtained should be considered reliable. 2. The number of categories played a significant role. For two categories, Krippendorff’s Coefficient takes higher values (K2 = 0.904, K3 = 0.902) than for three categories (K1 = 0.849). The way categories are created does not significantly affect the stability of the rating. 3. The alpha coefficient can undoubtedly be considered a good, convenient measure of monitoring the reliability of the measurement system because, by giving a specific numerical rating, it allows to control the effectiveness of implemented improvement actions.

101

REFERENCES

1. COHEN, J. (1960). “A coefficient of agreement for nominal scales”, Educational and Psychological Measurement 20, pp. 37–46. 2. DRURY, C.G., KARWAN, M., VANDERVARKER, D.R. (1986). “The Two- Inspector Problem”, IIE Transactions 18(2), pp. 174–181. 3. FLEISS, J. L. (1971). “Measuring nominal scale agreement among many raters”, Psychological Bulletin 76(5), pp. 378–382. 4. GIESKO, T., MAZURKIEWICZ, A., ZBROWSKI, A. (2011). “Optomechatroniczny system do automatycznej kontroli jakości wyrobów w przemyśle”, Problemy eksploatacji 4, pp. 103–114 (http://yadda.icm.edu.pl/yadda/element/bwmeta1.element.baztech-article-BAR0- 00600085 pdf]) 5. HAMROL, A. (2007). Zarządzanie jakością z przykładami. PWN, Warszawa 6. HAMROL, A., KOWALIK, D., KUJAWIŃSKA, A. (2011). “Impact of Selected Work Condition Factors on Quality of Manual Assembly”, Human Factors and Ergonomics In Manufacturing 21(2), pp. 156–162. 7. HAYES, A.F., KRIPPENDORFF, K. (2007). “Answering the Call for a Standard Reliability Measure for Coding Data”, Communication Methods and Measures 1(1), pp. 77–89. 8. HOLSTI, O. R. (1969). Content analysis for the social sciences and humanities. Reading, MA: AddisonWesley 9. KREJTZ, K., KREJTZ, I. (2005). “Rzetelność w analizie treści”, pp. 217–230 in: Stemplewska-Żakowicz, K., Krejtz, K. (red.) Wywiad psychologiczny. Wywiad jako postępowanie badawcze (2005), Pracownia testów psychologicznych polskiego towarzystwa psychologicznego, Warszawa 10. KREJTZ, K., KREJTZ, I. (2005). “Wybrane statystyki zgodności miedzy sędziami w analizie treści”, pp. 231–249 In: Stemplewska-Żakowicz K., Krejtz K. (red.) Wywiad psychologiczny. Wywiad jako postępowanie badawcze (2005), Pracownia testów psychologicznych polskiego towarzystwa psychologicznego, Warszawa 11. KRIPPENDORﬀ, K. (2004). “Reliability in Content Analysis: Some Common Misconceptions and Recommendations”, Human Communication Research 30(3), pp. 411–433. 12. KRIPPENDORFF, K. (2004). Content analysis: An introduction to its methodology. Second Edition. Thousand Oaks, London, New Delhi: SAGE Publications 13. KRIPPENDORFF, K. (2011). “Computing Krippendorff’s Alpha-Reliability”, fromhttp://repository.upenn.edu/asc_papers/43 14. KRIPPENDORFF, K. (1980). “Validity in Content Analysis”, Computerstrategien fur die Kommunikationsanalyse 3, pp. 69–112 15. LOMBARD, M., SNYDER-DUCH, J., & BRACKEN, C.C. (2002). “Content analysis in mass communication research: An assessment and reporting of intercoder reliability”, Human Communication Research 28, pp. 587–604. 16. NEUENDORF, K. A. (2002). The content analysis guidebook. SAGE Publications, Thousand Oaks 17. SCOTT, W.A. (1955). “Reliability of content analysis: The case of nominal scale coding”, Public Opinion Quarterly 19(3), pp. 321–325. 18. SCOTT, W. A. (1955). “Scott’s π (Pi). Reliability for nominal scale coding”, pp. 347– 349 In: Krippendorff, K., Bock, M.A. The Content Analysis Reader (2009), Los Angeles, London, New Delhi, Singapore

102

19. ROSENFIELD, G.H., FITZPATRICK-LINS, K. (1986). “A coefficient of agreement as a measure of thematic classification accuracy”,Photogrammetric engineering and remote sensing 52(2), pp. 223–227. 20. WEBER, R. P. (1990). Basic content analysis. SAGE Publications, Newbury Park, Londorn, New Delhi 21. VOGT, K., KUJAWIŃSKA, K. (2015). “Human factors in visual quality control”, Management and Production Engineering Review 6(2), pp. 25–31. 22. VOGT, K., KUJAWIŃSKA, K. (2010). “Analiza wpływu wybranych czynników pracy na skuteczność kontroli wzrokowej”, Inżynieria Maszyn 18(1), pp. 40–50.

Streszczenie: Wykorzystanie współczynnika Krippendorffa do określania wiarygodności oceny ludzkiej w wizualnej kontroli jakości procesów produkcji mebli. W artykule zaprezentowano wyniki badań skuteczności ludzkiej wizualnej kontroli jakości procesów produkcji mebli. Eksperyment miał na celu sprawdzenie wiarygodności ocen wg. trzech elementów: wiarygodności oceniającego sędziego, instrukcji na podstawie której określano jakość poszczególnych przedmiotów wykonanych z laminowanej płyty MDF oraz oceny jakości przez człowieka. Jako miarę skuteczności kontroli wizualnej zastosowano współczynnik Krippendorffa. Wyniki eksperymentu pozwolił stwierdzić, że zastosowanie ludzkiej wizualnej kontroli jakości procesów produkcji mebli może być dobrym rozwiązaniem. Wartości wszystkich trzech współczynników K1, K2 oraz K3 dały alpha > 0,8, co uznaje się za gwarancję wysokiej zgodności. Sędzia JD okazał się, więc osobą rzetelną i kompetentną, zaproponowana instrukcja jasna i wystarczająca, a otrzymane dane liczbowe wiarygodne. Okazało się, że wartości Krippendorff’s Coefficient różnią się w zależności od ilości kategorii (dla dwóch kategorii K2=0,904, K3=0,902; dla trzech kategorii K1=0,849). Sposób tworzenia kategorii nie wpływał na stabilność oceny. Współczynnik alpha można niewątpliwie uznać za dobrą, wygodną miarę monitorowania wiarygodności systemu pomiarowego, gdyż dając konkretną ocenę liczbową, pozwala kontrolować skuteczność wprowadzanych działań doskonalących

Corresponding authors:

Katarzyna Śmietańska, Piotr Podziewski, Faculty of Wood Technology SGGW, Department of Mechanical Processing of Wood, ul. Nowoursynowska 159, 02-776 Warsaw, Poland e-mail: [email protected] e-mail: [email protected]

ORCID ID: Śmietańska Katarzyna 0000-0001-8705-3700 Podziewski Piotr 0000-0002-2628-5062

103

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 104-110 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

The effect of the addition of primary fibers on the papermaking ability of wastepaper

ALEKSANDRA NISKA, EDYTA MAŁACHOWSKA Faculty of Wood Technology, Warsaw University of Life Sciences – SGGW, 159 Nowoursynowska Str., 02-776 Warsaw, Poland

Abstract: The effect of the addition of primary fibers on the papermaking ability of wastepaper. The rapid increase in paper production and consumption at the end of the 20th century contributed to the increasing difficulties connected with assurance of required resources base for paper industry and the utilization of the growing amount of paper waste. The wastepaper salvage and application of secondary pulps for paper production make a significant contribution to solving these issues, and these activities are justified based on ecology and economy. It must be said, however, that the addition of certain amount of primary fibers is necessary to ensure strength and others usable properties of produced paper. The authors have set themselves the goal of the examination of primary fibers addition on papermaking ability of wastepaper. Studies have shown that the addition of primary fibers to wastepaper has a positive impact on all examined strength properties of paper.

Keywords: wastepaper, cellulosic pulp, wood, paper properties

INTRODUCTION The rapid increase in paper production and consumption in industrialized countries caused a shortage of wood in many areas of the world in the second half of the 20th century, which is a serious problem, because the global demand for fibrous raw materials is still growing. At the beginning of the 20th century, the production of paper products in the world amounted to about 5 million tons [1], while in 2016, on a global scale, it reached the level of 410 million tons [2, 3], and it shows a constantly increasing trend [4] (Fig. 1). The consumption of paper products per capita is therefore around 57 kg per year [5].

450

400 n to 350

300

million , , 250

200

150 consumption

100 Paper Paper 50

0 1900 1920 1940 1960 1980 2000 Years

Figure 1. Graph of the paper consumption in the 20th and 21st century The paper industry is therefore making efforts to expand its raw material base. The search for alternative raw materials for paper production is additionally fueled by the growing ecological awareness of society manifested in the protection of the natural

104 environment, including forest resources, the pressure of ecologists concerned about the destruction of forest animal habitats, the rising prices of typical wood fiber materials and the possibilities of today's technology. Activities aimed at expanding the resource base of the paper industry include, among others, building plantations of fast-growing trees [6, 7], using non-wood fibrous raw materials [8, 9] and increased use of waste paper [10]. The secondary pulp currently accounts for 50% of the total amount of pulp used in the world [11], while in Europe the indicator is 40% [12]. They can be divided into recycled paper and dry broke produced in paper mills and paper processing plants. Recycled paper is the most commonly used secondary raw material. Secondary pulp is cheaper than the original one, so thanks to its use, the costs of paper production are reduced. The use of recycled paper in the paper production process significantly contributes to the expansion of the raw material base in the paper industry and to the utilization of the growing amount of paper waste [13]. Not only does it save space in rubbish dumps, but also an extremely valuable raw material for paper production - wood. Unfortunately in Poland, only 42% of wastepaper is recycled [14], while in Europe in 2014 this figure came up to about 70% [15]. Along with the increased use of recycled paper, the technology of converting it into secondary pulp is also developing. Depending on the type of recycled paper and its processing methods, we can obtain pulps with different characteristics. However, it should be taken into account that in order to ensure the strength and other properties of the produced paper, some primary fibers must be added. Therefore, this work examines the impact of adding recycled paper to virgin fibers on the papermaking potential of the produced paper.

MATERIALS AND METHODS The following materials were selected for research:  an industrial air-dried and bleached pine sulfate pulp in the form of sheets  wastepaper: - mixed wastepaper composed of unsorted wastepaper; (ranked as 3.19 according to EN643 “List of European standard types waste paper”) - white wastepaper, including products made from bleached pulps; (ranked as 3.04 according to EN643 “List of European standard types of waste paper”). The sheets of paper were produced in laboratory conditions from rewetted all pulp samples (22.5 g d.w. samples were soaked in water for 24 h) that were subjected to disintegration using the laboratory JAC SHPD28D propeller pulp disintegrator (Danex, Katowice, Poland) at 23000 revolutions, according to ISO 5263-1 (2004). Additionally, the waste papers were screened using a membrane screener PS-114 (Danex, Katowice, Poland) equipped with 0.2 mm gap screen to remove impurities. The disintegrated pine pulps were concentrated to the dry weigh content of 10% and refined in a JAC PFID12X PFI mill under standard conditions, according to ISO 5264-2 (2011). After refining of pine pulps and screening of waste paper, the three mixtures were prepared from waste papers and pine pulps. The mixtures were prepared with the addition of 15%, 30% and 50% of the primary pine fibers. After preparing the mixtures, the following properties of the pulps were evaluated:  The Schopper-Riegler freeness – the measurement of refining degree was performed using Schopper – Riegler apparatus (Danex, Katowice, Poland) in accordance with PN-EN ISO 5267-1 (2002).  WRV(Water Retention Value) – the water retention within fibers was examined centrifugal method which was developed by Jayme and Rothamel, according to ISO 23714 (2014). According to this method, the pulp sample is subjected to centrifugation using the acceleration of 3,000g for 15min and determining ratio of water to weight of bone dry sample [16].

105

The next step was forming sheets of paper in the Rapid-Koethen apparatus. The formation of paper sheets was performed in accordance with PN-EN ISO 5269-2 (2007). Each laboratory paper sheet was described by basis weight of 80 g/m². Only sheets with a base weight that ranged from 79 to 81 g/m² were accepted for further investigation. Test sheets were conditioned for 24 h at 23 ± 1°C and 50 ± 2% relative humidity, according to ISO 187 (1990). The paper properties were examined as follows. The roughness and the air permeability were measured on a Bendtsen apparatus (Kontech, Lodz, Poland). Mechanical measurements were performed on a Zwick Roell Z005 TN ProLine tensile testing machine (Zwick-Roell, Ulm, Germany), according to PN-EN ISO 1924-2 (2010):  IB – breaking length [m]; b  T – width related force with break [N/m]; W  T – force at break index [Nm/g];  T – strain at break [%]; b 2  WT – energy absorption [J/m ]; W  WT – energy absorption index [J/g];  Eb – tensile stiffness [N/m];  Ew – tensile stiffness index [Nm/g];  E* – Young’s modulus [MPa];  Flow – being of Young’s modulus [N];  FB – tensile force at break [N].

RESULTS AND DISCUSSION Properties of examined pulps are shown in Table 1. The Schopper-Riegler freeness determines the ability of the pulp to dehydration under standard conditions. A freeness is expressed in Schopper-Riegler degrees (°SR) on a scale from 0 to 100°SR [17]. The freeness of papers produced form mixed and white wastepaper was comparable. No significant differences have been observed between secondary pulps obtained from mixed and white wastepaper, but the freeness increases with the addition of primary fibers (Tab. 1).

Table 1. Properties of pulp WRV Freeness Raw material (with fines) °SR % 100% wastepaper 15 93 15% pine pulp 15 111 85% wastepaper Mixed wastepaper 30% pine pulp 17 117 70% wastepaper 50% pine pulp 17 130 50% wastepaper 100% wastepaper 14 103 15% pine pulp 14 116 85% wastepaper White wastepaper 30% pine pulp 15 124 70% wastepaper 50% pine pulp 17 135 50% wastepaper

106

The WRV indicator determines the amount of water retained in the pulp. The essence of this method is removing free water from the test pulp (contained between the fibers) and determining the content of stopped water (water contained inside the fibers) [18, 19]. The paper with the largest content of pine fibers achieved the highest value of WRV, for both mixed and white wastepaper (Tab. 1). Figure 2 shows a comparison of breaking length of papers received from wastepaper with different addition of pine pulp. The indicator is one of the fundamental static tensile properties of paper. It is the length of paper strip, if suspended vertically from one end, would break by its own weight [20]. Breaking length is generally used in the paper trade to describe the inherent strength of paper. It constitutes a very good basis for comparing the strength of papers made from different materials and of different basis weight. The breaking length is a widely used indicator, because it allows to estimate the usable properties of many products. This is particularly important for the evaluation of usefulness of packaging and newsprint papers. The breaking length depends on the average fiber length, among other things, which explains why the papers with increased participation of pine fibers have higher strength (Fig. 2, Tab. 2). 50% addition of primary fibers caused the increase of breaking length level of 70% in relation to the paper made only from mixed wastepaper. In white wastepaper, the increase of indicator was 33%. The indicator for papers made from white wastepaper is 12 ÷ 43% higher than in the mixed wastepaper (Fig. 2). Also for other mechanical properties, the strength grew with increasing the addition of primary fibers, irrespective of the waste paper type (Tab. 2).

4000

3700

3400

3100

2800 Mixed 2500 wastepaper White 2200 wastepaper

Breaking length, m Breaking 1900

1600

1300

1000 0 15 30 50 Primary fibers participation, %

Figure 2. The effect of primary fibers addition for wastepaper on changes in the breaking length of paper. Roughness of the paper (or its smoothness) is a very important usable property, because it determines the surface susceptibility to printing [21]. This property depends on, among others, the composition and properties of pulp or the papermaking process conditions. The value of primary fibers addition did not show significant effects on the roughness

107 of examined samples. The papers produced with mixed wastepaper showed higher roughness that ranged between 323 and 361 ml/min, probably due to the higher content of impurities in the pulp (Tab. 2). The content of pine pulp also did not show significant effects on the air permeability of papers (Tab. 2). Air permeability is related to the porosity of the product and depends on the structure density. The measurement of air permeability is an important indicator of the production process control, because it indicates the porosity of the product and strength properties, absorbency of the product or dielectric properties associated with porosity [22]. High air permeability is a desirable property in dustproof cartons and filter papers. However, many wrapping paper, and especially food packaging paper, should have a very low air permeability. The air permeability of papers made from mixed wastepaper ranged between 4712 and 4921 ml/min. The indicator greater than 5000 ml/min was unchanged for papers produced from white wastepaper (Tab. 2).

Table 2. Properties of paper The W b w *

b b W W WT E E E F F R* AP** part of T T T T low B primary 2 Pulp N/m Nm/g % J/m J/g N/m Nm/g MPa N N ml/min ml/min fibers

0% 1584 19.62 1.56 16.23 0.20 257100 3185 2337 2.30 23.50 358 5000

15% 1849 22.68 2.20 27.67 0.34 212600 2608 1932 1.38 27.62 323 4851

30% 2372 29.55 2.77 43.68 0.54 281000 3502 2555 2.05 32.71 332 4712 Mixed wastepaper 50% 2710 33.26 2.98 62.53 0.77 310500 3810 2821 2.35 38.36 361 4921

0% 2189 27.92 1.93 27.87 0.36 309900 3957 2817 2.76 33.55 313 5000

15% 2296 29.22 2.43 37.3 0.47 321400 4090 2922 2.85 33.22 297 5000

30% 3021 36.71 2.83 58.15 0.71 353500 4289 3212 2.68 44.16 292 5000 White wastepaper 50% 2954 37.18 3.13 61.91 0.78 332800 4184 3026 2.82 44.33 301 5000

*R – roughness *AP – air permeability

CONCLUSIONS 1. The addition of virgin fibers to wastepaper has a major influence on the strength properties of papers. 2. The addition of virgin pine pulp in secondary pulp does not significantly affect roughness and air permeability of paper, however, the smoothness and air permeability are higher for papers made form white wastepaper compared to mixed wastepaper. 3. The type of wastepaper does not significantly influence the pulp properties (freeness and WRV), but the fiber swelling degree is growing as the addition of primary fibers rises. 4. The availability of a constant, year-round supply of fiber is a primary concern for paper mills. It is right and necessary to permanent developing technology of the

108

wastepaper processing owing to the shortage of wood raw materials and the rising prices of typical wood fiber materials.

