1 INSTYTUT TECHNOLOGII DREWNA WOOD TECHNOLOGY INSTITUTE DREWNO PRACE NAUKOWE DONIESIENIA KOMUNIKATY WOOD RESEARCH PAPERS RESEARCH REPORTS ANNOUNCEMENTS Vol. 55 POZNAŃ 2012 Nr 188
2 Wydanie publikacji dofinansowane przez Ministerstwo Nauki i Szkolnictwa Wyższego w ramach programu Index Plus. The journal is financially supported by Polish Ministry of Science and Higher Educations under the Index Plus programme. Recenzenci vol. 55 (Reviewers) prof. Ing. Pavlo Bekhta, Dr. Sc.; dr Ricco Emmler; prof. RNDr. František Kačík, PhD.; prof. Ing. Jozef Kúdela, CSc.; prof. mgr Jurai Ladomersky, CSc.; prof. Robert Lanouette; prof. Wojciech Lis; dr inż. Andrzej Mazur; prof. dr. Peter Niemz; dr inż. Adam Niesłochowski; prof. Pekka Ollonqvist; prof. dr hab. Stanisław Proszyk; prof. Roger M. Rowell; dr. hab. Jerzy Świgoń, prof. UP; Ing. Marek Trenciansky, PhD. Publikacje indeksowane są w bazach danych (Publications are indexed in the databases): Agro Science Citation Index Expanded BazTech SCOPUS DREWINF The Central European Journal of Social Sciences and Humanities W 2010 roku czasopismo znalazło się na tzw. liście filadelfijskiej (ISI Master Journal List) z obliczonym Impact Factorem (IF). W 2010 roku czasopismo zostało uhonorowane Medalem im. Michała Oczapowskiego. Artykuły polskojęzyczne zawierają streszczenia w języku angielskim, a obcojęzyczne w języku polskim. Spisy treści, streszczenia i pełne teksty artykułów są dostępne na stronie Wersja pierwotna papierowa. In 2010 the journal was indexed on ISI Master Journal List with calculated Impact Factor (IF). In 2010 the journal was honored with Michał Oczapowski Medal. Polish language articles have summaries in English language, and foreign language articles have summaries in Polish language. Tables of contents, summaries, and full versions of the articles are available at The original version paper. Wydawca (Publisher): Instytut Technologii Drewna ul. Winiarska 1, Poznań, Polska (Poland) Adres Redakcji (Editor s address): Instytut Technologii Drewna ul. Winiarska 1, Poznań tel.: , , fax: , Copyright by Instytut Technologii Drewna w Poznaniu Poznań 2012 ISSN Projekt okładki (Cover design): Piotr Gołębniak Skład komputerowy (Computer typesetting) oraz druk (Print): Studio Poligrafia, ul. Bułgarska 10, Poznań, tel.: Nakład (Edition): 520 egz.
3 SPIS TREŚCI CONTENTS Prace naukowe Research papers Adam Krajewski, Piotr Witomski, Piotr Bobiński, Aleksandra Wójcik, Magdalena Nowakowska: An attempt to detect fully-grown house Longhorn Beetle larvae in coniferous wood based on electroacoustic signals (Próba wykrywania wyrośniętych larw Spuszczela Pospolitego w drewnie iglastym na podstawie sygnałów elektroakustycznych)... Jan Bocianowski, Krzysztof Joachimiak, Adam Wójciak: The influence of process variables on the strength properties of NSSC birch pulp. Towards the limits of optimization: part one the effect of liquor ratio (Wpływ zmiennych czynników procesowych na właściwości wytrzymałościowe brzozowych mas półchemicznych (NSSC). W kierunku granic optymalizacji: część pierwsza wpływ modułu cieczy warzelnej)... Iwona Frąckowiak, Karolina Mytko, Ryszarda Bendowska: : Content of formaldehyde in lignocellulosic raw materials for particleboard production (Zawartość formaldehydu w surowcach lignocelulozowych do produkcji płyt wiórowych)... Maciej Wilczyński, Krzysztof Warmbier: Elastic moduli of veneers in pine and beech plywood (Moduły sprężystości fornirów w sklejce sosnowej i bukowej)... Zbigniew Czech, Magdalena Witczak, Agnieszka Kowalczyk, Jagoda Kowalska: The influence of residue monomers on selected properties of acrylic pressure-sensitive adhesives (Wpływ nieprzereagowanych monomerów na wybrane właściwości poliakrylanowych klejów samoprzylepnych)... Monika Zajemska: Analiza teoretyczno-eksperymentalna wpływu współspalania odpadów drzewnych w formie paliw reburningowych z gazem ziemnym na skład chemiczny spalin (Theoretical and experimental analysis of the impact of wood co-combustion, as a reburning fuel, with natural gas on the chemical composition of flue gases)... Elżbieta Mikołajczak: The profitability of converting sawmill by-products into energy (Opłacalność przerobu tartacznych produktów ubocznych na energię)
4 Doniesienia naukowe Research reports Kinga Szentner, Izabela Ratajczak, Bartłomiej Mazela: Improvement of water- -based wood coating performance MTMOS reactivity with wood (Poprawa parametrów wodorozcieńczalnej powłoki do drewna reaktywność MTMOS z drewnem)... Jakub Gawron, Magdalena Szczęsna, Tomasz Zielenkiewicz, Tomasz Gołofit: Cellulose crystallinity index examination in oak wood originated from antique woodwork (Badanie indeksu krystaliczności drewna dębowego pochodzącego z zabytkowej stolarki budowlanej)... Csilla Vanya: Damage problems in glued laminated timber (Problemy uszkodzeń w tarcicy klejonej warstwowo)
5 Drewno. Pr. Nauk. Donies. Komunik. 2012, vol. 55, nr 188 PRACE NAUKOWE RESEARCH PAPERS Adam Krajewski, Piotr Witomski, Piotr Bobiński, Aleksandra Wójcik, Magdalena Nowakowska 1 AN ATTEMPT TO DETECT FULLY-GROWN HOUSE LONGHORN BEETLE LARVAE IN CONIFEROUS WOOD BASED ON ELECTROACOUSTIC SIGNALS Initial comparative tests of a method for the detection of house longhorn beetle larvae (Hylotrupes bajulus L.) were carried out by means of recording the acoustic effects caused by these insects, as well as another instrumental method based on the use of X-ray pictures. Insofar as the X-ray method brings completely reliable results as regards the detection of the presence of fully-grown house longhorn beetle larvae in wood, the accuracy of situation assessment based on the electro-acoustic method is estimated to be approximately 70-80%, based on the results of the research presented herein. In this research the accuracy of situation assessment is understood as the correct determination of an estimated number of larvae and their condition. The differences in situation assessment accuracies using both methods were statistically verified by means of the Czebyszew method. Since the tests were initial in nature without the use of standard patterns, as these have not been developed yet, the obtained results are considered highly satisfactory. Keywords: woodworm, wood, longhorn beetle, electroacoustic, instrumental detection, detection of insects, Hylotrupes bajulus L. Adam Krajewski, Warsaw University of Life Sciences SGGW, Poland Piotr Witomski, Warsaw University of Life Sciences SGGW, Poland Piotr Bobiński, Warsaw Technical University, Poland Aleksandra Wójcik, Warsaw University of Life Sciences SGGW, Poland Magdalena Nowakowska, Warsaw Academy of Computer Science, Management and Administration, Poland
6 6 Adam Krajewski, Piotr Witomski, Piotr Bobiński, Aleksandra Wójcik, Magdalena Nowakowska Introduction The answer to the question whether wood contains the living larvae of xylophagous insects which are destroying it, or only abandoned insect galleries, is a basic issue when it comes to taking action concerning the protection of wooden structures and objects. Most often the traditional method for situation assessment is used. Decisions are made based on a series of symptoms, which are often subjectively formulated by the person assessing. According to all handbook publications, amongst these symptoms are the following: occurrence of new outlets on the wood surface, so-called wood meal spilling out from the outlets, making little mounds or trickles on the wood surface, the presence in buildings or at storage yards of living or dead adult forms of a given pest species or its parasites and predators, in the case of some insect species the sounds of wood boring by larvae, which is audible to the naked ear. These symptoms would often be unreliable, therefore attempts to apply instrumental methods to the assessment of wood infestation by insects have been made. The instrumental methods for the detection of xylophagous insects in wood have aroused great interest for many years. However, not all the instrumental methods may be used in practice, especially in field conditions. X-ray photography and CAT scanning, which are examples of effective methods, cannot be classified as field test methods. Video-endoscopy does not enable an insight into the wood inside, but offers only an image of inaccessible structural parts, which must be examined during a traditional examination. Previous laboratory attempts to apply electro-acoustics to the detection of feeding house longhorn beetle larvae were made in Europe, i.e. in Germany [Kerner et al. 1980; Plinke 1991; Esser et al. 1999] and Poland [Bobiński et al. 2006; Krajewski et al. 2011]. Even in the case of such an unquestionable symptom as the sounds of wood boring by house longhorn beetle larvae, there are some classification issues nonetheless the insects do not make a sound when the evaluator assessing the feeding grounds would like them to. This may be a result of thermal conditions or periodic inactivity of the larvae due to other reasons. One should also take into consideration that so-called background may overlap with the electro-acoustic record of the vital activity of the larvae. If the house longhorn beetle larvae are large and make a lot of noise, sometimes they may create effects that may be misinterpreted if heard by the naked ear. The development and perfecting of a method which would enable the recording of some of the vital activities of xylophagous and under-the-bark insects and the interpretation of the number or behaviour of these animals based on the obtained results, would also (besides being of practical use) enable basic research on some aspects of the biology of some species. Based
7 An attempt to detect fully-grown house longhorn beetle larvae in coniferous wood based on... 7 on the promising results of the recording of the electro-acoustic signals of the house longhorn beetle [Bobiński et al. 2006], an assessment of the possibilities of detecting the larvae of this species, using a simplified simulation of conditions in a large-dimensional wooden structure, was carried out at SGGW in Warsaw in the Faculty of Wood Technology in the Wood Protection Department. Materials and methods The effects of the movements of house longhorn beetle larvae s maxillae and bodies in the feeding ground were recorded electro-acoustically. Based on the recordings, an attempt to determine the activity of these insects in wood was made together with an attempt to define the presence and approximate number of individual organisms based on the activity determination. The research was carried out in the form of laboratory tests simulating field conditions, taking into account constraints resulting from the lack of developed standards and thus the lack of objectivity showed by the people classifying the presence of the larvae in the galleries. The house longhorn beetle larvae (Hylotrupes bajulus L.) used in the research were fully-grown individual organisms of a mass ranging from 30 to 190 mg. A simplified simulation of conditions found within a large-dimensional structure was used. The insects were placed in 200 mm blocks (sections) of structural wood with a cross-section of mm. The experiment was carried out on Scots pine (Pinus sylvestris L.) and silver fir (Abies alba Mill.). A larvae-free Norway spruce (Picea abies (L.) Karst.) was also used as the equivalent of a placebo used in medical science. All the blocks were randomly marked with letters from A to L. The larvae were placed in pine and fir wood, 10 individual organisms per block. All the blocks, including the larvae-free ones, had holes bored to place the larvae, holes which were plugged with compacted wood meal left by feeding insects. In the case of larvae-free blocks, wood meal was artificially put in the holes and compacted. The aim of simulating the places seemingly bored by the larvae was to avoid a situation, where evaluators are influenced by the initial number of the test material. The larvae were placed in the holes bored in the blocks fronts. The holes were twice as deep as a larva s body length, and a diameter approximately 0.5 mm larger than that of a larva s body-width at the widest point. The larvae placed in the wood were left there for 4 weeks in an incubator at a temperature of 28 C, which gave the insects time to bore into the wood. The research aimed at an objective evaluation of the effectiveness of house longhorn beetle larvae detection in wood. The lack of knowledge about the blocks contents, which was characteristic of evaluators assessing the recording of electro-acoustic activity of the larvae, was achieved in the following way. Individual blocks with the larvae were subjected to an asphyxiating atmosphere of nitrogen for various periods with a view to causing a different death rate of these insects. The time needed to ensure mortal effects from a low-oxygen at-