Acknowledgement. This work was financially supported by the National Centre of Research and Development in Poland (Project Number POIR.01.01.01-00-0084/17and POIR.04.01.04-00-22/18). The authors are also grateful to Natural Fibers Advanced Technologies for allowing the access to the company R&D laboratory.

REFERENCES

1. PRZYBYSZ K., PRZYBYSZ Z., DRZEWIŃSKA E., 2006: Wykorzystanie surowców roślinnych we współczesnym przemyśle papierniczym – stan i perspektywy rozwoju. Przemysł Chemiczny, 85(8-9): 1303-1306. 2. Rynek papieru i tektury w roku 2016, Rynek Papierniczy. pl. 2019. http://www.rynekpapierniczy.pl/artykul/rynek-papieru-i-tektury-w-roku-2016-1087. (Access 06.11.2019) 3. GODLEWSKA K., JASTRZĘBSKI M., 2017: Zużycie i produkcja papieru i tektury w Polsce w 2016 roku na tle krajów europejskich. Przegląd Papierniczy, 73(9): 593. 4. RENNER M., 2015: Paper Production Levels Off. Vital Signs. Global trends that shape our future. 5. Statistics and facts about the global paper industry. https://www.statista.com/topics/1701/paper-industry/. (Access 06.11.2019) 6. CHRISTERSSON L., 2008: Poplar plantations for paper and energy in the south of Sweden. BiomassBioenerg, 32: 997-1000. 7. ZAMORA D., WYATT G., APOSTOL K., TSCHIRNER U., 2013: Biomass yield, energy values, and chemical composition of hybrid poplar in short rotation woody crops production and native perennialgrasses in Minnesota, USA. Biomass Bioenerg, 49: 222-230. 8. GIROUARD P., SAMSON R., 2000: The potential role of perennial grasses in the pulp and paper industry. Resource Efficient Agricultural Production (REAP) - Canada. Ste. Anne de Bellevue, Qs. Canada https://www.researchgate.net/publication/289089257_Potential_role_of_perennial_gra sses_in_the_pulp_and_paper_industry/.(Access 07.11.2019) 9. ZALESNY JR. R. S., CUNNINGHAM M.W., HALL R. B., MIRCK J., ROCKWOOD D. L.,STANTURF, J., AND VOLK T. A., 2011: “Woody biomass from short rotation energy crops,” Sustainable Production of Fuels, Chemicals, and Fibers from Forest Biomass, J. Zhu et al. (ed.), American Chemical Society, Washington, DC, pp. 27-63. DOI: 10.1021/bk-2011-1067.ch002 10. FIETZ M., 2016: Makulatura-pochodzenie, przerób, wykorzystanie. Nauki Inżynierskie i Technologie, Uniwersytet Ekonomiczny we Wrocławiu, 2(21) s.9-27. 11. GODLEWSKA K., JASTRZĘBSKI M., 2016: Zużycie i produkcja papieru i tektury w Polsce w 2015 roku na tle krajów europejskich. Przegląd Papierniczy, 72(9): 527-535. 12. HOLIK H., 2013: Paper and Board Manufacturing – an Overview. In: Handbook of Paper and Board, 2nd ed. Willey-VCH, Germany. 13. DOLIWA M., 2009: Makulatura w praktyce, Przegląd Papierniczy, nr 65 (11), s. 660- 661. 14. Recykling papieru (makulatury), Wszystko o recyklingu. https://www.oostdam.pl/recykling-papieru-makulatury/. (Access 07.11.2019) 15. Paper recycling. Monitoring Report: 2014. European Declaration on Paper Recycling, Brussels. 16. SZWARCSZTAJN E., 1963: Technologia papieru, część I, WPLiS, Warszawa.

109

17. SUREWICZ W., 1971: Podstawy technologii mas włóknistych. Surowce włókniste przemysłu celulozowo-papierniczego i ich składniki chemiczne. Wydawnictwo Naukowo – Techniczne, Warszawa. 18. BÄCKSTRÖM M., KOLAR M., HTUN M., 2008: Characterization of fines from unbleached kraft pulps and their impact on sheet properties. Holzforschung 62:546– 552. 19. CHENG Q., WANG J., MCNEEL J., JACOBSON P., 2010: Water retention value measurements of cellulosic materials using a centrifuge technique. BioResources 5:1945-1954. 20. PRZYBYSZ K., 1997: Technologia celulozy i papieru – Technologia papieru. WSiP, Warszawa. 21. KLEIN R., SCHULZE U., 2007: Znaczenie podstawowych właściwości papieru dla oceny drukowności papierów niepowlekanych we wklęsłodruku. Przegląd Papierniczy 63, 5, str. 293-299. 22. MODRZEJEWSKI K., OLSZEWSKI J., RUTKOWSKI J., 1985: Metody badań w przemyśle celulozowo – papierniczym. Wydawnictwo Politechniki Łódzkiej, Łódź.

Streszczenie: Wpływ dodatku włókien pierwotnych na zdolność papierotwórczą makulatury. Szybki wzrost produkcji i zużycia papieru przyczyniły się pod koniec XX wieku do narastających trudności związanych z zapewnieniem odpowiedniej bazy surowcowej dla przemysłu papierniczego oraz utylizacji rosnącej ilości odpadów papierowych. Odzyskanie makulatury i wykorzystanie wtórnych mas włóknistych do produkcji wytworów papierowych w znacznym stopniu przyczyniają się do rozwiązania tych trudności i są jak najbardziej uzasadnione względami ekologicznymi oraz ekonomicznymi. Należy jednak wziąć pod uwagę fakt, że w celu zapewnienia wytrzymałości i innych właściwości użytkowych wytwarzanego papieru konieczny jest dodatek pewnych ilości włókien pierwotnych. Za cel niniejszego artykułu autorzy postawili, zatem zbadanie wpływu dodatku włókien pierwotnych na zdolność papierotwórczą masy makulaturowej. Stwierdzono, że dodatek włókien pierwotnych do masy makulaturowej pozytywnie wpływa na wszystkie badane właściwości wytrzymałościowe papieru.

Corresponding author:

Edyta Małachowska Faculty of Wood Technology, Warsaw University of Life Sciences – SGGW, 159 Nowoursynowska Str., 02-776 Warsaw, Poland email: [email protected] phone:+48 22 59 385 45

ORCID ID: Małachowska Edyta 0000-0002-0291-9278

110

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 111-118 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Trends in employment and labour productivity in the woodworking industry in selected EU countries

EMILIA GRZEGORZEWSKA, JUSTYNA BIERNACKA, IZABELA PODOBAS Department of Technology and Entrepreneurship in Wood Industry, Institute of Wood Sciences and Furniture, Warsaw University of Life Sciences, Poland

Abstract: Trends in employment and labour productivity in the woodworking industry in selected EU countries. Labour productivity and employment levels are among the most important factors determining the development of enterprises, regardless of the nature of their economic activity. From the point of view of the furniture industry, whose significant position in the ranking of producers and exporters is influenced by the woodworking industry, the main supplier of raw material for production, it seems important to analyse the dynamics and structure of employment, as well as to assess the use of human resources in wood processing companies. The main objective of the paper was to compare selected aspects of the labour market and the efficiency of human resources use in the European Community, with particular emphasis on the countries belonging to the Visegrad Group. The research was supplemented by a detailed analysis of men employed in the woodworking industry and their belonging to particular age groups. On the basis of the conducted analyses it was indicated that there are differences in the aspect of production potential, employment level and labour productivity.

Keywords: wood industry, employment, labour productivity, EU countries

INTRODUCTION Nowadays, the labour market in the EU faces many problems and dilemmas; an important one is the issue of efficient use of human resources in the wood industry. Crucial aspects are the adjustment of the level and structure of employment to the existing conditions and technical parameters, as well as the development of the competence of employees adequate to the implemented, innovative technical and organizational solutions. Wood industry is an important branch of the EU economy, and the European Community plays a special role in the global market of wood and wood products. This is confirmed by a high position of some EU countries, in particular the old member states, in the world ranking of producers and exporters of wood products. It is also worth noting that Poland is one of the leaders in terms of the value of production and furniture export (Ochman-Nowicka, Lis 2004; Adamowicz, Wiktorski 2006; Grzegorzewska, Stasiak-Betlejewska 2014; KPMG Report 2017), and among the most important semi-finished products used in the production of furniture are wood-based panels manufactured by wood processing enterprises. The analysis of the furniture manufacturing sector conducted by Gołaś (2017) revealed that there are significant differences between EU countries in terms of labour productivity measured by added value. Overall, the highest labour productivity was recorded in the furniture industry in the old EU member states, while the countries that joined the EU after 2003 had the lowest productivity. Moreover, the regression model revealed that the technological equipment used by employees was the factor causing the greatest diversification in the EU furniture industry (Report EU furniture; Landmann 2004; Gołaś 2017). Labour market flexibility is its ability to adapt quickly to changing conditions and technologies. According to the macroeconomic approach, this means a way to achieve a balance in the labour market that can be distorted by different fluctuations in supply and demand or structural changes. The manner and pace of rebalancing depend on the degree of flexibility of the labour market, including labour demand, labour supply and wage flexibility. According to the microeconomic approach, labour market flexibility is the ability of enterprises to adapt to the scope of human resources policy and to satisfy the living needs of

111 both employees and job seekers. Among the EU28 economies, a strong differentiation can be observed on the “core” and “peripheries” (Landesmann et al. 2015), which is reflected in the growing disparity in labour productivity (Filippetti and Peyrache 2013). The general context in which the productivity gap in the EU is growing is defined as “great divergence”. (OECD 2018). Labour market can be analysed with the use of many indicators. However, one of the most important factors affecting the development of economic entities and particular sectors of the economy, including the wood industry, is the effective use of labour resources. Achieving high labour productivity contributes to reducing costs, increasing the supply of cheaper goods and services, thus translates into an increase in the purchasing power of societies, their wealth and competitive abilities (Gołaś, Kozera 2008).

MATERIALS The main objective of the research was to analyse selected aspects of employment and labour productivity in the woodworking industry in the countries belonging to the Visegrad Group (i.e. the Czech Republic, Poland, Hungary, Slovakia). The comparative analysis was carried out against the background of all EU member states (EU28), as well as the group of countries that joined the European Union after 2003. (EU13). The temporal scope of the research was adopted for the years 2010–2017. During the research, a horizontal analysis (allowing to determine the dynamics of selected economic and financial categories) was carried out, as well as a vertical analysis (determining the importance of individual countries in creating the value of sold woodworking industry and jobs in this industry). In order to determine the level of relative differentiation of the studied characteristics, the coefficient of variation was used, which is the relation of standard deviation and the average value of a given characteristic, taking into account the whole period covered by the research. Further parts of the research focused on the analysis of employment in the wood industry and investigated the profile of employment by calculating the number of men employed in the wood industry and referring this value to the total number of employees in this sector of industry. In order to make the analysis of employment more detailed, the percentage of men employed in the woodworking industry was examined and assigned to the adopted age groups, namely: 15–39 years of age, 40–59 years of age and over 60 years of age.

RESULTS According to Eurostat data, in 2010 the sold production of the wood industry in the EU28 countries amounted to 112.0 billion euro, of which only 16.0% was produced in the new Member States (Table 1). This confirms a much greater importance of the EU15 countries in the sold production of wood and wood products. Among the analysed countries of the Visegrad Group, Poland was the largest producer of wood and wood products. The value of sold production of the wood processing industry in this country in 2010 was 6.1 billion euro; this accounted for 5.4% and as much as 34.1% of the product value of wood industry in, respectively, the EU28 and EU13. The Czech Republic should be mentioned as the next largest producer of the wood industry. The value of sold production of wood and wood products in 2010 amounted to 3.1 billion euro, which constituted 2.8% of the value of production of this industry produced by all EU Member States and 17.3% of the production achieved by the countries included in the EU13. At the beginning of the analysed period, a significantly lower value of sold production of the wood industry was observed in Slovakia (1.2 billion euro) and Hungary (0.7 billion euro). The share of these countries in creating the value of wood and wood products was 1.1 and 0.6%, respectively. Between 2010 and 2017, the sold production of the wood industry in the EU countries increased by 12.2% and at the end of the analysed period amounted to 125.7 billion euro.

112

Such a slight increase within a seven-year period was caused mainly by a decrease in the dynamics of the discussed phenomenon in years 2011–2012, which was also a consequence of long-term effects of the economic crisis which affected the majority of sectors of the national economies of the EU countries.

Table 1. Sold production and employment in the wood industry in selected EU countries in years 2010–2017 Itemisation 2010 2011 2012 2013 2014 2015 2016 2017 2017/2010 V* Sold production value [billions euro] UE-28 112.0 118.1 114.0 112.2 118.5 122.2 121.5 125.7 112.2 4.2 UE-13 17.9 19.6 19.5 20.3 22.2 23.2 23.0 24.9 138.9 11.0 Czechia 3.1 3.4 3.2 3.0 3.0 3.2 3.3 3.5 111.8 4.9 Hungary 0.7 0.7 0.7 0.7 0.8 0.8 0.9 1.0 135.2 13.7 Poland 6.1 6.6 6.7 6.9 7.7 8.1 7.6 8.5 140.3 11.4 Slovakia 1.2 0.9 0.8 0.8 1.1 1.2 1.0 1.1 95.1 16.0 Structure of sold production value [%] 2017–2010 V* UE-28 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 UE-13 16.1 16.6 17.1 18.1 18.8 19.0 18.9 19.8 3.7 7.4 Czechia 2.8 2.9 2.8 2.7 2.6 2.6 2.7 2.8 0.0 3.4 Hungary 0.6 0.6 0.6 0.6 0.6 0.7 0.7 0.8 0.2 9.9 Poland 5.4 5.6 5.9 6.2 6.5 6.6 6.3 6.8 1.4 7.9 Slovakia 1.1 0.8 0.7 0.7 1.0 1.0 0.8 0.9 -0.2 14.9 Number of people employed [thousand] 2017/2010 V* UE-28 1 060.9 1 041.0 1 001.8 967.9 982.6 975.8 972.4 970.8 91.5 3.6 UE-13 385.0 387.7 376.5 369.6 381.8 387.5 387.6 385.3 100.1 1.7 Czechia 61.2 60.4 59.2 55.4 54.4 54.3 54.5 53.8 87.9 5.4 Hungary 18.5 17.6 17.2 17.0 17.4 18.0 18.4 18.3 98.9 3.2 Poland 124.8 121.8 115.4 113.9 120.8 123.8 128.3 127.9 102.5 4.3 Slovakia 28.1 27.9 22.9 20.1 23.4 24.7 21.9 23.9 85.0 11.4 Structure of people employed [%] 2017–2010 V* UE-28 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 UE-13 36.3 37.2 37.6 38.2 38.9 39.7 39.9 39.7 3.4 3.4 Czechia 5.8 5.8 5.9 5.7 5.5 5.6 5.6 5.5 0.0 0.0 Hungary 1.7 1.7 1.7 1.8 1.8 1.8 1.9 1.9 3.7 7.4 Poland 11.8 11.7 11.5 11.8 12.3 12.7 13.2 13.2 0.0 3.4 Slovakia 2.6 2.7 2.3 2.1 2.4 2.5 2.3 2.5 0.2 9.9 *V – Coefficient of variation Source: own elaboration based on Eurostat.

In 2017, the share of the EU13 countries in the production of wood industry was 19.8%, i.e. 3.7 p.p. higher than at the beginning of the analysed period. This resulted from a higher growth rate of the value of sold production of the wood processing industry in this group of countries than in the EU15 (38.9 compared to 12.2%). It should be noted, however, that it is still the member states that joined the EU before 2004 that played a greater role in creating the sold production of wood and wooden products, although the difference between the distinguished groups of countries decreased. The highest increase in the production of the wood industry in terms of value was observed in the case of Poland – by 40.3% to the level of 8.5 billion euro. Favourable economic tendencies in this country caused an increase in its

113 importance in creating the value of sold production of wood and wood products in the EU (from 5.4 to 6.8%). The Czech Republic was ranked next in terms of the value of production of wood and wood products among the Visegrad Group countries (3.5 billion euro), which recorded a slight increase in this indicator (11.8%). This contributed to maintaining the country’s share in generating the value of sold production of the wood processing industry in the EU at the same level. The situation was diversified in the case of other Visegrad Group countries of lesser importance for the creation of EU wood industry production. A significant increase (by 35.2%) in the value of sold production of wood and wood products was observed in Hungary – 1.0 billion euro at the end of the analysed period amounted. In Slovakia, on the other hand, there was a drop by 4.9%, which contributed to a decrease in the importance of this country in creating the EU value of sold production of the wood industry. The highest level of the coefficient of variation in the value of sold production of this industry, showing a relative diversity of this feature, was observed for the Slovak wood industry (16.0%) and the lowest for the Czech Republic (4.9%). As shown in Table 1, in 2010, the woodworking industry in the EU28 countries employed 1060.9 thousand workers, of which 385.0 thousand in the group of new member states, which constituted 36.3% of the total employment in this industry in the European Union. Again, among the Visegrad Group countries, the first place was taken by Poland, where 124.8 thousand people found employment in the production of wood and wood products, which constituted 11.8% of the employed in the EU28 and as much as 32.4% of the employed in the EU13. The level of employment in the Czech Republic was 50% lower than in Poland and in 2010 it was 61.2 thousand people, which constituted 5.8% of the employed in the wood industry in all of the EU member states. Significantly lower employment was observed in Slovakia (28.1 thousand people) and Hungary (18.5 thousand people). The share of these countries in the creation of jobs in the EU wood industry was 2.8 and 1.6%, respectively. It should be noted that the number of people employed in the wood industry in relation to the total number of people employed in this industry in the Czech Republic, Poland and Slovakia was at a similar level of 5–6%. In Hungary, on the other hand, the indicator was almost twice as low (2.8%). Between 2010 and 2017 employment in enterprises manufacturing wood and wood products in the EU decreased by 8.5% and at the end of the analysed period was at the level of 970.8 thousand people, 385.3 of whom were employed in companies located in the EU13. In this respect the significance of the new member states increased by 3.7 p.p. to the level of 39.7%. Among the analysed countries of the Visegrad Group, only in Poland there was a subtle increase in the number of employees (by 2.5%) to the level of 127.9 thousand people. This contributed to the slightly increasing role of Poland in creating jobs in the woodworking industry. In other countries covered by the analysis the employment level decreased by 15.0% in Slovakia, 12.1% in the Czech Republic and 11.1% in Hungary. The highest rate of the variation coefficient of the number of employees, which determines the relative diversity of the feature, was observed for the Slovak wood industry (11.4%), and the lowest one for Hungary (3.2%). One of the basic measures of labour productivity is the value of production per employee. As can be seen form the data presented in Figure 1 at the beginning of the analysed period, it amounted to 105 thousand euro on average in the EU member states, and it was more than twice as low for the EU13 countries. This proves that the labour productivity in the woodworking industry in the EU13 countries is lower than the average in the EU member states. Among the countries that are members of the Visegrad Group, the highest values of sold production of wood and wood products per employee were reached by the Czech Republic (50.9 thousand euro) and Poland (48.5 thousand euro). Moreover, those two

114 countries also exceeded the average of the new member states. Lower values of this indicator were recorded in Slovakia (42.5 thousand euro) and Hungary (39.0 thousand euro). Between 2010 and 2017, the value of sold production of the wood processing industry per employee increased by 22.5% to 129.4 thousand euro in the EU28. In the new member states, this indicator increased by 38.8%. At the end of the analysed period, it was at the level of 64.7 thousand euro, but it was still significantly lower than the average in the EU28. Among the countries covered by the analysis, the highest labour productivity was recorded in Poland. The abovementioned rate increased by 37.0% to 66.4 thousand euro. The Czech Republic (64.7 thousand euro) ranked next in terms of labour productivity of the woodworking industry measured by the value of production per employee. Significantly lower rates were observed in Slovakia (47.5 thousand euro) and Hungary (53.3 thousand euro), although it should be noted that the latter country recorded a significant rate of growth in the value of woodworking industry production per employee (36.6%).

Chart 1. Value of sold production of wood industry per employee in selected EU countries in 2010–2017 Source: own elaboration based on Eurostat.

The analysis of employment in the wood industry was developed by analysis of the structure of employment depending on the age group of employees. Data on the classification of men employed in the woodworking industry by age group was presented in Table 2. Between 2010 and 2017, in EU28 countries, in all three age groups, men employed in the woodworking industry accounted for more than 80% of all employees. In the age groups 15–39 and 40–59, the share of employed men among all employees ranged from 83.5% to 87.0% and from 81.5% to 83.9%, respectively. The greater variability in the number of all men employed in that industry was observed in the group of employees over 60 years of age (3.2%). Additionally, in the last year of the analysis, this group was characterized by a slight increase in the share of the structure of employment in the woodworking industry. Only a larger share than in 2017 can be seen for 2012, when the share of men in total employment reached 90.2%. A significant share of men in the total number of the employed can also be found in the analysis of the discussed figure in the EU13. In the group of people aged 15–39, the share of men ranged from 83.4–87.7%. A similar structure of men among all employees was observed in the group of people aged 40–59 (from 79.1% in 2012 and 2014 to 80.9% in 2016). As in

115 the case of EU28 countries, the lowest share of employed men was observed in the two youngest age groups of employees – the highest value was again observed for people over 60 (3.7%). Moreover, this age group recorded a decrease of 2.5 p.p. in 2017 in comparison to 2010.

Table 2. Percentage of men employed in the woodworking industry by age group in selected EU countries 2010– 2017 2017– Itemisation Age group 2010 2011 2012 2013 2014 2015 2016 2017 2010 V* [p.p] 15–39 84.2 85.6 84.7 84.9 83.5 87.0 86.2 83.8 -0.4 1.4 UE-28 40–59 81.5 82.0 83.9 82.6 81.9 82.0 82.6 82.0 0.5 0.9 60 or over 86.9 86.1 90.2 86.7 83.6 81.4 85.1 88.2 1.3 3.2 15–39 83.4 85.3 84.4 84.7 84.8 87.7 86.0 83.7 0.3 1.6 UE-13 40–59 79.7 79.5 79.1 78.4 79.1 80.7 80.9 80.3 0.6 1.1 60 or over 92.8 91.7 92.9 92.3 90.9 83.6 86.5 90.3 -2.5 3.7 15–39 85.4 87.0 80.0 87.0 88.1 92.1 78.4 84.0 -1.4 5.2 Czechia 40–59 77.3 75.5 79.2 78.8 78.2 83.6 77.1 81.4 4.1 3.3 60 or over 91.7 95.5 100.0 81.5 96.3 90.0 95.5 97.5 5.8 6.2 15–39 80.5 87.0 90.4 92.2 96.8 88.7 89.5 90.7 10.2 5.2 Hungary 40-59 86.0 88.8 86.9 83.2 77.0 70.0 71.3 83.2 -2.8 8.9 60 or over 100.0 88.6 88.9 100.0 85.7 100.0 93.3 86.4 -13.6 6.8 15–39 83.5 86.4 86.7 83.7 81.8 88.6 89.4 87.7 4.2 3.1 Poland 40–59 82.8 82.3 81.0 81.7 83.7 79.4 84.9 81.1 -1.7 2.1 60 or over 85.4 92.5 98,8 100.0 91.5 79.2 85.7 95.0 9.6 7.9 15–39 92.4 93.9 89.1 80.7 92.6 92.6 75.3 76.7 -15.7 9.0 Slovakia 40–59 79.5 81.1 87.8 89.3 81.7 77.7 77.5 92.1 12.6 6.7 60 or over 100.0 100.0 87.5 100.0 83.3 100.0 100.0 100 0.0 7.1 *V – Coefficient of variation Source: own elaboration based on Eurostat.

Analysing the share of male employees in the woodworking industry in the Visegrad countries, it can be observed that the distinguished age groups were characterized by a greater variability than in the EU13 countries, to which the Visegard countries belong. In the Czech Republic, the age group that showed the greatest variability was the + 60 years old. The coefficient of variation for this group was at the level of 6.2%, and the values in the analysed years ranged from 81.5% to 100.0%. This value may result from a small number of employed women whose age corresponds with that age group. The analysis of the figures obtained for the abovementioned age group shows that during as many as 7 years of the study, the share of male employees was over 90%. The percentage of men employed in other age groups in the Czech Republic ranged from 77.3% to 92.1%. In contrast to the Czech Republic, the highest variability in the share of the male employees (8.9%) in Hungary was observed in the 40–59 age group. The percentage of men employed in woodworking companies ranged from 70 to 88.8%, while the lowest values were recorded in the final years period covered by the analysis, which indicates an increasing share of women in the number of employees in woodworking companies. In the case of Poland, the percentage of men employed in wood processing companies was the lowest in the 40–59 age group. There were significantly more men in this group – from about 80% to almost 85%. In 2017, however, the share of men decreased to 81.1%,

116 which also proves that the number of employed women in this group is growing. In other classification groups of the employed, the number of male employees was equally high and ranged from 81.8% in the age group 15–39 to even 100.0% in the age group 60 and more. In contrast to the previously analysed countries of the Visegrad Group, Slovakia was characterized by the highest coefficient of variation of the examined figure in the group of people aged 15–39 – the value of this indicator was 9.0%. In the abovementioned group employed men constitute between 75.3 and 93.9% of the total number of employed people.

CONCLUSIONS The research allows the following conclusions to be drawn: 1. Among the analysed countries of the Visegrad Group, Poland played a leading role in the creation of production value and jobs in the wood industry. Almost one third of wood industry production in terms of value in the EU13 countries was achieved in Polish companies. Over 32% of people employed in the wood industry in the new EU member states found employment in Polish enterprises. A lower level of employment, as well as a lower value of sold production of wood and wood products, was observed in Hungary and Slovakia. 2. The efficiency of the use of labour resources is much lower in the EU13 than the average in the European Community as a whole. The highest productivity measured as the value of sold production per employee was recorded in Poland and the Czech Republic. 3. The high share of men in particular age groups of the total number of employed may result from the specificity of the woodworking industry. The percentage of men employed, regardless of their age group, in both the EU13 and EU28 countries, was over 80%. 4. In the Visegrad countries, it can be observed that the employment level in the selected age groups was more volatile than the average in the EU13 group of countries to which they belong. 5. The share of women in the 15–39 and 40–59 age groups in the total number of the employed has increased in recent years in both the EU28 and the EU13. On the other hand, in the Czech Republic and Slovakia, the proportion of women in each age group has decreased in recent years. The share of the employed women in the 15–59 age group in Poland increased at the end of the analysed period.

REFERENCES

1. ADAMOWICZ M., WIKTORSKI T., 2006: Kondycja i perspektywy rozwoju polskiego przemysłu meblarskiego, Annals of Warsaw Agricultural University – SGGW Forestry and Wood Technology No 56, 2006, Warszawa 2. FILIPPETTI A., PEYRACHE A., 2013: Labour Productivity and Technology Gap in European Regions: A Frontier Approach, Regional Studies 49(4); pp. 532–544. 3. GOŁAŚ Z., KOZERA M., 2008: Strategie wydajności pracy w gospodarstwach rolnych, Journal of Agribusiness and Rural Development, Vol. 1 (7); pp. 73–87. 4. GOŁAŚ Z., 2017: The diversification and determinants of labour productivity in furniture industry in the EU countries, Intercathedra 3 (33); pp. 30–36. 5. GRZEGORZEWSKA E, STASIAK-BETLEJEWSKA R., 2014: The influence of global crisis of financial liquidityand changes in corporate debt of furniture sector in Poland, Drvna Industrija, Vol. 65, No. 4; pp. 315–322.

117

6. KPMG REPORT, 2017: Rynek meblarski w Polsce, Warszawa, https://assets.kpmg.com/content/dam/kpmg/pl/pdf/2017/06/pl-Raport-KPMG-Rynek- meblarski-w-Polsce-2017.pdf. 7. LANDESMANN M., LEITNER S., STEHRER R., 2015: Competitiveness of the European economy, EIB Working Papers from European Investment Bank (EIB) No 2015/01 8. LANDMANN O., 2004: Employment, productivity and output growth. Employment Strategy Papers, International Labour Organization, No. 17, Geneve. 9. OCHMAN-NOWICKA J., LIS W., 2004: Znaczenie eksportu dla rozwoju polskiego meblarstwa, Przemysł Drzewny, No. 3; pp. 11–14. 10. OECD 2018: Development Co-operation Report 2018, https://www.oecd.org/development/development-co-operation-report-20747721.htm 11. RAPORT EU FURNITURE: The EU furniture market situation and a possible furniture products initiative. Final report, Submitted to the European Commission DG Enterprise and Industry, Brussels, November 2014, http://ec.europa.eu/growth/sectors/raw-materials/industries/forest-based/furniture/ 12. SZYMAŃSKA A., 2017: The Labour Market in the Visegrad Group Countries – Selected, “Olsztyn Economic Journal” 12 (3); pp. 289–306.

Streszczenie: Tendencje zmian w zatrudnieniu i wydajność pracy w przemyśle drzewnym w wybranych krajach UE. Wydajność pracy i poziom zatrudnienia stanowią jedne z ważniejszych czynników determinujących rozwój przedsiębiorstwa, niezależnie od charakteru prowadzonej przez nie działalności gospodarczej. Z punktu widzenia przemysłu meblarskiego, na którego znaczącą pozycję w rankingu producentów i eksporterów wywiera wpływ przemysł drzewny, główny dostawca surowca do produkcji, istotną kwestią wydaje się dokonanie analizy dynamiki i struktury zatrudnienia, a także oceny wykorzystania zasobów ludzkich w zakładach zajmujących się przerobem drewna. Głównym celem artykułu jest porównanie wybranych aspektów dotyczących rynku pracy i wydajności wykorzystania zasobów ludzkich we Wspólnocie Europejskiej, ze szczególnych uwzględnieniem krajów należących do Grupy Wyszehradzkiej. Uzupełnienie badań stanowi szczegółowa analiza zatrudnionych w przemyśle drzewnym mężczyzn i ich przynależność do poszczególnych grup wiekowych. Na podstawie przeprowadzonych analiz wskazano, że występują różnice między wyróżnionymi krajami w aspekcie potencjału produkcyjnego, poziomu zatrudnienia i wydajności pracy.

Corresponding author:

Grzegorzewska Emilia Institute of Wood Sciences and Furniture Warsaw University of Life Sciences – SGGW 159 Nowoursynowska str.; 02-776 Warsaw, Poland [email protected]

ORCID ID: Grzegorzewska Emilia 0000-0002-7532-9287 Biernacka Justyna 0000-0003-3407-1280 Podobas Izabela 0000-0001-8315-0386

118

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 119-127 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019) The trends in employment and labour productivity in the pulp and paper industry in the selected EU countries

JUSTYNA BIERNACKA, EMILIA GRZEGORZEWSKA, IZABELA PODOBAS Department of Technology and Entrepreneurship in Wood Industry, Institute of Wood Sciences and Furniture, Warsaw University of Life Sciences, Poland

Abstract: The Trends in Employment and Labour Productivity in the Pulp and Paper Industry. Despite many challenges, the pulp and paper industry is a dynamically developing branch of industry in the world. It seems that especially growing competition from less developed countries, such as Indonesia, and the falling demand for paper may be reasons for the difficulties of pulp and paper factories, i.e. decreasing profitability of paper production and number of employed in this industry in recent years. From the perspective of the employment volume and related work efficiency, it was interesting to examine and assess employment and the use of human resources in pulp and paper industry, especially in the countries of the Visegrad Group, which economies have been characterized by stable GDP growth in recent years. The paper analyses the value of pulp and paper industry production and its structure in the Visegrad Group countries relative to EU28 countries and employment and age structure of the employed.

Keywords: pulp and paper industry, employment, labour productivity, EU countries

INTRODUCTION Forests are increasingly valued for their environmental functions, such as protecting the water quality of headwaters and rivers and regulating its quantity over time, preventing soil erosion, protecting human settlements from avalanches, filtering airborne pollutants, harbouring biodiversity and providing space for recreation (Agriculture, forestry and fishery statistics… 2018). However, forests are still the primary source of raw material used in the wood industry and pulp and paper industry. The pulp and paper industry is one of the largest industries in the world (Bajpai 2014; Rizk et. all 2004). It is also a dynamically developing industry in the world consisting of nearly 5000 pulp and paper mills around the world with annual production capacity of almost 400 million tons of paper (Vaez, Ziloue 2020). It is dominated by North American, Northern European and East Asian countries. Latin America and Australasia also have significant pulp and paper industries. Over the next few years, it is expected that both India and China will become the key countries in the industry’s growth (Bajpai 2014). The European pulp and paper sector has undergone significant consolidation since the turn of the century. European producers are facing increased competition from Asia, which is sometimes driven not only by market fundamentals, but also output and employment targets (Roh et al. 2016). Challenges, such as increasing competition from low-income countries such as Brazil and Indonesia, an increase in the use of digital media communications leading to a decline in paper demand, as well as EU environmental policy have led to stalled growth and a decline in profitability in recent years of this industry, which in turn resulted in a 30% decrease in the number of companies in Europe in the years 2000–2017. These problems have an impact on the employment rate in the European pulp and paper industry – it is estimated that since 2000 the number of employees has decreased by 37% (Brunnhofera et. all 2020). As compared to other European countries, Slovakia, Czech Republic and Poland have relatively large timber resources. The forest coverage of these countries amounts to 40.3, 34.5 and 30.8% (Forestry 2018). Hungary showed the lowest forest cover among the analysed countries. The area of forests in this country is 1940 thousand ha, which is 22% of the land area. However, mentioned above countries belonging to the Visegrad Group are characterized

119 by stable GDP growth year-to-year, which was an important premise for an attempt to analyse the situation of pulp and paper industry companies in terms of possible improvements in fields of employment and labour productivity (IMF 2018). The upward demand for the pulp and paper industry products, both in relation to the domestic consumers and foreign buyers, means that companies in this industry have to face challenges relating to the development of production potential (Grzegorzewska, Stasiak- Betlejewska 2019). From this point of view, not only the production potential which results from the available raw wood material and owned machine park is important. Increase of competitiveness of individual countries on the international market also results directly from an effective use of labour resources. Achieving high labour productivity contributes to reducing costs, increasing the supply of cheaper goods and services, and thus translates into an increase in the purchasing power of societies, their wealth and competitive abilities (Gołaś, Kozera 2008).