8 8 Adam Krajewski, Piotr Witomski, Piotr Bobiński, Aleksandra Wójcik, Magdalena Nowakowska mosphere was determined in separate research devoted to the effectiveness of combating xylophagous insects with nitrogen and argon, the results of which will be published in the future. Due to these actions, the number of living larvae in particular blocks was not known to any of the observers interpreting the results of the experiment. The adopted method for misinforming the observers interpreting the record of the blocks contents, enabled a reduction in the possibility of them being influenced (while evaluating the larvae s activity) by the knowledge of the number of the larvae placed, as indicated by the number of holes bored in the front of each sample. After gas treatment, the block containing the larvae were mixed with the larvae-free blocks, which also had 10 bored holes in their fronts plugged with wood meal. The presence of larvae in wood was recorded using a prototype apparatus built at Warsaw University of Technology at the Faculty of Electronics and Information Technology in the Electro-Acoustic Institute. The apparatus was previously used to identify the acoustic characteristics of a single feeding larva of xylophagous insects [Bobiński et al. 2006]. This apparatus processes all and any acoustic waves carried by wood into electric signals. The apparatus is equipped with, among other things, an acceleration sensor, the signal of which was strengthened by a charge preamplifier. Signals were recorded by a sound card and presented in the form of a graphical record on a computer monitor. The signals were recorded as a level of signal amplitude in relation to the maximum amplitude that can be registered by the sound card, i.e. an amplitude, which may be achieved by the maximum regulation of an analogue-digital converter. The average density of the pine blocks was 440 kg/m 3 and their average ring width was 4.2 mm. The average density of the fir blocks was 448 kg/m 3 and their average ring width was 5.0 mm. To record the electro-acoustic activity of the larvae, the sensor was placed in the middle of a block s front mm (situation α), therefore at its cross-section. For comparative reasons, in the case of most blocks a recording of the electro-acoustic effects transversely to the grain was carried out as well, i.e. the sensor was placed in the middle of a sample s surface mm (situation β). The air (ambient) temperature during the recording of the acoustic signals ranged from 20 C to 24 C: for samples A-D approximately 21 C on average and for samples H-L approximately 20 C on average. The acoustic signals were recorded for 24 hours. The obtained results were classified independently by three evaluators. The evaluators were marked with numbers from 1 to 3. The following recording times were analysed: 24 hours, 2 hours and 20 minutes of recording time. In order to present multi-hour records graphically, the recording effects were compressed by computer. To assess the activity and contents of the feeding grounds, the records of electro-acoustic activity of the larvae, recorded when the sensor was placed in the middle of a block s front mm (situation α), were used. The re-
9 An attempt to detect fully-grown house longhorn beetle larvae in coniferous wood based on... 9 cords taken when the sensor was placed in the middle of a sample mm (situation β) did not substantially differ from the first variant. The assessments of particular evaluators are given in table 1. It was assumed that the recording of electro-acoustic activity of the entomological contents of the feeding grounds characterised the presence and number of the larvae in the wood. The following scale of the electro-acoustic assessment of the feeding grounds activity, which should stem from the following number of living larvae in a block, was applied: no record of activity lack of larvae in the wood or all larvae are dead, low activity recorded the equivalent of the presence of a small number of living larvae in the wood (1 4 individual organisms), high activity recorded the equivalent of the presence of a large number of living larvae in the wood (5 10 individual organisms). Table 1. Evaluation of electroacoustic signals as a result of the presence, activity and abundance of old house borer larvae (marked in bold are accurate assessment) Tabela 1. Ocena elektroakustycznej aktywności, będącej wynikiem obecności, czynności i liczebności żywych larw spuszczela pospolitego w drewnie iglastym (pogrubionym drukiem zaznaczono trafne oceny) Oznaczenie klocka, gatunek drewna i teoretycznie zakładana aktywność larw (na podstawie zawartości klocka) Determination of sample, wood species and theoretically expected activity of the larvae (based on the content of the sample) Liczba żywych larw w momencie prowadzenia nasłuchu The number of live larvae at the time of listening Aktywność larw w drewnie według obserwatora nr 1 Activity of larvae in the wood according to observer 1 Aktywność larw w drewnie według obserwatora nr 2 Activity of larvae in the wood according to observer 2 Aktywność larw w drewnie według obserwatora nr 3 Activity of larvae in the wood according to observer A, sosna, aktywność mała A, pine, little activity B, sosna, aktywność duża B, pine, substantial activity C, jodła, aktywność mała C, fir, little activity 2 żywe larwy 2 live larvae 10 żywych larw 10 live larvae 1 żywa larwa 1 live larva brak no duża substantial brak no mała little duża substantial brak no mała little duża substantial brak no