MATERIALS The main objective of the research was to analyse selected aspects of employment and work efficiency in the pulp and paper industry in the countries belonging to the Visegrad Group (i.e. the Czech Republic, Poland, Hungary, and Slovakia). The comparative analysis was carried out against the background of all EU countries (EU28), as well as the group of countries that joined the European Union after 2003. (EU13). The temporal scope of the research was adopted for the years 2010–2017. During the research a horizontal analysis was carried out, allowing to determine the dynamics of selected economic and financial categories, as well as a vertical analysis, determining the importance of individual countries in creating the value of sold production of pulp and paper industry and jobs in this industry. In order to determine the level of relative differentiation of the studied characteristics, the coefficient of variation was used, which is the relation of standard deviation and the average value of a given characteristic, taking into account the whole period covered by the research. Further part of the research focused on the analysis of employment in the pulp and paper industry and investigated the profile of employment by calculating the number of men employed in this industry and referring this value to the total number of employees in this sector. In order to make the analysis of employment more detailed, the percentage of men employed in the pulp and paper industry was examined, ranking the employed with regard to age to the adopted groups, namely: 15–39 years of age, 40–59 years of age and over 60 years of age.

RESULTS According to Eurostat data, in 2010 the sold production of the pulp and paper industry in the EU28 countries amounted to 159.8 billion euro, of which only 8.4% was produced in the new Member States (Table 1). This confirms the much greater importance of the EU15 countries in the sold production of paper and paper products. Among the analysed countries of the Visegrad Group, Poland was the largest producer of paper and paper products. The value of sold production of the pulp and paper industry in this country in 2010 was as follows 6.1 billion euro, which accounted for 3.8% and as much as 45.6% respectively of the value of the production of this industry produced by the EU28 and EU13. The Czech Republic should be mentioned as the next largest producer of the pulp and paper industry among the analysed countries. The value of sold production of paper and paper products in 2010 amounted to 2.3 billion euro, which constituted 1.4% of the value of production of this industry produced by all EU Member States and 17.2% of the production achieved by the countries included in the EU13. At the beginning of the analysed period, a significantly lower value of sold production

120 of the pulp and paper industry was observed in Slovakia (1.2 billion euro) and Hungary (1.2 billion euro). The share of these countries in creating the value of sold production in this industry was equal and amounts to 0.8%.

Table 1. Production sold and employment in the pulp and paper industry in selected EU countries in the years 2010–2017 Itemisation 2010 2011 2012 2013 2014 2015 2016 2017 2017/2010 V* Production value [billion euro] UE-28 159.838 169.528 165.020 164.973 166.753 173.593 171.806 180.000 112.6 3.7 UE-13 13.395 14.680 18.570 15.562 16.182 17.397 17.976 19.449 145.2 12.3 Czechia 2.257 2.375 2.347 2.343 2.432 2.665 2.713 3.011 133.5 10.2 Hungary 1.231 1.332 1.289 1.304 1.375 1.480 1.586 1.653 134.2 10.7 Poland 6.105 6.918 7.074 7.610 7.981 8.519 8.564 9.498 155.6 13.9 Slovakia 1.195 1.210 1.312 1.287 1.184 1.268 1.265 1.267 106.0 3.7 Structure of production value [%] 2017–2010 V* UE-28 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 UE-13 8.4 8.7 9.0 9.4 9.7 10.0 10.5 10.8 2.4 9.0 Czechia 1.4 1.4 1.4 1.4 1.5 1.5 1.6 1.7 0.3 6.6 Hungary 0.8 0.8 0.8 0.8 0.8 0.9 0.9 0.9 0.1 7.3 Poland 3.8 4.1 4.3 4.6 4.8 4.9 5.0 5.3 1.5 10.8 Slovakia 0.8 0.7 0.8 0.8 0.7 0.7 0.7 0.7 -0.1 4.5 Number of persons employed [thousand] 2017–2010 V* UE-28 649.1 655.9 649.5 639.4 641.1 643.5 649.1 660.0 101.7 1.1 UE-13 129.9 129.1 128.7 130.0 132.3 138.9 142.0 146.4 112.7 5.1 Czechia 19.5 19.2 19.4 19.3 19.0 20.1 20.6 21.2 108.3 3.9 Hungary 11.3 11.0 10.8 11.3 11.5 14.2 14.9 14.9 131.9 14.7 Poland 53.8 54.8 54.1 55.0 56.9 57.8 58.1 60.9 113.0 4.3 Slovakia 7.7 7.2 6.9 6.7 6.8 6.9 7.0 7.3 94.1 4.6 Structure of persons employed [%] 2017–2010 V* UE-28 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 UE-13 20.0 19.7 19.8 20.3 20.6 21.6 21.9 22.2 2.2 4.8 Czechia 3.0 2.9 3.0 3.0 3.0 3.1 3.2 3.2 0.2 3.5 Hungary 1.7 1.7 1.7 1.8 1.8 2.2 2.3 2.3 0.5 14.3 Poland 8.3 8.4 8.3 8.6 8.9 9.0 8.9 9.2 0.9 4.0 Slovakia 1.2 1.1 1.1 1.0 1.1 1.1 1.1 1.1 -0.1 4.9 *V – Coefficient of variation Source: own elaboration based on Eurostat.

Between 2010 and 2017, the sold production of the pulp and paper industry in the EU countries increased by 12.6% and at the end of the analysed period amounted to 180.0 billion euro. Such a slight increase within a seven-year period was caused mainly by a decrease in the dynamics of the discussed phenomenon in the years 2012-2013. In 2017, the share of the EU13 countries in the production of pulp and paper industry was 10.8%, i.e. 2.4 p.p. higher than at the beginning of the analysed period. This resulted from a higher growth rate of the value of sold production of the pulp and paper industry in this group of countries than in the EU15 (10.8 compared to 9.6%). It should be noted, however, that it is still the member states that joined the EU before 2004 that played a greater role in creating the sold production of

121 paper and paper products, although the difference between the distinguished groups of countries decreased. The highest increase in the production of the pulp and paper industry in terms of value was observed in the case of Poland – by 55.6% to the level of 9.5 billion euro. Favourable economic tendencies in this country caused an increase in its importance in creating the value of sold production of paper and paper products in the EU (from 3.8 to 5.4%). The Czech Republic was ranked next in terms of the value of production of paper and paper products among the Visegrad Group countries (3.0 billion euro), which recorded an increase in this indicator (33.5%). In the case of other Visegrad Group countries of lesser importance for the creation of EU pulp and paper industry production, the situation was diversified. A significant increase (by 34.2%) in the value of sold production of paper and paper products was observed in Hungary, which at the end of the analysed period amounted to 1.7 billion euro. In Slovakia, on the other hand, there was a significantly smaller increase (by 6.0%), which contributed to a decrease in the importance of this country in creating the EU value of sold production of the pulp and industry. The highest level of the coefficient of variation in the value of sold production of this industry, which shows the relative diversity of this feature, was observed for the Polish pulp and paper industry (13.9%), and the lowest for Slovakia (3.7%). In 2010, as shown in Table 1, the pulp and paper industry in the EU28 employed 649.1 thousand persons, of which 129.9 thousand in the group of new member states, which constituted 20.0% of the total employment in this industry in the European Union. Again, among the Visegrad Group countries, the first place was taken by Poland, where 53.8 thousand persons found employment in the production of paper and paper products, which constituted 8.3% of the employed in the EU28 and as much as 41.4% of the employed in the EU 13. The level of employment in the Czech Republic was three times lower than in Poland and in 2010 amounted to 19.5 thousand persons, which constituted 3.0% of the employed in the pulp and paper industry in all the EU member states. Significantly lower employment was observed in Hungary (11.3 thousand persons) and Slovakia (7.7 thousand persons). The share of these countries in the creation of jobs in the EU pulp and paper industry was 1.7 and 1.2%, respectively. Between 2010 and 2017 employment in pulp and paper industry in the EU increased by slightly1.7% and at the end of the analysed period was at the level of 660.0 thousand persons, 146.4 thousand of whom were employed in companies located in the EU13. In this respect the significance of the new member states increased by 0.2 p.p. to the level of 22.2%. Among the analysed countries of the Visegrad Group, with the exception of Slovakia, an increase in the number of employees was observed. This led to the slightly increasing role of Poland in creating jobs in the pulp and paper industry. In other countries covered by the analysis the employment level decreased by 15.0% in Slovakia, 12.1% in the Czech Republic and 11.1% in Hungary. The highest rate of the variation coefficient of the number of employees, which determines the relative diversity of the feature, was observed for the Hungarian pulp and paper industry (14.7%), and the lowest one for the Czech Republic (3.5%). The highest growth rate of employment was observed in Hungary (by 31.9%), followed by Poland (13.9%) and the Czech Republic (8.3%), however, the importance of these countries in creating jobs in the pulp and paper industry did not change significantly. One of the basic measures of work efficiency is the value of production per employee. As can be seen form the data presented in Figure 1 at the beginning of the analysed period, it amounted to 246 thousand euro on average in the EU member states, and at the same time for the EU13 countries it was more than twice as low. This proves that the labour productivity in the pulp and paper industry in the EU13 countries is lower than the average in the EU member states. It should be emphasized, however, that in all analysed countries this indicator was at a higher level than the average in the group of EU13.

122

Chart 1. Value of sold production of pulp and paper industry per employee in selected EU countries in 2010– 2017

Among the countries that are members of the Visegrad Group, the highest value of sold production per employee in pulp and paper industry was reached by Slovakia (154.7 thousand euro). It resulted from the lowest value of sold production in pulp and paper industry accompanied with the lowest employment rate. These conditions resulted in the best labour productivity rate among the analysed countries. Lower values of this indicator were recorded in the Czech Republic (115.6 thousand euro) and Poland (39.0 thousand euro). Between 2010 and 2017, the value of sold production of pulp and paper industry per employee in the EU 28 increased by 10.6% to 272.2 thousand euro. In the new member states, this indicator increased by 28.9% and at the end of the analysed period was at the level of 132.9 thousand euro, but it was still significantly lower than the average in the EU28. Among the countries covered by the analysis, the highest labour productivity was recorded Slovakia. The abovementioned rate increased by 12.7% to 174.4 thousand euro. Poland (64.7 thousand euro) ranked next in terms of labour productivity of the pulp and paper industry measured by the value of production per employee. Significantly lower rates were observed in Hungary (110.7 thousand euro), where the growth rate of the value of sold production in pulp and paper industry per employee was clearly lower than in other countries of the Visegrad Group. The next stage of the research was the study of employment structure by determining employees’ belonging to the appropriate age group. Data regarding the classification of men employed in the pulp and paper industry companies to particular age groups was presented in Table 2. The share of men employed in pulp and paper mills in the EU28 in all age groups remained at a similar level throughout the entire analysed period. In the age group of 15–39 years old, in the years 2010–2017 men employed in pulp and paper factories constituted from 71.7% to 76.5% of the total number of employees, so the value of coefficient of variation for this age group was not very high at 3.0%. The age group of 40–59 years old had even more stable levels of male participation in the total number of employees. In these age limits, the share of male employees ranging between 71.2–75.2% was observed. A slight decrease of 0.5 p.p. in the share of employed men in the discussed age group in 2017 compared to 2010 was

123 observed. The highest variability in the share of men in the total number of employees was observed in the group of employees over 60 years of age. The coefficient of variation reached the value of 7.2% in the discussed group.

Table 2. Employment of men in the pulp and paper industry by age groups in selected EU countries in 2010– 2017 2017– Itemisation Age group 2010 2011 2012 2013 2014 2015 2016 2017 V* 2010 15–39 71.7 73.2 70.5 73.4 75.7 72.8 76.2 76.5 4.9 3.0

UE-28 40–59 71.7 72.7 73.1 73.4 75.0 72.6 75.2 71.2 -0.5 1.9 60 or over 74.7 78.5 77.4 74.9 76.7 75.8 61.7 77.5 2.8 7.2 15–39 64.9 66.3 64.7 71.7 68.8 68.9 71.3 69.5 4.6 4.0

UE-13 40–59 49.8 56.6 56.5 57.7 57.6 58.9 63.5 55.7 5.9 6.6 60 or over 62.2 72.5 87.8 69.4 70.3 74.0 68.6 75.7 13.5 10.2 15–39 52.9 67.6 54.2 59.6 60.0 62.2 58.4 66.3 13.4 8.6

Czechia 40–59 52.5 52.8 50.6 58.9 48.5 46.0 45.9 46.2 -6.3 9.0 60 or over 71.4 85.7 88.9 64.3 100.0 85.7 100.0 75.0 3.6 15.4 15–39 60.0 58.2 70.0 71.7 61.1 59.6 68.0 66.2 6.2 8.2

Hungary 40–59 12.2 50.5 37.0 49.4 50.5 34.6 23.3 41.4 29.2 37.1 60 or over 33.3 25.0 60.0 81.0 33.3 50.0 33.3 56.3 23.0 40.3 15–39 81.0 69.8 70.9 72.4 75.4 72.6 79.2 70.6 -10.4 5.6

Poland 40–59 52.8 61.3 65.5 63.8 60.6 69.7 76.3 63.1 10.3 10.7 60 or over 64.3 83.3 99.5 68.8 77.8 77.3 80.0 72.7 8.4 13.7 15–39 65.8 74.3 73.0 70.7 76.3 83.7 78.7 77.7 11.9 7.2

Slovakia 40–59 73.0 73.7 19.7 53.1 69.5 55.0 63.6 63.3 -9.7 29.8 60 or over 100.0 100.0 99.9 100.0 100.0 100.0 100.0 99.8 -0.2 0.1 *V – Coefficient of variation Source: own elaboration based on Eurostat.

It was interesting to verify whether the share of employed men in EU13 corresponded to the levels observed for all EU countries. Discussed phenomenon data for the EU13 were presented in the subsequent rows of Table 2. In the group of youngest employees, the share of men ranged from 64.7–71.7% and was characterized by the smallest degree of differentiation among all employees age groups. The highest variability was observed, similarly to the EU28, in the group of persons over 60 years of age – the observed value was even higher and reached 10.2%. The share of men classified to the middle age group in the EU13 reached from 49.8% in 2010 to 63.5% in 2016. An analysis of the share of employed men in the total employment in the Visegrad Group countries shows, that the highest values differences were noted in the group of the oldest employees. The highest observed share of men employed could be found in Slovakia – in this country, in most of the analysed years, men constituted for almost 100% of employed. This means that in this age group employed women constituted just a symbolic percentage. Even small fluctuations in the share of women in all employed, especially when the number of all employed persons is relatively small, may be responsible for the observed high variability of the discussed indicator in the age group over 60 years. In the Czechia, Poland and Hungary, analysed oldest age group of employees also noted a high share of men in all persons employed in the pulp and paper industry. The analysis of the share of classified in the 40–59 age group men showed, that in all the Visegrad Group countries, values of analysed indicator were on average lower than the values

124 recorded in the EU-28 countries. The highest average values of the share of men in this age group were recorded in Poland (values higher than average in EU13), in the remaining countries the analysed indicator was lower, which indicated a higher share of working women in this age group.

CONCLUSIONS The research allowed to define the following conclusions: 1. Among analysed countries of the Visegrad Group, Poland played a leading role in production value and jobs creating in the pulp and paper industry. The value of Polish pulp and paper companies production accounted for almost half of EU13 production. Of all employed in the pulp and paper industry in EU13 countries, almost 42% were employed in Polish companies. Employment in other EU13 countries reached levels from around 5% of the EU-13 level in Slovakia, to around 10% of employed in Hungary and almost 15% of employed in the Czechia. 2. The value of industrial production per person employed in the pulp and paper industry in the EU13 countries accounted for less than 11% of the EU28 countries industrial production, which indicated a significantly lower degree of labour resources usage. The highest productivity measured by the value of sold production per employee was recorded in Poland, labour productivity in Hungary and Slovakia reached similar level. 3. The high share of men in particular age groups employed could have been the result of the specificity of the pulp and paper industry. The Czechia, Hungary and Poland were characterized by the largest variability in the percentage of employed men in total employed in the age group over 60 years old. 4. The 40–59-year-old age group working in pulp and paper industry in the Visegrad Group countries was characterized by a lower share of employed men in relation to the total employed than average in EU28. 5. The share of men in each of the analysed age groups in EU28 countries remained at a similar level. EU13 was characterized by on average lowest level of men share in all age groups employed than in EU28. 6. On average, the highest share of employed women in relation to the total employed was observed in the 15–39 and 40–59 age groups in the Czechia, Hungary and Slovakia. In Poland the average share of employed women in all employed in all age groups was within 22–36%.

REFERENCES

1. BAJPAI P., 2014: Recycling and Deinking of Recovered Paper, Recycling and Deinking of Recovered Paper, Elsevier Insight; pp. 1–19. 2. BRUNNHOFERA M., GABRIELLA N., SCHÖGGL J., STERNC T., POSCH A., The biorefinery transition in the European pulp and paper industry – A three-phase Delphi study including a SWOT-AHP analysis, Forest Policy and Economics 110 (2020) https://www.sciencedirect.com/science/article/pii/S1389934118303691?via%3Dihub 3. EUROSTAT 2018: AGRICULTURE, FORESTRY AND FISHERY STATISTICS, https://ec.europa.eu/eurostat/documents/3217494/9455154/KS-FK-18-001-EN- N.pdf/a9ddd7db-c40c-48c9-8ed5-a8a90f4faa3f (20.10.2019) 4. EUROSTAT 2018: Annual enterprise statistics for special aggregates of activities (NACE Rev. 2)[sbs_na_sca_r2]; https://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=sbs_na_sca_r2&lang=en

125

5. GOŁAŚ Z., KOZERA M., 2008: Strategie wydajności pracy w gospodarstwach rolnych, Journal of Agribusiness and Rural Development, Vol. 1 (7), pp. 73–87. 6. GRZEGORZEWSKA E, STASIAK-BETLEJEWSKA R., 2019: The production potential of the pulp and paper industry – the case of Poland; 12th International Scientific Conference WoodEMA “Digitalisation and Circular Economy: forestry and forestry based industry implications”; Varna, Bulgaria September, 11th–13th, 2019; pp. 257–262. 7. IMF 2018: https://www.imf.org/external/pubs/ft/weo/2018/01/weodata/download.aspx (9.10.2019). 8. MARTIN J., HAGGITH M., 2018; The state of the global paper industry, Environmental Paper Network; pp. 1–89. 9. RIZK N., MARTEL A., D’AMOURS S., 2004: Production Planning in the Pulp and Paper Industry, Research Consortium on e-Business in the Forest Products Industry, Kanada; pp. 1–17. 10. ROTH S., ZETTERBERG L., ACWORTH W., KANGAS H.L., NEUHOFF K., ZIPPERER V., 2016: The pulp and paper overview paper. Sector analysis for the Climate Strategies Project on Inclusion of Consumption in Carbon Pricing, Climate Strategies; pp. 2–43. 11. STATISTICAL YEARBOOK OF FORESTRY, 2018: Zakład Wydawnictw Statystycznych, Central Statistical Office, 2018; https://stat.gov.pl/files/gfx/portalinformacyjny/en/defaultaktualnosci/3328/12/1/1/stati stical_yearbook_of_forestry_2018.pdf (5.10.2019) 12. VAEZ E., ZILOUEI H., 2020: Towards the development of biofuel production from paper mill effluent, Renewable Energy 146 (2020); pp. 1408–1415; https://reader.elsevier.com/reader/sd/pii/S0960148119310791?token=1739F8E03271AB9F67 6B4091685A585165285D02343D88727EBE2EDBFD87924ED1E185B8EB45C4890586056 5DDEAB113

Streszczenie: Tendencje zmian w zatrudnieniu i wydajność pracy w przemyśle celulozowo- papierniczym w wybranych krajach UE. Przemysł celulozowo – papierniczy jest, pomimo wielu wyzwań, dynamicznie rozwijającą się gałęzią przemysłu na świecie. Wydaje się, iż zwłaszcza rosnąca konkurencja ze strony krajów słabiej rozwiniętych, takich jak Indonezja oraz spadek popytu na papier mogą być przyczyną trudności zakładów celulozowo- papierniczych, tj. spadku rentowności produkcji papieru oraz zmniejszenia się w ostatnich latach liczby zatrudnionych w tej gałęzi przemysłu. Z punktu widzenia wielkości zatrudnienia i związanej z nią wydajności pracy interesującym jest zbadanie zatrudnienia i ocena wykorzystania zasobów ludzkich w zakładach przemysłu celulozowo-papierniczego, szczególnie w krajach należących do Grupy Wyszehradzkiej, których gospodarki w ostatnich latach charakteryzuje stabilny wzrost PKB. W pracy przeanalizowane zostały: wartość produkcji przemysłu celulozowo-papierniczego oraz jej struktura w krajach Grupy Wyszehradzkiej w odniesieniu do krajów należących do UE28, a następnie zatrudnienie i jego struktura związana z przynależnością do danej grupy wiekowej.

126

Corresponding author:

Justyna Biernacka Institute of Wood Sciences and Furniture Warsaw University of Life Sciences - SGGW 159 Nowoursynowska str.; 02-776 Warsaw, Poland [email protected]

ORCID ID: Grzegorzewska Emilia 0000-0002-7532-9287 Biernacka Justyna 0000-0003-3407-1280 Podobas Izabela 0000-0001-8315-0386

127

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 128-134 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

The influence of temperature on selected strength properties of furniture particleboard

PIOTR BORYSIUK, ANNA TETELEWSKA, RADOSŁAW AURIGA, IZABELLA JENCZYK-TOŁŁOCZKO Institute of Wood Sciences and Furniture, Warsaw University of Life Sciences – SGGW, 159 Nowoursynowska St., 02-787 Warsaw

Abstract: The influence of temperature on selected strength properties of furniture particleboard. As a part of the study, the influence of temperature on selected properties of furniture particleboard was tested. P2 type industrial particleboards in three finishing options: raw boards (1), boards covered with melamine film in white (2) and black (3) (10 samples per variant) have been subjected to temperatures from -20oC to +120oC, at 10oC intervals. The Time of exposure for individual temperatures was 7 days. MOR, MOE and IB were determined for tested boards. It has been shown that temperatures above 50oC have a negative effect on strength properties of boards. A large decrease in all tested parameters was observed in the temperature range from +60oC to +120oC. It was also noted that finishing boards with melamine film did not improve their durability.

Keywords: furniture particleboard, temperature, strength properties

INTRODUCTION The conditions of use show a significant impact on the properties of wood-based panels that are the basic component of many products, e.g. furniture. Poland is the third largest exporter of furniture in the world, after China and Germany. In 2018, Poland exported furniture worth PLN 45.6 billion (www.biznesmeblowy.pl). In terms of categories, most of furniture produced in Poland is used in table rooms and living rooms, bedroom furniture and kitchen furniture (https://businessinsider.com.pl). One of the basic raw materials used in production of furniture is three-layer particleboard (type P2 according to EN 312:2011) covered with melamine film. In 2018, a total of 4.85 million m3 of particleboard was produced in Poland (http://www.fao.org). During use, furniture particleboard may be exposed to changing environmental conditions: relative air humidity and temperature. Kozakiewicz and Matejak (2013) report, among others, that in facilities, especially in summer, there is a high probability of subjecting equipment elements to temperatures above 40oC. The first research on the effect of temperature on lignocellulosic materials was mainly focused on solid wood. At the beginning of 20th century, Baumann (1922) and Vorreiter (1938) showed that frozen wet wood has greater strength when compared to wood with a positive temperature. Kollmann (1940 and 1942) stated that irrespective of humidity, wood at 20°C has about 30 to 50% lower compressive strength along fibres when compared to wood at -42°C. At the same time, he showed that with temperature increasing in the range from -200°C to +200°C, wood’s resistance to forced shape changes decreases and its compressive strength along fibres is also reduced. Similar results were obtained by Kozakiewicz (2010) examining, among others, effect of temperature in the range from -40°C to +80°C on compressive strength along fibres of selected types of wood of various density and anatomical structure. He also showed that with changes in temperature typical for closed spaces (temperature changes in domestic utility rooms), i.e. between +7°C and +38°C, changes in strength can be considered insignificant. An increase in temperature also negatively affects the properties of wood materials (Bekhta et al. 2003). Authors stated, among others, that temperature changes from +20°C to +140°C (at 20°C intervals) with 1.0 hour of exposure

128 time to a given temperature affects the reduction of MOR by a maximum of 40% for particleboard, 37% for MDF and 30% for OSB. The observed changes were linear. In turn, Sonderegger and Niemz (2006), subjecting wood materials to 1.5 hours exposure to temperature in the range from -20oC to +60oC (at 20°C intervals) showed that the smallest decrease in strength parameters (MOR and MOE) was recorded for plywood (respectively 12 and 14%) and the largest for blockboards (39 and 46%, respectively). As a result of research, particleboard was characterized by a decrease in MOR of 26% and a MOE in the range of 37 to 43% (for boards with a thickness of 16 and 18 mm). Ayrilmis et al. (2010) examining the mechanical properties of plywood, OSB and MDF exposed for 48 hours to temperature in the range of -30°C to +30°C (at 10°C intervals), found a decrease in their strength, with the largest changes observed in the temperature range from -10°C to +10°C. Borysiuk et al. (2018) examining MFP boards found that an increase in temperature in the range of -20°C to +120°C causes a decrease in MOR by an average of 30% and a decrease in MOE by an average of 26%. The aim of the work was to determine the influence of temperature on strength properties of industrial furniture particleboard (type P2). The scope of work included exposure of boards to temperatures in the range of -20oC to +120oC (at 10oC intervals) during 7 days. The particleboard used in this study was in three variants of finish: raw boards (1), boards covered with melamine film in white (2) and in black (3).

MATERIALS AND METHODS The study was conducted with the use of industrial furniture particleboard (type P2) in three different finishes: raw boards – PB, boards covered with melamine film in white – PB_w and black – PB_b. Boards covered with melamine film differed only in colour. They were characterized by a nominal thickness of 18 mm and an average density: 674 kg/m3 for raw panels and 720 kg/m3 for panels covered with melamine film. Samples of boards with dimensions of 50 mm by 300 mm (10 for each variant) were exposed to temperatures between -20oC and +120oC at 10oC intervals (15 temperature variants in total). Samples were exposed to the specified temperature for 7 days. Each time the process was carried out at normal pressure without controlling relative air humidity. Immediately after exposure to temperatures samples were tested: - MOR and MOE – in accordance with standard EN 310:1994. The length of the specimens for test was 300 mm – not 360 mm plus 50 mm, as it should be done according to standard. The span between the supports during tests was 280 mm; - IB – in accordance with the standard EN 319:1999. To compare the obtained results of strength properties, one-factor analysis of variance (ANOVA) was carried out using Statistica 13.1 software.

RESULTS AND DISCUSSION Results of the research are presented in Fig. 1, 2 and 3. Generally it can be stated, that increase of temperature has the effect on decrease in strength properties of particleboards type P2. Analogous dependency was noted by many authors (among others Bekhta et al. 2003, Sonderegger and Niemz 2006, Ayrilmis et al. 2010, Kulman et al. 2015, Borysiuk et al. 2018) in relation to other wood materials, which was also presented in the introduction. The dependencies obtained as a part of the study are linear (R2 in the range from 0.5037 to 0.9392). It confirms data presented by Bekhta et al. (2003). Both in case of MOR and MOE, the largest decrease in value (for temperatures in the range from -20oC to +120oC) was recorded for boards finished with melamine film (Fig. 1 and Fig. 2). It was from 44% to 50% for MOR and from 36% to 49% for MOE for PB_w and PB_b boards. For raw particleboard (PB), the decrease in MOR and MOE in the full range of tests was 31% and 27%,

129 respectively. These values generally are comparable to data given in the literature for particleboard (Bekhta et al. 2003, Sonderegger and Niemz 2006, Borysiuk et al. 2018). It should be noted, however, that literature data usually refers to other types of particleboard exposed to temperatures in narrower ranges of temperature variations and significantly shorter exposure times. In the temperature range most commonly found in utility rooms, i.e. +10oC ÷ +30oC, slight fluctuations in strength values were observed for both raw (PB) and melamine film finished boards (PB_w and PB_b). PB_w and PB_b boards in the same temperature range were generally characterized by greater variability of strength parameters in relation to PB boards. At -20oC ÷ +50oC, the differences between selected MOR values of the tested boards are statistically significant – it can be specified in this range: 4 homogeneous groups of MOR values for PB boards, 3 homogeneous groups of MOR values for PB_w boards and 2 groups of homogeneous values MOR for PB_b boards (table 1). Similarly, in case of MOE, in temperature range -20oC ÷ +50oC, statistically significant differences between selected values were noted – it can be specified in this range: 5 groups of homogeneous MOE values for PB boards and 2 groups of homogeneous MOE values for boards PB_w and PB_b (table 1). It should be noted that in this temperature range all boards (type P2) meet the requirements of EN 312:2010 regarding MOR (above 11 N/mm2). Regarding MOE, tested boards met the requirements of above standard (above 1600 N/ mm2) in the entire tested temperature range (-20oC ÷ +120oC). A clear decrease in MOR and MOE values for tested boards was recorded at temperatures above +50oC. It is worth adding that the differences in MOR and MOE values of tested boards in the temperature range +60oC ÷ +120oC are generally not statistically significant (table 1). Kulman et al. (2015) while examining the influence of temperature in the range of +20oC ÷ +80oC on MDF properties showed significantly greater decreases in MOR and MOE values than in case of tested particleboards in the same temperature range.

22 20 18 16

] 14 2 12 10

MOR [N/mm MOR 8 6 PB PB w PB b 4 Liniowy (PB) Liniowy (PB w) Liniowy (PB b) R² = 0,8987 R² = 0,8588 R² = 0,8846 2 0 -20 -10 0 10 20 30 40 50 60 70 80 90 100 110 120 Temperature [oC]

Figure 1. Dependence of particleboards MOR on the exposure temperature

130

4500

4000

3500

3000

] 2 2500

2000 MOE [N/mm MOE 1500 PB PB w PB b 1000 Liniowy (PB) Liniowy (PB w) Liniowy (PB b) R² = 0,6513 R² = 0,9392 R² = 0,8939 500

0 -20 -10 0 10 20 30 40 50 60 70 80 90 100 110 120 Temperature [oC]

Figure 2. Dependence of particleboards MOE on the exposure temperature

1,0

0,9

0,8

0,7 ] 2 0,6

0,5 IB [N/mm IB 0,4

0,3 PB PB w PB b 0,2 Liniowy (PB) Liniowy (PB w) Liniowy (PB b) R² = 0,5037 R² = 0,6159 R² = 0,5842 0,1

0,0 -20 -10 0 10 20 30 40 50 60 70 80 90 100 110 120 Temperature [oC]

Figure 3. Dependence of particleboards IB on the exposure temperature

As in the case of MOR and MOE, also IB values of boards decrease with increasing temperature (Fig. 3). In tested temperature range from -20°C to +120°C, the decrease in IB value for boards finished with melamine film was PB_w - 22% and PB_b - 30%, respectively (Table 1). For raw particleboard (PB), a decrease of IB in the full range of research was 9%. In case of IB (similar to MOR and MOE), a decrease in the value for temperatures above +50oC was noted, but it was not so clear (greater variation in homogeneous groups than for

131

MOR and MOE) (Table 1). It should be added that in the whole temperature range, boards meet the requirements of EN 312:2010 regarding IB (above 0.35 N/mm2).

Table 1. The relative values of the studied strength properties (MOR, MOE, IB) in relation to the boards obtained at 20°C with regard to one-factor analysis of variance of examined strength characteristics in reference to temperature exposure (a, b, c, d, e, A, B, C, D, E, 1, 2, 3, 4, 5 – homogeneous groups). Temperature MOR [%] MOE [%] IB [%] [°C] PB PB_w PB_b PB PB_w PB_b PB PB_w PB_b -20 111 e 91 BC 110 3 121 e 114 D 114 4 98 bcde 90 BCDE 106 45 -10 108 de 101 C 110 3 114 de 113 D 120 4 103 cde 80 ABCD 104 345 0 107 cde 94 BC 115 3 105 bcd 110 D 113 4 108 e 92 CDE 105 345 10 108 de 93 BC 115 3 112 cde 111 D 116 4 104 cde 94 DE 106 45 20 100 bcd 100 C 100 23 100 abcd 100 C 100 3 100 bcde 100 E 100 345 30 101 bcd 85 BC 103 23 100 abcd 94 C 100 3 106 de 90 BCDE 98 2345 40 97 bc 79 B 99 23 89 a 96 C 100 3 99 bcde 94 DE 102 345 50 96 b 80 B 90 2 98 abc 93 C 99 3 108 de 83 ABCDE 117 5 60 82 a 53 A 68 1 95 ab 81 B 89 2 84 abc 78 ABCD 87 1234 70 83 a 53 A 65 1 90 a 75 AB 82 12 87 abc 71 A 86 123 80 80 a 48 A 69 1 85 a 69 A 83 12 89 abcd 74 ABC 80 12 90 77 a 47 A 63 1 88 a 69 A 82 12 76 a 73 AB 90 1234 100 81 a 45 A 67 1 92 ab 66 A 83 12 95 bcde 76 ABC 89 1234 110 80 a 49 A 62 1 95 ab 72 AB 84 12 84 ab 70 A 90 1234 120 80 a 47 A 60 1 94 ab 65 A 78 1 89 abcd 68 A 76 1

The decrease in strength of boards in the studied temperature range results from impact on both wood particles and glue joints. During heating of boards, elementary wood particles forming particleboard shrink, as a result of which microcracks in the material that may weaken its internal structure may arise. Green et al. (1999) report that permanent changes in the strength of wood caused the impact of higher temperature are a consequence of hydrolysis of acetyl and formyl groups of hemicelluloses, as a result of which acetic and formic acids are formed. An analogous process of degradation of wood material and glue lines also occurs as a result of the impact of hardener residues present in the joints. With regard to gelled glue, it should be noted that temperature itself up to 200oC does not affect degradation of UF and MF glue joints (Hirata et al. 1999).

CONCLUSION As a result of research of both raw and finished with melamine film furniture particleboards exposed to temperature in the range from -20oC to +120oC, the following conclusions were formed: 1. The increase in temperature causes a decrease in the mechanical properties of P2 type furniture particleboard. 2. Particleboards covered with melamine film are characterized by a greater range of decrease in strength properties as a result of an increase in the temperature of use compared to unfinished boards. 3. The increase in the temperature of use of furniture particleboard finished with melamine film in the range from -20°C to +120°C causes a decrease in MOR and MOE by over 40% and IB by over 20%.

132

4. P2 type furniture particleboards meet the requirements of EN 312:2010 for MOR in the temperature range from -20°C to 50°C, and for MOE and IB in the full temperature range from -20°C to 120°C.