10 10 Adam Krajewski, Piotr Witomski, Piotr Bobiński, Aleksandra Wójcik, Magdalena Nowakowska Table 1. Continued Tabela 1. Ciąg dalszy D, jodła, aktywność duża D, fir, substantial activity E, świerk, brak aktywności E, spruce, no activity F, jodła, aktywność duża F, fir, substantial activity G, sosna, aktywność duża G, pine, substantial activity H, świerk, brak aktywności H, spruce, no activity I, świerk, brak aktywności I, spruce, no activity J, sosna, aktywność mała J, pine, little activity K, jodła, brak aktywności K, fir, no activity L, świerk, brak aktywności L, spruce, no activity Udział prawidłowych ocen sytuacji The share of the correct assessment of the situation 10 żywych larw 10 live larvae nie obsadzono larw larvae not inserted 8 żywych larw 8 live larvae 6 żywych larw 6 live larvae nie obsadzono larw larvae ot inserted nie obsadzono larw larvae not inserted 2 żywe larwy 2 live larvae brak żywych larw w wyniku duszącego działania azotu no live larvae as a result of choking action of nitrogen nie obsadzono larw larvae not inserted duża substantial brak no duża substantial mała little brak no brak no brak no brak no brak no duża substantial brak no duża substantial duża substantial brak no brak no brak no brak no brak no duża substantial brak no duża substantial duża substantial brak no brak no brak no brak no brak no 67% 83% 83%
11 An attempt to detect fully-grown house longhorn beetle larvae in coniferous wood based on Following this, X-ray photographs of the samples were taken in order to see the positions of the larvae, and then the wood was chipped. The X-ray photographs were verified by the assessment of the larvae s condition by means of direct observation after the blocks were split into very small pieces of wood in which no larva could hide. The movement was considered an attribute of a living larva the insects brought out from the wood were regarded as alive if they showed liveliness, either spontaneously or when mechanically irritated. The results of the actual entomological contents of the 12 blocks, accurately imaged in the X-ray photographs, were compared with the assessments carried out by each of the 3 evaluators drawing conclusions based on the recorded electro- -acoustic signals. If the evaluator s assessment of the feeding ground conditions (the presence of living larvae and their number within the assumed limits), carried out separately for each block based on the electro-acoustic signals, was correct, then it was marked with the number 1, if the assessment was wrong, then it was marked with the number 0. In the case of the X-ray photographs, all 12 results of the interpretations were correct and their assessment number was 1. Hence, these results became an element of the verification of the accuracy of the assessment carried out by means of classifying the electro-acoustic recordings as particular categories. For the 12 cases of assessment carried out by each evaluator separately, an arithmetic mean was calculated. Using Czebyszew inequality, the significance of the difference between an average from a given evaluator s assessments of the blocks contents, carried out by means of recording the electro-acoustic activity of the larvae, and the average from classifications made based on the X-ray photographs, which corresponded with the visual assessment of the blocks contents carried out after they were split, was determined. Using this method, the absolute difference between the arithmetic mean from the assessments carried out separately by each evaluator and the mean from classifications based on the X-ray photographs were compared, with a triple standard error in the difference of these arithmetic means: while: (1) where: xr xn i n an average assessment based on the X-ray photographs, classification of the contents of block no. i based on the X-ray photograph, an average assessment based on the interpretation of the electroacoustic effects carried out by evaluator N, classification of the assessment of evaluator N concerning block no. i, number of compared assessments (corresponding with the number of blocks). (2)
12 12 Adam Krajewski, Piotr Witomski, Piotr Bobiński, Aleksandra Wójcik, Magdalena Nowakowska In the case that the first inequality (or equation) is satisfied, the difference between the means is significant, otherwise the difference may be incidental. In order for the method to hold promise for practical use, the difference between the average assessment carried out by each evaluator and the average assessment resulting from the X-ray photographs (which corresponds with the actual situation evaluated after the blocks are split) should be insignificant. Results and discussion The levels of amplitudes of the recorded electro-acoustic impulse range from approximately -20 to 0 db on average in relation to a maximum amplitude, which may be recorded by the sound card. These are levels recordable by the sound card. The 20-minute recordings, cut from a multi-hour recording, proved to be the most useful for an assessment of the presence of the larvae in the wood and determination of their number. The examples of the recordings of the electro-acoustic activity of the numerous larvae in the wood and of the lack of larvae, but with background interference, are presented in fig. 1. The results of the assessment of the presence and number of the larvae in the wood are given in table 1. Comparing the absolute difference between the arithmetic means obtained from the assessment carried out using the recording of electro-acoustic effects and X-ray photographs with a triple value of the standard error between them, it was observed that there was no statistically significant difference between the assessment of the condition of the blocks contents carried out using the recording of the electro-acoustic activity of the larvae and the actual condition. In the case of evaluator no. 1 it was 0.33 < 0.43, and in the cases of evaluators no. 2 and no. 3 it was 0.17 < Previously only the recordings of the electro-acoustic activity of a single house longhorn beetle larva were published, without any classification concerning the effectiveness of the determination of the larvae number in wood [Kerner et al. 1980; Plinke 1991; Bobiński et al. 2006; Krajewski et al. 2011]. It was observed that the sound signals emitted by house longhorn beetle larvae during feeding encompass the frequency range hovering at 10 khz, while in the case of house furniture beetle larvae (Anobium punctatum De Geer) this range hovers at approximately 20 khz [Esser et al. 1999]. Despite the lack of a previously developed reference scale, in the case of the test results described herein, in most cases the three evaluators correctly classified the approximate number of larvae in the wood. However, none of the evaluators achieved 100% classification accuracy, although the eavesdropped larvae quite often may be audible to the naked ear. Difficulties in the interpretation of the records might have been caused by the very limited activity of some larvae in the