REFERENCES

1. AYRILMIS N., BUYUKSARI U., AS N., 2010: Bending strength and modulus of elasticity of wood-based panels at cold and moderate temperatures. Cold Regions Science and Technology 63, pp. 40–43. 2. BAUMANN R., 1922: Die bisherigen Ergebnisse der Holzprüfungen in der Materialprüfungsanstalt an der Technischen Hochschule Stuttgart. Heft 231 der Forschungsarbeiten auf dem Gebiete des Ingenieurwesens. Berlin. 3. BEKHTA P., ŁĘCKA J., MORZE Z., 2003: Short-term effect of the temperature on the bending strength of wood-based panels. Holz als Roh- und Werkstof 61, pp. 423– 424. 4. BORYSIUK P., KOZAKIEWICZ P., NURCZYK T., 2018: Wpływ temperatury na wybrane właściwości płyt wiórowych. Biuletyn Informacyjny Ośrodka Badawczo- Rozwojowego Przemysłu Płyt Drewnopochodnych w Czarnej Wodzie, 1/2, pp. 6–13 5. EN 310:1994 Wood-based panels. Determination of modulus of elasticity in bending and of bending strength. 6. EN 312:2010 Particleboards. Specifications. 7. EN 319:1999 Particleboards and fibreboards. Determination of tensile strength perpendicular to the plane of the board 8. GREEN D.W., WINANDY J.E., KRETSCHMANN D.E., 1999: Mechanical Properties of Wood – Chapter 4 in: Wood handbook – wood as an engineering material. Forest Products Laboratory USDA Forest Service. Madison, Wisconsin USA. 9. HIRATA T., KAWAMOTO S., OKURO A., 1991: Pyrolysis of Melamine – Formaldehyde and Urea – Formaldehyde Resins. J. Appl. Polym. Sci., 42. 10. http://www.fao.org/faostat/en/#data/FO (electronic document, as of November 10, 2019) 11. https://businessinsider.com.pl/wiadomosci/produkcja-mebli-w-polsce-w-i-kwartale- 2019-roku/5z8480v (electronic document, as of November 05, 2019) 12. KOLLMANN F., 1940: Die mechanischen Eigenschaften verschieden feuchter Hölzer im Temperaturbereich von –200 bis + 200 oC. Forschung auf dem Gebiete des Ingenieurwesens 403 (11), pp. 1–18. VDI-Verlag. Berlin. 13. KOLLMANN F., 1942: Über das Gefrieren und den Einfluß tiefer Temperaturen auf die Festigkeit der Hölzer. Mitteilungen der Hermann-Göring-Akademie der Deutschen Forstwissenschaft. 2. Jahrgang, Band 1. J.D. Sauerländer Verlag. Fankfurt am Mein. 14. KOZAKIEWICZ P., 2010: Wpływ temperatury i wilgotności na wytrzymałość na ściskanie wzdłuż włókien wybranych rodzajów drewna o zróżnicowanej gęstości i budowie anatomicznej. Trzysta siedemdziesiąta pozycja serii - Rozprawy Naukowe i Monografie Wydawnictwo SGGW, Warszawa. 15. KOZAKIEWICZ P., MATEJAK M., 2013: Klimat a drewno zabytkowe - dawna i współczesna wiedza o drewnie. Wydanie IV – zmienione. Wydawnictwo SGGW. Warszawa. 16. 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 Forestry and Wood Technology, 91: pp. 101–106

133

17. SONDEREGGER W., NIEMZ P., 2006: Der Einfluss der Temperatur auf die Biegefestigkeit und den Elastizitätsmodul bei verscheidenen Holzwerkstoffen. Holz als Roh- und Werkstoff 64, pp. 385–591. 18. VORREITER L., 1938: Biege- und Druckfestigeit gefrorenen Fichtenholzes. Tharandter Forestl. Jahrbuch Bd. 89: pp. 491–510. 19. www.biznesmeblowy.pl/rynek_mebli/111/jaki_byl_rok_2018_w_branzy_meblarskiej _zobacz_najnowszy_raport,16390.html (electronic document, as of November 5, 2019)

Streszczenie: Wpływ temperatury na wybrane właściwości wytrzymałościowe meblowych płyt wiórowych. W ramach pracy zbadano wpływ temperatury na wybrane właściwości meblowych płyt wiórowych. Przemysłowe płyty wiórowe typu P2 w trzech wariantach wykończenia: płyty surowe (1), płyty pokryte filmem melaminowym w kolorze białym (2) i w kolorze czarnym (3) (po 10 próbek na wariant) zostały poddane działaniu temperatur od -20oC do +120oC, z od stopniowaniem, co 10 K. Czas oddziaływania poszczególnych temperatur wynosił 7 dni. Dla badanych płyt oznaczono MOR, MOE. IB, odporności laminatu na pękanie. Wykazano, że temperatura powyżej +50oC wywiera negatywny wpływ na właściwości wytrzymałościowe płyt. Zaobserwowano duży spadek wszystkich badanych parametrów w przedziale temperatur od +60oC do +120oC. Stwierdzono również, że wykończenie płyt filmem melaminowym nie wpływa na poprawę ich wytrzymałości.

Corresponding author: Piotr Borysiuk, PhD., D.Sc. Nowoursynowska Str. 159 02-787 Warszawa, Poland email: [email protected] phone: +48 22 59 38 547

ORCID ID: Borysiuk Piotr 0000-0002-7508-9359 Auriga Radosław 0000-0001-5627-2425 Jenczyk-Tołłoczko Izabella 0000-0002-1172-1386

134

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 135-139 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Sinoxylon anale as wood borer insect and its parasites

ADAM KRAJEWSKI1, PIOTR WITOMSKI1, ANNA OLEKSIEWICZ2, 1 Department of Wood Science and Wood Protection, Warsaw University of Life Sciences – SGGW 2 Department of Physics, Warsaw University of Life Sciences – SGGW

Abstract: Sinoxylon anale as wood borer insect and its parasites. A piece of furniture imported to Warsaw from India was used for the investigations. The wood borer insect was defined as Sinoxylon anale Lesne. The wood was hardly damaged by the larvae. There is 4.8 cm3 of wood for one adult who has left wood. In the conditions of high population density of the local population of S. anale only few holes of adults of Chalcidoidea have been found that are parasites. The death of adults in exuvium was probably caused by microorganisms.

Keywords: Sinoxylon anale, wood borer insect, parasites

INTRODUCTION Sinoxylon anale Lesne, 1897 (Coleoptera, Bostrichidae) is very cosmopolitan species recorded in different countries (Baker and Berry 1978, Allen et al. 1997, Dominik and Starzyk 2004, Bajpai 2007, Gumovsky 2010, Lykidis et al. 2016). Their xylophagous larvae live in timber and wood packaging materials, destroying them so much that they crumble during unloading (Allen et al. 1997, Dominik and Starzyk 2004). The presence of Sinoxylon anale was also noted in Poland (Dominik 1970, Śliwa 1971, Krajewski and Mazurek 2010). S. anale can settle in heated house interiors in winter and destroy wooden furnishings (Dominik 1970, Śliwa 1971). The larval tunnels in wood can be very numerous. Wood damage can be very severe. Under these circumstances, there may be parasites and predators that destroy S. anale. A list of S. anale potential parasites was also made (Gumovsky 2010).

Figure 1. The wood sample that was used in the study

The “rose wood” investigations with S. anale tunnels were done. This wood was imported from India to Poland. The number and size of insects that left in the wood, the number of insects that died in the wood, and the presence of predators and parasites were examined. The purpose of the research was to pre-determine the possibility of natural reduction of S. anale in wood that is imported into Poland.

MATERIAL AND METHODS A piece of furniture was used for the investigations, which was imported to Warsaw from India (Figure. 1). The furniture was imported to Poland in winter.

135

The furniture was made of an indeterminate wood species. This wood has been designated under the trade name as “rose wood”. The furniture showed a high degree of damage. The wood borer insect was designated as Sinoxylon anale Lesne (Figure 2).

Figure 2. Adult of Sinoxylon anale Lesne Measurements of the wood sample size were made. The volume of wood was calculated. The number of imagines holes was calculated. Hole diameters were measured. Imagines’ holes are ranked in size classes. Based on the number of holes, the number of beetles per 1 cm3 of wood was calculated. 1/3 part of the wood was split into very small pieces. The number of dead insects that were find in the wood were counted. Their condition was assessed.

RESULTS AND DISCUSSION The condition of the wood interior is shown in Figure 3.

Figure 3. The condition of the wood interior

All the wood was heavily riddled with holes. The number of holes and diameters of holes are shown in the Table 1. Larvae droppings do not have a geometric shape, neither do those of Lyctus sp. After defecation, etched wood sawdust by S. anale tends to collect into small clusters. This phenomenon does not occur in Lyctus sp. Lyctus sp. was included in the Lyctide family (Dominik and Starzyk 2004), which is currently the subfamily Lyctine in the Bostrychidae family (Borowski and Węgrzynowicz 2007). 136

Table 1. Number of imagines holes and diameters of holes The side The total number of The size class of hole The number of holes in holes in side diameters class A 70 1.1 – 1.5 3 1.6 – 2.0 1 2.1 – 2.5 14 2.6 – 3.0 48 3.1 – 3.5 4 B 43 1.1 – 1.5 0 1.6 – 2.0 1 2.1 – 2.5 4 2.6 – 3.0 36 3.1 – 3.5 2 C 31 1.1 – 1.5 1 1.6 – 2.0 6 2.1 – 2.5 6 2.6 – 3.0 16 3.1 – 3.5 2 D 42 1.1 – 1.5 1 1.6 – 2.0 7 2.1 – 2.5 9 2.6 – 3.0 19 3.1 – 3.5 6

The results of hole diameter measurements in classes are presented in Figure 4.

Figure 4. The total number of holes in classes for damaged piece of furniture There is 4.8 cm3 of wood for one adult who has left wood. The amount of wood per individual is actually even smaller because some of the insects were dead before leaving the wood. It is not known how many adults left the wood after being brought to Poland. At least 15 adults left the wood in the laboratory. The adults’ body length is determined to be “about 5 mm” (Dominik and Starzyk 2004). This value was found through the measurements made during the experiment. The diameter of adults was measured to compare with the diameter of the holes. The share of hole sizes in individual classes does not differ from the normal distribution. Imagines’ hole diameters are generally defined as “up to 3 mm” (Dominik and Starzyk 2004). It seems that the holes with a diameter below 1.6 mm are not made by S. anale. Probably the holes with a diameter less than 1.6 mm were made by predators or

137 parasites of S. anale. Several species of Chalcidoidea are known as S. anale parasites. An annotated list of 6 known chalcidoid parasitoids (Hymenoptera: Chalcidoidea) of bostrichid beetles is given (Gumovsky 2010): Family Chalcididae: Tanycoryphus conglobatus Steffan, 1950, Tanycoryphus criniger Steffan, 1950, Family Pteromalidae: Cheiropachus quadrum (Fabricius, 1787), Cerocephala aquila (Girault, 1920), Family Eurytomidae: Endobia donacis Erdos, 1964, Family Eulophidae: Entedon stephanopachi Hedqvist, 1959.

In this case, the species of Chalcidoidea could not be identified because no individual was found in the wood. In Europe, only Entedon stephanopachi Hedqvist (Eulophidae) is expected to be a bostrichid specialist rather than an associate of other xylophagous beetles (Gumovsky 2010). Dead adults in exuvium were spotted in the wood (Figure 5). The death of these adults was probably caused by microorganisms.

Figure 5. Dead adults in pupa exuvium from wood interior

Microorganisms are potential perpetrators of death for larvae and adults at very high population density of Sinoxylon anale in wood.

CONCLUSIONS The winter period did not prevent adults from leaving the wood in a heated room. In the conditions of high density of the local population of S. anale, quite numerous dead young adults in pupa exuvium have been found in the wood. The reason for their death has not been clearly explained. The content of larvae borings is not a feature that makes it possible to identify the species as S. anale (or Sinoxylon in general). There are no droppings in the holes that have a characteristic shape. Only few holes (3–4) of adults of Chalcidoidea have been found that are parasites of S. anale.

138

REFERENCES

1. ALLEN E.A., HUMBLE L.M., DAWSON J.L.M., BELL J.D., 1997: Exotic interceptions from wooden dunnage and packing material. North American Plant Protection Organization Bulletin No. 15, Abstracts of the 21st annual meeting and colloquium on quarantine security. Seattle, WA Oct 20–24, 1997. 2. ARGAMAN Q., 1987: Synoxylon anale – a new destructive wood bored in Israel. Phytoparasitica, 15(3): 257. 3. BAJPAI R., 2007: Damage of Sinoxylon anale Lesne on timber of Albizia lebbeck. IndianJournal of Agroforestry, 9(2): pp.128–129. 4. BAKER J.M., BERRY R.W., 1978: Exotic timber insect species intercepted in the UK since 1945. International Research Group on Wood Preservation, IRG/WP/182: p. 5 5. BOROWSKI J., WĘGRZYNOWICZ P. 2007: World Catalogue of Bostrichidae (Coleoptera), Mantis. 6. DOMINIK J., 1970: Z obserwacji nad niektórymi gatunkami owadów obcego pochodzenia przywożonych do polski wraz z wyrobami z drewna, Sylwan, 1, pp. 35– 39 7. DOMINIK J., STARZYK J.R., 2004: Owady uszkadzające drewno, Warszawa, PWRiL, 2004, p. 550 8. MOVSKY A.V, 2010, A record of Sinoxylon anale Lesne in Ukraine with notes on false powder-post beetles (Coleoptera:Bostrichidae) and their chalcidoid parasitoids (Hymenoptera). Ukrainska Entomofaunistika 1(2), pp. 1–8 9. LYKIDIS Ch.T, NARDI G., PETRAKI P.V., 2016: First record of Sinoxylon anale and S. unidentatum in Greece, with an updated account on their global distribution and host plants (Coleoptera: Bostrichidae), Fragmenta entomologica, 48 (2): pp. 101–121 10. ŚLIWA E., 1971: Sinoxylon anale Lesne – szkodnik zawleczony z Pakistanu do Polski, Sylwan, 9, pp. 51–54

Streszczenie: Sinoxylon anale, jako szkodnik drewna i jego parazytoidy. Do badań wykorzystano mebel, który sprowadzono do Warszawy z Indii. Ksylofagiczny owad, który zniszczył drewno, oznaczony został, jako Sinoxylon anale Lesne. Drewno było bardzo mocno uszkodzone przez larwy. Na jednego dorosłego osobnika, który opuścił drewno, przypada 4,8 cm3 drewna. W warunkach wysokiej gęstości miejscowej populacji S. anale stwierdzono jedynie kilka otworów dorosłych osobników Chalcidoidea, które są pasożytami. Śmierć imago w exuvium była prawdopodobnie spowodowana przez mikroorganizmy.

Corresponding author: Piotr Witomski Department of Wood Science and Wood Preservation Institute of Wood Sciences and Furniture Warsaw University of Life Sciences – SGGW Nowoursynowska Str. 166 02-787 Warsaw, Poland e-mail: [email protected] phone: (+48) 22 59 38 655

ORCID ID: Krajewski Adam 0000-0002-6009-6441 Witomski Piotr 0000-0002-8735-2214

139

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 140-147 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

The study of the impact of in situ polymerisation with styrene or acrylates on water absorbability and swelling of thermomechanically densified poplar wood

MARTA GNACIŃSKA, ANDRZEJ RADOMSKI Department of Wood Science and Wood Protection, Institute of Wood Science and Furniture, Warsaw University Abstract: The study of the impact of in situ polymerisation with styrene or acrylates on water absorbability and swelling of thermomechanically densified poplar wood. Black poplar samples, which were previously subjected to thermomechanical densification, were tested for an improvement on the field of water resistance. Series of samples were additionally thermally treated in a nitrogen atmosphere, and then series of densified only or densified and thermally treated samples were treated with monomer mixtures, containing styrene or acrylates, and co-monomers reactive toward cell wall of wood, followed by thermally induced radical polymerisation. All samples were tested by prolonged soaking in water, while volume swelling and absorbability were determined. Densified wood proved to be suitable for modification by in situ polymerisation. Thermally treated densified wood was found to be significantly more compatible with polymers used, as a decrease in its swelling was observed as dominant effect, while absorbability changes were less clear.

Keywords: in situ polymerisation, styrene, acrylates, wood absorbability, wood swelling

INTRODUCTION Wood is a polymer composite composed mainly of cellulose, hemicelluloses and lignin (Krzysik 1978). They form cell walls, which are responsible for most of the physical and chemical properties exhibited by wood. Wood has been used for centuries as a building and engineering material. It possesses however several properties that negatively affect such applications. These include shrinkage and swelling of the wood, causing cracks and creating conditions for fungal growth (Krzysik 1978). The factor responsible for these phenomena is the interaction of wood material with water. Various types of methods are applied to minimise these undesirable phenomena. It is possible to use chemical compounds that react with the cell wall causing its hydrophobisation (Rowell 2014) or mechanically protect against water access, e.g. by oiling wood or through the application of creosote (Shiraishi 2001). Filling the lumen with a polymer is another method of mechanical protection against moisture (Rowel 2005). Such a filling blocks the fastest way of transporting water within wood tissue. Direct polymerisation in lumen, called in situ polymerisation, is the best method of filling, as the method does not demand following solvent evaporation (Hill 2006). As wood material is hydrophilic while most of the polymers, including vinyl and acryl ones, is hydrophobic, adhesion of the filling to cell wall is poor and some space can occur between them. It is possible to use substances, which on the one hand are able to become embedded into the polymer chain, and on the other hand chemically bind with wood (Li et al. 2013). In this study the method of in situ polymerisation using styrene and a mixture of acrylates was employed. Wood modified in this way has greater dimensional stability and resistance to biotic agents (Li et al. 2013), with some mechanical properties such as bending strength and hardness being improved, as well (Sejati et al. 2017). In some methods, e.g. using furfuryl alcohol treatment wood also significantly changes colour, approaching the appearance of exotic species (Bartkowiak et al. 2015).