13 An attempt to detect fully-grown house longhorn beetle larvae in coniferous wood based on blocks, larvae which beforehand were subjected to the asphyxiating atmosphere of nitrogen for a longer time. The insects probably needed a relatively long time to get over the modified atmosphere with the reduced content of oxygen in favour of nitrogen. In such cases, the presence of a living larva might not be revealed as a recording of electro-acoustic activity corresponding with a healthy insect. Fig. 1. Varying examples of electroacoustic activity of old borer house larvae in wood: F large, G small and E zero, when 24-hour recordings are compressed and 2-hour and 20-minute fragments are isolated from them Rys. 1. Przykład zapisu różnej intensywności elektroakustycznej aktywności larw spuszczela pospolitego w drewnie: F dużej, G małej i E zerowej przy skompresowaniu wyników odsłuchiwania w czasie 24 h oraz wyjęciu z nich okresów 2 h i 20 min
14 14 Adam Krajewski, Piotr Witomski, Piotr Bobiński, Aleksandra Wójcik, Magdalena Nowakowska Research on the electro-acoustic method for the detection of insects in wood are continuing in the Wood Protection Institute in co-operation with the Electro- -Acoustic Institute, with a view to applying it in practice and in discovering some aspects of the biology and behaviours of both xylophagous and under-bark species. It should be added that research on the use of this method for the detection of the presence of the larvae of a dangerous species of Asian longhorn beetle (Anaplophora glabripennis) (carried to Europe) in the wood of living and dying trees [Becker 2000] and also in the wood of pallets and crates [Fleming et al. 2005], as well as for the detection of various species of earth termites, arouses great interest all over the world. Hitherto the research on termites was carried out as laboratory [Lemaster et al. 1997] and field tests [Fujii et al. 1999; Mankin et. al. 2002; Dunegan 2005; Mankin, Benshemesh 2006]. Conclusions Although the tests were relatively little advanced, the results obtained using the method of electro-acoustic recording of the activity of the grown larvae of the house longhorn beetle, were satisfying in terms of the assessment of the contents of the feeding grounds of this insect. It can be assumed that this method holds great promise of success, provided it is further improved, and may probably be applied to the detection of wood pests and to the entomology used in wood science and forestry when some aspects of the biology of xylophagous (and probably also cambium-feeding) insects are researched. The average accuracy of the assessments of 3 observers evaluating the activity of the house longhorn beetle larvae hovered at approximately 70-80%, and the average results of the assessment of the feeding ground contents, obtained by this method, were not statistically different from the actual contents of the tested wood samples determined by examination of the X-ray photographs and by evaluation. Acknowledgements Research financed from the funds for science in the years granted in the form of a research project no. N N References Becker H. : Asian Longhorned Beetles, Agricultural Research, June 2000, usda.gov Bobiński P., Krajewski A., Witomski P. : Acoustic properties of xylophagic insect activity. Annals of Warsaw Agricultural University : Dunegan H. L. : Detection of movement of termites in wood by acoustic emission techniques,
15 An attempt to detect fully-grown house longhorn beetle larvae in coniferous wood based on Esser P., van Staalduinen P., Tas A. : The Woodcare project: Development of detection methods for Death watch beetle larvae and fungal decay, prepared for the 30 th Annual Meeting, Rosenheim Fleming M. R., Bhardwaj M. C., Jankowiak J. J., Shield J. E., Roy R., Agrawal D. K., Bauer L. S., Miler D. L., Hoover K. : Noncontact ultrasound detection of exotic insects in wood packing materials, Forest Product Journal : Fujii Y., Yanase Y., Yoshimura T., Imamura Y., Okumura S., Kozaki M. : Detection of Acoustic emission (AE) Generated by Termite Attack in Wooden House, prepared for 30 th Annual Meeting, Rosenheim Kerner G., Thile H., Unger W. : Gesicherte und Zerstörungsfreie Ortung der Larven holzzerstörender Insekten im Holz, Holztechnologie, : Krajewski A., Kozyra K., Wójcik A., Witowski P., Bogusław A., Bobiński P. : The use of electro-acoustics in the evaluation of the effectiveness of old house borer in wood with p-dichlorobenzene. Lisowe Gospodarstwo, Lisowa, Papierowa i Dieriewoobrobna Promisłowist [37.1]: Lemaster R. L., Beall F. C., Lewis V. R. : Detection of termites with Acoustic Emission, Forest Product Journal : Mankin R. W., Osbrink W. L., Oi F. M., Anderson J. B. : Acoustic detection of termite infestations in urban trees, Journal of Economic Entomology [5/95]: Mankin R. W., Benshemesh J. : Geophone Detection of Subterranean Termite and Ant Activity, Journal of Economic Entomology, [1/99]: Plinke B. : Akustische Erkennung von Insektenbefall in Fachwerk. Holz als Roh und Werkstoff : 4 PRÓBA WYKRYWANIA WYROŚNIĘTYCH LARW SPUSZCZELA POSPOLITEGO W DREWNIE IGLASTYM NA PODSTAWIE SYGNAŁÓW ELEKTROAKUSTYCZNYCH Streszczenie Przeprowadzono wstępne badania porównawcze metody wykrywania larw spuszczela pospolitego (Hylotrupes bajulus L.) za pomocą rejestracji efektów akustycznych powodowanych przez te owady z inną metodą instrumentalną, opartą o wykorzystanie zdjęć rentgenowskich. O ile metoda rentgenowska daje całkowicie pewne wyniki w stosunku do wykrywania obecności wyrośniętych larw spuszczela pospolitego w drewnie, to poziom trafności oceny sytuacji przy użyciu metody elektroakustycznej szacuje się na podstawie rezultatów niniejszych badań na ok %. Jako trafność oceny sytuacji rozumiano tu prawidłowe określenie szacunkowej liczby larw i ich stanu. Różnice w trafności oceny sytuacji za pomocą obu metod poddano weryfikacji statystycznej przy użyciu metody Czebyszewa. Ponieważ były to wstępne badania bez użycia standardowych wzorców, które nie zostały jeszcze wypracowane, uznaje się uzyskane wyniki za bardzo zadowalające. Słowa kluczowe: drewno, spuszczel pospolity, owady, elektroakustyka, metody instrumentalne, wykrywanie owadów, Hylotrupes bajulus L.