140

Polymers are macromolecular compounds, which are characterised by a kind of inhomogeneity, despite the fact that specific repeatability occurs in their structure. Unlike other compounds, they do not consist solely of the same molecules. They often do not have such a precisely defined chemical structure. Some of them can be used as additive in wood materials, especially as sealants, filling lumen in wood tissue. Poly(butyl methacrylate) is a soft and flexible polymer (Ciabach 2001) while coatings are susceptible to dirt. Crosslinking undergoes when exposed to light which causes loss of solubility and is significant disadvantage from the viewpoint of polymer application. Poly(butyl methacrylate) is used in wood conservation as structurally consolidating agent (Lazzari and Chiantore 2000). Solutions of the polymer are used as wood adhesives or binder in wood putty for filling small defects. Styrene is proposed as a cheap polymer precursor (Ermeydan et al. 2014), however its toxicity and carcinogenicity constitute severe problems (Weast and Grasselli 1989). Styrene is, in normal conditions, a colourless or slightly yellowish liquid with a characteristic, faint, unpleasant odour. On the other hand, styrene can be obtained with a good yield as the product of thermal depolymerisation of wastes, like expanded polystyrene packagings. This may be alternative source of styrene, in line with the principles of sustainable development. The same method of depolymerisation can be used to wood-styrene composites disposal. Styrene undergoes polymerisation quickly and easily (Rowel, 2005). Paraloid B72 is a trade name of copolymer of ethyl methacrylate (EMA) and methyl acrylate (MA) (Fig.1) at 70:30 ratio (Ciabach 2001). It is sold in two forms: as a 100% granulated resin and as a 50% solution in toluene. Paraloid B-72 is widely used for structural reinforcement of damaged wood items, with reduced strenght. This method is used, for example, in the conservation of historic wood (Ciabach, 2001).

Figure 1. Paraloid B-72 polymer structure Many compounds capable of reacting with wood are known, some of them, like isocyanates, are used in adhesives and varnishes. Styrene easily undergoes copolymerisation with maleic anhydride. Maleic anhydride, in turn, displays high reactivity to hydroxyl compounds, and can therefore react with alcohols and phenols present in wood (Fig. 2). Another type of reactive monomer is glycidyl methacrylate, capable of copolymerising with other acrylic monomers, at the same time its epoxy ring may be reactive towards wood (Fig. 3) but in this case the reaction often requires elevated temperature or the application of catalytic agents.

Figure 2. Maleic anhydride and its reactivity toward wood

141

Figure 3. Glycidyl methacrylate and its reactivity toward wood

Aspen and poplar wood has relatively low durability or density, hence they are rarely used as a construction material for furniture and floors (Li et al. 2013). On the other hand, poplars, especially the fast-growing ones, achieve considerable annual growth and allow for obtaining large quantities of cheap material. Improving the quality of this material may cause new fields of application to emerge for poplar wood. Wood densification, which can be achieved by thermo-mechanical way is a promising method of its improvement. There are many reports on wood densification, e.g. alder, beech, birch, and pine wood veneer (Bekhta et al. 2017). Thermo-mechanical treatment was found to improve strength, surface properties (e.g. contact angle with water, roughness), and reduce equilibrium moisture content of wood as well (Vasconcelos and Del Menezzi 2013). Improvement of mechanical parameters was confirmed for black poplar wood as well (Ülker and Burdurlu 2016). There is lack of reports connected to the modification of densified wood, especially of poplar, by in situ polymerisation in lumen. Due to changes in surface properties, interaction between polymer and wood cell wall may lead to different behaviour in contact with water. The aim of this study is to determine the improvement in water absorbability and dimensional stability of densified black poplar wood (Populus nigra L.), also additionally thermally treated, after in situ polymerisation with various monomers.

MATERIALS AND METHODS Large samples (bars) were cut from a single board of black poplar wood (Populus nigra L.). The samples, of original dimension of 20×20×300 mm were subjected to densification at heated plate press at Versal sp. z o. o. company, as a part of the BIOSTRATEG2/298241/10/NCBR/2016 research project. Some of the samples were then thermal treated for 6 hours at 190 °C, in an atmosphere of nitrogen. Then the bars, both of densified only wood (DW) and thermally treated wood after densification (TTDW) were cut to 20 mm long samples, having dimensions about 20×20×6 mm, while thickness and width of the samples resulted from densification process, without further machining. 40 samples of densified wood (DW) and 40 samples of thermally treated densified wood (TTDW) were selected for the study. The samples were numbered, measured with accuracy to 0.01 mm and weighed with accuracy to 0.01 g. The samples were divided into 4 groups, dried at 105 °C to constant mass and subjected to in situ polymerisation. The samples were saturated with a solution of the following composition: 1. 115 cm3 of styrene and 1.85 g of maleic anhydride (MAn). 2. 2 cm3 of glycidyl methacrylate (GMA) in 115 cm3 of butyl methacrylate (BuMA). 3. 2 cm3 of glycidyl methacrylate (GMA) in 80 cm3 of ethyl methacrylate (EMA) and 35 cm3 of methyl acrylate (MA). In each case, 0.25 g of benzoyl peroxide (BPO) initiator was dissolved as well. The last two sample groups were the control one, consisting of the DW and TTDW samples respectively, but dried only, without monomer treatment and further polymerisation. The sample groups destined for modification were placed in glass vessels and loade with glass stopper to prevent floating. Appropriate monomer mixtures containing the initiator were poured into the vessels. Then the vessels with the samples were placed in a larger vacuum vessel and the air was evacuated using a vacuum pump for 30 minutes. The valve was then opened in the vacuum vessel to increase the pressure and fill up the pores in wood with

142 the monomers. After 30 minutes of soaking the process of air evacuation and pressure reduction was repeated. The next step involved removing the excess solution and placing the samples in polyamide pressure vessels, which were then hermetically sealed and placed in an oven. The samples were thermally hardened for 48 hours at 120 °C. Then samples were removed from the oven and pressure vessels, and weighed to determine the weight percentage gain (WPG). WPG was measured based on the mass of the samples before polymerisation (md) and the mass of the samples after modification (mm) using the following formula: m  m WPG  m d 100 % m d Then the samples were placed in a vessel filled with water and loaded against floating. The samples were removed from water after 3, 11, 24, 48 and 100 hours of soaking, excess of water was removed by filter paper and then samples dimensions and mass were measured. Water absorbability (W) was measured on the basis of the mass of the samples before soaking (m0) and the mass of the same samples after soaking (mw) based on the following formula: m  m W  w 0 100 % m 0

Swelling (KV) was calculated on basis of the volume of the samples. Due to the wood densification process, it was not possible to determine the swelling in fixed anatomical directions of wood, so the calculations were limited to the volume before soaking (V0) and the volume of the same samples after soaking (Vw) based on the following formula:

V V K  w 0 100 % V V 0 RESULTS The average density of densified poplar wood was 395.5 ± 0.03 kg/m3, and after thermal treatment the density dropped to 350 ± 0.03 kg/m3. This result indicates that the process of thermal treatment causes the densification to be partially inversed, probably due to interaction with steam or water contained in wood tissue.

A B 150%

15%

V K

100% 10%

50%

volume swelling, 5% water absorbability, water

DW+styrene/MAn DW+styrene/MAn TTDW+styrene/MAn TTDW+styrene/MAn

0% 0% 0 50 100 0 50 100

time, t/h time, t/h

Figure 4. Swelling (A) and water absorbability (B) of wood treated with styrene/MAn

143

WPG averaged between 30 and 40 % for both DW and TTDW wood, thus thermal treatment of densified wood is supposed not to significantly influence its capacity of monomers used. In the case of the control sample series, after 24 hours of soaking in water, the average swelling was found 15.5 % for DW and 9.7 % for TTDW, while average water absorbability 111 % and 76 % respectively. Figure 4 shows water absorbability and swelling found for the first series of modifications, using styrene and maleic anhydride. After the first hour of DW soaking, the water absorption was approx. 71%, while TTDW reached 27% at the same time. Swelling was found to be 8% and 2% respectively. Then, after 48 hours the absorption of water increased to 102% in the case of DW and 70% in the case of TTDW, while swelling increased to 14% and 7% respectively. The amount of water absorbed continued to increase, reaching 132% and 129%. The final swelling value was determined to be 16% for DW and 8% for TTDW. Over the entire experiment, TTDW swelling was about two times lower than DW one. The amount of water absorbed by DW increased twice as fast in the case of TTDW at the beginning, but after two days these values began to converge.

A 150% B

20%

W K 100%

10%

50%

volume swelling, water absorbability, water

DW+BuMA/GMA DW+BuMA/GMA TTDW+BuMA/GMA TTDW+BuMA/GMA 0% 0% 0 50 100 0 50 100

time, t/h time, t/h

Figure 5. Swelling (A) and water absorbability (B) of wood treated with BuMA/GMA

In the Figure 5 water absorption and swelling are presented, which were determined for treatment with BuMA/GMA (modification No 2). Within the first hour of soaking, absorbed water amount was reached approx. 65% (DW) or 31% (TTDW). Swelling was found to be 8% and 2% respectively. After two days, absorption increased to 102% (DW) and 77% (TTDW), while swelling; increased to 14% and 7% respectively. Finally, the amount of water absorbed reached 134% and 128%, while swelling was 21% and 8%, respectively for DW and TTDW. In this case, the swelling value was twice as high for DW than TTWD as well. The values of absorbed water as in the first modification after 48 hours began to converge. Figure 6 shows water absorption and swelling determined for treatment with the EMA/MA/GMA mixture (modification No 3). In this case the water absorption was approx. 55% (DW) and 30% (TTDW) after the first hour of soaking. Swelling at the level of 1% was found for both DW and TTDW. After two days, absorption increased to 117% for DW and to 94% for TTDW, while swelling increased to at the levels 14% and 6% respectively. In the end, the amount of water absorbed reached 165% and 137%, and swelling was 15% and 6% for DW and TTDW respectively. Initially, the swelling in both cases was the same and did not change after 5 hours of soaking; after 15 hours DW increased 10 times and then kept

144 gradually increasing. In this case, swelling of polymerised TTDW was the lowest among all the series tested.

B A 200%

15% W

V 150% K

10%

100% volume swelling,

5% absorbability, water 50% DW+EMA/MA/GMA TTDW+EMA/MA/GMA DW+EMA/MA/GMA TTDW+EMA/MA/GMA 0% 0% 0 50 100 0 50 100 time, t/h time, t/h

Figure 6. Swelling (A) and water absorbability (B) of wood treated with EMA/MA/GMA

In general, the tested methods of wood improvement demonstrated a great impact on the volume swelling of treated wood, while changes in water absorbability were found to be much lower. Strong interaction between cell wall and polymer can be thus concluded, reducing ability of wood to swell, in spite of still significant amount of water permeating into wood tissue. Changes in the swelling of densified wood were low in comparison to native wood of black poplar, which was investigated by Żmuda and Radomski (2018). At the same time, significant improvement of dimensional stability was found for densified and then thermally treated wood, as its swelling is about two time lower. The decrease in absorbability was observed mainly at the initial stage of soaking. This proves that the cell wall is still ready to interact with water. At the same time, the rate of change is strongly reduced, which can be linked to filling up the lumen with the polymer, and thus closing the fastest way of transporting water inside the wood. Such modified wood has a slightly lower water capacity and is more resistant to short-term exposure to wet conditions, but it is still able to adsorb and desorb water, providing a healthy indoor microclimate. On the other hand, its outdoor use is limited due to its relatively low dimensional stability, although it is much better than in the case of untreated wood (Li et al. 2010).

CONCLUSIONS Based on the obtained results of the tests of treated wood, it can be concluded that the application of in situ polymerisation reduces the hygroscopicity of wood. The application of a mixture of acrylates improves the wood. However, better results were obtained with the use of styrene and maleic anhydride. The following particular conclusions can be drawn as well:  Samples with higher WPG values were characterized by higher absorbability values on average.  Wood modified with the in situ method displays greater dimensional stability, while the addition of maleic anhydride additionally enhances its properties.  In this case, a strong time dependence was established, as the improvement of wood properties is high in a short period of time, but much lower in the case of long-term exposure.

145

 In all cases of tested modifications water absorption was found to be slightly lower in the case of densified and thermally treated wood samples, especially in a short period of time, than in densified wood. In contrast, the reduction of swelling for densified and thermally treated wood samples is much more conspicuous.  It is worth subjecting densified wood to thermal modification, as such wood has a more aesthetically pleasing and desirable colour and can be used as an exotic wood imitation, while simultaneously achieving much better dimensional stability.

Acknowledgement. The study was conducted as a part of research project founded by the National Centre for Research and Development BIOSTRATEG2/298241/10/NCBR/2016 „Intelligent systems for breeding and cultivation of wheat, maize and poplar for optimized biomass production, biofuels and modified wood.”

REFERENCES

1. BARTKOWIAK M., DOCZEKALSKA B., STRZELECKI S. 2015: Modification of wood with furfuryl alcohol catalysed by mixture of acid ahydrides. Ann. WULS- SGGW, For. And Wood Technol. 92; 26-29 2. BEKHTA P., PROSZYK S., KRYSTOFIAK T., SEDLIACIK J., NOVAK I., MAMONOVA M. (2017): Effects of short-term thermomechanical densificationon the structure and properties of wood veneers, Wood Material Science & Engineering, 12:1,40-54 3. CIABACH J., 2001: Właściwości żywic sztucznych stosowanych w konserwacji zabytków, UMK; 4. ERMEYDAN M.A., CABANE E., GIERLINGER N., KOETZ J., BURGERT I., 2014: Improvement of wood material properties via in situ polymerization of styrene into tosylated cell walls, RSC Advances, 4, 12981 5. HILL C., (2006); WOOD MODIFICATION. CHEMICAL, THERMAL AND OTHER PROCESSES, WILEY, BELGIUM 6. KRZYSIK F., 1978: NAUKA O DREWNIE, PWRiL, Warszawa; 7. LAZZARI M., CHIANTORE O., 2000: Thermal-ageing of paraloid acrylic protective polymers, Polymer, 41 (17), 6447 8. LI Y.F., LIU Y.X., WANG X.M., WANG F.H., 2010: Improvement of Durability of Wood by Maleic Anhydride, World Academy of Science, Engineering and Technology, 41, 67-70 9. LI Y., LIU Z., DONG X., FU Y., LIU Y. 2013: Comparison of decay resistance of wood and wood-polymer composite prepared by in-situ polymerization of monomers. International Biodeterioration & Biodegradation 84; 401-406 10. ROWELL R., 2005: Handbook of Wood Chemistry and Wood Composites Chapter 14. Chemical modification of wood., CRC Press, Boca Raton, London, New York, Washington, 11. ROWELL R., 2014: Acetylation of wood – A review, International Journal of Lignocellulosic Products, 1 (1), 1 12. SEJATI P.S., IMBERT A., GÉRARDIN-CHARBONNIER C., NANDIKA D., PRIADI T., GÉRARDIN P. 2017: Tartaric acid catalyzed furfurylation of beech wood. Wood Sci Technol 51; 379-394 13. SHIRAISHI N., Wood Plasticization w: HON D. (ed.), 2001: Wood and cellulosic chemistry, Mercel Dekker Inc., New York; 146

14. WEAST R.C, GRASSELLI J.G., ED(S), 1989: CRC Handbook of Data on Organic Compounds, 2nd Editon, CRC Press, Inc., Boca Raton, FL 15. ÜLKER O., BURDURLU E. (2016) Some Mechanical Properties of Densified and Laminated Lombardy Poplar (Populus nigra L.) Wood Research, 61 (6), 959-970 16. VASCONCELOS R., Del MENEZZI C. (2013) Utilization of a Three-Step Thermo- Mechanical Treatment to Modify Wood Properties, International Conference on Composite Materials-ICCM19, Montreal, Canada 17. ŻMUDA E., RADOMSKI A., (2018) Water resistance and swelling of black poplar wood (Populus nigra L.) modified by polymerisation in lumen with acrylate polymers, Annals of Warsaw University of Life Sciences - SGGW. Forestry and Wood Technology, 104, 345-352

Streszczenie: Badanie wpływu polimeryzacji in situ z użyciem styrenu lub akrylanów na nasiąkliwość i pęcznienie zagęszczonego termomechanicznie drewna topoli czarnej (Populus nigra L.).Próbki topoli czarnej poddano zagęszczeniu termomechanicznemu, a następnie przebadano je pod kątem poprawy odporności na wodę. Seria próbek została dodatkowo poddana modyfikacji termicznej w atmosferze azotu, a następnie próbki drewna tylko zagęszczonego oraz zagęszczonego i termowanego zostały zmodyfikowane metodą polimeryzacji mieszaniny monomerów, zawierających styren lub akrylany z dodatkiem komonomerów reagujących ze ścianą komórkową drewna. Wszystkie próbki badano przez długotrwałe moczenie w wodzie, określając spęcznienie objętościowe i nasiąkliwość. Drewno zagęszczone okazało się podatne na modyfikację metodą polimeryzacji in situ. Stwierdzono, że drewno zagęszczone i termowane jest znacznie bardziej kompatybilne z zastosowanymi polimerami, ponieważ zauważono silny spadek spęcznienia i mniej wyraźny spadek nasiąkliwości. Corresponding author:

Marta Gnacińska ul. Krzyżówki 13/40, 03-191, Warsaw, Poland email: [email protected] phone: +48 514 735 575

Andrzej Radomski Department of Wood Science and Wood Protection, Institute of Wood Sciences and Furniture, Warsaw University of Life Sciences ul. Nowoursynowska 159, 02-787 Warszawa email: [email protected]

ORCID ID: Radomski Andrzej 0000-0003-4437-009X

147

Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology № 108, 2019: 148-155 (Ann. WULS - SGGW, For. and Wood Technol. 108, 2019)

Studies on the suitability of oxidizing agents for discolouring lime and poplar wood in the first stage of transparent wood forming process

DARIA KAŹMIERCZAK, ANDRZEJ RADOMSKI Department of Wood Science and Wood Protection, Institute of Wood Sciences and Furniture, Warsaw University of Life Sciences

Abstract: Studies on the suitability of oxidizing agents for discolouring lime and poplar wood in the first stage of transparent wood forming process. Series of lime and poplar wood samples were prepared and subjected to oxidising agent in order to decolourise wood in bulk. Sodium chlorite solution in the environment of diluted acetic acid and alkaline hydrogen peroxide solution were used as different treating agents, followed by intense rinsing in water and drying the samples. The effect of wood delignification conditions such as time of treatment and reagent used was investigated. Changes in mass and dimensions of the samples were measured, and thus density changes were calculated. Colour changes were measured with colorimeter in CIE Lab colour space. In the case of lime wood swelling of the samples was observed at the first stage, along with mass loss, leading to density decrease by 10 % after 20 h exposure. In the case of poplar wood, shrinking of the samples was observed, but due to severe mass loss, final density was similar to lime wood. Colour changes correlated mainly with lightness parameter of the samples. Significant colour differences were found even at the shortest time of treatment.

Keywords: transparent wood, delignification, lime, poplar, wood colour

INTRODUCTION The properties of wood and its suitability for industrial purposes depend mainly on wood anatomical features, the chemical composition and the structure of cell wall (Krzysik 1974). Wood is one of the most abundant resources in the bio-based industry and yet it is also one of the most complex materials. Wood is essentially composed of cellulose, hemicelluloses, lignin, and extractives. The relative composition, however, varies greatly for various wood species (Stenius 2000). As technology advances, wood does not lose its significance. Development of new technologies allows modifying wood properties and thus the range of wood applications is still wider. Natural wood is not transparent in the visible spectra range and this is due to strong absorption and scattering of light. To make wood transparent, both absorption and scattering effects need to be eliminated. Light absorption is strongly dependent on the chemical composition and wood brownish colour can be connected to the presence of light-absorbing components such as lignin, low-molecular weight phenols and tannins. Among these, lignin accounts for around 80-95% of light absorption in wood (Müller et al. 2003). Lignin has originally yellow-brownish colour due to its phenolic character – responsible for aromatic ring absorption in the visible spectrum (Li et al. 2016). Transparent wood is a composite of delignified wood and transparent polymer, like poly(methyl methacrylate) (PMMA). The first reported study on transparent wood the modification was conducted in order to facilitate wood morphology studies (Fink 1984). Later on, engineering use of transparent wood was suggested by combining mechanical strength considerations with optical transmittance studies (Li et al. 2016). Many studies followed later around similar engineering use considerations (Li et al. 2019). Transparent wood shows lower density than glass (approx. 1.2 g/cm³), high optical transmittance (over 80%) and haze (over 70%), good mechanical performance, and potential for multi-functional modiﬁcations (Li et al. 2017b). Wood materials have good properties and are suitable for large-scale structures. Transparent building structure is such an

148 exemplification towards the future use of transparent wood, where light transmittance can be designed so that artificial light can be partially replaced by sunlight. Transparent wood roofs can be designed for some buildings, which will provide more uniform and comfortable illumination compared with conventional glass (Li et al. 2019). Another application for transparent wood is furniture making, where both aesthetical and functional properties are sought in modern homes and offices. There is possibility of diffusing luminescence by embedding quantum dots in a transparent wood panel. This is advantageous for planar light sources and luminescent construction elements or furniture (Li et al. 2017b). Currently the established and most commonly used approach to prepare wood templates for transparent wood is a delignification process with sodium chlorite (NaClO2), which causes oxidative aromatic ring-opening reactions to form acidic groups and make lignin degradation products soluble in water (Li et al. 2017a). This process usually takes place at a temperature around 80 °C, while samples thickness usually does not exceed 1 mm. Another delignification method consists of lignin modification process through alkaline hydrogen peroxide (H2O2) treatment, which selectively reacts and removes chromophore structures, while the bulk lignin is preserved (Ramos et al. 2008). In other studies, the wood samples were submerged in the lignin-modifying solution at 70 °C, until wood became white (Li et al. 2017a). According to earlier studies on pinewood, it was noted that the process of delignification generally leads to a higher porosity of wood, as the entire density was reduced from 440 kg/m³ in the reference samples to 330 kg/m³ after delignification (Frey et al. 2018). Delignification process typically removes around 25 % of wood tissue mass (Li et al. 2017b), which is consistent with the above values of density, when no changes in dimensions take place. The purpose of this work was to investigate the action of delignifying and oxidizing agents on the following wood parameters: mass, volume, density and colour. This is the initial stage in the development of transparent wood production technology, preceding the use of delignified wood material as filler in transparent plastics. In order to conduct the research, lime and poplar wood were selected, having lignin content of about 18% and 20% respectively, according to literature (Prosiński 1984). There are many reports on bleaching methods used in papermaking (e.g. Biermann 1996), but very few scientific reports on colour determination of bleached solid wood. Csiha & Papp (2013) investigated bleaching beech wood using sodium percarbonate and observed changes in all colour parameters of CIE Lab space. General information on wood bleaching can be found in a report by Forest Products Laboratory (1967). Kadir & Jantan (2016) studied colour changes of rubberwood under the bleaching, and found colour difference (E) in the range of 5 to 18 units.

MATERIALS Wood discs were taken from European lime (Tilia cordata Mill.) and white poplar (Populus alba L.) trees at the breast height. The cubic samples having dimensions of 15×15×15 mm were cut from the respective discs. Two solutions were prepared for each wood species: delignifying and bleaching ones. The mass and the volume of the samples were measured at room temperature before treatment. The bleaching solution was prepared in conical flasks by mixing 0.08 g of sodium hydroxide (NaOH) and 200 cm3 of 30 % hydrogen peroxide solution. The delignifying solution was prepared in conical flasks by mixing 3 3 200 cm of distilled water, 10 g of sodium chlorite (NaClO2) and 0.08 cm of glacial acetic acid (CH3COOH). The samples of both species were placed in the appropriate flask, supplied with a reflux condenser, and placed in a water bath heated to 80 °C. The treated samples of both species were collected at subsequent time intervals 1, 3, 6 and 20 hours of treatment.

149

The samples were then washed three times by boiling in distilled water, while the water was exchanged every hour. Reference samples were not treated by solutions, but were boiled in water as well, for comparison and determination the effect of washing. All the samples were dried at 105 ± 3 °C for 48 h before further measurements. Mass and dimensions of each sample were measured, and then their density was calculated. Colour changes of the samples were measured using BYK Spectro-guide gloss 45/0 spectrophotometer in CIE Lab colour space.

RESULTS Average mass loss of the samples in series is shown in the Figure 1. In the case of poplars wood, the difference in mass loss between both agents tested was significant over the entire period of experiment, reaching over 28 % in chlorite solution. In peroxide bath after initial loss to about 15 % within 8 hours, the changes in the next stage were negligible. The samples of poplar wood in chlorite bath showed constant mass loss in the same time. In the case of lime wood mass loss trends with both agents were more similar to poplar wood in peroxide bath. The changes in peroxide bath did not reach a plateau, featuring small, but still increase. On the other hand, in chlorite bath lime wood mass loss reached the level of poplar wood in chlorite bath and similarly the stage of stabile mass was observed.

Lime wood (peroxide) Lime wood (chlorite) Poplar wood (peroxide) Poplar wood (chlorite) 30

Mass loss [%] loss Mass 10

0 0 10 20 Treatment time [h]

Figure 1. Dependence of changes in mass of the samples on treatment time.

Clear differences in the behaviour of both tested wood species were observed when volume changes were measured. In the case of lime wood, a significant increase in sample volume was observed at the initial stage, which can be associated with swelling of the test material. In the case of poplar wood, however, volume loss was observed from the very beginning. Thus, despite the swelling, the mass loss of poplar wood was large enough to be accompanied by a volume loss. In the case of lime wood treated with hydrogen peroxide solution, at the initial stage there was a strong increase in volume, while at prolonged treating the volume of samples dropped, reaching the initial average value within 20 h. In the case of the sodium chlorite solution, the volume of lime wood samples increased at the initial stage of treatment and remained near unchanged until the end of the experiment. In the case of poplar wood treated with the hydrogen peroxide solution, overall volume loss was observed about 2 to 3 %, while after 3 hours of treatment a temporary deviation towards larger decrease was found. In the case of the sodium chloride solution, the volume loss of poplar wood samples

150 was very strong, in spite of changes of the trend, which was observed for the first 6 hours. 12% volume loss found was the biggest decrease among all the series tested.

Lime wood (peroxide) Lime wood (chlorite) Poplar wood (peroxide) Poplar wood (chlorite)

0 10 20

-5 Volume . change [%]

-15 Treatment time [h]

Figure 2. Dependence of changes in volume of the samples on treatment time.

The changes in density of wood samples, as the result of mass and volume changes, are presented in the Figure 3. Significant decrease at the first stages of the treatment and subsequent slowing down the changes was found in all cases, irrespectively of the direction of volume changes. The decrease of lime wood density was greater than of poplar wood, while the difference was greater in the case of peroxide treatment. Lime wood density decrease was generally uninfluenced by treatment method, and final decrease of approximately 10 % was observed. In the case of polar wood faster decrease was found for chlorite bath, reaching near 9 %, while in peroxide bath decrease did not exceed 7 %.

Lime wood (peroxide) Lime wood (chlorite) Poplar wood (peroxide) Poplar wood (chlorite) 0

-4

-8 Density change Density [%]

-12 0 10 20 Treatment time [h]

Figure 3. Dependence of density changes of the samples on treatment time.

Lightness (L) was found to gradually increase in most samples (Figure 4), except for the poplar wood in the hydrogen peroxide solution, which decreases slightly after initial growth. In contrast, Poplar in chlorite solution showed the highest level of brightness with

151 constant growth. Lime wood showed a similar increase, with hydrogen peroxide solution being the highest, reaching 0.7 units lower than poplar wood in chlorite solution.

Lime wood (peroxide) Lime wood (chlorite) Poplar wood (peroxide) Poplar wood (chlorite) 90

70 Lightness, L* .

50 0 1 3 6 20

Treatment time [h]

Figure 4. Dependence of lightness (L) changes of the samples on treatment time.

Figure 5 shows the average colour difference measured between the series reference samples and the treated ones. Such difference allows to better estimate the changes occurring due to the treatment, without the effect of soaking in water. Systematically increasing colour difference was found for lime wood treated with hydrogen peroxide. The same wood treated with chlorite solution featured an increase in colour difference as well, while the changes were significantly lower. In the case of poplar wood, treatment with hydrogen peroxide wood led to a clear difference for the first series of samples, collected after an hour of treatment, while in following measurements these differences were less significant. In the case of poplar wood treatment with chlorite solution, there was a growing trend of colour difference dependence on treatment time, in spite of a lower value of the ΔE parameter found after 6 hours. Due to the heterogeneity of the wood material, it can be argued that there are individual colour differences between the samples, which in turn can be sufficient explanation of trend fluctuations.

Lime wood (peroxide) Lime wood (chlorite) Poplar wood (peroxide) Poplar wood (chlorite)

10 Colour Difference, ΔE . Colour Difference,

0 1 3 6 20

Treatment time [h]

Figure 5. Dependence of changes of colour difference (E) treatment conditions

152

25 Lime Wood Poplar wood R2 = 0.9777

15 R = 0.9922 Colour Difference, ΔE Colour Difference,

5 72 78 84 Lightness, L*

Figure 6. Correlation between colour difference (ΔE) and lightness (L*) for both wood species tested

Figure 6 shows the correlation between lightness and colour difference for both wood species. An excellent correlation was obtained for poplar wood, with an R2 coefficient of 0.9922, while the correlation for lime wood was very good as well, with R2 of 0.9777. It can be concluded that changes in colour are primarily associated with the brightening of samples, while their hue changes were of minor significance, both for red-green, and yellow-blue directions.

CONCLUSIONS Based on the research results, the following conclusions can be drawn. Agents used for both wood species caused rapid weight loss in the initial stages and in later handling the weight loss stabilized. The maximum mass loss was 28 % for poplar wood treated with chlorite solution, while Li et al. (2017) found mass loss during the delignification process about 30 wt% of wood tissue for various wood species. Both tested wood species show different behaviour as the volume of lime samples increased while poplar wood samples shrinked. Their behaviour overtime was also different when using different oxidise agents. In the hydrogen peroxide solution, the lime wood mass after initial growth dropped continuously and in the case of poplar wood, such a situation occurred with the use of sodium chloride solution. This treatment method caused severe deformation of the samples, causing analytical difficulties. When a potential application is considered, possibility of dimensional change and deformation should be taken into account and appropriate method of further machining should be provided. Surprisingly, the density of the samples comparably decreased in each case. Larger mass loss was compensated by larger volume drop, while poplar wood showed a little higher density decrease in average. Non-uniform changes were found in wood colour during treatment. In general, wood of both species was still brighter over the experiment, in spite of some fluctuations. The colour differences, measured at the various stages of treatment against blind probes are clear except for samples of poplar wood in hydrogen peroxide solution. The colour difference for all samples after the first series was above five units of ΔE, which means that the average observer has the impression of a different colour from the reference samples (Budzyński et al., 2015). Another significant colour change is observed between 6 hours and 20 hours exposure of poplar wood treated with the hydrogen peroxide solution and poplar wood treated with sodium chlorite solution. Comparison of E values with directly measured lightness of

153 the samples showed excellent correlation and confirmed a naked-eye observation, that all samples became brighter without significant change in their hue.

REFERENCES

1. BIERMANN Ch. J., 1996: Handbook of Pulping and Papermaking (Second Edition), Academic Press 2. BUDZYŃSKI Ł., ZAJKOWSKI M., 2015: Modyfikacja parametrów kolorymetrycznych w oprawach oświetleniowych ze źródłami LED. Przegląd Elektromechaniczny. 7: 67-71. 3. CSIHA Cs., PAPP É.A., 2013: Investigation on bleaching Beech wood using environment friendly agent. Proceedings of the XXVIth International Conference Research for Furniture Industry (ed.: Szmardzewski, J.), Poznan, Poland, 19-20 September 2013. pp. 7-15 4. FINK S., 1984: Transparent wood a new approach in the functional study of wood Structure. Holzforschung. 46: 403-408. 5. FOREST PRODUCTS LABORATORY, 1967: Bleaching wood, Res. Note FPL-0165, Madison, Wisconsin. 6. FREY M., WIDNER D., SEGMEHL J., CASDORFF K., KEPLINGER T., BURGERT I., 2018: Delignified and Densified Cellulose Bulk Materials with Excellent Tensile Properties for Sustainable Engineering. ACS Applied Materials & Interfaces. 10(5): 5030-5037. 7. KADIR R. & JANTAN M.D., 2016: Enhancement of Hevea brasiliensis properties through chemical application. Anais da Academia Brasileira de Ciências, 88(4), 2081- 2092 8. KRZYSIK F., 1974: Nauka o drewnie. Państwowe Wydawnictwo Naukowe: 15-17. 9. LI Y., FU Q., ROJAS R., YAN M., LAWOKO M., BERGLUND L. 2017a: A new persective on transparent wood: Lignin – retaining transparent wood. ChemSusChem 10: 3445–345. 10. LI Y., FU Q., YANG X., BERGLUND L., 2017b: Transparent wood for functional and structural applications. Philosophical: 376. 11. LI Y., FU Q., YU S., YAN M., BERGLUND L., 2016: Optically Transparent Wood from a Nanoporous Cellulosic Template: Combining Functional and Structural Performance. Biomacromolecules. 17: 1358−1364. 12. LI Y., VASILEVA E., SYCHUGOV I., POPOV S., BERGLUND L. 2019: Optically Transparent Wood: Recent Progress, Opportunities, and Challenges. Optical Mater. 6: 1-14. 13. MÜLLER U., RÄTZSCH M., SCHWANNINGER M., STEINER M., ZÖBL H., 2003: Yellowing and IR changes of spruce wood as result of UV- irradiation. J. Photochem. Photobiol. 69: 97-105. 14. PROSIŃSK S., 1984: Chemia drewna. Państwowe Wydawnictwo Rolnicze i Leśne: 56. 15. ROMOS E., CALATRAVA S., JIMENEZ L. 2008: Bleaching with Hydrogen Peroxide. A Review. Afinidad. 65: 366–373. 16. STENIUS P., 2000: Forest Products Chemistry. Helsinki Fapet Oy: 50-56.

154

Streszczenie: Badanie przydatności środków utleniających do odbarwiania drewna lipowego i topolowego w pierwszym etapie wytwarzania przezroczystego drewna. Przygotowano serie próbek drewna lipowego i topolowego i poddano je obróbce utleniającej w celu odbarwienia drewna w całej objętości. Jako reagenty zastosowano roztwór chlorynu sodu w środowisku rozcieńczonego kwasu octowego oraz alkaliczny roztwór nadtlenku wodoru, po czym próbki poddawano intensywnemu płukaniu w wodzie i suszeniu. Zbadano wpływ warunków delignifikacji drewna, takich jak czas obróbki i zastosowany odczynnik. Mierzono zmiany masy i wymiarów próbek, i obliczono zmiany gęstości. Zmiany koloru mierzono za pomocą kolorymetru w przestrzeni kolorów CIE Lab. W przypadku drewna lipy zaobserwowano pęcznienie próbek na pierwszym etapie, wraz z utratą masy, co prowadziło do zmniejszenia gęstości o 10% po 20 godzinach obróbki. W przypadku drewna topolowego stwierdzono kurczenie się próbek, ale z powodu znacznej utraty masy gęstość końcowa była zbliżona do drewna lipowego. Zmiany koloru korelowały głównie z parametrem jasności próbek. Istotne różnice barwy stwierdzono nawet po najkrótszym czasie obróbki.

Corresponding author:

Daria Kaźmierczak, ul. Kujawska 3/44, 00-739, Warsaw, Poland email: [email protected] phone: +48 694 951 479

ORCID ID: Radomski Andrzej 0000-0003-4437-009X

155