17 Drewno. Pr. Nauk. Donies. Komunik. 2012, vol. 55, nr 188 Jan Bocianowski, Krzysztof Joachimiak, Adam Wójciak 1 THE INFLUENCE OF PROCESS VARIABLES ON THE STRENGTH PROPERTIES OF NSSC BIRCH PULP. TOWARDS THE LIMITS OF OPTIMIZATION: PART ONE THE EFFECT OF LIQUOR RATIO The aim of this work was to establish the best mechanical and strength properties of NSSC pulp using operating conditions at which the lowest (optimal) spent liquors, cooking temperature and time of treatment can be obtained. Minimal and maximum limits of independent variables were proposed using real mill conditions as starting points. The analyzed variables were: cooking time (13 to 15.5 min), cooking temperature (172 to 179 C), liquor- to- wood ratio (from 1.2 to 2.2). In spite of the extremely narrow ranges of controlled cooking variables, a large database and combined statistical methods (analysis of variance, parallel coordinates, principal component analysis) allowed the distinction of the optimal range limits of studied technological factors determining the tested pulp s properties. The mechanical and strength testing of the pulp s sheets showed that the analyzed time, temperature and liquor- to- wood ratio influenced the CMT, SCT, Tear strength and to some extent Burst strength. Keywords: NSSC, pulp, liquor ratio, statistical methods Introduction Neutral sulfite semi chemical (NSSC) pulp accounts for approximately 5-7% of the overall production of paper pulp. The main advantage of this type of pulp are that it provides fibers of a very high quality for special types of paper, especially the so-called microwave for multilayer cartons (corrugated medium). NSSC pulp may be used as such or it may also partially replace other pulp from wood, Jan Bocianowski, Poznan University of Life Sciences, Poland Krzysztof Joachimiak, Poznan University of Life Sciences, Poland Adam Wójciak, Poznan University of Life Sciences, Poland
18 18 Jan Bocianowski, Krzysztof Joachimiak, Adam Wójciak lignocellulosic residues or old paper during the manufacture of various products such as printing and writing paper, tissue paper, cardboard, bag grades and other products [Farrington, Hickey 1989; Odom 1991; Myers et al. 1996; Area et al. 1997; Ahmadi et al. 2010]. Although the relatively traditional process of neutral sulfite pulping maintains its industrial importance, the question of the recovery of spent liquor chemicals still needs to be resolved. Most producers use the cross recovery system with Kraft spent liquors, while others have fluidized bed combustion systems. In every case, the problem is the relatively high inorganic to organic ratio in spent liquors and that is why the liquor- to- wood ratio is (among others) such an important technological factor [Area et al. 2001a, 2001b]. The pulping process of various wood and non-wood plants has been analyzed with the use of many mathematical models and statistical methods allowing the estimation of pulp quality in terms of process variables and determining optimal operating conditions. Most of these models were based on delignification kinetics and only a few took into consideration the influence of process variables on pulp quality [Jimenez et al. 1999, 2000, 2004]. Moreover, most of these models were established by applying the results of laboratory scale experiments only and with arbitrary admitted ranges of variables. The aim of this work was to examine the influence of extremely low changes in industrial operating variables (liquid/solid relation, temperature, pulping time) on the mechanical and strength properties of the NSSC pulp paper sheets (CMT, SCT, Burst and Tear strength) obtained, using mathematical and statistical methods. Materials and methods Industrial chips obtained from birch trees growing in Poland (Betula verrucosa) were used as the raw material. The chips included the following fractions: >O45 1.1%, >II8 5.7%, >O7 91.0%, >O3 2.1%, the rest 0.1% (O and II mark holes and slots, diameter in mm according to Brecht-Holl classification) [SCAN CM 40:01]). The birch wood contained 82.4% holocellulose, 27.31% α-cellulose, 31.44% lignin by dry matter weight. All cooking experiments were carried out on a real industrial NSSC production line with a controlled capacity and with an 82% average process yield. Pulping: continuous cooking, a vapour-gas phase technique, pulp kappa no 120±10. The most important units composing the pulp line were: an atmospheric chip bin, a plug screw feeder, a pressurized impregnator, a continual digester and a cooked chip disintegration refiner. After the release of air at the chip bin, the wood was squeezed by the screw feeder and then pushed to the pressurized impregnator. The impregnator was connected directly with the digester and that construction had the
19 The influence of process variables on the strength properties of NSSC birch pulp. Towards the same operational pressure 9.2 bar. The initial cooking liquor concentration was the same during all the experiments (165g/dm 3 Na 2 SO 3 ; 50g/dm 3 Na 2 CO 3 ). After processing at different conditions of liquor-to-wood ratio, at a fixed temperature of 178/179 C and cooking time of 14.5 min, the chips were defibred by a double- -disc disintegration refiner with a concentration of 36% (Andritz Sprout Bauer, 2.5 MW, optimal energy consumption 130 kwh/mg). The retention time of the line was calculated for approx. 1 hour. For each studied technological parameter (liquor-to-wood ratio, time and temperature of cooking), the obtained pulp samples were refined in a PFI laboratory mill to reach four Schopper-Riegler degrees: 20 ºSR, 25 ºSR, 30 ºSR, 35 ºSR. After Schopper-Riegler freeness tests from all pulp samples, hand sheets were made. From the dried and conditioned paper sample, four strength properties were examined: SCT short crush test [EN/ISO 9895], CMT Concora medium test [EN/ISO 7263], Tear strength [EN 21974] and Burst strength [EN/ISO 2758]. Liquor-to-wood ratio Pulp samples were collected after cooking with 1.4, 1.5, 1.7, 1.8 and 2.2 liquid- -to-wood ratios (l/w). The other technological variables (temperature and cooking time) were determined with l/w ratio 1.8 and The total number of observations used for l/w ratio analysis: CMT (373), SCT (642), Burst strength (442), Tear strength (150). The number of observations resulted from the different number of paper sheets qualified as convenient for strength analysis. In particular, the compression strength of paper (CMT, SCT) is measured at high grammage 130 g/m 2 and all the paper sheets which did not have an appropriate weight per unit area and caused the risk of bad readings were not analyzed. Statistical analysis Firstly, the normality of the distribution of the CMT, SCT, Tear and Burst strength was tested using Shapiro-Wilk s normality test [Shapiro, Wilk 1965]. A two-way analysis of variance (ANOVA) was carried out to determine the effects of liquid- -to-wood ratio, Schopper-Riegler numbers and interaction of liquid-to-wood ratio Schopper-Riegler numbers on the variability of CMT, SCT, Tear and Burst strength development. The least significant differences (LSDs) for each variable were calculated and, on this basis, homogeneous groups for the analyzed variables were determined. The relationship between CMT, SCT, Tear and Burst strength were estimated on the basis of correlation coefficients. The relationship between the analyzed properties was presented in the form of scatter-plot [Kozak et al. 2010]. The application of principal components analysis (PCA) made it possible to find on the plane the graphic dispersion of pairs of liquid-to-wood ratio
20 20 Jan Bocianowski, Krzysztof Joachimiak, Adam Wójciak and Schopper-Riegler degrees characterized with regard to all properties treated together. The parallel coordinate plot is proposed as an efficient tool for liquid-to- -wood ratio Schopper-Riegler freeness interaction visualization [Kozak 2010]. Analysis of the data was performed using the statistical package GenStat v [GenStat 2007]. Results and discussion Because of the variety of effects of chemical and mechanical treatment on final product properties, it becomes especially important for the papermaking process to understand how the different technological factors affect the technological practice of NSSC pulping. Little differences among the controlled operating conditions on the industrial-scale production line (liquor-to-wood ratios, temperature, cooking time) make the choice of both appropriate statistical tools and the possession of a sufficient database very important. As the most appropriate indices for the evaluation of the technological efficiency, first of all CMT [N] and SCT [kn/m]were admitted, then Burst [kpa] and Tear [mn]. Literature in this area is very limited but it is generally accepted that the most important property of fluting papers is the compression strength: CMT and SCT. CMT is a measure of opposition to the crushing of the flute after it is developed by the corrugator. The use of the SCT strength index has been found to be an accurate indicator of refining effect and is widely used because of the simplicity of testing. Burst strength is the result of a combination of many factors, depending amongst others on the amount and preparation of the fiber present. It is a useful indicator of strength property regardless of paper grade. The tearing resistance, which is a way of evaluating the crack sensitivity of the paper, is a strength dimension of central importance for all paper [Ek et al. 2009]. Initially, the significant differences were analysed between all the results of the strength properties measured, which were obtained for different l/w ratios and different Schopper-Riegler (SR) freeness numbers. The essential point for further studies is that the analysis of variance (table 1) showed that when all the studied strength properties (CMT, SCT, Burst and Tear) were tested together, they also differed significantly with regard to l/w ratio and SR freeness. The same statistics allowed an evaluation of the possible interaction between the studied technological parameters and analytical data e.g. l/w and SR freeness. Although, in relation to SCT values, the interaction was weaker than in the case of CMT, Burst and Tear, the analysis of variance proved that l/w and PFI refining (SR tests) together influenced the pulp s strength properties (table 1).
21 The influence of process variables on the strength properties of NSSC birch pulp. Towards the Table 1. Mean squares from the analysis of variance for CMT, SCT, Tear and Burst strength Tabela 1. Średnie kwadraty z analizy wariancji dla oznaczeń CMT, SCT, oporu przedarcia i przepuklenia Source of variation Źródło zmienności l/w moduł SR SR l/w SR moduł SR Residual Błąd df ss CMT CMT ms śk df ss SCT SCT ms śk df ss Burst Przepuklenie ms śk df ss Tear Przedarcie *** *** *** *** *** *** *** *** *** ** *** *** ms śk l/w liquid-to-wood ratio moduł moduł cieczy warzelnej SR freeness in SR (Schopper Riegler) SR smarność w stopniach SR (Schopper Riegler) df degrees of freedom ss stopnie swobody ms mean squares śk średnie kwadraty ** significant at 0.01 level ** istotne na poziomie 0,01 *** significant at level *** istotne na poziomie 0,001 Table 2. The correlation matrix for CMT, SCT, Tear and Burst strength Tabela 2. Macierz współczynników korelacji dla oznaczeń CMT, SCT, oporu przedarcia i przepuklenia CMT CMT SCT SCT Correlation Korelacje Burst Przepuklenie Tear Przedarcie CMT CMT 1 SCT SCT 0.538* 1 Burst Przepuklenie 0.839*** 0.665** 1 Tear Przedarcie * * 1 *,** or ***show increasing significance of correlation between analysed strength properties. Minus before value indicates reverse correlation. *,** lub *** oznaczają wzrastającą istotność korelacji pomiędzy analizowanymi właściwościami wytrzymałościowymi. Znak ujemny przed wartościami wskazuje korelację odwrotną.
22 22 Jan Bocianowski, Krzysztof Joachimiak, Adam Wójciak Fig. 1. Scatter plot matrix for CMT, SCT, Tear and Burst strength Rys. 1. Wykresy rozproszenia dla oznaczeń CMT, SCT, oporu przedarcia i przepuklenia In order to recognize more closely the mutual tendencies and interactions of the studied technological factors and pulp properties, correlations between the pairs of tested strength features (table 2, fig. 1) were analyzed. As seen in table 2 and on the scatter plot (fig. 1) the strongest correlation appeared between the CMT and Burst (r = 0.839). A weaker correlation showed SCT and Burst (r = 0.665). The less significant (P < 0.05) correlations were observed for SCT and CMT (r = 0.538), then Burst and Tear (r = ) and at the end for SCT and Tear (the two last pairs showed negative correlations). The most important point is that in spite of the scattering data, a positive correlation between the increasing CMT, SCT and Burst values (fig. 1) is confirmed. The correlations were independent of the refining energy applied during sample preparation. The pulp freeness is directly related to several processes that occur during refining such as external fibrillation, fiber shortening and fines creation. The other known result of refining is the internal fibrillation reflecting the increase in fiber swelling, which is caused by the delamination of cell walls and the general growth
23 The influence of process variables on the strength properties of NSSC birch pulp. Towards the of the capacity of pulp to retain water [El-Sharkawy et al. 2008]. During the next step significant differences were observed between the separate results of strength properties gained for the whole range of studied l/w ratios, and with regard to the increasing SR freeness numbers (table 3). The repeating letters (capital for mean values) inserted according to the results of the pulp strength analysis prove the lack of significant differences. As could be expected, the CMT and SCT values generally increased with the higher SR numbers [Kim, Jo 2000]. Thus, refining in a PFI mill increased the tested paper sheet strength by influencing the surface area of the fibers allowing their optimum papermaking properties to develop. The mean values (calculated for all SR freeness range) of CMT showed significant differences between the 1.4 and 1.5 l/w ratios used. Table 3. Mean values and coefficients of variation for CMT, SCT, Tear and Burst strength Tabela 3. Wartości średnie i współczynniki zmienności dla CMT, SCT, oporu przedarcia i przepuklenia l/w moduł mean śr CMT [N] CMT SRº cv wz mean śr cv wz mean śr cv wz mean śr cv wz mean śr gh abcdef abcdef abcde B bcdef abcd abc abcde A defg gh abcdef ab B h efgh abcd ab B fgh abcdef cdef a AB mean śr 231T 242.8S 251.7R 258.8P SCT [kn/m] SCT SRº l/w moduł mean śr cv wz mean śr cv wz mean śr cv wz mean śr cv wz mean śr efg defg bcd cdef B fg efg cdef defg B g efg defg cde B efg defg abcd abcd B cdef a abc ab A mean śr 4.452S 4.737R 4.965P 4.981P Burst [kpa] Przepuklenie
24 24 Jan Bocianowski, Krzysztof Joachimiak, Adam Wójciak Table 3. Continued Tabela 3. Ciąg dalszy l/w moduł SRº mean cv mean cv mean cv mean śr wz śr wz śr wz śr cv wz mean śr h cdef abcde abc B fg def abcde abcd AB bcde fg abcd a AB g def abcd ab B efg abcde abc ab A mean śr 422.4T 459.8S 497.7R 518.2P Tear [mn] Przedarcie SRº l/w moduł mean śr cv mean cv mean cv mean wz śr wz śr wz śr cv wz mean śr a abc cdef bcde A abcd ef ef f C def def def ef C abc def def bcde BC ab ef cdef cdef B mean śr 809.3P 732.4R 725.2R 736.2R l/w liquid-to-wood ratio l/w moduł cieczy warzelnej SR freeness in SR (Schopper Riegler) SR smarność w stopniach SR (Schopper Riegler) abc repeated small letters show no significant difference between variables abc powtarzające się małe litery wskazują na brak istotnych różnic pomiędzy zmiennymi ABC repeated capitals show no significant difference between mean values of variables ABC powtarzające się duże litery wskazują na brak istotnych różnic pomiędzy średnimi wartościami zmiennych mean mean value śr wartość średnia cv coefficient of variation wz współczynnik zmienności In opposition to this, the average SCT values differed significantly at higher l/w ratios between 1.8 and 2.2. Taking into consideration the preferences for aiming at high CMT pulp indices as seen in industrial practice, the results seem to show the possibility of pulping with a low l/w ratio. However, when plotting the analyzed data as a function of l/w ratio the same tendencies for these proper-